\. ^ ^ frit ^ ^ ^^ '"^ ^ I I ■ ^ ^ Ir ' ^"^i mmw ^ ^ ,l| f' «'l 11 x^ Y, '\IW '^-^## 'W J> '^"i;5l!||il! rlllllil ^ 4'^ -^^^ V THE JOURNAL OF GEOLOG\ A Semi-Quarterly Magazine of Geology and Related Sciences / Editors T. C. ClIA.MliKRLIX R. D. SALISBURY R. A. F. PENROSE, Jr. Geographic Geology Economic Geology J. 1'. IDDINGS C. R. VAN RISE Petrology '■. Pre-Camhrian Geology W. H. HOLMES, Anthropic Geology Associate Editors SIR ARCHIBALD GEIKIE Great Brifai 77 H. ROSENBUSCH Germa7iy CHARLES BARROIS France ALBRECHT PENCK Austria HANS REUSCH N' or way GERARD DE GEER Sweden GEORGE M. DAWSON Canada O. A. DERBY, Brazil JOSEPH LE CONTE University of Calijoriiia C. D. WALCOTT U. S. Geological Suri'ey G. K. GILBERT Washington, D. C. H. S. WILLIAMS Vale University j. C. BRANNER Lelafid Stanford University I. C. RUSSELL University of Michigan WILLIAM B. CLARK fohns Hopkins Unii'ersity VOLUME V CHICACiO C1)c sauiberisitp of iSTiiicago IJress iSg; PRINTED AT tibe innlverslt^ of Cbtcago press Contents of J^olume P NUMBER I. PAGE Comparison of the Carboniferous and Permian Formations of Kan- sas AND Nebraska. Charles S. Prosser . . - - - i Evidences of Recent Elevation of the Southern Coast of Baf- fins Land. Thomas L. Watson ------- 17 PlALIAN PeTROLOGICAL SKETCHES. III. ThE BrACCIANO, CeRVETERI, AND TOLFA Regions. Henry S. Washington - . . - 34 Mode OF Formation of Till as Illustrated by the Kansas Drift of Northern Illinios. Oscar H. Hershey ----- 50 The Geology of the San Francisco Peninsula. Harold W. Fair- banks ------------ 63 Editorial - . - - 77 Reviews : Manual of Determinative Mineralogy, with an Introduction on Blowpipe Analysis, by George J. Brush (review by O. C. P^arrington), 86; The Dinosaurs "of North America, bv Othniel Charles Marsh (review by E. C. C), 87. --------- 85-88 Abstracts : Papers read at the Washington Meeting of the Geological Society of America; Glacial Observations in the Umanak District, Greenland, by George H. Barton, 89 ; The Origin and Relations of the Grenville and Hastings Series in the Canadian Laurentian, by Frank D. Adams and Alfred E. Barlow, 92 ; The Origin and Age of the Gypsum Deposits of Kansas, by G. P. Grimsley, 95 ; The Cor- nell Glacier, Greenland, by Ralph S. Tarr, 94 ; Unconformities in Martha's Vineyard and Block Island, by J. B. Woodworth, 96; Homol- ogy of Joints aiid Artificial F'ractures, by J. B. Woodworth, 97; Notes on Rock-weathering, by George P. Merrill, 98 ; Notes on the Potsdam and Lower Magnesian F^ormations of Wisconsin and Minnesota, by Joseph F. James, 99 ; Preliminary Note on the Pleistocene History of Puget Sound, by Bailey Willis, 99; The Leucite Hills of Wyoming, by J. ¥. Kemp, 100; The Pre-Cambrian Topography of the Eastern Adiron- dacks, byj. F\ Kemp, loi; Stratigraphy and Paleontology of the Laramie and Related Formations in Wyoming, by T. W. Stanton and F. H. Knowlton, 102 ; A Complete Oil-well Record in the McDonald FMeld between the Pittsburg Coal and the Fifth Oil Sand, by I. C. White, 103 ; A Note on the " Plasticity " of Glacial Ice, by Israel C. Russell, 104 ; Work of the U. S. Geological Survey in the Sierra Nevada, by H. W. Turner, 105; Shore Lines of Lake Warren and of a Lower Water Level in Western-central New York, by H. L. Fair- child, 106 ; Principal Features of the Geology of Southeastern Wash- ington, by Israel C. Russell, 107; Old Tracks of F:rian Drainage in Western New York, by G. K. Gilbert, 109 ; The Physical Nature of the Problem of General Geological Correlation, by Charles R. Keyes, no ; Modified Drift in St. Paul, Minnesota, by Warren Upham, iii. 89-112 iv CONTENTS NUMBER 11. I'AGE Professor Geikie's Classification of the North European Glacial Deposits. K. Keilhack - - - - - - - - - ii3 The Average Specific Gravity of Meteorites. Oliver C. Farrington 126 Drift Phenomena in the Vicinity of Devil's Lake and Barakoo, Wisconsin. Rollin D. Salisbury and Wallace Walter Atwood - 131 Comparison of the Carboniferous and Permian Formations of Nebraska and Kansas. II. Charles S. Prosser - - - - 148 The Geology of the San Francisco Peninsula. Andrew C. Lawson - 173 Note on the Geology of Southwestern New England. William H. Hobbs ------------ 175 Studies FOR Students : Deformation of Rocks. V. C. R. Van Hise - 178 Editorial - - - - - - 194 Reviews : Geology and Mining Industry of the Cripple Creek District, Colorado, by Whitman Cross and R. A. F. Penrose, Jr. (review by Arthur Winslow), 197; Glacier Bay and Its Glaciers, by Harry Field- ing Reid (review by Israel C. Russell), 203 ; Water Resources of Illinois, by Frank Leverett (review by W. H. Norton), 206 ; The Geology of Santa Catalina Island, by William Sidney Tangier Smith (review by F. L. Ransome), 208 ; Geology of the Castle Mountain Mining District, Montana, by W. H. Weed and L. V. Pirsson (review by H.Foster Bain), 210; The Ancient Volcanic Rocks of South Mountain, Pennsylvania, by Florence Bascom (review by J. P. Iddings), 213. ------------ 197-216 Abstracts : Upper Cretaceous of the Northern Atlantic Coastal Plain, by Wm. B. Clark, 217 ; Age of the Lower Coals of Henry County, Mis- souri, by David White, 218; Crater Lake, Oregon, by J. S. Diller, 219; Nipissing-Mattawa River, the Outlet of the Nipissing Great Lakes, by B. F. Taylor, 220 ; The Grain of Rocks, by Alfred C. Lane, 222 ; A Study of the Nature, Structure, and Phylogeny of Dasmonelix, by E. H. Barbour, 223 -------- -217-224 Recent Publications, ---------- 224-228 NUMBER III. Glacial Studies in Greenland. X. T. C. Chamberlin - - - 229 Italian Petrological Sketches. IV. The Rocca Monfina Region. Henry S. Washington --------- 241 Are the Bowlder Clays of the Great Plains Marine? George M. Dawson ------------ 257 The Bauxite Deposits of Arkansas. J. C. Branner - ^"^ - - - 263 Editorial ------------ 290 Reviews : Some Recent Papers on the Influence of Granitic Intrusions upon the Development of Crystalline Schists (review by Frank D. Adams), 293; Glaciers of North America, by Israel C. Russell (review by T. C. CONTENTS V PAGIi C), 302; Former Extension of Cornell Glacier near the Southern End of Melville Bay, by Ralph S. Tarr (review by T. C. C), 303; Report on the Valley Regions of Alabama, Part'I, by Henry McCalley (review by Stuart Weller), 307; Final Report on the Geology of Minnesota, Vol. Ill, Part II (review by Stuart Weller), 308; Bulletins of American Pal- luontology. Vol. I (review by Stuart Weller), 309; Eocene Deposits of the Middle Atlantic Slope in Maryland, Delaware, and Virginia, by Wil- liam Bullock Clark (review by Charles R. Keyes), 310; The Elevated Reef of Florida, by Alexander Agassiz, with Notes on the Geology of Southern Florida, by Leon S. Griswold (review by J. Edmund Wood man), 312; Correlation of Erie-Huron Beaches with Outlets and Moraines in Southeastern Michigan, by F. B. Taylor (review by C. H. Gordon), 313; Elementary Geology, by Ralph S. Tarr (review bv Henry B. Klimmel), 317. - - - - - - - - - 293-318 Abstracts : The Solution of Silica under Atmospheric Conditions, by C, Willard Hayes, 319 ; The Crystalline and Metamorphic Rocks of Northwest Georgia, by C. Willard Hayes and Alfred H. Brooks, 321 ; T he Age of the White Limestone of Sussex County, New Jersey, by J. E. Wolff and Alfred H. Brooks, 322 ; Erosion at Base-level, 322, and Origin of Certain Topographic Forms, by M. R. Campbell, 323; Dikes in Appalachian Virginia, by N. H. Darton, 324. . . . . 319-324 NUMBER IV. The Last Great Baltic Glacier. James Geikie ----- 325 The Post-Pleistocene Elevation of the Inyo Range, and the Lake Beds of Waucobi Embayment, Inyo County, California. Charles D. Walcott ---------- 3^0 Italian Petrological Sketches. V. Summary and Conclusion. Henry S. Washington --------- 3^^ Variations of Glaciers. H. Harry Fielding Reid - . . - 37}j A Sketch of the Geology of Mexico. H. Foster Bain - - - 384 Editorial- --------.... 3gi Reviews : Some Queries on Rock Differentiation, by G. F. Becker (review by C. F. Tolman Jr.), 393 ; An Introduction to Geology, by W. B. Scott (review by R. D. S.), 398 ; Missouri Geological Survey, Vol. XI ; Clay Deposits, by H. A. Wheeler (review by H. Foster Bain), 399 ; The Uni- versity Geological Survey of Kansas, by Erasmus Haworth and Assistants (review byS. W.), 400; Preliminary Report on the Mar- quette Iron-Bearing District of Michigan, by Charles Richard Van Hise and William Shirley Bayley, with a Chapter on the Republic Trough, by Henry Lloyd Smyth (review by U. S. Grant), 402. - - 393-404 Abstracts : Geological Atlas of the United States : — P^olio, 30 ; Vellowstone National Park, Wyoming, 1S96, 405 ; Folio 24, Three Forks, Montana, 1896, 407 : Folio 29, Nevada Cit}-, special folio, California, 1896, 409 ; Folio 28, Piedmont, West Virginia-Maryland, 1896, 411; Folio 23, Nomini, Maryland-Virginia, 1896, 413; Folio 26, Pocahontas, Vir- ginia-West Virginia, 1896, 414; P^olio 25, Loudon, Tennessee, 189b, 416; P'olio 27, Morristown, Tennessee, 1896,417. - . . . 405-419 Recent Publications 419-420 VI . CONTENTS NUMBER V. PAGE Moraines of Recession and Their Significance in Glacial Theory. Frank Bursley Tajdor --------- 421 The Eruptive Rocks of Mexico. Oliver C. Farrington - - - 466 The Stratigraphy of the Potomac Group in Maryland. William Bullock Clark and Arthur Bibbins ------- ^yg Studies for Students : Comparative Study of Palaeontogeny and Phylog- eny. James Perrin Smith --------- 507 Editorial ------- ^25 Reviews : The Bedford Oolitic Limestone in Indiana, b}' T. C. Hopkins and C. E. Siebenthal (review by J. C. Branner), 529 ; Ancient Volcanoes of Great Britain, by Archibald Geikie (review by J. P. Iddings), 531 ; The Submerged Valleys of the Coast of California, U. S. A., and of Lower California, Mexico, by George Davidson (review by W.S.Tan- gier Smith), 533. - - " - - - - - - - - 529-534 Recent Publications ---------- 535-540 NUMBER VI. The Newark System of New Jersey. H. B. Iviimmel - - - - 541 The Topography of California. Noah Fields Drake - - - - 563 A Comparative Study of the Lower Cretaceous Formations and Faunas of the United States. Timothy W. Stanton - - 579 Correlation of the Devonian Faunas in Southern Illinois. Stuart Weller ------------ 525 Editorial -----..---.- 606 Reviews: Geological Survey of Canada, Annual Report, Vol. VUI, 1895, by G. M. Dawson, Director, Ottawa, 1897 (review by T. C. C), 641 ; Iowa Geological Survey, Vol. VI (review by T. C. C), 642 ; Geology and Natural Resources of Indiana, Twenty-first Annual Report (review by T. C. C), 644 ; Geological Survey of Alabama, by Eugene Allen Smith, State Geologist (review by T. C. C), 646 ; First Report of the Geological Commission of the Colon}' of the Cape of Good Hope (review by T. C. C), 647 ; North Carolina and Its Resources, issued by the State Board of Agriculture (review by T. C. C), 648 ; Proceedings of the Iowa Academy of Sciences (review by T. C. C), 648 ; Bulletin of the Minnesota Academy of Natural Sciences, Vol. IV,> No. I, Part I, C. W. Hall, Editor (review by T. C. C), 648 ; Proceedings of the Davenport Academy of Natural Sciences, Vol. VI, 1889 to 1897 (review by T. C. C), 649 ; Stone Implements of the Poto- mac-Chesapeake Tidewater Province, by William H. Holmes (review by T. C. C), 649 ; Glacial Observations in the Umanak District, Greenland, by George H. Barton, Report B. of the Scientific Work of the Boston Party of the Sixth Peary Expedition to Greenland (review by T. C. C), 650 ; Seventeenth Annual Report of the United States Geological Survey, Parts I, II and HI, by Charles D. Walcott, Director (review by T. C. C), 651. --------- 641-652 CONTENTS VI 1 NUMBER VII. PAGE A GROII' ok IIVI'OTHESES BEARING ON Cl.IMATIC CHANGES. T. C. Chamberlin ----------- 653 An Analcite Basalt from Colorado. Whitman Cross - - - 684 Studies on the So-called Porphyritic Gneiss of New Hampshire. Reginald Aldworth Daly --------- 694 The Measurement of Faults. J. Edward Spun- ----- 723 The Drift and Geologic Time. H. M. Bannister - . - - 730 On the Presence of Problematic Fossil Medus.t^ in the Niagara Limestone of Northern Illinois. Stuart Weller - - - 744 Editorial ------------- 752 Reviews : The Unpublished Papers of the Geological Survey of Brazil (Boletim do Musen Paraense), Vol. II, No. 2, 756; The Devonian Fauna of the Reo Maecuru, by Dr. F. Katzer (review by John C. Bran- ner), 757; Report of the United States Deep Waterways Commission, James B. Angell, John F. Russell, Lyman E. Cooley (review by F. L.), 758; The Former Extension of the Appalachians acict-s Mississippi, L-ouisiana, and Texas, by Professor J. C. Branner (review by A. H. Purdue), 759 ; Maryland Geological Survey, Vol. I, by William Bullock Clark, State Geologist (review by R. D. S.) 760. - - - 756-761 Recent Publications ---------- 761-766 NUMBER VIII. The Geologic Relations of the Martinez Group of California AT the typical locality\ John C. Merriam - . . - 767 Studies in the So-called Porphyritic Gneiss of New Hampshire. II. Reginald Aldworth Daly 776 Supplementary Hypothesis Respecting the Origin of the Loess of the Mississippi Valley. T. C. Chamberlin - - - - 795 Cryptodiscus, Hall. Stuart Weller ------- 803 A Note on the M!(;ration of Divides. Wm. Sidney Tangier Smith - 809 Discovery of Marine Jurassic Rocks in Southwestern Texas. ¥. W. Cragin ----------- 813 Andendiorite in Japan. C. Iwasaki ------- 821 SlUDIES IN THE DrIFTLESS REGION OF WISCONSIN. G. H. Squicr - 825 Studiks for Students: The Method of Multiple Working Hypotlieses. r. C. Chamberlin ---------- 837 Ediiokial .-.--------■ 849 viii CONTENTS PAGE Reviews : The Glacial Lake Agassiz, by Warren Upham (review by T. C. C), 851 ; Catalogue of the Tertiary Mollusca in the Department of Geology, British Museum (Natural History); Part I, The Australasian Tertiary Mollusca, by George F. Harris (review by Wm. B. Clark), 853 ; Transactions of the American Institute of Mining Engineers, Vol. XXVI (review by C. F. Tolman, Jr.), 854; The Law of Mines and Mining in the United States, by Daniel Moreau Barringer and John Stokes Adams (review by T. C. C), 858 ; The Science of Brick- making, by George F. Harris (review by Wm. B. Clark), 85Q. 851-859 THE JOURNAL OF GEOLOGY JANUARY-FEBRUARY, 1S97 COMPARISON OF THE CARBONIFEROUS AND PER- MIAN FORMATIONS OF NEBRASKA AND KANSAS. INTRODUCTION. After having devoted three summers to the study of the Upper Carboniferous and Permian formations of Kansas, the writer recently reviewed quite rapidly the similar formations of southeastern Nebraska, and in this paper will give an outline of their general correlation. No particular formation was traced from Kansas into or across Nebraska, nor were the stratigraphic details worked out carefully in Nebraska; still the work done, limited as it was in character, showed the geologic age and general relation of the rocks of Kansas to those of Nebraska, and as various opinions are entertained regarding these relations it seems important to give a brief description of this work. NEMAHA COUNTY. Wabaunsee formation. — It is desirable to select the three counties — Nemaha, Otoe, and Cass — in which the greater part of the field work was done, for the general subdivisions of this paper and then under each county to describe its geologic formations, beginning with the oldest. The exposures in the eastern part of Nemaha county along the Missouri River were very fully described by Meek,' sections 'Final Rep. U. S. Geol. Sur. Nebraska and adjacent Territories, Pt. II, Paleontology, pp. 109-1 14. Vol.. v., No. I. I 2 CHARLES S. PROS SEE were given, together with lists of fossils from the vicinity of the towns, Peru, Brownville, and Aspinwall,^ and the following species were reported by him from these localities : 1. Neuropteris hirsuta Lesqx. 2. Neuropteris Loschii Brgt. ^ 3. Fusulina cylmdrica Fischer. 4. Productus nebrascensis Owen. 5. Productus sejnireticulatus (Martin) de Kon. 6. Derbya crassa (M. & H.) Hall & Clarke. 7. Meekella striata -costata (Cox) White & St. John. 8. Spirifer ca?neratus Morton. 9. Spirifer {JVIartinid) piano- convexus Shum. 10. Spiriferina kentuckensis Shum. 11. Athyris (^Seminuld) subtilita (Hall) Newb. 12. Nuculana bellistriata (Stevens) Meek. 13. Myalina perattenuata M. & W. 14. Aviculopecten IV/iitei Meek. 15. Edniondia aspinwallensis Meek. 16. Alloris?7ia subcuneatitvi M. & H. 17. Bellerophon percarinatus Qoxi. 18. Straparollus {Euompkalus) catilloides (Con.) Y^eye?, = Eiiompkaltis riigosus Hall. ig. Nautilus occidentalis Swallow. 20. Deltodus (?) angularis N. & W. 2 1 . Productus longispinus Sowb. = Nor. & Pratt. 22. Chonetes granulifera Owen. 23. Bellerophon carbonarius Cox. 24. Bellerophon Alontfortianus Nor. & Pratt. 25. Productus pertenuis Meek. 26. Productus cor a d'Orbigny. 27. Myalina subguadrata Shum. 28. Aviculopecten occidentalis (Shum) M. & W. The fossils of the above list are nearly all well-known Carbonif- erous species as was stated by Meek.^ Marcou referred the rocks along the Missouri River in this county to the Dyas 3 (= Permian) but this view was thoroughly 'Fin. Rep. U. S. Geol. Sur. Nebraska, etc., pp. 109, no, 112, and the "Tabular list, illustrating the geological and geographical range of the fossils of eastern Nebraska," pp. 124-127. ^ Ibid., p. 114. 3 Bull. Soc. Geol. France, 2^ serie, t. XXI, 1864, p. 134. CA RB ONIFER O US A ND PERM I A N FORMA TIONS 3 disproved by Meek, who criticised Professor Marcou's corre- lation in the following words : " but upon what evidence he [Professor Marcou] does not say, nor is it apparent to anyone who regards fossils as any guide in identifying rocks, as those found here consist of the same forms constituting the group so often mentioned as characterizing the Coal Measures of the Western States."' Meek regarded the exposures along the Missouri River in Nemaha county as geologically above the Nebraska City beds, for he said : "At Nebraska City, and below there, at Otoe City, Brownville, and Aspinwall, there are, perhaps, altogether, near 150 to 200 feet of additional strata, all holding probably a posi- tion above the geological horizon of the top of the boring at Nebraska City."^ On the contrary, Dr. Hayden regarded the Aspinwall section as somewhat older than that at Nebraska City, for he said: "The rocks at Aspinwall are all geologically at a little lower horizon than the Nebraska City beds, and mostly beneath the Brownville beds." 3 In the Aspinwall section Meek reported a stratum of coal one foot ten inches in thickness, eighteen feet above low water of the Missouri River and another six-inch stratum eleven and one-half feet higher. No oppor- tunity was afforded me to study this section, hence I am unable to state whether the coal lies at the base of the Wabaunsee formation or not ; but all the rocks of this county exhibited along the bluffs of the Missouri River belong to that geologic series usually termed the Upper Coal Measures, for which Dr. Keyes has proposed the very appropriate name of Missourian.'' 'Fin. Rep. U. S. Geol. Sur. Nebraska, etc., p. 114. "Ibid., p. 137 ; also see p. 123, where a similar statement occurs. 3 Ibid., p. 17. •• Am. Geol., Vol. XVIII, Jul}'' 1896, p. 25 ; also see Iowa Geol. Surv., Vol. 1, 1893, p. 85. The writer would suggest the following classification for the Upper Coal Measures of the Western Interior province : Cottonwood formation (Prosser) Missourian series (Keyes), - - - -j Wabaunsee " " The formations below the Wabaunsee have been recently studied by Drs. Keyes and Haworth, to whom the writer leaves the selection of names for the lower part of the series. 4 CHARLES S. PROSSER With the exception of the region adjacent to the Missouri River, the geology of Nemaha county was not described by Meek or Hayden. About twelve miles northwest of Aspinwall and some 235 feet above Nemaha City, which is near the level of the Missouri River, and near the center of Nemaha county, is Auburn, the county seat. To a large extent, the rocks of this region are con- cealed by the thick deposit of loess, but there are occasional exposures along the streams or at favorable places on the bluffs or hills. In the vicinity of Auburn are a few outcrops of rocks which are referred to the upper part of the Wabaunsee forma- tion. Along the highway, one and one-fourth miles directly west of Auburn, the rocks show for a short distance. The top of the outcrop is some thirty-five feet higher than the Auburn hotel, or approximately 270 feet above the Missouri River, and is just west of the Sheridan cemetery on the east side of the South Fork of the Nemaha River. The following beds occur in descending order : Feet 5. Rough yellowish limestone - - - - - -. - i =65 4. Soft light gray to yellowish shales - - - - - 2 = 64 3. Drab hard limestone that weathers to a light gray color. A few fossils — Aviculopecten occidentalis (Shum.) M. & W. and frag- ments of other species - - - - - - - -2^62 2. Light gray and greenish shales, the lowest very white - [ , , I. Covered to the level of South Fork of the Nemaha River - - S These rocks are considered to belong in the upper part of the Wabaunsee formation ; for on the upland two and one-half miles farther west and perhaps fifty feet higher is a ledge of light gray limestone composed to a considerable extent of specimens of Fusiilina cylindrica Fischer which the author regards as the Cottonwood limestone of Kansas. There are a number of out- crops and quarries of this limestone on the upland in the western central part of Nemaha county, as well as a few exposures of the underlying rocks, but as they are only a few miles northwest of this eastern outcrop of the Cottonwood limestone they may be described in connection with it. CA RB ONIFER O US A ND PERM J A X FORMA TIOXS 5 Cotto/nvood formation.— The most eastern outcrop of the Cot- tonwood limestone seen in Nemaha county is the one which is known as the Van Court or Keyes quarry and is three and three- fourth miles directly west of Auburn. Unfortunately my barom- eter at the time of my visit was not reading accurately ; but its altitude is estimated as some 345 feet above the Missouri River. The Cottonwood limestone forms a massive light gray stratum four feet thick, moderately hard and filled with large numbers of Fiisidina. When freshly broken the Fiisulinas though slightly darker in color than the limestone are less prominent than on the weathered surface and are only conspicuous when seen through a magnifying glass. On the weathered surface of the rock the shells resist decay longer than the limestone and stand out prominently. One who has visited the large quarries of Cottonwood limestone in northern and central Kansas would be immediately impressed with this very striking similarity. Few, if any, "fossils other than FusuH?ia cylindrica Fischer occur in this^ limestone ; only an occasional fragment of a spine of ArchcBocidaris or a bit of shell was noticed. The section at the Keyes quarry is as follows : Kt. In. Ft. In. 3. Shaly limestone containing Athyris {Scininula) subtiUta (Hall) Newb. and a few other fossils - - - - ' - i = Q 2. Light gray shales to shaly limestone - - - i 9 = 5 9 I . Cottoniuood limestone, light gray massive Fusulina limestone 4 = 4 The Nemaha county quarry which is the most extensively worked is the next well-exposed section of the Cottonwood limestone and is one and one-fourth miles west of the Keyes quarry. It is located by the side of the Burlington and Mis- souri River Railroad, f^ve miles due west of Auburn and one mile north of Hickory Grove. The average of three barometric readings makes the bottom of the quarry 130 feet higher than Auburn or approximately 365 feet above the Missouri River. This is twenty feet higher than the base of the Keyes quarry which gives a dip directly east of twenty feet in one and one- fourth miles or at the rate of sixteen feet per mile. The follow- 6 CHARLES S. PROSSER ing section of the Nemaha county quarry gives a clear idea of the character of the Cottonwood limestone and associated rock^ as they appear in Nebraska. SECTION OF THE NEMAHA COUNTY QUARRY. Ft. In. Ft. In. 5. Soil 3 = 12 10 4. Somewhat shaly light gray limestone used for riprap. Fossils common, especially Athyris {^Seminuld) subtilita (Hall) Newb. - 2 10 = 9 10 3. Light gray shaly limestone changing to shales on weath- ered surface, used for railroad ballast. From i foot i inch to I foot 9 inches in thickness. Fossils not so common as in upper limestone - - - - -19= 7 2. Cottonwood limestone, of light gray color, containing immense numbers of Fiisulina cylindrica Fischer. In two layers, the upper 2 feet and the lower i foot thick, 3 =^5 3 I. Light gray to slightly yellowish limestone, which con- tains some almost white streaks and in places is bluish in color. Very few specimens of Fusiilina. Bottom of the quarry - - - - - - - -23=23 As stated above, No. 2 of the section contains immense numbers of Fusiilina cylindrica Fischer with an occasional broken spine of Archceocidaris sp.; while in No. i there are very few specimens of Fnsidi?ia. This character agrees with the Cotton- wood limestone at its typical localities in Kansas where the Fusulifias are only found abundantly in the upper part of the limestone. On this account it seems advisable to the writer to regard both Nos. i and 2 as representing the Cottonwood limestone of Kansas which will give it a thickness of five feet three inches in the Nemaha county quarry. The shaly lime- stones and shales above the Cottonwood limestone — Nos. 3 and and 4 of the section — contain a moderate number of fossils although none of the species are abundant. The following were collected : -\ Athyris {^Seminiild) subtilita (Hall) Newb.=^. argentea (Shep.) Keyes (c).^ Productus semireticulatus (Mart.) de Kon. (c). ^ The relative abundance of the species is indicated in the following manner : CARB ONIFER OUS A ND PERM I A N FORMA TIONS 7 Derbya crassa (M. & H.) H. & C. (rr). Pinna peracuta Shum. (?) only a fragment (rr). Allorisma sp. fragment of an impression (rr). Archceocidaris sp. fragment of spines (r). Above the Cottonwood limestone in northern and central Kansas are about fourteen feet of yellowish, calcareous shales, the lower seven feet of which contain abundant fossils. In Nebraska no similar shales have been found above the Cotton- wood shales. Perhaps the shaly limestones and shales — Nos. 3 and 4 of the above section — represent the Cottonwood shales, although the fossils are not nearly as abundant. If this sup- position be correct, then they belong in the Cottonwood forma- tion Again it is possible that the Cottonwood shales are repre- sented by the thin shale or shaly limestone, No. 3, which is between one and two feet in thickness; while No. 4 represents the shaly limestones at the base of the Neosho formation m Kansas.' The Gilbert quarry region is six miles west and two and one- fourth miles north of Auburn or about two and one-half miles northwest of the Nemaha county quarry. Near the head of a small stream and by the highway are several quarries. One on the east side of the highway, just south of the Gilbert quarry, gives the following section : Feet Feet c Coarse grayish shales to rather shaly gray limestone at the ^' ^^gg . - - - - 3 = io|< 4. Massive gray limestone, fragments of fossils numerous (2 feet in Gilbert quarry) - - " ' ' " - lyi - 1 A 3. Rather coarse, grayish shales (shaly limestone m Gilbert quarry) - - - 2. Cottonwood limestone, massive Fusulina limestone with small amount of flint and an occasional critwid segment, and ^^ ^ fragment of ^//^J^" - ' ' " ' " - 3/2 — 5 I. Grayish to slightly buff limestone without /^«^/^/«/a^-. Bot- tom of quarry a = abundant ; aa = very abundant ; c = common ; r = rare ; rr = very rare, when but one or two specimens are found. ' See Journal of Geology, Vol. Ill, 1895, P- 766. CHARLES S. PROSSER The Cottonwood limestone and the succeeding Nos. 3, 4 and 5 of this quarry are fairly well shown in Fig. i. The Fiisulina limestone is filled with specimens of Fiisidina cylindrica Fischer all the way through the massive stratum ; ' ^^^^9 M 1 K ^j ■**■- 1 • * • J '^- ,^< 1 ^^^^^^ ^31 a ^3 W^^'-^rr „ V'W^****?' - - ' ^^^^m^Bt K' ' ■ m Wjk B ^M I H ■ 1 ''^B^B ? ■ H am ^ I. View of Cottonwood limestone just soutli of tlie Gilbert quarr)', northwest of Auburn. The Cottonwood limestone is the heavy layer at the bottom. Then in ascending order are shown Nos. 2, 3 and 4 of the section. but as in the Nemaha county quarry both Nos. i and 2 are regarded as representing the Cottonwood limestone of Kansas. The base of this quarry is forty feet higher than that of the Nemaha county quarry or approximately 405 feet above the Missouri River. Since it is two and one-half miles northwest of the Nemaha county quarry it gives between the two a dip of sixteen feet per mile to the southeast. In a small run a quarter of a mile north of the Gilbert quarry and some seventy feet lower, the outcrop shows'a ledge of light gray to buff rather hard limestone with shaly layers above and below. Fossils are few. Aviculopecten occidcntalis (Shum.) M. & W. and Cythere iicbrasccnsis Geinitz (?) occur in the shaly CA RB ONIFER O US A ND PERM I A N FORMA TIOXS 9 layers and in some of them are large numbers of this minute Crustacean or a closely related species. The Carlisle quarry is located on the northern edge of the rather steep bluff some distance south of the Nemaha River. It is nearly one mile north of the Gilbert quarry, and six miles west and 3 + miles north of Auburn. This is at the northern edge of the upland south of the Nemaha River and the farthest north that the Cottonwood limestone was found. In this quarry the base of the Cottonwood stone is about twenty feet higher than in the Gilbert quarry. The section of the G. W. Carlisle quarry is as follows : Feet Feet 4. Shaly limestone - - - - - - - - i yi = 7 % 3. Massive limestone - - - - - - - 2^5^ 2. Shale - - - - - - - - - - iX = 3^ I. Coito}i'WOod \\me.s,t.oi\t. Bottom of quarry - - - 2>^= 2j^ Below the Cottonwood stone on the slope of the hill are thin ledges of smooth limestone alternating with shales. There is one stratum of limestone which on a weathered surface is rough and cellular something like the "dry bone" limestone in Kansas. There are also some reddish shales and the rocks, which for sixty feet below the Cottonwood limestone are partly exposed, resemble to some extent the upper rocks of the Wabaunsee forma- tion in Kansas. These rocks are somewhat fossiliferous and in a bluish-gray shaly limestone from twenty to thirty feet below the Cottonwood limestone the following species were collected : Aviciilopecten occide7italis (Shum.) M. & W. (r). Pletiropho7'ns subcostatiis M. & W. (?) Two rather large specimens which resemble the figures of this species (juite closely (rr). Allorisma {Sedgwickia) cf. topekaensis (Shum.) (Meek) (rr). Edinotidia sp. (rr). Bellerophon sp. (rr). In the region west of Auburn the highest rock found in place is the Cottonwood limestone with the shaly limestones imme- diately on top. The country to the west of the Nemaha county (juarrv rises from lOO to 125 feet higher, but a somewhat hasty search failed to reveal anv ledges of rock in place, all being 10 CHARLES S. PROSSER quite deeply covered by drift and loess. On this high country seven and one-fourth miles west and one mile north of Auburn is a Burlington and Missouri River R. R. cut of ten feet which only shows the recent deposits, so that it would appear to be a difficult undertaking to find the bed rock on the uplands. This difficulty was experienced by Dr. Hayden who stated that : " From Tecumseh [the county seat of Johnson county, twenty- two miles west of Auburn] to the source of the Nemaha, about forty-five miles, I did not discover a single exposure of rock, and I could not ascertain that any had ever been observed by the settlers." ^ From the facts stated above it seems reasonably certain that the massive limestone west of Auburn may be correlated with the Cottonwood limestone of Kansas. Lack of time prevented the actual tracing of this limestone south to the exposures of Cottonwood limestone in northern Kansas, yet its biologic and lithologic characters are so similar to those of the Kansas stone that it appears quite certain they belong to the same for- mation. In Kansas, the Cottonwood limestone is reported by Professor Knerr in the northeastern part of Marshall county where he states that it " disappears under the drift about five miles north of Beattie in Marshall county."- This locality is about fifty miles southwest of the Nemaha county quarry of Cotton- wood limestone in Nebraska. JOHNSON AND GAGE COUNTIES. Wabaimsee formation. — Johnson county lies directly west of Nemaha county to the west of which is Gage county which extends south to the state line and is crossed by the Big Blue River. At Tecumseh in Johnson county about fifteen miles west of the Nemaha county quarry and from lOO to 200 feet lower, as far as I can judge from the data at hand. Dr. Hayden reported a 'Fin. Rep. U. S. Geol. Sur. Nebraska, p. 34. "^ Univ. Geol. Sur. Kansas, Vol. I, 1896, p. 142. Also see "A Geologic Map of Kansas (preliminary)," PL XXXI, in the above work on which the line of outcrop of the Cottonwood limestone is represented. CARBOXIFEROUS AND PERMIAN FORMATIONS I I stratum of coal from ten to fifteen inches in thickness/ In this connection it is interesting to state that Professor Knerr noted the occurrence of "a shale bearing a four-inch stratum of coal " which is given as from 50 to 100 feet below the Cottonwood limestone.- Again, in the vicinity of Tecumseh Dr. Hay den stated that "in a bed of limestone, holding a high position in the hills, the following fossils were found : Spirifev cameratus, Athyris S7ibtilita, Sy?itnlasma heniiplicata \_Enteletes Jiemiplicatus\, Prodiictus semiretiailatus." ^ In Kansas, where the waiter has carefully studied the stratigraphy and palaeontology of the Upper Car- boniferous and Permian rocks along the Kansas and Cottonwood valleys where they, as far as known, are more clearly exposed than at any other locality in the two states, Spirifer cameratus Morton has not been found above the top of the Wabaunsee formation, and from this fact the writer is inclined to refer the limestone mentioned by Dr. Hayden to the Wabaunsee forma- tion. Permian. — In Gage county, which lies next west of Johnson county, Dr. Hayden reported grayish and yellowish argillaceous limestones which he stated — "are undoubtedly of Permian or Permo-Carboniferous age, though they contain fossils common to both Permian and Carboniferous rocks."'* He was not confi- dent of the presence of Permian rocks in Nebraska for he said : "It is not certain that the true Permian beds, as recognized in Kansas, extend northward into Nebraska, though thin beds may occur in some of the southern counties." ^ And he further said that the Permian rocks " pass beneath the water level at Beatrice," the county seat of Gage county, and westward are the yellowish and dark brown Cretaceous sandstones, now known as the Dakota sandstone. In 1886 Professor Hicks published a short paper about the rocks along the Blue River in Gage county which he designated provisionally as Permian and said that they were " distinguished ■ Fin. Rep. U. S. Geol. Sur. Nebraska, p. 34. ^ Ibid., p. 28. ^ Univ. Geol. Sur. Kansas, p. 142. ^ Ibid., p. 28, footnote. 3 Kin. Kep. L'. S. Geol. Sur. Nebraska, •jj. 34. 12 CHARLES S. PROSSER from the underlying Coal Measures by the absence of coal and black shales, and by the prevailing magnesian character of its limestones, by the presence of certain characteristic indurated marls and oolitic limestone, as well as by the new and distinct types of animal life." ' These rocks are undoubtedly of Permian age and it is probable that the Neosho formation and possibly a part of the Chase occurs in Gage county. This supposition is supported by the statement of Professor Knerr that in Marshall county, Kansas, which adjoins Gage county on the south, there are about 250 feet of Permian above the Cottonwood lime- stone.^ OTOE COUNTY. HISTORIC REVIEW OF THE GEOLOGY OF THE COUNTY. To the north of Nemaha and Johnson counties is Otoe county which extends from the Missouri River on the east to Lancas- ter county on the west. This is an important county in the history of the geology of southeastern Nebraska because the cliffs near Nebraska City have been fully described by several geologists. Owen in 1852 gave some account of the rocks along this part of the Missouri River, referring them to the Carboniferous. He briefly described the rocks in the bluff near Fort Kearney (the old name for Nebraska City) and reported Productus costatiis, P. Flemingii = P. corn, and Fusulina cylindrica, which he says "previous to this discovery was only known in Ohio." 3 Swallow in 1855 mentions strata at Fort Kearney and the mouth of the Little Nemaha which he referred to the "Upper Coal Series " of the Upper Carboniferous. ^ In 1855 Marcou published a geological map of the United States on which the rocks along the Missouri River from the mouth of the Big Sioux to that of the Kansas River are colored 'Am. Naturalist, Vol. XX, p. 882. ^Univ. Geol. Sur. Kansas, Vol. I, p. 144. 3 Rep. Geol. Sur. Wisconsin, Iowa and Minnesota, pp. 133, 134. See sections 34 and 35 M. ■* 1st. and 2d. Ann. Rep. Geol. Sur. Missouri, p. 79. CARBONIFEROUS AND PERMIAN FORMATIONS I 3 as belonging to the "Terrain du nouveau Gres rouge" (Triassic system.)' Marcou republished this map as a frontispiece of his Geology of North America, \n which occurs the statement that "beds of Nezv Red Sandstone .... cover and form the majority of the immense prairies bordering the rivers Missouri, Platte, Arkansas, and Red River of Louisiana." ^ In 1857 Dr. Hayden published a geological map of Nebraska on which the rocks along the Missouri River valley from about fifty miles north of the mouth of the Platte River, south to the Kansas River in Kansas, are colored as belonging to the Carbon- iferous age. 3 The following year Dr. Hayden published a second edition of the above map on which the Carboniferous area remains about the same. To the west of the Carboniferous, the Permian system, which was not indicated on the earlier map, is mapped. This system is represented as beginning at a point a number of miles northwest of Nebraska City and then extend- ing southward increasing in breadth to the southern part of Kansas. The base of the system is represented as crossing the Republican and Smoky Hill rivers several miles west of Ft. Riley, while its upper boundary crosses the Grand Saline and Smoky Hill rivers a number of miles west of the present city of Salina. In Kansas, small areas of Permian are represented on the high divides to the east of the main Permian area. Imme- diately west of the Permian or the Carboniferous where the Permian is absent, rocks are represented that are referred to the Lower Cretaceous.'* In 1863 Marcou and Capellini studied the Missouri River section along the eastern border of Nebraska and the 'following January Marcou published quite a full description of the rocks 'Carte Geologique des Etats-Unis et des Provinces Anglaises de 1' Amerique du Nord. In Bull. Soc. Geol. France, Vol. XII. ^Op. cit., Zurich, 1858, p. 11. 3Proc. Acad. Nat. Sci., Philadelphia, Vol. IX, opposite p. 109 and on p. no the description of the area of the system. ^ Ibid., Vol. X, p. 139. For a statement of the distribution of the Permian, see p. 144. 14 CHARLES S. PROSSER in the vicinity of Nebraska City and referred them to the Upper Dyas or Permian.^ The following year Meek criticised the correlations of this paper, stating that "all the rocks seen by Mr. Marcou on the Missouri, from St. Joseph to the Cretaceous above Bellevere, belong to one unbroken series of Upper Coal Measures, as was first shown by Professor Swallow ; with possibly the exception of some of the highest outcrops near Nebraska City, where there is a downward undulation, that may have left portions of the Per- mian on the high parts of the country." ^ In ] 866 Geinitz described the fossils collected by Marcou in Nebraska together with some from the Permian of Kansas, 3 and also concluded that the rocks in the vicinity of Nebraska City belonged to the Dyas.'^ Meek in 1867, reviewing at length Professor Geinitz's work, failed to agree with him in many instances concerning the identity of Nebraska species with those of Europe ;5 he also reaffirmed his previous statement that the Nebraska City rocks with possibly the exception of the highest beds, belonged in the Upper Coal Measures.^ ' Bull. Soc. Geol. Fiance, 2.^ ser. Jan. 1864, Vol. XXI, pp. 134-137. Marcou's con- clusion, based on the fossils he collected being expressed as follows : " Les fossiles que j'ai trouves dans cette section [Nebraska City] m'ont rappele tout a fait le faune dj'asique du Zechstein de la Saxe, et je regarde ces couches de Nebraska-City comme appartenant et repr^sentant en Amerique la partie superieure du dyas d'Europe," p. 137. ^ Am. Jour. Sci., 2d ser., Vol. XXXIX., March 1865, p. 165. 3M. d. K. Leop. -Carol. Akad. d. Naturl. — Carbonformationund Dyas in Nebraska. Dresden, pp. vii-l-91. 5 plates. '' Ibid., p. 89, where he says : " Die bei Nebraska-City vorkommenden Versteine- rungen gehoren einer Zone an, welche den untersten bis mittelren Schichten der deutschen Zechsteinformation (oberen Dyas) entspricht." s Am. Jour. Sci., 2d ser., Vol. XLIV, Sept. 1867, pp. 170-188 ; Nov., pp. 327-340. ^Meek's statement was as follows : Those [rocks] by both of them [Marcou and Geinitz] referred to the Upper Dyas at Wyoming and Bennett's Mill and Nebraska City, y^i'Cci possibly the exception of divisions C and D [the higher beds] at the latter place, belong to the horizon of the Upper Coal Measures. The only point in regard to which there can be any reasonable doubt is, whether the divisions C and D at Nebraska City belong more properly to the horizon of the rocks Dr. Hayden and I termed Permo-Carboniferous in Kansas, or to the Coal Measures proper." {Ibid., pp. 336-337.) CARBOA'IFEROUS AND PERMIAN FORMA TIONS I 5 In 1868 Professor Marcou reaffirmed the Dyassic age of the Nebraska City rocks, stating that: "In Nebraska the Dyassic rocks form the right bank of the Missouri River from Aspinwall to Plattesmouth and Aureopolis, that is to say, all the bluffs of the counties of Nemaha, Otoe, and Cass The best sec- tion of the Dyas of Nebraska and the most easy to be studied, is that formed by the bluff at the Nebraska City landing, at Otoe City, at Peru, and at Brownville, where the strata are higher in the Dyassic series than at Nebraska City ; whilst at Rock Bluff, Plattesmouth, and Aureopolis, on the contrary, we find the lower layers forming the base of the Nebraska Dyas." ^ Following the above is a detailed section of the rocks at the Nebraska City landing accompanied by lists of fossils which were identified by Professor H. B. Geinitz. In 1892 Professor Marcou still regarded the Nebraska City rocks as of Lower Dyassic age.- In the summer of 1867 Meek and Hayden spent about two months in carefully studying the rocks along the Missouri River from Omaha to Kansas. In 1872 an excellent work based upon this study was published in which Hayden discussed the general geology of southeastern Nebraska, and Meek gave a detailed account of the stratigraphy of the Missouri River region accom- panied by a very careful description of the fossils. 3 This work has become a classic in paljEontology for the Upper Carbonifer- ous of the Mississippi Valley. The conclusion in reference to the age of these Missouri River beds agrees with the former opinion of Meek and is clearly expressed in the following sen- tences : "From all of the facts, therefore, now determined, it must, I think, be clearly evident that all of these strata under consideration along the Missouri, that have been by some referred in part to the Mountain limestone, in part to the Per- mian or Dyas, and in part to the Coal Measures, really belong entirely to the true Coal Measures ; unless the division C, at 'Trans. Acad. Sci., St. Louis, Vol. II, p. 562. 'Am. Geol., Vol. X, pp. 369, 373. 3Final Rep. U. S. Geo. Sur. Nebraska and portions of the adjacent Territories, pp 245, 11 plates. 1 6 CHARLES S. PROSSER Nebraska City, and some apparently higher beds below there on the Missouri, may possibly belong to the horizon of an inter- mediate series between the Permian and Carboniferous, for which, in Kansas, Dr. Hayden and the writer proposed the name Permo- Carboniferous It is true that in first announcing the existence of Permian rocks in Kansas, we also, upon the evidence of a few fossils from near Otoe and Nebraska cities, resembling Permian forms, referred these beds to the Permian ; but on after- wards finding that these fossils are there directly associated with a great preponderance of unquestionable Carboniferous species ; and that there is also in Kansas a considerable thick- ness of rocks between the Permian and Upper Coal Measures containing, along with comparatively few Permian types, numer- ous unmistakable Carboniferous forms, we abandoned the idea of including these Otoe and Nebraska City beds in the Permian. And all subsequent investigations have but served to convince us of the accuracy of the latter conclusion."^ It is to be noted in reference to this correlation of the Upper Palaeozoic rocks of Nebraska with the Upper Coal Measures, that Meek did not intend to include the rocks in Kansas which he and Hayden had called Permian,^ a fact which has been misap- prehended by certain writers on the geology of this region. Since the report of Meek and Hayden, no contribution of importance has appeared relating to the geology of the Upper Palaeozoic of Nebraska, consequently it is especially interesting to compare their conclusions with our present knowledge which has been enriched by the labors of the last quarter of a century. Charles S. Prosser. 'Fin. Rep. U. S. Geol. Sur. Nebraska, etc., pp. 130, 131. = Trans. Albany Inst., Vol. IV, 1858, p. 76; Proc. Acad. Sci. Phil., Vol. XI, 1859, pp. 20, 21 ; and Am. Jour. Sci., 2d ser., Vol. XLIV, 1867, p. 37. (To be continued.) EVIDENCES OF RECENT ELEVATION OF THE SOUTHERN COAST OF BAFFIN LAND.^ CONTENTS. Introduction. Big Island. Location, description and topography. Kind of rock. Proofs of elevation in raised beaches. Proofs of elevation in differential weathering and unlike surface conditions. Palseontological evidence of elevation. Southern coast of Baffin Land about twenty miles north of Ashe Inlet. Raised beaches. Fossils. Weathering. Icy Cove — Southern part of Meta Incognita. Raised beaches. Niantilik Harbor — Cumberland Sound. Raised beaches. Evidence of present rising of the land on Big Island around Ashe Inlet region, and at Niantilik Harbor. Degree of the rapidity of the uplift. Conclusions. Partial bibliography of recent elevations in Northern America. Introdiictioii. — This paper is the outgrowth of the opportunity afforded for studying the lands at several places in and north of Hudson Strait, during the past summer, while a member of the Cornell Greenland party, with the sixth Peary expedition. Four stops were made in all along the coast of Bafifin Land, three going up, as follows : Big Island ; the mainland, just north of the island ; and Icy Cove on Meta Incognita. The fourth land- ing was at Niantilik Harbor in Cumberland Sound on the return homeward. 'The writer is greallv indebted to various members of the expedition for valuable suggestions and help, especially Mr. Bonsteel, who stopped with him on the island ; but, whatever value this paper may contain is largely due to Professor R. S. Tarr, who kindly directed the work throughout. To all the writer wishes to express his thanks and add his most grateful acknowledgments. 17 1 8 THOMAS L. WATSON BIG ISLAND. Locatio?i, description and topography . — The location of the island is immediately off the southern coast of Baffin Land, in Hudson Strait, and separated from the mainland by a narrow channel of water, ten to twenty miles wide, known as White Strait in north latitude 62° 30' to 63° and west longitude 70° to 71° 10'. It is some twenty-five to thirty miles in the direction of its longest axis, which is northwest and southeast, and has an aver- age width of from five to ten miles. The coast is a steep and irregular one, being much cut up by fiords and embayments. The highest land reached on the island was 470^ feet above sea level. Its surface has been deeply incised by interlocking fiordic^ valleys, which are quite broad at their tops, with the ridges or divides between, of a typical moutonneed form. These, of course, are on their tops narrow in proportion as the valleys are wide. The rise and fall of the tides is about thirty feet. The topography shows marked signs of glaciation, though, in places, it has been greatly modified by weathering, which has been chiefly of the mechanical kind and on a large and rapid scale. Notwithstanding the great amount of mechanical weather- ing, due almost entirely to frost action, chemical disintegration is distinctly noticeable. Many sections which have been but recently uncovered by ice are rough and angular, with nearly every trace of glaciated form obliterated. Kind of rock.— Th^ rocks consist of regularly banded horn- blende-biotite gneiss, complexly folded and gray in color, the intensity of which varies according to the amounts of the dark minerals present. The gneisses are intersected by numerous pegmatite veins composed of the same minerals. Proofs of elevation in raised beaches. — In nearly every valley studied, one of its most prominent and striking characteristics was the occurrence of shore lines in the form of distinct beaches. Sometimes a full and complete series of half Mozen or more of these would be found in a single valley at different elevations, 'All altitudes were measured with an aneroid barometer. = The term ''fiordic'" is here used in the sense of "fiord-like." EVIDENCES OF RECENT ELEVATION 19 while in others only one or two would occur. These were, in most cases, well developed, level topped, and composed of well- rounded, water-worn material, four-fifths of which was distinctly f\\ ^ Sketch Map of the Hudson Bay Region. Baffin Land and Labrador. ftG\JRt£ RCPRESENT T~t LOC >T,ON OF nA.SEO DC^CHU X. LoCAl-'T.tS IN ' """'" " Thomas L Watsdn derived from the local gneisses. However, thev contained as well quite a noticeable quantit}- of foreign material, c. g., shale, lime- stone, quartzite, etc. This material was variable in size in the lower beaches, and also in several of the higher, though noticeably uni- form and coarse for the higher ones. With reference to the size of the materials composing these beaches, several different types 20 THOMAS L. WATSON were represented, from the true sand and gravel, on the one hand, to the typical bowlder beach on the other, with all grada- tions between these two extremes. Except for the materials being covered with lichens, these are as fresh and perfect in every respect as though they had been formed but yesterday. In the majority of cases the beaches extended entirely across the val- leys from side to side, although it was not uncommon to find them thinning out at one end, and only reaching from one- half to two-thirds the entire distance. They were variable in dimensions, in width from lo to 50 yards, and in length from 60 to 1 10 yards. Their length depended upon the width of the valleys in which they were formed. In elevation they ranged from 270 feet above, down to sea level ; and, so far as studied, could be correlated throughout. The best developed and most uniform and regular series of beaches found were in a valley ^ which began at the north end of Ashe Inlet, with a direction' S. 13°. 5 W. The divide in this valley is located nearer the southwest end, and about 1000 yards from its northern terminus in Ashe Inlet. Unlike most of the other divides on the island, which are composed of loose material, either glacial or beach deposits, this one is formed of the gneissic rock, in situ, and is exposed for the entire width of the valley with an elevation of 185 feet above sea level. The first beach is built immediately against this rocky divide at an elevation of 175 feet above sea, with an average width of some forty feet. The second beach is 165 yards beyond the first one, northward, at 125 feet above sea, and is the best developed one of the series in this valley, with a width of some one hundred feet. Between these two slight fragmentary ones are scat- ' It was on the south side of this valley, only a short distance from Ashe Inlet, that the Hudson Bay Company established their scientific station, or Observatory No. 3, in 1884 ; and it was in their house that we camped during our stay on the island. - Through the kindness of Mr. G. R. Putnam, of the U. S. Coast and Geodetic Survey, the writer has been enabled to give all bearings in terms of true North and South readings. Mr. Putnam states that the compass needle is rather unstable in these regions ; also, there may be daily changes of several degrees, and the effect of local attraction is likely to be great. H ° if- o > " Z 22 THOMAS L. WATSON tered, within a few feet of each other. The third and fourth beaches are found at the same elevation, lOO feet, and about equally distant from the opposite sides of a rock-basin lake which has been formed in the intersection of this, with a second valley, whose direction is N. 53°. 5 E. The fifth beach is found at an elevation of 75 feet, and at a distance of about 165 yards from the fourth one, with an average width of some 60 feet. This is the last well -developed beach in this valley, though there are two fragmentary ones found at the respective eleva- tions of 55 and 50 feet above sea. These are located on the north side of and at a short distance from a second small lake, whose surface is 60 feet above sea and 75 yards from the fifth beach. In the southwest half of the valley, which has been mentioned above as crossing the one just described, are found two well- developed beaches at the respective elevations of 50 and 75 feet, which are correlated with the two corresponding beaches in the above series. In the next valley immediately beyond, eastward, and approximately parallel to the one trending S. 13°. 5 W., is found the largest, and by far the best-developed, beach seen on the island. It is distinctly a sand and gravel beach, 40 feet high, with its crest 175 feet above sea, and about 120 feet wide by 330 long. It serves as the divide in its valley, and is the correlative of the 175-foot beach in the first series. Proofs of elevation in differential weathering and unlike surface con- ditions. — The fact of recent elevation of this island does not rest alone upon the evidence of raised beaches, though this, to be sure, is entirely satisfactory in itself. It is confirmed by other geolog- ical evidence of a very strong nature. Apparently there exist on this island two sharp and well-defined zones, whose surface conditions, in nearly every respect, are very markedly different from each other. The first zone, which begins at present sea- level, and has its upper limit about 300 feet above sea, includes all the land below that level. This zone includes the hilltops for a distance of from two and one-half to three miles back from the sea, and the bottoms of all the major valleys observed on EVIDENCES OF RECENT ELEVA TION 23 the island. Excepting the valley bottoms, which are filled with quantities of unusually large and well-rounded bowlders, this belt has been stripped of its loose material in the form of glacial drift ; consequently the bare and naked rock is exposed on the hilltops. The attack of the agencies of weathering upon the surface of this area has been in progress, more or less, ever since the rock was exposed, but the effect is far less than that over the areas above an elevation of 300 feet. This would naturally follow, since it will be shown that up to this elevation the waters have but recently subsided or fallen. By far the larger quantity of loose materials which are scattered here and there over this zonal surface is rounded and waterworn. Mechanical weath- ering has in places shown its effect, and angular masses are seen scattered about somewhat sparingly, in the form of small talus deposits. This zone will embrace at least three-fourths of the total land area of that part of the island visited. The second zone has its lowest level and beginning at the 300-foot elevation and includes all the surface above, including an area which is not continuous, being merely the tops and sides of the hills for two-thirds of their distance downward. In this zone the bedrock is seldom seen, but is covered to an unknown depth with very large and loose angular blocks ; in a few places, however, the bedrock outcrops at the surface in the shape of small knolls of somewhat decayed rocks. These angular rocks are clearly derived from the local gneisses, pre- sumably largely, if not entirely, by frost action. In some cases they are weathered to a thoroughly crumbled condition. The scarcity of glacial bowlders in this angular mass was very strik- ing, yet a few, which proved in each instance to be of foreign source were seen. Along the sides, at an elevation of from 50-75 feet above the valley bottom, which was 220 feet above the sea, were noted patches of pebbles and bowlders, mostly the latter. These were deposited in small channel ways which had been carved by temporary streams flowing down the valley sides. Fully a half dozen of these were seen at different places on the same hill- 24 THOMAS L. WATSON side and at the same elevation. A short distance above this the loose angular material commenced to cover the surface. These conditions would seem to indicate that the waters of the sea had their level near this elevation, when these deposits were formed in what then were the mouths of the present channel ways. Such loose and angular material as may have extended below that line was subjected to sea action. It was ground up and distributed in the usual way over the sea bottom. It might be of interest to mention that in this valley was found a series of beaches, four in number, which were distinct and perfect pebble beaches, deposited on a shelf rising some 50 feet above the valley floor. For the sake of a brief comparison let us note the salient features of the two zones. The first zone is characterized by a seaward strip of land, some two and one-half to three miles wide, reaching an elevation of some 300 feet, with deeply incised val- leys (a feature common to both areas) and occupied by raised beaches. This zone skirts the higher interior land area, which has been termed the second zone. Furthermore, this area has its rock bare and more or less polished by glacial action with but little material strewn over its surface, which, for the most part, is waterworn, with occasional talus slopes of angular rock. The second zone, which includes all the land above 300 feet, is cov- ered deeply with large angular blocks, has its bedrock exposed in a few places only, and all glacial form mostly destroyed. The contact between these two areas is marked in places by pebble and bowlder patches along the hillsides in some of the valleys. PalcBontological evidence of elevation. — In one of the valleys, 270 feet above sea level, was found a large deposit of well-pre- served shells, representing two genera living at present, Macojna calcarea, Chemnitz,' and Mya tnmcata, Linn. (?). These were not found in direct association with the beaches, but were only a short distance from one series, and were taken from a small area of black mud, not covered by vegetation. Dawson refers to several cases in southern and eastern Can- 'The writer is indebted to Mr. E. M. Kindle for the identification of species. E I 'JDEXCES OF RECENT ELE VA TION 2 5 ada where vertebrate remains, especially the whale, are associ- ated with raised beaches. In speaking of the lower St. Law- rence in the neighborhood of Little Meta,' he s^P;"^;"^^^^ large whales occasionally occur on this terrace." After describ- ing a beach on the island of Anticosti,^ he says, "The bones of a whale were found on this beach." He further states that the same condition is observed along the shore of the St. Lawrences and at Smith's Falls,^ Ontario. The beach at the latter place has an elevation of 420 feet above the level of the sea. Packard 5 refers to the same association on the lower Savage Islands In each case invertebrate shells representing several genera 'and species were found occurring with the vertebrate remains, but in greater abundance, as would be expected. Sorcthern coast of Baffin Land, about twenty miles north of Ashe /.^/.^^- Three stops were made on the mainland at different places, and at each one proofs of recent elevation were seen m the form of raised beaches. At the first landing, which was across White Strait to the north, and opposite the middle of Big Island, about twenty miles from Ashe Inlet, the beaches were associated with other forms of evidence, above mentioned, as being present on the island; viz., fossils and difference m degree of weathering above certain heights. The rock of this pan of Baf^n Land is a fine-grained, garnetiferous gneiss, and thus differs from the rock of the island. The land is low near the coast, but rises into a series of hills which at a distance of a mile or more back from the sea reach an elevation of 700 feet, continuing to rise inland. Raised beaches.-T\.^^^ did not attain so perfect a degree of development as those on the island. They were formed in very narrow valleys, crescentic in shape and concave seaward, dammmg back small ponds or lakelets. In one place, where this condition ^The Canadian Ice Age, 1894, P- 65- ' ^t!^f^' '^\ 4 Ibid., p. 203. ^Jbld., p. 159- TT <. sMemo.rs Bost. Soc. Nat. Hist., 1867, I, Part II, p. 226. eThe writer was landed on Big Island and d.d not v,sU th.s pa.t «* J - ^^ ^l^^' For what follows he is indebted to Professor R. S. Tarr, who has ver> k>ndh fur- nished him with all the facts. 26 THOMAS L. WATSON was especially noticed, it looked in every way as though it were artificial ; the crescentic beach, which held up a lake behind it, had been so regularly constructed that had it been in an inhab- ited region Professor Tarr states that he should have ascribed it to the hand of human beings instead of to nature's handiwork. These crescentic lines were composed of large bowlders, weigh- ing from fifteen to twenty pounds, and in length were from lOO to 125 feet, rather narrow topped, probably six feet across, but several times this width at base. The exact counterpart of these was seen in process of formation in one place at sea level, in which the ice was an important factor in their construction. The ice, moving in strong tidal currents, bore along bowlders and ground them against the coast, forming a bowlder pave- ment of a very perfect kind.^ Professor Tarr states that, due to the narrowness of the valleys, it would be impossible for these to form without the aid and action of the ice, for no waves could exist here which would transport and pile up such an accumula- tion of bowlders, particularly when below the zone of ice action the bottom is clayey. Fossils. — The following genera of living shells were found: Mya, Saxicava, Pecten, Terabratula, Balajius, and several other living species. These were found in a blue mud, some patches of which were fifty to sixty feet across, in some cases covered with moss, and in others not. Fossils were also found at lower levels. Weathering.- — Professor Gill independently suggested recent elevation purely on the basis of the weathering of the rock in place. In chipping and breaking off petrographical specimens he found a striking difference in degree and intensity of weather- ing, which he placed at an elevation of from 300 to 400 feet. The rocks below this elevation were found to be much less ' For action of similar kind described by Packard and"FEiLDEN along the Labrador coast see references in Bibliography. ^I am indebted to Professor A. C. Gill for kindly furnishing me with this fact. The exact elevation of contact marking the difference in degree of weathering was not determined. EVIDEXCES OF RECENT ELEVA TION 2/ affected or changed bv the weathering agencies than those above this height. Icy Cove, southern part of Mcta Incognita. — Our second land- ing place was on the southwest coast of a peninsula lying between Frobisher Bay and Hudson Strait, and about sixty miles east of Big Island. This land is known as Meta Incognita, and the landing was at "Icy Cove," where the only Eskimo settlement in the Straits was found, called "Noogla." The topography here was much the same as that of the other two places — very rugged — and like Big Island the coast was steep and rough, indented with embayments and rocky capes or headlands. The rock is a very coarse-grained granitic gneiss. Raised beaches. — Two beaches were noted, composed of coarse, rounded material. No elevations were taken, but these looked to be about 50 and lOO feet, respectively, above sea level. A bench between the two beaches was seen, which appeared to be a wave-cut terrace, at an elevation of about 75 feet, Niantilik Harbor, Cn>nberland Sound. — This was our last landing on the Baffin Land side. Niantilik Harbor is a fiord on the south side of Cumberland Sound. A stream of consider- able size enters at this point, having its head waters in a series of true rock-basin lakes of rather large size. Raised beacJies. — It is along the west side of this valley that we have a series of unusually large and well-de\'eloped beaches. Unlike those described from the other localities, these are com- posed of fine material, excepting the topmost one, which consists of coarse shingle. The direction of the two principal beaches is approximately parallel to the stream, N. 75° 55' W. The first one is at an elevation of 1 10 feet above low tide, 200 yards long, with an average width of 30 yards, and is composed of sand, gravel and pebbles. The second beach is from 50-100 feet above the first one, fully three-ciuarters of a mile long and with an average width of 75 feet, and is built of very large bowlders. It is not so well preserved as the lower one, as it lies against a \^ery high and steep scarp, with its flat-top[)ed condition seen in only a few 28 THOMAS L. WATSON places, and for rather short distances, the rest being almost entirely masked by the piling up of the products of weathering. An intermediate stage is represented by a beach some 50 yards long and as many wide, composed of very fine material, mostly sand and gravel, and acting as the divide in the very shallow valley in which it is built. The two large beaches grade or run into rock-basin lakes at their eastern ends. The cemetery of Black Lead Island is built on a well-defined beach composed of sand and gravel, and is at an estimated ele- vation of between lOO and 125 feet above sea level. Beaches were noticed at several other places, but time would not admit of their study. A condition, unlike that been at any of the other places, was noticed on all of the lands enclosing this harbor, which, in itself, would have a tendency to indicate or suggest elevation. The condition was that of a form of rocky headland or cape of pecu- liar development, cut out of the solid rock, primarily, by wave- cutting and perhaps, subsequently, by ice erosion to an unknown extent. They were very numerous, extended seaward for quite a long distance, were very narrow — only a few yards at widest ■ — and were of a remarkably level-topped condition, rising five to ten and twenty feet above sea level. At about'the same level notches of wave-cut origin were more or less distinctly notice- able, and while time would not admit of their study, they appar- ently were in correlation with the capes. Partial evidence was found which seemed to indicate recent elevation of some 50-100 feet above the highest beach mentioned. Evidence of presejit rising of the land on Big Island around Ashe Inlet region and at Niantilik Harbor. — At each of these places, in nearly every valley studied, was found a beach built of fine material, sand and gravel, at an elevation of from five to ten feet above high tide. The evidence of present upward movement at Niantilik is made stronger by the peculiar type of rocky headland, extending seaward. BelP has shown evidence of a like kind indicating a similar ' Canad. Geol. Survey, Rept. of Prog. 1882, 1883, 1884, pp. 26, 31, 33 and 35, DD. EVIDENCES OF RECENT ELEVATION 29 uplift of the lands to the west of the region herein described, and with Tyrrell' has proven the raised or elevated condition along the west and southwest shore of Hudson Ba}-. Again, Bell" has produced sufficient evidence, although doubted by Tyrrell,^ that the Hudson Bay region has been elevated in historic times ; the elevation being believed to be in progress at present. No landing was made on the lands along the south side of the straits, but during the summer of 1884 Dr. Robt. Bell of the Canadian Geological Survey was sent out as the geologist by the Canadian government, through the straits and into Hudson Bay, and he has described raised beaches on some of the islands to the west and southwest of Baffin Land. For convenience I quote from Dr. Bell's report : Speaking of Cape Prince of Wales,* he says : " Beaches of shingle, as fresh looking as those on the present seashore, except that the stones are covered with lichens, may be seen at all levels, up to the tops of the highest hills in this vicinity The materials of the raised beaches above referred to con- sist principally of gneiss with milk quartz from the veins of the neighborhood, together with a few fragments of yellowish gray dolomite, with obscure fossils, a hard and nearly black variety of siliceous clay slate, with an occasional bowlder of dark, hard crystalline diorite." Concerning Digges Island.s he says : " Between this and the western extremity of the island the hills have a rounded out- line, and raised beaches, composed mostly of coarse shingle, form a prominent feature on their slopes, all the way from high tide mark to their summits, the highest of which is between 300 and 400 feet." Mansfield Island,'^ he says : " For many miles, the whole of the eastern slope of the island presents a succession of steps or ' Geological Magazine, Decade 4, Vol. I, 1894, p. 398. = Am. Jour. Sci., Vol. I, Fourth Series, 1896, pp. 219-228. ^Ibid,Yo\. II, Fourth Series, 1896, pp. 200-205. For other relerences to this region see Bibliography. ■tCanad. Geol. Sur. Kept. Prog. 1882, 1883, 1884, p. 26. DD. 5 Ibid, p. 31, DD. * //"""', P- 33. t)l>- 30 THOMAS L. WATSON small terraces, mostly too low to be distinctly counted, but there might be a hundred of these between the sea level and the high- est parts of the island visible. These appeared to be partly ancient beaches, and partly the outcropping edges of nearly horizontal strata." Marble Island,^ he says: " Even the bowlders and coarse shingle forming the raised beaches remain quite white, and these beaches appear as conspicuous horizontal lines against the dark vegetable matter." Degree of rapidity of the iiplift. — It is strikingly noticeable from the description of the beaches given above, as also from their study in the field at the various localities in which they occur, that the conditions suggest a difference in the rapidity of movement with which the land was raised above the waters at the successive stages and levels. The movement seems to have varied in intensity or rate for the same locality. In the case of the two highest beaches on Big Island and at Niantilik harbor, the conditions point very strongly indeed to a uniformly slow change in level. The interval between the two beaches at each of these places is marked by intermediate fragmentary lines. Materials are strewn thickly over the area between in an interlocking manner. This condition is strikingly absent from the land areas between the lower beaches. Thus the change in level from the second highest beach downward was sudden and rapid, and is better described as having taken place by jumps, so^to speak ; while above this line the change in level must have been less sudden and violent, and in character slow and gradual. At Icy Cove and the mainland to the north of Ashe Inlet, the con- ditions indicate the same sudden or rapid jwnping movement as in the lower levels at Niantilik and Ashe Inlet. Going still farther westward the lands along the west coast of Hudson Bay have been described as containing raised beaches, thus indicating recent elevation in that region. In speaking of the raised beaches in the Aberdeen Lake region, Mr. Tyrrell^ 'Canad. Geo). Sur. Kept. Prog. 1882, 1883, 1884, p. 35 DD. ^Geol. Mag., 1894, Vol. I, decade 4, p. 398. EVIDENCES OF RECENT ELEVA TION 3 I says that the}' are found at the following elevations above the lake, 290, 220, 180, 150, 105, 90 and 60 feet; also, "Similar raised beaches are found in favorable localities all along the shore of Hudson Bay." In Mr. Tyrrell's' account of his first expedition through the barren lands of northern Canada, he mentions raised beaches in two localities, one at Doobaunt Lake with an elevation of some 400 feet above sea level ; the second at the mouth of Chesterfield Inlet and on the south side. In Tyrrell's" second trip through these regions raised beaches are mentioned near Ferguson Lake at an elevation of from 400 to 500 feet above sea level, and on the southwest side of Churchill River in the region of Deer River with an elevation of some 600 feet. Conclusions. — i. The evidence favoring recent elevation of from certainly 270 to 300 feet above present sea level on the lands along the south and southeast coast of Baffin Land has been shown to be of the most conclusive character, and can be briefly summed up under three general headings. a. In the form of raised beaches. b. Unlike surface conditions intimately associated with a dif- ference in degree of weathering at a well-defined elevation. c. In the form of extinct life. The remains of several genera and species of living shells were found to be in greater or less degree directly associated with the beaches. Furthermore, that the conditions attending this upward movement at least show that the rate of movement was not alike for all the localities studied, but rather indicates that for some the uplift was sudden and rapid, rising by jumps or strides, while for others it was more uniformly slow and gradual. 2. Conditions strongly favor a present movement on Big Island and in Cumberland Sound. This is shown in beaches found in a great number of the fiordic valleys, which are at ])resent out of the reach of high tide by some five to ten feet, but so recently formed that not a sign of vegetation has commenced ■ Geog. Jour. (London), 1894, July-Dec, Vol. IV, pp. 444-447. "Ibid., 1895, July-Dec, Vol. Vl, pp. 445-447. 32 THOMAS L. WATSON to grow on the beach materials. Also, by the peculiar type of rocky headland and wave-cut notches above described, and at about the same elevation as the lowest beaches. 3. It would further appear that the uplift along south Bafifin Land was coextensive with that described by Bell and Tyrrell in the Hudson Bay region. Thomas L. Watson. Fellow in Cornell University. Ithaca, N. Y., December 10, 1896. PARTIAL BIBLIOGRAPHY OF RECENT ELEVATIONS IN NORTHERN AMERICA. Bell, Robert, Observations on the Geology, Zoology, and Botany, of Hudson's Strait and Bay made in 1885. Canad. Geol. Sur. Ann. Rept., 1885, N. S., i, pp. 7, 8, II and 12 DD. Report on an Exploration of Portions of the At-ta-wa-pish-kat and Albany Rivers, Lonely Lake to James Bay. Geol. and Nat. Hist. Sur., Canada Ann. Rept, 1886, 2, N. S., pp.34-38 G. Report on the Country Between Lake Winnipeg and Hudson's Bay. Canad. Geol. Sur. Rept. Prog., 1877-8, 25 CC. Report on the Exploration of the East Coast of Huason's Bay in 1877. Canad. Geol. Sur. Rept. Prog., 1877-8, 32 C. Report on Explorations on the Churchill and Nelson Rivers and around God's and Island Lakes, 1879. Canad. Geol. Sur. Rept. Prog. 1878-9, pp. 20, 21 C. On Glacial Phenomena in Canada. Bull. Am. Geol. Soc, 1890, i, 308 (287- 310). Proofs of the Rising of the Land around Hudson Bay. Am. Jour. Sci., 1896, 4 S., I, 219-228. Dawson, J. W., Canadian Ice Age. Montreal, 1894, pp. 61-70; chap. 5, pp. 151- 206. De Range, C. E., Arctic Geology. Nature, 1875, n. PP- 447-449. 493. 5o8. Feilden, H. W., and De Range, C. E., Geology of the Coasts of the Arctic Lands Visited by the Late British Expedition under Capt. Sir Geo. Nares. Quart. Jour. Geol. Soc, London, 1878, 34, 556-567. De Geer, Baron Gerard, On Pleistocene Changes of Level in Eastern North America. Proc. Boston Soc. Nat. Hist., 1891-2, 25, 454-477 ; particularly PP- 473. 474- Geikie, James, The Great Ice Age. 1894. Chap. 43, p. 781. Kane, E. K., Arctic Explorations. Vol. II, p. 81. Low, A. P., Preliminary Report on an Exploration of Country between Lake Winnipeg and Hudson Bay. Canad. Geol. Sur. Ann. Rept., 1886, 2, N. S., 16 F. EVIDENCES OF RECENT ELEVATION 33 Report on Explorations in James Bay and Country East of Hudson Bay, Drained by the Big, Great Whale and Clear-water rivers. 1887-8. Canad. Geol. Sur. Ann. Rept.. 1887-8, 3, N. S., 26, 27 J ; particularly pp. 26, 29, 30, 31, 34, 35, 58, 59, 62 J. Murray, A., On the Glaciation of Newfoundland. Proc. and Trans. Roy. Soc, Can- ada, 1882-3, Sec. 4, I, 55-76. Rise of the Land, Terraces and Shells, PP- 57-59- Packard, A. S., Observations on the Glacial Phenomena of Labrador and Maine, with a View of the Recent Invertebrate Fauna of Labrador. Memoirs, Boston Soc. Nat. Hist., II, 1867, i. Part II, 210-303; particularly pp. 119, 120, 219, 222, 224, 225, 226, 227, 229, 230. Salisbury, R. D., The Greenland Expedition of 1895. Chicago, 1895, Jour. Geol., Ill, 875-902; particularly pp. 899-902. Tyrrell, J. B., Notes to Accompany a Preliminary Map of the Duck and Riding Mountains in North Northwestern Manitoba. Canad. Geol. Sur. Ann. Rept. 1887-8, N. S., 3, 10, II, 12 E. Notes on the Pleistocene of the Northwest Territories of Canada, Northwest and West of Hudson Bay. Geol. Mag., 1894, i, decade 4, 394-399- An Expedition Through the Barren Lands of Northern Canada. The Geog. Jour. (London), 1894, 4, pp. 444, 447. A Second Expedition Through the Barren Lands of Northern Canada. The Geog. Jour. (London), 1895, 6, pp. 445, 447. Is the Land Around Hudson Bay at Present Rising ? Am. Jour. Sci., 1896, 4 S., II, 200-205. Wright, G. F., and Upham, VV., Greenland Ice Fields and Life in the North Atlantic. D. Appleton & Co., New York, 1896, pp. 31, 43, 44 ; also chap. 12, pp. 310-333. Wright, G. F., The Ice Age in North America. D. Appleton & Co., New York, 1891, 3d ed. ITALIAN PETROLOGICAL SKETCHES. III. THE BRACCIANO, CERVETERI AND TOLFA REGIONS. Bibliography. — The modern papers dealing with these three regions petrographically are extremely few, and since some of them are not confined to the description of only one they will be noticed together here. We begin, as usual, with vom Rath,' who devotes parts of one Italian " Fragment" to Bracciano and to Tolfa. The descrip- tions are largely topographical, though in the Tolfa paper the "trachyte" is described and an analysis given, and considerable space devoted to the alum mines of the district. With excep- tion of the paper just cited and a few stray notices of rocks in Rosenbusch's " Massige Gesteine," practically all the other articles on the regions are by Italians. Of these the following are the only ones which need be named here. Struever^ published in 1885 an account of the ejected blocks and their minerals which are found to the east of Lake Bracciano, but does not touch upon the eruptive rocks proper. Many of the eruptive blocks, and enclosures of Lake Bracciano, are described by Lacroix.^ In the same year Tittoni'' describes the so-called Agro Saba- tino, which includes the trachytic. hills immediately to the west of Lake Bracciano, the region southwest of it, and the masses of eruptive rock near Cerveteri.s He gives a good geological map on a scale of i : 50,000. The first half of the paper is 'Vom Rath, Zeit. d. d. geol. Ges. XVIII, 561-576, 585-607, 1867. = Struever, Atti Ace. dei Lincei, Series 4, I, i, 1885. 3Lacroix, Enclaves des Roches, Macon, 189. v *TiTTONi, Boll. Soc. Geol. Ital. IV, 1885, 337-376. 5 He states that he was the first (in 1879) to discover the eruptive character of these hills. 34 ITALIAN PEIROLOGICAL SKETCHES 35 devoted to the sedimentary terranes, while the latter half is taken up with brief megascopic descriptions of the eruptive rocks and their tuffs, and with their distribution. He considers that all of the eruptives are post-Pliocene. The rocks of the Agro Sabatino collected by Tittoni, as well as others of the same region from other sources, were examined by Bucca' who also later^ adds a note on the enclosures in the eruptive rock of Monte Calvario. His papers are entirely petro- graphical, and will be referred to later on in detail. Busattis in The same year gives a description of a trachyte from Tolfa, to which Lotti adds some remarks on the genetic relationships. De Stefani,4 later, also gives an account of the regions of Tolfa and Cerveteri.5 It will be evident from the above that the literature of the regions in question is of a very scanty description, the important leucitic rocks of Lake Bracciano being practically untouched, and the others not having been studied very fully. The regions in question are shown on the Tolfa and Bracciano sheets (Foglii 142 and 143) of the Italian Geological Map (scale i : 100,000). THE BRACCIANO REGION. Topography.— The center of this region is the large Lake of Bracciano. This is almost circular in shape and with a diameter each way of 9'^'"^ The surface of the lake is 164" above sea level. On the north at Trevignano its symmetry is broken by a small bay which apparently represents a small explosive crater. The depth of the water is not stated, though it seems to be much deeper than Lake Vico. There are no islands in the lake. 'BUCCA, Boll. Com. Geol. Ital., 1886, 212-223. ^BuccA, Loc. cit., 377-379- 3BUSATTI, Proc. verb. Soc. Tosc. 4, Luglio. 1886. 4De Stefani, Boll. Soc. Geol. Ital. X, 4^7-499, iSQI- s In a paper received just before going to press, P. Moderni (Boll. Com. Geol. Ital. 1896, Nos. I and 2), describes the Bracciano center. He regards the mass as the product of eruptions from about 50 centers surrounding the lake, and this not as a true crater-lake but as due to a sinking in of the surface. He does not touch on tue petrography. 36 HENRY S. WASHINGTON Surrounding it is a circle of hills, whose steep inner sides come down close to the water's edge, leaving only a narrow shore mar- gin. These hills are highest on the north, where they reach their maximum elevation in Monte di Rocca Romana (602""), and from this gradually diminish in height around the lake toward the southern shore, where their height at Monti is only 336"" above sea level. From this circular crest the land here, as at Bolsena and Vico, slopes gradually down on all sides at a low angle, and presents much the same characteristic features. To the east of the lake are three maar-like craters described by vom Rath. The largest of these is the dry circular Valle di Baccano, about 3^" in diameter. Between this and Lake Brac- ciano lie Lake Martignano (whose water level is 43™ above that of Lake Bracciano), and the Stracciocappa Marsh. These are all surrounded by ridges of tuff. The walls and sides of the Bracciano Volcano are built up of leucitic lava flows and tuff beds, except on the west. On this side we find two small non-leucitic centers. About 3""" due west of Bracciano is the group of low hills which include Monti Oliveto and San Vito. These are partially covered by leucitic tuffs, so that they are older than some of these eruptions. To the north of them is a small solfatara, whose floor is white through decomposition of the rocks, and where sulphurous vapors are abundantly given off. To the north of this again is Monte Cal- vario,' a domal mass of eruptive rock. According to Tittoni this rests on Pliocene beds. Petrography. — In this region my stay was of very short dura- tion, so that the specimens collected and the observations made were few. A study of the geological map and my own observa- tions are, however, sufficient to show that we have to deal here, as in the other centers, with two prominent groups of rocks, a non-leucitic and a leucitic. The group of phonolites proper seems to be lacking, or perhaps is present in only small amount. ' It may be noted that BucCA speaks of this last eruptive center as Monte Virginio. The hill of this name, which lies just north of Monte Calvario, is composed of tuffs enclosing blocks of leucitite, as shown on the geological map. ITALIAN PETROLOGICAL SKETCHES 37 Toscanitc. — The non-leucitic eruptive rocks of this region resemble the vulsinites and ciminites in containing basic plagio- clase, as well as orthoclase, and are consequently quite rich in lime. They differ, however, in being much more acid, with SiO., from 63-72, and sometimes contain quartz. They, there- fore, occupy a place intermediate between the rhyolites and the dacites. They correspond, in fact, very closely, both mineral- ogically and chemically, with the rocks of Monte Amiata described by J. F. Williams.' They also resemble the quartz- trachytes of Campiglia^ and Roccastrada.3 As these earlier known localities are in Tuscany (Ital. Toscana) this group of acid effu- sive rocks, characterized mineralogically by the presence of basic plagioclase, as well as orthoclase, with occasional quartz, and chemically by high silica and alkalies and (for the acidity) high lime, and low alumina, may be called tosca?iite. It may be mentioned that they also resemble certain rhyolites from Ponza and from the Euganean hills near Padua. They thus occupy the place in Brogger's^ table filled by the group of quartz-trachyte- andesites, and in some cases are so acid as to fall in with his del- lensite (dacite-liparite). They approach this especially in their low alumina. Analyses of typical toscanites will be found in Table I. The rock of Monte Calvario is very much decomposed, so much so that good fresh specimens are dif^cult to find. I finally obtained some which are quite, though not entirely, fresh at a quarry on the southeast side where work was going on. The rock is rather coarse-grained and resembles many of the por- phyries of our western states. The groundmass is light gray, and glassv feldspar phenocrysts abound, which are colored light yellow by the infiltration of ferruginous water. They are chiefly of sanidine with a smaller quantity of acid labradorite. Some 'Williams, Neu. Jahr. B. Bd. V, 381, 1885. ^VoM Rath, Zeit. d. d. geol. Ges. XVIII, 639, 1866. Also Dalmer, Neu. Jahr. 1887, II, 206. 3MATTEUCCI, Boll. Com. Geol. Ital. 1890, 284 ff., and Boll. Soc. Geol. Ital. X, 670 ff., 1891. ^Brogger, Eruptionsfolge bei Predazzo, Kristiania, 1895, 60. 38 HENRY S. WASHINGTON quartz grains are present. Biotite flakes are not rare, though these and the augite phenocrysts have suffered much through weathering, being represented by brown limonitic spots in some of the specimens. Some large enclosures of a darker, fine- grained, vesicular rock were seen, which will be described pres- ently. In thin section the sanidine phenocrysts are seen to predomi- nate over those of plagioclase. Examination of suitable sec- tions of the latter shows it to be a labradorite of the approxi- mate composition AbgAn^. The}^ are clear, but inclusions are not uncommon, nearly always of brownish glass, and often in the shape of the host. A few instances of parallel growth were seen, a plagioclase core being surrounded by a border of alkali feldspar. This, however, forms one crystal individual with the plagioclase, and belongs to the same general period of crystal- lization, being sharpl}^ outlined and distinct from the ground- mass. It is thus of a different character from the alkali feldspar mantles already noticed. A few rounded cr^'stals of quartz are to be seen. The ferromagnesian minerals are represented by brown biotite and fewer pale green diopside crystals, both of which are much decomposed. The groundmass, which is almost holocrystalline, is made up chiefly of alkali feldspar flakes, very few laths being present, with some brown spots representing orig- inal pyroxene microlites. A little quartz may be present, but no plagioclase could be detected, though the rock is in such a con- dition that a satisfactory study of it was impossible. This rock is described by Bucca as a quartz-trachyte. His description agrees closel}^ with my own limited observations. He mentions, however, the presence of hypersthene, whose exis- tence in my specimens may have been concealed by its well- known liability to decomposition. The enclosures in this trachyte are apparently fragments of an early lava stream, brought up from probably no great distance below. They are dark gray, fine-grained and compact, and quite vesicular, though far from being scoriaceous. The line of junc- tion between the enclosed and enclosing rock is sharp, and no ITALIAN PETROLOGICAL SKETCHES 39 great amount of contact metamorphism is apparent. In the gray groundmass are large tabular glass}- sanidines and some rounded grains of quartz, but no ferromagnesian phenocrysts are to be seen. In many of the vesicles are slender brownish black needles of breislakite, and hexagonal scales of tridymite.' These enclosures are much less decomposed than the trachyte surrounding them, though the sanidines are stained slightly yellow. Under the microscope the structure is strikingly like that of a dolerite. Very numerous prisms of colorless diopside, slightly brownish on the edges, and longplagioclase laths, whose extinctions show them to be labradorite of the composition Ab^An,, with fewer orthoclase laths, lie in a holocrystalline mesostasis of orthoclase. Some comb-like skeleton forms of labradorite are also seen. Magnetite is present though not abundant. A few larger phenocrysts of violet augite are seen, and the few sections of the large tabular sanidine phenocrysts met with are clear and show only a few inclusions of glass and apatite. A narrow orthoclase mantle surrounds them. There were found a number of grains of the peculiar brown barkevikite-like hornblende noticed in a leucite-tephrite of Bolsena. The extinction angle of C on r was 17° and the pleochroism was identical. No olivine is present, but apatite needles are quite abundant. The vesicular structure of this rock shows that it is not a segregation proper, but an inclosure of an earlier solidified lava mass which had been erupted on the surface. The breislakite and tridymite were formed prior to the eruption of the toscanite, since they are present in cavities revealed by breaking open good sized masses of the enclosures. The rock corresponds mineral- ogically, and probably more or less closely chemically, with the vulsinites, though the structure is quite different. It probably represents one of the earliest outflows of the volcanic center, carried up by the later toscanite. Bucca {op. cit. 377) gives a very good description of these enclosures which agrees closely ' Bi'CCA also notes hypersthene and angite crystals. 40 HENRY S. WASHINGTON with the above. The hypersthene he speaks of as present is perhaps to be referred to the barkevikite just mentioned. The toscanite of Monte San Vito is structurally quite different from that of Monte Calvario. The predominating dark brown- ish black mass is highly vitreous, with a few irregular cavities which are lined with a light blue gray opal. Through this are scattered many quite large glassy tabular sanidines, in almost every case twinned according to the Carlsbad law. They are stained slightly yellow and many carry the bluish opal in the crevices. Some irregular quartz grains are also visible, but phenocrysts of ferro-magnesian minerals are wanting. The large sanidines are seldom met with in thin sections, but of the smaller phenocrysts those of feldspar are the most abun- dant. Those of plagioclase are in the majority over those of orthoclase, and Michel-Levy's method shows this to be labra- dorite of the composition Ab^Ang. The crystals, both of lab- radorite and of orthoclase, are automorphic and sharp in outline, showing the usual planes. They are clear and quite free from even incipient alteration. Inclusions of pale brown glass are common, often with a bubble and sometimes of the shape of the host. Pale diopside, apatite and magnetite are also included in the feldspars. These feldspars are often clustered together, but no distinct relative order of crystallization could be made out between the two. The diopside is in stout well-formed crystals, usually with a pyramid largely developed. It is almost or quite colorless, and inclusions are rare and almost wholly of brown glass. A number of small thick tables of brown biotite are present which invariably show a narrow border of fine-grained augite-magnetite aggregate. A few large grains of magnetite and some apatite needles complete the list. In my few speci- mens I could detect none of the hypersthene mentioned by Bucca as abundant. The groundmass is highly vitreous, con- sisting of a largely predominating light brown glass base, in which are sprinkled, with little evidence of flow structure, many minute orthoclase and fewer prismatic diopside microlites. Small magnetite grains are rare and are perhaps derived from / TA LI A N PE TR OL O GICA L SKE TCHES 4 1 altered biotite, as may also be the case with part of the diopside microlites. An analysis of this rock by Dr. Rohrig is given in Table I, No. i. It is seen to be as high in silica as a trachyte, low in alumina, with rather andesitic amounts of lime and mag- nesia, and in alkalies standing between the two groups, though the relative amounts of potash and soda are not what we might expect. The rock just described, as well as similar ones from other localities in the region, are called by Tittoni "trachytic retin- ites." Bucca describes them as augite-andesites, with which name his descriptions agree very well, though it seems to me, as it does to de Stefani, that he unreasonably neglects the abundant sanidines and quartz. He speaks of hypersthene as abundant, and gives extinction angles for the plagioclase' from which he concludes that they must be very basic — "from labradorite to anorthite." According to him orthoclase only occurs as the laro-e tabular phenocrysts, which is certainly not the case in my specimens. Leiicitite .—l:\\& leucitic rocks collected by me belong to three groups, leucitite, leucite-tephrite, and leucite-phonolite. Speci- mens of the first were obtained south of Lake Bracciano, from a flow at Crocicchie, from a similar, or the same, flow about one kilometer west of this and from the quarry at L'Uomo Morto southwest of the lake. A similar rock was seen also at Oriolo, northwest of the lake, but I have no specimen of it. They are all compact, dark gray, basaltic looking rocks, with some fresh well-shaped leucite phenocrysts (0.2 to i '="^) and a few smaller ones of augite. Under the microscope the phenocrysts present no remarkable features. The leucites are clear, somewhat cracked, show quite strong double refraction, and contain few inclusions. The augites are well shaped, pale green and not pleochroic, generally darker toward the center, and with an extinction angle of about 42°. Some are seen with a fringe of later dark green augite at the ends, which extinguishes at an angle of 50^ and includes some magnetite grains. 'He says that they vary from 35" to 59°. There seems to be some error here. 42 HENRY S. WASHINGTON The groundmass is almost holocrystalline, and is made up chiefly of small round leucites. In two cases these contain only few peripheral inclusions of augite and apatite needles, a peculiar feature being that around the leucite crystals proper as defined by these rings, is a late growth of leucite which extends irregu- larly and acts as a mesostasis for the other constituents. In rock from the second locality mentioned the inclusions are more numerous, and in spots due to skeleton growth, as has been previously described, though the skeleton forms are not as per- fect as some of those seen elsewhere. Between these leucites is an interstitial mass of green or greenish brown augite needles and grains, with some magnetite grains, with the latter being associated flakes of orange red hematite. Spots of a colorless feebly doubly refracting substance, giving bright grays of the first order in m}- ver}^ thin sections, are referred to alkali feldspar, as treatment with acid revealed no nepheline. No melilite was seen. Glass base is present in a very small amount. An analysis of one of these leucitites is given in Table I, No. 5. The leucitite of Santa Maria di Galera,' which belongs to this volcano, resembles the above very closely under the microscope, though it is much coarser in structure. A rock very similar to these is found as loose blocks in the yellow tuff of Monte Vir- ginio. This is very much finer grained than any of the preced- ing ; contains some plagioclase but no glass. Leucite -te phrite . — This seems to be quite common in the region though not so much so as leucitite. It is represented by a specimen from a lava stream in the crater wall just below the town of Bracciano. It is rather dark and fine grained, but rough in texture and not aphanitic. Scattered through the groundmass are very many small clear leucites and some very small black augites. Under the microscope it closely resembles the similar rock of the Bolsena region, having the same doleritic structure. The irregularly shaped leucites show strong double refraction, and ' RosENBUSCH, Mikr. Phys. II, p. 1233, 1896. ITALIAN PETROLOGICAL SKETCHES 43 carry few inclusions. The augite is in stout prismatic crystals, green in color and generally darker toward the center. Large grains of magnetite abound, which are accompanied by orange hematite flakes. Many lath-shaped crystals of a basic labrado- rite, with a little orthoclase, are present with these, and the interstitial colorless glass base contains only few augite micro- lites and some minute opaque grains. An analysis is given in Table I, No. 6. Lencite-pJionolite. — My only specimen of this was collected at a rather thick flow on the northwest shore of the lake close to the water's edge. The groundmass is compact and fine-grained, of a very light gray color and with a slightly greasy luster. Clear rather glassy leucites and minute black augites occur as phenocrysts. Examined with the microscope the large leucites show the twinned structure very finely, and the prismatic augites are of an olive-green color, with the pleochroism and other characters of aegirine-augite. In the holocrystalline groundmass are many small olive- green aegirine-augite prisms, a few colorless haiiynes showing the characteristic inclusions, and some magnetite grains. The feldspars are represented by stout crystals and grains of alkaline feldspar, which are occasionally twinned. There is also present a residual base of colorless, feebly doubly refracting nepheline, whose identity was established by treatment with acid and fuchsine. Plagioclase is absent. A peculiar feature of the groundmass is the presence of round spots of a clear substance, whose outlines are defined by rings of minute augite microlites with some magnetite grains. These seem at first sight to be groundmass leucites. Close examination under crossed nicols, however, reveals the fact that onl_y a small percentage of them is really of this mineral. The majority are composed of nephe- line, with some orthoclase grains and crystals, the latter occa- sionally in fan-shaped aggregates. We have here, then, a case of paramorphism of leucite into " pseudo-leucite " — a mixture of nepheline and orthoclase, such as has been observed in Arkansas,' 'J- F. Williams, Ark. Geol. Sur., 1890, II, 267. 44 HENRY S. WASHINGTON Brazil/ and Montana.^ It is difificult to understand why the small groundmass leucites should suffer this change, while the larger phenocrysts remain preeminently fresh and unaltered. Bucca does not describe any of the leucite rocks of the Bracciano region. THE CERVETERI REGION. This region 3 lies about lo''™ southwest of Lake Bracciano and north of the small town of Cerveteri, so famous for its Etruscan tombs. It consists of a small group of hills, extend- ing about 8^" W. N. W. and E. S. E. (about parallel with the coast line, and 4 to 5'^'" broad. They are of no very great alti- tude, the hills reaching their maximum height of 384 meters in Monte Cerchiara, in the center of the group. I could only spend the better part of one day at the eastern end, but as far as my observations permitted me to judge they are a mass of domal eruptions resting, as Tittoni points out, on Pliocene beds. They are composed almost exclusively of acid non-leucitic rocks and their tuffs. Some leucitite is met with at the eastern end, but these leucitic lava streams probably belong to the Bracciano volcano proper. Since this leucitite is met with beneath the "trachytic" masses, its occurrence is of great interest as showing that the earliest leucitic outflows of Bracciano are of an earlier date than those of Cerveteri, though they continued after these latter had ceased. There seems to me to be an intimate con- nection between the Bracciano center and that of Cerveteri, and probably also that of Tolfa, but my opportunities for observation were so few that I do not feel able to discuss this point at present. Toscanite. — To this group belong all the specimens collected by myself, and also apparently all those described by Bucca, except those of a few leucitites. Their prominent mineralogical and chemical characteristics have already been noted. ' HussAK, Neu. Jahrb. 1892, II, 146. ^ PiRSSON, Am. Jour. Sci., II, 194, 1896. 3 It is included on the Bracciano Sheet (Foglio 143) of the Italian Geologic Map. ITA LI A N PE TR OL O GICA L SKE TCHES 4 5 The rock of which Monte Cucco, at the extreme east end of the region, is composed, is a light gray porphyritic rock, many glassy feldspar phenocrysts and some of biotite lying in a rather vitreous brownish groundmass. Under the microscope there appear well shaped and uncorroded clear orthoclase and plagio- clase phenocrysts, the former in the majority. The plagioclase is shown by Michel-Levy's method to be labradorite of the com- position Abo An 3, the symmetrical extinction angles of the lamellae of the Carlsbad individuals being in one individual 26° and in the other 29°. There are also many tables of an olive green biotite, which shows no corrosion or alteration phenomena. A few large colorless diopsides also appear. There is no mag- netite, and no quartz could be detected. The groundmass is highly vitreous, the glass base being colorless or of a very pale brown. It shows perlitic cracking in great perfection. Through it are sprinkled small (o.0 1-0.02) microlites of diopside, and a few stout orthoclase microlites which often show "horns" at each end. An interesting feature is the occurrence in abundance of very small (about 0.05™™) forked and sheaf forms of orthoclase, which correspond to the so-called keraunoids in Ischian trachytes already described by the writer.' These keraunoids are of such minute dimensions that they exert only a feeble action upon polarized light, but examination of the largest shows that the axis of greatest elas- ticity a lies parallel to the length, and that therefore they are elongated in the direction of the axis a. The use of high powers proves that they are in every way identical with those described from Ischia, except that they are of much smaller dimensions and more delicate. Bucca's description of this rock closely agrees with the above, though he speaks of the biotite as being brown, and the small augites and keraunoids seem to have been lacking in the groundmass of his specimens. Though no quartz is present he is of the opinion that these rocks are to be referred to the same acid group as those of Monte Calvario. This opinion is con- ' Washington, Am. Jour. Sci., 1, 375, 1S96. 46 HENRY S. WASHINGTON firmed by my analysis given in Table I, No. 2, which may be taken as representative of the rocks of this region. The rocks of Monte Lungo and Monte Ercole, west of Monte Cucco, are rather darker and rougher in groundmass, and look like the Arsotrachyte of Ischia, minus olivine. The plagioclase is, however, rather more basic, having a composition Ab^An^, and the biotite rather browner in tone though still green, and colorless diopside phenocrysts rather more frequent. The glass base is rather dark brown, and the small orthoclase keraunoids are much more abundant, as are also the diopside microlites, which are here prismatic in habit. The rock of Belvedere del Principe south of Monte Ercole, is also closely similar in general characters, but the orthoclase keraunoids are here so abundant as to give the base a hyalopilitic structure. The glass is brown and the diopside microlites are prismatic in habit and show some flow structure. It may be noted that absolutely no magnetite is to be found in any of these rocks. Small apatite needles are found in all, especially as inclusions in the feldspars. Bucca mentions a brown hornblende as present, known by its prismatic angle and oblique extinction, but careful research failed to reveal it in my specimens. As has been said, analysis No. 2 of the Monte Cucco rock may be regarded as representative of them all, and its close resemblance to that of the Tolfa and Calvario rocks will be noted. Further remarks on these points must be reserved for the final paper. Leucitite. — This is represented by a specimen from a flow met with at the bottom, of the deep ravine immediately to the west of Monte Cucco, whose rocks overlie it, though tuffs and soil conceal the contact. It is almost identical with the leucitites from the south of Lake Bracciano, and is composed essentially of round and irregular leucites with insterstitial green augite needles. Some magnetite and a very little orthoclase are also present. The rock of which the ruined Castle Dannato is built was obtained from some now forgotten quarry in the neighborhood. ITALIAN PETROLOGICAL SKETCHES 47 It is also a Icucititc, and ol the same composition as the pre- ceding". It is, however, of much finer structure and closely resembles the leucitite of Sassi Lanciati near Bolsena. It carries segregations of coarse-grained, granitic masses, made up of large grains of green augite and a few well formed crystals of olixine, with much leucite which shows excellent twinning lamelLx, and which takes the place of the feldspar in typical gabbros. Some cavities are filled with calcite. These segrega- tions closely resemble the leucitic blocks briefly described by Lacroix from Lake Bracciano, and may be compared with the Missourite recently described by Pirsson.' THE TOLFA REGION. This region consists of a triangular group of hills, lying a little more than lo*"" northeast of Civita Vecchia and some 20''" west of Lake Bracciano. Its highest point is Monte Sassicari, 525"" above sea level. Since my visit of a few hours was con- fined to the southern angle I will not go into the topography, which moreover presents little of interest here. It may only be noted that the hills are composed essentially of toscanite, which in general is so extremely decomposed to alunite that a fresh specimen of rock is indeed a vara avis. De Stefani points out that the eruptions in all probability took place in post-Plio- cene time, differing in this from Ponzi and others, who would place them earlier. Petrography . — According to vom Rath there are two distinct kinds of "trachyte" represented in the region. The one is a "sanidine-oligoclase-trachyte" with light gray compact ground- mass, in which lie phenocrysts of sanidine, oligoclase and bio- tite. This is more abundant in the northeastern part of the region, often shows a bed- like parting, and is comparatively little subject to decomposition. The other is more like a pitchstone, and is almost universally decomposed. Its groundmass is dark brown and resinous, with phenocrvsts of sanidine, biotite and ■ i'lRsso.N, Am. Jour. Sci., II, 315, iSgb. 48 HENRY S. WASHINGTON augite. To these Busatti adds a third variety, though his description hardly seems to warrant the separation. I am inclined to agree with de Stefani in thinking that there is only one kind of rock, and that the differences noted are of very small importance. From vom Rath's analysis of the pitch- stone-like trachyte and from mine of the rock from Tolfa it will be seen that the two varieties closely resemble each other chemically and that the rock belongs to the group of Toscanites as already defined. The only fresh specimens which I obtained were collected at the hill on which stands the old castle above the village of Tolfa. In all other places I found it so decomposed as to be absolutely worthless for petrographical study. The groundmass is very compact and bluish gray, and speckled with minute black spots of augite. Very many large glassy feldspar phenocrysts, which often show twinning, with some biotites are scattered through it. Under the microscope this rock presents an appearance almost identical with that of the Monte Cucco rock. The rather abundant plagioclase is an acid labradorite having the composition Kh^ Auj, as determined by the extinction angles (22° and 23°) of the albite-twinned lamellae of the two individuals of a Carls- bad twin.^ The feldspar phenocrysts are quite often corroded. The hyalopilitic groundmass has a colorless glass base, which is thickly crowded with diopside prisms and laths, rather than keraunoids of orthoclase. De Stefani'' mentions a leucitic rock which he obtained at the base of Monte Elceto between the toscanite and the underlying Cretaceous rocks. Rosenbusch, to whom a specimen was sent for examination, reports that it is a "leucitite perfectly identical with the leucitic rocks of Albano and the vicinity of Rome." ^Busatti states that the plaglolase is oligoclase, but his determination seems to rest on insufficient data, as he only mentions its " polysynthetic structure " but gives no angles, as neither does vom Rath. ^ = De Stefani, Boll. Com. Geol. Ital., 1888, 224. SiOs .. AUO3 . Fe^O, FeO .. MgO . . CaO. . NagO . KjO.. H„0 .. Tibo.. Sp. Gr. ITALIAN PETROLOGICAL SKETCHES 49 TABLE I. I 2 3 4 s 6 64.04 66.24 65.19 67.61 47.89 49-73 14.48 15.64 16.04 14.04 18.25 19.20 1-73 1. 16 1. 16 4-93 5-50 4-35 2.19 2.48 5-40 3-64 2.41 1.03 0.89 0.99 0.65 3-68 2.63' 4.00 2.17 2.92 3-71 8.70 7.96 4.14 2.05 2.26 5-50 2.60 1.99 3-65 6.60 6.11 2.41 8.23 9-39 2.06 3-25 1.S5 2.28 0.65 1. 19 0.28 100.19 0.77 99.76 99.00 101.60 99-34 100.00 2.542 2.455 2.509 2.537 2.781 2.655 55.87 21.82 2.34 1. 10 0.48 3-07 4.81 10.49 0.34 100.32 2.551 1. Toscanite, Monte Calvario, Bracciano, A. Rohrig anal. 2. Toscanite, Monte Cucco, Cerveteii, H. S. Washington anal. 3. Toscanite, Castle Hill, Tolfa, H. S. Washington anal. 4. Toscanite, Tolfa, vom Rath, op. cit., 596. 5. Leucitite, Crocicchie, Bracciano, H. S. Washington anal. 6. Leucite-tephrite, Bracciano, H. S. Washington anal. 7. Leucite-phonolite, Lake Bracciano, H. S. Washington anal. Henry S. Washington. ' By difference, through loss of Mg precipitate. MODE OF FORMATION OF TILL AS ILLUSTRATED BY THE KANSAN DRIFT OF NORTHERN ILLINOIS. Perhaps at no time during the Quaternary era were the climatal conditions of Illinois of such a nature as to originate a glacier independently of the ice introduced into the territory by outflow from the vast ?ievc to the northeast. Consequently, when near the culmination of the Kansan epoch, a very early representative of the Lake Michigan glacier overspread nearly the whole of the territory of the present state of Illinois, it advanced across a region in which the indurated rock was but thinly covered by a nearly continuous mantle of residuary material, mostly clay, sand and angular gravel ; and the ice thereupon proceeded to manufacture this into till. In a part of northwestern Illinois, especially in Stephenson county, the glaci- ation was of short duration and never repeated, and this district therefore, presents one of the best fields for the study of the contact phenomena between the base of the glacier and the preglacial land surface. There are scattered over the area incomplete deposits representing every stage in the process from solid rock to typical till. By a careful study of these imperfectly formed masses of till we may infer the process by which the ice manufactured the residuary material and underlying rock into the various types of deposits due directly to glacial action. The necessarily limited nature of this paper will compel me to state the hypothesis which seems to best explain the phenomena known to me, and simply refer to the localities where each stage is illustrated, mentioning a few of their most significant fea- tures. The ice, in advancing across Stephenson county, moved westwardly, and within fifteen miles of the Driftless Area, in a decidedly northwesterly direction. During this time of general 50 MODE OF FORMA TION OF TILL 5 I ach'ance, glacial abrasion was at a minimum, and had the ice then melted from this district, without further movement, its drift phenomena would be insignificant and uninteresting. But it so happened that, during the general recession of the glacial front, there were repeated slight readvances, forming a peculiar variety of incipient moraine, consisting of knolls and short ridges of angular limestone debris and of stratified water-worn gravel and sand. Between the marginal accumulations of a distinctly morainic type, there are other deposits of somewhat similar nature, but which belong to the so-called ground moraine. In other words, the period of most pronounced glacial action in the Kansan epoch in northern Illinois, occupied a position considerably later than the culmination of the epoch. This I attribute to a milder climate, causing a recession of the ice-front, but yet giving more free movement to the glacier and softer material to work upon. I have gone into this short explanation of the glacial history of Stephenson county, so as to exactly locate the age of the deposits the significance of which I propose to discuss. We may gain some idea of the condition of the surface pre- vious to the arrival of the ice by a study of a north-south belt along the eastern boundary of Stephenson county, and about thirty miles back from the glacial boundary. Here we will find frequent exposures of the semi-decayed upper portion of the Galena limestone with its overlying undisturbed residuary clay.^ The Galena formation is usually a heavy-bedded, sub-crystalline dolomite, shaly in certain thin layers and extensively jointed and fissured. Within ten to fifteen feet of the surface, weather- ing has opened certain deposition-planes, separating the rock into layers from two to four inches in thickness. As the rock is composed of small rhomboidal crystals, the next stage of decay consists of a solution of the bond between the crystals resulting in a loose mass of angular grains, macroscopically ' This is finely displayed in a road cutting one mile south of Egan in Ogle county, and in an excavation of the Chicago and Great Western railway, two miles east of German Valley in Stephenson county. 52 OSCAR H. HERSHEY resembling fine yellow sand. By a continuance of the process the crystals are dissolved, leaving a slight residue which, accumu- lating to a thickness of two to ten feet, becomes a highly oxidized, dark red, very fine-grained, structureless clay. Certain layers of the dolomite abound in chert, which breaks up into angular fragments, averaging about the size of a walnut. These abound in the residuary clay, and on hillsides often pass hori- zontally from the solid rock into the red clay, with but little disturbance of the lines of stratification. This is one of the strongest proofs of the undisturbed nature of the residuary clay at the localities previously mentioned. In fissures and pit- shaped holes in the surface of the rock, the red clay may extend down from ten to thirty feet. This particular form of its occur- rence can be observed in almost any rock-cutting in the county.' I . The first effect of the contact between the base of the glacier and the mantle of residuary clay, was a rearrangment of the latter. It was pushed forward and downward, but as its down- ward progress was limited to a slight compression, the effect was a crushing or kneading of the mass. The faint evidence of stratification in the undisturbed deposit furnished by the layers of chert, was totally destroyed and the fragmental chert scattered indiscriminately through the mass. All deposits of rearranged residuary clay are referred to this stage when they contain no foreign material whatever. As they, in many cases can be. proven to have been moved but a very short distance, their transportation and deposition were clearly exclusively sub- glacial.^ . ^ For a thorough discussion of the residuary material accumulating over the Galena limestone in unglaciated areas, the reader is referred to Chamberlin and Salisbury's excellent report on the superficial geology of the Wisconsin Driftless Area in the Sixth Annual Report of the U. S. Geological Survey. "A careful study of the county of Stephenson would probably afford several hun- dred localities — of very limited extent, however — where th^ process of till making was stopped in this first stage. Good exposures can be found by following the C. G. W. R. R. from South Freeport to Egan, or the I. C. R. R. from Feeeport to Everts. Another, exposed in a small ravine on the south side of Yellow creek, is important from the fact that the rearranged residuary cherty clay has been pushed off of its original rock ridge onto the surface of a lake deposit. MODE OF FORMA TION OF TILL 5 3 2. The process in the second stage is merely a continuation of the last. The material has been transported farther, kneaded lono-er and more thoroughly commingled. In addition, a small percentage of foreign drift, mostly well-rounded pebbles of small size, has been worked into the deposit. Still, at the com- pletion of this stage, the fine-grained, dark red clay basis and predominance of angular white chert indicate the close relation between this very incomplete till and the undisturbed residuary material. This is the nature of the ground moraine over practi- cally the entire northern half of the county. But in looking for exposures of it, it is necessary to remember that the upper three feet of the typical or completed till in this district have been highly oxidized during the following interglacial epoch, so that in color they often resemble the deposits of stage No. 2. The latter never were of any other color than dark red and reddish brown.' The mode of transportation of this imperfectly formed till appears to have been exclusively subglacial. In a very large proportion of exposures in Stephenson county, there is a con- tinuous section from an undisturbed preglacial residuary clay, through every gradation to what may be considered typi- cal of the " semi-residuary " class of drift. There is absolutely not an iota of evidence that any of this material has been lifted from the earth's surface and enclosed in ice. Clear evidences of a kneading or rolling over of the mass are abundant. 3. There existed under the glacier, certain areas where depo- sition was being carried on at the same time that abrasion was active in others. So that, while in many portions of the county, the manufacture of till was carried into the second stage only, on certain closely adjoining areas this was but the beginning or early stage of the glacial action. In these latter, the ice after hav- ino- removed the red clay, proceeded to attack the loose sand-like mass of dolomite crystals, which it quickly disposed of by incor- porating with the deposits of stage No. 2, furnishing their first ' This red til! is best exposed in several cuttings of the I. C. R. R., about three miles southeast of Winslow. in Stephenson county. 54 OSCAR H. HERSHEY supply of calcareous material. A few sections show it inter- stratified with till, and it can sometimes be traced back to its original undisturbed position/ Having been transported but a score or at most a few hundreds of feet, there is no room for anything but subglacial action. We have disposed for the present of all the loose material on the surface of the solid rock, and we now come to the most inter- esting part of the process. 4. Stephenson county is a comparatively hilly region, although the hills are not high, steep, or close together. A sec- tion throusfh a hill would show the Galena limestone as a series of practically horizontal layers, not cemented, but held in posi- tion by gravity and the projections on the upper and under sur- faces of the layers. The ice in moving westwardly across the country, upon ascending the eastern slope of a hill from which the loose material had been removed, exerted a powerful pres- sure on the edges of these layers. Near the top of the hill, the pressure overcame the friction, and the upper layers of the rock were pushed forward, sliding on a lower unmoving stratum. In many cases, these transported rock ledges were not broken up, but moved forward as a solid mass, being found in that condi- tion to day. They may be ten or fifteen feet in thickness, and several hundred feet in length, and perfectly horizontal, so that the fact of their not being insitu^owXdi not be known were it not for their unusual position, partially obstructing valleys, produc- ing a topography radically different from the preglacial ; and from their overlying drift of various kinds, including stratified water-worn gravel ; and more often than otherwise underlain by loose angular gravel from ten to thirty feet in thickness.^ In transportation they were clearly pushed in front of or under the extreme marginal portions of the ice. None of them have been removed far from their original position, and this with much ' Small exposures of it, after being disturbed but befoi'e being combined with the red clay, may be seen at nearly every outcrop of the latter, notably in the railway cut- tings of the southeastern part of the county. ^ Locality where best exposed — six miles east southeast, of Freeport, in a high ridge, on the north side of the I. C. R. R. MODE OF FORMA TION OF TILL 5 5 other evidence, negatives the idea that the basal portion of the glacier "plucked " these huge masses of rock from the ridges and transported them in its body. Sometimes the pressure of the ice resulted onlv in a slight folding, or a tilting and faulting of the rock strata without mov- ing them to a distance from their original position. This may affect them to a depth of twenty or thirty feet, and ma}- include some of the solid, heavy-bedded, unweathered layers. In these cases, at least, no question can remain that the action was extra- glacial or marginally subglacial. Their importance lies chiefly in the fact that they prove beyond dispute that even the Kansan ice-sheet could, at times, exert powerful forward pressure on the rock over which it moved. 5. The forward movement of the unbroken rock mass was rarely continued many scores of vards before some obstruction was encountered, often in the form of an upward slope of the land, increasing the friction to such a degree that the transported ledge was unable to withstand the pressure and general fracture resulted.' In a few cases, the impression made on the observer is that the mass remained unbroken while the stress accumulated, until, becoming too great, the entire ledge was suddenly frac- tured into innumerable small angular pieces, under a well-known principle recognized as a condition of the formation of certain limestone breccias in various portions of the earth. Usually, however, there is clear evidence of a kneading or rolling over and crushing of the mass, producing an internal structure which cannot be simulated by the ])roduct of an}^ other known process. When a series of semi-decayed or loosely compacted strata being forced forward in front of the ice, are checked by some obstruc- tion, they tena to corrugate in a manner somewhat similar to the Appalachian type of mountain building ; and the disturbance is greatest in the vicinity of the obstruction. But when the ice 'The deposits representing this stage are quite numerous in the Pecatonica basin, over an area about three miles in diameter, lying immediately southwest of the village of Dakota, and another west of the village of Everts. They are usually in the form of dome-shaped knolls and short ridges, and forming conspicuous features of the drift topography, it is unnecessary to more minutely describe their localities. 56 OSCAR H. HERSHEY overrides the deposit, if the basal friction is great enough, it will turn up the strata in contact with it, twist them over on to the as yet undisturbed strata, and inaugurate a motion in the mass which may be not inaptly compared to the movement of a roller under a heavy weight. Rarely, however, will the actual action be as perfect as it should theoretically be, but usually there is combined with the rolling process a forward and downward thrust which crushes the mass. The more resistant portions of the ledge break into huge angular blocks, the semi-decayed lay- ers into smaller fragments intermingled with a great quantity of calcareous sand. The rolling is obviously produced by the "drag" of the forwardly moving glacier, while the thrust is the result of its great weight. This kneading process is not only theoretically probable, but has given rise, as I have already intimated, to phenomena in Stephenson county, Illinois, which are explainable only under the supposition of having been pro- duced by its action. 6. By a continuance of the kneading process, the larger blocks are crushed, all evidences of the original lines of stratifi- cation are destroyed, and the deposit for the first time bears a slight resemblance to certain very stony tills. Few of the par- ticles are smaller than the crystals of which the original lime- stone was composed, and, therefore, the deposit may still be clas'^ed with the "transported rock ledges." Moreover, in this stage there are practically no foreign rock fragments although a few Canadian pebbles may be found imbedded in it far from the surface, proving its glacial age.' 7. The process of manufacture between the last stage and the typical glacial till may proceed along several lines, all simi- lar in the general features of the method, but differing in details. Several deposits along the Galena road, one mile northwest of Freeport, and along the C. G. W. R. R. between German Valley and Egan, seem <-o represent a stage slightly in advance of that just described. Here the angular limestone debris is com- ' A deposit representing tiiis stage is finely exposed in a gravel pit, one mile north of Freeport, where the angular gravel overlies water-worn drift gravel. MODE OF FORMA TION OF TILL 5 7 mingled with a considerable quantity of rounded drift of which a large portion consists of Canadian bowlders. The foreign material may reach as high as 50 per cent, of the mass, but usually falls below 10 per cent. The bowlders belong to the "englacial " drift, and the deposits containing them are marginal, but I introduce them here because by a marked readvance of the glacial front, they may be reworked into till. 8. In other cases, the ice was transporting the red, imper- fectly formed till, described under stage No. 2, contemporane- ously with the angular limestone debris, and they often came in contact, either becoming thoroughly commingled or perhaps only interstratified. When the intermingling is complete, the deposit has the appearance of a very stony till of a light brown color, and the material of which is evidently mainly of local derivation.^ 9. Just previous to the incursion of ice into the Pecatonica basin, in Stephenson county, its area was occupied, at the lower levels, by several successive extra-glacial lakes, which deposited glacial silts, some of which resemble typical loess, and others a modified or water deposited till. During the glaciation of the region, the abrasion of these comparatively soft silts was remark- ably small, yet sufficient to leave an impress on the character of the till which now overlies them. For it is observed that, while the ground moraine over the northern half of the county, where these silts were poorly developed and rarely reached by the ice, is very similar in color and texture to the residuary clays, over the southern third of the county, where the land lies lower and the extra-glacial silts were strongly developed, the ground moraine is in general of a quite different character. The marginal deposits are similar in both districts (consisting largely of angular limestone debris and stratified water-worn gravel and sand) , but on the inter-incipient-morainic tracts in the southern district, there is often seen a till identical in all important respects with that which is exposed in long and repeatedly glaciated regions 'Typically developed along the road between Egan ami Lightsville in Ogle county, about two miles north of the latter village. 58 OSCAR H. HERS HEY and which I have referred to as "typical till." It is a stiff, unstratified, light yellowish gray clay, containing irregularly dis- seminated subangular blocks, largel}' of local derivation, but also an appreciable percentage of foreign rock fragments, many of which are Canadian in origin. The Niagara limestone pebbles are next in abundance to the Galena limestone and chert, and are often beautifully striated. A belt of this variety of glacial clay occupies the country between Yellow creek and a high upland area about four miles south of it. Its eastward and westward limits are indefinite, but its length may properly be included within ten miles. It can be observed best by proceed- ing due south from Freeport, where at the distance of about four miles, it will be seen to assume distinct drumloid characters. In particular, one of these pseudo drumlins, while it has a core of stratified sand and gravel, is overlain by an eight-foot stratum of yellow till so nearly identical in constitution with the typical till of a highly glaciated region, that no appreciable difference could be detected. This drumlin (?) is also the best locality in this county for securing finely striated stones. Now, a careful study of the composition of this till belt shows that, [a) the mass of its clay may be referred to abrasion of the silt formations which it frequently overlies : [b) its foreign rock fragments must have been brought into the district by the ice-sheet, but independently of the accumulation of the till, their occurrence in which is an accident of deposition ; [c) the iron in the till was derived from the residuary clay of the region, as was also the large amount of angular and subangular white chert ; and [d ) the subangular Galena limestone blocks which are quite abundant but generally overlooked because of the more attractive appearance of the for- eign drift, must have been derived from the country rock, very close to its present position, by the process which has been described in this paper. It may be objected to the first proposition^hat the difference between the till in this limited belt and that in the remainder of the county, may be due to a greatly increased introduction of foreign material along some line as the broad, shallow basin MODE OF FORMATION OF TILL 59 which occupies the southern third of the county. The existence of this basin I recognize as the cause of the much larger propor- tion of foreign drift pebbles in its vicinit}', for it favored sub- glacial transportation of them, but I do not think it alone can explain the phenomena connected with the above described belt of typical till. For (i) when we go east on the line of glacial movement, the characteristics of this till disappear, and in the extreme eastern portion of the county, what little till we do find in the Pecatonic basin, is mostly of the very imperfectly formed "semi-residuary" variety; and (2) if we study the limits of this characteristic till belt and the direction of ice movement, in con- nection with the limits of the main body of the extra-glacial lake clays and silts, we find such a relation between them as to clearly indicate the derivation of a large portion of the former from the latter. In short, the composition of the till in the belt now under discussion, points to its being only a stage in advance of that described under stage No. 8, this advance being due chiefly to the introduction of a large amount of clay and silt, the result of an accidental passage of the ice over a semi-plastic lake deposit, which resisted erosion comparatively well, but yet suffered some abrasion at its surface. Therefore, as the yellow till occurs over and in the immediate vicinity of the stratified clays, there is no room for an englacial transportation, which is, moreover, quite unnecessary, and in opposition to the comparative abundance of striated pebbles. In summarizing the evidence presented in the preceding pages, I would subdivide the product of direct glacial action, in the central Pecatonica basin, into classes as follows : Class I. Semi-residuary drift. Till composed largely of rearranged red residuary clay and chert, with comparatively few foreign pebbles. Class II. Angular local limestone debris. Varies between contorted, tilted, and transported unfractured rock ledges, and fine limestone breccia. Class III. Consisting largely of a combination of the two 6o OSCAR H. HER SHE Y preceding with the addition generally of Canadian pebbles and small bowlders. Usually seen in marginal deposits. Class IV. Typical till. Light yellow gray calcareous clay containing pebbles and small bowlders of local and foreign derivation, many of the latter beautifully striated. The divisional lines between these classes, in the Pecatonica basin, must always be arbitrarily placed, as there are gradations from those deposits which are typical of one class to those which are typical of another. There is, also, no definite rule in their distribution for, while the four classes evidently represent four successive general stages in the process of manufacturing typical till, and should, therefore, naturally be expected to occupy suc- cessive belts from the outer glacial boundary back or east, they are in reality scattered indiscriminately over the entire district. However, Class I prevails in the northern half of the county ; Class 11 in a north and south belt which, extending across the county, has its western boundary on a meridianal line which passes a few miles west of Freeport, and dies away to the east near the Winnebago county line ; Class III prevails over a belt several miles in width extending diagonally from the south- eastern corner of the county to near Freeport ; and Class IV is principally developed south of Yellow creek, over and near to the remnants of the Lake Pecatonica clays. Perhaps the most remarkable feature of the drift of this county is its anomalous distribution. Over the western one- third of the county, the drift, although comparatively thin, is generally distributed. Over the central belt, the evidence of vigorous glacial action is strongest. While in the extreme eastern one-fourth of the county, which was glaciated longest, there is very little drift of any kind, and there are areas, several square miles in extent, of nearly bare rock, and others where the preglacial residuary clay still remains undisturbed. It is evident that the conditions which controlled glacial action varied locally within wide limits, so that excessive abrasion [of the rock surface might occur with a less weight of ice and a less length of glaciation, than on a neighboring area with similar MODE OF FORMA TION OF TILL 6 1 topography would not suffice even to remove the residuary clay. When I make the statement that the four classes of glacial deposits discriminated in this district, occupy hills equally as high, as narrow, and as steep, also occupy valleys equally as YiQ. I.— The Kansan ice at work in northwestern Illinois. deep and broad, and trending in the same direction, we are pre- pared to accept the following conclusions : I That certain glaciers, notably that which glaciated north- western Illinois, in passing over a slightly hilly region, exerted very unequal pressure on the land surface, this inequality not always being directly due to the topography of the immediate vicinity. 2. In crossing hills of moderate height, they sometmies strongly abraded the crests, while on closely adjoining areas they deposited ground moraine on hills of similar height and shape. 3. The areas of maximum and of minimum glacial action were generally permanent, or approximately so, throughout the time of glaciation. 4. In studying areas in the Kansan drift region or at least that portion of it which is in northwestern Illinois, the relative length of the glaciation in different localities cannot be even inferred from the apparent severity of the glacial action. In conclusion, I will state in more definite language what I conceive to have been the position with relation to the mass of the ice, of the unstratified drift of Stephenson county, Illinois. The derivation, transportation, and deposition of the material occurred mainly under the peripheral portion of the glacier. Mere, through the thinning of the edge, the weight was less, and the force of the ice movement having decreased, there was a 62 OSCAR H. HERS HEY tendency of the extreme marginal portion to turn up, producing a base of a form somewhat as in the accompanying figure. There was a struggle for the mastery between the advancing ice and the accumulating debris. The incipient morainic or marginal deposits of Stephenson county, including the transported rock ledges, belong to the extreme outer border of the ice and were transported least. Those which are most thoroughly kneaded belong to a position at about b. The drumlin (?) area south of Freeport may be placed at c, while much of the ground moraine was formed and deposited still farther back under the ice. By recession and repeated slight readvances of the ice-front, with a partial reworking of old deposits, the apparently indiscriminate distribution of the drift of this county may be accounted for. When the ice melted away entirely, the englacial bowlders which had become superglacial through ablation of the border portions of the glacier, and are so represented in the figure, came to rest upon the surface of all the other drift of the district and the areas of nearly bare rock represented at a. If the advance was continued far, the ice overrode the coarse angular gravel at ^, crushed it finer, brought up "semi-residuary" till and a little foreign drift from farther back, and combined them into typical till. This gradually fell behind the advancing border of the ice, and was redeposited as ground moraine, locally accumulating into drumlins and smoothly undulating drift areas, while the extreme outer portion of the glacier was forming a new deposit of angular local limestone debris. Oscar H. Hershey. Freeport, III., January 5, 1897. THE GEOLOGY OF THE SAN FRANCISCO PENINSULA. The Coast Ranges of California embrace an extensive region in which are presented complicated but exceedingly interesting geologic problems. Much attention has been given to these mountains for many years, but with the exception of a stud_y of the quicksilver deposits, it is only recently that we have had presented in a thorough manner the results of the detailed exami- nation of local areas. Professor Lawson's recent report upon the geology of the San Francisco Peninsula^ is perhaps the best yet made of any local area in the Coast Ranges and illustrates well what modern methods of research can accomplish in a complicated field. Notwithstanding the excellencies of the report, there is an infe- licity displayed in the discussion of several problems which is regrettable in a study of this kind. This has, however, neces- sarily resulted from the method pursued, in that the investigator has given his attention almost exclusively to a narrow field of complex geology and has failed to make use of the results of the work of others, concerning questions which the phenomena in that field did not illuminate. It is absolutely necessary for the appreciation of many facts in any local area, and for the philo- sophic discussion of the history of that area, that the student should have a general knowledge of the relations existing over the region as a whole. As a result of some experience in the Coast Ranges I feel called upon in the interest of geological progress to express most profound objections to a number of conclusions reached by Professor Lawson concerning some of the vital questions involved in the geological history of this region. Consequently in answer to his frank request for friendly criticism, I will take up the dif- ' U. S. Ge(jl. Sur., 15th Annual Report, pp. 405-476. 63 64 HAROLD IV. FAIRBANKS ferent points which I feel are open to question in the general order in which they occur in the report. There are several statements in a synopsis of this report/ published a year and a half ago, which will be included in the criticism. There is a failure, which is without doubt due to an oversight, to give recog- nition to some contemporaneous and earlier work in the same general field and upon the same topics. - Passing over the Montara granite and the associated marbles which exhibit the same relations and have with- out doubt the same history as similar rocks in the Gavilan and Santa Lucia ranges, we come to the author's Francis- can series, the oldest uncrystalline terrane. Professor Lawson divides it into five petrographic divisions. The lowest consists of conglomerate, sandstone, shale, etc., and is well exposed at Point San Pedro. He considers that this division may possibly be older and underlie the strata north of San Pedro valley uncon- formably, because fragments of shale similar to that at the point occur in the sandstone north of the valle}^ The fact seems not to have been noticed that fragments of similar shale are found in the basal conglomerate on the point. The conglomerates are identical in character with those at the base of the Golden Gate series on the Monterey coast, and are without much doubt of the same age. The " foraminiferal limestone" and "radiolarian chert" form perhaps the most interesting portions of the series. They are dwelt upon in detail, particularly the " cherts." According to Professor Lawson the latter are hard siliceous rocks of varying degrees of purity, and are "prevailingly of a dull brownish red color, although other shades occur." He says that "in many 'Am. Geol., June 1895. ''Bull. Geol. Soc. of Am., Vol. XI, pp. 71-102. The results of this work were •■ead before the Geological Society of America nearly a year^in advance, and published six months prior to the first of the papers under discussion. The subjects concerned were the position and character of the marbles of the Coast Ranges, and especially the rocks constituting the Golden Gate series (Franciscan series of Professor Lawson), the nature of the sandstones and the radiolarian origin of the jaspers, as well as the geologic position of the series. GEOLOGY OF SAN FRANCISCO PENINSULA 65 cases thev are true jaspers and have been so designated in some o the earner deser.ptions of them." After discuss.ng the.r variability he again says : " In view of this variation nt petro- : phic character, it has been deemed best to refer to these focks by the old and familiar name of 'chert.- It seems to me that, on the contrary, the designation "chert" is^ not only not as appropriate for rocks of this character but it does away wrth ..an old ind familiar name" for no suflfieient reason. T'- desig- nation "jasper" was used by Blake, Newberry, and Wh.tney. According to Geikie, "Chert is a name applied to .mpure cal- careous varieties of Hint in layers and nodules which are found among the Pateozoic and later formations, especially but not exclusively in limestones." The siliceous bands in the foraminiferal limestone which a.e referred to by Professor Lawson as "veins" I do not consider such in any true sense of the word ; they are more properly cherts or phthanites according to the original use of the terms, but in regard to the great body of siliceous rocks occurring inde- pendently there seems to be no use in making a change in terms The nature of the so-called " veins" inthelimestone seems not to have been clearly understood. On exposed surfaces of many feet in extent these siliceous bands stand out with grea distinct- ness Some of them are as even and regular as the limestone strata while others are discontinuous and more uneven. Ihey sometimes blend into the limestone but more commonly are quite sharply distinguished. These bands of chert are contem- 'aneous'd'eposits, being always conformable to the stratifica- Uon of the limestone, and differing most markedly from the veins of secondary origin. Thin slides prepared from a number of specimens show them to be thickly filled with organic remains of radiolarian character. . After a description of the jaspers several theories o their origin are considered. The theories are as follows : (l) siliceous springs in the bottom of the ocean, similar to those well known in vofcanic regions; (.) radiolarian and other siliceous remains which mav have become entirely disssolved in sea water, (3) 66 HAROLD IV. FAIRBANKS volcanic ejectamenta which may have become similarly dis- solved. The two latter are rejected and the first received, as having the most to support it, in the following words: "The hypothesis of the derivation of the silica from siliceous springs and its precipitation in the bed of the ocean in local accumula- tions, in which the radiolarian remains became imbedded as they dropped to the bottom, seems, therefore, the most adequate to explain the facts, and there is nothing adverse to it as far as the writer is aware." Some time since I proposed a theory^ to account for the origin of the jaspers substantially equivalent to the second given above. Professor Lawson's chief objection to the view of the organic origin of these rocks consists in the fact that they occur in lenticular masses instead of evenly bedded deposits. In his petrographic description it is stated that the "cavities of the radiolaria have been filled with chalcedonic silica and are in definite contrast with the non-chalcedonic matrix." With this last statement my experience is not often in accord. I have found every gradation in the specimens from those in which the radiolaria are distinctly marked, as Professor Lawson says, to those in which they are only faintly distinguishable from the matrix, or apparently absent. In my opinion this state of things gives good ground for the view that, owing to possible trans- formations through the action of sea water, and the secondary changes which are known to have taken place, there is no valid reason for denying the organic origin even when no organic remains are distinguishable. Professor Lawson has failed to recognize that the siliceous bands in the limestone must have had an origin similar to those occurring in aggregates by themselves. If the theory of forma- tion by springs is applicable to one it is to the other. The occurrence of these radiolarian jaspers interstratified with the limestone is a most suggestive fact. Similar conditions of sedi- mentation must have obtained in the one case as in the other, the only difference being that at one time calcareous layers were ' Bull. Geol. Soc. Am., Vol. VI, p. 85. GEOLOGY OF SAN FRANCISCO PENINSULA 67 formed, and .t another siliceous. Tire loeal bands »£ J''l'';='- ^ L fee perhaps in lateral extent, wholly surrounded by hme- one and sharply d.fferentiated from rt could not m all proba- b li V have been formed through the action of s.hceous spr.ngs^ The large lens-shaped bodies of mass.ve jasper ar also .bruptlv marked off from the inclosing sandstone o"- e. Heach lens were due to the action of one or more sprmgs the ct,rrents Tst of necessity have been weal, near the edges of the depos and could not possibly be conceived as "sufficient to deflect "diment-laden ounter-currents." If sedimentation were going o r lund the springs it is impossible that the silica cot. have been precipitated without more or less commingling with the sand The presence of similar radiolarian ,asper m the hn e- one IS suggestive of the view that to whatever cause the Tnticular fofm of the latter is due the same cause conditioned the similar outline of the former. Turner, says of the limestones in 'he KnoxviUe a Moun Diablo and of the older limestones in the Sierras that each "Tcare'ous layer is rather -eries of lerUicular od.s tha„_a rras::u:r;thr:h:-l.mi:;:neTodies are hundreds of feet in diameter." The belt of foraminiferal limestone described by Professor Lawson is equally bunchy, reaching a V "consid table thickness in places, as on the northern slope o Bl ck Mountain, and then contracting to very .m.ted propo - ons. It thus appears that as far as the shape of t^e deposits concerned the spring theory of ongm is equally applicable "^^Tirrements in the report concerning the condition of the formation of the jaspers do not exactly -"i- ^ F- inst on oaire 426, Professor Lawson says. It the springs fong the cu rents engendered might in some places have bee sufficfent to deflect sediment-laden counter-currents, and his way s fve to explain the general absence of clastic material in the chert." On page 466 he says : " At different more or less pro- 'Bull. Geol. Soc. Am., Vol. Ill, P- 394- 68 HAROLD W. FAIRBANKS longed periods during the accumulation of the series the bottom of the sea sank sufficiently rapidly to be out of the reach of lit- toral sediments." According to the observations of the author the body of the limestone is as free from detrital matter as the jasper, and except for the thickly scattered calcareous remains shows no more traces of organic origin than portions of the jasper. I cannot see that the origin through springs has anything whatever to support it. It is undoubtedly true that a subsidence or change in ocean currents gave rise to conditions favorable to the accumulation of beds of jasper and limestone. It is, how- ever, rather difficult to believe that this movement, which could not have been of a catastrophic kind, should have accorded exactly with the flow of hundreds, if not thousands, of springs over the sea bottom, which at one period were purely siliceous, and at another deposited nothing but pure carbonate of lime. If the currents were as strong as the author supposes, it does not seem possible that the radiolaria should have settled so thickly as we frequently find them, and besides the springs possibly being fresh would not form a congenial place for marine organisms. A short discussion is given by the author to deposits which he terms silica-carbonate sinter,^ the true nature of which seems not to be understood. He says : "Its occurrence in extensive sheets, roughly parallel with the bedding, suggests that it is a contemporaneous deposit, but it may possibly be a vein forma- tion. Its occurrence in the Aucella sandstones elsewhere, and in the San Francisco sandstones of the peninsula is of interest as a possible factor in the correlation of these formations." This is a case in which wider familiarity with the Coast Ranges and their mineral deposits would have readily settled a very simple question. These deposits of sinter are almost always associated with quicksilver ores forming Hheir gangue. The quicksilver deposits are known to date from post-Miocene times, and owing to their recent formation it is to be expected that ' Am. GeoL, June 1895. GEOLOGY OF SAN FRANCISCO PENINSULA 69 thev would occur at all geological horizons. Consequently the correlation of the KnoxviUe with the Franciscan series (Golden Gate series) by this means is out of the question. The sinter is very similar all through the Coast Ranges, and any deposit is a possible source of quicksilver. Professor Lawson's discussion of the structure of the older uncrystalline rocks shows, although he minimizes the importance of the disturbances which they have undergone, that it is often difficult of elucidation. The earlier observers have all remarked upon the difficulties connected with a study of the so-called metamorphic rocks, and we cannot admit that all of their work is "superficial." While the author is probably right in asserting that the larger structural features are comparatively simple in the area under discussion, yet it seems to me that there is good reason for believing that the structure is very complex in detail. This is shown by the fracturing and folding of the jasper bands ; the frequent occurrence of crushed shale, and thin-bedded sand- stone in a ruptured condition ; and the presence of innumerable cracks and shear planes in the more massively bedded sand- stones. Local areas occur, it is true, totally free from the effects of strain, but they do not dominate. A comparison of the sand- stone at Point San Pedro with that north of the valley of the same name affords a good illustration. Both in the cliffs and on top of the point the sandstone weathers out in large blocks in a manner closely simulating the sandstone in the Cretaceous and older Tertiary in the Coast Ranges. On the contrary the similar sandstone in the hills north of San Pedro Valley breaks up in angular fragments ; being permeated in many places with veins of quartz and calcite, or linear seamlike cavities. This fracturing and veining is also almost everywhere to be noted in the jaspers and limestone. Professor Lawson considers that because the number of parting planes decreases downward from the surface they are due in great part to atmospheric agencies alone, although he recognizes some shear planes. I believe, however, that we have good reason for holding that a large part of these planes which separate the sandstone into 70 HAROLD W. FAIRBANKS angular fragments are not due to atmospheric agencies alone, but that they are capable of being produced in thick-bedded rocks as well as in thin-bedded which have not been greatly sheared, but subjected to a rending strain. Such rocks under proper conditions can become recemented and apparently as massive as before. The cracks may thus become completely closed, or left slightly open and filled with calcite or quartz. Subjected to atmospheric agencies lines of weakness are soon developed, and the rock crumbles in angular pieces. A similar rock which has not undergone this breaking strain will weather either in large rounded knobs, or, if it be soft and argillaceous, break up into fragments more or less conchoidal. Professor Lawson remarks further upon this subject as fol- lows : "The superficial study of this phenomenon has led to grossly exaggerated views as to the amount of disturbance (shattering) to which the Coast Ranges have been subjected. The sharply marked alternation of wet and dry seasons, com- bined with the treeless character of many of the ranges, is peculiarly favorable t) this disintegration." It is only necessary to compare the most of the sandstones of the Golden Gate series with those of the Cretaceous or Ter- tiary to see the vast difference in the manner of weathering. As a result of my experience through nearly the whole of the Coast Ranges, I feel satisfied that peculiarities of climate have not been the cause of the phenomenon under discussion. The abun- dance of cracks and shear planes in the older rocks is one of the chief reasons why it is so difificult to obtain good building stone from them, although the sandstones are characteristically thick bedded. I think that Professor Lawson has underestimated the amount of disturbance to which the Golden Gate series has been subjected. It is not necessary for the folds to be involved or intricate, although they certainly are in many places, for it to have undergone a large amount of strain, fracturing, and shear- ing. Professor Lawson professes to be entirely ignorant of the GEOLOGY OF SAN FRANCISCO PENINSULA 7 1 "erogenic movements which effected the deformation and fault- ing of the Franciscan series" as well as of the relative sequence of these disturbances and the peridotitic intrusions. This is, of course, excusable in one studying only a narrow field where a part of the record is wanting, but too much is at present known of the wider field of Coast Range geology for one to plead ignorance on these questions. The initial disturbance of the Golden Gate series, with the exception of that produced by the contemporaneous intrusions and flows, dates from the post-Jurassic upheaval. Leaving out of account the serpentine, there is evidence that many of the eruptive bodies associated with this series were subsequently formed. The intrusive nature of the diabase at Hunters point is recognized by Professor Lawson, and he is certainly correct. To the presence of these eruptives, I believe, is to be attributed the extreme disturbance in local areas. As to the age of the Franciscan series^ of Professor Lawson nothing more definite is advanced than the evidence of a few imperfect fossils, which cannot be determined specifically and in many cases not even generically. These fossils are supposed to favor the old view of a Cretaceous age. He says : " Evidence, such as it is, is confirmatory of the opinions of Whitney and Becker The series as a whole is very probably older than the Knoxville Aucella horizon of California. The writer has no doubt upon this point." Professor Lawson gives no reason for assuming the series to be older than the Knoxville, and since Mr. Stanton's^ work places the Knoxville Aucella horizon at the base of the Creta- ceous, it is difficult to understand how these pre-Knoxville rocks can be included in the Cretaceous. 'This designation (American Geologist, June 1895) embraces the same aggregate of strata to which I have given the name Golden Gate series (Journal of Geology, May-June 1895) from its characteristic exposures at the entrance to San Francisco Bay. As to which name shall finally be accepted I am, for my part, willing to rest the case, although another claim might with great justice be added, on the truth or falsity of my published statements as to its age and stratigraphic position. = Bull., No. 133, U. S. Geol. Sur. 72 HAROLD IV. FAIRBANKS Again Professor Lawson argues for the post-Jurassic age of the granites in the Coast Ranges, and in order to account for the fact that the pre-Knoxville rocks rest on the granite with a basal conglomerate, he is obliged to assume that they are a part of the Cretaceous. The correctness of one view argues much for the other ; if one falls both must be considered invalid. Again he says: "Fairbanks has combated the views of Whitney and Becker, and has pronounced the series to be of pre-Cretaceous age. He has not yet, in the writer's opinion, established the correctness of his contention." I wish to call especial attention to this statement because of the fact that Pro- fessor Lawson has not advanced one particle of evidence in refutation of my published statements as to the existence of a nonconformity between the Knoxville and these lower beds. 'It is hardly probable that we shall soon find any fossils in good enough condition to be decisive concerning the question at issue, and the main dependence must be placed upon stratigraphy. The fact that these rocks (Golden Gate series) lie unconform- ably beneath the Knoxville is shown by the most strongly marked contrast, structurally as well as lithologically, in addition to the stratigraphic break which I have described in previous publications. In closing his description of the stratigraphy and structure of the Franciscan series. Professor Lawson says in regard to the disturbance of the strata : "The most of it seems to have long ante- dated the uptilting of the Montara fault block, so that the latter differs from most tilted blocks with which we are familiar. These are commonly tilted blocks of strata previously undisturbed, and the tilting is recognized by the attitude of the strata and the presence of fault scarps. In the present case, however, the region has been moderately folded and profoundly faulted with local sharp plication." As far as I am conversant with the geology of California and adjacent parts of Nevada, the fault blocks are commonly not tilted blocks of previously undisturbed strata, but exactly the opposite ; that is, the strata have been more or less highly folded and tilted previous to the faulting. GEOLOGY OF SAX FRANCISCO PEXIXSULA 73 much of which now recognizable, is of quite recent date, geo- logically speaking. In the last sentence quoted Professor Lawson seems to recog- nize quite fully the complicated structure of the Franciscan series. The serpentine in the area covered by Professor Lawson's report is divided into three linear tracts. One extends across the city of San Francisco from Fort Point to Hunters Point, a distance of ten miles, with a width of one and one-half miles. Another large tract extends from San Andreas Lake to San Mateo Creek, having a length of eleven and one-half miles and a maximum width of one mile. According to the description these bodies occur as laccolites, laccolitic sills, or dikes. From a careful perusal of the report, and an examination of several of the more important areas covered by these eruptives, I cannot understand the reason for applying to them the designation laccolite. This term was originall}' given by Gilbert to an eruptive mass which in the course of being forced upward, instead of reaching the surface, finally spread out between the strata, forming a thick lens, and arching them over it in dome form. Professor Lawson is certainly right in the statement that the main bodies are not true dikes, but it seems to me that the term sheet or sill which he uses quite frequently is the really proper one for these eruptives. If the term laccolite has any exact meaning it is certainly not synony- mous with sheet or sill. I have carefully examined the so-called laccolites of Hunters Point, Potrero and Fort Point and can find no evidence that these eruptive masses were ever covered by an arched roof. In my opinion the field relations indicate that they cooled as sheets or dikes. They have no appearance of being lens shaped, and if it is true, as Professor Lawson supposes, that the}' are all con- nected through the distance of ten miles, we have then a long sheet inclined at an angle of 35° to 40°; and though in a general way intruded along the bedding planes of the sandstone and shale, yet, owing to the marked irregularities of the surface of the erup- 74 HAROLD W. FAIRBANKS tive sheet, the inclosing strata have been much disturbed and show locally a marked variation in strike and dip. The remnant of a supposed laccolite roof on Hunters Point I would interpret to be a body of jasper and sandstone or shale inclosed in the serpentine, as a portion of it occupies a sag between higher serpentine ridges, and apparently extends down into a small ravine. It is fully as difficult to believe that the Fort Point occurrence is a laccolite. The stratum of shale, sandstone, and occasional bodies of jasper, which appears in the cliffs and along the beach south of the fort, dips into the cliffs at an average angle of not less than 30°, and it seems to me that it can be nothing else than an inclusion between two sheets of serpentine. The exposures at the Potrero are good and bear out the opinion which I have already expressed. That the two dominant ridges, Montara Mountain and San Bruno Mountains, are fault blocks of such importance, or their diastrophic history so clear as Professor Lawson outlines it, does not seem evident to me. The San Bruno Mountains are said to be the older block, but the southern slope facing the supposed fault is remarkably bold and steep, not showing an advanced stage of degradation, and in addition the northern slope is almost as abrupt. The supposed fault of 7000 feet appears almost incred- ible, and it does not seem at all necessary to postulate it in order to account for the position of the Merced series. That this series should once have existed over the whole of the northern end of the San Francisco peninsula appears very problematical at least. That a series over a mile in thickness, and of so late an age, should have been so completely removed as not to leave a trace north of the supposed fault it is not easy to believe. I cannot find any evidence either of the supposed great fault on the southern slope of Montara Mountain. The topography as a whole does not support the idea, and that the ridge is a simple tilted block is not supported by the evidence which Professor Lawson adduces of important faults on the northern slope. Pro- fessor Lawson's suggestion in another place that Montara Moun- GEOLOGY OF SAN FRANCISCO PENINSULA 75 tain has been forced up through the strata in the manner of a tele- scopic thrust appears to me to have an important element of truth in it. This thrust probably resulting from lateral compression pro- duced the elevation of the granite, partly, at least, through shearing. It would appear that the evidence in favor of the successive appearance of the "two dominant fault blocks" is exceedingly slight. While the San Bruno Mountains are considered to have been above the sea level and the Pliocene undergoing erosion, there is believed to have been no such subaerial period for Mon- tara Mountain until the final uplift resulting in the terrace deposits. This is opposed to the observation of Mr. Ashley' as well as to my own views concerning the general post-Pliocene elevation of the coast. By this elevation I do not mean that to which the terraces are due, but an earlier one resulting from the same influences which deformed the Pliocene sediments the whole length of the California coast. In regard to the age of the granite Professor Lawson ^ says : "The simplest and most natural hypothesis that suggests itself is that the granite corresponds in age with that of the Sierra Nevada, and this hypothesis has not yet been exhausted of its strong probability of truth. The granites of the Sierra Nevada, in so far as their age is known, are clearly post-Jurassic. Gran- ites of about this age are extensively developed along the west coast of North America from Alaskasouthward. The granites of the southern and northern Coast Ranges seem to be geologically continuous with those of the Sierra Nevada. The fact that the Sierra are separated from the Coast Ranges by the valley of Cal- ifornia is immaterial to the discussion, since the latter is clearly a delta-filled geosyncline of late Tertiary or post-Tertiary origin. There is therefore a strong presumption in favor of the view that the granites of the Coast Ranges and those of the Sierra Nevada are of common origin and common history. This pre- sumption must be steadily kept in view till it is negatived by positive evidence." 'Neocene Stratigraphy of the Santa Cruz Mountains, p. 334. = Am. Geol., June 1895. 76 HAROLD W. FAIRBANKS It does not seem to me that there is any validity whatever in the above reasoning, and in the light of the true position of the Golden Gate series there is a strong presumption in favor of an opinion exactly opposite to that just quoted. In fact the pre- sumption is so strong that it amounts almost to a certainty that the granite in the Coast Ranges is older than the main body of the granite in the Sierras. What we at present know of the position of the Golden Gate series points to the fact that its first upheaval was contemporaneous with the last great upheaval recorded in the rocks of both the Sierras and Klamath Moun- tains. The Mariposa beds involved in this upheaval in the Sierras are held to be Upper Jurassic. The Knoxville beds deposited after this upheaval are believed on the best authority to be Lower Cretaceous ; and if we make the granite in the Coast Ranges the same age as that in the Sierras, we must crowd into the break between the Mariposa beds and the Knoxville, a series of beds thousands of feet in thickness and separated from the Knoxville by a break as profound as that between the Knoxville and the Jurassic. There is every reason for assuming that the granitic rocks of California are not all of the same age. Granitic bowlders occur in the Mariposa beds south of Colfax, a fact pointing to a pre- Jurassic granite body in that region. The youngest fossiliferous rocks associated with granite in southern California are probably of Carboniferous age, and while the extension of the crystalline basement rocks of that region northwestward into the Coast Ranges is not likely to be much younger than the Carboniferous, for all that we know at present it may be much older. Harold W. Fairbanks. Berkeley, California. Editorial. The success of the recent meeting of the Geological Society of America was undoubtedly due to the fact that it was held in Washington. No other city in the country offers so many attractions to geologists in the winter time as the national capital. Containing as it does the largest body of geological investigators to be found in any one place in the world, it has become a center of geological activity and the repository of many valuable collections. Located within easy reach of the universities of the East and South and of the Middle West it has become a favorite rendezvous for geologists scattered through- out these parts of the country. For these reasons the sugges- tion made by Mr. Walcott, Director of the United States Geological Survey, that the society hold all its winter meetings in Washington and its summer meetings elsewhere, is an excellent one, and was heartily endorsed by the retiring president. Pro- fessor Le Conte. Since the present method of entertaining the society would prove a burden upon the Washington members if repeated annually, it would be proper to allow the visiting members to share the expense by dividing W\q co?,l per capita. It is to be hoped that the proposition to meet annually in Washington be considered seriously by the society. The necessity of more expeditious methods of conducting the presentation of papers and the need of a livelier sense of obligation to fellow-members on the part of those misusing the time of the meeting were apparent. It ought not to require lono- reflection to convince one that beyond a reasonable use of time the tax on the patience of an audience defeats the object of an address, and is more than a waste of energy. It is to be noted that the present unsatisfactory process of 78 EDITORIAL conducting the annual election resulted in the only regrettable incident connected with this otherwise highly successful meeting. The present method should be changed speedily. It has been suggested that a process should be adopted b}^ which the secre- tary should send to every member blanks for nominations for each officer ; each member to nominate one person for each office. Upon receiving these nominations the secretary should ascertain those having the two highest numbers of nominating votes, and mail to each member a ballot placing these in nomi- nation, and requesting a vote. The extent of the interest taken in the meeting is shown by the number of papers presented, which reached forty-nine, and by the fact that discussions were quite generally participated in. The presence of ladies at the banquet and the efforts of the committee on entertainment and those of the toastjnaster. Pro- fessor Emerson, combined to render this feature of the meeting most enjoyable. J. P. I. *** In the October-November number of the Journal of Geology (p. 8 1 1 ) Mr. J. B. Tyrrell's paper on the " Genesis of Lake Agassiz " includes incidental reference to the application of the name "Laurentide Glacier," upon which it may be useful to make some comment. When it had been shown by the writer that the Cordilleran ice-sheet of the West was self-contained, it became evident that the great eastern division of what had previously been referred to as a "continental glacier" required some distinctive appellation. The name Laurentide glacier was then proposed in the following terms : "Recognizing, however, the essential separateness of the western and eastern confluent ice-masses, and the fact that it is no longer appropriate to designate one of these the 'Continental glacier' the writer ventures to propose that the eastern mer de glace may appropri- ately be named the Laurentide glacier while its western fellow is known as the Cordilleran glacier y^- Thus it appears that the ^American Geologist, Sept. 1890. See also Trans. Royal Soc, Canada, Vol. VIII, Sec. IV, p. 56. EDITORIAL 79 term Laurentide glacier was intended to designate an ice-sheet developed upon the entire Laurentian })lateau or protaxis — the Canadian shield of Suess — which extends itself, as a relatively elevated tract of U-shaped form around the depression of Hudson Bay, from Labrador on the east to the Arctic Ocean near the mouth of the Mackenzie on the west. Mr. Tyrrell's investigations relate particularly to that arm of the protaxis which lies to the west of Hudson Bay, and upon the northern part of this he has been able to define a center of dispersion of glacier-ice to which he has applied the name " Keewatin glacier." He suggests that the name Laurentide glacier may now be relegated to the similar center of radiation of ice shown b}- Mr. A. P. Low to have existed on the Labrador peninsula, and supposes that ice from the latter gathering-ground (at a date subsequent to that of the greatest spread of the Kee- watin glacier) crossed the southern part of Hudson's Ba}- to the Winnipeg basin as well as to that part of the continent in the vicinity of the Great Lakes. It is, however, tacith' assumed that no general occupation of the Laurentian plateau by glacier- ice occurred at a period antecedent to that in which the Kee- watin glacier became defined; although, to the writer, it seems probable that at one time this plateau or axis was thus covered by ice in all its length and that this ice moved down from higher to lower levels in conformity with the normal slope in all direc- tions. A second assumption (to which, however, several writers on this subject besides Mr. Tyrrell have given credence) is that implied in the suggested passage of glacier-ice originating in the Labrador region across the watershed to the north of the Great Lakes. The actual evidence on this point appears to the writer to be slight and inconclusive in its nature. Our knowledge of the rocks of the great region involved is, it is believed, insuffi- cient to prove that any erratics from the basin of Hudson Bay have actually crossed this watershed to the south or southwest. Nor can the observed directions of striation be taken as definitely indicating ice-movement in a southerlv sense in all 8o EDITORIAL cases, for at the time of observation of many of these, it cannot be doubted that a southerly rather than a northerly motion was naturally assumed, in consequence of hypotheses then unques- tioned, and without the critical examination called for in view of later discoveries. So great an extension of an ice-sheet origi- nating on the Labrador peninsula, is all the more doubtful in the light of Mr. Chalmer's investigations, which show that no glacier- ice from the northward crossed the highlands to the south of the St. Lawrence below Quebec. These highlands developed in New Brunswick and northern Maine, an independent center of dispersion, which Mr. Chalmers suggests may be referred to as the Appalachian system of glaciers or the Appalachian glacier.^ If, therefore, we may admit the existence of a Laurentide glacier of early date and such as to envelop the whole Laurentian plateau, the Keewatin glacier of Mr. Tyrrell, with that which may be named the Labradorian glacier,^ as known to us by Mr. Low's work, may be regarded as relatively local ( although still very important) centers of dispersion connected with a dimin- ishing stage of the glacial period. In this case the overriding of the area at one time covered by the southward extension of the Keewatin glacier by ice from the eastward ( as demonstrated by Mr. Tyrrell ) may be attributed merely to a still later reex- tension, due to climatic changes, of that part of the Lauren- tide glacier situated to the east of Lake Winnipeg. If, on the other hand, it should ultimately be shown that no continuous ice-sheet ever covered the entire length of the Laurentian pla- teau, the several great glacier masses actually known to have been developed upon these highlands may still be appropriately considered as constituting together a Laurentide group, as distinguished from the great western Cordilleran ice-mass. In the case of the Keewatin and in that of the Labradorean glacier, it has been shown by Messrs. Tyrrell and Low that the centers of dispersion changed, giving rise to superposed series of striations, and it appears probable that in the study of this ' American Geologist, Vol. VI (1890), p. 325. - This has, I believe, already been proposed by Mr. F. B. Taylor. EDITORIAL 8 1 migration of gathering-grounds of glacier-ice on the Laurentian highlands, the explanation of many otherwise anomalous circum- stances will yet be found. Any inquiry into the causes of such migration opens a wide field of discussion, including particularly meteorological and possibly changed orographical conditions. Other things being equal, it is obvious that the places in which the accumulation of glacier-ice began would also in all proba- bility be those where the last centers of its dispersion continued longest and ultimately failed. It is very important to trace the migration of these centers from which the ice flowed, and if possible to ascertain the conditions which, in all probability, may have interfered with the coincidence of the initial radiant areas and those marking the close of the epoch of glaciation. G. M. D. *** It would appear from the papers of Professors Tarr and Barton read at the recent meeting of the Geological Society at Washington, abstracts of which appear in this number of the Journal, that Professor Salisbury and myself must have invited misinterpretation of our views respecting the former extension of the inland ice-sheet of Greenland by infelicities of expression not altogether evident to us. What we have thus far published has consisted, in the main, of brief statements intended rather to show the bearing of the topography of the coast upon the larger questions of glacial prevalence than to indicate the precise local extension of the ice. In the series of papers entitled " Glacial Studies in Greenland" (not yet complete, having been interrupted to give earlier place to contributions awaiting pub- lication), the specific subject of former glacial extension has not yet been reached. When it shall be, however, we shall not wish to be understood as attempting to map the precise limit of former glaciation on the thousand miles of borderland along which we coasted, but merely as endeavoring to indicate its general limita- tions. We believe, neverthless, that our observations and infer- ences are thoroughly trustworthy in the general sense in which thev are intended to be received, and that thcv are decisive in 82 EDITORIAL their bearing upon two general hypotheses of wide interest ; the one, that the ice-cap of Greenland formerly stretched across Baffins Bay and Davis Strait and became the source of the main- land ice-sheet ; the other, that, in a former supposed state of elevation, the ice-cap pushed out into the heart of Baffins Bay or into the Atlantic until it reached adequate conditions of wast- age by flotation or by low-level extension. In either case, the borderland of Greenland must, presumably, have been effect- ively glaciated for a long period, and should manifest this in the subjugation of its topographic contours. In its bearings upon these general problems, an advance of a few miles, more or less, an ineffectual overtopping of a few heights, more or less, are relatively inconsequential. Our language is to be interpreted in the light of the major questions whose solution we sought. There does not appear, however, to be any essential lack of harmony between the data of Professors Barton and Tarr and the interpretations of Professor Salisbury and myself, unless it be in relatively unimportant details. Professor Salisbury makes the following statement relative to the region (this Journal, Vol. III., pp. 876-7): On the whole, judging from topography alone, it seemed more probable that the coast from about latitude 70° north to the end of the Nugsuak penin- sula [the southern one] had not been recently smothered in ice, though it is well possible that the ice-cap may have once extended beyond its present limits, and that isolated glaciers occupied the valleys leading down to the sea. The northwest end of the peninsula bears the marks of the passage of ice over a considerable part of the coastal front. North of Nugsuak peninsula, and from that point to the south side of Melville Bay, the topography of the coast, so far as seen, indicated general, though not universal, glaciation. Thus the southwestern end of Svarten Huk peninsula (71° 30') has a topog- raphy denoting the absence of glaciation. This embraces both of the districts in question. Relative to the region of Professor Barton's studies, I made the following (unpublished) note on July 18, 1894, at the con- clusion of detailed notes on Hare island, the Nugsuak peninsula, (the one south of Umanak fiord), Ubekyendt and Upernivik islands, Svarten Huk peninsula, and Disco Island : EDITORIAL 83 I infer that the inland ice once pushed out into Baffins Bay through the Waigat and through Umanak fiord and overlapped the adjacent lands, but did not overtop the highest parts of Disco, Upernivik and Ubekyendt islands and Svarten Huk peninsula. The ice border rose 2000 feet above the present sea surface, but probably not 3000 feel on the coastal line, quite certainly not 4000 or 5000 feet. I have traced on the chart a theoretical outline of the farthest ice. This of course does not include local ice on the uncovered land. It, on the contrary, presumes it. This hypothetical line starts with the northwestern part of Disco Island and swings outside of Hare island and some distance off the extremity of Nugsuak peninsula, touches the outer side of Ubekyendt Island, and thence connects with the southern portion of Svarten Huk peninsula. The outer curve of this line is sixty miles azvay front the present border of the i?ila7td ice and embraces the entire territory of Professor Barton's special studies, as I understand them. I have no notes on the northern Nugsuak peninsula which was the special field of Professor Tarr's studies, the Falcon hav- ing turned away from the coast somewhat south of it to attempt the "middle passage" of Melville Bay, and, failing in this, returned to the vicinity of the coast a little north of the peninsula. On the region next south I made the following note on July 19 : With the exception of Sanderson's Hope and a few other prominences south of it, the whole region shows rounded contours, so far as seen. The small islands are well-rounded domes of rock that appear entirely bare. I could not detect any bowlders or other debris upon them. They are simple roches moutontices, but not strikingly typical as such. The reduction has not completely subordinated them to glacial types. I have no doubt the ice pushed out well into the bay here. Some of the higher peaks may have remained as nunataks. These small islands are rather farther away from the edge of the inland ice than the similar ones described by Professor Tarr. It appears, therefore, that we explicity recognized predom- inant glaciation reaching out from 30 to 60 miles, and trespass- ing on the borders of Baffins Bay, and that we excepted only some of the higher points. It is only by demonstrating that these were submerged by the inland ice that any notabl}^ greater 84 EDITORIAL extension can be established, and the most of these points were not visited by the members of the recent expedition, nor were comparable heights of like situation near the coast line studied. Sanderson's Hope is charted as 3467 feet; a point on Svarten Huk peninsula, as 5230; a point on Disco Island as 51 10, and many other points range from 3000 feet upwards. At most, therefore, the differences between us are merely matters of minor detail so far as observational determinations are con- cerned. This advance of 30 to 60 miles seemed to one party to call for expressions of amplitude, to the other, instinctively making comparison with the many hundreds of miles of ice invasion of the mainland, to call for diminutives. Here is, indeed, a wide psychological difference, and we can contribute nothing to mini- mize this difference. Under the widest permissible interpreta- tion of the facts as seen by either party the advance seems to us emphatically small, and very significant in its smallness. The reliefs of the region seemed to the earlier party to belong to the semi-subdued type. The later party appear to have supposed them to represent the imsiibdued type of the ear- lier party. The determinations of the later party are, however, in close accord with the classification of the earlier party. It is possible that Professors Tarr and Barton unconsciously transferred to this region our rather emphatic descriptions of certain markedly serrate tracts farther south, and of certain ragged islands lying to the north, e.g., Dalrymple Island (figured in this Journal, Vol. II, p. 661), and Cone Island (Fig. i.. Vol. III., p. 772). As indicated above, we did not class the topog- raphy of the region in question under the markedly ragged and serrate type but under an intermediate one of partial sub- jugation. The figures referred to illustrate our standard of the former class. The courteous suggestion that the differences of interpretation were due to our point of view on the leeside of the prominences is inapplicable to the important case of Disco Island, for Professor Salisbury passed on the stoss side. It is only measurably applicable to the rest of the territory, as the EDITORIAL 85 oblique views of approach and retreat gave traversing lines of vision that partially reached the stoss contours. This rather extended note has its purpose in a desire to make clear the narrow limits of such differences of observation as may exist and to show that the two main questions of general interest are essentially unaffected by them. Glacialists may feel sure that the conclusions derived from the topographic study of a thousand miles of the coast, even though cursory, and limited by conditions, will be found decisive on the main issues. T. C. C. Reviews. Manual of Determinative Mineralogy , zvith an hitroduction o?i Bloiv- pipe Analysis. By George J. Brush. Revised and enlarged by Samuel L. Penfield. New York : John Wiley & Sons, 1896. Since its first appearance in 1874, ^xnsW?, Defennijiative Mineralogy and Blowpipe Analysis has filled the demand for a text-book in this science almost to the exclusion of others. The tables, especially, proved to furnish an exceptionally ready and accurate means of iden- tifying mineral species. The use of the book has of late years, how- ever, been confined largely to the tables,, since its opening chapters, while valuable for reference, were hardly adapted for class instruction, if, indeed, they were intended to be so used. This lack in the book Professor Penfield has most satisfactorily filled in the new form in which he now presents it. In its present form it furnishes not only a complete text-book for class use, but also a work which should enable even the novice without further aid to gain a practical knowledge of determinative mineralogy. Opening with a statement of the chemical principles especially applicable to mineralogy, for the author remarks that mineralogy is chiefly a chemical science, the next chapter treats, with excellent illus- trations, of the apparatus and reagents to be employed and of the nature and uses of flames. Several new experiments incorporated here add much to the value and interest of the chapter. In the following chapter the reactions of the elements, arranged in alphabetical order, are given with great fullness, while carefully described experiments show how many of the typical reactions can be obtained. Every detail is here given with a care and precision that might at first sight seem superfluous, but the necessity for each. Professor Penfield, it may be believed, has fully proved in his long experience as a teacher of the subject. The student will at least be impressed with the need of care and accuracy in making the tests and will not be led to 86 REVIEWS 87 expect satisfactory results if they are not so performed. If there be any criticism to be made of this portion of the work it is to wish that more mention had been made of possible interfering elements, for the field mineralogist rarely has cabinet specimens to deal with. In the next chapter the blowpipe and chemical reactions are tabu- lated in a form which will be recognized as a distinct improvement on those of the earlier book. The tables are models of clearness and com- pleteness. With an introduction to the use of the tables, the revised portion of the work closes, but promise is made of an early revision of the tables, including the introduction of new species and many changes and corrections in the chemical formulje. To express the hope, which will be general, that this promise may soon be fulfilled, is to bestow the highest praise on the work which has been already done. O. C. Farrington. TJie Dinosaurs of North America. By Othniel Charles Marsh. Sixteenth Annual Report of the United States Geological Survey, Part I, p]>. 133-414. Under this title Professor Marsh has published, in an article of loi pages of text, a resume of his long series of papers on the Dinosauria of North America. Intended as it is for the general reader, there are few new facts brought out, and the specialist finds little of value to himself beyond the convenience of a condensed statement of Professor Marsh's views, and the large collection of illustrations which form so important a part of the paper. There are eighty-four plates and sixtv- six figures in the text. Most of the plates are familiar to the readers of the American Journal of Science, as they are enlarged copies of the ones accompanying the author's papers in that journal. The author's aim is " to give the general reader a clear idea of some of the type specimens of one great group of extinct animals that were long the dominant forms of life in this continent." Of especial interest to such a class must be the restorations attempted by Professor Marsh, and it is well to issue a warning against receiving as entirely accurate restorations which, made in many cases from incomplete material, can only represent the author's idea of the most probable form of the skeleton. In the same spirit of warning we must call attention to the statement that "the best authorities regard them (the Dinosauria) as constituting a distinct subclass of Reptilia;" indeed, it should be 88 REVIEWS remembered that some of the "best authorities" are among those who assert "that these reptiles do not form a natural group, but belong to divisions remotely connected and not derived from a common stock." It is, to say the least, a very open question whether the groups defined as the orders Theropoda, Saiiropoda, and Predentata are not distinct and unrelated. The Dinosaurs are discussed in the order of their appearance in time. The Triassic Dinosaurs come under Part I and are followed by the Jurassic and Cretaceous forms in Parts II and III. These parts are confined to descriptions of the anatomical characters of the various forms and their distributions. The author discusses briefly in this connection the European forms and their relations to the North American forms. Part IV is taken up with a general discussion of the relationship of the Dinosauria to the Aetosauria, Crocodilia, Belodontia, and Aves, to all of which groups he finds many points of similarity. Part V is devoted entirely to a classification of the group which he regards as a distinct subclass Dinosauria, with three orders Theropoda, Sauropoda, and Predentata. The order Theropoda contains ten fami- lies and the four suborders Cceluria, Compsognatha, Ceratosauria, and Hallopoda. The order Sauropoda contains six families and no subor- ders. The order Predentata contains eleven families and the suborders Stegosauria, Ceratopsia, and Ornithopoda. Professor Marsh has retained the opinions which he has expressed in his former publications on the same subject, many of which have been criticised and controverted in current scientific literature, so that, while the work will find its chief, and a great value, to the reader who has not access to scientific periodicals, he must not lose sight of the fact that the taxonomic position of the animals comprising the hetero- geneous group called Dinosaurs is still an unsettled question. E. C. C. Abstracts, PAPERS READ AT THE WASHINGTON MEETING OF THE GEOLOGICAL SOCIETY OF AMERICA. Glacial Observations in tJic Ihnatmk District, Greoiland. By Geo. H. Barton. A part}' consisting of six persons under the direction of Professor Alfred E. Burton, of the Massachusetts Institute of Technology, was landed by the Sixth Peary Expedition at the village of Umanak, Lat. 70° 30', on an island in Umanak fiord, August 5, 1896. On Septem- ber 9 the party was again taken on board for the return home. Between these dates observations were carried on upon the marginal area of the continental ice-cap to a distance of fifteen miles inland, and upon the glacial tongues passing down from the ice-cap by the various valleys leading into the Karajak and Itivdliarsuk fiords. The region of the Umanak District was selected for investigation because of the facilities offered for much to be seen in the short time at the disposal of the party. Easy access to the edge of the inland ice by means of the larger fiords, the many glaciers large and small descending near or into the waters of the fiords, the great number of icebergs constantly passing out into the open waters of Baffin's Bay, and finally the fact that this region had not been visited by Ameri- cans, caused the selection. Access to the surface of the inland ice was obtained from the nunatak lying between the Greater and Smaller Karajak glaciers. Considerable difficulty was encountered in getting upon the surface owing to the precipitous, generally vertical, sometimes overhanging character of the edge, rising from ten to forty feet. In only three places for a distance of over twenty miles did the edge present an inclined surface sufficiently gentle in slope to allow of ascent or descent. In attempting to pass inland upon the ice the crevassing of the Greater Karajak glacier was found to extend far backward so that our course had to be deflected for several miles and finally to pass along the neutral ground between the two Karajaks. The elevation of the 89 90 ABSTRACTS ice surface where first reached was about 1500 feet above sea level, while our fartherest point inland was about 2950 feet in elevation. For a distance of about two miles the surface of the ice is very- rough, consisting of a series of billows, very much like a frozen short, choppy sea. Over this it was very difficult to drag the sledge which constantly upset. This area was also full of dust holes from the size of a pencil to three feet in diameter and an average depth of about two feet. These were filled nearly to the surface with water and had a layer of dust in the bottom, an inch'or two in thickness. Beyond this we passed through a crevassed area for two or three miles and then came to a broadly rolling, undulating area which stretched away as far as the eye could reach looking like the Ameri- can prairies in winter. The surface here was hard and crisp so that walking was very easy. One lake was seen with several small streams flowing into it and one large stream flowing from it toward the land. Small streams were constantly encountered except in the crevassed area all flowing toward the land or toward the lake. One large stream, twenty to thirty feet across, and twenty to twenty-five feet deep, including five feet of water, was found at the farthest point reached flowing directly away from the land toward the interior. No detritus was found on the inland ice except the dust in the dust hole area. Outside of this area the water of the streams was clear and their channels showed only clear ice. The two important glaciers studied were the Greater Karajak and the Itivdliarsuk glaciers. The first of these has a width of about five miles and a length of from ten to fifteen miles from the edge of the inland ice to the waters of the fiord at its frontal face, but the crevas- sing of its current passes backward several miles into the mass of the inland ice. The elevation of the surface of the glacier as it first leaves the margin of the inland ice is about 1500 feet above sea level while directly at its frontal face it is about 500 feet. This gives a gradient of about 1000 feet in a distance of about ten miles, or 100 feet to the mile. The rate of motion was carefully studied by Professor Burton dur- ing a period of nearly thirteen days. The result showed that there was an eddy in the ice at the point where the obse^rvations were made, as the ice near shore. was moving slightly upstream. At a distance of 1708 feet it was stationary, while at a distance of 3396 feet the motion amounted during this time to 30.20 feet or 2)/}, feet per day. Up to this distance stakes had been set in the ice. At a distance of 2.4 miles a ABSTRACTS 9 1 peculiarly shaped pinnacle furnished a point of observation. This was found to move during the same time a distance of 236 feet or about 19 feet per day. This was the only case in which the actual rate of motion was determined. The Itlvdliarsuk, three miles in width, was crevassed completely to the shore margin so that it could not be traversed with any safety, and no attempt was made to determine the rate of the smaller ones. Neither of these large glaciers carry any detritus either superglacial or englacial, except in the case of the Itivdliarsuk. Small nunataks near its source furnish two medial moraines which can be traced nearly throughout its length on the surface but are largely incorporated as englacial material before reaching the terminus. The edge of the larger glaciers like that of the inland ice is almost always nearly vertical with a height of from ten to forty. The struc- ture is distinctly banded usually, apparently due to a shearing of the upper portions over the under from the central portions of the stream shoreward, the layers being inclined away from the shore. Lateral streams flowing along the margin often cut caverns into the ice by means of which opportunity is furnished to study subglacial action. Small bowlders caught between larger ones or between a larger one and a projecting ledge are often broken into small frag- ments or ground to powder. In the case of rapid motion of the ice, distinct vacant spaces are left between the lee of a boss of rock and the ice, but where the motion is slow the ice is kept pressed against the lee side of the boss taking on a fanlike structure as it is pressed downward. Lateral moraines along the sides of the glaciers are always present. They consist of a mixture of angular, subangular and rounded frag- ments, usually with considerable sand and a little clay, and occasionally largely made up almost entirely of till- like material. From the edge of the inland ice and also from the detached ice- caps of the Nugsuak peninsula a large number of smaller glaciers descend the more or less narrow gorges, some reaching sea level and discharging icebergs, but the majority terminating from 100 to 1000 feet above the sea. The gradient of descent varies strongly in the various ones. In some it is very steep, in some very gentle, while many have a great change in various portions of their own length. With two exceptions all the glaciers show evidence of diminu- tion in len;j^th and depth, in older terminal moraines and glaciated surfaces farther down the valleys and older lateral moraines higher up the sides. 92 ABSTRACTS The two exceptions, near Sermiarsut on the Nugsuak peninsula seem to be overriding terminal moraines of not great age but these were seen toward the last of my work and were not carefully studied. Everywhere throughout the regions visited there is evidence of the former greater extension of the inland ice, it having covered all the highest peaks, passed out over the ends of the promontories, filling the fiords and passing into the waters of Baffin's Bay/ The peaks which present sharp serrated edges toward Baffin's Bay are distinctly rounded on the sides toward the inland ice; the highest points visited show abundant roche moutonnee forms, and everywhere erratics are freely scattered. The Origin and Relations of the Grenville and Hastings Series in the Canadian Laurentian. By Frank D. Adams and Alfred E. Barlow, with observations by R. W. Ells."^ This paper may be regarded as a continuation of two former papers ^ by Adams, treating of the Laurentian of Canada and presenting the results of further work. It deals more especially with the probable origin of the Grenville series in the light of recent studies by the authors of a veryjarge Laurentian area in central Ontario, along the margin of the protaxis and north of Lake Ontario. The northwestern portion of the district in question is underlain by the Fundamental Gneiss, which is believed to form part of the original crust of the earth and to be of igneous origin. The south- eastern portion is occupied chiefly by the thinly bedded calcareous rock of the Hastings series, a series of undoubtedly sedimentary ori- gin, but of unknown age, separated, however by a long erosion inter- val from the overlying Cambro-Silurian, and having certain petro- graphical resemblances to the Huronian of the typical area north of Lake Huron. Between these two series of rocks is an irregular belt of the Grenville series, identical in all respects with that of the original Grenville area in the province of Quebec. The character of these several series is described and their resemblances and points of differ- ence indicated. 'See paper by writer in American Geologist, Vol. XVIII,,pp. 379-384, 1896. = The paper appears in the American Journal of Science for March 1897. 3Ueber das oder Ober Laurentian in Canada. Neues Jahrbuch flir Mineralogie Beil. Bd. VIII, 1893. A Further Contribution to our Knowledge of the Laurentian.— Am. Jour. Sci. July 1895. ABSTRACTS 93 The Grenville series differs from the Fundamental Gneiss in that it contains certain rocks whose composition marks them as highly- altered sediments. These rocks are in part limestones and in part certain peculiar gneisses, rich in sillimanite and garnet, having the composition of shales, or very rich in quartz and passing into quart- zite, having thus the composition of sandstones. These rocks, as has been shown in one of the papers above referred to, usually occur in close association with one another, and are quite different in composi- tion from any igneous rocks hitherto described. These rocks it is which are considered as characterizing the Grenville series. They usually, however, form but a very small proportion of the rocky com- plex of the areas in which they occur, and which, owing to their pres- ence, are referred to the Grenville series. They are associated with, and often enclosed by, much greater volumes of gneisses and amphi- bolitic rocks, identical in character with those of the Fundamental Gneiss. The limestones are also almost invariably penetrated by great masses of coarse pegmatite, and in some cases large bodies of the limestone are found imbedded in what would otherwise be supposed to be the Fundamental Gneiss. The whole thus presents the character of a series of sedimentary rocks, chiefly limestones invaded by great masses of the Fundamental Gneiss, and in which possibly some varie- ties of the gneisses present may owe their origin to a partial admixture of sedimentary material with the igneous rocks by actual fusion. There is, however, no reason to believe, from the evidence at present avail- able, that any considerable proportion of the series has originated in the last mentioned manner. From the relations of the several series, as displayed in central Ontario, it is believed that the sedimentary portion of the Grenville series is but a modified part of the Hastings series, that in fact the present relations of the Grenville series are due to a commingling of portions of the Hastings series with the Fundamental Gneiss along their line or zone of contact. The Fundamental Gneiss upon which the Hastings series was originally laid down having at a subsequent time been softened by the influence of heat, and having under the influence of dynamic action eaten into and fretted away the overlying Hastings series, giving rise to an intermediate zone of mixed rocks which constitutes the Grenville series. The relations of the two series being thus very similar to that shown by Lawson and Barlow to exist between the Huronian and the Fundamental Gneiss in the districts about Lake Superior and Lake Huron. 94 ABSTRACTS The appended remarks by Dr. Ells deal with the district to the east and northeast of that studied by Adams and Barlow, and show that here also the same series can be recognized and that the same . relations between them obtain. The Origin afid Age of the Gypsum Deposits of Kansas. By G. P. Grimsley. Among the minerals of economic importance in the state of Kansas, gypsum occupies a prominent place, and it has attracted the attention of geologists for many years. At the present time the state stands first in the Union in value of its.gypsum product. The deposits of ecomic value occur in a belt trending northeast-southwest across the state with a length of 230 miles and in width varying from five miles at the north to thirty-six miles iiear the southern line. There are three important areas : northern, central, and southern. Intermediate deposits connect the northern and central areas, while the interval between the central and southern is occupied by salt beds. The topography increases in ruggedness southward. In the northern area three companies are engaged in plaster manufacture. There are six mills in the central area, and two in the southern. The rock is white or a gray mottled, with a saccharoidal texture in the upper portion, but becomes more compact below. The dip is toward the west. In the central area occur a number of secondary gypsum dirt deposits, which form the basis of the greater portion of the plaster manufac- ture. This material is a granular dirt of an ash-gray color. It is soft, and readily shoveled into cars, so that it is ready for calcining with less labor and expense than in the case of the solid gypsum. At the present time four of these deposits are worked. Under the microscope the dirt is seen to consist of a mass of small angular gypsum crystals of varying size. Mingled with the gypsum crystals are small quartz crystals, some calcite poorly crystallized, and traces of organic material. The deposits are found in low, swampy ground near small creeks, and strong springs occur in most of them. Usually there is a ledge of rock gypsum at the same level, or ten to twenty feet below. The gypsum beds are all Permian, ranging from the Neosho epoch to the close, and they rise geologically to the south. The northern and central rock gypsum deposits appear to have been formed in the same gulf cut off from the western Permian sea ; while the gypsum dirt ABSTRACTS 95 deposits are secondary and of recent age. The southern deposit was formed in a shallow gulf cut off from the Permian sea at a later period not far from the close. Salt appears to have been deposited with the gypsum, but now it is only found farther out in the old gulf where it was thicker. The Cofficii Glacier, Greenland. By Ralph S. Tarr. In this paper attention is called to the recently published conclu- sions of Professors Salisbury and Chamberlin that glaciation on the Greenland coast has not in recent times extended much further than the present ice margin. The evidence upon which this is based is stated, and it is pointed out that another interpretation of the rugged topography of the high parts of the Greenland coast is possible. This is, briefly, that the peaks seen from a passing vessel present the face which is least likely to be glaciated ; also that they have been longer exposed to denudation, both preglacial and postglacial, and that, being higher than the surrounding land, they were more rugged before the ice came, and reaching up into the glacier they were less scoured by the ice than is the case with the lower lands. The description of the geological conditions and evidence of glaciation upon the Upper Nugsuak peninsula (latitude 74° 10'- 15') follows, and it is pointed out that the higher points in this region have all been covered by ice notwithstanding the fact that, as seen from the sea, the topography of the peaks is as rugged as that of other parts of the Greenland coast. Small beds of till and numerous trans- ported pebbles, whose origin is somewhere within the ice-sheet, are found on the highest points. The Devil's Thumb, rising 2650 feet above the sea, has been glaciated ; Wilcox, 1400 feet high, and twenty- five miles from the margin of the ice, has also been ice covered, and the glacier has extended over the Duck Islands, which are eight or ten miles further to sea than this. Adding to the land height the depth of the water in the neighboring bays, it is certain that at the Devil's Thumb the ice-sheet was not less than 3000 feet thick ; that at Wilcox Head, twenty to twenty-five miles from the present ice margin, there was glaciation to a depth of certainly not less than 2000 feet, and that at the Duck Islands, thirty or thirty-five miles from the present glacier, the ice-sheet had a depth of not less than 600 feet. In other words, in this part of the Greenland coast all the land now visible has been recently glaciated, and so far as evidence here is concerned, there is no 96 ABSTRACTS reason to place a limit to its advance. The presence of shells in a recently abandoned moraine, as well as in the moraine now being made, and in the ice itself, appears to prove a recent retreat of the glacier, for these shells are of species novv living in the waters of the fjord. They are found at a distance of five miles from the sea and at an elevation of 586 feet above it, and at various levels below this. The description of records left near the margin of this glacier is given, with the idea of putting forward facts which will permit accurate measurement of the change of ice front in the future. This seems all the more desirable, since evidence is abundant that the glacier in this part of Greenland is now retreating at a notable rate, and that it has recently withdrawn from a considerable area. The evidence of this is found in the presence of moraines at a distance of 100 or 200 feet from the present margin of the glacier, and so recently abandoned that no vegetation has begun to grow upon the soil. Moreover, freshly scratched bedrock has been so recently uncovered by the glacier that lichens have not beafun to arrow. Unconformities in Marthas Vineyard and Block Island. By J. B. WOODWORTH. Beginning below, plant-bearing beds of Cretaceous age appear in both islands, without their base being exposed. On Marthas Vineyard marine Cretaceous strata overlie the non-marine beds ; although the contact has not been worked out, unconformity is inferred from the occurrence of lignitic fragments in the overlying marine beds. Above the Cretaceous at Gay Head, and on an eroded surface rests the Miocene of Lyell and Dall, composed of the osseous conglomerate and the greensand. Overlying these beds is a yellowish greensand, pos- sibly Miocene but probably Pliocene. Fragments of a Pliocene formation have been identified at Gay Head by Dall. There was erosion in the area during or just after Eocene time, again between the greensand and the osseous conglomerate, and between the time of the greensand and the latest Neocene, the yellowish greensand. A bowlder formation now appears, composed of rocks derived from the mainland on the north. It was preceded by local folding of the underlying beds and by erosion along narrow channels, the Miocene being locally swept away. On Block Island the earliest Pleistocene rests upon the Cretaceous at Clay Head, the Miocene being entirely wanting. Succeeding the bowlder deposits there are from twenty-five ABSTRACTS 97 to thirty feet of granitic gravels and sands, the deposition of which was brought to a close by profound folding over Marthas Vineyard and Block Island. The Tisbury beds on Marthas Vineyard and the Mohegan Bluff beds on Block Island were now laid down upon the eroded surface of the upturned beds of older strata. These horizontal beds are evidently of glacial origin. After their deposition there came a long period of fluvial erosion, the Vineyard subepoch, in which the islands were deeply denuded, as pointed out by Shaler in 1888. On xMarthas Vineyard the capture and beheadal of streams is shown in the existing topography as the work of this time. Then followed the last glacial epoch with the deposition of moraines and sand-plains. The older Pleistocene deposits denominated Weyquosque by Shaler in 1888 appear to be the equivalent of the Columbia which in the region of the New England Islands is divisible into an upper and lower horizon, separated by unconformity of dip and erosion marking the Gay Head diastrophe. The author is unable to accept the hypothe''- sis of glacial thrust for the deformation on the islands named because in each case the fold preceded the evidence of glaciation found in overlying bowlder deposits. Those who accept the ice-thrust hypothe- sis, it is insisted, must consider the work to have been done' by two advances of an ice-sheet long anterior to the last glacial epoch. In the discussion Professor Wm. B. Clark called attention to the evidence of unconformity found by him in New Jersey between the non-marine or Potomac beds and the marine Cretaceous, and also to the meagerness of the horizons in this northern extension of the coast- plain as compared with more southern fields. He suggested that an examination of the foraminiferal greensand might show that, as in New Jersey, the Miocene foraminiferal casts had at least in part been derived from preexisting Cretaceous beds. Mr. Cxilbert noted that the long erosion interval on the islands, preceding the last glacial epoch, wa& consistent with the history made out for the rest of the United States. Ho7nology of Joints and Artificial Fractures. By J. B. Woodworth. A synopsis of the structure of a typical joint as described in a previous paper {Proc. Boston Soc. Nat. Hist., Vol. XXVII, 1896, pp. 163-183) showed a central area of fracture termed the joint-plane, which is traversed by feather-fracture. Marginal to this is a border of overlapping fractures, traversed by cross-fractures, forming the "joint- 98 ABSTRACTS fringe." This system of fractures en echelo>i pervades the entire joint. Specimens were exhibited showing the joint structure above described, and also examples of fracture produced by blasting in a quarry, in which specimens there was a central warped surface corresponding to the joint-plane, and a system of marginal fractures comparable to the joint-fringe. The oblique imbricating plates formed by this fracture suggest a comparison with the "diagonal fissility" of Van Hise. The obliquity in the case of joints is found of value in determining the axis of breaking tension, whatever the nature of the force which pro- duced the joint or fracture. This axis is a normal to the oblique marginal fractures. In mapping joints, a short straight line like the "strike" line crossed by short oblique lines having the obliquity of the observed marginal fractures of the fringe was proposed as a symbol. (Not to be published in the Bulletin of the Geological Society.) Notes on Rock Weathering. By George P. Merrill. This paper was in direct line with those read at previous meetings of the Society, and had to do with the ultimate products of decompo- sition as shown in the transition of a fresh micaceous gneiss to the condition of a bright red residual soil. The results of physical and chemical analyses were presented, and on the assumption that the alu- mina had remained essentially constant during, the decomposition, it was shown that 44.67 per cent, of original matter had been lost through the solvent and leaching action of water. The writer stated that on the basis that such results were rather under than above the actual amounts he had from a large number of analyses been led to the con- clusion that an ordinary crystalline siliceous rock of a granitic or gneissoid type in passing by decay from its fresh condition to that of soil might lose about 50 per cent, of its original constituents. The cause of the red color of the residuary material was discussed. The writer agreed with Professor Crosby in that the color was wholly superficial, but thought that such color might be due not wholly to dehy- dration of the ferric oxides present, but also to the presence of a large amount of coloring matter, the iron oxides having been shown to be largely insoluble, and hence accumulating in the residue along with the alumina. The subject of zeolites in soils, and their efficacy as con- servators of potash was discussed, the conclusions being that not only is there no proof of the formation of zeolites, during the ordinary proc- esses of true weathering, but also that even did zeolites exist in the ABSTA\4CTS 99 soils, such could hardly be of interest for the reason stated, since potash is not, on the whole, a prominent constituent of zeolitic minerals. In this connection Mr. Merrill suggested that the term weathering as applied to rocks should be limited to those processes going on within the zone of oxidation and resulting as a rule in the destruction of the rock was as a geological body, while the more deep-seated processes which result in the production of new minerals of the nature of zeolites, chlorite, epidote, etc., and which are really of mineralogical rather than geological moment, should be looked upon as metamorphic and desig- nated as products of hydrometamorphism. In discussing the probable condition of the soluble constituents of soil, a brief table was given showing that in rocks of as diverse types as granite, phonolite, diabase and basalt, the percentage amount of soluble constituents in both fresh and decomposed materials may be very nearly equal. JVb^es on the Potsdam and Lower Magnesian Formations of Wisconsin and Minnesota. By Jcseph F. James. The observations recorded in this paper were made on a trip through Wisconsin and Minnesota during 1889. Starting from Mad- ison, Wis., the route was through Lodi, Devil's Lake, Baraboo, Able- man's, Ripon, Winneconne, New London, Morritan, Hudson and River Falls, Wis., to Stillwater, Winona, and Dresbach, Minn. Natural and artificial, exposures, especially in quarries, were studied at all of these places and others, and nearly all of them are illustrated by sections drawn to a scale. The most extensive section is that made between Knapp and Wilson stations on the line of the Chicago, Milwaukee, St. Paul and Omaha R. R. The distance between these two places is about five miles. There is a continuous up grade with many cuts and a number of exposures. The section extends from well down in the Potsdam up to and into the Lower Magnesian, with an estimated vertical range of about 140 feet. Two slabs of worm trails from the Potsdam of Dresbach and the Lower Magnesian of Prairie du Chien are also illustrated. Preliminary Note on the Pleistocene History of Paget Sound. By Bailey Willis. During the past season the drift deposits about the southeastern edge of Puget Sound have been studied in some detail. They are found to consist of several beds of till, separated by stratified deposits 100 ABSTRACTS of clay, sand and coarse gravel, together with widely distributed lignite beds. The character and extent of the glaciation of the Puget Sound region are indicated in these deposits, and it is found that the principal flow of ice was rather from the north than from the moun- tains on the southeast. Two problems are presented by the phenom- ena : (i) The sequence of glacial advance and retreat and the extent and duration of climatic changes indicated by the presence of lignites ; (2) the bearing of the peculiar conditions of glacial development upon the physiography of the sound. Either the deeper valleys of the sound have been eroded during a period of high level from the once more extensive sheets of drift or, as suggested by Russell, the chan- nels represent the beds repeatedly occupied by glaciers which in their advance and retreat built up the plateau-like eminences of the region, probably upon the divides of the preexisting topography. The past condition of Puget Sound under confluent glaciation is probably now represented by the Malaspina glacier and its attendant phenomena. The Leucite Hills, Wyoming. By J. F. Kemp. The paper opened with a brief statement of our earlier knowledge of the distribution of leucite, and of the special interest that attaches to it. Outside of the European mainland localities were cited on the Lipari Islands, the Cape Verde Islands in Asia Minor, Siberia, Java, New South Wales, Kilima-Njaro, Argentina, Brazil, New Jersey, Arkansas, Lower California, Yellowstone Park, Montana, and British Columbia. The presence of the rich alkaline magmas from Montana to the Mexican line, and in or near the eastern Rocky Mountains, was commented on. The previous observations of Zirkel on the rock of the Leucite Hills were briefly summarized so as to show that the only minerals observed by him in the very limited materials gathered by the geologists of the fortieth parallel were leucite in great abun- dance, small prisms of augite, biotite, apatite, and magnetite, and that the remarkable absence of feldspar, haliyne, and nepheline, was a cause of surprise to him^. The geographical situation of the Leucite Hills and their areal distribution was then shown by lantern slides. There are three mesas or buttes in the main group, of which the southern, or Leucite Hills proper, is the largest. Orenda Butte, two or three miles north, is next, and Black Rock Butte, five miles northeast, is the smallest of all. The two former consist of several lava sheets resting on Laramie sandstones. From them arise cones 300 feet and less in ABSTRACTS 10 1 height that resemble craters, but that are solid lava. Black Rock Butte has no cone. The lava sheet is more or less dissected, but the fragments are found but a very short distance from the escarpment. Specimens from all the buttes, and from various parts of the southern one, show that while the leucite is almost always present it varies greatly in amount and may be replaced almost entirely by sanidine. Haiiyne was observed, and in Black Rock Butte large augites with surrounding borders of the peculiar yellow biotite. All these points of petrographical interest were illustrated by lantern slides from photo-micrographs. The two published analyses were cited, and in view of the abundant sanidine the rock was called a leucite-phonolite rather than a leucitite. The cones were explained as due to the upwelling of a viscous lava in the final phases of activity. No tuffs or dikes were met. Pilot Knob, an isolated butte lying twenty miles further west and north of Rock Springs, was also described, and the rock was shown to consist of small needles and larger crystals of augite in an isotropic groundmass that was regarded as glass. The determinations of the rock as trachyte in the Fortieth Parallel Survey must have been due to a confusion of slides. T/ic Pre-Cambriafi Topography of the Eastern Adirondacks. By J. F. Kemp. The paper opened with a review of the geological formations in the Lake Champlain Valley. It was shown that Palaeozoic sediments formed all the eastern shore except for a small area at the south. In Vermont the Georgian strata of the Cambrian are met, but Acadian representatives are lacking. The Georgian is only seen at a distance from the crystalline nucleus of the Adirondack Archrean (Algonkian) Island, whereas the Potsdam and Ordovician lie now well up on its flanks. Prolonged search has failed to show on the west shore of Lake Champlain anything below the Potsdam. Mr. Walcott has already pointed out these relations. The writer stated that the crystallines had all been formed and greatly metamorphosed long before the Cam- brian times, whose strata on their flanks lie flat and show no meta- morphism whatever. He further stated that the crystallines must have been land with an encroaching shore line during the Cambrian, and that the land was deeply carved by erosion, whose results he freely admitted had been greatly modified by faulting. Then, with a series of lantern slides based on the topographic maps of the United States 102 ABSTRACTS Geological Survey, he traced the relations of the Palaeozoic rocks to the pre-Cambrian strata, showing that they set up into embayments at Willsboro, Essex, Westport and Port Henry. The most interesting of all are, however, at Crown Point and Ticonderoga. Small outlines of Potsdam were shown in the valley of Trout Brook, west of Rogers Rock on Lake George, and over a high divide ; from the lake and in the valley of Lake George itself. The old depressions up which the encroaching sea of the late Cambrian set are still discernible. Then, with a succession of adjacent maps, the speaker traced out an old depression back from Lake Champlain through Crown Point, with several scattered outliers of Potsdam, thenqe southward past Penfield Pond, and up a branch in central Ticonderoga, where another Pots- dam outlier is found ; thence through the valley of Paragon and Paradox lakes into the Schroon Lake valley, where a few acres of Calcif- erous flinty limestone remain under Schroon Lake P. O. The last named exposure is nearly twenty miles from Lake Champlain, and is forty miles from the next outcrop down the Schroon and Hudson rivers. Attention was called to the little outlier containing Potsdam, Calciferous, Trenton, and Utica strata at Wells, on the Sacondaga River, as recently figured by Mr. Darton. All these outliers are quite flat, seldom reaching fifteen degrees, and as a rule dip northwest and strike northeast. The original depressions were developed especially where there is crystalline limestone. Stratigraphy and Patceontology of the Laramie and Related Forma- tions in Wyoming. By T. W. Stanton and F. H. Knowlton. The separation, in recent years, of the Denver, Arapahoe, and sim- ilar formations in Colorado, and of the Livingston in Montana, from the coal-bearing Laramie series, has called in question the unity of the beds usually referred to the Laramie in Southern Wyoming. Sev- eral facts in the reported stratigraphy and palseontology of the region suggested the possible presence of the Denver beds. The same hori- zon was suggested by the vertebrate fauna of the " Ceratops beds" in Eastern Wyoming as described by Messrs. Marsh and Hatcher. In view of these facts and queries the authors undertook the examination of the Ceratops area in Converse county, and of the coal -bearing localities along the Union Pacific Railroad, giving special attention to the stratigraphy and to the branches of palseontology — invertebrates and plants — in which they are respectively interested. The areas studied in ABSTRACTS 103 Converse county, the Laramie Plains, and Bitter Creek valley are described somewhat in detail, and more general observations are given on other localities treated, including Evanston and Hodge's Pass, Wyo- ming ; Coalville, Utah, and Crow Creek, Colorado. The results are nega- tive so far as the Denver and later related formations are concerned. That is, no division of the beds between the base of the Laramie and the Fort Union was found possible, and the Laramie age of the Ceratops beds of Converse county, as well as of the Black Buttes coal horizon, was confirmed. On the Laramie Plains most of the coal-bearing localities are on a horizon lower than the Laramie, and overlain by marine Fox Hills beds. The coal-bearing series at Point of Rocks is also in a similar position, and the considerable flora it has yielded must now be referred to the Montana formation instead of to the Laramie. In this connection all the non-marine invertebrates and plants now known from the Montana formation are brought together in lists, and the more general questions of the upper and lower delim- itation of the Laramie are discussed. A Complete Oil- Well Record i// the McDonald Field between the Pittsburg Coal and the Fifth Oil Sand. By L C. White. Geologists have long desired a careful and accurate measurement of the rocks passed through by the borings for oil in southwestern Pennsylvania. Mr. Carll, during the life of the second geological survey of Pennsylvania, had several careful records kept of the borings in Clarion and adjoining counties, but as the strata dip to the south- west, and thicken somewhat, it was very desirable to have a new standard of comparison in southern Pennsylvania. The writer hap- pily found in Mr. T. J. Vandergrift of Jamestown, N. Y., a successful oil operator of many years' experience, the right man to undertake the work purely as a labor of love in the interest of geology. The well in question was drilled by the Woodland Oil Company, of which Mr. Vandergrift is president, on the S. B. Phillips farm in the famous McDonald field, near the line between Allegheny and Washington counties, Pa., about twenty miles southwest from Pittsburg. The pro- ducing rock of this region is the lowest member of the Venango Oil .Sand group, or what is known to oil operators as the Fifth Oil Sand. This rock, it will be remembered, has yielded petroleum in greater quantity than any other yet found on the American continent, since some of the wells of the McDonald region produced oil for a short 104 ABSTRACTS time at the rate of 15,000 barrels daily. It is no slight task to keep one of these deep-well records with the necessary care, since to be reliable a sample of the drillings must be washed, dried, and properly labeled every time the tools are withdrawn from the well. Then, too, as all rope measurements are not reliable to within ten to fifteen feet, steel-line measurements must be taken frequently, all of which was done in the present instance, so that the 2342 feet of strata, penetrated from eighty-six feet above the Pittsburg coal down to seven feet under the base of the Fifth Oil Sand, is represented by 570 samples of drill- ings and fifty-five steel-line measurements, from which data a very complete account of the rocks from the Pittsburg coal down nearly to the top of the Chemung may be obtained, for all of which geologists are certainly under great obligations to Mr. Vandergrift. The writer will publish the log of the well, as noted by the party keeping the same, and also a condensed record in more strictly geological terms. A Note on the ''Plasticity'' of Glacial Ice. By Israel C. Russell. Experiments by McConnell, Kidd, and Miigge have demonstrated that a bar of ice cut from a single crystal, with the optic axis perpen- dicular to two of the side faces, when subjected to a bending stress, will bend freely in the plane of the optic axis, but not at all in a plane perpendicular to that axis. In the bent crystals the optic axis in any part is normal to the bent face in that part. As stated by McConnell, the crystals behave as if they were composed of an infinite number of thin sheets of paper, normal to their optic axis, and attached to each other by some viscous substance which allowed one to slide over the next with great difficulty. The greatest freedom of movement was found to occur when the ice experimented on was near the melting point, and became less and less with a decrease of temperature. In certain of the experiments referred to the freedom of movement at — 2° C. was twice as great as at — 10° C. When thin sections of gla- cial ice are examined by means of polarized light, as shown by Deeley, Fletcher, and others, it is seen that the granules of which it is com- posed are fragments of ice crystals, or perhaps imperfectly formed crystals which interlock one with another so as to resemble the struc- ture of coarsely crystalline dolomite. The optic axes of the granules are without definite arrangement, but have all directions. Pressure brought to bear on such granular ice in a definite direction would cause movement in the crystal frag- ABSTRACTS 105 inents or granules, whose optic axes were in the plane of pressure, but would not change the forms of the crystal fragments not thus oriented. The cr3'stal fragments having their optic axes in the plane of pressure would be changed in shape, as in the experiments cited above, by movements along gliding planes. The stresses in glaciers are mani- festly in various directions, and owing to the flow of the ice, inequali- ties of the rock surfaces beneath, etc., must be exceedingly complex. As the crystal fragments of which glaciers are composed are without orderly arrangement it is evident that some of them will be properly oriented to be deformed by a differential movement of their parts along gliding planes, by pressure acting in any direction. Under diverse stresses the resultant motion would be in the direction of least resistance. That is, the granular ice would behave like a plastic solid. The yielding of glacial ice to pressure, we conceive, is due to movements along gliding planes in the granules of which it is com- posed. If this process is continued the granules will evidently be destroyed by being divided along the planes on which movement occurs, but the resulting subdivisions, or "plates" perhaps we may term them, are reunited probably by the process of regelation. The plates, which after uniting have the same orientation, would form new granules. The longer this process is continued, or, in other words, the farther a glacier flows the greater the chances that a large number of "plates" will come together with similar orientation and the larger will be the resulting granules. Under the hypothesis here suggested the granules of glacial ice are considered as an inheritance from the granules in the ice formed by partial melting and refreezing of neve snow. The granules at first are minute, but under the pressure of ice at higher levels they yield along gliding planes, and a resultant motion similar to that of plastic solids under like conditions is initiated. The granules are destroyed by this process, but progressive motion leads to the union of a constantly increasing number of the plates into which they are divided, and the growth of larger granules. JVork of the U. S. Geological Survey in the Sierra Nevada. By H. W. Turner. To the north of the Fortieth Parallel Mr. Diller has published one folio on the Lassen Peak area, and has a very extensive series of notes which are as yet unpublished. The gold deposits of the Sierra I06 ABSTRACTS Nevada are, with minor exceptions, pretty closely confined to the belt of Palaeozoic and Juratrias slates with their associated greenstones and included granitic masses. The area covered by these rocks forms the real gold belt of the Sierra Nevada. It is gratifying to announce that the geological work covering this area is now nearly completed. The Marysville, Smartsville, Sacramento, Placerville, and Jackson folios are published, and the Bidwell Bar, Downieville, Truckee, Pyramid Peak, Big Trees, and Sonora sheets are soon to be published. Atten- tion was called to the very extensive beds of Juratrias tuffs in the foot- hills, indicating enormous volcanic activity in Juratrias time. Later in the Tertiary the scene of volcanic activity was transferred to the crest of the range, and nearly the whole area of the gold belt was flooded with lavas chiefly of a rhyolitic and andesitic character. Shore Lines of Lake Warren and of a Lower Water Level in Western- Central New York. By H. L. Fairchild. This paper describes the eastward extension of the Warren shore line as discovered and traced by the author during the past summer. The beach, in excellent form, with the various shore-line phenomena, as cliffs, bars, spits, and hooks, has been traced from Crittenden to Lima, somewhat east of the meridian of Rochester ; while sufficient evidence of static water at a corresponding level is found much further eastward. The altitude of the beach between Batavia and LeRoy is 880 feet above tide. The altitude at Lima, subject to possible correc- tion of datum, is 877 feet. This water plane would lie just about 500 feet over the Iroquois plane and would apparently lie below the plane of Watkins Lake or of Lake Newberry, the early local glacial waters in the Seneca embayment. It therefore seems certain that the Warren waters never found outlet by the Horseheads channel. The comparison of the beach phenomena east of Indian Palls with the same phenomena at Crittenden and westward is difficult to make on account of the dif- ference in the topography of the country. East of Indian Falls the beach is transverse to the drumloid molding of the land surface, and the beach is not of so mature a character as south of Lake Erie, but nevertheless indicates a long period of wave action. Another beach at an altitude of 700 feet has been found at Geneva, New York, and traced westward continuously nearly to the meridian of Canandaigua. Abundant evidence of the same water plane is found in terraces and cliffs nearly to the Genesee River. The phenomena of this shore line ABSTRACTS 107 are nearly as strong as those of the Warren waters in the Genesee region. This beach is specially interesting as being entirely new and unexpected, and indicating a long pause of the waters at an altitude between the Warren and the Iroquois plane, while the correlating out- let is unknown. The body of the paper is a detailed description, for record, of the phenomena of these two beaches. Principal Features of the Geology of Southeastern Washingto)i. By Israel C. Russell. A brief statement of the results of a six weeks' reconnoissance in the southeastern portion of the state of Washington, made for the United States Geological Survey. Practically all of that portion of Washington which lies south of the Big Bend of the Columbia is occupied by a succession of basaltic lava flows. This is a portion of a vast lava-covered region embracing northern California, central and eastern Oregon, central and south- eastern Washington, and southern Idaho. The great fissure eruptions which supplied the Columbia lava, as the basalt is termed, occurred in the Miocene. The Columbia lava in Washington to the west of the Columbia, south of the Big Bend, as ascertained during a previous reconnoissance, is broken by extensive faults and the blocks thus formed variously tilted. In the region here treated, however, the basalt is horizontal over extensive areas, and deeply dissected by Snake River and its tributaries. Many lava sheets, one resting on another, were seen. Between some of the flows there are widely extended sheets of lacustral clay, sand, gravel, volcanic dust, and lapilli. In some instances leaves, and the silicified stumps and trunks of trees occur in these layers. The Columbia lava flowed about the bases of the mountains of eastern Washington and the adjacent portion of Idaho, in a series of inundations which covered the low country to the south. The level basaltic plateau meets the mountains of metamor- phic rock in much the same manner that the sea joins a rugged and deeply indented coast. The lava entered the valleys and gave them level floors of basalt; the deeply sculptured ridges between the valleys were transformed into capes and headlands; outstanding mountain peaks became islands in the sea of molten rock. After the last of the lava sheets was spread out, the rivers flowing from the mountains began the excavation of channels across the basaltic plateau and have deeply dissected it. The most important of these channels is the one [08 ABSTRACTS excavated by Snake River. From the mouth of^Snake River to Lewis- ton, the stream is a narrow^, deep-sided canyon, about 2000 feet deep. Where Snake River forms the boundary between Washington and Idaho, its gorge is about 4000 feet deep and fifteen miles broad. Within this vast canyon there are many lateral ridges, and a great vari- ety of topographic forms due to erosion. This portion of Snake River canyon compares favorably with even the most magnificent parts of the Grand Canyon of the Colorado, except that it lacks the gorgeous col- oring to which so much of the charm of its southern rival is due. The thickness of horizontally bedded basalt exposed in the walls of Snake River canyon and in the adjacent Blue Mountains, is in the neighbor- hood of 5000 feet, but the maximum thickness is not exposed. In the walls of the canyon at three localities, the summits of steep, angular mountain ranges are revealed. One of these buried peaks rises about 2500 feet above the river and is covered by fully 1500 feet of horizon- tally bedded basalt. The sheets of clay, sand, and gravel interleaved with the basalt, especially near its junction with the bordering moun- tains, furnish conditions favorable for obtaining artesian water. A number of flowing wells derive their water supply from this source. The Blue Mountains, at least at their northern extremity, consist of a low, flat-topped dome of Columbia lava, which has been deeply dis- sected by consequent streams. The surface of the basaltic plateau is covered with residual soil which has an average depth of sixty to eighty feet over thousands of square miles. The soil is exceedingly fine, dark brown or black in color, and of wonderful fertility. This is the soil of the celebrated wheat lands. The subsoil is fine, light yellowish in color, without stratification, and in many localities traversed by minute, irregular, but, in general, vertical tubes. In many ways the subsoil closely resembles loess. Its origin from the disintegration and decay of the underlying basalt is clearly manifest. Topographically the Columbia lava presents great diversity. In the Blue Mountains there is an intricate series of sharp-crested ridges, separated by a labyrinth of canyons, having in general a depth of about 3000 feet. Between the Blue Mountains and Snake River there are broad remnants of the nearly level plateau, separated by narrow, steep-sided canyons, in general 2000 feet deep. North of Snake River there is a vast area without deep canyons, but diversified by short hills from fifty to eighty feet high, none of which, however, rises above a certain general level. Along the eastern portion of this hilly plateau, or rolling prairie as it was before cultivation began, there are a few ABSTRACTS IO9 prominent, island-like buttes of quartzite, which rise through the basalt. Among the details noted in the Columbia lava are certain hori- zontal joints which cut the vertical columns of basalt and may be traced for several miles. The large vertical columns of lava when weathered sometimes show that they are composed of small horizontal columns or prisms which radiate from a confusedly jointed central core. The joints which bound the large vertical columns furnished the cooling surfaces for the rock they enclose. The bases or ends of the radiating columns are frequently revealed on the surfaces of the slightly weathered vertical columns by a network of lines resembling shrinkage cracks. A report on the observations outlined above will be published by the United States Geological Survey. Old Tracks of Brian Drainage in Western New York. By G. K. Gilbert. The glacial Lake Warren occupied the Erie and parts of the Huron and Ontario basins. In the next important stage Lake Erie was established as a non-glacial lake, being separated from the glacial Lake Algonquin at the west and the glacial Lake Iroquois in the Ontario basin. As Warren drained westward and all the lakes of the succeed- ing stages drained eastward, it was inferred as probable that the revo- lution was initiated by the opening of an eastward passage for Warren waters. Such eastward passage was sought and found in western New York. In fact several passages were found, which succeeded one another as the northward retreat of the ice-front exposed lower and lower cols on the north-south ridges which diversify the general north- ward slope of that region. Between the ridges there were lakelets, walled by ice on the north, and these lakelets received the material scoured out by the waters in crossing the ridges, so that the surviving phenomena are chiefly channels and scourways through the higher lands, with gravel deltas at their eastern ends. Most of the drainage lines descend to the Iroquois level, but some terminate at higher levels, and it is thought probable that for a short time a glacial lake occupied the middle Mohawk valley, being doubly dammed by ice lobes on the east and northwest. The district traversed by the channels extends from Le Roy to Chittenango, and the relation of the channels to one another and to the Iroquois shore show that during the channel epoch the trend of the ice-front in that region was approxinjately east-west. no ABSTRACTS It was found also that when the water first fell to the level of the Niagara escarpment, so as to cascade over its edge, there were five points of discharge. Three of these were short-lived, but Lockport for a considerable period divided the water with Niagara. The Physical Nature of the Problem of General Geological Correlation. By Charles R. Keyes. The main object of this communication has been to formulate briefly certain suggestions which have arisen in the course of recent attempts to parallel some of the geological formations in the Missis- sippi valley. They appear to have a much more than local bearing, and to affect the whole problem of general geological correlation, and perhaps also even our present system of geological classification. Regarding as the principal function of correlation the establish- ment of a practical scale of geological succession to which may be referred the strata of all parts of the globe, the critical criteria adopted became essentially the real basis of geological classification or historical geology. A rational classification of geological phenomena reflects the genesis of the events recorded, and this is manifestly the ultimate aim of all methods of paralleling strata. In the correlation, or com- parison of geological formations, experience has shown that the sub- ject may be viewed from at least four very different points of vantage. The aspects presented are: local, provincial, regional, or general. With the various methods which have been followed from time to time in correlative inquiry the almost universal practice has been to attempt to base the broader generalizations upon criteria that are in reality applicable only to limited areas. Passing by a review of the various methods of correlation which have been used, a little consideration shows that all are necessarily local in application and not general. This distinction between the two sorts of correlation which is usually entirely overlooked should be clearly borne in mind. When widely applied the almost total failure of the several methods that have been tried is only too apparent. The inherent weakness of the most common criteria — the fossils — is readily gleaned from the writings of Huxley, Irving, Van Hise, McGee, Wal- cott. Brooks and others. The problem is manifestly a physical and not a biological one. We must look therefore to physical criteria for its solution. The nearest approach to an absolute foundation for a system of geological synchrony is to be found in the work of Irving, ABSTJ^ACTS III in his work on the pre-Cambrian crystallines of the northwest, in which unconformities are given great prominence; by McGee in his investi- gations of the coastal plain deposits of the Middle Atlantic slope, in which similarity of genesis, or homogeny, is the governing factor; and by Davis and others in physiographic work, in which periods of base- leveling are made the all-important features in the cycle of land degradation, and inferentially the consequent sedimentation in adjoin- ing, seas. Without considering all the available physical and biotic criteria of correlation in these various local and general phases it may be said that general correlation finds a fertile suggestion for a rational foun- dation in the factors which govern sedimentation. Its chief criterion is a function of continental growth and decline, and is dependent upon orogenic movement. This has recently be encalled correlation by mountain-building cycles, or orotaxis. The division planes cutting the geological columns into systems, series, or smaller parts, are actu- ally, as well as theoretically, the lines of unconformities and their rep- resentatives. In the case of the more extensive ones they do, no doubt, represent base-leveled surfaces, or peneplains. Modified Drift in St. Paul, Minnesota. By Warren Upham. Within the limits of St. Paul (fifty-five square miles) are numerous plateaus of drift gravel and sand, 225 to 240 feet above the Missis- sippi River, and about 100 feet above the highest plains and terraces of similar modified drift in the Mississippi valley. The courses of marginal moraines in this city and its vicinity, the glacial strise, and diverse origin of the glacial and modified drift on the east and west, show that the slopes and currents of the ice-sheet converged to this area from the northeast and the northwest. On account of the sig- moid course of the Mississippi valley here, during the final melting and retreat of the ice-fields the tract where they had been confluent was occupied by a small glacial lake named Lake Hamline, which extended over the greater part of the western half of the area of St. Paul, having an outlet toward the southwest and south, across the present watershed between the Minnesota and Mississippi rivers, to Rich Valley and the Mississippi. In this temporary lake the modified drift plateaus were accumulated, their material being thought to be supplied from exceptionally abundant englacial drift brought to the area of glacial confluence. The surface of the glacial lake during the 112 ABSTRACTS early part of its existence, as shown by the Hamline and Como pla- teaus, was about 250 feet above the present river, or 930 to 940 feet above the present sea level. A little later, when the plateau a mile east of Lake Como was formed, the lake level had fallen apparently five or ten feet. At the time of formation of the Summit Avenue plateau, it had been further lowered several feet, having then appar- ently an elevation of about 915 feet; and still later and smaller plateaus show that this lake finally was reduced to 875 or 870 feet above the present sea level. In one section it was observed that the till, exposed to a depth of twenty to twenty-five feet, at an elevation of about 875 to 900 feet above the sea, has throughout that thickness an imperfect stratification. The author attributes this to the action of the water of Lake Hamline, receiving the till from a previously engla- cial and superglacial position in and upon the melting ice-sheet. It is an observation almost exactly like that made by him a year earlier in the drift section of the Lake Erie shore in Cleveland, Ohio; and here, as there, he concludes that the volume of the englacial drift is represented by the th „kness of the obscurely bedded or laminated till. In another section a sheet of till ten to twelve feet thick, resting on striated Trenton limestone, is shown, as the author thinks, to have been wholly englacial, this being indicated by deflection of the glacial striae. Many of the striae, intersecting the southward courses of ear- lier glaciation, run N. 6o°-75° W., perpendicularly toward the neigh- boring Lake Hamline and its long plateaus of water-deposited drift. (Abstracts to be continued.) THE JOURNAL OF GEOLOGY FEBRUARY-MARCH, iSq-j PROFESSOR GEIKIE'S CLASSIFICATION OF THE NORTH EUROPEAN GLACIAL DEPOSITS. In Vol. Ill of this Journal, Professor James Geikie gives a concise account of his present ideas concerning the classification of the north European glacial formations. His statement involves a comparison of the drift formations of various north European localities with one another, and with those of the Alps. He differentiates six glacial epochs, separated by five interglacial epochs, which he names after typical localities, as follows : First glacial epoch, Scanian. First interglacial epoch, Norfolkian or Elephas meridiojialis stage. Second glacial epoch, Saxonian. Second interglacial epoch, Helvetian or Elephas antiquus stage. Third glacial epoch, Polandian. Third interglacial epoch, Neudeckian. Fourth glacial epoch, Mecklenburgian. Fourth interglacial epoch. Lower Forestian. Fifth glacial epoch, Lower Turbarian. Fifth interglacial epoch. Upper Forestian. Sixth glacial epoch, Upper Turbarian. I. Sca)iia?i stage. — The oldest glacial formation of north Europe appears in Schonen and points to a Baltic glacier. In England perhaps the Chillesford Clay and Weybourn Crag with Vol. v., No. 2. 113 114 K. KEILHACK their arctic moUuscan fauna belong to this stage. Here also Geikie places the oldest Alpine glacial deposits as well as the old "Diluvium" of the plateau of central France. In another part of his work he expresses the opinion that the oldest ground moraine in the Baltic region of Germany belongs to this stage. 2. Norfolkian. — To this stage belongs the Forest bed of Norfolk, during the formation of which a climate prevailed at least as temperate as that of today. In Alpine districts the lignite of Leffe and elsewhere, as well as the interglacial deposits of the Hotting breccia, indicating a climate warmer than that of the present day, correspond to this stage. 3. Saxonian. — In this epoch the ice reached its greatest expansion. In north Europe the formations of this epoch extend to the borders of the Carpathians, the Sudetic Mountains, the Erzgebirge and the Thuringian Mountains. In the Alps, the corresponding formation covers a more extensive area than that of the Scanian epoch, while in Great Britain the corresponding sheet of drift is more widespread than that of any preceding or following epoch. To this stage belongs the lower bowlder clay of the British Isles, the Lower Diluvium of Holland and north Germany, the outer moraines and the associated gravels of Alpine lands, as well as the older moraines of the numerous imountain chains of middle and southern Europe. 4. Helvetian. — The character of the flora and fauna is vari- able, being here more arctic and there more temperate. To this stage belongs the interglacial deposits in Lanarkshire, Ayrshire, Edinburghshire, etc., and the Hessle gravels of East Anglia, the beach deposits of Sussex, and certain cave accumulations of mammalian remains, the interglacial beds of Holstein and Kott- bus, the sands of Rixdorf, the interglacial beds of Moscow the interglacial deposits of Cantal, as well as numerous old river deposits of the Thames, Seine, Rhine, etc. 5. Polandian. — To this stage belong the glacial and fluvio- glacial deposits of a Scandinavian nier de glace, which was smaller than the second, and the similar deposits of Great NORTH EUROPEAN GLACIAL DEPOSITS 1 1 5 Britain, the Alps and other districts. This stage includes the upper bowlder clay of the British Islands, the Upper Diluvium of central north Germany, Poland and central west Russia, the ground and terminal moraines of the "inner zone" of the Alps, together with the attendant gravels, and the younger valley moraines in various mountain chains. 6. Neiideckiaii. — The deposits of this interglacial stage are best observed in the southern coast lands of the Baltic. They originated partly in salt, and partly in fresh water, and are inter- calated between two ground moraines, which are designated the lower and upper bowlder clay respectively. The fauna points to a temperate, not to an arctic climate. 7. Mecklejiburgian. — To this stage belong the ground mor- aines and end-moraines of the latest Baltic glacier, and it reaches its southern extremity in the terminal moraine of the Baltic Hohenrucken. Of the same age as these north German deposits (Upper Diluvium of northern north Germany) are the moraines of the first postglacial stage in the Alps, the great valley glaciers of the British Islands, the Yoldia deposits of Scandinavia, the 1 00-foot beach of Scotland with its arctic fauna, and certain arctic plant beds below the Turbaries of Great Britain, Denmark and Scandinavia. 8. Lower Fore stia?i. — To this stage belong the deposits of the large fresh water lake (Ancylus-beds) filling a part of the basin of the Baltic, the older buried forests under the peat bogs of northwest Europe, and to some extent the Scandinavian Littorina beds. In the Alps no equivalent is known. The land in Europe possessed at that time a greater extent and a warmer climate than at the present day. 9. Lower Tiirbarian. — Marked by expansion of the sea, moister and colder climate, glacial formations in Scotland and Norway, where some of the valley glaciers extended to the sea, though most terminated at a considerable distance from it. In the Alps, the deposits of the second postglacial stage, the moraines situated in the inner valleys, correspond to this stage. This stage is represented in Britain by certain peat beds, in ii6 K. KEILHACK Scandinavia by calcareous tufa, and certain Littorina beds and in Scotland by beach lines. 10. Upper Fore stian. — -In northwestern Europe a second for- est bed, overlying 8, represents this stage. The area of land is greater than in the preceding glacial stage, but is still less then in the preceding interglacial epoch. The flora and fauna indicate a temperate climate drier than that of the 9th stage. 1 1 . Upper Turbarian. — Characterized by a new advance of the sea on the land. The shores are no longer reached by the ice, but it is highly probable that the moraines in the upper parts of the valleys of Scotland and Norway, and belonging to the last glacial period, are of the same age as the lower beach lines. With the close of this glacial epoch recent conditions begin. This is the new scheme by which Geikie would correlate the whole of the European glacial deposits. The following is given in tabular form, for the sake of simplicity. Stage. Great Britain. Sgandinavia. Alps. North Germany. I. C hi lie sf ord Deposits of the Oldest ground- Deck ensch otter clay. Wey- oldest Baltic moraine of and moraines bourn crag. glaciers in Schonen. the Baltic. belonging to it. 2. Forest-bed of Norfolk. Lignite of Lef- fe. Hotting breccia. 3- Lower Bowlder Moraines and Lower Diluvi- Outer moraine. clay. fluvio- glacial formations. um. High terrace gravels. 4. Marine deposits of Lanark- shire, Ayr- shire, etc.; Hessle grav- el, beach de- posits of Sus- sex. Peat bed in Holstein and Lign ites of Switzerland and Alt gau near Kottbus. Sands of Rix- dorf. 5- Upper bowlder Inner moraine Upper Diluvi- clay. and lower terrace grav- els. um south of the Baltic ter- minal mo- raine. NORTH EUROPEAN GLACIAL DEPOSITS 117 Stage. 6. Great Britain. Scandinavia. Alps. 7- Valley glaciers Yoldia clay. Moraines of the and 100-foot first post-gla- beach of cial stage. Scotland. 8. Lower forest- Ancylus beds, bed. Littorinabeds (in part). 9- Terminal mo- Littorina beds iMoraines of the raines in the (in part). second post- valleys. glacial stage. North Germany. Marine depos- its in West Prussia. Upper Diluvi- um north of the Baltic terminal mo- raine. r 0. Upper forest bed. II. Terminal mo- raines in the upper parts of the val- leys. The high authority of Professor Geikie on all questions con- cerning the European Ice Age has caused this classification, of the correctness of which Geikie himself is not thoroughly assured, to be received and used as final by many German writers who are not in a position thoroughly to appreciate the signifi- cance of the questions involved. Thus R. Credner^ writes : Folgen wir den Anschauungen, zu welchen neuerlich einer der hervorra- gendsten Glacialgeologen James Geikie, auf Grund vergleichender Unter- suchungen sammtlicher europaischer Vergletscherungsgebiete, vor Allem der britischen, der alpinen und der skandinavischen, gelangt ist, so haben wir fiir unser baltisches Becken vier durch Interglacialzeiten von einander getrennte Eisausbreitungen anzunehmen ; und . . . . es entstand schliesslich, den Rand des letzten baltischen Eisstromes andeutend, der Zug echter End- moranen, wel:her in Gestalt wallartig gestalteter Blockschiittungen nordi- schen Ursprunges von Preussen bis nach Schleswig-Holstein hinein den Landriicken kront. In opposition to the preceding, and in full concert with my col- league in the Royal Prussian Geological Survey, who with me is ' Entstehung der Ostsee, pp. 540 and 646. 1 1 8 K. KEILHA CK concerned in the geological exploration and mapping of the north German plain, I am bound to say that the fourfold classi- fication of the north German glacial formations, given by Geikie, in no way corresponds to our observations, nor to the statements published in the " Explanations to the geological special map of Prussia and the Thuringian states," or in the annual report of the Royal Prussian Geological Survey, and in other places. After many years of careful work in the territory referred to, we find no conclusive reason for ascribing the ground moraine, designated by us as the upper bowlder clay, to more than one ice epoch. On the contrary, all of my colleagues who are or were occupied in the territory of the terminal moraine of the Baltic range, are, like myself, firmly of the opinion that the youngest ground moraine in front of and behind the terminal moraine, was deposited at one and the same time, by one and the same mer de glace distinct from and younger than that which deposited the upper bowlder clay of the territory of middle north Germany. I must next consider the reasons which guided Geikie in his reference of the Upper Diluvium (drift) of north Germany to two glacial periods. They are stated in the second edition of the Great Ice Age, and seem to me to be essentially traceable to the four following points of view : 1. In different localities the ground moraine of the epoch of most extensive glaciation (Saxonian stage) contains (on account of the different directions of movement, radiating from the north of Scandinavia) bowlders quite different from the ground moraine of the younger inland ?ner de glace, which moved in the direction of the Baltic. According to Zeise the lower ground moraine of Schleswig-Holstein, is, so far as concerns its con- stitution, in no way different from the lower ground moraine of the moraine district, so the former cannot be the equivalent of the lower bowlder clay of the Mark, Posen, etc. 2. In Finland two systems of glacial striae are developed ; an older one, which thus far has been observed only outside the territory enclosed by the terminal moraine, and a younger one NORTH EUROPEAN GLACIAL DEPOSITS I 19 which thus far has been observed only inside the territory enclosed by the terminal moraine. The latter owes its forma- tion to a glacier which moved in the direction of the Baltic, and which cannot have exceeded the line occupied by the terminal moraine. The ground moraines before and behind this moraine must belong to two different ice epochs. 3. The upper bowlder clay of Great Britain contains Scan- dinavian bowlders, and the Scandinavian mer de glace, in this third ice epoch came in contact with the Scottish one. But the ice depositing the Upper Diluvium of the Cimbric peninsula did not quite reach the North Sea ; therefore this Upper Diluvium cannot be of the same age as the Scottish Diluvium of the third ice period, but must belong to a later fourth ice period. We come now to a comparison, if we suppose that the ter- minal moraine of the Baltic range represents the outermost edge of an independent glaciation, differing in point of time from that left behind by the Upper Diluvium of middle north Germany. 4. The separation is also shown by the appearance of the interglacial formations at Neudeck in West Prussia. On the other hand we must remark ( i ) that continual study of bowlders in the different ground moraines has convinced Ger- man geologists, more and more, that no fundamental difference in this respect, is to be found; that, in other words, no single stone can be considered as a guide stone of the ground moraine of a definite ice epoch throughout the whole of its duration. Therefore we must form no far reaching conclusion from the local differences between bowlders belonging to two ground moraines lying one over the other. 2. Geikie believes that the Finnish terminal moraine is iden- tical in point of time with that of the Baltic range. The line of union which he draws from east Prussia to Finland is, to my knowledge, purely hypothetical, and supported by no observa- tions. I look for the easterly continuation of the Baltic terminal moraine much farther south, in the interior of Russia, and am of the opinion that the Finnish terminal moraine belongs with that 120 K. KEILHACK of middle Sweden and southern Norway, a view which is shared by Herr Vogt in Christiania. 3. It seems to me unnecessary entirely to repudiate the opinions of the north-German geologists in order to explain the occurrence of Scandinavian bowlders in the upper bowlder clay of Great Britian. It remains also to be proved that these Scan- dinavian bowlders have not been taken from the terminal moraines of the second ice period and incorporated in the drift of the third, and it may also be questioned whether the ground moraine of Scotland with northern bowlders really belongs alto- gether to the third glacial period, or whether the connection of the Scandinavian with the Scottish ice in the third glacial period is more than imaginary. I am prompted to raise these ques- tions, because there are extremely weighty reasons against the acceptation of the Baltic terminal moraine as the outermost extremity of a separate (fourth) glacial epoch. On the con- trary, the similarity in age of this moraine with the so-called upper bowlder clay of the Mark, Posen, etc., seems undisputed. I will state these reasons more definitely. Of the terminal moraine of the Mark-Brandenburg, seventy- five square kilometers have been carefully mapped (scale: : 25,000) by the geological survey, and the investigation and surveying of the bordering territory on both sides reaches almost to Stettiner Haff in the north, and nearly to the northern edge of Lausitz in the south ; therefore the territory investigated stretches in round numbers seventy-five kilometers from the terminal moraine in both directions. I have myself mapped a strip of country in the territory of the Baltic range between the Oder and Vistula extending thirty-four kilometers from east to west, and lOO kilometers from north to south. The terminal moraine passes through this strip for a distance of forty-five kilometers, and a territory twelve to twenty-four kilometers in breadth south of the terminal moraine is included within it. I have also by gen- eral surveying of the 500 kilometers of terminal moraine between the Oder and the Vistula, crossed the range in many places. All this work, and particularly the special mapping of about NOR TH E UR OPE A N GLA CIA L DEPOSITS I 2 I 150 field sheets (i : 25,000) has now shown that the upper- most bowlder clay north and south of the terminal moraine are identical, that they belong to one glacial epoch, and that one sees in the terminal moraine' not the external margin, but only a stacre of retreat of the latest glacier of these countries. In numer- ous places the ground moraine passes smoothly under the terminal moraine and in this manner several bridges are formed between the inner and outer ground moraine, and their identity is so firmly established, that stronger arguments than Geikie's are required to overturn the results of long years of careful, special inquiry. Those parts of the ground moraine passing under the terminal moraine are joined together in large masses to the north and south with the extensive ground moraines which Geikie maintains belong to two different ice epochs. 4. By mentioning as proof of his position, the existence of marine interglacial formations at Neudeck in West Prussia, Giekie plainly reaches a false conclusion, for he assumes what has still to be proven, that the last ground moraine but one of Neudeck is of the same age as the last (uppermost) ground moraine south of the Baltic. Without this proof the marine layers of Neudeck have no demonstrative significance, especially as their underlying beds are not known, and no observations have been made as to the number of ground moraines beneath them. Granted, however, that Geikie's view is right, that the Baltic ter- minal moraine is the southern limit of a distinct glaciation, one cannot understand why each of the other terminal moraines of north Germany may not also represent the edge of the ice during a distinct glacial period. On this basis the so-called " last gla- cial epoch" would have to be divided into four if not five epochs, so that even the most fanatical advocate for as many glacial periods as possible would be terrified. I see a farther argument against Geikie's classification in the great difficulties of his comparison and in the inequality of the layers placed in the same stage. While with regard to the older ' I think Dr. Keilhack here uses " terminal moraine " in the German, not in the American sense (see p. 136 this number of this Journal). R. D. S. 122 K. KEILHACK glacial periods there is a satisfactory conformity between the ice expansions of Great Britain and those of north Europe and the Alps, such conformity is altogether wanting in Geikie's assumed fourth epoch. The Baltic glacier supposed by him to belong to the fourth period is still of such immense size, and is so little inferior to those of the earlier epochs, that it is almost impossible to correlate it with the valley glaciers of the Alps which Penck has described, and it seems to me with perfect right, as repre- senting only postglacial episodes. Hensen has regarded the similar deposits of Norway as epiglacial projections of existing glaciers. A graphic representation, such as that of Fig. i shows better than words can, the unnaturalness of Geikie's classification. If we suppose the expansion of the north European and Alpine glaciations to be expressed by lines which represent the extent of glaciation from the northern end of the Gulf of Bothnia, and from the central Alps respectively, and if the lines which repre- sent the greatest extremity of ice be represented by unity in both cases, we obtain the proportions expressed by the follow- ing figures : Alps, scale i : 3,400,000. North Europe, scale i : 23,500,000. I- I— 11. I— III. ! — IV, I— I. I 1 II. I — — I ni. I —1 IV. I 1 If, however, one considers the Baltic terminal moraine as well as all others which lie further south, only as stages in the retreat of a single ice-sheet, as the Prussian Geological Survey has done, and so unites Geikie's No. 3 and No. 4 into No. 3, all difficulties of comparison at once disappear, i, 2 and 3, in north NORTH EUROPEAN GLACIAL DEPOSITS 123. Europe correspond respectively with i, 2 and 3 in the Alps, and the small extension of ice in the fourth and fifth ice epochs, recog- nizable in the Alps, Norway, and Scotland, and of approximately equal extent, are represented in north Germany, not by the reap- pearance of a mcr dc glace, but by climatic depressions only. On the other hand I can agree with Geikie in his view that the principal glacial period (Saxonian stage) is the second gla- cial epoch, which had a predecessor in the so-called Scanian stage. The undeniable proof of three extensive glaciations of the Alps must awaken the suspicion that the north European glacial period also possesses a threefold division, and this sus- picion would be increased still more in the mind of the present writer, by a number of other phenomena. The reason, however, why the special surveying has so far produced no conclusive evidence of a pre-Saxonian glacial period, lies simply in the fact that the mapping has been confined almost exclusively to dis- tricts in which no ground moraines of the inland ice of the Scanian stage exists, but where the deposits of this epoch are almost exclusively fluvio-glacial, and these for the most part fine. It is known, morever, that such formations in north Ger- many found due appreciation much later than those in the Alps, and that today even eminent geologists, such as H. Credner, will not acknowledge that they possess any demonstrative power. What has led me to recognize in the lower sands and clays of the Diluvium of middle north Germany the fluvio-glacial equiv- alent of a glacial period older than that which deposited the lower bowlder clay of the Mark, is the fact that between these layers are to be found a flora and fauna which point to a mild, temperate climate like that of today, if not indeed a warmer one. Because, however, the underlying layers contain certain northern material, such as feldspar, fragments of bryozoans, flints, and occasionally even large bowlders, the conclusion is not to be gainsaid that the ice lay at no very great distance from the territories in which those northern sands were deposited. The superimposed layers, however, contain a forest vegetation with deciduous trees, a water vegetation with plants of southerly 124 K. KEILHACK character, such as Trapa natans, and even Cratopleura. Such, liowever, can never thrive in a flat country, if the latter is partly covered with glacier ice and the climate of that country arctic. Therefore a period with a warm climate must necessarily have existed between the deposition of the oldest northern sands and those of the ground moraine of the lower bowlder clay, and there- fore for north Germany a third oldest period must be assumed besides the two glacial periods which deposited the upper and lower bowlder clay. The ground moraines of this oldest glacial period I recognize not only in Schonen, but also in the deepest ground moraines of the Baltic range, especially in that part of it lying to the east of the Oder. On the other hand, no obser- vations have hitherto been made which point to an extension of these ground moraines in the district south of the range. If I attempt at the conclusion of these remarks, which I con- sider necessary for the verification and defense of the stand- point adopted in the official survey in north Germany, to give a classification of the north German diluvial deposits in tabular form, I beg that this attempt may be considered only as a pri- vate opinion, which I should like to submit to a wider circle, for criticism and examination. Preglacial. — Not yet determined with certainty. No deposits between the Miocene and the first glacial epoch certainly recog- nized. First glacial epoch. — Oldest ground moraines in the region of the east Balt>c Lake district. Fluvio-glacial formations, reach- inid Pliylogcny of Dcemonelix. By E. H. Barbour. Additional expeditions to the Utemonelix region have added new data showing the apparent steps in the phylogeny of this anomalous group. The simplest expression of Dcemonelix seems to be a fiber found in the sand rock, which shows unmistakable plant structure, and is, in every respect, like the fiber found in all the Daemonelix series. The author's present belief is that the various forms of the Daemonelix group result from the aggregation of these simple fibrous, fresh-water seaweeds into variously shaped bunches, clusters, and spirals. 2 24 RECENT P UBLICA TIONS Ascending to higher beds one comes next to Daemonelix "cakes," which are about the size of common camp griddle-cakes. So far as can be determined by the eye or the microscope they are nothing more nor less than colonies, or aggregations of Dsemonelix fibers. Vertical range twenty-five feet. Next above the "cakes" come the Daemonelix "balls," which resemble in size and shape the old New England codfish ball. These are but aggregations or bunches of Dsemonelix fibers. Vertical range about twenty-five feet. Next come slender forms of nearly vertical and somewhat spiral habit, called Daemonelix "cigars." The weathered and broken ends of these occur in immense abundance. They are about the size and length of an ordinary walking stick. These, too, are but aggregations of the simple Daemonelix fiber. Though practically confined to a range of twenty nor more feet they occur, in decreasing numbers, almost through the upper Daemonelix beds. The forms encountered next are frail and slender, scarcely thicker than the wrist, yet positively spiral and vertical in habit. They are viewed as the immediate progenitors of Daemonelix regular. They are but aggregations of the primitive Daemonelix fiber. Vertical range scarcely twenty feet. Above all, comes the " Devil's Corkscrew," the first forms of which are smaller, more regular, and more mathematically exact than are the larger and strangely modified forms characteristic of the topmost beds. Some are free spirals, some are fixed about an axis, some have no transverse trunk, others have one, two, or three ; others, called " twin screws," have, in each case, a large screw and transverse trunk, ending in a smaller reversed screw and trunk. These are thought to be the first complete specimens of Daemonelix. Continued studv makes it only the more apparent that these magnificent screws are but spiral aggregations of the simple Daemonelix fiber first encountered. More than one hundred micro-sections have been cut from all parts of all forms. Without exception all show precisely the same simple, cellular, non-vascular structure, to be likened only to sea- weed. Numerous photomicrographs of these have been made and are ready for publication. '^ RECENT PUBLICATIONS. — American Museum Natural History. — Bull., Vol. VIII, 327 pp. New York, 1896. RECENT P UBLICA TIONS 225 -Barbour, E. H. Deposits of volcanic ash in Nebraska. — Pub. V, Nebraska Acad. Sci., Proc, 1894-5, pp. 12-17. Diatomaceous deposits of Nebraska. — Pub. V, Nebraska Acad. Sci., Proc, 1894-5, pp. 18-23. Progress made in the study of Daemoneli.x. — Pub. V, Nebraska Acad. Sci., Proc, 1894-5, pp. 24-28. -Barton, George E. Evidence of the former extension of glacial action on the west coast of Greenland and in Labrador and Baffin Land. — Amer. Geol., XVIII, 379-384, 1896. Glacial origin of chan- nels on drumlins. — Bull. Geol. Soc. America, VI, 8-12, 1894. -Berghell, Hugo. Bidrag till Kannedomen om sodra Finlands Kvartara NivaforJindringar. — Bull. Comm. geol. de la Finlande, No. 5, 64 pp. Helsingfors, 1896. -Bertrand, Marcel. Etudes sur le bassin houiller du Nord et sur le Boulonnais. — Ann. des Mines, 1894, pp. 1-71, pl. X-XI. Paris, 1894. -Bittner, a. Geologisches aus dem Pielachthale nebst Bemerkungen liber die Gliederung der Alpinen Trias. — Verh. der k. k. Geol. Reich- sanstalt, 385-418. Wien, 1896. -Branner., J. C. Bibliography of clays and the ceramic arts. — Bull. U. S. Geol. Survey, No. 143, 114 pp., 1896. -Brigham, A. P. Glacial flood deposits in Chenango Valley. — Bull. Geol. Soc. America, VIII, 17-30, I pl. Rochester, 1897. -Calvin, Samuel. The Buchanan gravels ; an interglacial deposit in Buchanan county, Iowa. — Amer. Geol., Vol. XVII, pp. 76-78. Pl. IV-V. Minneapolis, 1896. -Chapman, Frederick. On some Pliocene Ostracoda from near Berke- ley. — Univ. Calif., Bull. Geol. Dept., Vol. II, No. 2, pp. 93-100, pl. III. Berkeley, 1896. -Club, Alpin Frana^is. — Bull. Men., No. i, Jan. 1897, 24 pp. Paris, 1897. -Comitato Geologico d'ltalia. — Bol., vol. ventiseesimo (6° dell 3^ serie), N. I a 4. Roma, 1895. -CusHiNG, H. P. On the existence of pre-Cambrian and post-Ordovician dikes in the Adirondacks. — Trans. N. Y. Acad. Sci., Vol. XV, pp. 248-252, 1896. -Dale, T. Nelson. Structural details in the Green Mountain region, and in eastern New York. — Sixteenth Ann. Rept. \J. S. Geol. Surv., pt. I, 543-570- Washington, 1896. -Dept. Mines and Agriculture, New South Wales. — Ann. Rept. 1895, 191 pp. Sidney, 1896. -Darton, N. H. Catalogue and index of contributions to North Ameri- can geology, 1732-1891. — Bull. U. S. Geol. Surv., 127, 1045 pp., 1896. -Dawson, William J. Additional notes on fossil sponges and other organic remains from the Quebec Group at Little Metis, etc. — Trans. Roy. Soc, Canada (2), II, Sec IV, 91-121, 4 pl. Ottawa, 1896. -Ells, R. W. Report on a portion of the Province of Quebec, etc. — Geol. Surv. Canada, Vol. VII, pt. J, 157 pp., map. Ottawa, 1896. -Fairbanks, Harold W. The geologv of Point Sal. — L^niv. Calif., Bull. Dept. Geol., ^•(ll. II, pp. 1-92, pl. f-III, Berkeley, 1896. 226 RECENT P UBLICA TIONS — Farr, Marcus S. Notes on the osteology of the White River horses. — Proc. Amer. Philos. Soc, Vol. XXV, pp. 147-175. Philadelphia, 1896. — -Frech, Fritz. Das Profil de Grossen Colorado-Canyon. — Neu. Jahr. f. Min., etc., Bd. II, 153-156, taf. Ill, 1895. Uber das Devon der Ostal- pen III. Die Fauna des Unterdevonischen Riffkalkes I. — Pam. 446- 479, pi. XXX-XXXVII. Berlin, 1894. Uber den Gebirgsbau der Radstadter Tauern. — Stzber. d. k. Preus. Akad. d. Wiss. z. Berlin, XLVI, 1-23, 1896. liber palaeozoische Faunen aus Asien und Nor- dafrika. — Neu. Jahr. f. Min., etc., Bd. II, 47-67, 1895. — Frech, Fritz und W. Dames. Uber unterdevonischen Korallen aus den Karnischen Alpen. — Zeit. d. Deutsch. geol. Gessell., 1 19-201, 1896. — Frosterus, Benj. Ueber einen neuen Kugelgranit von Kangasniemi in Finland. — Bui. de la Comm. geol. de la Finlande, No. 4, 38 pp., 2 pi. Helsingfors, 1896. • — Geological Society of Washington. — Presidential Address Constitution, etc., 60 pp. Washington, 1897. — Geological Survey of Georgia. — Administrative Report of State Geolo- gist, October 1894-October 1896, 45 pp., 4 pi. Atlanta, 1896. — Geological Survey of Canada. — Ann. Rept., N. S., Vol. VIII, 1894. Ottawa, 1896. — Gresley, W. S. Observations regarding the occurrence of anthracite, with a new theory of its origin. — Amer. Geol., Vol. XVIII, pp. 1-2 1, pi. I. Minneapolis, 1896. — HovEY, E. O. Catalogue of meteorites in the collection of the American Museum of Natural History to July i, i8g6. — Bull. Am. Mus. Nat. Hist., Vol. VIII, Art. VIII, pp. 149-155. New York, 1896. Notes on the artesian well sunk at Key West, Fla., in 1895. — Bull. Mus. Comp. Z06I., Geol. Ser. Ill, 65-91, 1896. — Howard, J. W. Good paving essential to the success of a city. — Reprint, Muncipal Eng. Mag., 8 pp. New York, 1896. — Iowa Academy of Science, Proc. 1895, Vol. Ill, 230 pp., 15 pi. Des Moines, 1896. Containing: The Le Claire limestone; The Buchanan gravels, etc., by Samuel Calvin; Recent discoveries of glacial scorings in southeastern Iowa ; Some facts brought to light by deep wells in Des Moines county, Iowa, by F. M. Fultz ; Recent developments in the Dubuque lead and zinc mines, by A. G. Leonard ; The area of slate near Nashua, N. H. ; Notes on the geology of the Boston basin, by J. L. Tilton ; Note on the nature of cone-in-cone ; Two remarkable cephalopods from the Upper Paleozoic, by Charles R. Keyes ; Variation in the position of the nodes on the axial segments of pygidium of a species of encrinurus, by William H. Norton ; A theory of the loess, by B. Shenick. — Iowa Geological Survey. — Administrative Reports, 1896, 31 pp., i pi. Des Moines, 1897. — Jentzsch, Alfred. Ber. ii. d. Verwalt. d. ostpreussischen Provinzial- museums Phys.-Okon. GeselL, 1893-5. Konigsberg i. Pr., 1896. — Keyes, Charles R. The Bethany limestone of the western interior coal field. — Amer. Jour. Sci. (4), Vol. II, pp. 221-225. New Haven, 1896- RECENT P UBLiCA TIONS 2 2 7 —Knight, W. C. The geology and technology of the Salt Creek oil fields. —School of Mines, Univ. Wyoming, Petroleum Ser.,Bull. i, pp. 1-22. Laramie, 1896. — Knowlton F. H. a new fossil Hepaticafrom the lower Yellowstone in Montana— Bull. Tor. Bot. Club, Vol. XXI, pp. 458-459- 21Q pi. 1894. Description of a new problematical plant from the lower Cretaceous of Arkansas.— Bull. Tor. Bot. Club, Vol. XXII, pp. 387-39°- 1895. Description of a supposed new species of fossil wood from Montana. Bull Tor. Bot. Club, Vol. XXIII, pp. 250,251, 271. pi. 1896. The Tertiary floras of the Yellowstone National Park.— Amer. Jour. Sci. (4), Vol. II, pp. 51-58. New Haven, 1896. — Krantz, F. W. Ueber thorhaltige Mineralien und ihre Bedentung flu- die Gasgliihlicht- Industrie.— Sitzber d. Niederrhein. Gesell. f. Natur. u. Heilkunde zu Bonn, pp .42-50, 1896. —La Commission Geologique de la Finlande : Beskrifning till Kartoladen 27-31. Helsingfors, 1895-6. — Le Conte, Joseph E. Elements of Geology, 4th ed., 670 pp. D. Appleton & Co. : New York, 1896. — Leonhard Richard und V^ilhelm Volz. Das Mittelschleische Erd- veben vom 11 Juni 1895.— Sonderabd. d. Jahrber d. Schles. Gesell. f. vaterl. Cultur, 1-71. Breslau, 1895. — Leverett Frank. Water resources of Illinois.— Seventeenth Ann. Kept., U'. S. Geol. Survey, Pt. II, 1-155. pl- CVIII-CXIII. Washmg- ton, 1896. — Lievre Daniel, Une Eruption Volcanique au Japan, Higashi Kiri- shima.— Bull, de la Soc. de Geog. Commerciale. 30 pp. Havre, 1896. — Maitland a. Gibb. The geological structure of extra-Australasian Artesian basins.— Proc. Royal Soc. Queensland, Vol. XII, 20 pp., 1896 —Marsh, O. C. Stylinodontia a suborder of Eocene Edentates.— Amer. Jour.'sci. (4), in, 137-146, 1897. — Margerie, Emm. de. Catalogue des Bibliographies Geologiques redi^e avec le concours des Membres de la Commission Bibliographie que^'du Congres.— Cong. Geol. Inter. 5= Ses. Washington, 1891. 733 PP- Paris, 1896. — McCalley, Henry. Valley regions of Alabama, Pt. I. The Tennes- see Valley region.— Geol. Surv. Alabama, E. A. Smith, state geologist, 436 pp., 9 pl. Montgomery, 1896, —Phillips, W. B. Iron making in Alabama.— Alabama Geol. Survey, pam. 164 pp. Montgomery, 1896. —Ramsay, Wilhelm. Till FrSgen om det senglaciala hafvets utbredning i sodra Finland.— Bull. Comm. geol. de la Finlande, No. 3, 44 PP- Helsingfors, 1896. Urtit ein basisches Endglied der Augit-syenit- Nephelin-syenit-Serie.— Geol. Foren. i Stockholm Forhandl., Bd. 18, hft. 6, 459-468, 1896. — RiES Heinrich. Pottery industry of the United States.— Seventeenth Ann. Rept., U. S. Geol. Surv., Pt. HI, 842-880, pl. XI, XII, 1896. — Ries Heinrich and Lea McJ. Luquer. "Augen" - gneiss area, pegmatite veins and diorite dikes at Bedford, N. Y.— Amer. Geol., XVIII, 239-261, 2 pl., 1896. 228 RECENT P UBLICA TIONS — Royal Geographical Society, Australia. — Journal, Vol. VI, No. i, 23 pp. Newtown, 1896. — RuCKER, A. W. Summary of the results of the recent magnetic survey of Great Britain and Ireland, conducted by Professors Riicker and Thorpe, Terrestrial Magnetism, Vol. I, No. 3, pp. 10, 105-146. Chicago, 1896. — Sederholm, J. J. Ueber einen metamorphosviten prsecambrischen Quarz-Porphyr von Karvia in der Provinz Abo. — Bull. Comm. geol. de la Finlande, No. 2, 16 pp. Helsingfors, 1895. — SCHWARZ, Ernest H. L. Coccoliths. — Ann. and Mag. Nat. Hist. (6), XIV, 341-346, 1894. Descent of the Octopoda. — Jour. Marine Z06I. and Microscopy, I, 87-92, 1894. Spirula peronii Lam. — Jour. Marine Z06I. and Microscopy, II, 26-30, 1895. The Aptychus. — Geol. Mag., December, IV, 454-459, 1894. — Schriften der physikalisch-okonomischen Gesellschaft zu Konigsberg in Pr., 1895. — Slosson, E. E. The analysis of the Salt Creek petroleum. — School of Mines, Univ. Wyoming, Petroleum Ser., Bull. i,pp. 22-47. Laramie, 1896. — Smith, E. A. Geol. Surv. of Alabama, Bull. V, 197 pp., 3 plates. Mont- gomery, 1896. Containing: Prelim, report on the upper gold belt, by W. M. Brewer ; Sup. notes on the most important varieties of meta- morphic or crystalline rocks, by E. A. Smith, Geo. W. Hawes, J. M. Clements, and A. H. Brooks. — Smith, George Otis. Geology of the Fox Islands, Maine. — Pam., 76 pp., I pL, map. Skowhegan, 1896. — Smith, J. Perrin. Classification of marine trias. — Journal Geology, Vol. IV, pp. 385-398. Chicago, 1896. — Steinman, Gustav. Die Spuren der letzten Eiszeit im hohen Schwartz- walde. — Separat-Abd. aus d. Freiburger Universitats-Festprogramm, pp. 187-226. Freiburg, 1896. — Taylor, Frank B. Correlation of Erie-Huron beaches with outlets and moraines in southeastern Michigan. — Bull. Geol. Soc. America, VIII, 31-58, I pi., 1897. — Utah University Quarterly, Vol. Tl, No. 2, pp. 73-136. Salt Lake, 1896. Containing : The great Salt Lake, past and present, J. E. Talmage ; some of the crystalline rocks of Salt Lake and Davis counties, Utah W. D. Neal. — Walcott, Charles D. Seventeenth Ann. Rept., Director U. S. Geol. Survey, 1895-6. — Administrative papers, 200 pp., 1896. — Weeks, F. B. Bibliography and index of North American geology, palaeontology, petrology, and mineralogy for 189I. — Bull. U. S. Geol. Surv., No. 135, 141 pp. Washington, 1896. — Whitney, Milton. Texture of some important soil formations. — U. S. Dept. Agri., Div. of Agri. Soils, Bull. 5, 22 pp., 34 pi. Washington, 1896. — Woodworth, J. B. Ice sheet in glacial Narrangansett Bay. — Amer. Geol., XVIII, 391-392, 1896. On the fracture system of joints, with remarks on certain fractures. — Proc. Boston Soc. Nat. Hist., XXVII, 163-183, 5 pi., 1896. Retreat of the ice sheet in the Narrangansett Bay region. — Amer. Geol., XVIII, 150-168, i pi., 1896. THE JOURNAL OF GEOLOGY APRIL-MAY, iSg-j GLACL\L STUDIES IN GREENLAND. X. The Bozvdoi/i glacier. — The Bowdoin glacier is a lobe of the great inland ice- sheet which descends from the north into the head of the valley of Bowdoin Bay. It is joined on the east by ice pushing in through the intervals between a row of nunataks that lie on the plateau border. On the west it is con- fluent with the trunks of the Tuktoo and Sun glaciers which debouch into the broad flat valley at the head of McCormick Bay. This is only separated from the Bowdoin Bay valley by nunataks of which the Sierra and the Sentinel are the most conspicuous. These lie in the lower part of the common valley and have already^^been mentioned in connection with the Tuktoo glacier (see sketch map, p. 668, Vol. III). No very precise point can be fixed upon as the place of parting of the Bowdoin glacier from the main ice-cap, but it may fairly be regarded as having a length of six or eight miles. In its lower half it is joined on the east by the Obelisk glacier which comes in below the North nunatak and by the East Branch glacier which pushes in more directly from the east on the south side of the Obelisk nunatak. The East Branch glacier appears to terminate by lat- eral wastage almost at the point of joining the Bowdoin glacier and apparently never becomes really confluent with it. But the Obelisk glacier blends with and becomes a part of the Bowdoin glacier without being distinguished from it even by a medial moraine. The Bowdoin glacier in its lower part has a breadth Vol. v., No. 3. 229 230 T. C. CHAMBERLIN of about two miles. It is the master glacier of the vicinity both in respect to dimensions and activity. The descent of its floor is apparently between 2000 and 3000 feet. This is accomplished by a somewhat steep fall from the Fig. 64. Transverse view of Bowdoin glacier seen from the heights on the east looking northwesterly. The glacial movement is from right to left. The dark longi- tudinal line represents the sunken medial moraine. The undulations of the surface are fairly well indicated. The slope just beyond the crevassed area in the right fore- ground and between that area and the medial moraine inclines backward, i. e., to the right, and a brook flows in that direction. The notable feature of the surface is the crevassing. The nunatak beyond the Bowdoin glacier on the right is the Sierra, that on the left the Sentinel. The Tuktoo glacier is seen between these and in the dis- tance a glimpse of the Sun glacier is obtained. The ice-capped heights beyond are a part of the border of Prudhoe Land. Lieutenant Peary's first route across the ice- cap lay along these heights. Photograph by Professor Wm. Libbey, Jr. upland to the valley, and then by a notable but unequal decline throughout its remaining course. The inequality of the decline is indicated not only by the undulation of its profile, but by intensive crevassing which distinguishes it from most of the val- ley glaciers of the vicinity. The crevassed condition is well shown in Fig. 64 and is worthy of thoughtful consideration when the relatively slight undulations and the relatively slow movement of the glacier are taken into consideration. Such pro- GLACIAL STUDIES IN GREENLAND 231 nounced crevassing emphasizes the slight ductility of the ice. The failure of the ice to fill up the inequalities and form a con- tinuously declining surface after the fashion of liquids is a fur- ther feature of interest. The undulations are so great as to Fig. 65. The end of Bovvdoin glacier. The head of Bowdoin Bay is seen at the right and the crevassed end of the glacier in the center and left. The dark line rep- resents the medial moraine. The absence of other debris is notable. A portion of the ice-cap and the Mirror glacier are dimly seen in the distance. (The sky has been a little penciled.) The rounded embossment on the right is Bartlett Mountain, about 2600 feet high, 'he name being specifically applied to the frontal promontory. Sugar Loaf is seen at the left of Mirror glacier. give rise to irregular surface drainage. In one instance at least, to which I was guided by Lieutenant Peary, the drainage is reversed. A considerable brook runs in a direc- tion opposite to that of the motion of the glacier. It flows northward while the tjlacier flows southward. The northward 232 T. C. CHAMBERLIN ' slope may be seen at the right hand of Fig. 64 a little this side of the medial moraine. The Bovvdoin glacier debouches at its lower extremity into the head of Bowdoin Bay and discharges icebergs of consider- able dimensions, Fig. 65. This discharge seems to vary very greatly with the nature of the summer season. My observa- tions happened to fall upon a season of relative quiescence.. From the sixth to the twenty-third of August I occupied, through the kindness of Lieutenant Peary, a lodge overlooking the sea wall of the glacier and so near at hand as to be within easy hearing distance of any notable disruptive action but dur- ing this time I was not fortunate enough to witness the dis- charge of a single' notable mass of ice, or even to hear it. A more substantial evidence- of quiescence is found in the fact that within less than 200 yards of the extremity of the glacier on the east side there was an ice bridge spanning the lateral stream and connecting the glacier with a spur of rock. Lieutenant Peary informed me that the bridge had remained intact for two years and had been used as a means of communication with the glacier whose vertical face at other points rendered access diffi- cult. It would be an error, however, to draw conclusions froni these facts, for during the succeeding season the bridge was not only broken away but extensive discharges of icebergs took place along the sea face of the glacier. I presume that this unequal and somewhat spasmodic action represents the habit of the glacier. It seems probable that during a succession of severe seasons the motion of the ice is quite slight and the extremity of the glacier makes but little advance. But an exceptionally warm season following such a succession of severe ones, during which thickening and steepening of the gradient may have taken place, probably causes a notable thrusting for- ward of the glacial foot accompanied by corresponding iceberg discharge. On its west side the Bowdoin glacier does not present an extensive vertical wall in accordance with the prevailing fashion of the region. In its upper part the glacier runs side by side GLACIAL STUDIES LY GREENLAND 233 with the trunk of the Tuktoo glacier, as already observed. Below this it rubs hard against the Sierra and Sentinel nunataks and its activity is such as to prevent the development of the fossae which often lie between nunataks and adjacent glaciers. Between the two nunataks named the glacier protrudes westward to meet the reciprpcal eastward protuberance of the Tuktoo glacier, and the two jointly impound two small triangular lakes in the angles between themselves and the nunataks. Where the protuber- ances are opposite to each other their extremities are not verti- cal, though steep and somewhat stepped. This is perhaps due to the absence of the reflecting frontal plane which I have thought might be one of the factors in the development of the ver- tical faces. Each glacier standing opposite to the other covers the ground which would otherwise act as a reflecting surface. On the east side of the Bowdoin glacier, however, there is the usual vertical face with its accompanying sharp triangular valley between the ice and the adjacent rock slope. This extends from the extremity of the glacier to the point of its junction with the East Branch glacier, a distance of perhaps three miles. At a few points spurs of rock close the valley. East Branch hill projects notably into the path of the glacier and not only closes the lateral valley but crowds the border of the ice out of its path. At these points of interruption the stream, which as usual runs between the glacier and the valley side, is forced to tunnel under the glacier. After a cold day or two, the water in the stream becomes low and the tunnels then afford a means of penetrating to limited distances beneath the ice. Fig. 66. The phenomena so disclosed, however, do not essentially differ from those seen to better advantage in the ver- tical sides. The amount of debris appears more scant because it is not concentrated by surface melting nor exaggerated by surface wash. In some instances where blocks of ice had recently fallen away from the roof, opportunity was given for the study of exceptionally fresh unweathered ice which had not been sub- jected to the sun's rays, except as they reached it through con- siderable depths of ice. Even here the ice is far from being 234 r. C. CHAMBERLIN really compact and transparent. It varies from translucency in the more dense layers to nearly complete opacity in the more porous layers. This is an expression of lack of solidity and may be noted as specially instructive since the ice is here near Fig. 66. Entrance to a tunnel under the side of Bowdoin glacier, looking down- stream. The shoulder of rock which caused the undercutting of the stream is par- tially shown at the left. The peculiar arching of the beds is an interesting feature. Photograph by Professor Wm. Libbey, Jr. the end of its career as a part of the glacier. As it was derived from the main ice-cap it may possibly have had as long a glacial history as often falls to such ice. The stratification and the basal loading of the ice exhibited along the extended vertical face are much the same as in the glaciers previously described, but differ somewhat in the fact that the layers of ice are more warped and contorted, a result doubtless of the greater inequalities of the glacier's bed. Very notable dips of the glacial layers, both up stream and down stream, were frequently observed and sometimes reached large GLACIAL STUDIES LY GREENLAND 235 angles, in one instance 70°. These dipping planes were some- tintes marked by debris (though more commonly not) showing that they were genuine planes of acquired stratification. But the distortion often went far beyond mere changes of dip, how- FiG. 67. Upthrusted and contorted ice layers on the east side of Bowdoin glacier just north of the East Branch hill. Lines of shearing, sharp flexures, localized belts of contortion and thrust, are among the notable features. ever great or abrupt. It amounted to crumpling and even to faulting. A special instance is illustrated in Fig. 67. The debris does not rise so high in the Bowdoin glacier as in some of the others already described. This fact is perhaps worthy of special note as this is a tongue of the great ice cap and has descended over an undulatory bottom nearly to the sea level. It is perhaps further worthy of note because this glacier discharges into the sea. The basal layers are not forced 236 T. C. CHAMBERLIN upwards at their extremities to the same degree that was observed in some glaciers that have less free termination. A measurement of the upper limit of the debris-bearing la3^ers at a point about two miles distant from the end of the glacier gave a height of only twenty-eight feet. Between this point and the extremity of the glacier the debris was usually confined to less heights, and at some points near the end it was not visible at all in the lateral face of the glacier. The amount of debris even within the limited range indicated was often relatively small and never very heavy. The accompanying photographs sub- stantiate this. The bowlders of this glacier were usually more rounded than those previously described, and this rounding was of such a nature as to imply very considerable wear. The debris was composed wholly of crystalline rock, so far as observed, no admixture of clastic material being seen. The considerable rounding of the material, the small amount of debris and its low position at the base of the glacier are facts Worthy of special note. They will be seen to have special sig- nificance when it is remembered that this is one of the larger (though not by any means the largest) tongues of the great ice-cap. . The rubbing of the glacier against shoulders of rock project- ing from the side of the valley gave opportunity for observing some of the special phenomena of such situations. At one point the process of "plucking" was well indicated (though not actually observed) on the lee slope of a spur of gneissoid rock. Blocks ranging up to three or four feet in width and length and one or two feet in thickness had been detached in considerable numbers. The process involved much breaking and bruising with relatively little wear. Corners and angles were broken off and heavy bruise marks were observed both on the blocks and on the sides and edges of the cavities from which they had been removed. At some points considerable^ crushed rock was observed. On the other hand, systematic grooves and striae were not abundant nor pronounced. The dynamic impression given was that of a forceful tearing out of blocks by the action of a GLACIAL SrC'DlES LV GREENLAND 237 relatively rigid agency . which did not press the blocks hard upon the lee slope after their removal. At the point where the glacier grinds against the projecting side of the East Branch hill, the ice was notably crevassed, the fractures being most open at the side and gradually closing as thev passed away toward the axis of the glacier. They were directed obliquely up stream according to the general law of lateral crevasses. Immediately against the obstructing projec- tion the ice was broken up into numerous blocks and fragments and somewhat piled up against the side of the hill or tumbled back upon the upper surface of the glacier. The same thing occurs on the opposite side where the glacier rubs hard against the Sierra nunatak. just north of the point of forceful contact with the East Branch hill, a very interesting instance of contortion and upward thrust was observed. The accompanying figure (67) shows the nature of this better than any verbal description. It may be observed that near the base there are two quite pro- nounced lines which have the appearance of special shear planes. The contortion of the central belt in being forced obli(|uely upward is well displayed in the photograph. While the cause of the phenomena cannot be positively stated, it associated itself in my mind with the obtrusion of the projecting point of East Branch hill immediately below it. Whether this be the cause or not, the phenomena impress me as a clear case of forceful localized thrust with resulting foliation and shear planes. At the point where the East Branch glacier joins the Bow- doin glacier, a large amount of debris had accumulated from a medial moraine derived from the Obelisk nunatak lying between the East Branch and the Obelisk glacier. The accumulated rubbish was the terminal dump of the medial moraine. There appeared to be much ice beneath this accumulated debris. It was obvious that both it and the extremity of the East Branch glacier had become essentially stationary. The more active Bowdoin glacier (or the Obelisk glacier, for the two are con- 238 T. C. CHAMBERLIN tinuous) had forced itself over the edge of this debris. The result was the bending upwards of the edges of the layers of the Bowdoin glacier. Accompanying this there was a phenomenon which seemed at the time to be a clear and unequivocal expres- sion of the shearing of these layers over each other, due to the resistance of the morainic material. Subsequent studies, as else- where remarked, led to some skepticism as to the validity of this interpretation. This point was not visited after the skep- tical spirit arose, and hence I am only able to reproduce the observations and interpretations as they impressed themselves at the time. On the nearly vertical face of the glacier where it was most strained in being thrust over the morainic mass, there were as many as nine distinct projections of upper layers over under layers. The over-projection was sharp and definite and at once attracted the notice of my companion, Mr. E. B. Bald- win, who called my attention to it. The ground for skepticism regarding the interpretation of phenomena of this class is the presence of debris bands which may give rise to differential melting. There is no question that projections of a kind much like those here observed are due to such unequal melting. It seems to be demonstrable also that overthrust shearing takes place in this way. Each case must therefore be judged by itself. The limited amount of debris in this glacier encourages less skepticism regarding overthrust here than in some other cases. My notes state that pebbles and bowlders were some- times lodged between these layers and that they appeared to have been pushed along the face of the layers in the shearing process, sometimes making grooves in the ice. Both notes and memory imply scantiness of debris between the layers. Unfor- tunately my photographs at this point were not a success. The overjutting of the layers ranged from a few inches to eighteen inches. They were most pronounced where the strain resulting from the obstruction would naturally be greatest and they died away in either direction. At another point somewhat farther on the phenomenon was repeated, but here there were but five chief planes of displacement in the vertical section. The maxi- GLACIAL STUDIES IN GREENLAND 239 mum displacement of two adjacent layers, however, exceeded two feet. The adjacent faces appeared to be very distinctly fluted by the movement. Skepticism with reference to the interpretation of the fluting was subsequently raised on the Fig. 68. Sunken medial moraine on the back of Bowdoin glacier ; seen from a point near the moulin looking northerly. The ice motion is toward the foreground. The width of the moraine is about four rods and its depth below the general surface ten or fifteen feet. The stream's trench is shown at the left. It falls into the moulin at the lower right-hand corner. Glimpses of the main ice-cap are caught beyond the heights. ground that it might perhaps be due to unequal melting by water trickling down the face of the ice, but, as elsewhere noted, in one instance at least, the debris in the junction plane was cor- rugated. It was noted here that a shearing plane was offset ; that is, the shearing plane passed from the contact between two given planes to a similar contact between two planes below. The offset was downward and to the right or up-stream side. The overthrust was about six inches and the offsets, two of which were sketched, about ciijht inches. 240 T. C. CHAMBERLIN The back of Bowdoin glacier is marked by an interesting medial moraine that takes its rise far up toward its junction with the main ice-cap, Fig. 68. I did not see its origin, but Lieu- tenant Peary informed me that it came to the surface from beneath the ice at a point north of the north nunatak, but south of the descent of the ice from the main plateau. On account of this he could not be certain whether it was derived from a nunatak that lies still farther north on the border of the main plateau, or from some concealed embossment. Undoubtedly the material was picked up from a prommence on or near the edge of the plateau. The debris is so thinly spread that it catches the sunlight and by conversion and conduction makes its heat available to the ice. As a result the moraine is sunken ten or fifteen feet below the surrounding surface. In breadth and form it is not unlike a depressed street strewn with coarse debris. On one ■ side of it a stream has cut a notable trench through which a considerable body of water flowed during the warm days of August. Fig. 68 illustrates the moraine and its depressed condition, as also the channel on the west side. The latter terminates in a moulin at the right-hand lower corner of the illustration. The moraine is also illustrated in the general transverse view of the glacier shown in Fig. 64. T. C. Chamberlin. ITALIAN PETROLOGICAL STUDIES. ]V. THE ROCCA MONFINA REGION. The volcano of Rocca Monfina, was long celebrated as one of the classical examples brought forward in favor of von Buch's hypothesis of Craters of Elevation. Its historical interest is equaled by its interest from a petrological point of view, since, apart from the character of the rocks of which it is built up, their order of succession forms an exception to that met with elsewhere among the Italian volcanoes. My visit to the locality was extremely brief and was confined to a trip from Teano on the eastern border, past the central mass of Mte. Sta. Croce to Conca, with one or two short excursions. I must therefore rely for most of my knowledge of the volcano on the work of others — a resource which in this case is somewhat meager. Rocca Monfina has been in fact one of the most neglected of Italian volcanoes. The early works, chief among which are those of Abich^ (who gives views and a map), Dauben}-^ and Pilla 3 are devoted to an exposition of their views in favor of von Buch's theory and the examination of the volcano from this standpoint. They are therefore chiefl}' of historical value, though Abich gives some descriptions of the rocks. Vom Rath'' devotes a few pages of Part IV of his " Italian Fragments" to a description and analyses of two of the rocks, which we shall have occasion to notice later. A few others are briefly described by J. Roth, 5 and Bucca's descriptions are quoted by Roienbusch. The only modern writers who deal at all fully with the region 'Abich, Vulk. Ersch. der Erde. Braunschweig, 1S41. ^ Daubkny. Volcanoes, London, 1848. 174. 3PILLA, cf. Neu. Jahrb. 1845, 843, and references in Moderni. •> Vom Rath, Zeit. d. d. geol. Ges. 243, 1873. ? Roth, Geologic, IT, 245, 275, 354. 241 242 HENRY S. WASHINGTON are Bucca' and Moderni.^ The former gives petrographical des- criptions of the rocks collected by Moderni ; and the latter describes the region from a geological point of view, giving a quite detailed geological map on a scale of 1:100,000. Deecke 3 also touches upon the tuffs of the region in an article on the tuffs of Campania. Topography. — The volcano of Rocca Monfina lies about 70 kilometers northwest of Naples, west of the main railroad line from Naples to Rome. In the center is the trachy-andesitic dome of Mte. Sta. Croce, 1005 meters above sea level, with the smaller similar dome of Mte. Lattani (847 meters) adjoining it to the northeast. These are steep well-wooded hills, made up of solid masses of eruptive rock, with no tuffs nor signs of separate lava flows. At the eastern foot of Mte. Sta. Croce is the small town of Rocca Monfina, and extending around the two domes, especially to the east and south, is the plain of the same name. This plain (from 500 to 650 meters above sea level) is bounded by the remains of a large crater ring, which is almost circular in form and with a diameter of about 6 kilometers. At the west and south the ridge, which is called Mte. Corti- nelli, is quite high and sharp, reaching its maximum elevation on the west at Mte. la Frascara (926 meters). Its inner slope is very steep, while on the exterior it is very gentle, between 6° and 10^. On the northeast and southeast the plain is bounded by a series of hills which become less high as they are farther from the center. This part of the crater ring has suffered much from erosion, being furrowed by many deep radiating ravines. In regard to the age of the volcano Moderni says that little can be stated positively, except that at various places volcanic material is seen resting on Mid-'^ ^.ocene beds. Deecke notes the superposition of Rocca K^nhna tuffs on the gray Campanian tuffs. It is probable that like the other volcanoes previously ^BuccA, Boll. Com. Geol. Ital., 1886, 245. ^Moderni, Boll. Com. Geol. Ital., 1887, 74. 3 Deecke, Neu. Jahrb. 1893, I, 62, 65. ITALIAN PETROLOGICAL SKETCHES 243 described the eruptions date from the Pliocene. The date of extinction of the volcano is likewise uncertain, though probably quite recent. Some lead objects and ancient buildings, with frescoes have been found beneath the tuffs, and the date 269 B.C. has been proposed for this period — on what authority I am unable to state as Moderni's reference is inaccessible to me. Moderni however, taking into consideration the silence of the ancient writers, the want of exact data, and the liability of objects being buried by the easily transported tuff, rejects this view and leaves the question an open one.' The hot springs of Sujo '^ are the last S3'mptoms of volcanic activity manifest at the present day. Petrography. — Moderni recognizes three phases of activity in the volcano, distinguished by different products of eruption: (i) the leucitic, the oldest, which is subdivided into two sub-phases characterized by leucitites and leucite-tephrites ; ( 2 ) the trachy tic ; (3) the basaltic, which is the youngest. As my own observa- tions were not extended enough to enable me to judge of this point for myself I shall accept Moderni's views. The volcano is thus seen to form a notable exception to most of the composite volcanoes of Italy — the order of succession of the leucitic and non-leucitic rocks being reversed. Before entering on the petro- graphical descriptions it will be as well to give a brief resume of Moderni's work. While it is difficult to unravel certain portions of the history of the volcano one fact is certain, that the first lava erupted was leucitite. This is shown by the fact that it is found beneath all the others. The flows of this rock are the greatest in amount of any of those of the region. They form a large part of the ridge of Mte. Cortinelli, with flows at many points elsewhere round the crater ring which it is unnecessary to enumerate here. There are certain eruptive centers on "^^ - flanks of the volcano which gave vent to this type of rock, but ix. j.st flows of the western half seem to have been poured out of the central crater. Lying above ' This is also tlie conclusion to which Daubeny comes in the second edition of his Volcanoes, London, 1848, p. 178. - Johnston-Lavis, South Italian Volcanoes, Naples, 1891, 73. 244 HENRY S. WASHINGTON the leucitites, and covering almost all the western part of the cone, are leucite-tephrites, whose amount, while great, is not equal to that of the leucitite. These also appear as small flows in the northern and eastern parts. Belonging to th'is period are certain tuffs, the most extensive of these being a gray lithoidal tuff which covers the flanks and extends for a great distance to the north, east, and south of the volcano. Moderni estimates that the volcano at the close of this period had an altitude of nearly 3000 meters. The second or trachytic phase was inaugurated by the blow- ing out of the large central cavity, which involved the destruc- tion of the greater part of the northern and eastern wall. It was during the first part of this phase that the double dome of Mti. Sta. Croce and Lattani was formed. These form, as already observed, a homogeneous compact mass, and are a good example of a "domal" eruption of a pasty magma which was not suf- ficiently fluid to flow to any great distance. The material of this eruption was of a trachy-andesite approaching the vulsinites already described, while the other eruptions of this period were of a more acid and more nearly trachytic rock and were all flank eruptions. While stating that it is impossible to decide definitely whether the central domal or the flank eruption took place first, Moderni thinks it more probable that the former was the earlier. He bases this conclusion on the ground that the large domal eruption choked up the main central vent, so that any further ejection of volcanic material had to take place through the flanks of the volcano. It is needless to mention in detail the various trachytic flows, which moreover are not nearly as large in amount as the flows of the first phase. With the trachyte there was also erupted a small quantity of leucitic rock. Several areas of tra- chytic tuffs also belong to this phase. The third and last phase of activity was characterized by the eruption of basaltic rocks in which leucite is entirely wanting. These eruptions were of small amount, the quantity of basalt being even less than that of the trachytes. They formed a num- ber of small parasitic cones, the lava poured out being often ITALIAN I'ETROLOGICAL SKETCHES 245 scoriaceous. Most of these small cones were formed on the outer flanks of the volcano, but three are found in the interior, one being in the valley of Pratolungo between Mte. Sta. Croce and Mte. Lattani. The tuffs belonging to this phase are few and of small amount. That the eruption of the "basalts " was posterior to those of the leucitites and trachytes is made evident by sev- eral facts. The small cones formed by them are the best pre- served of all, in many cases quite intact and covered with lapilli and bombs and with their lava flows quite bare. At Sipicciano on the northwest flank of Mte. Cortinelli a basaltic flow covers the leucitites and leucite-tephrites ; while in the central eruption at Pratolungo the flow of "basalt" is seen to be superposed on the trachy-andesite of Mte. Lattani. Leucitite. — This is represented by only one specimen in my collection, from a flow below Preta, on the outer northern side of the volcano. This shows a very compact, fine grained, dark gray groundmass, through which are scattered many small but quite glassy leucites, with rare specks of augite. In thin sections these large leucites are seen to be perfectly clear, and they show very weak double refraction for such large fresh crystals. Over the greater part of their area they carry almost no inclusions, but at the edge of almost all is a very narrow peripheral line of small augite microlites, which in turn is surrounded by a narrow mantle of leucite. This shades into the groundmass, and is quite anal- ogous to the alkali feldspar mantles so frequently noticed in the preceding papers. Apart from the leucites only rare light green augite phenocrysts are seen, which also carry no inclusions. The groundmass is fine grained and composed essentially of small leucites with interestitial greenish gray augite prisms and anhedra and a little magnetite. These small crystals are imbed- ded in a colorless isotropic substance. It is possible that this is o-lass, but, judging from the leucitic mantles around the leucite phenocrysts and from the fact that there is no difference in refractive index between this base and the small leucites, it seems probable that it is leucite. It would thus correspond to the nephe- line base of certain phonolites and other nepheline rocks, such 246 HENRY S. IVASHINGTOA as have been noticed previously. A very few flakes of alkali feldspar are also met with. Leucite-tephrite. — -One specimen of this rock was collected at Mte. San. Antonio (707 meters high), which forms one of the girdle of hills around the plain and lies due north of Mte. Lat- tani. The rock shows a fine grained gray groundmass in which, with the exception of very rare augites, the only phenocrysts visible are leucites. These are extremely abundant, making up a large portion of the rock. They are apparently fresh, with a waxy luster ; the smaller ones pale gray, while the larger show a core of dark gray and an outer shell of light gray. This rock is extremely tough. In thin sections the large leucites are very prominent. They are clear and show strong double refraction. Inclusions are abundant, of small crystals of anorthite and green augite. They include also many small slender needles of diop- side, minute grains of magnetite and spots of glass, which are very commonly clustered toward the center of the crystal, thus accounting for the dark gray cores. Beside the leucites only one or two dark green augite phenocrysts were seen in the slides. The groundmass is composed largely of irregularly shaped greenish yellow augite grains, a few small leucites, considerable magnetite, many small crystals of a plagioclase which is shown by its extinction angles to be anorthite, and some small flakes and laths of alkali feldspar. A little residual glass is also present. A peculiar leucite-tephrite is that, from what appeared to be a small flow on the road to Conca, below Orchi. It is very dark gray and fine grained, but rather rough in texture. A few pheno- crysts of augite and leucite are visible. Under the microscope it shows the doleritic structure which is so common to the Italian leucite-tephrites. None of the phenocrysts were met with in the single section which was made of this rock. In the groundmass stout columnar, pale green augites are very abundant, the largest showing cores of darker green ; while these are clear and contain few inclusions (of magnetite and glass), their outlines are very irregular, most of the prisms showing deep embayments due to corrosion. A little brown barkevikitic amphibole and a few ITALIAN PEIROLOGICAL SKETCHES 247 flakes of brown biotite are also present. There are also numer- ous grains of magnetite and a few apatite needles. Leucite is present in abundance, but shows an unusual habit. While often in the form of small, rounded or trapezohedral crys- tals, it generally occurs in irregularly shaped masses, occasionally showing crystal planes here and there, or with small, definitely shaped crystals attached to the larger mass. These leucites are extremely clear, and show a remarkably faint double refraction. Indeed so faint is the action of these leucites on polarized light, that in most of the crystals recourse must be had to the quartz wedge before any such action can be perceived, a proceeding not often necessary in the Italian leucitic rocks. Later than the leucites, and often enclosing them micropoi- kilitically,' are flakes and large patches of a plagioclase with generally well developed twinning lamellae, whose high sym- metrical extinction angles refer it to anorthite. There are also some patches of a feldspar, which, though not showing any signs of twinning, may also be referred to anorthite on the ground of the equality of its refraction with that of the striated anorthite. There is also considerable colorless base, which, while in general isotropic, shows in places a very faint double refraction. Since treatment of the powdered rock with acids produces abundant gelatinous silica, this base may be regarded as nepheline, or at least as a glass corresponding to nepheline in chemical com- position. An analysis of this rock made for me by Dr. A. Roling of Leipzig is here inserted : SiOo 47.40 AI2O3 19.84 Fe,03 2.72 FeO 4.40 MgO 4-23 CaO 9.88 NaoO 2.93 KoO 5.91 HoO 1.66 Tib. 0.30 99.27 ' The marked difference in refractive indices between the leucite and anorthite serve to bring the structure out very nicely in ordinary light, while the dark spots of 248 HENRY S. WASHINGTON t A leucite-tephrite from Mte. San Antonio is described by vom Rath,' which seems to be more like the Conca tephrite than the other just described. Megascopically only few phenocrysts of augite, leucite and feldspar are visible, while under the micro- scope the groundmass resembles that of many leucitites already described. Sanidine phenocrysts are few, but smaller ones of plagioclose are abundant. Vom Rath's analysis is inserted here : SiO^ 58.48 AI203 19.56 FeO 4.99 MgO 0-53 CaO 2.60 NasO 3-14 K2O 10.47 H2O 0.24 100.01 Leucite -trachyte. — The occurrences of this rock which I observed seemed to belong to the second (trachytic) phase of the volcano, though on this point I cannot speak with certainty. My specimens may be referred to two different varieties, an augitic and a biotitic. To the former belongs the rock of a well marked flow at Acqua Rotta, about a kilometer and a half north- west of Teano. The groundmass of this is very compact and fine grained, and of a dark gray color. Leucite phenocrysts are not abundant, but many small black augites and a few flakes of biotite are visible. Under the microscope a few large leucite sections met with are clear, with quite strong double refraction, and show few inclusions. The pale green augite phenocrysts are clear, carry inclusions of magnetite, especially toward the edges, and generally show corroded outlines. The biotites present are all altered to an almost opaque, finely granular mass of augite and magnetite, generally to such an extent that but little of the original mineral remains. The groundmass resembles that of many leucitites, being composed largely of small round leucites, leucite show out with striking clearness against the background of bright anorthite between crossed nicols. ' Vom Rath, op. cit. 243. ITALIAN PETROLOGICAL SKETCHES 249 with interstitial prisms of green augite, many magnetite grains, and flakes and laths of alkali feldspar. Some larger crystals of anorthite are also seen. No glass seems to be present. To the biotite-leucite-trachytes belong specimens from Tuoro west of Acqua Rotta, and from below Orchi on the northern flank of the volcano. Their groundmass is lighter gray than that of the preceding, small leucite and augite phenocrysts are not very common, but there is an abundance of flakes of biotite of a dark brown color and bronzy luster. In thin section they very much resemble the preceding. Augite phenocrysts are, however, rare, their place being taken by biotite. In the rock from Tuoro its color is of a deep orange red, while in that from Orchi it is much paler. It is much altered at the edges to the usual fine (rrained aggregate. In the groundmass of the former the alkali feldspar is in the form of laths, in the latter it is rather in the form of flakes and interstitial cement. Some basic plagioclase is present, but in small amount. A rock which occupies a position between the above and the following group is represented by a specimen also from below Orchi. Megascopically this closely resembles the other, the only important difference being that here leucite phenocrysts are entirely wanting, and replaced by glassy sanidines. There is also considerable augite along with the biotite. The rock in thin section resembles more one of the vulsinites to be described presently than a leucite-trachyte. The sanidine phenocrysts are rarely met with. Large greenish gray augites are not uncommon, which carry inclusions of brown glasS and magnetite and are somewhat corroded. The biotites are all profoundly altered, usually so much so that the crystal is almost completely dis- inteo-rated. The groundmass is made up of flakes and laths of an alkali feldspar, with a considerable number of laths and basic plagioclase which are usually surrounded by mantles of orienta- ted alkali feldspar. Small grayish green irregularly shaped and corroded augite crystals with magnetite grains are common. There are also present many small round spots of leucite, recog- nized by their characteristic inclusions and by their analogies 2 50 HENRYS. WASHINGTON with other occurrences. A scanty glassy base is also present. No nepheline was detected in any of the above rocks, Biotite-vulsiiiite. — In the descriptions of the volcanic centers which form the subject of the preceding papers we have had occasion to examine certain members of a group of effusive rocks which, both chemically and mineralogically, stand in a position intermediate between the trachytes and the andesites. The Rocca Monfina region resembles the others in furnishing — and very prominently — members of this group. We find here in fact effusive rocks showing the mineral combination of alkali feld- spar and basic plagioclase, which approximate closely to the ciminite of the Viterbo region in chemical composition. Miner- alogically, however, they come closer to vulsinite of Lake Bol- sena, no olivine being present, but augite and especially biotite being abundant representatives of the ferromagnesian minerals. So much indeed do they resemble vulsinite that they were con- sidered at first to belong to this group, forming a species which would be called a biotite-vulsinite, as was briefly noticed in the paper on the Bolsena Region.^ A chemical analysis, however, which I completed after the printing of that article and which will be found on page 252, shows that this determination is not quite correct. It will be seen that the silica is notably lower than in the vulsinites, and furthermore that lime, magnesia and iron are very much higher, while the alkalies are considerably lower. Comparison with the analysis of the Viterbo ciminite (which is inserted for convenience) will show that the trio are chemically almost identical. We have then a rock which is chem- ically a ciminite and mineralogically, a biotite-vulsinite. In regard to the name by which they should be called there may be some doubt. From a mineralogical standpoint they are obviously not ciminite, nor chemically can they strictly be called vulsinite. Since, however, in the schemes of classifica- tion in general use at the present time, the mineralogical com- position takes precedence over the chemical, and bearing in mind the unadvisability of adding new names to the already 'Jour, of Geol., IV, pp. 551-553, 1896. ITALIAN PETROLOGICAL SKETCHES 25 I overburdened nomenclature, I shall designate these rocks as biotite- V ulsi/i itc . The most important occurrence of biotite-vulsinite is that of Mti. Sta. Croce and Lattani, which are, as far as can be seen, entirely formed by a domal eruption of the rock just noted. This was called " trachy-dolerite " by Abich,' he considering it a link intermediate between trachyte and dolerite, while Pilla and von Rath consider it a trachyte. Bucca calls it an augite-andesite, and refers the greater part, if not all, of the feldspar to plagio- clase. Otherwise his description agrees closely with mine. The Santa Croce rock is trach3?tic in appearance, having a fine grained, but not very compact, light gray groundmass of a slightly reddish tone. Through this are scattered very many small, dark brown flakes of biotite, some small augites, and a few small feldspars. The groundmass of the Mte. Lattani rock is rather darker and of a bluish gray color, more compact, biotite flakes are less abundant, and feldspars more common. My specimen of this latter is not quite fresh. In thin section the feldspar phenocr3^sts are quite common. They consist of an alkali feldspar, apparently a soda-orthoclase, and a larger num- ber of plagioclase, which is shown by its extinction angles to have the composition AbgAng. These are well shaped, but con- tain many inclusions of glass with small opaque grains and aug- ite microlites. The biotites are of a greenish brown color, quite fresh in the interior, but surrounded by a thin alteration border of fine augite and magnetite grains. The large, irregularly shaped augites are pale green. The groundmass is trachytic in character and shows well-marked flow structure. There is an abundance of small laths, chiefly of alkali feldspar, with some of plagioclase. Magnetite grains and small diopside microlites are present in abundance, and there is a residue base of alkali feld- spar. The rock of Mte. Lattani closely resembles the above, though there are certain differences, the rather more abundant feldspar- phenocrysts are identical, as is also the augite. The biotites are, ' Abich, Vulk. Ersch., Brunswich, 1841, 100. 252 HENRY S. WASHINGTON however, all completely, or almost completely, altered, the prod- uct being rather coarsely granular. The grains are very slightly coherent with one another, and show in a very striking way the distribution of such augite and magnetite grains through the groundmass of certain rocks, as has been already pointed out.^ The groundmass is holocrystalline, and composed largely of flakes of alkali feldspar, with few laths. The augite and mag- netite grains resulting from the disintegration of the biotite are very abundant. Bucca refers this Mte. Lattani rock to the trachytes, though, as will presently be seen, he is struck by their andeistic character. An analysis of the rock of Mte. Santa Croce (I) made by myself is given below, II being an analysis of the same occur- rence by vom Rath, ^ and III being an analysis of the ciminite of Fontenu Fiesole, near Viterbo,^ inserted for convenient com- parison. Abich (p. 114) gives the silica percentage as 57.41 and the specific gravity as 2.79. I II III SiOa 55-69 55-08 55-44 AI3O, 19.08 17-25 18.60 Fe^Oa 4.07 2.09 FeO 3.26 9-33 4.48 MgO 3-41 2-77 4-75 CaO 6.87 7-34 6.76 NaoO 2.89 1.86 1-79 K3O 4.41 5-32 ,6.63 HgO 0.17 0.17 0.25 TiOo tr. 0.16 P2O5 ti. MnO tr. 99-85 99,12 100.75 Sp. gr. 2.713 Belonging to the same group is a specimen from a flow met with in the valley east of Casi, to the west of Teano. This is a ' Cf. Jour, of Geol., IV, 270, 1896. = Vom Rath, Zeit. d. d. geol. Ges., XXV, 245, 1873. ^Yi. S. V^ASHINGTON, JOUR. OF GEOL., IV, , 1896. / TA LI A N PE TR OL O GICA L SKE TCHES 253 medium gray compact rock showing small biotitc, augite, and feldspar phenocrysts. In thin sections the phenocrysts of alkali feldspar and basic plagioclase are present in about equal amounts. They contain many glass inclusions. The biotite phenocrysts are all deeply altered, and a few large augites met with show no noteworthy features. The groundmass is trachytic and com- posed of flakes and laths of alkali feldspar, with laths of plagio- clase, diopside needles, and magnetite grains. There is little or no glass present. Rocks which represent transitional forms between the above and true trachytes are found in specimens from the ravine at Molino di Casa Fredda, from a flow northwest of Tuoro, near Teano, and from a loose block in the valley of Casi. They resemble the rocks already described, but plagioclase is much less abundant, and in some almost entirely wanting. The groundmass of the Casi rock is much darker and finer grained than in the other cases ; feldspar phenocrysts are abundant, but few of ferromagnesian minerals are seen. Some good examples of microperthite seen in it indicate that the alkali feldspar con- tains considerable soda. In the specimen from Tuoro a note- worthy feature is the presence in one slide of a single, large, well-shaped crystal of colorless diallage, showing its character- istic dusty inclusions and parting parallel to the orthopinacoid These, as well as other rocks similar to these, are all described by Bucca under the head of trachytes, his descriptions agreeing very closely with the characters of my specimens. He closes, however, with the significant remark (p. 257) that they "approach so closely to the andesites that I was tempted sev- eral times to refer them to these." Trachyte. — A rock from a small quarry on the road to Conca, on the northern outer flank of the volcano, belongs to the trachytes proper rather than the vulsinites. It is the same as a rock described by Bucca (p. 255), whose description agrees very well with the characters of my specimens. The main mass of the rock is compact and light gray, showing few phenocrysts of augite and feldspar. Scattered through this are rounded 2 54 HENRY S. WASHINGTON black masses, which often attain diameters of five centimeters. These are coarsely crystalline and not very compact, with miarolitic and vesicular cavities. Under the microscope the only phenocrysts visible are of pyroxene. These are usually well formed, but occasionally quite fragmentary. The interior con- sists of a colorless diopside, which is uniformly surrounded by a rather narrow border of yellowish green non-pleochroic augite; The substance of this border corresponds exactly with that of the small groundmass augites, so that it is due to the same late period of growth to which they belong. The large augites are hence a good instance of the growth and enlargement of crystals brought up from below by the late accretion of isomorphous sub- stance of somewhat different composition. Apart from the not very abundant small augites and magnetites, the groundmass is made up of feldspar and some residual glass. The feldspar forms laths with somewhat ragged edges, and they, with the augites, show well-marked flow structure. The greater part of these laths are of orthoclase, or at least an alkali feldspar ; only very few showing twinning striations and extinctions which would refer them to plagioclase. No suitable sections of these latter were found by which their approximate composition could be determined, though I am inclined to think that they belong to the middle of the plagioclase series rather than to the basic end. The powdered rock, on treatment with acids, furnishes abundant gelatinous silica, which probably comes from the glass base. The dark spots referred to above are seen under the microscope to be holocrystalline, and composed of a few large diopsides with augite borders and very many smaller prismatic augites lying in a cement of orthoclase or alkali feldspar. The line of junction with the surrounding rock is quite sharp. They are true segregations and are to be classed with the enclaves homcBoghies of Lacroix.^ Bucca also mentions briefly the rock of Mte. Ofelio, near Sessa, on the southwest flank of the volcano, as a true trachyte. It seems to be much decomposed. ' Lacroix, Enclaves des Roches, Macon, 1893, 8. ITALIAN PETROLOGICAL SKETCHES 255 Basalt. — Of the rocks of this class described by Moderni and Bucca, I was unfortunately unable to obtain specimens. This is the more to be regretted, inasmuch as the feldspar basalts are only sparingly represented along the main line of Italian volca- noes, and at Rocco Monfina they are, according to Moderni, the last product of volcanic activity. Bucca describes a number of them, and his observations may be summarized here so as to complete as far as is possible, the petrographical description of the region. The basalts of the region are apparently quite feldspathic, and are quite free from leucite or nepheline. In the majority of cases they are olivine-bearing, this mineral being generally phenocrystic, and seldom forming part of the groundmass. It is usually some- what altered, especially on the borders, to a dark red substance. A few specimens are free from olivine and approach the augite- andesites, but are classed by Bucca with the basalts on account of their basic character. These last are of a light gray color, while the olivine basalts are dark. Augite is abundant, of a slightly bluish green, and not pleochroic. In two specimens a dark reddish brown biotite is present, which is for the most part largely altered to the usual augite-magnetite aggregate. The plane of the optic axes is perpendicular to the plane of sym- metry. Feldspar which is referred to plagioclase rarely appears as phenocrysts, but is abundant as laths in the groundmass. Its optical characters are not noted, so that we are unable to judge of its place in the series. Small magnetite grains are abundant and a colorless glass base is usually present. Although Bucca constantly refers to the feldspar as plagio- clase, yet he does not mention twinning lamellae, and indeed treats it in a rather cursory way. Certain facts, indeed, incline me to the belief that there is some, if not quite a good deal, of orthoclase in the rock. In the first place all the rocks which I have examined from this and the other volcanic centers are eminently rich in potash, even the phonolite of Viterbo con- taining 9.14 per cent, of it. It would then be quite anomalous to find here such a rock as a normal basalt containing a mini- 256 HENRY S. WASHINGTON mum amount of this alkali. If the feldspar is all plagioclase, as Bucca's description would indicate, there would be no min- eral capable of taking up any notable amount of potash, as its percentage in biotite is never over 10., and the amount of this mineral in these rocks is small. Furthermore we find at Radicofani, which is on the same main line, " basalts" containing very considerable orthoclase along with the plagioclase, and showing in one instance : SiO^ = 55.13 ; K30 = 2. 43, and Na30 = 2.07. Again Abich (^Z. cit. page 114) gives the silica percentage of the "basalt" from the foot of Santa Croce as 54.62, and classes it as a trachy-dolerite, though a rather basic one resembling the dolerites, while the rock of Monte Santa Croce is more acid and like the trachytes. It seems then very probable that these "basalts " of Rocca Mon- fina are in reality not normal plagioclase basalts, but rather approach the ciminites in composition both mineralogically and chemically. Henry S. Washington. ARE THE BOWLDER CLAYS OF THE GREAT PLAINS MARINE? Several trains of evidence show that the western plains, as well as the CordiUeran region, have been affected by great changes in elevation relatively to the sea level and to that of the eastern parts of the continent in later Tertiary and Pleistocene times. Facts bearing upon these changes have been detailed by the writer in previous papers and more particularly in those entitled respectively "Later physiographical geology of the Rocky Mountain region," and "Glacial deposits of southwestern Alberta in the vicinity of the Rocky Mountains.'" The observations made are in effect such as to lead the wri- ter to believe that the bowlder clays and other deposits of the glacial period covering a large part, at least, of the Great Plains in Canada, are glacio-natant deposits, not directly due to an ice- sheet and not calling for an extension of glacier ice as such to this part of the continent. He has further ventured to suggest that the water covering the western plains at this time may have been at the level of that of the sea and in more or less direct communication with it. The present note relates, however, to the discovery in the bowlder clays of the Great Plains of marine organisms which appear to be contemporaneous with their deposi- tion, and the general observations above alluded to need only be mentioned in introducing the subject. Some time ago Mr. T. Mellard Reade, writing in comment on the paper last referred to above, and in the light of his own investigations and those of Mr. Joseph Wright on the bowlder clays of Great Britian,^ suggested that a search should be made for 'Trans. Royal Soc. Can., Vol. VIII, Sec. 4 (iSQo). Bull. Geol. See. Am., Vol. VII, = Cf. Present Aspects of Glacial Geology, T. M. Reade, Geol. Mag.. (IV,) Vol. Ill, p. 542. Bowlder Clay a Marine Deposit, J. Wright, Trans. Geol. Soc. Glas- gow, Dec. 1894 and May 1895. 257 258 GEORGE M. DAWSON micro-organisms in the bowlder clays of the plains. In con- junction with Mr. B. W. Thomas, I had some years ago exam- ined several of these clays microscopically, but, as pointed out by Mr. Reade, the material examined in this case, was, in con- sequence of the mode of preparation adopted, that resulting from the elimination of all lighter particles by successive decan- tations, and not likely to include any foraminiferal forms still unfilled with mineral matter, which as a rule contain sufificient air to float to the surface of the water employed. In respect to the existence of forms contemporaneous with the deposit of the bowlder clays, the results arrived at by me were scarcely more than negative, but foraminifera evidently derived from the Cretaceous strata of the region were found in some of the clays from the Northwest.^ Mr. Wright having very kindly offered to examine some of the western bowlder clays by the methods found appli- cable by him to those of Great Britian, several specimens were collected for the purpose and submitted to him. The results arrived at form the subject of this note, which is, however, essen- tially of a preliminary character, and is intended to be followed by further investigation as soon as it may be possible to obtain additional material. The specimens sent to Mr. Wright were from the following places : Nos. I and 2. Saskatchewan River, twelve miles below Victoria, col- lected by Mr. R. G. McConnell. These represent a bed of bowlder clay about fifty feet thick, the first being from its upper, the second from its lower part. Present height above the sea level about 1850 feet. No. 3. Bowlder clay from near Victoria, Saskatchewan River, one and one-half miles up Egg Creek, also collected by Mr. McConnell. Height about 1900 feet. No. 4. Bowlder clay from Selkirk, Red River, Manitoba. Collected by Mr. J. B. Tyrrell. Height above sea, 720 feet. To these specimens from the West was added one collected by Dr. H. M. Ami, at Ottawa (No. 5). It is not certainly known whether the bowlder clay repre- sented by the first three specimens is the "lower" or "upper" ' Bull. Chicago Acad. Sci., Vol. I, No. 6 (1885). A RE BOH 'LDER CLA YS MA RINE ? 259 bowlder clay elsewhere recognized in the western part of the plains,' but it is probably the latter. Microzoa were found only in the three samples from the Saskatchewan Valley. ' In giving the results of his examinations Mr. Wright writes as follows : "In the clays from Victoria (1,2 and 3) I find foraminifera (and Radiolaria) and I am of opinion that they are contempora- neous with the clay and not derived from Cretaceous strata — I judge by the general resemblance of the foraminifera to those we find in British bowlder clay. The foraminifers in the Creta- ceous rocks of Canada may possibly be different to those which occur in the rocks of this age with us — I have never seen Creta- ceous microzoa from Canada and so can give no opinion on this subject. " Our chalk foraminifera are invariably of a dull white chalky appearance, the tests alone being calcareous, the interior being usually siliceous. On the other hand, our bowlder clay foraminif- era differ in no respect from recent specimens, except in being usu- ally smaller in size, the species being such as are now met with in shallow water around our coasts. " All the species which I have been able to identify in the clays you sent me, are referable to recent species, and with the exception of Cristellaria Italic a and Rotalia orbiailaris , have been found in British bowlder clay. No7iio7iina deprcssnla is the most abundant form in our bowlder clay, and it is instructive to find the species, so common with us, also occurring in your clay. "• Bolivina IcBvigata, Cristellaria Italic a, as also some of the other specimens, have the clear hyaline luster of recent specimens. If Cretaceous, we would expect to find Globigcrina Cretacea and Tex- tularia globidosa plentiful." The above references to Cretaceous foraminifera, are explained by the fact that Mr. Wright's attention had been called, when the specimens were sent, to the probable existence of such forms in the bowlder clays. In replying to the letter from which I have just quoted, half a dozen specimens of Cretaceous foraminiferal material from the • Cf. Bull. G. .S. A., Vol. VII, p. 60. 26o GEORGE M. DAWSON Canadian Northwest, collected by Mr. J. B. Tyrrell, were sent to Mr. Wright, and allusion was also made to the report by Messrs. A. Woodward and B. W. Thomas on the "Microscopic Fauna of the Cretaceous in Minnesota, Nebraska and Illinois." ' In this report, all the foraminifera found in bowlder clays, as well as those actually obtained from Cretaceous rocks, are classed together as Cretaceous. After carefully examining the Cretaceous material sent, and preparing lists of the forms represented, Mr. Wright notes the occurrence in it of a great preponderance of the two species already mentioned by him as likely to be characteristic. He further points out that these Cretaceous foraminifera are filled with calcite, differing in that respect from most of those of the same age in Great Britain, but none the less stony and unlikely to float during the treatment of the clays. In Yorkshire he has met with clays containing about equal proportions of Creta- ceous (derived) and Pleistocene (contemporaneous) foraminif- era, but found no great difficulty in separating the two lots by the criteria already alluded to. Referring to Messrs. Woodward and Thomas' report, he expresses the belief that it really com- prises a mixed fauna of the same kind, stating that of twenty- nine species recognized by these gentlemen, ten had not before been recorded from rocks of Cretaceous Age, according to Brady's monograph in the Challenger report. One of the localities mentioned by Messrs. Woodward and Thomas for foraminiferal bowlder clay, that of South Chicago, lies so far from known Cretaceous outcrops and away from the line of any recognized drift from such outcrops, that I ventured to address a question on the subject of the probable origin of the microzoa to Professor T. C. Chamberlin. The foraminifera found in this bowlder clay, appear to be in part, at least, undoubt- edly Cretaceous in age. In reply. Professor Chamberlin quotes observations made in northern Wisconsin which tend to show the existence of Cretaceous outliers there, as well as perhaps beneath the northern part of Lake Michigan, or even further east. He ' Geology of Minnesota, Vol. Ill, Part I (1895). ARE BO WLDER CLA YS MA NINE f 261 writes: "Taking the ev^idence all in all, I do not think there is any serious difficulty in accounting for Cretaceous forms in the drift of this region" (Chicago). After referring to the interest attaching to Mr. Wright's observations, he adds the following interesting suggestion concerning them: "It has occurred to me to raise the question whether a certain number of marine micro- scopic forms are not to be expected in any slow-accumulating deposit like a clay, in the interior of the continent, having been borne there by the wind with other dust picked up from marine flats on the windward side of the continent." The purpose of this communication is accomplished in stating as above, briefly, the new facts which appear to bear upon the question asked in its title. It seems to be at least very prob- able that, in addition to derived Cretaceous foraminifera often found in the drift deposits of the Great Plains, we have con- temporaneous forms of the sea of the glacial period, still unfilled with mineral matter, unaltered, hyaline in aspect, and represent- ing the same species elsewhere commonly found in deposits of this period. Should further investigation confirm the contempo- raneous and autochthonous character of this fauna, it will greatly assist in enabling the formation of definite hypotheses respecting events of the glacial period in the western part of the continent. Mr. Wright's notes on the specimens of bowlder clay from the Saskatchewan, are as follows : — No. I. Bowlder clay, twelve miles below Victoria. Weight 4 lbs. 4.5 oz Troy. After washing — Fine i lb. 3.7 oz. Course 0.7 oz. Stones mostly rounded, some angular. Gandryifia Sp., very rare. Bu/imina piipoides D'Orb., frequent. Pulvmulina Karsteni {Kss.),'v&ry rare. Nonionina depresstila (W. & J.), very rare. Rotalia orbicularis, D'Orb., very rare. Radiolria, frequent. Sponge spicules, rare. No. 2. Bowlder clay, twelve miles below Victoria. Weight 2 lbs. 6.2 oz. Troy. After washing — Fine S.goz. Coarse 0.5 oz. Stones mostly rounded, some ancrular. 262 GEORGE M. DA WSON Bulimina piipoides D'Orb., rare. Bolivina Icevigata (Will.), very rare. Cristellaria Italica, Defr. (young), very rare. Truncatnlina Sp., very rare. Rotalia orbicularis , D'Orb., very rare. Radiolaria frequent. Sponge spicules, rare. No. 3. Bowlder clay near Victoria. Weight 3 lbs. 1.6 oz. Troy. After washing — Fine i lb. o.i oz. Coarse 0.6 oz. Stones mostly rounded, some angular. Gaudryina Sp., frequent. Bulimina piipoides, D'Orb., rare. Rotalia orbicularis, D'Orb., frequent. Nonionina depressula (W. & J.), very rare. " scapJia (F. & M.) ? very rare. Radiolaria, frequent. Sponge spicules, rare. Ostracod, very rare. It may be of interest to add for comparison, the species actu- ally recognized by Mr. Wright in the several, small samples of Cretaceous material supplied to him. These are, nearly in order of relative abundance, as follows : — Textularia globulosa Ehr., ver}' common. Globigcriiia Cretacea, D'Orb., very common. " digitata, Brady, rare. Aiiomolina ammonoides, (Rss.), rare. Nodosaria Zippei, Rss., very rare. George M. Dawson. Geological Survey of Canada, March 10, 1897. THE BAUXITE^ DEPOSITS OF ARKANSAS. CONTENTS. Previous publications. Definition and composition. Structure and appearance. Geologic age. Origin. Forms of the deposits. Methods of mining. Uses. Bauxite as a refractory material. Markets. Bibliography. The geological survey of Arkansas was begun in June 1887. In that month I discovered some of the bauxite beds of Pulaski county, but the nature of these deposits was not announced until January 1891, when I gave a short account of their distri- bution and character.^ The many inquiries concerning the Arkansas bauxites, both from geologists and manufacturers, furnish sufficient reason for the present more extended account of them. This is made still more necessary by the fact that my report on the clays, kaolins, and bauxites of Arkansas has not yet been published, and it seems improbable that it ever will be published by that state. In view of the fact that Owen was State Geologist of Arkan- sas from 1857 to i860, and that other surve3^s were carried on 'This is also sometimes written beauxite. The word is derived from Les Baux or Les Beaux, the name of a town in the south of France. Pronounced bozile, not bawksite. ^ K letter on this subject was addressed to Governor James P. Eagle and was printed in the Arkansas Gazette and the Arkansas Democrat of January 8, 1891, and reproduced in the Arkansas Press in several numbers from January 18, 1891, until 1893, and also in the third and fourth biennial reports of the Commissioner of Mines, etc., of the state of Arkansas. A brief notice of the deposits was also published in the American Geologist iox y\.'A.xc\\ 1891, pp. 181-183. 263 264 /. C. BRANNER from 1 87 1 till 1874/ it is somewhat remarkable that the baux- ite deposits were not sooner discovered. The rocks were seen and mentioned by two geologists, but their true character and importance were not suspected. As long ago as 1842 Dr. W. Byrd Powell published a brief paper on the geology of Fourche Cove, near Little Rock, in which he says:^ "There is an exten- sive amygdaloid formation within the cove, and also upon the eastern side of it (page 1 1 j . . . . The amygdaloid proper is of a light brick color. In some localities it is darker, and in others lighter .... At one locality the amygaloids are small, resem- bling a mass of peas, .... but each amygaloid, or pea-like body, is hollow, having a shell not thicker than that of an ^^^. At another the amygdaloids are filled, but they, as well as the cement or gangue, are earthy and more or less friable." There can be no doubt that some of these so-called amygdaloids are the bauxites. It is evident, however, that some of the tertiary conglomer- ate beds were regarded by Dr. Powell as variations of these "amygdaloids," for he says that some of them consist of jasper (p. 12). Owen also mentions this "ferruginous amygdaloid of rather a peculiar character," and says that "the amygdules are very globular, so that the rock has much the appearance of peastone, the cavities being mostly empty." ^ The description of the rock and the locality given make it clear that the bauxite rock is here referred to, and yet there is no word to show that either Powell or Owen knew or suspected the true nature of the material. That these observers did not recognize this mineral is prob- ably due to the fact that bauxite was, at that time, but little 'See the Geological Surveys of Arkansas, by J. C. Branner. Jour, of Geol. II, November-December 1894, 826-836. ^A geological report upon the Fourche Cove and its„immediate vicinity, by W. Byrd Powell, M.D., Little Rock, 1842, 11-13. This is a very rare pamphlet; I knowf of but one copy of it, and that is in the library of the U. S. Geological Survey. 3 Second report of a geological reconnoissance of Arkansas. By David Dale Owen, Philadelphia, i860, 70. Jour. Geol., Vol. V., No. 3. Plate I. T1AP OF "THE LittleRockIauxite Iistrict Op-lc rrvilo IBa,u.-K[tl Z^neous J'aleozoiC 266 /. C. BRANNER known even in Europe, and it was altogether unknown in America. At the time when it was found that this rock was bauxite (1887) it was being used extensively for making the roadbed of the Little Rock and Sweet Home turnpike, especially along the part of the road near one of the bauxite beds. It had also been recommended by a former state geologist, Mr. W. F Roberts, Sr., as a " pisolitic iron ore," and attempts had been made to mine some of the highly colored Saline county beds for iron. Definition and compositioii. — It is practically impossible to draw a sharp line of division between bauxite as defined miner- alogically and what appears in a hand specimen to be the same thing, but which on analysis is found to contain one or another impurity in such proportions as to throw it out of the bauxite list. Mineralogically bauxite is a hydrate of alumina,^ but it is never found without certain impurities, the common ones being iron, silica, potash, soda, and titanium. All these impurities are found in the Arkansas bauxites, and in some cases the iron is so abundant that the beds have been prospected with a view to using the material as an ore of iron. The following analyses of some of the more ferruginous varieties show its iron contents : ANALYSES OF FERRUGINOUS PISOLITIC BAUXITE. Number Percentage of iron Phosphorus Silica I 66.85 54-2 59-2 54-9 ■043 .048 .031 .10 not det. 2 not det. -l 4 1.99 No. 4 was collected by the writer in 2 S. 14 W. section i, northwest quarter ; the other numbers are from the Saline ' St. Claire Deville regards it as a variety of Diaspore (AlgOgHgO) ; Dufrenoy thinks it is close to Gibbsite (AI0O33H3O), while Liebrich shows it to be closely related to or identical with Hydrargillite. Jour. Geol., Vol. V., No. 3. Plate II. 268 /. C. BRANNER county deposits, but they were collected by prospectors for iron ore, and their exact locations are not known. In other cases the bauxite contains so much silica and so little water that it is not to be distinguished by analysis from ordinary kaolm. And these varieties grade so insensibly into each other that no line of demarkation can be drawn between them. The following analyses show the similarity in composi- tion between a kaolin and a variety of bauxite, which, properly speaking, is simply a pisolitic kaolin. ANALYSES OF PISOLITIC AND ORDINARY KAOLINS. Silica, SiOg Alumina, AlgOg . . Ferric oxide, Fe^O Lime, CaO Magnesia, MgO . . . Potash, KgO Soda, NasO Water Total Pisolitic kaolin I S. 13 W., sec. 10 N. E. of N. E. 48.05 38.92 1. 19 0.58 0.45 0.18 0.28 10.86 100.51 Pisolitic kaolin Same sec, S. E. of N.W. 13 100.98 Washed kaolin from Brandywine Summit, Pa. 47.24 37-27 1.94 0.52 trace 0.22 0.13 13.62 100.94 Water at II0°-II5°C 0.46 % 1.4 In the following table are brought together all the analyses of the Arkansas bauxite, and a few analyses of representative bauxites from other parts of the world. ^ These analyses are of individual examples, however, and must not be' accepted as if made of car-load lots. ^ In the article published in the A7nerican Geologist for March 1891 I gave an average of fourteen partial analyses of bauxite from France, Austria, and Ireland. I wish here to express my disapproval of such a method of making comparisons. Owing to the extreme variability of bauxites, even from the same 156ds, such an average can scarcely fail to be misleading. If material is wanted with a low percentage of silica the high percentage in some of the samples will so increase the silica in the average that one may be led to infer that none of it is low enough to be available — a conclu- sion entirely unwarranted. THE BAUXITE DEPOSITS OF ARKANSAS 269 ANALYSES OF ARKANSAS AND FOREIGN BAUXITES. I 2 3 4 5 6 7 8 Q 10 II Silica Alumina Iron Titanic oxide TiO^ Water SiO, A!,0^ Fe„03 H,0 10.13 55-59 6.08 28.99 11.48 57.62 1.83 28.63 3-34 58.60 9. II 28.63 4.89 46.44 22.15 26.68 5.II 55-^9 19-45 17-39 33-94 44.81 1-37 2.00 17.88 2.00 62.05 1.66 3-50 30.31 10.38 55-64 1-95 3-50 27.62 16.76 51.90 3.16 3-50 24.86 45.20 37.60 3-00 13-54 21.70 58.10 3.00 3-20 14.00 2.80 57.60 25-30 3-10 10.80 6.29 64.24 2.40 some 25-74 Color Light brown Gray Light red Brick red Black Gray, surface Pink Surface f Lime 0.89 ) Potash .06 I Soda .69 L White, pisolitic White Red 1. I north, 12 west, section 24, north side of the southeast quarter. On Little Rock Sweet-Home turnpike, cut near road. 2. I north, 12 west, section 25, southwest corner, and sec- tion 36, northwest corner. Tarplay's. 3. 2 south, 14 west, sections 9 and 10; extending also from 10 into northwest of 15. 4. 2 south, 14 west, section 3, southeast of the southwest. 5. 2 south, 14 west, section 3, southeast of the southwest. 6. 2 south, 14 west, section 16, northeast corner of the south- west quarter near Sol. Nethercut's. 7. I south, 12 west, section 9, northwest quarter of the north- east quarter, at the end of the Arch Street Pike, and just north of the fork of the road at the point mentioned. 8. I south, 12 west, section 4, middle of the south side of the northwest of the southwest quarter on the west side of the Arch Street Pike leading south from Little Rock. P^xposure in the field, a stone's throw from the road. 9. I south, 12 west, section 9, northeast quarter of the north- west quarter, west of the pike and west of small stream, about TOO feet south of bridsfe. 270 J. C. BR ANNE R 10. I south, 13 west, section 10, southeast quarter of the northwest quarter. 11. Baux, near Arle, France (Sainte-Claire Deville) . 12. Revest, near Toulon, France (Sainte-Claire Deville). 13. Near Feistritz, Styria. The analysis must determine the value of bauxite, but it should not be forgotten that varieties not available for one pur- pose may sometimes be used for some other purpose. As a rule, however, silica, iron, and titanium are the objectionable ingre- dients. When the percentage of silica reaches that in kaolin or clay the bauxite has no advantage over kaolin or clay for the purposes for which it is used. Mr. McCalley, in his valuable paper upon Alabama bauxite, calls attention to the fact that the surface material contains more silica than samples taken at a depth.' This had escaped atten- tion in my brief examination of the Arkansas bauxite, but since reading Mr. McCalley's paper I recall the fact that this view is borne out by the analyses of the Arkansas materials so far as they have been made. If this is a fact that can be depended upon as constant, it is one of great importance in mining bauxite. In the ferruginous, earthy, and kaolin-like varieties the piso- litic structure is always more or less pronounced in the Arkansas bauxite. Structure and appeara^ice. — Bauxite is very light ; its specific gravity is about 2.4. In gross structure, color, texture, and general appearance Arkansas bauxite varies greatly. The colors are red, pink, brown, black, gray of various shades, white and yellow, and these colors are also more or less mixed in the same deposit, and even in the same hand specimen. The several classifications or subdivisions proposed by Coquand, Laur and others will, in all probability, hold with the Arkansas deposits, but inasmuch as these divisions all grade insensibly into each other it seems unnecessary to give those classifications or to lay any stress upon them. Chemical analyses and the practical availability of the different varieties can alone be 'Proc. Ala. Ind. and Sci. Soc, Vol. II, 1892, p. 29. THE BAUXITE DEPOSITS OF ARKANSAS 271 depended upon. I may add, however, as a suggestion of possi- ble utility that the red varieties are not all high in iron as one might suppose ; some of the reddest examples found in Saline county contain very little iron. Fig. I. Pistolitic Bauxite from near Little Rock. Some of the pisolites are solid and some of them are hollow ; sometimes they are scattered through a more compact and homo- geneous groundmass, and at other times they make up the body of the rock and resemble a great mass of peas cemented together. The accompanying illustration (Fig. i) shows a specimen of the more pisolitic variety taken from the deposits in township I south, 12 west, section 4, southwest quarter. 2/2 J. C. BRANNER The pellets are in some place much larger, the biggest being the size of one's two fists or even larger. Bauxite is sometimes compact. Mention should be made of the fact that while one often sees small pieces of the compact variety, there are, so far as I know at present, no beds or con- siderable deposits of the compact kind in Arkansas. I have seen in the Royal College of Sciences in Dublin specimens of the compact bauxite of Ireland. That material looks very like a compact and homogeneous clay, and has no evidences of pisolitic structure. Bauxite of this variety I have not found in Arkansas. The heavy beds in Saline county are in some respects differ- ent from those at the other Arkansas localities. In small frag- ments this material is not distinguishable from that found near Little Rock, but the beds are, in places, made up of what seem to be cobbles or rolled and waterworn lumps of the same mate- rial. Some of these lumps are as large as a man's head. Under the microscope one specimen shows the pisolites to be concentric in structure, while here and there through them are thin bands or veins of quartz. This suggests that the silica found by analysis is sometimes free." Geologic age. — Bauxite has been found in Arkansas only in Tertiary areas and in the vicinity of eruptive syenites with which I believe them to be genetically related. This statement con- cerning areal distribution must be accepted simply as a state- ment of fact, but one that may be of considerable importance in Arkansas, and possibly of none whatever in other bauxite regions. The evidence of the Tertiary age of the Arkansas bauxite is not abundant, but it all points in one direction. At the village of Ridgewood, about five miles south of Little Rock, a well dug on lot 1 1 exposed the following section : Section of well on lot 1 1 , village of Ridgewood. — 3 feet surface soil. - 7 feet pisolitic bauxite. , ^For a good microscopic view see paper of C. W. Hayes in the l6th Ann. Rep. U. S. G. S., Part III, Plate XXIII. THE BAUXITE DEPOSITS OF ARKANSAS 273 2 inches iron (limonite.) 6 inches pink and white clays. The pink and white clays at the bottom of this well belong with the horizontally stratified Tertiary beds of the neighbor- hood. There are no fossils, however, in these clays, and they are considered as Tertiary solely upon lithologic and strati- graphic grounds. At Tarplay's place in the northwest corner of section 36, i north, 12 west, the bauxite has a thin bed of Tertiary or post- Tertiary sandstone overlying it. At Mabelvale in i south, 13 west, section 10, northwest quarter, the same kind of ferruginous sandstone overlies bauxite. It should be added also that the bauxite at several places overlies the eruptive syenites or so- called "granites," while at several others it is very probable that there are syenites beneath them. These syenites are apparently of late Cretaceous or post-Cretaceous Age. It is not to be inferred, however, that bauxite is confined to Tertiary rocks, except perhaps in Arkansas. I see no reason why the conditions that produce this mineral might not exist in one age as well as in another, though these conditions are likely to occur in one age in one place and in another age at another place. The statement of Laur that " the phenomena which gave rise to the bauxites in Europe occurred with great intensity toward the end of the Cretaceous epoch, and has never been repeated"' is probably unwarranted. In southern France bauxite is said to form the parting between the Cretaceous and the Jurassic. Coquand says it is in the Lychnus beds of the Upper Cretaceous.^ Collot shows that it lies between the Urgonian below and the Cenomanian above, that is, about the junction between Lower and Upper Cre- taceous. 3 Rouville says the pisolitic iron of Herault is in the "Oxfordian" or Upper Jurassic, if by " Oxfordian " he means ■Trans. Amer. Inst. Min. Eng. 1896. ^Bull. Soc. Geol. 1870-1, XXXVIII, iii. 3Compt. Rend. 1887, CIV, 129-130. 274 J. C. BRANNER the Oxford clays of the English geologists.' Fabre found it in crevices in Jurassic rocks, ^ but deposited in early Tertiary times, while that referred to by Dr. Raymond in a corundum mine in North Carolina is believed to be in eruptive rock, 3 the age of which is not stated. The Irish bauxites are associated with eruptive rocks of Tertiary Age.'* The Alabama bauxite deposits are said^ by McCalley to be of Lower Silurian Age, but Hayes thinks^ the Alabama and Georgia deposits were formed "toward the close of the Eocene." These cases are cited simply to show that bauxite is not con- fined to rocks of any particular age, except perhaps in a given region. I have seen it stated that bauxite has been found recently near the Maumelle Pinnacles, about fifteen miles up the river from Little Rock. This would bring the deposits into rocks of Carboniferous Age This report lacks confirmation; I know of neither bauxite nor syenite in that region. It is important to notice, however, that if the writer's theory of the origin of these deposits is correct, search for them in Arkansas should be confined to the neighborhood of the eruptive syenites, ^ though not all these syenites have bauxite deposits in their vicinity. The eruptive rocks at Magnet Cove are mostly syenites, but no considerable bauxite deposits have been found associated with them. Some small fragments were found by me in the Cove on the north slope of the hill just south of the Baptist church. Origin of the Arkansas bauxite deposits. — In searching for new deposits and in determining the limits of those already known, we must be guided to a certain extent by a knowledge of the method ' Bull. Soc. Geol. de France XXV, 1867-8, 935. ^BuU. Soc. G^ol. de France, 1869-70, XXVII, 518. 3 Trans. Amer. Inst. Min. Eng. VII, 86. ''Memoirs of the Geol. Survey [of Ireland] to accompany sheets 7, 8, 14, 20. By R. G. Symes. 5 Alabama bauxite. By Henry McCalley. Proc. Ala. Ind. and Sci. Soc, 1892, II, 21. ® Bauxite. By Charles W. Hayes, i6th Ann. Rep. U. S. Geol. Sur., Part III, 592. Washington, 1896. ^ For the distribution of eruptive rocks in Arkansas see Vol. II o f the annual report of the Geological Survey of Arkansas for 1890. THE BAUXITE DEPOSITS OF ARKANSAS 275 by which the deposits have been made, for it is only in this way that we can anticipate the peculiarities of its distribution. There is therefore given below, as briefly as possible, the several theories that have been advanced to explain the method by which bauxite deposits have been made. The Arkansas beds appear to have been laid down in water near the shore, but the material does not seem to have been car- Fig. 4. Ideal Section through Fourche Cove. ried far from the spot at which it originated, or to have been widely distributed by the water. They are all at or near the contact between the palaeozoic sediments and the eruptive sye- nites. Several of them, however, have no palaeozoic rocks exposed near at hand, and one of them has no syenite expo- sures. These conditions are shown in part by the following ideal section illustrating the relations of the rocks about the Fourche Cove. The concretionary structure of the bauxite suggests that it has been formed in some such manner as the oolites or sprudel- steins of the well-known hot springs at Carlsbad. At this last named place the oolites are made of carbonate of lime. The water is charged with lime, and as it issues the lime collects about centers. These masses are kept in motion in the rising waters until by accretion they become so heavy that they sink to the bottom. Oolites are also made in various other ways, such as the roll- ing of grains upon a shore and in waters heavily charged with lime, by lime secretions of certain algae ^ and other organisms," and possibly by insect eggs as suggested by M. Virlet d'Aoust.3 'On the formation of oolite. By Dr. A. Rothpletz. Amer. Geol. X, 279-282. ='The formation of oolite. By E. B. Wethered. Quar. Jour. Geol. Soc, LI, 1895, 196-209. 3 The Geologist (I), 1858, 72-73- 2/6 J. C. BRANNER Attention is directed to the peculiar nature of some of the great beds of Saline county ; there the strata are composed of rolled or waterworn lumps as large as one's head which, when broken open, show the same pisolitic structure as that found at other places. It looks as though the material had been depos- ited in water near the shore, and that it had been partly uncov- ered at low tide or broken up by storm waves, and that its earthy material had been broken and rolled by the waves, and finally left at or near where it had originally lain. None of the beds examined show the lamination or thin bedding planes so characteristic of sedimentary rocks. These facts seem to point to an origin for bauxite very sim- ilar to that of calcareous pisolites, and its association with the syenites suggests that the latter have something to do with the matter. I am of the opinion that the explanation offered by Coquand, Auge, de Rouville,^ Virlet d'Aoust,^ Daubree,^ and Hayes is the correct one so far as hot waters are concerned, but I am unable to see why they should have been geyser waters. Auge has cited* from Hayden's report of 1878, Part II, p. 416, what he considers a case of bauxite actually forming in a gey- ser. This, however, is a mistake ; I have examined in the United States National Museum the material referred to ; it has no resemblance to bauxite, and Professor George H. Merrill tells me that it is a geyser mud or kaolin mechanically churned up by the water and from which the silica has been removed. Strangely enough this error has been extensively copied and is found in many of the papers on bauxite, even as late as that by M. Lauer published in 1895. Dr. Genth holds that the bauxite may be derived by hydra- tion from corundum, but frankly admits that he cannot explain the transformation. s ■ Bull. Soc. Geol, de France, 1867-8, XXV, 935. ^Op. cit., XV, 199; XXII, 418-420. 30p. cit, XXVI, 915. -i Bull. Soc. Geol. de France, 1888, XVI, 345-350. 5 Proc. Am. Phil. Soc. Vol. XII, 373 and 405. I THE BAUXITE DEPOSITS OF ARKANSAS 277 Coquand says:' "The aqueous origin of the bauxites is as well marked by their structure as by their stratification and alternation with sandstones, limestones, and clays. It is evi- dent that the sedimentation (at the time the deposits were made) began at the bottom of a lake by the deposition of alu- minous and calcareous matter brought in from mineral sources and to which a certain movement of the waters gave a pisolitic form." It is not altogether clear what M. Coquand means by the "mineral sources" to which he refers several times and to which he attributes the material of the bauxites. His paper ends with the conclusion that the several bauxite localities of France "are of the same age and fall under the head of irregular deposits of geyser origin," from which it must be inferred that he thinks the hot waters of geysers are the "mineral sources" from which the bauxite has been derived. M.Stanislas Meunier holds Mhat salt water penetrating to great depths can, on account of its high temperature and the pressure upon it, dissolve the ferruginous shales and form chlo- rides of alumina and perchlorides of iron. When these chlo- rides come to the surface and spread over limestones, as they usually do, a change of bases takes place, alumina and peroxide of iron are precipitated, chloride of lime is carried away and the carbonic acid of the limestone is set free, while the iron and alumina is deposited as bauxite. Auge, as stated in the article already referred to, was led to reject Meunier's theory by finding bauxite resting on rocks other than limestones. To this Meunier replies that the chem- ical changes demanded by his theory might be produced by the waters carrying the material passing over limestones and depositing the bauxite further on in their course. Daubree says^ that while generally associated with sedimen- tary beds, the bauxite beds show their relations to deep-seated emanations by the presence of anhydrous peroxyde of iron or I Bull. Soc. Geol., 2™« sen, 1870-1, XVIII, 98 ff. = Bull. Soc. Geol., s'"^ sen, XVI, 1888, 345-346. 3 Bull. Soc. Geol. de France, XXVI, 1868-9, 9i5- 278 J. C. BRANNER oligiste, which generally colors them, and by its ramifications generally penetrating underlying beds and by the juxtaposition to granite. R. G. Symes thinks that the bauxite at Estertown, County Antrim, Ireland,^ "seems to be a mud lava." The most exhaustive study yet made of bauxite is that of Dr. A. Liebrich, the title of whose treatise is given in the bib- liography at the end of this paper. He quotes Streng as hold- ing that the Vogelsberg bauxite is derived from basalt by decomposition, and, in general terms, Liebrich endorses this view. Further on, however, he states that it is "not the decomposition product of an underlying basalt, but of a com- pletely disintegrated anamesite lying above a compact basalt." Again he savs : "There is nowhere bauxite containing a kernel of stone which is not a decomposition product." Again "the transition between bauxite and basaltic hematite may be traced step by step in thin sections." Also; "it is to be regarded as a concretionary formation which has originated in the clay formed by decomposition of rock, and that it is therefore not a simple process but several different processes following each other." He thinks, however, that other kinds of basalt than anamesite can yield bauxite. He also concludes from his micro- scopic and chemical studies of bauxite that it is the same as hydrargillite. The chemical process and the method of form- ing the pisolites, he says, is unknown. Of the theories above mentioned the only one that appears to be applicable to the Arkansas deposits is that of hot waters. This however is not necessarily in conflict with the theory of Liebrich that they are decomposition products, though they are not, in the present case, associated with basalts, and are cer- tainly not formed by the decomposition of any rock in place. It is my opinion that before the eruptive syenites had cooled they were sunk beneath the Tertiary sea, and that either by the con- tact of the sea water or by the issuing of springs whose waters ' Memoirs of the Geological Survey of Ireland, Mem. accompanying Sheet 20, p. 12. THE BAUXITE DEPOSITS OF ARKANSAS 279 had been in contact with the hot syenites the aluminous mate- rials were se;^gregated as pisolite and sank near where they were formed. Thus if the process is one of decomposition it is decomposition in the presence of and due to high temperature. The irregular forms of the deposits, their variable thicknesses and characters, and, in fact everything known about them, is in keeping with this theory of their origin. Stress has been laid by Meunier upon the fact that so many bauxites rest upon^ or are associated with limestones ; and this fact was used by him to explain the precipitation of the material from chlorides of aluminium in hot sea water. But, as Auge has pointed out, lime- stones are not always present,^ and there are certainly none associated with the Arkansas deposits. Forms of the deposits. — From what has been said it is evident that the bauxite deposits must be very irregular in form. Although associated with marine sediments of wide and regular distribution, the bauxite deposits are local, irregular and of uncertain extent, for the influences that produced them were local, and the rocks (syenites) with which they are associated are in all probability concealed in many places. The figure given under the head of "origin" on page 275 will give as definite an idea as can be had of the forms of the deposits with- out prospecting. If, at any time, it should become necessary to prospect for other deposits, the prospecting should be confined to the region in which bauxite is now known and to the soft Tertiary beds. It does not now seem advisable to look for this mineral above an elevation of 350 feet or below 250 feet above mean tide. MetJiod of mining. — Stripping and quarrying in open cuts is the method to be used in nearly all the deposits known. Of course when the cover becomes too thick it cannot be removed economicall}^, and such deposits will either have to be aban- doned or the mining will have to be done by drifts. Persons experienced in the driving of tunnels and mine timbering should ' Comptes Rendus, 1S83, XCVI, 1737-1739. » Bull. Soc. Gdol. de France, 1888, XVI, 346. 280 J. C. BRANNER have charge of the work in this stage. The roof is likely to be soft and to require lagging. Furthermore, unless discretion is used in opening quarries and driving tunnels, the operators are likely to have difificulty with the draining of the mines. THE DISTRIBUTION OF BAUXITE IN ARKANSAS. In thickness the bauxite beds vary greatly, the greatest found being forty feet, and even in this case the full thickness of the bed is not exhibited. In vertical distribution it has a range of about sixty feet, lying, so far as observed, between 260 and 320 feet above mean tide level. This observation, however, refers only to the Pulaski county deposits, no observations having been made on the vertical range of the Saline county beds, which, however seem to be at, or near, the same elevation. It is not supposed that the exposures now known are the only ones in the state, for it has not been possible to make detailed search throughout the area in which the deposits may reasonably be expected. The number of exposures will probably not only be considerably increased, but it should be added, that inasmuch as sedimentary beds overlie or have overlain some of the known deposits, it is quite possible that there are others yet uncovered by natural processes of erosion. These bed should be sought only in the areas of soft sands, clays, and gravels and in the neighborhood of the eruptive rocks of Saline and Pulaski counties. THE USES OF BAUXITE. Many attempts have been made to use bauxite as an ore of iron, but with poor success or with no success at all. The kao- linic varieties may be found available as kaolins, though I know of no attempts to use them for such purposes. Bauxite has been successfully used for the manufacture of the following materials : 1. Alum. 2. Sulphate of alumina. 3. Aluminum (the metal). THE BAUXITE DEPOSITS OF ARKANSAS 28 1 4. Refractory wares, such as furnace linings. 5. It is also available for increasing the refractoriness of fire clays used in the manufacture of fire bricks, furnace linings and the like. I shall say nothing of the methods employed in the utiliza- tion of the raw material. Some of them are described in the works mentioned at the end of this paper, others are guarded as trade secrets, or are covered by patents. There is one use, however, for which bauxite is available to which I wish to direct especial attention, and that is as a refractory material in the manufacture of iron and steel. Bauxite as a refractory material. — Bauxite is one of the most refractory materials used in the arts. It is especially val- uable for lining blast furnaces where it outlasts the best artificial fire bricks. It is used alone and also as an ingredient for increas- ing the percentage of alumina in other refactory materials. In his " Feuerfesten Thone," Bischof speaks as follows of bauxite:^ "This natural aluminium hydrate which has as yet been found only in a few places, .... when not impure on account of the admixture of foreign substances, especially of iron, which generally occurs in considerable quantities in compounds of alumina, is extremely refractory. . . . The addition of the varieties free from iron, or the white ones, to other refractory cla^'S offers the only importafii means k?iozv7i of increasing their percentage of alumina, and at the same time their refractoriness. On account of a large percentage of chemically combined water this material shrinks considerably in burning, a fact to be noted in using it." In another place Bischof says:' "Bricks made from calcined bauxite are especially useful in the production of iron and steel in the Siemens revolving furnace. "3 Bruno Kerl, an excellent authority on this subject, says :* " Bauxite bricks and crucibles of bauxite, recommended by 'Die Feuerfesten Thone, C. Bischof, pp. 193-4, = 0p. cit., p. 277. 3See also Dingler's Polytechnisches Journal, Vol. CXCVIII., p. 156; ibidem, Vol. CCX., p. 109. •t Thonwaarenindustrie, p. 526. 282 J. C. BRANNER Gaudin as long ago as 1858, should, when low in iron, withstand heat which would fuse all other refractory materials. "Bauxite is a compound having a composition intermediate between diaspore and limonite, and consists of hydrate of alumina combined with hydrated oxide of iron. "On account of the large quantity of water present this min- eral shrinks greatly on heating, is usually refractory, and, if it does not contain too much iron, when added to fire clays, increases their refractoriness. " Siemens used bauxite bricks with 50 per cent, of alumina, 35 of iron oxide and 3 to 5 of silica, which lasts five or six times as long as Stourbridge first-brick. Schwarz recommends for crucibles for manufacture of cast steel a composition of one to two parts Goettweiher clay and two parts of burnt bauxite from Wochein. The bricks of the Compagnie Parisienne at the Vienna Exposition withstood the heat of molten platinum, yet their fracture was like that of stoneware." Sir William Siemens tested bauxite as a furnace lining and says^ of it: "A series of experiments to form solid lumps by using different binding materials have shown that 3 per cent, of argillaceous clay suffice to bind the bauxite previously calcined. To this mixture about 6 per cent, of plumbago powder is added, which renders the mass practically infusible, because it reduces the peroxide of iron contained in the bauxite to the metallic state. Instead of plastic clay as the binding agent, waterglass or silicate of soda may be used, which has the advantage of setting into a hard mass at such a comparatively low temperature as not to consume the plumbago in the act of burning the brick. A bauxite lining of this description resists both heat and fluid cinder in a very remarkable degree, as I have proved by lining a rotative furnace at my sample steel works at Birmingham, partly with bauxite, and partly with carefully selected plumbago bricks. After a fortnight's working the brick lining was reduced from six inches to less than half an inch ; whereas the bauxite lining was still five inches thick and perfectly compact. It is ' The scientific works of Sir Wm. Siemens, I., London, 1889, 296. THE BAUXITE DEPOSITS OF ARKANSAS 283 also important to observe that bauxite when exposed to intense heat is converted into a solid mass of emery of such extreme hardness that it can hardly be touched by steel tools, and is capa- ble of resisting mechanical as well as the calorific and chemical actions to which it is exposed. "The bauxite used in the above mentioned lining was of the following composition : AI2O3 - - - 53.62 FczOs - - - - 42.26 SiOa - - - - 4.12" Speaking of the value of bauxite as a refractory material Pro- fessor Thomas Egleston of the Columbia School of Mines says that it " lasts five or six times as long as the best Stourbridge bricks. Nothing has yet- been found which resists the corrosive action of basic slags so well."^ In a letter to the author regarding bauxite as a refractory material Professor Egleston writes : "If the material is pure there would be a very large demand for it. With the introduction of basic processes, the demand for basic lining has increased steadily, but on account of the uncertainty of the composition of bauxite, it is being very gen- erally replaced by carbonate of magnesia, which is found in several localities in Europe and is imported both to England and this country." The results of several analyses of Aakansas bauxites were sent Professor Egleston when he wrote : "The subject is a very interesting one and may be of great value to the state if it should prove that any of these are alumi- nates. In some of them I fear there is too much silica, but in any case I think valuable fire bricks could be made of them. I have often tried to interest the fire-brick people in the new proc- cesses for the manufacture of these bricks which have been developed within the past ten or fifteen years in Europe, and in the hope of so doing have published several articles on the sub- ject, but the manufacturers have been generally unwilling to 'Transactions Am. Inst. M. E., Vol. IV., 261-2. 284 J. C. BRANNER invest any more capital in their business, which, they say is already very much cut up. If a new industry were started, it could be started on altogether a different basis and I think would easily compete with the old manufactories. This is all the more true since the development of the basic open hearth and Bessemer process in the South calls for a higher grade of fire- brick, I had, while recently in Europe, some important inter- views with the proprietors of the magnesite quarries in the west with regard to the introduction of that material into the United States. If your material should prove to be aluminates you could easily compete with them." Markets. — The processes by which bauxite is manufactured are in some cases patented and the parties owning the patents are alone entitled to use or to dispose of them ; in other cases the processes are guarded as trade secrets. Partly for these reasons and partly because the utilization of bauxite is confined to but few companies, the public knows but little of the uses to which the raw materials are put, or of the processes employed in their manufacture. One thing that has thus far prevented the Arkansas bauxite getting into the market is the fact that the samples sent away have been selected without a knowledge of the composition required or of the material sent. As has been stated, iron, silica, and titanium are the objectionable ingredients, and the percentages of these cannot be determined by simple inspection, though familiarity with analyses of types will enable one to form an opinion of value on this subject. Another matter of importance is that the freight rates charged by the railways out of Little Rock are so high as to prevent its profitable shipment. Still another is that extravagant ideas of the value of the ore have induced those who would otherwise have done the mining and shipping to expect very large profits from it. As a matter of fact the value of bauxite at the place of production in the United States during the year 1895 averaged about four dollars^ 'Engineering and Mining Journal, Jan. 2, 1897, 3. THE BAUXITE DEPOSITS OF ARKANSAS 285 a ton. The question for owners and miners is whether the market price will leave them a resonable margin of profit. The cost of plants for the utilization of bauxite is so great, and the local market for manufactured products so small, that, in my opinion, it is useless to think of factories being established in Arkansas. The factories are all in the north, ^ and the probabil- ities are that they will remain there for some years at least. In the meantime, if the bauxite beds are to be utilized, the material must be mined and transported cheaply. Labor and teaming are low, but railway freights are high at present. This neces- sarily prevents Arkansas' competition with the Georgia and Alabama deposits. It is to be hoped that the railways may see their way to offering rates that will allow of bauxite mining in Arkansas. BIBLIOGRAPHY OF BAUXITE. Anonymous: Bauxite in puddling furnace. Iron and Steel Institute, 1882, I, 304-305- AUDOIN and Gaudin : Notizblatt des deut. Vereins f. Fabr. etc., V, 446. AUGE : Note sur la bauxite, son origine, son age et son importance g^ologique. Bui. Soc. G^ol. de France, 3""= sdr. 1888, XVI, 345-350. Bauer, Max: Lehrbuch der Mineralogie, 1886, 353. Berthier, R : Analyse de I'alumine hydratee des Baux, departement des Bouches-du-Rhone. Annales des Mines, 2'"'= ser. 1821, VI, 531. BiscHOF, C: Analysis of bauxite. The Metallurgical Review, 1878, II, 523- Abstr. Berg- u. Hiittenmannische Zeitung, 1878, 196. : Die feuerfesten Thone, Leipzig, 1876, I93-I95- : Dingler's Poly. Journal 1881, B, 239, 469. Blackwell,G. G.: (Bauxite), Trans. Manchester Geol. Soc. 1894, XXII, 525-527- Blake, W. P.: Alunogen and bauxite of New Mexico. Trans. Am. Inst. Min. Engs. 1894, XXIV, 571-573- Abst. Amer. Geol. Sept. 1894, XIV, 196. Branner, J. €.: Preliminary report to the governor of Arkansas on bauxite and kaolin. Arkansas Gazette and Arkansas Democrat, LiUle Rock, Jan. 8, 1891. : (Bauxites of Arkansas). Third biennial report, Bureau of Mines, Manu- factures, and Agriculture of the state of Arkansas for 1893 and 1894, Little Rock, 1894, 1 19-125; also fourth biennial report of same. Little Rock, 1896, 105-110. \ : Bauxite in Arkansas. Amer. Geologist, March 1891- Abst. Eng. and Min. Jour., New York, Jan. 24,1891, 114; Trans. Federated Inst. Min. Engs. Ill, 1057 ; Jour. Iron and Steel Institute, 1891, I, 275. Science, XVII, 1891, 17- ^he Pennsylvania Salt Company, Natrona, Pa. The Solvay Process Company, Syracuse, N. Y. The Pittsburg Reduction Company, Pittsburg, Pa. 286 J. C. BRANNER Brewer, W. M.: The Warwhoop bauxite bank, Alabama, Eng. Mg. Jour. 1893, LV, 461. Cole, G. A. J.: The rhyolites of the County Antrim (Ireland) with a note on baux- ite. Sci. Trans. Roy. Dublin Soc, ser. II, 1896, VI, 105-109. CoLLOT L.: Age de la bauxite dans le sud-est de la France. Compt. Rendus 1887, CIV, 127-130. Abstract Neues Jahrb. f. Min. 1888, I, 452. : Age des bauxites du sud-est de la France. Bull. Soc. Geol. de France, 3"'® ser. XV, 331-345. Abstr. Trans. North of Eng. Inst. Min. Eng. 1888, XXXVII, 4. : Sur la bauxite d'Olli^res. Description g^ologique des environs d'Aix en Province, 1880, 84 CoQUAND : Sur les bauxites de la chaine des Alpines (Bouches-du-Rhone) et leur age geologique. Bull. Soc. Geol. de France, 2""= ser. 1 870-1, XXVIII, 98-115. Abstr. Neues Jahrb. fiir Min. 1871, 940-941. Jahresbericht der Chem. 1871, 1144. D'AOUST, ViRLET : De la formation des oolithes et des masses nodulaires en general. Bull. Soc. Geol. de France, 2'"'= ser. 1857-8, XV, 187-205. ■ — : Sur le mineral de fer alumineux pisolithique de Mouries, dit aussi des Baux. Bull. Soc. Geol. de France, 2™® ser. 1864-5, XXII, 418-420. Daubree, a.: Sur I'existence de gisements de bauxite dans les departements de I'Herault et de I'Ariege. Bull. Soc. Geol. de France, 2^^ ser. 1868-9, XXVI, 915- 918. : Note sur un silicate alumineux hydrate, depose par la source thermale de Saint-Honor^ (Nievre) depuis I'epoque romaine. Comptes Rendus, 1876, LXXXIII, 421. • — : Les eaux suterraines aux epoques anciennes. Paris, 1887. (Bauxite, 96.) De Launay : See Fuchs. D AMOUR, A.: Note sur un hydrate d'alumine ferrugineuse trouve dans Pile d'Egine, Grece. Bull. Soc. Geol. de P>ance, 1864-5, XXII, 413-416. Deville, H. Sainte-Claire : De la presence de vanadium dans un mineral alumineux du Midi de la France. Ann. de Chimie et de Phys., 3™'=^ ser. 1861, LXI, 309-342. : Analyse d'une bauxite du Paradon. Ann. de Chim. et de Phys., 3™* ser. LXI, 309. Abstr. Jahresbericht der Chemie, 1861, 980. DiEULAFAiT : Les bauxites, leur ages, leur origine. Comptes Rendus, 1881, XCIII, 804-807. Ditte, a.: Sur la preparation de I'alumine dans I'industrie. Comptes Rendus, CXVI, 509-510. : Preparation industrielle de I'alumine. Ann. Chim. et de Phys., 1893, XXX, 280-282. Drechsler E. : Analyse des Bauxits aus der Wochein. Dingler's Pol. Jour. 1872, CCIII, 479-481. Egleston, T.: (Refractory material). Iron and Steel Inst. 1877, II, 533-534. Fabre, G.: Note sur les failles et fentes k bauxite dans les environs de Mende. Bull. Soc. Ge'ol. de France, 1869-70, XXVII, 5 16-5 18. Flechner, a.: Zeitschrift der geologischen Gesellschaft, 1866, XVIII, 181. THE BAUXITE DEPOSITS OF ARKANSAS 287 FUCHS, Ed. ET DE Launay, L.-: Trait(5 des gites min^raux et metallifferes. Paris, 1893, I. 595-599- Hauer, F. von : (Bauxit, Krain). Jahrb.der K. K. Geol. Reichsanstalt 1866, XVI, 457. Abstr. Jahresbericht der Chemie, 1866, 923. Hayes, C. W.: Bauxite. Mineral Resources of U. S. 1893, 159-167- Washing- Ion, 1894. : Geological relations of the Southern Appalachian bauxite deposits. Trans. Amer. Inst. Min. Eng. 1894, XXIV, 243-254. : Bauxite. Sixteenth An. Rep. U. S. Geol. Survey for 1894-5, Part III, 547- 597. Washington, 1895. Heintz : (Analysis). Stahl u. Eisen VII, 430; Journ. Iron and Steel Inst. II, 1887, 281. Hunt, A. E.: Bauxite. (Discussion.) Trans. Amer. Inst. Min. Eng. 1894, XXIV. 855-861. : The properties, uses, and processes of production of aluminum. Tech- nology Quarterly, 1891, IV, 1-35. (Bauxite 20.) Jannettaz, Ed.: La composition d'une variety pisolithique de bauxite. Bull. Soc. Geol. de France, 1S77-8, VI, 392. : Gibbsite et beauxite de la Guayane fran9aise. Bull. Soc. Min. de France, I, 70-71, Paris, 1879. John, K.: Verh. Geol. Reichsanstalt, 1874, 289 ; abstr. Jahresbericht der Chemie, 1874, 1246. Lang, J.: Ueber Bauxit von Langsdorf. Bericht der Deutschen Chemischen Gesell., XVII, 1884,2892-2894. Abstr Neues Jahrb. f. Min., 1886,11,342. Ab:,tr. Jour. Iron and Steel Inst., I, 1886, 293. Laur, F.: The Bauxites; a study of a new mineralogical family. Trans. Amer. Inst. Min. Eng., 1894, XXV, 234-242. *■ Lencauchez: (Dephosphorizing iron by bauxite) Societe des Ingenieurs Civils. Abstr. Iron and Steel Institute, 1880, I, 334-335- LIEBRICH, A. von: (Bauxite). Zeitschr. fiir Kryst. u. Mineral., XXIII, 296. : Bauxit. Bericht der Oberhess. Gesellscliaft fiir Natur.- u. Heilkunde, XXVIII, 57-98. : Beitrag zur Kenntniss des Bauxits vom Vogelsberge. Giessen, 1892. McCallev, H.: Alabama bauxite. Proc. Ala. Indust. and Scientific Society, 1892, II, 20-32. •: Bauxite mining. Science, Jan. 19, 1894, 29-30. : Bauxite. The Mineral Industry, 1893, II, 57-68, N. Y. 1894. Abstr. Jour. Iron and Steel Inst., II, 1894, 303- Meunier, S.: Sur I'existence de la bauxite k la Guayane fran^aise. Comptes Rendus, i872,LXXIV, 633-634. Abstr. Jahresbericht der Chemie, 1872, 1099. : Sur I'origine et le mode deformation de la bauxite et du fer en grains. Compl. Rend. 1883, XCVI, 1737-1740. : Reponse a des observations de M. Augd et de M. A. de Grossouvre sur I'histoire de la bauxite et des minerals siderolithiques. Bull. Soc. Gdol. de France, 1889, XVII. 64-67. 288 J. C. BRANNER MiERZlNSKi, Dr.: Die Fabrication des Aluminiums. Vienna, 1885. MiNET, A.: L'Aluminium; fabrication, emploi, alliages. Paris, 1893. Nichols, Edward: An aluminium ore. Trans. Amer. Inst. Min, flng., XVI, 905-906. Abstr. Iron and Steel Inst. 1888, II, 228-229. Packard, R. L.: Aluminum. Sixteenth Ann. Rep. U. S. Geol. Survey, Part III, Bauxite, 542-544. Washington, 1895. Percy, John: Metallurgy. London, 1875. (Bauxite, 130-133.) Petersen, Th.: Ueber den Anamesit von Rudigheim bei Hanau und dessen bauxitische Zersetzungsproducte. Jahresbericht d. Phys. Ver. zu Frankfurt a. M., 1891-2, 10. : Ueber Bauxitbildung (Bei. d. XXVII. Vers. d. Oberrhein. Geol. Ver. 2 S. 1893). Abstr. Neues Jahrb. f. Min., 1894, Ref. 460. Prziwoznik, E.: (Analyses of bauxites) Berg- u. Hiittenmannisches Jahrb. der k. k. Bergakad., XXXVIII, 415-416; XL, 564. Richards, J. W.: Aluminum: its history, etc. Philadelphia, 2d ed. 1890. Baux- ite, 47-49, 88, 109-115. Roth, J,: Allgemeine u. chemische Geologie. Berlin, 1879. (Bauxite, I, 141, 142,557-) Roth, LuDW.: Der Bauxit u. seine Verwendung zur Herstellung von Cement aus Hochofenschlacke. Wetzlar, 1882. Roule, Louis: Sur les gisements et Page de la bauxite dans le sud-est de la France. Compt. Rend., 1887, CIV, 383-385. Abstr. Neues Jahrb. fUr Min., 1888, 1, 452. : Recherches sur le terrain fluvio-lacustre inf^rieur de Provence. Ann. des Sciences Geol. XVIII, 1885, 138; abstr. Neues Jahrb. fiir Mineral., I, 1887, 98-100. Saemann: (Sur la bauxite des Baux) Bull. Soc. Geol. de France, 2""^ ser. XXII, 416-417. Schnitzer, Guido: (Analysis) Dingl. Poly. Jour. 184, 329. Chem. Centrb., 1868, 32. Jahresbericht der Chemie 1867, 981. Schwartz: (Analyses of bauxite) Dingler's Poly. Jour. V, 198, 156. Jahresbe- richt der Chemie, 1870, 1133-II34. Chem. Centrb., 1870, 740. Seger, H.: Zusammensetzung von Bauxit aus Irland. Dingler's Poly. Jour. 1880, ccxxiv, 334. Sena, J. da Costa: Note sur hydrargillite des environs d'Ouro Preto (Brazil). Bull. Soc. Min. de France, 1885, Vll, 220-222. Siemens, C. Wm.: The scientific works of C. Wm. Siemens. Vol. I, Heat and Metallurgy. London, 1889 (Bauxite in the manufacture of iron and steel, 296-298, . 304, 217, 218. : (Furnace lining) Iron and Steel Inst. 1879, I, 153-155. Simmerbach, B.: Gliikauf, XXXI, 234-236; 251-252. Abstr. Iron and Steel Inst, 1895, II, 397. Spencer, J. W.: Geol. Sur. of Georgia; the palaeozoic group. Atlanta, 1893, 210- 230. : How aluminum is obtained from its ores. Science, February 16, 1894 XXXIII, 89. THE BA UXI TE DEPOSI TS OF A RKA NSA S 289 Stolzel, E.: Gewinnung der Metalle (aluminum from bauxites), Vol. II, 1591 BoUey's Handbuch der chemischen Technologic. Braunschweig, 1863-1886. Sutherland, James : The preparation of aluminum from bauxite. Eng. and Min. Jour., October 3, 1896. LII, 320-322. Symes, R. G.: Memoir to accompany sheets 7 and 8 of the maps of the Geol. Survey of Ireland. Bauxite 24; mem. to sheet 14, p. 21 ; mem. to sheet 20, pp. 12 and 29. TissiER, Ch. et Alex.: Guide pratique de la recherche de I'extraction et de la fabrication de I'aluminium et des metaux alcalins. Paris, 1863. Wagner, R.: Handbook of Chemical Technology. New York, 1889. (Alumi- num from bauxite, p. 1 13; alum from bauxite, 259-260. •: Ueber die Bedeutung des Bauxits fiir die chemische Industrie. Berg- u. HUttenmannische Zeitung, 1865, 264. Kunst- u. Gewerbeblatt, 1865, 68. Deutsch. Industrie-Zeitung, 1865, 113, 125. Wagner's Jahresbericht iiber die Fortschritte der chem. Tech, 1865, XI, 323-336. Moniteur Scientifique, 1865, 343. Wedding : Notiz iiber den Beauxit (Niederrhein. Gesellsch. fiir Natur- u. Heil kunde zu Bonn. Sitzg. V. 8. April, 1863. Abstr. Neues Jahrb. fur Min. 1863, 723. Williams, C. H.: (Bauxite) Trans. Manchester Geol. Soc. XXII, 521. Williams, J. F.: (Age, origin and distribution of Ark. bauxite) Igneous rocks of Ark. Little Rock, 1891, 124-125. Maps II and III. Will, W.: Bericht der Oberhess. Gesellschaft fiir Natur- u. Heilkunde. 1883 XXII, 314. : Revue Universelle des Mines et de la Metallurgie No. 2, IX, 1881 (on bauxite in dephosphorization of iron). Abstr. Iron and Steel Inst. I, 1881, 331-332. J. C. Branner. Stanford University. Editorial. As THE time for the meeting of the International Congress of Geologists approaches interest in the series of events con- nected with it increases. The several circulars that have been issued have made evident the extent of the preparations that are being made at St. Petersburg and the largeness of the generosity of His Majesty the Emperor of Russia. Opportunity has been offered for all geologists taking part in the congress to visit many of the most important regions of Russia, presenting widely different geological phenomena- — the crystalline rocks of Fin- land and the glacial phenomena in the north ; the vast expanse of slightly disturbed and little-altered Palaeozoic strata in the central portion of the country ; the far-famed mining districts of the Urals, and the grand scenery and varied features of the Caucasus and of Transcaucasia. Seldom may geologists have such an opportunity to contrast within the brief period of a few weeks the forest-clad, moist north lands with the barren, arid regions of the south, or experience in quick succession the sen- sations produced by the boundless horizon of the steppes and by the deep gorges and lofty summits of the Caucasian Alps. The liberality of the Czar in presenting the geologists with first-class tickets over all the railroads under Russian control is an act of hospitality fitting the ruler of so great an empire, and one fully appreciated by the geologists of all parts of the world. That it should have been accepted by many who would not otherwise have been able to travel so extensively was to have been expected. And it is not surprising to hear that many per- sons not having claims to the title of geologists have applied for the privileges, to the embarrassment of those in charge of the 290 EDITORIAL 291 management of the great excursion. It is proper that a scrutiny of the lists of applicants should be undertaken for the purpose of limiting the numbers to those who are actually geological workers or teachers of the science, and who will derive the most benefit from the instruction which will be given by the geolo- gists acting as leaders of the several excursions. It would be unfortunate if, by the presence of many untrained excursionists, the object of the expeditions should be defeated. It is reason- able, then, that the committee in charge at St. Petersburg should propose to test the geological knowledge of all those wishing to join the excursions, when their attainments in the science of geology is not alread}- known. We understand that those not known to the committee to be geologists will be required to pass an examination in various branches of geology before being permitted to take part in the excursions. A RECENT circular from St. Petersburg calls attention to the fact that there was established at the London congress a per- manent committee that should have charge of the selection and elaboration of questions to be submitted to each subsequent ses- sion, as well as of the preparation of reports relating to such questions. After citing the report of the committee relative to the unification of stratigraphical nomenclature, and after noting the fact that the matter received no attention at the Washington and Zurich meetings, the committee of organization at St. Petersburg suggests that the question of the general principles involved in any stratigraphical classification be discussed, first, as to whether it be artificial, based solely on historical data, or /latitral, based as well on general physiographic changes as on faunal data ; second, as to laws that should govern the introduc- tion of new terms into stratigraphical nomenclature. It is also suggested that the principles that should govern petrographical nomenclature ought also to be considered. The confusion in these nomenclatures which is constantly increasing owing to rapid accession of new facts and to lack of 292 EDITORIAL system in the creation of new terms makes apparent the need of some organization of terms and of the establishment of some general principles of nomenclature. The committee, realizing the magnitude of the undertaking and the shortness of the time of a single session of the congress, nevertheless expresses the hope that enough may be accomplished at the coming session to lead to the satisfactory solution of these problems. Certainly the discussion of the more general and fundamental principles of nomenclature by which a common usage of terms may be brought about is the proper function of an international con- gress of geologists, and it is to be hoped that the proceedings of the coming session may be along these lines. J. P. I. Reviews. SOME RECENT PAPERS ON THE INFLUENCE OF GRAN- ITIC INTRUSIONS UPON THE DEVELOPMENT OF CRYSTALLINE SCHISTS. Michel-Levy : Contribution a V etude du Granite de Flanianville et des Granites Franfais en general. Bull, des Services de la carte geol. de la France, No. 36. Paris, 1893, pp. 41. L. DuPARC and L. Mrazec : Noicvelles Recherches sur le massif du Mont-Blanc. Archives des Sciences Physiques et Naturelles. Tome XXXIV, 1895, pp. 39. L. UuPARC : Le Mont-Blanc au point de viie geologique et petrogra- phiquc. Ibid. 1896, pp. 8. J. Vallot and L. Duparc : Sur un Synclinal schisteux ancien forniant le coeur du massif du Mont-Blanc. Comptes Rendus des Seances de r Academic des Sciences. Paris, Mars 1896, pp. 3. J. Horne and E. Greenly : On Foliated Granites and their relation to the Crystalline Schists in Eastern Sutherland. Quarterly Journal of the Geological Society. London, Vol. LII, 1896. A. Sauer : Geologische Specialkarte des Grossherzogthums Baden, Erlauterungen zu Blatt Gengenbach. Heidelberg, 1894, pp. 87. A. Sauer : Ibid. Erlauterungen zu Blatt Oberwolfach-Schenkenzell. Heidelberg, 1895, pp. 76. F. ScHALCH : Ibid. Erlauterungen zu Blatt Petersthal-Reichenbach. Heidelberg, 1895, pp. 82. A. Andrew and A. Osann : Ibid. Erlauterungen zu Blatt Heidelberg. Heidelberg, 1896, pp. 60. G. Klemm : Beitrage zur Kenntniss des krystalliiien Grundgcbirges im Spessart. Abh. der Gross. Hessischen Geologischen Landesan- stalt zu Darmstadt. 1895, pp. 87. In one of the oldest and best known of the German universities there is delivered annually a very able and exhaustive course of lectures on petrography, in which no less than fourteen lectures are devoted to a presentation of our knowledge of the single rock, granite, but it is 293 294 REVIEWS nevertheless somewhat depressing to consider on how many important questions connected with this, the most common and ordinary of all the plutonic rocks, there are still grave differences of opinion among those justly considered as authorities. Thus all do not agree even as to the order of the crystallization of the constituents of the rock, some holding that there are two generations and others that there is but one, while again a marked difference of opinion exists concerning the effects produced by granite upon the rocks through which it is intruded. As the result of a whole series of careful studies on various contact zones, chiefly in Germany, Austria, and Scandinavia, it is commonly believed in these countries that the granite magma, by its heat, pres- sure and escaping vapors, causes a recrystallization of the country rock, the process being one of diagenesis, the granite giving nothing to the rock through which it breaks, except in places, perhaps, a small amount of boracic acid. In France, however, and everywhere within the French "sphere of influence" different opinions prevail, and it would actually appear from the studies made in these parts of Europe that the laws of nature changed upon crossing the political boundaries. Contact zones have been described by Barrois, Michel-Levy, Delage, and other French petrographers, in which the country rock adjacent to the granite has become completely "granitized" by the transfusion of granitic material into it, and in a well-known paper by Michel-Levy, which appeared in 1887,' he stated it as his belief that by this piocess gneisses, leptynites, dolomitic schists and amphibolites, indistinguishable from those of the Archaean, are produced, and that in fact the so-called primitive rocks have really originated in this way, by the intrusion of igneous rocks into clastic sediments, which sediments have undergone a profound metamorphosis with the addition of an immense mass of material, the process being essentially one of a metasomatosis. In the first section of the Bulletin, whose title is given above, Michel-Levy describes an additional contact zone of this kind occurring about the granite of Flamanville, which granite cuts shales, sandstones, and quartzites, chiefly of Silurian and Devonian Age. On approaching the granite the shales present successively the usual zones of the spot- ted clay slate, micaceous clay slate, and hornstone, but in the vicinity of the actual contact they are broken up, eaten into, and partially dis- ' Sur I'orijine das terrains cristallins primitifs. Bull. Soc. Geol. de France, ^'T s.Siy, 103. REVIEWS 295 solved bv the granite, which holds many inclusions and also injects itself into the rock in narrow veins — lit par lit — along its plains of lamination or foliation, altering it intensely and at the same time giving to it a granitic character. In the second half of the Bulletin the author presents some observations on granites in general, more especially in relation to their contact effects, reafifirming and some- what enlarging upon the views put forward by him in 1887. He holds that the conclusions arrived at by Rosenbusch and commonly held by petrographers, that feldspar is not usually produced in contact zones except in comparatively small amount by the diagenesis of the altered rock, and that there is no transfusion of granitic material into the invaded rock, although true of the individual contact zones investi- gated, are not true of contact zones in general, but that there is fre- quently developed immediately along the contact a zone in which an intimate admixture of the granitic material with that of the injected rock is a dominant characteristic. This admixture is brought about in part by the injection of the granite in thin layers — lit par lit — into the stratified rock parallel to its lamination, in part by a transfu- sion in some obscure way by means of mineralizing solutions of the elements of quartz and feldspar through the schists, causing these minerals to crystallize out through the substance of the altered rock, and in part by the actual solution of the injected rock in the granite magma. In this zone, which in ordinary granite intrusions is usually but a few yards in width, there can be found all the rocks characteristic of the great regions of crystalline schists — mica schists, granulites, gneisses, amphibolites, etc., formed by the action of the granite on the ordinary sedimentary strata of the earth's crust. This zone, moreover, although often narrow as exposed, where the deeper seated parts of the "appareils granitiques" are laid bare to our studv bv the process of denudation, is found to become much wider and often of great stratigraphical importance. Barrois is cited as having found in Brittany cases where there has been an undoubted transformation of whole districts of Cambrian schists into gneiss by the process of " granitization " above referred to, it being possible to follow bands of quartzite which resist the general " feldspathization " from the margin of the area into parts of the district where the asso- ciated schists have been completely transformed into gneiss. The work of Duparc and Mrazec, on the massiv of Mont Blanc is also cited as affording conclusive demonstration of similar transforma- 296 REVIEWS tion, while the gneissic zone about the granulite of the Saxon Granu- litgebirge is cited as another case in point. It is beHeved that the granite magma first rises along lines of fracture in the crust. Its presence leads to a heating of the rock into which it is injected, and its intrusion is accompanied by a ^'circulation intense'''' of mineralizing fluids, probably rich in alkalis. These produce at first a transference of quartz from one part of the mass to another and the development of biotite, which is a marked feature in contact zones. Then follows "feldspathization," which commences by the development of little strings of quartz and feldspar following for the most part the schistos- ity of the invaded rock, and which grow in size until the whole mass of the schist is transformed into granite, the texture of the schist being broken down and its elements set in motion to form with the trans- fused material new combinations. The granitic magma or emanations thus slowly dissolve, alter or incorporate, whichever we may choose to call it, the wall rock, transforming it first into a gneiss, then into a gneissic granite, and finally into a granite. The original intrusion thus slowly enlarges its boundaries and increases its volume. This process, we are told, is at work wherever granitic magmas come in contact with clastic rocks in the deeper parts of the earth's crust, and it is thus that, as before mentioned, the crystalline schists are produced. The granite does not therefore, as Suess has supposed, fill great cavities in the earth's crust which have been produced by tan- gential stresses, thus giving rise to batholites, but starting from some line of fracture eats its way into the surrounding rocks and develops itself largely at their expense in the way above described. According to Professor Duparc and his associates, this process of granitization plays a very important role in the development of the crystalline rocks of the Mont Blanc massiv. This massiv has usually been considered as composed of protogine, that is of a somewhat altered granite, massive in the center and progressively more gneissic or schis- tose as the outer portions are approached, the whole enveloped by a mantle of mica schists. These mica schists contain bands of amphib- olite, eclogite, and other similar rocks found in corresponding posi- tions about other protogine masses elsewhere in the Alps. Messrs. Vallet and Duparc have however found that the central part of the massiv is composed largely of various micaceous gneisses and crystalline schists, associated with and invaded by the protogine and even passing into a protogine gneiss. Some of these included rocks REVIEWS 297 are very basic and cannot therefore, it is thought by the authors, be in any way considered to be derived from the protogine by dynamic action, but are to be considered as sediments, depressed by a syncHnal fold and bounded by the protogine on either side. The whole series of rocks, both protogine and surrounding schists, are penetrated by a series of more recent granite veins or dykes, and these it is believed have brought about the profound metamorphism of the surrounding rocks, injecting and "granitizing the schists everywhere in the vicinity of the protogine, so that the gneissic zone which immediately borders the protogine is not in any way connected, genetically, with the pro- togine itself, but results from the profound alteration of the mica schists surrounding the protogine by these newer granite dykes. The varying character of the different schistose rocks in this gneissic zone is considered to be due to the varying resistance offered to this "granit- izing" action by the different beds in question. Thus, for instance, the eclogites retain their basic character and have not been transformed into orthoclase gneiss, because they are too compact to allow of -a free circulation through them of the solutions producing the alteration. In the paper by Duparc and Mrazec, a number of analyses of the several varieties of protogine and granitized schists are given. The crystalline schists of eastern Sutherland, described by Messrs. Home and Greenly, consist of a series of gneisses, granulites, mica schists, etc., some few members of which show conclusive evidence of a sedimentary origin while the origin of others is doubtful. The whole series has been intensely deformed. Not a cubic inch can be found which has not suffered deformation, but distinct cataclastic structure is not seen, so that recrystallization must have taken place during or after the movements. The series is invaded by masses of intrusive granite, which have broken across the schists, anastomozing through them and often penetrating them as a series of thin leaves, parallel to their foliation, in the manner termed by the French writers /it par lit injection. The boundary between the injecting granites and the schis- tose series is often rather ill-defined, owing to the fact that the granitic constituents seem to interlock with those of the wall rock with which they are in contact. The granite never shows any finer grained sahl- band, indicating injection to a cold rock, but is usually coarse-grained and pegmatitic on the borders. It seems reasonable to infer, therefore, that the igneous material was introduced when earth movements were in progress and when the countrv rock was at a high temperature. 298 REVIEWS On approaching large masses of granite, the schistose series becomes more highly altered, sillimanite and other contact minerals making their appearance, but the invaded rocks often take upon them- selves a character so closely resembling that of the invading granite as to "amalgamate the two rocks into one great gneissose complex." Thus the foliation of the invading granite, which can often be seen to be parallel to that of the invaded gneisses, is in many cases certainly due to the biotite folise of the latter, having retained their original position, while the associated "quartzo-feldspathic elements have been incorporated with those of the granite," as every gradation can be traced from inclusions retaining their natural orientation to the merest trains of mica fiakes in a granitic rock. In other cases, however, the foliation of the invading granite does not coincide with that of the invaded gneiss but cuts it. Powerful movements were the "initial cause of the whole series of phenomena With regard to the granites, it is difficult to believe that they are wholly foreign matter; though here it is necessary to observe the utmost caution, the chemical difficulties being so great." Although Messrs. Home and Greenly are guarded in their statements, their studies being rather general in char- acter, it is clear that they believe the processes at work to be very similar, if not identical with, those described by Michel-Levy and Duparc. The views put forward in these papers lead us back to the time of Hutton, who, in his Theory of the Earth states that the kind of gran- ite which shows banding and foliation is probably an altered sediment, the foliation being a survival of the bedding of the original rock. This fact, however, does not by any means discredit the view as manv of Hutton's opinions after long neglect have finally proved to be correct. The views also have certain features in common with the crenitic hypothesis of Hunt. The whole process is, however, very recondite and mysterious in character. One of the great difficulties in the way of the acceptance of these views is the absence of chemical proof. In those contact zones on which accurate chemical work has been done, it has been shown that no considerable transference of material has taken place. In these other cases where this enormous transference of material is assumed^ no accurate chemical work seems to have been carried out to support the contentions. In Duparc's work, as has been mentioned, a number of analyses of various normal and "granitized" rocks are given, but REVIEWS 299 no attempt is made to follow out the changes undergone by a single bed, and it is impossible to make out in how far the differences in composition shown to exist, are primary differences in the composition of the rocks analyzed. The chemical evidence adduced is, therefore, by no means conclusive. The question also arises as to the ultimate source of the enormous amounts of silica and alkalis required for the conversion of hundreds of cubic miles of the miscellaneous rocks of a sedimentary system into granite. It is furthermore a question as to how far dynamic action is respon- sible for many of the phenomena described. When, for instance, a schist is shattered and granite is intruded into the cracks and fissures, masses of the invaded rock being found scattered through the granite, and, after cooling, the whole complex is stretched or rolled out by dynamic movements, as is usually the case in districts where ciystalline schists occur, the injected arms of the granite, great and small, become pulled apart and eventually appear as little discontinuous strings and lumps of quartz and feldspar in the enclosing schists, following the line of movement, while a schistose structure parallel to these strings is given to the whole rock by the same movements. In certain parts of the Laurentian of Canada, schists and gneisses are found full of such strings and lumps of quartz and feldspar, pre- senting exactly the characters described by the French petrographers as resulting from the granitization of sedimentary rocks. The Canadian rocks, however, have undoubtedly been produced in the way just described, every possible transition from the massive injection to the foliated complex being observed in a hundred different cases. In the Lepontine Alps, moreover, to the east of Mont Blanc, where the Schie- ferhulle of the several protogine masses have been very carefully studied by Heim, Schmidt, and many other observers, the phenomena attributed to "granitization" by Duparc are everywhere considered to be the results Of crushing under the influence of such movements, with, in certain cases, the infiltration of secondary cracks and rifts by materials deposited from ordinary terrestrial waters, which in such positions would probably be more or less heated. Even in the Saxon Granulitgebirge, cited by Michel-Levy as a case where transference of material could be distinctly observed, and where certainly the granulite does seem to have eaten its way into the schists, only however for a short distance back from the immediate contact, the appearance pre- sented bearing a striking resemblance to that, very familiar to the tyro 300 REVIEWS in assaying, when his slag being too basic has eaten its way into the clay crucible appearing when the crucible is broken through its substance here and there in spots and streaks ; Lehmann, who has made a most exhaustive study of the whole region points to the fact that this zone is not always present, and states it to be his belief that the granitic material forming the flammen in the schist has not been derived from the granulite magma, but is due to the deposition of the granitic material in spaces formed by the separation of the foliae of the schist, under the great stresses to which the region has been subjected, the granitic material in question having been derived from later intrusions which elsewhere can be clearly seen to cut the granulite. From per- sonal observation, however, I must say that appearances are strongly in favor of Naumann's view, that along the narrow zone of the imme- diate contact the granulite magma has eaten into the schist to a certain distance, a phenomenon which is quite intelligible and perhaps in cer- tain cases to be expected, but which is quite distinct from the whole- sale transformation of the schist into granite by the mysterious process of " granitization." In how far this process of transfusion which is considered by Michel- Levy and other French geologists to play so important a part in the origin of the crystalline schists is really active, must be determined by detailed studies of the deeper seated granite contacts and of the so-called Archean areas in various parts of the world. Such studies in the case of Archean areas are "presented in the recent maps, with accompanying explanatory texts, issued by the geo- logical survey of Baden, and whose titles are given above. This survey, following the lead of those of Prussia, Saxony, and Hessen, was con- stituted in 1888 for the purpose of mapping in detail the Grand Duchy of Baden, an area of 5843 square miles, on a scale of -^-^\-^-^- For this purpose the territory in question has been divided into 170 sections. Work was begun in 1889 and maps of twelve sections have already been published. As about one-quarter of the Grand Duchy of Baden is underlain by Grundgebirge including the well-known area of the Black Forest, ample opportunity is given for a thorough study of these ancient rocks. Six of the maps already published are in areas of the Grundgebirge ; of these three have been mapped by Dr. Sauer, to whom we are already indebted for his valuable contributions to our knowl- edge of the ancient crystalline rocks of Saxony. The maps are among the best which have yet appeared of any Archean region, and serve to REVJEWS 301 bring out clearly the complex relations of the several members of the system. Two distinct classes of gneissic rocks are recognized in the areas examined, in addition to the numerous intrusions of igneous rocks of various kinds. To each of these classes a collective name has been assigned, taken from a locality where it is well exposed; the first being known as the Rensch Gneiss and the second as the Schapbach Gneiss. A few of the most notable varieties of each class of rock are distinguished in mapping, but no attempt is made to map separately the bewilderingly numerous and minute petrographical variations of the gneissic rocks attempted in the survey of Saxony. The Rensch gneisses consist chiefly of orthoclase, biotite, and. quartz. The mica is usually abundant, and sillimanite is a characteristic accessory con- stituent, often occurring as a paramorph after andalusite. Garnet and sphene are seldom found. The rock often shows in the arrangement of the constituents a structure similar to that seen in the hornstones of contact zones. The presence of small lenticular segregations of quartz, or of quartz and orthoclase, scattered through the rock is also a char- acteristic feature of the gneisses of this class. Conformably inter- banded with these Rensch gneisses are subordinate masses of quartz schist, graphitoid schists and gneisses, pyroxene gneisses and amphib- olites. The Schapbach gneisses are much more uniform in character, usually poorer in mica, and have a marked tendency to assume a granitic aspect, often passing over into granulites. Quartz lenses and fine-grained, highly quartzose bands are absent, but garnet and orthite are frequently present. The only inclusions found in the gneisses of this class consist of a gabbro-like amphibolite. The gneisses of the two classes sometimes seem to pass into one another along the contacts, but the distinction is usually sufficiently well marked to enable them to be properly separated in mapping. Although no direct expression of opinion concerning the origin of these gneisses is given in the publications in question, it seems to be the opinionof the survey that the two classes probably differ in origin, the Rensch gneisses representing highly altered sediments and the Schapbach gneisses being of igneous origin. The chemical evidence afforded by a number of analyses of typical gneisses of each series which are given, tend to support this view as does also the structure of the rocks, and the character of the subordinate intercalated masses in 302 REVIEWS the case of the Rensch Gneiss. The evidence one way or the other will, however, be greatly extended as the mapping with concomitant chemical investigation progresses, and the Director of the survey. Pro- fessor Rosenbusch, evidently desires to await this further evidence before making any decided statements concerning the genetic relation- ships of the complex. If the Rensch gneisses prove to be altered sedimentary rocks their high content of feldspar and the presence in them everywhere of lenticular masses and strings of quartz and feldspar will certainly be cited by the French authorities as evidences of "grani- tization." But two questions remain to be decided — first, whether the high content in feldspar is not due to a high content of alkalis in the original sedimentary rocks, these having been perhaps of the nature of feldspathic sandstones, arkoses and greywackes, and secondly, whether the strings and lenses of quartz or quartz and feldspar do not fill spaces opened by the dynamic movements to which the rocks have been subjected, quite independent of any granitic intrusion. Whether in fact any mysterious cementation-like transfusion of granitic material through these rocks has really taken place. The detailed chemical work which is now being carried out will, when completed, undoubtedly decide whether the supposed altered sediments have or have not a composition which can be attributed to a sedimentary series. A similar twofold origin is claimed by Klemm for the crystalline Grundebirge of the Spessart, although here the sedimentary portion is believed to be of late Palaeozoic age and is possibly equivalent to a series of schistose hornstones, graphite schists and garnet rocks, quite distinct from the gneissic series of the Black Forest, which were found by Andrese and Osann in the Odenwald to the north of Heidelberg. These studies bearing upon the vexed question of the origin of the crystalline schists have at present an especial interest for petrographers in America, where such enormous areas of these rocks are now under investigation. ' Frank D. Adams. Glaciers of North America, a Reading Lesson for Students in Geog- raphy and Geology. By Israel C. Russei l. Boston : Ginn & Co., 1897. The preparation of a work of this high grade by a busy university professor of large professional experience and demonstrated investi- gative ability, as a reading lesson for students of geography and RE VI J:: IV S ' 303. geology, is worthy of special note as one of the signs of the educa- tional times. It is significant both as an indication of a demand and as exemplifying a supply. It is a gratifying mark of progress that there should have grown to be a place for a work of this character as a supplement to the usual treatises on geography and geology. It is a not less gratifying mark of progress that such a demand should be appreciated and met by a careful and competent scientist of high position. The work opens with a clear and brief statement of the nature of glaciers, and of their varieties and of the work done by them. Their distribution in North America is then sketched comprehensively, after which individual glaciers and glacial districts are described in detail. It is in the study of these glaciers individually, aided by the numerous photographic illustrations, that the real characters of glaciers will come to be realized by the students. The average reader will doubtless be surprised at the number, variety and instructiveness of American glaciers. They very greatly surpass those of all other accessible conti- nents. Following the individual descriptions are chapters on the climatic changes indicated by the glaciers of North America, upon the cause and mode of glacial motion, and upon the life history of a glacier. The discussions of theoretical questions are conservative and judicial in tone, and manifest a notable tendency to eclectic conclusions. Pro- fessor Russell's comprehensive statement of the various hypotheses of glacial motion will doubtless be found one of the most interesting sections of the volume by advanced glacial students. The work is heartilv commended to teachers and general readers as well as students. T. C. C. Former Extoision of Cornell Glacier near the SoJithern end of Melville Bay. By Ralph S. Tarr. Bull. Geol. See. Amer., Vol VIII, pp. 251-268. Plates XXV to XXIX, March 1897. An abstract of this paper was given in the January-February num- ber of this Journal, pp. 95-96. An editorial relative to it appeared in the same number, pp. Si-85. C(5mmunications in reference to it have also appeared in Science, Vol. V, No. 113, February 26, p. 344; No. 114, March 5, pp. 400-401, and No. 117, March 26, pp. 515-516. This further notice is introduced mainly for the purpose of presenting RE VIE I VS 305 to the readers of the Journal the chief photograph of the paper for comparison with those previously presented in this Journal in the series entitled "Glacial Studies in Greenland." It need only be added that Professor Tarr regards this photograph as presenting a "rugged, angular topography," and offers it with similar photographs in sub- FlG. I. Dalrymple Island, an illustration of unglaciated topography. stantiation of his claim that topographic contours cannot be trusted as indices of glaciation. As glaciation seems to the reviewer to be expressed with much clearness and definiteness in the contours of the promontory here cited as a proof to the contrary, it seems the fairest mode of review to reproduce the photograph and permit geologists to judge for themselves. For comparison there is added a photographic illustration of Dalrymple Island, which has been previously published in this JOURNAL as an illustration of unglaciated asperities. The reviewer does not see how anyone trained in glacial topography can fail to see glacial modification in the one and the absence of it in the other. The humorous feature of the issue raised in the paper and the outgrowing discussion is that contours of the type illustrated by Professor Tarr's photographs were identified as moderately glaciated by those whose conclusions he seeks to overthrow and his observa- 306 REVIEWS tions of the drift and other positive indices confirm the correctness of their identifications. It is infelicitous to call the promontory in the photograph the Devil's Thumb. The author remarks in a footnote: "This is the Devil's Thumb as given on the Danish and British Admiralty charts. Fig. 2. Devil's Thumb, S.E.^E. (True) S.S.W.>^W. (Mag=.) (Legend on British Admiralty Chart.) The real Devil's Thumb of the Arctic explorers is some forty or fifty miles north of this" (p. 254). The true Devil's Thumb is, however, sketched on the British Admiralty chart and the sketch there given is herewith reproduced. It is topographically an object of a very dif- ferent order from the promontory of the photograph. As it has been cited in the articles in this Journal in its bearings upon the limita- tions of glaci'ation, it seems unfortunate to introduce another Devil's Thumb of so different a nature. Confusion has already arisen by rea- son of this. An error in the location of the Devil's Thumb, in a region where the charts are confessedly inaccurate, does not seem to us to justify the transfer of the name to the false location. The author makes passing mention of the driftless area in the Inglefield Gulf region, and although he declines to discuss it, as it was not seen by him, he remarks in a footnote that "he cannot let this opportunity pass without raising the query whether the topography in the neighborhood of the Greenland driftless area is not such that an area of this sort would naturally be expected. Was not the move- ment of the ice outward and the main stream down the Inglefield Gulf ? And is not the driftless area located in the place where the high Red Cliff peninsula would naturally have clogged the ice and hence prevented its action of erosion and notable transportation ?" The driftless area is part of the same ancient peneplain as the summit of the Red Cliff peninsula (Jour. Geol., pp. 205-206, Vol. Ill, 1895). It lies on the east side of Red Cliff peninsula (see map on p. 668, Vol. II, Jour. Geol.). It Xio.'s, between it and the great ue-cap. It is separated from the peninsula by the valley of Bowdoin Bay about two miles wide and 2000 feet deep. How an isolated part of a peneplain can protect from glaciation another part of the same plain lying between REVIEWS 307 it and the source of the glacial motion, and several miles distant, is not easily understood. The suggestion does not seem to have sprung from a serious consideration of the conditions of the problem. T. C. C. Report 071 the Valley Regions of Alabama { Pa Icsosoic Strata^. Parti. Oj) the Tennessee Valley Regions. By Henry McCalley. The palaeozoic area in northern Alabama may be divided into two regions. That to the northwest, drained chiefly by the Tennessee River and its tributaries, is characterized by the nearly horizontal and undisturbed condition of the strata. The region to the southeast, drained by the Coosa River, is a region of greater disturbance with the geologic features much complicated by the folding and faulting of the strata. The first of these regions is described in the present report; part two of the report, not yet published, will treat of the second or Coosa valley region. The report is divided into two sections, the first of which is a gen- eral treatment of the physical features, geology, natural resources, soils, agricultural features, timber, waterpower, climate, rainfall, drainage, and health. Much the larger part of the report is devoted to the sec- ond section which treats of county details. Chapter two of the first section entitled Geology describes the stratigraphy of the region and is the only portion of the report of gen- eral interest. The following table of formations is given : (8) Tertiary (k) Lafayette (7) Cretaceous (j) Tuscaloosa (6) Carboniferous (i) Coal Measures 200-500 feet (5) Upper Sub-Carboniferous \ f\ Bangor limestones, 200-450 feet ^■^' '' ((to) Hartselle sandstones 150-400 feet r(f) Tuscumbia or /NT c 1 r- I f I St. Louis limestones 75-200 feet (4) Lower Sub-Carboniterous - / n t j 1 1 ^ ' \ (e) Lauderdale or 1^ Keokuk chert 100-250 feet (3) Devonian (d) Black shale 0-45 feet , -, TT fi • \ (c) Red Mountain or (2) Uiiper bilunan -, ^ ' r-i- ^ /m- \ , r ^ ^ ^ '' / Chnton (Niagara) 3-350 feet f (b) Pelham or ( ■,\ T ^„.^,. c;i,,,.;o^ Trenton (Nashville, 700-1000 feet (i) Lower hdurian ■<^ , > , ' ' ^ I (a) Siliceous (Knox) [^ dolomite and chert 2000 feet 308 REVIEWS The physical characters of each of these formations is briefly described, though no lists of fossils are given by which the faunas may be compared with those of the supposed equivalent strata elsewhere. The remainder of the report will have its chief interest and useful- ness among the local geologists of the region described. Stuart Weller. Fijial Report 071 tJie Geology of Minnesota. Pal(2o?itology . Vol. Ill, Part II. Minneapolis, Minn., 1897. Part II of the PalcEontology of Minnesota, a volume of about 600 pages, illustrated by forty-eight plates besides 133 figures, has just appeared. Like its companion volume, Part 1, it treats only of the Ordovician fossils. Besides the introduction, which is a paper by N. H. Winchell and E. O. Ulrich upon "The Lower Silurian deposits of the Upper Mississippi province ; a correlation of the strata with those in the Cincinnati, Tennessee, New York, and Canadian provinces, and the stratigraphic and geographic distribution of the fossils," the volume contains the following papers : 1. The Lower Silurian Lamellibranchiata of Minnesota. By E. O. Ulrich. 2. The Lower Silurian Ostracoda of Minnesota. By E. O. Ulrich. 3. The Lower Silurian Trilobites of Minnesota. By J. M. Clarke. 4. The Lower Silurian Cephalopoda of Minnesota. By J. M. Clarke. 5. The Lower Silurian Gastropoda of Minnesota. By E. O. Ulrich and W. H. Scofield. The first three of these papers were published separately in small editions and distributed during the period from June 16, 1894,10 Sep- tember 27, 1894. The last two papers appear for the first time with the publication of the complete volume. The volume supplies a long-felt want to students of the Ordovician faunas of the West, and will doubtless be the standard work upon these faunas in the Mississippi region for many years. The author of the chap- ters upon the Lamellibranchiata and Gastropoda should perhaps have been more conservative in establishing new genera and species; how- ever the classification of the greater number of the classes of organisms abundantly preserved as fossils in the palaeozoic rocks, is at present in a transition state, and any attempt to make more natural, and to give more definiteness to their classification is a step in advance. Stuart Weller. REVIEWS 309 Btilletins of American PalcBontology. Vol. I. Published by Pro- fessor G. D. Harris, Ithaca, N. Y. The publication of a purely palaeontological bulletin has been undertaken by Professor G. D. Harris, of Cornell University, and the first volume has just been completed. The undertaking of Professor Harris is truly a commendable one, and should receive the encourage- ment of American palaeontologists. Heretofore palaeontology has had, in America, no organ of publication purely its own. The'literature of the subject has been scattered through a score or more of proceedings or transactions of learned societies and periodicals. If, from now on, Professor Harris' bulletin meets with the cooperation of American investigators which it deserves, our literature will become more con- centrated and consequently much more accessible. The bulletin appears, not at stated intervals, but at such times as material may be ready for publication. Volume I consists of five numbers, which have appeared at intervals since May 25, 1895. 1. "Claiborne Fossils," by G. D. Harris. Pp. 1-52. Plate I. (May 25, 1895.) Part I of this bulletin is a "Synonymy of the Claiborne sand species of Conrad and Lea, as determined by an inspection of the type collections now at the Academy of Science of Philadelphia." This paper will be welcome to all students of the Eocene faunas of eastern America. The description of species from the Claiborne sands by the two authors, Conrad and Lea, during the same period of time, and without the slightest recognition on the part of either of the work being done by the other, brought about a most confusing state of synonymy. Professor Harris has straightened out this confusion by an inspection and comparison of the type specimens of both authors. Part n of the bulletin is a description of six new species from the Claiborne sand. 2. " New or little known Tertiary Mollusca from Alabama and Texas," by T. H. Aldrich. Pp. 53-82. Plates H-VI. (June 24, 1895.) 3. "Neocene Mollusca of Texas, or fossils from the deep well at Galveston," by G. D. Harris. Pp. 83-114. Plates VH-X. December 2, 1895.) This paper was noticed in this Journal, Vol. IV, p. 126, and needs no further comment. 4. "The Midway Stage," by G. D. Harris. Pp. 115-270. Plates XI-XXV. (June 11, 1896.) 310 . REVIEWS In this paper it is shown that between the basal Eocene deposits, or Midway Stage, and the uppermost Cretaceous, there is in the south- ern states a decided break, both stratigraphic and faunal, so that not a single species is known certainly to have crossed from one formation to the other. These initial beds of the Eocene are treated both geo- logically and faunally by Professor Harris. The field investigations were carred on in the states of Texas, Arkansas, Tennessee, Mississippi, Alabama, and Georgia. From his studies Professor Harris is led to believe that a considerable time interval elapsed between the close of the Cretaceous deposition and the beginning of the Eocene deposition in the Mississippi basin, and that wherever good contact exposures are found, there may be found, on careful study, ample evidence of non- conformity. This initial Eocene fauna is discussed at length, all the old species of Mollusca and many new ones are described and figured. 5. "A reprint of the palseontological writings of Thomas Say; with an introduction, by G. D. Harris. Pp. 271-354. Plates XXVI- XXXII. (December 7, 1896.) The republication of these papers, originally published from 1819 to 1825, long since out of print and accessible only in the larger libraries, will be appreciated by all those who have had occasion to refer to such literature, and have been unable to find access to it. The following papers are republished, word for word, line for line, and page for page, as written and punctuated by the original author : I and 2. " Observations on some species of zoophites, shells, etc., principally fossil. Am. Jour. Sci., ist ser., Vols. I and II (1819-1820). 3. "Fossil shells found in a shell mass from Anastasia Island." Jour. Acad. Nat. Sci., PJiil., ist ser.. Vol. IV (1824). 4. "An account of some fossil shells of Maryland." Jour. Acad. Nat. Sci. Phil., ist ser.. Vol. IV (1824). 5. "On two new genera and several species of Crinoidea." Jour. Acad. N'at. Sci. Phil., ist ser., Vol. IV (1825). Stuart Weller. Eocene Deposits oj the Middle Atlantic Slope in Maryland, Delaware arid Virginia. By Wm. Bullock Clark, U. S. Geol. Sur., Bull. 141, 167 pp., 40 pis., 1896. During the first half of the century the Tertiary formations of eastern United States were among the most carefully studied and best REVIEWS 311 known deposits of the country. However, for nearly two generations little has been done to extend our knowledge of these interesting beds. Since the brilliant work of Rogers and Conrad ceased practically nothing has been attempted in the way of keeping the information regarding these strata abreast of the times. It is, then, with peculiar pleasure that the recent revival of interest in these formations is noted ; and none of the late contributions is more welcome than the one just issued, on the Eocene of the Middle Atlantic region. The memoir contains a complete review of the literature and results of past observations in this field. The author exhaustively investigates both the stratigraphy and the fauna of this important member of the coastal plain series. He traces the limits of the forma- tion from its most northern occurrence in Delaware across Maryland into Virginia, where it gradually becomes buried beneath later forma- tions. A detailed study of the 300 feet of Eocene deposits in the central portion of the district shows two distinct faunas, which are named the Aquia Creek and the AVoodstock faunas, the former occupying a sequence of beds extending some 60 feet from the base of the forma- tion, while the latter apparently does not reach quite to its upper limits. The Aquia Creek stage, which contains an assemblage of forms closely allied to the middle Lignitic, probably stands, with its under- lying poorly fossiliferous zone, as an equivalent, in a broad way, of the whole of the Lignitic of the Gulf ; while the Woodstock stage, which contains a group of forms closely allied to the Ostrea sellceforviis zone of the Claiborne, stands, with the overlying and underl3ing beds, as the equivalent, in a broad way, of the Buhrstoneand Claiborne, yet it is not assumed that the lower and upper beds are exactly synchronous with the lowest portions of the Lignitic and the highest portions of the Claiborne. The much slower accumulation of the Atlantic coast materials is shown in the fact that in Alabama more than 600 feet of deposits are found between the two fossiliferous horizons above cited, while in Maryland and Virginia a thickness of but little over 100 feet is found, and without the differentiation into the fossiliferous zones which characterizes the Gulf area. The middle Atlantic Slope Eocene, therefore, represents, according to the author, only the Lignitic, Buhr- stone, and Claiborne of the Alabama geologists. These results, together with others lately obtained in a study of the Cretaceous strata throw much light upon the character of sedimentation 312 REVIEWS along the Atlantic coast during the late Mesozoic and early Cretaceous time. Regarding the criteria of correlation which were followed the author significantly remarks : "As the different methods of correlation are examined in retrospect, the interdependence which exists between the various classes of physical and biological criteria becomes clearly manifest. "The faunal and floral characteristics of a formation find their full interpretation only as the physical factors are clearly understood, since the geological and geographical range of forms is determined to a large extent by conditions of sedimentation. The physical characters of a formation therefore bear a close relationship to its contained fossils, and cannot be ignored in the correlation of the deposits. "Although the most trustworthy correlations are based upon palseontological data, the possibilities of variation in the succession of organic forms, in two distant areas, are so great that detailed correla- tions can seldom be satisfactorily attempted, even where general equivalence is recognized. "The geologist, therefore, must take into consideration both the geological and the palseontological criteria in the correlation of the sedimentary rocks. No class of facts can be ignored." Charles R. Keyes. The Elevated Reef of Florida. By Alexander Agassiz. With Notes 071 the Geology of Soutliern Florida. By Leon S. Griswold. Bull. Mus. Comp. ZooL, Vol. XXXVIII, No. 2 (pp. 29-62, 26 plates). The work upon which this double paper is based consists of a trip made by Mr. Agassiz somewhat over two years ago, and one made by Mr. Griswold early in 1896. The purpose of the latter was to clear up if possible, some of the obscurity which surrounds the geology of the Everglades. The paper opens up much that is new in the story of the organic portion of the peninsula; and it is to be regretted that low water prevented Mr. Griswold from reaching Long Key, which was one of the most important goals. The reef has been elevated from six to twenty feet, the amount decreasing southward. At Key West the coastal plain is found at a depth of 50 feet (Pliocene), while Eocene strata are 700 feet from the REVIEWS 313 surface. At Key Largo the reef has "a probable width of at least nine miles from the outer reef patches." Very great weight is given to solu- tional action. Not only are the Everglades in great measure explained bv it, but the sounds and mud flats between the line of keys and the mainland. The disconnection is increasingly great southward, until between Ke}- West and Cape Sable the mud flats and mangrove islands "are all that give evidence of the former continuity of the land." Thus Florida is constantly losing territory, instead of gaining it, as the older views maintained. The Marquesas, formerly considered an atoll built upon a raising bank, are, according to the author, merely a very perfect sound, and the "lagoon" is an area of greater solution. Mr. Griswold, however, does not carry the solutional theory to the extent to which it is borne by the senior author. In manv places on the keys the elevated reef is capped by an oolite. Mr. Agassiz states positively its Eeolian origin ; but Mr. Griswold, who was last on the ground, says that, for several reasons, which he enumerates, "the topography favors an origin for the limestone in water." This guarded statement does not seem too strong for the evidence presented. This oolite was traced by Mr. Griswold much farther into the Everglades than any geologist has penetrated hereto- fore. Mr. Agassiz interprets the formation as modified dune sand, obtained from the beaches of the elevated reef, and blown inland by prevailing east winds. J. Edmund Woodman. Correlation of Eric-Hiiron Benches with Outlets and Moraines i?i Southeastern Michigan. By F. B. Taylor. Bull. Geol. See. America, Vol. VIII, pp. 31-58, with map. In this paper there is given an account of observations in a region hitherto little studied but containing evidences of a most interesting glacial and postglacial historv. As stated in the title, the paper deals with features observed in southeastern Michigan chiefly in the north- ward projecting portion between Saginaw Bay and the lower end of Lake Huron, commonly called the "Thumb." Having spent two months during the past season in geological investigations in this region the reviewer is prepared to lend corroborative evidence as to the general conclusions of the paper and to testify to the scientific acumen displayed in the interpretation of the phenomena observed. The paper constitutes the first published account of detailed observa- 314 REVIEWS tions of glacial phenomena in this portion of the state. Almost the entire region is heavily covered with glacial drift. Very few outcrops of country rock occur. All the moraines of the region show two dis- tinct phases. Above a contour of approximately 200 feet, which marks the upper limit of submergence, they have an irregular rolling surface, while below this line they lose their irregular features and are in fact so subdued as often to be distinguished with difficulty from the ground moraine. These water-laid moraines, however, are distinct though faint topographic features and mark the former position of the ice front as surely as the hilly land laid form. In the paper cited a description is given of all the different moraines and beaches formed in this part of the state though the chief interest centers in those found in Sanilac and adjoining counties. In the vicinity of Imlay City a channel occupied by a swamp and bounded by beach lines was discovered extending north and west past this place and North Branch, then southwestward toward Flint. This, called the Imlay Outlet, is regarded as the outlet of a lake (unnamed) at the time the ice foot rested on its northern side. The position of the ice front is marked by a rugged moraine — the Toledo moraine. On the retreat of the ice from this position there was next formed the Detroit moraine which, except for a tract near Yale is nearly all water laid. North of Yale near Melvin it is represented by kames and kame moraines. In this area which the author of the paper did not examine we have observed a well-marked series of kame ridges mark- ing the northernmost limit of the eastern limb of the Detroit moraine. Northward the land fades away into the great swamp which extends north from Capac past Valley Center and Brown City to Shabona Post Office and into the almost imperceptible divide between this and the Black River swamp toward the east. The last moraine is the Port Huron Saginaw moraine which extends from a point six or eight miles northwest of Port Huron along the east side of Black River to Tyre, thence curving around toward the southwest to Vasser. From the 200 feet contour, around the north- ern and eastern side of this moraine, the land slopes evenly away toward the present shore. West of the eastern limb of this moraine lies the great Black River swamp. As the ice front retreated over this swamp region to its last halting stage at the Port Huron-Saginaw moraine, the water from the wasting glacier evidently found its way westward along the ice front, the posi- J^E VIEWS 315 tion of several of these outlets being noted. The last of these, the Ubly channel, was the most important. The position of this channel is now marked by the north branch of Cass River, a small stream flow- ing in a wide valley filled with valley gravels. At Tyre, a branch was received from the southeast. The relation of a great ice dam and the outlet of the waters is here admirably shown. The evidence of reversed drainage in the case of Black River is unmistakable. Previ- ous to the opening of the Ubly channel when the ice-front was on the south side of the north branch of Cass River the outlet which Taylor calls the Cumber spillway was close along its edge. This position is now marked by a long narrow swamp extending from the valley of the Freiburger channel north of Cumber, parallel with the North Branch, southwest across the South Branch of Cass River and on toward Deford. In this swamp is the famous "stone wall " consisting of a low embankment of earth and bowlders extending for a considerable distance parallel with the edge of the swamp. Locally it is regarded as the work of pre-historic man. The "wall" is about eighteen inches high and was until recently obscured by vegetable mold. It was exposed by fires in clearing the swamp. Mr. Taylor refers to its resemblance to ice beaches but regards the explanation as unsatisfac- tory. We spent some time in this vicinity and from a study of the surroundings were convinced that the ice-beach theory is an adequate and the only adequate explanation of the phenomena. It is of interest as constituting the only instance known to us of a bowlder wall formed by ice-push in a lake now wholly extinct. Our own observations on the beach lines of Sanilac county shows a marked northward elevation corresponding to previous observations in the lake region, the result probably of resilience following the withdrawal of the ice and to which was evidently due the reversal of the Black River drainage. This river now flows south and then east into the Saint Clair River at Port Huron, An indication of reversed drainage is seen in the direction of the branches of Black River as, for example, that of Elk Creek which rises in the southwest part of Sanilac county and flows northeast to its junction with Black River which here flows nearly due south. While Mr. Taylor has carefully described all the glacial features observed by him, and also by G. K. Gilbert who accompanied him on a part of the trip, he has described no eskers. A well-marked exam- ple occurs about ten miles northeast of Marlette. It consists of a well- 3l6 REVIEWS marked gravel ridge (or "hogback") extending for several miles in a westward direction bearing to the north from the southeast corner of Lamotte township. This was evidently formed at the time the ice front rested on the Detroit moraine. Other interesting features not observed by Mr. Taylor, though suggested by him, are the well-marked kame deposits in the southern part of Sanilac county. In the vicinity of Melvin and Peck they consist of hills and ridges of gravel and sand thirty to fifty feet in height. The name Erie- Huron is offered for the whole series of lakes rep- resented by all the beaches, while Lake Warren is restricted to one of the separate lakes wliich most closely corresponds to the original idea of Spencer who proposed the name. Taylor concludes that if the relations of the moraine and beaches have been correctly inter- preted then he has found outlets for three and possibly four stages, and these so related to the topography of the region and to the moraines that in at least three cases there is no reasonable doubt as to the contemporary place of tlie ice front. Tlie continental ice-sheet was obviously the great dam that held all these waters up. While testifying to the general excellence of the paper we find our- selves unable to follow the author in the degree of confidence he seems to repose in his barometer. As the correlation of some of the beach lines depends upon the identification of beaches having a ver- tical interval of twenty to forty feet only, and several miles apart and was made by means of a common aneroid barometer, it must be said that the conclusions in some cases are not without an element of doubt. Thus, for example, the Belmore beach is identified near Applegate at an elevation of 770 feet. Forrest Beach six or seven miles east of this has an elevation of about 735 or 740 feet. Near Cass City Mr. Taylor identifies the Du Plain beach which he says is twenty to thirty feet above the Forrest beach. The Belmore and Du Plain though so nearly related are regarded by Mr. Taylor as distinct, while between them he postulates still another stage represented by the Arkona beach which is said to have an elevation of 755 feet north- east of Spring Hill. Allowing for northward rise this beach would have about the elevation given the Belmore beach near Applegate. The question then arises which beach does the Applegate beach represent, the Belmore, Arkona or Du Plain, and what is the actual relation of these to the Forrest beach ? Beaches twenty to forty feet above the Forrest beach have been identified with each of these. Our own REVIEWS 317 barometer readings westward from Croswell gave 765 feet for the alti- tude of a shore line stated by Mr. Taylor to be 780 feet above sea level. In an area so little diversified, more exact measurements are needed to settle some of the questions raised. Notwithstanding these questions, however, we regard the paper as an important contribution to the literature of the glacial recession. C. H. Gordon. Elementary Geology. Ralph S. Tarr.B.S., F.G.S.A. The Mac- millan Company, 1897. Pp. xx+499. $1.40. This is the most attractive elementary text-book on geology which the writer has seen. The style is generally simple, the illustrations numerous, well selected, and with but few exceptions clearly repro- duced, and the mechanical execution, including the binding, is unsur- passed. The wealth of illustration is shown by the twenty-four full page plates and 268 figures, which adorn its pages. This work differs from most other text-books on geology in the greater stress laid upon the dynamic aspect of the subject. The stratigraphic and historic phases are treated briefly and concisely[in 100 pages as against 275 pages given to the dynamic and 92 to the structural side. The develop- ment of life during the various periods is outlined, but no attempt has been made to teach palaeontology, or to give long lists of fossils. The chapter on the life development precedes the chapter on the evolution of the land and the geography of the different periods. This depar- ture from the more common custom of considering first the geography and then the life of each period will perhaps find favor with some teachers but not with all. Whatever be the success of this arrange- ment, there can be but little doubt but that the author has done well to lay the greater stress upon the dynamic phases of the subject in a book meant for high-school pupils. In judging this book due regard must be had for the author's pur- pose,/. (?., to "furnish a companion and adjunct" to his Physical Geog- raphy. It is his hope that the two books will be used together, the geology being taken after a study of the air and ocean, and before the physiography. Knowing the author's purpose, the omission of some topics, such as "river cycles" and "base levels" can be understood, since they have been treated in the Physical Geography. In spite, however, of some omissions, there is considerable repetition in the two books — something hardly to be avoided. The Geology is a more ele- 3l8 REVIEWS mentary book than the Geography, whereas the present practice in high schools, one recommended by the Conference on Geography appointed by the Committee of Ten, is to place physical geography in the earlier part of the curriculum and geology, when studied at all, near the end. For these reasons, and because of the size of the two books, it is questionable whether they will ever be widely used together. Where such use is attempted, considerable culling and dove- tailing will be necessary. This would be easier for the teacher if the paragraphs were numbered. But doubt as to the feasibility of the author's plan does not mean condemnation of the book as a whole. It is admirably adapted for high-school pupils, and it is to be hoped that it will be widely used. There is so much that is to be commended in it that it is to be regretted that a few slips have passed uncorrected. On page ii8 the language implies that the term "oxidation" embraces all the chemical changes caused by percolating water; and the expression " limy shells " (p. 132) as applied to corals is hardly accurate. The statements in the text concerning normal and reversed faults (pp. 293, 324) do not correspond with the diagrams (p. 292) and the following sentence, " Mountains are present in nearly all parts of the world " (p. 314) does not agree well with another statement that " the formation of mountains occurs in only a few comparatively small parts of the whole earth" (p. 328). It may seem hypercritical to refer to some of these points, but too much care cannot be taken in a text-book to be used by immature pupils, and perhaps by teachers, who have not a wide knowledge of the subject. But these are not vital and do not detract greatly from the value of the book. Professor Tarr can be con- gratulated upon the degree of merit to which the book has attained. Lewis Institute. Henry B. Kummel. Abstracts. The Solution of Silica under Atmospheric Conditions. By C. \\'illard Hayes. While it is well known that under conditions prevailing at con- siderable distances below the earth's surface silica is one of the more easily soluble substances entering into the composition of the earth's crust, it is commonly regarded as the mineral least liable to be affected by solvents under ordinary atmospheric conditions. Some recent observations, however, show that even quartz pebbles are not proof against chemical as well as mechanical erosion. At several widely separated points in West Virginia and Tennessee occur coarse conglomerates, in two cases Carboniferous and a third Cambrian in age, in which the projecting portions of the pebbles have been deeply etched, evidently by solution. The pebbles vary in size up to an inch in diameter and are composed chiefly of vein quartz and quartzite together with a few of chert and feldspar, embedded in a coarse sandstone matrix. The projecting portions of these pebbles, particu- larly those composed of quartz and quartzite, are deeply pitted with rough, irregular surfaces, in many cases as much as a third of the pebble having been removed. From the form assumed by most of the pebbles, it would seem that their interior was more easily soluble than the outer portions. The latter usually forms a sharp rim within which is a depression with a slight elevation at the center. The etching is not confined to the larger pebbles, for on closely examining the surface of the matrix it is seen that in many places the projecting portions of the Eand grains have been removed, leaving a more or less smooth mosaic. A comparison of these etched pebbles with the glypto- liths or facetted pebbles described by Woodworth and others shows clearly that thev are not produced by the same agency as the latter, namely wind driven sand. Everything, on the other hand, ])oints clearly to solution. Again, numerous geodes from the Carboniferous limestone in Tennessee, collected by Campbell and Taff, show etching similar to 319 320 ABSTRACTS that described above. They vary from three inches to a foot in diam- eter and beneath the rugose opaque surface are colorless and trans- lucent, resembling in texture extremely fine-grained quartzite or slightly granular vein quartz. Under the microscope this is seen to be made up of spherulitic aggregates of quartz grains with much coarse granular quartz but no cryptocrystalline or amorphous silica could be detected. Many of these geodes are deeply etched upon one or both sides, not only the opaque shell being removed but also portions of the translucent interior. From its microscopic character, the silica of which they are composed would appear to be but little more liable to solution than vein quartz or quartzite. Minute quantities of amorphous silica may, however, be present and by their solution facilitate the removal of the crystalline grains by solution or otherwise. Hence less importance is attached to the solution of these geodes than to the etching of the conglomerate pebbles. The wide separation of the localities at which these cases of the undoubted solution of silica occur, renders it improbable that they are due to some peculiar and exceptional conditions, such as the presence of thermal, alkaline waters. It seems rather that they must be attrib- uted to the action of widespread agencies working in these cases under more than ordinarily favorable conditions so that the effect is excep- tionally striking. All the cases occur in a heavily forested region where there is a thick layer of humus and doubtless an abundant sup- ply of the humus acids. Further, the region is one in which forest fires are frequent and considerable potassium carbonate must be sup- plied to surface waters by leaching of the resulting ashes. The con- ditions, therefore would appear to be favorable for a chemical process somewhat as follows : By the oxidation of the vegetable tissues in the process of decay, the humus acids are formed, chiefly humic and crenic. These absorb varying quantities of free nitrogen from the air forming the azo-huraic acids, which, in turn, combine with free silica. The resulting acids combine with alkaline carbonates, particularly potassium carbonate, to form easily soluble salts. In most cases the etched surfaces of the pebbles described above support a more or less abundant growth of cryptogamic vegetation which might facilitate the solution in two ways: first by supplying humus acids directly from its own decay, and, second, by absorbing solutions of those acids from other sources and keeping the rock surfaces moist so that their solvent action might be practically continuous. ABSTA'ACTS . 321 While it is readily granted that such agents seem inadequate to produce the effects observed, no others at all conriparable with them in efficiency suggest themselves. And if the above conclusion is correct, a solvent capable of removing a third of a quartz pebble an inch in diameter, while still imbedded in its matrix, must be an extremely important factor in gradation when acting under the much more fav- orable conditions prevailing in the humus layer itself where the surface exposed by the quartz grains is vastly greater in proportion to their bulk and where the solvent action is not interrupted as it must be on exposed rock surfaces. T/ie Crystalline and Mctamorphic Rocks of Northwcsf Georgia. By C. WiLLARD Hayes and Alfred H. Brooks. The region discussed in the paper extends southward loo miles from the Tennessee line and westward twenty to eighty miles from the Atlanta meridian. It embraces about 4000 square miles, forming a belt to the east and south of the Georgia-Alabama Palaeozoic area. The rocks of this region naturally fall into three groups : (i) the gran- ites, gneisses and crystalline schists of the basal complex— probably Archean; (2) the slates and conglomerates of the Ocoee series — prob- ably Algonkian ; (3) intrusives in the other two series. The first group includes the Acworth gneiss, Austell granitoid gneiss, Corbin granite and the Piedmont gneiss and crystalline schist. In the second or clastic group only four members are distinguished and formation names have not yet been assigned to them since this classification is not regarded as final. These four members are {a) basal conglomerate resting on the Corbin granite; {l>) black slate often graphitic; {c) a series of interbedded slates and felspathic sandstones; {d) garnetiferous slate probably an altered phase of {b) and (/). The intrusives, which are found in association with members of both the other groups, but most abundantly in a belt from ten to twenty miles broad along their contact, named in the order of their intrusion, are {a) amphibolite schists and diorite; (//) gabbro and basic greenstone schist ; and (c) Villa Rica granite and associated coarse pegmatites. The above classification as well as the map accompanying the paper is to be regarded as only preliminary since the necessary petrographic study of the rocks has not yet been completed. The region discussed is embraced within the limits of the Dalton, Cartersville, Marietta, 322 , ABSTRACTS Tallapoosa, and Anniston folios of the United States Geological Sur- vey which are now in course of preparation. The Age of the White Limestone of Sussex County, New Jersey. J. E. Wolff and Alfred H. Brooks. The principal new observations contributed to this well-known field by the authors were those indicating the presence of several longitudinal faults along the boundaries between the Cambrian limestone and sand- stone and the white limestone and associated crystalline rocks ; and the discovery, at a new quarry in Franklin, of a crevice in the white lime- stone which has been filled up by the overlying Cambrian arkose, which contains bowlders and fragments of the white limestone, and fragments of mica, feldspar and other minerals characteristic of the white limestone Or associated rocks. From this and other localities, they conclude that the Cambrian sandstone contains the debris of the white limestone granite and associated rocks and explain the apparent transition between the blue (Cambrian) and white limestones as due to fault brecciation and shearing, and the apparent interbedding of the white limestone and quartzite as due to the deposition of the latter on the irregular sur- face of the white limestone, with local faulting in some places. They conclude that the white limestone is of pre-Cambrian age and an inte- gral part of the gneissic formation. This paper will appear in full in the 1 8th Annual Report of the Director of the United States Geological Survey. Erosion at Baselevel. By M. R. Campbell. Some late observations by Mr. C. W. Hayes and the writer on the solubility of quartz under atmospheric conditions and the writer's own study of local baselevels in the Appalachian coal field seem to throw some light on the question of the ultimate result of undisturbed ero- sion, a result which has hitherto received but little attention from physiographers. The conditions which seem to facilitate the solution of quartz are those which would probably prevail during the final stages of the pro- cess of baseleveling to a much greater extent than under ordinary con- ditions of erosion ; consequently it seems probable that most, if not all, of the waste of the rocks which is washed in from the surrounding ABSTRACTS 323 slopes is removed by solution and does not accumulate on the surface of the plain. This conclusion is still further substantiated by the fact that in the local basins observed in the coal field, the surrounding slopes are sharply separated from the bottoms of the basins, showing that even in these small examples there is some action going on by which the waste is carried off in solution leaving the floor of the basin unobstructed and a nearly perfect plain. If these conclusions are correct, the baseleveling epochs of the past njay have been marked by very completely formed plains, and the slopes of the unreduced areas may have been sharply separated from, instead of blending with, the surface of the plain. Origin of Certain Topographic Forms. By M. R. Campbell. The recognition of apparently abnormal physiographic forms in limited areas has induced the writer to undertake an investigation of their probable cause. Since the abnormal types are limited in their geographical distribution, they are doubtless caused by local variations in some of the conditions governing erosion. Climate and rock char- acter are variable, but changes in them could not produce tne forms in question. Declivity must be accountable. Declivity may be changed by crustal movements, but those which operate over a broad area change the declivity as a whole, and hence cannot produce local varia- tion. Movements which are limited in their extent are the ones which must be looked to for local variation and the production of abnormal forms. Local uplifts may result in the production of normal faults, monoclinal folds, symmetrical folds, and dome-shaped uplifts ; each of these structural forms, when produced during the final stage of an erosion cycle, — for at that time alone will they introduce new erosion conditions, — will be marked by physiographic features peculiar to itself. The production of a scarp from a monoclinal fold is perhaps the most ren:arkable form, and it is due to a peculiar combination of slow, regular uplift, homogeneous rocks, and baseleveling conditions outside of the area affected bv the uplift. This principle is applied to the solution of the Blue Ridge scarp in North Carolina, and it is shown that the physical features of that region could have been produced by a monoclinal uplift in a broad 324 ABSTRACTS peneplain. On this supposition the Blue Ridge plateau was once con- tinuous with the Piedmont plain, but the latter remained at sea level while the former was slowly elevated 2000 feet. There were halts in this movement, during which partial peneplains were formed on the western side of Blue Ridge. The final result of such a combination of crustal movements and periods of baseleveling is that all of the pene- plains on the western side of the Blue Ridge, collectively, represent approximately the same time interval as that represented on the east- ern side by the Piedmont plain. This theory also explains the present condition of the New River- Roanoke divide, and why the former stream has not been captured by the latter, — a crucial test for all theories relating to the origin of this scarp. The same explanation is also suggested for the Black Hills of South Dakota, — a dome-shaped uplift in which the outer rim of Dakota sand- stone is in the zone of maximum erosion, and consequently is kept at baselevel, while the inner belt of limestone owing to its more rapid upward movement is unreduced. This is offered merely as a sugges- tion to geologists in the hopes that it will be entertained as a working hypothesis and its applicability tested by persons familiar with the field. Dikes in Appalachian Virginia. By N. H. Darton. An account was given of further discoveries of diabase dikes and of certain dikes of acidic eruptives among the Palaeozoic rocks near Mon- terey in Highland county, Virginia. The diabase dikes are similar to those described by Messrs. Diller and Darton some years ago, but the acidic rocks are very great novelties for this region. The rock is por- phyritic diorite containing quartz, mainly in rounded grains, with plagioclase, biotite, and hornblende phenocrysts, in a white groundmass. The dikes are of small extent and appear to be restricted to a limited territory. The area is one in which the most prominent anticline of central Appalachian Virginia reaches its culmination. The abstracts in this and the previous issue of this Journal were read before the Washington Meeting of the Geological Society of America. THE JOURNAL OF GEOLOGY MAY-JUNE, iSgj THE LAST GREAT BALTIC GLACIER. In a recent number of this Journal* Dr. Keilhack criticises certain views which I have been led to entertain as to the glacial succession in northern Europe. He does not agree with me that the great terminal moraines of the Baltic Ridge are the products of a separate and independent glacial epoch. On the contrary, he and his colleagues on the Royal Prussian Geological Survey are persuaded that the moraines in question simply mark a stage in the retreat of the third ice-sheet. In the last edition of the Great Ice Age I have stated somewhat fully the evidence for the conclusion which my critic now disputes, and a short outline of the subject subsequently appeared in this Journal^. Geologists who may wish to know what my views are will, I hope, take my own account of them rather than that which Dr. Keilhack has presented to his readers. Doubtless the discus- sion has been conducted by him in good faith and with every intention of being just, but, all the same, he has not succeeded in giving an accurate outline of the reasons which compelled me to dissent from the opinions maintained by him and his eminent col- leagues. Some of the more important evidence adduced he either ignores or slurs over, while in certain other matters he has strangely misunderstood and thus unwittingly misrepresents me. Having carefully considered Dr. Keilhack's paper, I find that his criti- 'Vol. V, No. 2, pp. 1 13-125. ^^Vol. Ill, p. 241. Vol. v., No. 4. 325 326 JAMES GEIKIE cisms go wide of the mark; he has failed to meet my argument, and I have nothing to modify or withdraw. Under these circum- stances I might well leave what I have already written on the subject, and simply appeal to it as a sufficient reply to Dr. Keil- hack. Silence on my part, however, might be misunderstood, and I would not appear to be discourteous to a geologist whom I highly esteem, and whose work no one can value more than myself. At the risk of being tedious, therefore, I shall again shortly state my reasons for the faith that is in me, and take due note of my critic's objections. It was while studying certain glacial phenomena in Britain that I was first led to inquire more fully into the grounds upon which German geologists base their belief that the terminal moraines of the Baltic Ridge were deposited during the retreat of the third ice-sheet (Polandian). The evidence which impressed me in Britain may be very shortly outlined. We have two well- recognized bowlder clays developed in the low grounds of our country — the lower of which belongs to the epoch of maximum glaciation (Saxonian), at which time the British and Scandina- vian ice-sheets were coalescent. The upper bowlder clay is the ground-moraine of a less extensive, but still immense, ice-sheet, which, like its predecessor, also occupied the basin of the North Sea, so that continuous inland ice extended as before between Norway and Britain. During the formation of the lower bowlder clay the Scandinavian ice invaded East Anglia and the midlands of England, but this invasion was not repeated by the succeeding ice-sheet. Nevertheless, we have no doubt that the British and Scandinavian ice-sheets were again confluent. Along the whole eastern seaboard of Britain, from Flamborough Head to the extreme north of the Scottish mainland — a region over which the upper bowlder clay is strongly developed — all the evidence indicates that the British ice, underneath which that clay accu- mulated, was steadily deflected to right and left as it passed outwards into the North Sea basin. North of the Firth of Forth the trend of all the stones in the upper bowlder clay, as well as the direction of drumlins, of roches moutonnees and striae are THE LAST GREA T BA L TIC GLA CIER 327 towards northeast, north, and, eventually, northwest. South of the Firth, on the other hand, the direction of glaciation is south- erly. In a word, it is demonstrable that during the formatio?i of the 2ipper bozvlder clay the British ice did not follow its 7iat7iral course a7id flozv directly outwards into the basin of the North Sea. It was deflected fust i?i the sajne zvay as the earlier itdand ice of the Saxo?na7i stage had been. So far, indeed, as Scotland was concerned, the ice-sheet of Polandian times was hardly less important than its predecessor. The upper bowlder clay is the chief glacial accumulation in our lowlands, and its surface is often concealed under more or less continuous sheets of gravel and sand, while here and there it is dotted over with large erratics and traversed by eskers. All these obviously belong to the period of melting. Nowhere, however, is a vestige of terminal moraine encountered' until the base of the uplands and mountains is approached. Here well- developed terminal moraines suddenly put in an appearance — the relics of large valley -glaciers and local or district ice-sheets. For a long time the general belief was that these moraines had been accumulated during the retreat of the last general ice- sheet (Polandian). Further investigation, however, proved that the outermost moraines had been laid down by advancing gla- ciers, which had everywhere ploughed into the upper bowlder clay, sweeping it out of the valleys, and so modifying its surface in the low grounds at the base of the mountains as to impress upon it a new configuration. The dimensions of the moraines, the extent of their fluvio-glacial gravels, and the amount of erosion experienced not only by the upper bowlder clay, but by the solid rocks themselves, all indicate that these valley-glaciers and local ice-sheets mark a distinct stage of the Glacial Period. Again, there is evidence to show that after the retreat of the ice-sheet of the upper bowlder clay, and before the valley-gla- ciers in the west of Scotland had descended to the coast, the maritime districts were slightly submerged. Loch Lomond at that time formed an arm of the sea, and was tenanted by a fauna which consisted for the most part of existing British forms. 328 JAMES GEIKIE Subsequently a large valley-glacier entered the fiord, filling it up, and ploughing out preexisting marine deposits. By this time, however, the climate had become arctic, as is proved by the character of the fauna in the contemporaneous undisturbed shelly clays of the lOO-feet terraces and beaches. Such, stated as shortly as may be, are the main facts that led me to conclude that our lOO-feet beach and the contem- poraneous morainic accumulations indicate a distinct stage of the Glacial Period. There was clear proof that a readvance of active glaciers had taken place after the ice-sheet of the upper bowlder clay had vanished from our low grounds, and from the lower reaches at least of our mountain valleys. But, as I remarked in my former communication to this Journal, it was not quite so evident that the readvance in question had been preceded by a definite and prolonged interglacial epoch. The only evidence of preceding relatively genial conditions are the shells which the Loch Lomond glacier took up from the old sea floor and included in its moraines. It was with these facts and inferences in view that I approached the examination of the Upper Diluvium of North Germany. Perhaps it may be objected that I had no right to expect that the glacial succession in Britain should correspond with that of the Continent, and, if we descend to details of minor importance, that is no doubt true enough. But so far as the chief stages of glacial accumulation are concerned, these must have followed in the same order throughout Europe. I do not suppose, for example, that any geologist will doubt that the epoch of maximum glaciation was one in Britain and the Conti- nent, and the same must be true of other well-marked divisions of the glacial system. I think, therefore, that I was justified in my expectation that the Upper Diluvium of Germany would prove no exception to the rule. If the valley-moraines of the British mountains, which I formerly took to-be the degenerate relics of our minor ice-sheet, mark in reality a recrudescence of glacial conditions, it was not improbable that the German Upper Diluvium would have a similar tale to tell. THE LAST GREA T BAL TIC GLA CIER 329 Now, in the first place, the Upper Diluvium of North Ger- many closely resembles, in all essential particulars, the British upper bowlder clay with its associated deposits. Along the southern limits reached by the former we encounter banks and undulating hillocks and sheets of gravel and sand, just as is the case with our upper bowlder clay. Like the latter, also, the upper GescliiebeleJmi is often covered with wide stretches of water- worn materials and sprinkled with erratics. So, again, no ter- minal moraines present themselves as we traverse the country from south to north, until we are suddenly confronted with the great moraines of the Baltic Ridge. Here, then, we cannot but recognize a general resemblance, at least, between the glacial phenomena of Britain and North Germany. In the published works of German geologists I could meet with no evidence to show that the bowlder clay in front of the Baltic Ridge moraines was identical with the bowlder clay behind them. This identity appeared to me to have been taken for granted. At all events, no doubt had been expressed upon the subject, and no attempt made to demonstrate that the two upper bowlder clays are one and the same. Even in the paper to which I am now replying Dr. Keilhack does little more than reiterate his and his colleagues' opinions that the bowlder clays in question are con- temporaneous. Apparently he thinks that their interpretation of the evidence ought to be accepted on the ground of their intimate knowledge of the glacial accumulations of their country. I need hardly say that the accuracy of their observations is not called in question. I doubt if their works have, out of Germany, gained the attention of a more admiring student than myself. Their descriptions, wherever I have been able to test them in the field, have proved, as might have been expected, full and exact. It was not without hesitation, therefore, that I found myself unable to accept their explanation of the evidence. I was consoled, however, by the reflection that they themselves are not agreed as to the precise mode of formation of the Baltic Ridge moraines. When geologists, who claim a long acquaint- ance with the same facts, cannot yet agree amongst themselves 330 JAMES GEIKIE as to how those facts are to be interpreted, I may be excused for forming an independent opinion. Having for many years mapped and studied glacial deposits in my own country — and such deposits are much the same in all well-glaciated lands that I have visited — I did not feel incompetent to weigh the evidence and judge for myself as to its meaning. Dr. Keilhack does not, in my opinion, strengthen his posi- tion when he states that in numerous places the ground-moraine in front of the great terminal moraines of the Baltic Ridge passes smoothly under these. If continuous sections showing such connection are in existence, I confess I have never seen or heard of them. Possibly Dr. Keilhack only infers that the bowlder clay which he sees passing under the moraines is the same as that which overspreads the region on the other side. Be that as it may, I have not said and do not maintain that all the bowlder clay lying north of the terminal moraines is younger than the bowlder clay which occurs south of them. I am quite prepared to believe that in many places the terminal moraines have been dumped upon the bowlder clay of the Polandian stage. Further, I do not doubt that the same bowl- der clay actually extends over wide areas behind the moraines, but, if such be the case, it must be largely worked up into or concealed underneath the ground-moraines and other accumula- tions of the great Baltic Glacier. I now come to Dr. Keilhack's criticism of certain facts which have been adduced by me in support of the conclusion that the terminal moraines of the Baltic Ridge are deposits of a separate and independent glacial epoch. (i) In Denmark and Schleswig-Holstein two bowlder clays have been for a long time recognized. Both of these occur over the eastern section of that region. To that region the upper bowlder clay is confined, while the latter extends west- ward to the shores of the North Sea. Both clays are charged with the same assemblage of erratics, the character of which shows that the ice which brought them moved out from the Baltic and traveled in a general westerly direction. In Holland THE LAST GREA T BAL TIC GLA CIER 3 3 I there is only one bowlder clay- — that of the Saxonian stage or epoch of maximum glaciation^ — and it is crowded with erratics which have come from the north. The lower bowlder clay of Schleswig-Holstein and Denmark cannot, therefore, have been formed at the same time as the Dutch deposit. The latter indicates an ice flow from north to south — the former an ice flow from east to west approximately. The "lower bowlder clay" of the Cimbric peninsula is not the product of the Sax- onian ice-sheet, but of its successor, the Polandian. This infer- ence is greatly strengthened by the fact that the so-called "lower bowlder clay" of Schleswig-Holstein is underlaid by well-marked interglacial beds, which in their turn rest upon a yet older bowlder clay — in all probability belonging to the Saxonian stage. Thus we have actually three bowlder clays in the Cimbric peninsula. The uppermost and youngest of the series is confined to the eastern or Baltic side of the peninsula, and is, I believe, the product of the last great Baltic Glacier — the terminal moraines of which appear to form its outer margin. It may be added that freshwater deposits, containing relics of a characteristic interglacial flora (indicative of genial climatic conditions) have quite recently been detected in the neighbor- hood of Copenhagen, and thus well within the area overflowed by the Baltic Glacier. The deposits referred to are covered by the morainic accumulations of that glacier, and are underlain by diluvial gravels and bowlder clay of an earlier glacial epoch. Dr. Gunnar Andersson remarks of these interglacial beds that they have certainly been accumulated during the time that elapsed between the melting of the great Scandinavian ice-sheet and the invasion of Zealand by the last Baltic Glacier.^ Dr. Keilhack's remarks on the evidence cited by me and the inferences I have drawn are somewhat vague, and he ignores the occurrence of the interglacial beds and underlying bowlder- clay, which are proved to occur below the so-called "lower bowlder clay" of Holstein. He states, what is quite true, that 'Bihang till K. Svenska Vet.-Akad. Handlingar. Bd. 22. Afd. iii. No. I 1896. 332 JAMES GEIKIE no single erratic can be taken as a guide to the direction in which an ice-sheet has flowed. But if by this he means that we cannot decide upon the direction of ice flow from a consideration of the general assemblage or facies of the stones in any given bowlder clay, I cannot possibly agree with him. I am well aware of the fact that every now and again erratics occur in bowlder clay, whose presence would seem to indicate a different direction of ice flow from that suggested by the others in their neighbor- hood. In regions which have been overflowed by ice at differ- ent times and in different directions, such occurrences must be expected. But in the case of the bowlder clays of the Cimbric peninsula it is not merely a few sporadic stones of eastern derivation that we have to deal with. The whole mass of the materials of the two bowlder clays which appear at the surface has traveled from east and northeast, So, again, the bulk of the stones in the Dutch bowlder clay have come from the north. (2) Following De Geer, I maintain that the strong belt of terminal moraines which extends from Hango Head in a northeasterly direction through Finland, are contemporaneous with those of the Baltic Ridge. Referring to the observations of the accomplished geologists of the Geological Commission of Finland, I cite the fact that two distinct systems of glacial striae are apparent in that country. "The striae of the one sys- tem run in parallel directions, and extend far east and southeast of the terminal moraines. The other and younger system, on the other hand, is bounded by these moraines — the later striae crossing the older series at various angles. When striae belong- ing to both systems appear on one and the same rock-surface, the younger are always the fresher of the two, the older ones being worn and abraded. The latter, according to Rosberg, are the products of a general mer de glace, which attained so great a thickness that minor inequalities of the ground had little or no influence in deflecting the ice flow, which extended far beyond the limits of the terminal moraines. The ice, under- neath which the younger system of striae came into existence. THE LAST ORE A T BALI IC GLA CIER 333 must have been relatively thin, in consequence of which the inequalities of the ground produced endless local divergences from the general direction of ice flow."' The small sketch map of the great Baltic Glacier given in my book is largely a repro- duction of De Geer's map, published some years ago. Accord- ing to Dr. Keilhack the line of union drawn from East Prussia to Finland is "purely hypothetical and supported by no observa- tions." Here, however, he is mistaken. Let him consult De Geer's recent work, Oftt Skandinaviens geografiska Utveckling efter Istiden," and he will find that a great terminal moraine, marking the edge of the Baltic Glacier, traverses the islands of Osel and Dago in a north-south direction, and can even be traced for some distance on the bed of the sea. This can hardly be other than a continuation of the terminal moraines (Salpausselka) at Hango Head. Dr. Keilhack says he looks "for the easterly continuation of the Baltic terminal moraine much further south, in the interior of Russia." Well, it is never safe to prophesy, but I shall be surprised if his anticipations prove correct. That much gravel and sand and many erratics will be met with in the regions he refers to I can well believe, but all these, I think, will be found to be similar in character and age to the diluvial gravels, etc., which overspread the bowlder clay in the lands to the south of the Baltic Ridge. According to Dr. Vogt, of Christiania, with whom Dr. Keilhack agrees, "the J-'lnnish terminal moraine belongs with that of middle Sweden and Norway." On the other hand, De Geer shows that the terminal moraines which occur in Norway, somewhat south of Lake Mjosen, represent the front of the Baltic Glacier during a stage in its retreat. He traces the position occupied by the ice edge right across Sweden in an easterly and northeasterly direction. The glacier is repre- sented as filling the upper section of the Baltic basin and reach- ing as far south only as Aland. The great gravel-ridge at Hameenkanga, in the middle of western Finland, would appear to be its terminal moraine in that country. This moraine occurs 'Great Ice Age, 3d edit., p. 474. 334 JAMES GEIKIE about 125 miles north of the older belt that extends from Hango Head to the north of lake Ladoga. (3) Dr. Keilhack's third criticism has already been met by the remarks I have made on the subject of the upper bowlder clay of Britain. My worthy critic could not have read my work with attention or he would hardly have attributed to me the strange statement that my belief in the confluence of Scandi- navian and British ice-sheets, during the formation of our upper bowlder clay, was suggested by and based on the occurrence in that deposit of Scandinavian erratics ! He says : "It seems to me unnecessary entirely to repudiate the opinion of the north German geologists in order to explain the occurrence of Scandi- navian bowlders in the upper bowlder clay of Britain." I have nowhere said that Scandinavian bowlders occur in our upper bowlder clay. So far as I am aware such erratics are confined to the lower bowlder clay of East Anglia and the midlands of England. Not one has been met with in any bowlder clay of East Britain north of Flamborough Head. The evidence which shows that our upper bowlder clay was formed at a time when "inland ice" filled up the basin of the North Sea, and the Scan- dinavian and British ice-sheets were confluent, has nothing whatever to do with the occurrence in our country of Scandi- navian erratics. The great extension of ice which characterized the formation of our upper bowlder clay is demonstrated, as I have shown above, in quite another way. When Dr. Keilhack has duly weighed our evidence he will no longer consider "whether the connection of the Scandinavian with the Scottish ice in the third glacial period is more than imaginary." The fact of that connection is as well established as one can expect it to be. (4) Believing as I do that the terminal moraines of the Baltic Ridge are the products of a distinct glacial epoch, I naturally hold that the superficial or youngest morainic accumu- lations which lie behind those moraines are of later age than the ice-sheet of Polandian times. Thus, I maintain that the upper bowlder clay throughout the region extending from the THE LA S T GREA T BALTIC GLA CIER 3 3 5 Baltic Ridge to the shores of the Baltic Sea, is a younger deposit than the bowlder clay that stretches south from the Baltic Ridge to the valley of the Elbe. Further, I hold that the youngest interglacial beds, which occur between the upper and lower bowlder clays of the Baltic coast lands, are neces- sarily of later date than the interglacial beds which are met with in the regions lying to the south of the Baltic Ridge. Dr. Keil- hack remarks that this is a "false conclusion," by which I sup- pose he means merely that he does not agree with me. It will be time enough, however, to call my conclusions false when he has succeeded in proving that the upper and lower bowlder clays of the Baltic coast lands are of the same age as the upper and lower bowlder clays of the Elbe valley. At present he merely assumes that they are, and expects me to accept his dictum in lieu of direct evidence or reasonable argument. Dr. Keilhack further remarks that if I am right in my view that the terminal moraine of the Baltic Ridge defines the southern limits of a distinct glaciation, he cannot understand why each of the other terminal moraines occurring to the north of it may not also represent the edge of the ice during a distinct glacial epoch. "On this basis," he says, "the so-called 'last glacial epoch' would have to be divided into four if not five epochs, so that even the most fanatical advocate for as many glacial periods as possible would be terrified." I am sorry if Dr. Keilhack cannot distinguish any difference between the position and character of the very pronounced terminal moraine of the Baltic Ridge and the distribution and character of the minor moraines of retreat which lie behind it. For myself I have no more difficulty in distinguishing between the two sets of moraines than I have in the case of the morainic ridges of the inner zone of the Alpine paysage morainique. In Britain, just as in the Alpine Vorland and in north Germany, the large moraines of our district- and valley-glaciers are succeeded by numerous smaller moraines of retreat. In North America, also, geologists find no diffi- culty in differentiating between the first zone of great moraines of the Wisconsin formation, which mark the edge of a sep- 336 JAMES GEIKIE arate ice-sheet, and the moraines of retreat that occur behind them. The southern limits of the Polandian ice-sheet in Britain and Germany alike is bordered by more or less thick banks and sheets of morainic gravels. These are the only terminal moraines connected with the ground-moraine of that ice-sheet. Although the surface of that ground-moraine is often abun- dantly strewn with fluvio- glacial detritus it shows no lines of ter- minal moraines comparable to those of the Baltic Ridge in Germany and of the valley -glaciers and local ice-sheets in Britain. The latter bear witness to vigorous glacial action — they have not merely been dumped upon the ground, for the local ice-flow has in many places ploughed into and pushed up preexisting bowlder clay and ground out hollows in the solid rocks. The terminal moraines of the Baltic Ridge no doubt con- sist largely of water-worn materials, just as is the case with the moraines of all sheet-like or Piedmont glaciers, and much of the material has been dumped. But they also yield evidence of glacial push, for ground-moraine, confusedly commingled with gravel and sand, not infrequently enters into their composition. The lines of moraines and the irregular mounds, banks, and sheets of water-worn materials which are distributed over the ground between the Baltic Ridge and the shores of the Baltic Sea are simply moraines of retreat and fluvio-glacial deposits, and I hope no "fanatical" giacialist will "terrify" either him- self or Dr. Keilhack by suggesting that they indicate "four if not five" separate glacial epochs. (5) Dr. Keilhack further objects to my conclusion that the terminal moraines of the Baltic Ridge mark the limits of an inde- pendent glaciation, because he thinks these moraines cannot be correlated, as I have supposed, with the Alpine valley-moraines belonging to Professor Penck's "first postglacial stage." The Baltic Glacier of my fourth glacial epoch .(Mecklenburgian) , according to him was of such immense size and so little inferior to the ice-sheets of earlier stages that such correlation as I have attempted is almost impossible. And he gives a graphic repre- THE LAST GREA T BA L TIC GLA CIER 337 sentation to show the " unnaturalness" of my classification. Such graphic representations, however, unless they are con- structed on sound principles, are apt to express rather the views of the draughtsman than the actual facts of nature. Dr. Keil- hack's diagrams are, in my opinion, constructed on a wrong principle, and were I to draw similar diagrams they would show a different result. But as my critic would probably be no more satisfied with my graphic representation than I am with his, I prefer to give one constructed a year or two ago by an indepen- dent authority — Dr. Du Pasquier, whose recent untimely death every student of glacial geology must deplore. For purposes of comparison Dr. Keilhack's diagrams are also given. Alps (Keilhack). I I r II H 1 III. I 1 IV I 1 North Europe (Keilhack). 11^ 1 II. I III. I IV. I ■ 1 Alps (Du Pasquier). X. \ 1 Y 1^ ^ Z I ^1 z I 1 The first thing that strikes one in comparing Dr. Keilhack's representation of the Alpine glacial system with that given by the eminent Swiss geologist is the discrepancy between the lines representing the last three glacial stages (II, III, and IV of Keil- hack ; Y, Z, Z\ of Du Pasquier) . Dr. Keilhack obtained his lines by simply measuring the length of individual glaciers on the north side of the central Alps, but Du Pasquier has followed a differ- 338 JAMES GEIKIE ent plan. He says: "La longueur des lignes representant les glaciations successives est proportionelle aux aires glaciees. Le rapport des aires glaciees aux differentes epoques X, Y, Z, Z^ est presque identique dans le Nord de I'Europe et dans les Alpes, et parait devoir etre exprime par: 0,7 (?): i : 0,66:0,27 en moyenne." ' If Dr. Keilhack, instead of estimating the extent of the several ice-sheets of northern Europe by drawing lines from the north end of the Gulf of Bothnia into Germany, will contrast the superficial areas of the successive glaciations he will prob- ably come nearer the truth. He will, at all events, ascertain that De Geer's great Baltic Glacier (Mecklenburgian) large though it was, yet covered a much smaller area than that occupied by the preceding Polandian ice-sheet. Nothing could be more absurd than to describe the last great Baltic Glacier as being "little inferior to those of earlier epochs." It was very much inferior, and quite comparable, as Du Pasquier has said, with the Alpine valley-glaciers which are represented in his diagram by the line Z^ In correlating the Alpine valley-moraines of Professor Penck's "first postglacial stage" with the terminal moraines of the great Baltic Glacier I was well aware that no interglacial beds had been observed underlying the former. I knew quite well that the valley-moraines in question were generally considered to mark pauses in the retreat of the great glaciers of the inner zone. Latterly, however, Du Pasquier had begun to doubt if such were really the case. Quite recently he had detected evidence to show that the formation of the moraines of the "first post- glacial stage" had been preceded by an interglacial interval of fluviatile or lacustrine accumulation. As Du Pasquier's observa- tions are the first of the kind recorded, I shall quote what he says on the subject : Au-dessus de I'etage glaciaire Z (see his diagram) encore deux ou trois systemes de moraines au moins, se terminant vers I'aval en nappes d'allu- vions precedent les moraines actuelles. Ces systemes, tres clairement marques ' See Bulletin de la Soc. Neuchat. de Geogr., VIII (1894-5). THE LA ST GREA T BAL TIC GLA CIER 339 dans les Alpes, se retrouvent en Ecosse ; ils ne paraissant pas avoir encore €te suffisamment eludies ailleurs. Ces moraines terminales, echelonnees le long des valines, marquent-elles simplement, comme on I'admet d'ordinaire, des moments d'arret dans la retraite generale des glaces? 11 nous parait bien douteux qu'il ne s'agisse que de cela. D'abord, comme nous I'avons dit en traitant des glaciers actuels, les moraines terminales ne se produisent guere que pendant I'arret qui s^pare une phase de crue d'une phase de decrue con- secutive. D'autre part, dans notre Jura en particulier, certaines localites nous presentent, entre les moraines Z et d'autres subs^quentes soit Z', des depots fluviaux ou lacustres assez considerables qui font penser que le drainage glaciaire avait fait place a un drainage fluvial. Ces moraines jurassiennes Z', nous paraissent etre equivalentes a certaines moraines alpines qui forment un ensemble bien caract6rise et qui seraient elles-memes equivalentes aux moraines dites baltiques de I'Europe septentrionale. Nous pensons done plutot que ces moraines Z% marquent une ^poque glaciaire distincte, sans pr^juger en aucune fa(;on de la valeur de I'intervalle interglaciaire qui les separe de I'epoque Z et qui dut, a en juger par I'importance relative des phenomenes d'alteration, etre bien plus court que les precedents. ' I have now I think taken note of all Dr. Keilhack's remarks and criticisms which seem to call for any reply. James Geikie. ' Loc. cit. THE POST-PLEISTOCENE ELEVATION OF THE INYO RANGE, AND THE LAKE BEDS OF WAUCOBI EMBAYMENT, INYO COUNTY, CALIFORNIA. The following notes are the result of observations made in the summers of 1894 and 1896. My routes were : first, from Alvord Station on the Carson and Colorado Railroad eastward on the Saline Valley road, passing through Waucobi Canyon and over the divide seventeen miles E. S. E. of Alvord ; second, from Alvord through Soldier Canyon, over the range to Deep Spring Valley. As seen from the foothills of the Sierra Nevada, looking across Owens Valley, the axis of the low portion of the Inyo- White Mountain Range, between Soldier Canyon and the ridge south of Waucobi Canyon, arches strongly to the eastward, and forms a broad embayment between Soldier Canyon on the north and the head of Waucobi Canyon on the south. (Fig. i.) The range from east of the divide at the head of Owens Valley to Owens Lake is practically one, but unfortunately it has been given the name of White Mountain to the north and of Inyo Range ^ to the south of the embayment. Inyo is here used to include the range south of Soldier Canyon. For the broad embayment formed I use the name Waucobi, and for the ancient lake in which the lake beds were deposited the name Waucobi is also adopted. It was the examination of the lake beds deposited in Waucobi embayment that led me to conclude, from their position, that a marked orographic movement had taken place in the Inyo Range since Pleistocene time. In crossing Owens Valley, from Alvord Station eastward, the lake beds are met with about a mile and a half east of the station, at an elevation of 100 feet above the railroad track. [a, Fig. 2.) They extend north of this point along the western ^ Am. Jour. Sci., March 1895, 3d series, Vol. XLIX, p. 169. 340 POST- PLEISTOCENE ELEVATION OF INYO RANGE 341 foot of the White Mountain Range for fifteen miles or more. Near Black Canyon they occur well up the slopes, and traces of them were seen north of a line passing through Bishop Station ; to the south they appear for a short distance beyond the entrance to Waucobi Canyon. As seen from Alvord Station, the Fig. I. Map showing the general relations of Soldier and Waucobi Canyons to Owens Valley and the Inyo Range. (Taken from Lieutenant Wheeler's map.) beds extend from the south side of Waucobi Can^'on north across the broad embayment to Soldier Canyon, and westward from three to ten miles, rising with the slope to the foot of the encircling ridge of Cambrian quartzites. {b, Fig. 2.) The beds at the Devils Gate in Waucobi Canyon are 2250 feet above the lowest bed exposed east of Alvord, and appear to be the same as those at the lower level ; they extend on up the canyon to a level 3100 feet above the valley bottom at Alvord Station and within about three miles of the stfmmit (below c, Fig. 2) at the head of the canyon. T 342 CHARLES D. WALCOTT The strata of the lake beds section three miles up the canyon are largely a fine calcareous deposit, with more or less arenaceous and argillaceous matter in the form of fine sand. Some of the white beds are made up almost entirely of the remains of fresh- water shells of the following genera, as identified by Dr. W. H. Dall : Valvata, Planorbis, Pisidium ?, and possibly Amnicola and Pampholyx. The species are undetermined, but resemble Val- FiG. 2. View of Waucobi embayment from the foothills of the Sierra Nevada. a Lake beds at the level of Ovi^en's Valley ; b, contact of lake beds with the Cambrian quartzites ; c, point above the highest exposure of the lake beds in Waucobi Canyon ; d, point of Inyo Range overlooking Deep Spring Valley on the east ; e, Waucobi Mountain south of Waucobi Canyon. vata siiicera Say and Planorbis parvus Gld. "Any of them might be recent or Pliocene ; my impression from the mass is that they are Pleistocene." As the beds approach the steeper slope of the mountain, about ten miles above the mouth of Waucobi Canyon, the sedi- ments become coarser and coarser, and brown arenaceous beds predominate over the drab and light gray sediments. Near the contact with the quartzites, a little below Devils Gate, bowlders of the quartzite a foot or more in diameter occur in the coarse sediments, and the contact of the lake beds and the Cambrian quartzites is finely shown on the south side of the canyon. At a point about two miles above Devils Gate, and 3000 feet above the lowest lake bed observed in Owens Valley, there is a POST- PLEISTOCENE ELEVATION OF INYO RANGE 343 fine exposure on the south side of the Saline Valley road, which is the highest seen in the canyon that might be referred without doubt to the lake beds. The band exposed is about six feet in thickness, formed of layers of white, very finely granular sedi- ment, which crumbles under strong pressure. It is capped by Fig. 3. Lake beds, Waucobi Canyon, Inyo Range. About five miles above Owens Valley. The dark Cambrian rocks of the White Mountain Range north of Soldier Canyon are shown on the left upper half of the plate. layers of fine conglomerate formed of small angular fragments of quartzite. The greatest thickness of the beds observed at an}- one point was estimated at 150 feet. The finer, light colored calcareous beds vary from sixty to seventy-five feet in thickness. Near the valley the average dip is 3° to 5°. About two miles up it increases to 10° for a short distance and then changes to from 344 CHARLES D. WALCOTT 3° to 5° in its continuation up the canyon. The rise of the can- yon bottom is nearly coincident with that of the lake beds. The upper surface of the lake beds throughout the Waucobi embayment is covered by a layer of debris formed of fragments of arenaceous limestone, siliceous shale, and quartzite that have been brought down from the mountain slopes. Numerous washes and canyons have cut through this mantle of drift and more or less into the lake beds beneath. The general character of the deposit is well shown in the accompanying figure (Fig. 3). The lake beds are of essentially the same character as those described by Mr. G. K. Gilbert as occurring in the Lake Bonne- ville basin, and by Professor I. C. Russell as occurring in the Lake Lahontan basin. ^ They were evidently deposited in the bottom of a lake, into which, near the shoreline, coarse material was washed from the mountains, the finer sand and silt being carried farther out and deposited with the calcareous sediment and remains of fresh-water shells. There may be no a priori reason why such deposits should not have been made upon a lake bottom sloping from 3° to 5°, but this is improbable, and it would presuppose the existence of a lake 3000 feet in depth over the site of the present Owens Valley. If such a lake existed, there must have been a barrier to the south of Owens Lake, of which no trace now remains. This is not at all probable. South of Owens Lake the divide is about 220 feet above the lake."^ There is no appearance, as viewed from the south end of the Inyo Range, of the remains of a great barrier between Owens Lake and the drainage basin to the south. A second conception is that the Inyo Range has been ele- vated, the range and the country to the eastward rising and tilt- ing the lake beds toward Owens Valley and the Sierra Nevada. The accompanying diagrammatic sketch (Fig. 4) illustrates the relations of the Sierra Nevada, Owens Valley, the Inyo Range, and the lake beds resting on the westward slope of the Waucobi embayment. ' Mon. U. S. Geol. Survey, Vols. I and XI. == Owens Lake, 3567 feet above sea level ; Hawai meadows, 3782 feet at divide. Wheeler). POS T- PLEIS TO CENE ELE VA Tl ON OF INYO RA NGE 3 4 S The view of the lake beds taken from the foothills of the Sierra Nevada, looking eastward across Owens Valley, shows the lake beds at a, Fig. 2, at the level of the valley, and various out- crops from point to point toward the foot of the mountain at b. SierrtgNei'ada Inyo Ffsngi Lake beds Oyve/vs Vs/Zey Fig. 4. Diagrammatic outline section from the Sierra Nevada to the summit of the Inyo Range, a little north of Waucobi Canyon. The lake beds continue up Waucobi Canyon to a point approxi- mately beneath the letter c, where, as previously stated, they are 3000 feet higher than the lowest beds at a. The view includes the greater part of the Waucobi embay- ment, but does not extend on the north (left) as far as Soldier Canyon. The high point at^ forms the summit of the ridge east of the Waucobi embayment and overlooks Deep Spring Valley to the northeastward. This part of the range, from d to the head of the Waucobi Canyon above c, forms the eastern portion of the block which appears to have been tilted toward the base of the range at a. On the northeastern side the slope of the range extends down to the level of Deep Spring Valley. It was on the southeastern side of this valley that I found evidence of a comparatively recent fault which is of great interest in con- nection with the view that the Inyo Range has been raised to the eastward and tilted to the westward within comparatively recent times. The best locality at which to examine the fault line is on the southern side of the valley, at a point about seventeen miles north of the head of Waucobi Canyon. Here there is evidence that the bottom of the valley is sinking in rela- tion to the mountains on the southeastern side. This is shown by the presence of a comparatively recent fault scarp at the foot of the ridge and the truncating of the spurs along the base of the ridge where the fault scarp is not otherwise defined. Great springs flow out along the line of the fault, and Mr. Lewis Pa3'son informed me that he had been unable to find bottom, by any 346 CHARLES D. WALCOTT means at his command, in the large pools into which the springs flow, and that where wagons formerly crossed, at the south- eastern end of the flat surrounding the springs, animals are fre- quently mired in the soft mud of the flat. The fault cuts through the Pleistocene, leaving a northward-facing wall from twenty to thirty feet in height overlooking the pools and bogs. A distant view of the ridge on the southeastern side of the valley is shown in Fig. 5. The fault scarp mentioned is directly beneath a, and the extension of the fault crosses the ridge a little to the right of a. The eastern slopes of the Inyo Range ten miles south of Waucobi Canyon, and south of the road which passes east from the divide to Saline Valley, are very steep and join the bottom of Saline Valley as abruptly as the ridge on the southeastern side of Deep Spring Valley meets the level valley bottom. (Fig. 5.) Until a good topographic map of this region is made, it will be impossible to trace and connect the relations of the various faults and evidences of comparatively recent disturbance ; but I think that there is sufficient evidence in the sinking of the southern margin of Deep Spring Valley, in the phenomena observed to the south in Saline Valley, and in the position of the Waucobi lake beds, to sustain fairly well the view that the range has been elevated to the eastward and tilted to the westward. The total amount of the uplift cannot be accurately deter- mined, as there was undoubtedly a slope at the bottom of Waucobi Lake from its margin toward the present site of Owens Valley. It was probably not much greater than, if as great as, the elevation of the present divide south of Owens Lake, which is about 220 feet. It is to be borne in mind, however, that perhaps in the tilt- ing of the range the western edge under Owens Valley has been depressed, and that the valley has been silted up by the wash brought down by the river and from the adjoining mountains, quite as rapidly as, if not more rapidly than, the tilting of the Inyo Range has carried down the floor of the valley. If this is correct, the total elevation of the range since the lake beds were deposited may be as great as or greater than the difference in the POST- PLEISTOCENE ELEVATION OF INYO RANGE 547 level of the lake beds at the margin of the valley and the highest point in Waucobi Canyon, or 3000 feet. There are so many factors that might be considered with more data, that only an approximation can now be made of the total displacement. We are justified, I think, in placing it at about 3000 feet, and Fig. 5. View of southeastern portion of Deep Spring Valley, showing pond near spring, and the north face of the ridge that extends northeastward from the Inyo Range, a, Position of fault cutting through the ridge ; b, southwestern end of the ridge where it unites with the Inyo Range. thus recording the fact that a movement of considerable magni- tude has occurred. That the movement is comparatively recent is proved by the characters of the lake beds and their contained fossils, which indicate the age of the deposits to be late Pliocene or Pleistocene. It is interesting to note in this connection the account of the earthquake that occurred in Owens Valley in 1872. This earth- quake, according to Professor J. D. Whitney,' originated in Owens Valley, and its occurrence was accompanied by a sinking of strips of land. Mr. G. K. Gilbert visited Owens Valley eleven years later, and in his observations on the subject he says 'The Owens Valley earthcjuake. Overland Monthly, Vol. IX, 1872, pp. 130-140 and 266-278. 348 CHARLES D. IVALCOTT that "the principal scarp produced by the earthquake follows the base of the alluvial foot slope of the Sierra Nevada, and has a maximum height of about twenty feet. Where this height is attained there is a companion fault scarp ten feet high facing in the opposite direction, so that the net displacement is about ten feet. At other points the main scarp is associated with others running nearly parallel and facing in the same direction."' Professor Whitney, in his discussion of the earthquake^ sug- gests that such disturbances might have their origin in the com- pression exercised by an enormous weight of material raised to a vertical height of two or three miles above the surrounding country. The extent of the tilting of the Inyo Range to the south of Waucobi Canyon is not readily determinable. The eastern face of the range, toward Saline Valley, indicates the presence of a fault line, and the steepness of the slopes near the valley that the faulting is of relatively recent date. On the Owens Valley side the slopes are also steep, but at the mouths of the canyons there are great accumulations of talus that extend far out into the valley, and the monoclinal character of the range is broken by the presence of arching masses of strata of Triassic age, dip- ping westward. North of the embayment, along the high, broad mass of the White Mountain Range, no evidence of recent elevation or tilt- ing was observed in the hurried trip through Owens Valley along the western foot of the range. In a paper now in preparation I shall describe certain types of faulting and tilting of monoclinal blocks of strata that are char- acteristic of the Great Basin area of Utah, Nevada, and south- eastern California. The principal illustrations will be taken from faulted slabs of limestones collected in Waucobi Canyon on the western slope of the Inyo Range, and it is anticipated that they will aid in explaining the dynamics of such a movement as has evidently taken place in that portion of the Inyo Range which is described in this paper. Charles D. Walcott. ^Mon. U. S. Geol. Survey, Vol. I, 1890, p. 361. ' Loc. cit, p. 276. ITALIAN PETROLOGICAL SKETCHES. V. SUMMARY AND CONCLUSION. The volcanoes which have been described in the preceding papers belong to the main Italian line which extends west of the Apennines from Tuscany to Naples, and which may be called the Bolsena- Vesuvius line. In a paper already quoted on the " Extinct Volcanoes of the Northern Apennines " de Stefani ^ calls attention to the fact that these volcanoes may be referred to two distinct types, distinguished not only by their structure, but by their eruptive rocks. One type is that of the great strato-volcanoes, represented at Bolsena, Viterbo, Bracciano, and the Alban Hills, as well as at Rocca Monfina and Vesuvius farther south. The main struc- tural features of these have been already noticed. Petrograph- ically these strato-volcanoes are characterized by two features. In the first place leucitic rocks are very abundant at all of them, and form indeed their dominant and most characteristic feature. These are accompanied in most cases, not always, by non-leucitic rocks — ciminites and vulsinites, with exceptionally phonolite at Viterbo and toscanite at Bracciano. True trachytes are met with in abundance in the Vesuvian region and in small amount at Rocca Monfina, but elsewhere seem to be rare. Their second characteristic is the variety of eruptive products at each center, both among the leucitic and the non-leucitic rocks, as well as the abundance of tuffs. Though the variety be great, yet the rocks of each bear a great resemblance to those of the other cen- ters, so much so that the whole line (including the centers of the second type) form an excellent example of a "petrograph- ical province," and it is clear that the products of all the centers are related to each other genetically. ' De Stefani, Boll. Soc. Geol. Ital., X, 449-555, 1891. 349 350 HENR V S. WA SUING TON To the northwest of Bolsena and between the line of vol- canoes just noticed and the coast are the eruptive centers form- ing de Stefani's other type. The most noteworthy are those at Campiglia, Rocca Strada, Monte Catini, Monte Amiata, Radi- cofani, Tolfa, and Cerveteri. These are all of relatively small extent, and are in general higher in proportion to their breadth than the strato-volcanoes. Some of them, as Tolfa and Cerve- teri, were superfusive and apparently formed by domal eruptions of a pasty magma. Monte Amiata was probably a true strato- volcano." According to Lotti^ the Campiglic mass was lacco- litic in character, being covered by arching Eocene beds. It is noteworthy that dikes, which are practically unknown elsewhere, are quite numerous here. As de Stefani points out, these smaller eruptions differ from the large volcanoes in two important petrographical particulars. First, each individual mass is made up substantially of one kind of eruptive rock, the rock type at each being practically per- sistent throughout the mass. This persistency of type at each is in marked contrast with the great variety of products found at each of the strato-volcanoes. In the second place these smaller eruptions seem never to have produced leucitic rocks — at least none are known with certainty to occur as their products. They are uniformly non-leucitic and trachydoleritic, and extreme in type — either acid or basic. The rocks of Campiglia, Rocca Strada, Monte Amiata, Tolfa, and Cerveteri belong to the tos- canites — acid trachydolerites with over 65 per cent, of silica with or without free quartz ; while those of Radicofani, and probably also those of Monte Catini, are basic in character, with SiOg about or below 55 per cent., and belong either to the cim- inites or the basalts. TRACHYDOLERITES. The intermediate potash-rich rocks which are found along the Bolsena- Vesuvius line cdixxy basic plagioclase-labradorite to 'De Stefani, Boll. Com. Geol. Ital., 1888, 223. ^LoTTi, cf. Dalmer, Neu. Jahrb., 1887, TI, 207. ITALIAN PETROLOGICAL SKETCHES 35 I anorthite — along with orthoclase, and such rocks will be called collectively in this paper by the name of trachydolerite, a name proposed by Abich' as far back as 1841. This term seems the more appropriate since one of his type rocks is that of Monte Santa Croce at Rocca Monfina, which has the mineralogical characters of one of the subgroups with the chemical composi- tion of another. For those intermediate effusiv^e rocks in which the plagioclase occurring along with orthoclase is acid — ande- sine to oligoclase — the name trachya?idesitc,'^ which is in use in France, will be reserved. Cimiidtc. — The rocks belonging to this group are porphyritic in structure, the phenocrysts being of augite, olivine, and some- times feldspar, and the rather light to dark groundmass being seen under the microscope to be generally a hoiocrystalline paste of feldspar with augite and magnetite grains, glass base being rare. The lamprophyric habit of some of the specimens has been remarked on. The ciminites are characterized mineral- ogically by the presence of alkali feldspar and a basic plagioclase, augite and olivine, with accessory mag^ietite and apatite. Biotite and hornblende (especially the latter) are either entirely absent or only present in small amount. Chemically they are rather basic, silica varying from 54 to 57 per cent.; alumina is mode- rately high, as are the iron oxides ; magnesia and lime quite high, respectively 3 to 6 and 5 to 9 per cent. Alkalis are low for the rocks of the region, but potash is uniformly higher than soda. These rocks do not seem to be very abundant, except at the Monti Cimini, where they apparently occur in large quan- tities. In Table I are given all the reliable analyses of these Italian rocks which are known to the writer, together with those of the rocks of Radicofani. Of these No. i of the flow at Fontana di Fiesole, near Viterbo, may be regarded as typical. Vom Rath's analysis (No. 2) of a similar rock from the same region closely agrees with this, the chief discrepancy being found in the alkalies, ' Abich, Vulk. Erscheinungen, Brunswick, 1S41, 100. ^This term has been used in a more general sense in the preceding papers. 352 HENR V S. WA SHING TON w H y-'\ z W s J C) PQ <; z H < 1-1 < H H OjOOi--< 6 O d o n QO NO p) o ^ " T 9 rt *-! ^'^ 9 9^ ro -^ M 00 s- t-^c» i-i tS N d O PO ON d On cq d o ON PJ 0^ O 00 O ro On • IN 11 "-1 PJ 00 ■ • oi • t^ u-1 ts in 1^ . . On ; j-iod pi p5 d I ; d o 00 o 00 PJ 00 00 l-l t^oo ^ d ^ ro ■ (^ 11 00 p) 00 r O O IN tJ Tl-OO OO Tt- M . . lo d- 'j- ro S tr>o N -<0i-i"-ic0i-crj- . dr^M'S-OMOciLOM ■ O p-i d d o CO •* COOO 0) CO • vO OO M >-H vr^ . . l>.d^^^d iMcocooNd ! ' 0^ q d o I-I c4 ro r^iocooosMcooi-N . o^vDou^or^u^mcooo • r^t-^dt-^di-Hi^i-HLOM ; lo i-i 0) d CO d o ci . d^od ivd ImcJ-^^Om ; ; d o 00 " M o " r^ ■ 00 00 t^oo -"J- 1) oooN-a-d IdcocJONdS d o ■3- CO - c L, I ITALIAN PE TR OL O GICA L SKE TCHES 3 5 9 NOTES TO TABLE II. 1. Bolsena, H. S. Washington anal., Jour.Geol., IV, 552, 1896. 2. Bolsena, Vom Rath, Zeit. d. d. geol. Ges., XX, 291, 1868. 3. Bolsena, Ricciardi anal., Klein, Neu. Jahrb. B. B., VI, 8, 1889. 4. Vetralla, near Viterbo, H. S. Washington anal., JouR. Geol., IV, 849, 1896. 5. San Magno, Latera, Bolsena, Ricciardi anal., Klein, loc. cit., 10. 6. Torre Alfina, Bolsena, Ricciardi anal., Klein, loc. cit., 3. 7. San Lorenzo, Bolsena, Ricciardi anal., Klein, loc. cit., 3. 8. Monte Amiata, Ricciardi, Gazz. Chim. Ital., XVIII, 1888. 9. Monte Santa Croce, Rocca Monfina, H. S. VVashington anal., JouR. Geol., V, 252, 1897. 10. Monte Santa Croce, Vom Rath, Zeit. d. d. geol. Ges., XXV, 245, 1873. In the last rock (9 and 10) we have an example of the mutual exclusion of biotite and olivine in the Italian and other trachydolerites, to which attention has already been called by Rosenbusch." It is mineralogically a biotite-vulsinite, in that it is free from olivine, but chemically a ciminite, the silica being low and magnesia and lime high. The explanation of this pecul- iarity may be looked for in the complexity of the biotite mole- cule. It was pointed out by Iddings^ that, since the biotite molecule may be regarded as made up of molecules of olivine 2(Mg, Fe) O, SiOg and feldspathoid molecules of the form (H, K) 2 O. AI2O3. 2 SiOg, under conditions where the complex biotite molecule would be unstable it would dissociate, resulting in the formation of crystals of olivine and of a potash-alumina silicate, either leucite or orthoclase, according to circumstances. This explanation has also been adopted by Backstrom^ in the case of the leucite-basanites of Volcanello, he considering them to be the effusive forms of magmas which would appear intru- sively as minettes or kersantites. The same idea has lately been alluded to by Pirsson,'* who adds : "This process would of course find its most natural expression in magmas rich in magnesia and potash," which, it will be observed, is the character of the cim- inite magma. ' RosENBUSCH, Mikr. Phys., II, 773, 1896. ' Iddings, Origin Igneous Rocks, Bull. Phil. Soc. Washington, XII, 166, 172, 1892. 3Backstr6m, Geol. Foren. Stockh. Forh., XVIII, 155, 1896. ^ PiRSsoN, Jour. Geol., IV, 687, 1896. 360 HENRY S. fVA SIZING TON Applying this idea to the rocks at hand it is evident that we may regard the olivine-bearing ciminites as formed from rather basic trachydoleritic magmas, rich in magnesia, which under other conditions of solidification would have assumed the form of biotite-vulsinites — thus adding another to the number of cases of magmas of the same chemical composition forming on solidification different mineral aggregrates. In this connection and as favoring this view of the dissociation of the biotite mole- cule may be mentioned the great paucity in biotite of all the leucitic rocks of Italy, and, it may be added, of the purely orthoclase-trachytes of Ischia. There seems to be, indeed, as demanded by the theory, a mutual exclusion of biotite and leu- cite. Olivine, it is true, is not very abundant in the leucitic rocks, except at Vesuvius, the magnesia generally entering into the pyroxene molecules, perhaps owing to the richness of the magma in CaO. Toscanite. — The rocks belonging to this group are highly porphyritic and a glassy groundmass is often met with, though holocrystalline forms occur. They are characterized by the presence, along with the alkali feldspar, of a plagioclase which is rather more acid than in the preceding groups, generally vary- ing from andesine to labradorite, though anorthite is found at Monte Amiata and elsewhere. The most constant and promi- nent ferromagnesian mineral is biotite, which is never wholly wanting, and at some localities, as at Campiglia and Rocca Strada, is the only colored constituent. In other places (Brac- ciano, Cerveteri, and Tolfa) atigite or diopside is abundant along with the biotite ; while again, as at Monte Amiata, hypersthene replaces augite as companion to the biotite. Quartz is often present, either in well-shaped crystals or as anhedra, but olivine is never found. Magnetite is rare or absent. Silica is much higher than in the other groups, varying from 65 to 73 per cent., alumina and iron are low, magnesia and lirrre high considering the acidity, and the alkalies also high, with potash higher than soda. We may then regard the toscanites as acid biotite-trachy- dolerites, and the rocks of Bracciano, Cerveteri, and Tolfa would ITALIA N PE TR OL O GICA L SKE TCHES 3 6 1 be augite-toscanites, while those of Monte Amiata would be hyperstheiic-toscanite , if such a subdivision be deemed advisable. In Table III are given the best analyses of the representa- tive Italian toscanites. Those from Bracciano, Cerveteri, and Tolfa (Nos. I, 2, 3, 4, and 5) have been already described. Attention need only be called to the richness of the San Vito rock in NagO relative to KgO, as shown by Dr. Rohrig's analy- sis, and confirmed by mine made later. In this respect it is notable among the rocks analyzed by me. The rocks of Monte Amiata, whose composition is given in No. 5, have been so fully described by J. F. Williams, that the reader is referred to his paper. ^ The rocks of Campiglia and Rocca Strada, Nos. 7, 8, 9, and 10, are purely biotite-toscanites and are all quartz-bear- ing. It will be observed that the pyroxene-bearing toscanites (1-6) are notably less acid than those which carry only biotite. Trachydolerites elsetvhere. — Having thus reviewed the main feature of the trachydolerites along the Bolsena- Vesuvius line, it will be of interest to see if such rocks are found elsewhere. Their geographically nearest counterparts are found in the Lipari Islands where vulsinites and ciminites occur according to the descriptions of Sabatini^ and Mercalli.3 It must be noted, however, that these are not very abundant and are associated with true rhyolites, andesites, and basalts, and not with leucitic rocks. Some of the rocks of Monte Ferru in Sardinia and of the Ponza Islands may also belong to this group, though there is doubt as to the character of the plagioclase, and they may be rather trachyandesites. Another example of such transition rocks is found at the Azores, where Hartung^ in i860 noted the presence of rocks intermediate between the trachytes and the basalts, which he called by Abich's name of trachydolerite. The later investigations of Miigge^ have established the presence of such rocks, though we are again left in doubt as to the char- 'J. F. Williams, Neu. Jahrb. B. B., V, 403 ff., 1887. = CoRTESEE Sabatini, Isole Eolie, Rome, 1892. 3MERCALLI, Giorn. Min., Ill, 97, 1892. ■• Hartung, Die Azoren, Leipzig, i860, 291, 319. sMiJGGE, Neu. Jahrb., 1883, II, 201. 362 HENRY S. WASHINGTON w H ^ <5 1— < < '"' U w 7} n hJ H W < Z h < h-1 < H roj-.OrO'-'f^O)-. • rt rt rt rt c^ HHHHH MM '^^^,^H^OM■^^Hd^,l-lt,►-< Om-O . MI-HOMO o^t-» -o ■r^ovt^OM O M ■jj-i-i .vO -MCSt^u-iO ^Om .00 -t^O^DOro 6-*;n loM-rj-ciM t^ M u-iMm00 1>vDOMCSO>*Di-i-* ONO-^U'^MvO u^o rr u 1-1 o 1 iA>OMoJ2MrjSM6t.do" MTf.O .tOMOMOO voo •■^ •or^LOTj-cs O M On Tf O OO . Qv CS vD m <-0 u-lO M oj ; d IN N O M VO M vO 1-1 . 1-1 ^ -^ ro M pi t^ in dv ri r^Ot-^MDOn-)Mi-ioo inOO 0^ O ^ O LTiOO O CS vO IT) ^odro^.n'rorO'i-i-' I I ! ! '. d\d ^O M ^ ITALIAN PETROLOGICAL SKETCHES 363 NOTES TO TABLE III. 1. Monte San Vito, Bracciano, H. S. Washington anal. 2. Monte San Vito, Bracciano, A. Rohrig anal., JouR. Geol., V, 49, 1897. 3. Monte Cucco, Cerveteri, H. S. Washington anal., JoUR. Geol., V, 49, 1897. 4. Castle Hill, Tolfa, H. S. Washington anal., JouR. Geol., V, 49, 1897. 5. Tolfa, vom Rath., Zeit. d. d. geol. Ges., XVIII, 596, 1866. 6. Monte .\miata, mean of 7 analyses, J. F. Williams, Neu. Jahrb. B. B., V. 446, 1887. 7. Campiglia, Tuscany, vom Rath, loc. cit., 640. 8. Campiglia, Dalmer, Neu. Jahrb., 1887, II, 213. 9. Sassoforte, Rocca Strada, Matteucci, Boll. com. geol. ital., 285, 1890. 10. Torniella, Rocca Strada, Matteucci, Boll. soc. geol. ital., X, 677, 1891. acter of the plagioclase. Some of the analyses by Bunsen which Hartung gives are very similar to those of Tables I and II and indicate that it is basic. It is probable that similar rocks occur on Madeira,' and Renard^ describes andesitic trachytes from Teneriffe and Ascension which may also belong here. A group of rocks apparently resembling the Italian trachy- dolerites from near Buda-Pesth is described by A. Koch 3 as "labradorite-trachytes." They are composed of augite, horn- blende, and biotite, and occasionally garnet, with labradorite (analyzed) as phenocrysts, and orthoclase only in the ground- mass. The analyses much resemble those of the Italian trachy- dolerites, silica ranging from 52 to 67 per cent, and lime from 7 to 2, though magnesia is very low. Alkalies are much lower on the whole, but still higher than in normal andesites, and with potash predominating over soda. The most interesting group, however, to compare with the Italian rocks is that of the absarokite-banakite series from the Yellowstone Park recently described by Iddings,'* the rocks of which closely resemble those of the ciminite-toscanite series. The American rocks are similarly characterized by the presence of orthoclase and labradorite, with augite as the most prominent ferromagnesian mineral. The frequent occurrence of borders of ' Hartung, Madeira und Porto Santo, Leipzig, 1864. "Renard, Petrology of Oceanic Islands, London, 1890, 7, 23. 3 Koch, Zeit. d. d. geol. Gesell., XXVIII, 293, 1876. BIDDINGS, Jour. Geol., Ill, 935, 1895. 364 HENRY S. WASHINGTON orthoclase around labradorite cores is also noteworthy; a feature which we have met with so often in the Italian rocks. The most basic of them — the absarokites with silica from 47 to 52 and very high magnesia and lime — carry olivine, as do the more acid shoshonites with silica from 50 to 56; these latter correspond to the Italian ciminites. The last of the series — the banakites with silica from 51.5 to 61 — carry no olivine, and more biotite than augite, and are rather poor in ferromagnesian minerals ; these would correspond to some of the vulsinites. It is especially worthy of note that in these last rocks "augite and olivine are more or less completely replaced by biotite" — the same mutual exclusion being observed here that we find in Italy. Chemically the absarokites are much more basic, while the analyses of the shoshonites and banakites closely approxi- mate to those of the ciminites and vulsinites, except in the alkalies. The similarity is indeed so great that the ciminites might with propriety be called shoshonites, and the vulsinites banakites. But there are certain considerations which seem to render such a course inadvisable. The difference in the alkalies has been already mentioned. While in the Italian rocks they are high and with potash very largely preponderating over soda, in the American rocks they are rather lower, and soda, while always lower in percentage, is much closer to the potash and molecu- larly often surpasses it. In Italy again we find the acid tos- canites which are not represented at the Yellowstone Park, while at the latter we have the absarokites and nothing corre- sponding to them in acidity among the trachydolerites of Italy. The center of acidity, so to speak, at the Yellowstone Park seems to lie quite low, while in Italy it is much higher. Furthermore the great abundance of leucite at the Italian vol- canoes is unparalleled in the American district, where the absaro- kite-banakites are genetically related to normal andesites and basalts, though leucite occurs to a small extent.^ From these 'RosENBUSCH, in the last edition of his Mikroscopische Physiographic (II, 1216, 1896), describes these rocks in connection with the leucite-tephrites. ITALIAN PETROLOGICAL SKETCHES 3^5 facts we are led to believe that the parent magma of the Italian rocks was not only more acid than at the Yellowstone, but was preeminently a potash magma, which is certainly not true of the other. It may be added that the occurrence of two such well-marked and generally similar series of orthoclase-plagio- clase rocks at widely distant localities forms a strong argument for the recognition of such types in accordance with Brogger's views. Correlation of the tracJiydolcrites. — Before passing on to the leucitic rocks it will be as well to examine the relationships of the trachydolerites and see what their position is in regard to other rocks of our classification. Brogger has so ably presented the arguments for the recognition of intermediate types that nothing need be added on that score, since all the arguments advanced by him for the recognition of the monzonites as an independent group of the same rank as the syenites and diorites apply with equal force to the trachydolerites and trachy- andesites. The chief objection which may be brought against such ideas is that their acceptance would tend to "over- burden petrological literature with new names." This is undoubtedly true to some extent, and many, perhaps most, of these names might be but temporary in their use, but they would yet fulfill their legitimate object of enabling us to comprise a given set of characters in one word, make our ideas of the vari- ous rock types more clear and precise, and thus lead us toward the solution of that vexed question — a rational and generally accepted classification of rocks, such as is found in the organic sciences.' The law of the survival of the fittest would hold good here as in animate nature. The needless names and types will be gradually discarded, and on what is left we may build a nomen- clature of which the terms will be concordant both with each other and with the facts of nature, whenever the broad principle underlying the relationships of rocks shall have been discovered, ' In these, by the way, names are far more numerous than in petrology, and it ma}' be added that it would seem that the advantage of having well-defined types would outweigh the disadvantage of putting a little greater tax on our memories. 366 HENRY S. WASHINGTON T3 ? 0) .^ ^"F 13 "'■£ "C vi) ^1) -w 'A .'S, " o T3 cr 1 13 m C/2 ' .9 5 o C pq .S > n D.-C O OJ i-J ^ ::: '^ M W n 2 oZ-09 -vj ®0IS 09-sS -vo ^oiS 35-efr -w ®ois u U 1£ D OJ _j) OJ .ti -o c 11 in u o O M ./: (33 -0 << I c Q < < <| oi-19 •»;> ^0!S I9-SS ■»:' ^OIS SS-St' •z'^ ^0!S ij _u •a ii "0 r-1 .ti Ol w t-iS ^ 1 "0 H Q TiJ c c C ^ COO •T3 O. O U5 Sis >1 ^ G CM" > m U -G "2 >, '-^." CS J= rii! C3 s ^-^ H Si-t'g -vj «oiS ^79-09 '^^ ^OIS 09-oS -vj ^0!S ij u 4J ■o '1 ii ■s- -a A 2 bi5 1 ^"it c S •a 5-E2 1 > <1 t-i — 3 D U cJ^ -^ n HH rt < tri Si-S9 -vj ^0!S 59-09 •»•' ^OIS 09-oS ■»;> ^ois 0) -. > j2 0;^ fv. r»-. -C Ph (In H < 08-^9 '^y ^0!S 99-09 -vj ^0!S 09-oS •z'^ ^0!S PPV uinipap^ Diseg ITALIAN PETROLOGICAL SKETCHES 367 as the doctrine of evolution now underlies the classifications of zoology and botany. It is through considerations of this sort that I have felt justi- fied in proposing the new names in the preceding pages, and that I sketched the following brief outline of a grouping of the feld- spathic effusive rocks. In Table IV Brogger's idea is carried out, and furthermore the two main divisions of the lime-soda feldspars — the basic and the acid — are recognized. The reasons for the rehabilitation of an earlier system of classification' need not be gone into here, since the scheme represented is not proposed for general adoption, but only for the purpose of showing the rela- tionship to well-established types of such rocks as those of the absarokite-banakite series and the ciminite-toscanite series. In Table IV, then, the rocks included are classified by their dominant feldspars and by their silica content. In the first col- umn we find the tracliytic series, which includes only the purely or predominantly alkali feldspar rocks. These include the rhyolites (both of the potash and soda series) and the quartz- pantellerites ; the trachytes proper, such as those of Ischia, Berkum, and Algersdorf, and the pantellerites ; while the most basic members of this series are unknown at present, or perhaps incapable of existence, their place being taken by certain leuci- tites.^ The tracJiyandesitic series would embrace those containing alkali feldspar and acid plagioclase (oligoclase-andesine) in approximately equal amounts. Their most acid representatives would include such rocks as the Icelandic rhyolites described by Backstrom^ and the vulcanite of Hobbs;* those of medium acidity include what are called oligoclase-trachytes, and are typically represented by the domite^ of the Auvergne and many trachytes of the Siebengebirge ; while their more basic members 'Cf. J. Roth, Gesteinsanalysen, Berlin, 1861. ' They would represent chemically and mineralogically in an effusive form the basic augite-orthoclase shonkinite of Weed and Pirsson. (Bull. Geol. Soc, VI, 415, 1895.) 3Backstr6m, Geol. Foren. Stockh. Forh., XIII, 637, 1891. ^HoBBS, Bull. Geol. Soc, V, 598, and Zeit. d. d. geol. Ges., XLV, 578, 1893. 5 This might be used as the name for the group of medium acidity. 368 HENRY S. WASHINGTON are found in many of the trachyandesites of the Auvergne described by von Lasaulx' and Fouque.^ Many of the rocks of the Euganean Hills also belong here. The series of trachydoler- ites has been already described. The andesitic series would include the dacites and andesites, with their definitions narrowed to their original limits, 3 so as to include only the rocks whose feld- spars are acid plagioclase — oligoclase to andesine. In the basaltic series, carrying only labradorite and anorthite, we find the most acid members well represented by the "pyroxene-andesites" of Santorini, the greater part of whose feldspars have been shown by the researches of Fouque'* to be basic plagioclase. These have a high percentage of silica (65 to 69), and for convenience such rocks might be called sa?itori?ntes. In the moderately acid part of the series would be found rocks corresponding to many of the labradorites of French petrographers, as well as many rocks classed as basalts, but which are more acid than the normal, such as those of the Lowenburg, Meissner, and the Vogelsberg. These may or may not carry olivine, as is also the case with the intermediate and basic trachydolerites and andesites. Toward the basic end we would have the basalts proper, chiefly anor- thite-bearing and with or without olivine ; while below these would come such rocks as the limburgites and augitites. It will be seen from the above that the trachytes are rich in alkalies and poor in lime ; the trachyandesites rather rich in potash and also in soda and lime ; the trachydolerites rich in potash and lime, but poor in soda ; the andesites rich in soda and lime, but poor in potash and the basalts very rich in lime and poor in alkalies, especially potash. A somewhat similar table is given by Fouque and Michel Levy on page 155 of their Mineralogie Micrographique (Paris, 1879). The effusive rocks are divided according as they con- ' Von Lasaulx, Neu. Jahrb., 1870, 187 1, 1872. ^^ Fouque, Etude des Feldspaths, Paris, 1894, 254-270.^ 3 Roth, Gesteinsanalysen, Beriin, i86i, p. XLV. Von Hauer und Stache, Geologic Siebenbiirgeiis, 1863, 70, 79. * Fouque, Santorini et ses Eruptions, Paris, 1879; also Etude des Feldspaths, 317-320. The smaller groundmass microlites are more acid. ITALIAN PETROLOGICAL SKETCHES 369 tain alkali feldspar, oligoclase, labradorite or anorthite, and these groups are subdivided according to their ferromagnesian min- erals, biotite, hornblende, or pyroxene. Neither trachyandesites nor trachydolerites are mentioned. It must also be noted that a somewhat similar scheme was proposed by H. O. Lang^ in 1891 and carried out by him in great detail. He bases his purely chemical classification entirely on the relative proportions of the percentages of lime, soda, and potash, neglecting the other components and the mineralogical composition, so that the present grouping and his differ dis- tinctly from each other. LEUCITIC ROCKS. Comparatively little can be said of these rocks here, since detailed descriptions of manv of them have been already given, and since a proper treatment calls for more space than is avail- able. As regards classification, while they are on the whole sharply separated from the trachydolerites, yet inter se they grade into each other to such an extent that the discrimination of definite types is more than usually difficult and subjective in character. It will also have been evident from the descriptions given that many of the varieties described differ from the types gen- erally accepted. Thus those rocks which have been called leu- cite-phonolite for want of a better name contain only a small amount of nepheline and are quite distinct from the t3'pical German leucite-phonolites. The unfortunate difference of opinion as to the use of the terms leucite-trachyte and leucite- phonolite has been noticed, and in other ways the need is made evident of some general revision of all this group of important rocks. At present, however, the material is not at hand for such an attempt, which it is hoped will be undertaken later ; and the names used are to be regarded as provisional only, and the whole group of leucitic rocks as the subject of future investiga- tion. In Table V, only a few of the many analyses are given, 'H. O. Lang, Min. Petr. Mitth., XII, 199, 1891. 370 HENRY S. WASHINGTON ^ 00 oo nO w-> On On 04 m • in 2 " « ■^ r^ d ; in d ro NO 04 t^ GO • vo • On • on o Ti- r^ -"J- . LO • O • li^O >-i t3- 0) • c> ; Tj- ; 6 oj oo 6 d : d 04 m 04' vO UO 6 OsDO .-^000040 • ror^O -i-hOnOoooi • vd Lo oi ; M ro ■4-od i-< .' 00 r--- On On ^ H-l IN . oo-ooq .M^^c>- ; NO 1 " oo 04' d M ; d in 04' 2 00 • i-H r^oo a; 00 tt On t~» -^ • oot^ooOi>.oqror^>-H • On 00 M i_ M 00 0006 w ■ 00 nO On On 00 nO 04 S oo • OlTfO'UOOt^i-iON'^ . ooooi-ju-^qoo^ro • " 04' " 2 d 00 rf d d ! ^ H " ■ 04 d m in 0)' t^Tj-NO •o^ooor^ONio OO"* .i-iTj-u-iOOOI>. nO oood ; oo^d i^ ■^ 00 m d nO in 04 o. 00 t^ 04 TT 0- t^OO r^ 04 LO qi-H-^ojNDOro-^r^i-H in 0006 d ^ d 04 04 d M in d On 04' 00 -^ • 6 : ON^majmoot~»u->NO • 1-OnO 00 On rj- oo • doiinSoor^oit^d ; 04 H 00 0" t^ .>-HdNi-i ; E-i d m in nD ^ O O ^ N O • rOOO CO '1 NO • ^ rnao r^ ^ . 04 oo o on^o • 04 On On ^ N ■ 00 • •n ^ 04 • 'I m M 00 04 ►-• . Tj- u-)vO -* 06 1-0 U-) 1 rj- On 04' NO d Tl- q • 00 r^04 On>-i-)00 OnI^OnnO M rOi-H QLO'sfO 000 inoiNo'dNooiM^oid 00 ON On ON 00 ■* 04° ^ 00 . ^ '. u-i On roOO 00 " On On q 00 q "* On 00 00 q ro o4_ NOModrO'H'MrOMd On d 00 04' « oo t--. r^ 6 w-irO'^'UOOOO'^LO ■ 04 0NNOOOt^N004NO • 06 TJ- 00 2 0006 oq 06 d ; ro On ON 00 04' » On • 04 06 00 • r^ 1^00 00 On • nO • NO invo 00 m ■ d : in d " no d : nO d 'in H ,0^ bi p D i ITALIAN PETROLOGICAL SKETCHES 37 1 NOTES TO TABLE V. 1. Leucitite, Capo di Bove, Bunsen anal. 2. Leucitite, Crocicchie, Bracciano, Washington anal. 3. Leucitite, Sassi Lanciati, Bolsena, mean o£ 2, Ricciardi. 4. Leucite-basanite, Toscanella, Bolsena, Ricciardi. 5. Leucite lavas, Vesuvius, mean of 49 anals.. Roth, Geologic, 11,268. 6. Leucite-tephrite, Rocca Monfina, Rohrig. 7. Leucite-tephrite, Bracciano, Washington. 8. Leucite-tephrite, Monte Cavallo, Bolsena, Washington.' 9. Leucite-tephrite, Monte Bisenzio, Bolsena, Ricciardi. ID. Leucite-tephrite, Montalto, Bolsena, Ricciardi. 11. Leucite-phonolite, Lake Bracciano, Washington. 12. Leucite-phonolite, Bagnorea, Bolsena, Washington.' 13. Leucite-phonolite (?) Bolsena, Vom Rath. 14. Leucite-trach3-te. San Rocco, Mte. Vico, Washington.' 15. Leucite-trachyte, Monte Venere, Viterbo, Washington. 16. Leucite-trachyte, Madonna di Lauro, Viterbo, Rohrig. 17. Leucite-trachyte (?), Monte. S. Antonio, R. Monfina vom Rath. 18. Leucite-trachyte, Viterbo, vom Rath. the most representative being selected. Many more must be made before these rocks can be adequately treated from a chem- ical standpoint. By far the best defined type is that of the so-called lencidtes, which are quite constant in structure and mineralogical com- position, as already described. The chemical composition of typical leucitites is given in Table V, Nos. i, 2, and 3. They are very basic rocks, with high iron and lime, quite high magnesia, and high alkalies. Lezicitc-basalts seem to be quite unknown in the Italian localities — perhaps owing to the rich- ness of the magma in lime, which makes a basic plagioclase- free rock almost an impossibility if any of the MgO is withdrawn from a possible diopside molecule to form olivine. Outside of the Vesuvian region, where they occur in abundance, lejicite- basa?iites are only met with in small amounts. Their analyses (Nos. 4 and 5) resemble those of the leucitites except that they are higher in magnesia and iron as well as in lime, and those of Bolsena low in alumina. The leiicite-teplirites form a group which is one of the most difficult to classify satisfactorily owing to the ' Made since publication of their respective papers. 372 HENRY S. WASHINGTON large number of transition forms toward the leucitites on the one hand and the leucite-trachytes on the other. In general they are characterized by a rather doleritic micro-structure and the plagioclase is basic ; chemically they are very basic in char- acter, their silica content scarcely running above 50 per cent. Leucite-tephrites of this type abound at Bolsena, Rocca Mon- fina, and Vesuvius, and typical analyses are represented by Nos. 6, 7, and 8. Other leucite-tephrites occur whose silica content is higher. Examples are the rocks of Montalto (No. 10), in which Klein reports the feldspar as anorthite, and that of Monte Bisenzio (No. 9). Such acid leucite-tephrites, however, have not come under my own observation, and the high silica is to be explained by the presence of considerable orthoclase. Indeed my observations and such analyses as I have been able to make lead me to think that a typical Italian leucite-tephrite (which is poor in orthoclase) has a silica percentage of 50 or less, and that the plagioclase is generally, if not normally, a basic one. When we reach the leucite-trachytes and leucite-pho7wlites we find a decidedly more acid group of rocks with silica ranging from 55 to 59. These shade off into the leucite-tephrites to some extent, but on the whole are distinctly separated from them. Of leucite-trachyte we have few analyses, Nos. 14-18 being the only reliable ones known to me.' They show high alumina, low magnesia and lime, and high alkalies — especially potash, though this last feature is perhaps not as marked as we might expect. The analyses of leucite-phonolite (Nos. II, 12, 13) resemble these very closely, though soda is somewhat higher, and are quite different from those of leucite- phonolites from Germany,^ which are much more basic and with soda much higher than potash. Indeed, so great is the resem- blance to analyses of leucite-trachyte that these Italian leucite- phonolites ought properly to be called nepheline-bearing leucite- trachytes. They all carry orthoclase in large amount, especially in the groundmass, and nepheline occurs as the last product of crystallization. Only one case of a purely leucite-nepheline ' ZiRKEL, Lehrbuch, II, 465, 1894. ITALIAN PETROLOGICAL SKETCHES 373 rock was observed, that at the Asteria di Biagio, near Orvieto.^ It may be noted here that in all these leucitic rocks the domi- nant feldspars are orthoclase and a basic plagioclase, that augite and diopside are the prevailing ferromagnesian minerals, that both biotite and hornblende are rare, and that they are rich in potash and also in lime. They thus resemble in their broad features the non-leucitic rocks — apart from the presence of leucite. Turning to Table V it will be observed that there is a break in the silica percentages between 50 and 55. Below the former are the leucitites, leucite-basanites, and most of the leucite- tephrites ; between the two very few analyses are seen, while between 55 and 56 there is a large number of analyses, and above this last a much smaller number, with silica running up to nearly 60, which is the extreme figure for leucitic rocks. This feature is best seen when all the analyses available are tabulated. Space considerations prevent this being done here, but the break between 50 and 55 is very evident. The number of analyses represented is so large, and covers so many varieties of rock, that this clustering of the silica about 49 and 56 may be reason- ably assumed to exist in fact and not to be due to chance in the selection of analyzed material, as might well be the case with few analyses. We may note also in the analyses of Table V that, while alumina is somewhat variable, it is in general higher in the basic group than in the more acid, as is also true of iron oxides Lime and magnesia show the greatest differences, there being more than double the amount of them in the basic than in the acid group. There is less difference to be observed in the alka- lies, they being very high in both groups, but perhaps more so in the acid one. Coinparisoti of the trachydoleritcs and leucitic rocks. — The general similarity of the two groups has already been noticed, and it is further of great interest to observe, on referring to Tables I, II, and III, that a clustering of the analyses about definite points is to be found in the analyses of the trachydolerites as well as 'Jour. Geol., IV, 558, 1896. 374 HENRY S. WASHINGTON in those of the leucitic rocks. The ciminites and vulsinites reach no higher than 6o per cent, of silica, and lie in general between 55 and 58 per cent., while the toscanites are no lower than 64 per cent., and are in general much higher. Correspond- ing differences may also be noted in the other constituents, especially FcgOg, FeO, MgO and CaO. It is also seen that transitional chemical forms are of limited occurrence, though the anal37ses are not as numerous as in the leucitic group. The analyses, then, of both groups are clustered about certain points and do not form a gradual series from the most basic to the most acid, as Brogger understands a series to be constituted. A similar clustering of the analyses may be seen at such volcanic localities as the Yellowstone Park, Montana, Cape Verde Islands, and yEgina and Methana ; and the study of complete series of analyses would probably reveal the same state of affairs else- where. Indeed the fact may be said to be generally known, though perhaps not definitely formulated. An explanation which may be suggested is that this clustering of analyses is due to a quite complete course of differentiation, assuming the correctness of the differentiation hypothesis. Each main group of rocks, therefore, may be subdivided chemically, and mineralogically to a certain extent, into two subgroups (classing the ciminites and vulsinites together), a basic and an acid. Further, while the leucitic subgroup is more basic in each case than the corresponding trachydoleritic one, yet the variations in each are similar. These are best seen on tabulating the means of the best of the various preceding analyses. It is scarcely worth while to do so in this place, but the main features may be briefly pointed out. There is a difference in each case of 8 to 10 per cent, of silica; and iron, magnesia, lime, and to a less extent alumina, are higher in the basic group, while alkalies show less regularity. The molecular ratio of soda to potash varies in my separate analyses from .3 to 1.2, being fairly constant (.5 to .6) in the leucitic groups, and with higher soda at the extremes of the trachydolerites. It is probable that with more numerous analyses the resemblances would be even ITALIAN PETROLOGICAL SKETCHES 375 greater since there are indications of a division of the basic leucitic subgroup into two divisions corresponding to the ciminites and vulsinites. These would include on the one hand the Icucitites and leucite-basanites, and on the other the leucite- tephrites. There exist then along the Bolsena-Vesuvius line at least two well-defined series with correlated members in each, the Ciminite-Vulsinite-Toscanite Series and the Leucitite-Leucite- tephrite-Leucite-trachyte Series. These two resemble each other very closely, except for the presence of leucite in the latter and the uniformly greater basicity of the correlated mem- bers of this series as compared with those of the former. There also occurs a typically trachytic group in the Vesuvian region, while the phonolites of Viterbo mineralogically stand apart, but chemically are related to the leucite-trachytes. These series are, however, not gradual in Brogger's sense of the term,' but form related but separated groups of rocks. Differentiation of the magma. — The question of the process or processes of differentiation and the composition of the parent magma is of great interest, but needs far more detailed and extensive knowledge of the regions involved and of the rocks than is yet available. The question is somewhat complicated by the presence of leucite in such large amounts, and by the fact already noted, that it is possibly derived from a potential biotite molecule, or at least that some connection seems to exist between the two minerals. It is pretty generally believed that leucite is essentially an effusive mineral, though some cases of intrusive leucitic rocks are known. Therefore, although the first impulse is to consider the leucitic rocks as distinct differentia- tion products from the trachydolerites, yet caution must be used, since it seems possible that their differences are due rather to differing conditions of extrusion than to differing secondary magmas. This, of course, would not be true in all cases, since leucite would be formed in a potash-rich magma poor in silica, while orthoclase would take its place in one more acid. This is seen clearly in the rocks with silica below 55 which are almost 'Brogger, Grorudite — Tinguait Serie, Kristiania, 1S95, 169. 376 HENRY S. WASHINGTON uniformly leucitic, while those with silica above 60 are as uniformly leucite free and orthoclase bearing. It is in the intermediate group with silica from 55 to 60 that we must look for evidences of the connection of leucite with extrusive conditions. These magmas are apparently in a nicely balanced chemical condition, which needs only a comparatively slight change of conditions to throw them into the one mineralogical group or the other. It is known* that the leucitic rocks are almost invariably met with as flows, while the trachydolerites more frequently take the form of domal extrusions ; that the leucitic rocks are the products of later eruptions, on the whole,. than the trachydolerites except at Rocca Monfina ; and that they have been ejected at central vents while the trachydolerites are more generally peripheral. This is suggestive of the idea that leucite has been formed in place of orthoclase during the simmering of the magma in the throat of the volcano under low pressure,' while the peripheral ortho- clase magmas were partially crystallized under greater pressure. In view of the above facts and those given in the pre- ceding section I am inclined to think that the two main groups of rocks which are found along the Bolsena- Vesuvius line do not represent two distinct primary differentiation products, but that their differences are due to diverse conditions of extrusion, solidification, and the like, these conditions being, of course, secondary to the chemical character of the differentiation prod- ucts, as has just been explained. In regard to the character of the parent magma we may feel confident that it was very rich in potash and lime,=' basing our judgment on the preceding descriptions and analyses. More than this we cannot postulate with any degree of certainty, but its general composition is perhaps shown by the mean of 'Cf. Backstrom, op. cit., p. 163; also Lacroix, op. cit., 637 ff. = The hypothesis advanced by Lavis (Natural Science, IV, 138, 1894) that the large amount of lime is due to absorption (osmosis) from-the subjacent limestone traversed by the magma must be mentioned, though the writer can only admit its influence in a most limited way, if at all. The researches of Brogger and Pirsson have shown conclusively that it does not hold good elsewhere, and various considera- tions, which need not be gone into here, lead me to minimize its effect in Italy also. ITALIAN PETROLOGICAL SKETCHES 377 the extreme differentiation products — the toscanites and basic leucitic rocks — or by the mean of the ciminites, vulsinites, and acid leucitic rocks, which closely resemble each other. On such a basis we could suppose it to have approximately a com- position as follows: 8103 = 57-58, Al303 = i7-i8, total iron oxides as FeO = 6-7, MgO = 2-3, CaO= 5-6.5, Na30=2-2.5, K20 = 7-8, HoO=i-i.5 per cent. Whether the original magma differentiated horizontally from north to south we are not yet in a position to discuss, but at each center we may suppose the body of magma to have been quite completely differentiated. We would thus have the tos- canites and the basic leucitic rocks (leucitites and leucite- basanites) representing the extreme products — the oxyphyric' and lamprophyric types, respectively. Apart from questions of conditions of solidification, the absence of leucite in the vulsinites may be explained by their higher silica,'' while in the ciminites and biotite-vulsinites the high magnesia conditioned the formation of olivine, leaving the remaining part of the magma sufificiently acid for the formation of orthoclase rather than leucite. That these suppositions are possible may be seen on comparing analyses i, 4, and 9 of Table I and i of Table II, with II, 12, 14, and 15 of Table V, which are practically iden- tical except for the silica in the first two, and the magnesia in the second two of the trachydolerites. All this is, however, admittedly speculative to a large extent, and these views are advanced in a provisional way to be tested, and perhaps greatly modified, by future investigations. Finally, it has perhaps have been noticed that little or no refer- ence is made to the succession of the rocks. The omission is intentional, because it does not seem to the writer, even o-rant- ing that the order of eruptions is of the importance often attributed to it, that our knowledge of the subject along the Bolsena-Vesuvius line is sufficient to be of much value. Henry S. Washington. ' PiRSSON, Am. Jour. Sci., L, 11 8, 1895. 'The anomalous acid leucite-tephrites observed by vom Rath and Ricciardi are against this view, and some of them need confirmation. VARIATIONS OF GLACIERS/ 11.^ The first annual report of the International Committee on Glaciers has been published,^ and gives the state of glaciers in various regions of the world so far as reports have been received, with references to some of the original sources of information. This committee was appointed to stimulate and record observa- tions on glaciers. The following is a summary of the report. No glaciers, except those of the Alps, have been under observations long enough to yield very definite results ; but these have shown a decided periodicity of about thirty-five years in their size ; it is not improbable that glaciers in other regions may have a similar periodicity. A period is the time during which a glacier goes through all its changes ; it begins at a minimum, continues through the phase of increase, the maximum., and the phase of decrease, and ends at the following minimum. The Alps. — The glaciers of this chain were for the most part in the phase of decrease from 1855 to 1875; since then a number of glaciers have entered the phase of increase, namely : all those of the Mont Blanc group, about a half of those of the Valais, not more than a quarter of those of the Bernese Oberland, a few in the eastern Alps, and none east of the Brenner pass ; so that the phase of increase at the end of the nineteenth century has been limited and not general ; and observations from 1893 to 1895 show that many of the glaciers which have recently been advancing are again in retreat. Of the glaciers observed in the Eastern Alps in 1895 there were about fourteen increasing, twenty decreasing, and five stationary. ' Read before the Geological Society of America at the Washington meeting, 1896. ^ See this Journal, Vol. Ill, pp. 278-288. 3 Archives des Sciences, Vol. II, pp. 129-147, Geneva, 1896. 378 VA RIA TIONS OF GLA CIERS 379 In the Swiss Alps where observations have been more general and have extended over a longer period, we find, in 189S, twelve glaciers increasing, forty-eight decreasing, seven stationary, and ten doubtful ; in addition, several glaciers were measured for the first time in 1895, so that we may expect results from a still larger number in the future. The great majority of the glaciers of the French Alps are decreasing in size. The Pyretiees. — T\\& eleven glaciers of this chain for which we have results, show five increasing, five decreasing and one stationary. Considerably over 200 French glaciers are now under observation. The Caucas7{s.—A number of glaciers in this chain have been observed, showing a fairly general retreat; in 1894 some of the neve fields were growing larger. Ce?itml Asia.— The glaciers are mostly in the Pamir, the Tian-schan, and the Alai mountains. They are of considerable size, and appear to be pretty generally in retreat. Nova Zembla. — Glaciation is increasing. The Scandinavian ^//5.— The Norwegian glaciers do not show evidence of having participated in the great retreat of 1 8 50-1 880. The state of the vegetation in the immediate neigh- borhood of the glaciers shows that in many cases they have either kept or reattained the dimensions they had a century or so ago. There have been some advances and some slight recent retreats ; there are no indications of any present advance.' The Swedish glaciers are too little known to yield any definite results as yet. The Himalaya.— '^'xx W. M. Conway reports the glaciers of this range in retreat, so far as observed, with the exception of the Bagrot glacier, which is beginning to advance. The New Zealand Alps.— Considerable attention has of late been given the fine glaciers of this region. They seem to be either stationary or decreasing. ' See especially Beobachtungen lib Gletscherschwankungen in Norwegen, by E. RiCHTER. Petermann's Mitth. Vol. XLII, p. 107. 380 HARRY FIELDING REID United States of America.'^ — The glaciers of the United States have for the most part been so infrequently visited that the information regarding them is very meager. Professor Russell, in 1892, collected the evidence showing that in general they are retreating.^ A few glaciers, however, give evidence of being in a state of advance. The Malaspina glacier occupies a large plateau on the south- ern side of the St. Elias Mountain range. Though in general receding, a part of it near the Yahtse River was advancing and destroying trees in 1886.2 Professor Russell states that the southeastern portion of the same glacier near Point Manby has recently advanced a distance of 500 meters and again retreated.'* The Frederika glacier, in the interior of Alaska (long. 142° 35' N., lat.6i°40' W.) was the only glacier in its neighborhood advancing in 189 1.^ Mr. John Muir writes me that a glacier at the southern end of the Fairweather range was advancing and destroying trees when he visited it in 1880. Muir glacier, Alaska, which has in general been receding for the last hundred years or more, made a temporary advance between 1890 and 1892 of nearly 300 meters, but in 1894 it had again retreated to its limit of 1890. This glacier reaches tide water and ends in a vertical cliff of ice, 2.75 kilometers (9000 feet) long and 50 to 65 meters (150-215 feet) high; on each side of this cliff the glacier rests on the land and ends like an ordinary alpine glacier. The oscillation mentioned applies only "This paper gives a more detailed description of the variations of the glaciers of this country than is contained in the Rep. of the Intern. Com. An excellent account of our present knowledge of these glaciers has been given by Professor Israel C. Rus- sell in his recently published book, "The Glaciers of North America." ^ Climatic Changes Indicated by the Glaciers of North America, by I. C. Russell. Am. GeoL, Vol. IX, 322-336. 3 Shores and Alps of Alaska, by H. W. Stetson Karr, p. 77. 4 Am. GeoL, Vol. IX, 329. The best map of this region is in Russell's Second Expedition to Mt. St. Elias, 13th Ann. Rep. U. S. Geol. Sur., p. 6, or his Malaspina Glacier, this Journal, Vol. I, p. 221. s An Expedition to the Yukon District, by Willard Hayes, Nat. Geog. Mag., Washington, 1892, Vol. IV, p. 153. VARIATIONS OF GLACIERS 38 1 to the part facing the water ; the sides have been steadily reced- insf. Mount Rainier is a volcanic cone, 4400 meters high, in the state of Washington, bearing on its steep slopes about a dozen glaciers from five to ten kilometers long. Information has reached me concerning the Carbon glacier on the northern, the Willis on the northwestern, and the Nisqually on the southern face of the mountain. All three of these glaciers are steadily receding. Mr. Otto J. Klotz made a photographic study of the end of Baird glacier, Alaska (long. 132° 50' N., lat. 57° 8' W.) in 1894.^ This will be the beginning of a careful record of the variations of this glacier. In conclusion, the very incomplete data indicate that, with few exceptions, the glaciers of the United States are shrinking at the present time. REPORT ON THE GLACIERS OF THE UNITED STATES FOR I 896.3 Cook' s Met.— K glacier on the Kenai Peninsular has receded about 250 feet between 1880 and 1895.^ Mr. Dall writes that he thinks all the glaciers on the Pacific coast which he has person- ally visited are retreating. Chilcat Pass. — The glaciers on the southern side of this pass are receding. (/. E. Spur.) Glacier Bay region. — Mr. John Muir reports that he found Rendu and Carroll glaciers, at the head of the bay, from three to seven kilometers (two to four miles) shorter in 1896 than they were in 1879. Muir glacier also continues to recede. ' Studies of Muir Glacier, Alaska, by Harry Fielding Reid, Nat. Geog. Mag. 1892, Vol. IV, pp. 33-42; and Glacier Bay and its Glaciers, i6th Ann. Rep. U.S. Geol Sur., pp. 440-442. - This Journal, 1895, Vol. Ill, pp. 512-518. I recommend this article ta observers as an example of how much of permanent value can be done in a short time by the photographic method. 3 A synopsis of this report will appear in the Second Annual Report of the Inter- national Committee. 4W. H. Dall, Bull. Am. Soc, 1896, XXXVII, 15, 382 HARRY FIELDING RE ID Mt. Rainier. — Professor I. C. Russell writes me: In company with Bailey Willis and George Otis Smith of the United States Geological Survey, I visited Mount Rainier, Washington, and spent two weeks, from July 15 to August i, in examining the glaciers on its sides. The Willis, Carbon, Winthrop, Emmons, Nisqually, and Cowlitz glaciers were visited. Each of these furnishes clear evidence of having recently been lowered by melting, especially in the lower courses. The extremities of the three first named were examined and in each case a recent and marked recession was manifest. The extremity of Carbon glacier, as judged by Willis, has receded about 1 00 meters since his former visit in 1 88 1 . The extremity of the glacier at the date mentioned was a vertical precipice of nearly clear ice, but now has a slope of 55° to 60° and is debris covered. Willis glacier is divided at its terminus by a rugged boss of rock, for which I suggest the name Division rock, the down-stream face of which is a rugged precipice by estimate 120 or 150 meters high. I am informed by Willis, who saw it from below, that in 1883 the glacier broke off not far behmd the summit of this precipice and formed walls of ice descending on each side of it. The ice did not cover the highest peak on Division rock at the date mentioned, and there are about ten small spruce trees growing on the apex. These trees are certainly more than fifteen years old. The upstream side of the rock, below the trees, is strewn with stones and dirt and has evidently been recently occupied by the glacier. At the time of my visit the ice in the central part of the glacier had receded 175 meters from the edges of the precipice. Fresh lateral moraines elevated from thirty to forty meters above the level of the glacier in i8g6, and extending fully three kilometers (two miles) above Division rock, agree approximately with the 1883 level of the ice as reported by Willis. Willis glacier now divides on reaching Division rock into two sharp- pointed tongues of debris-covered ice which end with low frontal slopes The extremities of these tongues are about abreast of the summit of Division rock. Where the glacier divides a pyramidal monument of angular stones about one and one-third meters high is built. This monument records the limit of the ice at the place where it divides, on July 31, 1896. Up stream from Division rock there is another similar eminence which might possibly be mistaken for it, if the glacier continues to recede. The rocky knob referred to is now a part of the right or northern wall of the glacier. Eight hundred and forty meters, as measured by pacing, above the monu- ment described above, there is an ice-fall in the glacier 130 to 135 meters hisrh. The descent of the surface of the srlacier from the base of the ice-fall VARIATIOXS OF GLACIERS 383 to the monument is 280 meters, by aneroid. The gradient becomes progress- ively steeper as one descends from the base of the ice-fall. No marks were made to record the extent of the other glaciers visited. Mount Hood, Oregon. — There are eight glaciers on this vol- canic cone; all of them are steadily diminishing in size. The Eliot glacier on the north side of the mountain is about four kilometers (two and one-half miles) long and about 0.8 kilome- ter (one-half mile) wide; and shows a marked recession in the last three years. Before that the face was much steeper.' In the central part of the glacier near the end the velocity of move- ment has averaged about fifteen meters (fifty feet) a year since 1890. Coe glacier, which lies next to the Eliot on the west, is about four kilometers (two and one-half miles) long and 0.4 kilometer (one-quarter mile) wide. {W. A. Langille.^ Illecellewaet Glacier, Selkirk Mountains, Canada. — As there is at present no one on the committee representing the British col- onies, I give what information I have concerning this glacier.^ Professor Charles E. Fay wrote me in 1895 that the glacier had receded since 1890, and markedly since 1894. Photographs taken by Messrs. Parker B. Field and Philip S. Abbot in 1895 and 1896, respectively, show that the recession continued last year. There are prospects that more systematic observations of some of the glaciers of the Pacific slope may be begun next summer. Harry Fielding Reid. Geological Laboratory, Johns Hopkins University, April 8, 1897. 'Probably due to an advance.— H. F. R. * Captain Marshall Hall, to whose interest and energy the international commit- tee owes its existence, represented the British colonies until his sudden death in April 1896. A SKETCH OF THE GEOLOGY OF MEXICO. Mexico is a country whose geology is very little known even by its nearest neighbors. It nevertheless offers a wide range of interesting problems. Its geology has been much misunder- stood and this must be the excuse for presenting here a resume of a portion of the very interesting report^ recently published by the Geological Institution of Mexico. Aside from the itiner- aries with detailed observations, the report includes a synopsis of the geology of Mexico by the Director Sr. Jose G. Aguilera. In the following pages is a very much abridged translation, or abstract of Professor Aguilera's paper. The original is accom- panied by a small scale map on which the geology is spread over the areas left blank on the older map of Castillo^ and the whole forms a notable contribution to the geology of the region as well as being perhaps the best general summary of the present state of our knowledge of the subject. Geographically Mexico may be considered as consisting of a central tableland sloping to the north and northeast, and inclosed between two ranges of mountains which are separated from the oceans by strips of lowland narrowing to the south. The two mountain ranges unite in the central portion of the country, ris- ing above the lower land in the form of a colossal V, the arms of which continue into the United States as the Rocky Mountains and Sierra Nevada. The united range continues south into Central America with the low tableland of Yucatan, to the east rising thirty to forty meters above the sea. The central tableland or Mesa de Anahuac with an area of about 666,000 kilometers and an average altitude of about 1700 ' Bosquejo geologico de Mexico, Bol. del Inst. Geol. de Mexico, Nums. 4, 5 y 6. 270 pp., S pi., map. Mexico, 1897. 2 Bosquejo de una Carta Geologica de la Republica Mexico formada par dis- posicion del Secretario de Fomento, i88g. 384 A SKETCH OF THE GEOLOGY OF MEXICO 385 meters, continues without interruption from the high plains of Texas and New Mexico to the valley of Toluca, which rises against the flanks of the Nevada de Toluca, reaching an altitude of about 2630 meters. This great meseta forms a geographic unit of the first order. Breaks in the bounding mountains afford easy communication with the coastal region and furnish outlets for the drainage. It is a continuation of the Great Basin region and has all the characteristics of that area. To the north it widens and decreases in altitude ; to the south it narrows and rises. At its vertex are situated the City of Mexico and the two great volcanoes of the country. In Archsan time the southern, and a part of the western coast of Mexico, rose above the waters and formed a series of islands or perhaps a single strip of land which, as with the northern por- tion of the continent, served as a point of initial deposition, and around which has been laid down in successive geologic times the beds which now make up North America. The rocks of this period are numerous and present many variations between different types. In the southern portion of Puebla, in Guerrero and Oaxaca where the greater number of exposures occur, their order of deposition was as follows : {a) Porphyritic gneiss sim- ilar to augen-gneiss, losing its schistosity below and passing into a species of granite; [b) Phyllite gneisses resting upon and grading into the preceding beds; {c) Mica schists some- what abundant, at certain points garnetiferous and comformable with the rock below; {d) Phyllites, very argillaceous in their upper portion, but with the clay gradually diminishing toward the base and with concordant structural variation from stratiform to schistose. These beds rest upon chlorite, sericite and amphi- bole schists, which in turn cover the phyllite gneisses. Later than the deposition of the argillaceous phyllites and before the end of the Palaeozoic, there were numerous eruptions in the following order: (i) Gneissic granite, passing into a porphyritic granite which cuts the mica schists without pene- trating the phyllites. These rocks are shown in the northwestern portion of the republic in the City of Caborca, Sonora. (2) 386 H. FOSTER BAIN Granite proper, cutting the mica schists and the phyllites and shown at most of the Archaean exposures as well in the north- west as the southern portion of the country. (3) Granulite, cutting all the rocks of the Archaean. (4) Hornblende-granite in frequent dikes and occasional stocks, cutting all the Archaean rocks. (5) Pegmatite, passing into graphic granite and occur- ring as dikes cutting the gneissic and true granites. There seem to have been two distinct periods of eruption of the pegmatite. The older, seen in Sonora, cuts only the gneissic granites, the mica and the amphibole schists. The later and more common type cuts all the Archaean rocks. (6) Greisen, associated with the granites, forming segregation veins. (7) Diorite dikes, later than the preceding rocks but earlier than the end of the Palaeozoic. These are very abundant in the southern portion of Puebla and the northern part of Oxaca and Guerrero. The Archaean forms considerable areas, and in addition to the points mentioned is found in Zacatecas near Fresnilo, in Guana- juato in the vicinity of the capital, in Sinaloa, near the crest of the Sierra Madre, and near Vera Cruz. The rocks also form the axis of the peninsula of Lower California. The Palaeozoic has few representatives in Mexico. The rocks of this system whose age is definitely fixed belong to the Car- boniferous. Although considerable areas have been referred heretofore to the Silurian, and fossils characteristic of this terrane and said to be from Mexico, may be found in collections, we know of no exposures proven to be of that age. In the collec- tion of the Institution is a piece of limestone holding beautiful specimens of Orthis testudinaria Dalman, and sent to Professor Castillo from the Cuesta de Santa Teresa near Cachuamilpa in Guerrero. Careful search in the locality by Professor Castillo failed, however, to reveal the source of the specimen. Many geologists have assigned to the Silurian the slates found at Guana- juato, Catorce and Zacatecas. Those at Catorce are Jurassic, and while at the other localities there are beds older than the Jurassic, there is no good reason for assigning to them so great an age as the Silurian. With regard to the Devonian Professor A SKETCH OF THE GEOLOGY OF MEXICO Z^7 Aguilera has succeeded no better than in searching for the Silu- rian. A careful study of localities said to have furnished Devonian fossils has only resulted in showing the presence of certain prc-Jurassic rocks whose age, because of the absence of fossils, and the metamorphism which the rocks have suffered, cannot be definitely fixed. Carboniferous rocks of undoubted authenticity occur along the Guatemala border directly below the Cretaceous. The rocks are compact ash-gray limestones, containing Productus semireticulatus. Large areas of rocks in the central and northern portion of the country, assigned by Frazer, Hall and others to Silurian, Devonian and Carboniferous, are now known to be either Cretaceous or of unknown age. In the absence of rocks belonging to the first two periods of the Paljeozoic, it seems probable that Mexico, which during the Archaean was reduced to a group of islands, or perhaps to a single narrow peninsula stretching from California to Tehuantepec and Chiapas, suffered during the Silurian and Devonian a continu- ation of the ascendant movement which began at the end of the Huronian. The complete absence of stratigraphic and palaeon- tologic data relative to the first subdivision of the Permo-Car- boniferous authorizes the belief that during this time the eleva- tion continued, and makes acceptable the hypothesis that it was during this time that the various islands became united and formed the skeleton of the country. The Mesozoic is represented in Mexico by beds of the Upper Triassic and Jurassic, as well as the whole of the Cretaceous. The Triassic beds indicate a period of depression. The sedi- ments accumulated in marshes and estuaries along the western coast to a thickness probably of lOOO meters; sediments 600 meters in thickness being now found in Sonora. The deposition was interrupted by minor elevations as is shown by lithological variations and the presence of basal conglomerates. The depos- its cover a considerable area and are composed mainly of gray, red and yellow sandstones and gray to black slates. In general the rocks outcrop on the crests of low hills and ridges, being 388 H. FOSTER BAIN uncovered and resting upon crystalline schists, granites and sim- ilar rocks, or are intercalated between the Huronian rocks and the arenaceous marls and slates of the Upper Jurassic. The lat- ter is especially true south of Acatlan and about Tezoatlan. In some instances the Triassic is covered only by the Cretaceous or the Tertiary. The position of the rocks along the Gulf of California and in the territory of Puebla and Oaxaca indicates that after their deposition they were subjected to an elevation which, continuing to the present time, has placed them more than a thousand meters above sea level in Puebla and more than two thousand in Oaxaca. In spite of the evidence of the invasion of the sea in the Triassic, the absence of marine sediments makes it impossible to trace the old shore lines, but the same absence indicates that the land then extended notably more to the west than at present and that the deposits then made along the coast have since been buried. While the deposits of the Triassic were made in marshes and lagoons, some of which perhaps communicated with the sea, those of the Jurassic are in general such as denote continental seas and deep water. Certain of the Jurassic rocks, however, seem to have been deposited in an interior sea of slight depth. At the close of the Triassic the northwestern and southern parts of the country were elevated, draining the marshes. At the same time the southeast sank beneath the Lower, and later the Middle Jurassic seas. At the close of the Middle Jurassic there was a further land movement and the Upper Jurassic sea crept in over large portions of Coahuila and Oaxaca. The Jurassic rocks are conformable with the Cretaceous and are intimately folded with them. The Cretaceous rocks cover much the greatest portion of Mexican territory. They include three well-marked series of beds, the Lower, Middle, and Upper Cretaceous. These corre- spond respectively the Lower to the Neocomian, the Middle to the Cenomanian, Turonian and a portion of the Senonian, and the Upper to the Danian, and a part of the Senonian of Europe. A SKETCH OF THE GEOLOGY OF MEXICO 389 The beds of the lower division are largely shales, clays, marls, and greensands. The middle formation is mainly made up of compact ash-gray limestones frequently magnesian, though not in general constituting true dolomite. The limestones are rich in fossils, though they have been much metamorphosed. The upper member occurs only in the northeastern portion of the country and is represented by fine to medium grained, gray to red and yellow sandstones, alternating with clay shales of gray to black colors. The advance of the sea begun in the Jurassic, continued until the country was converted into an archipelago at the end of the Middle Cretaceous. There was then a general elevation, carry- ing all but the northeastern portion of the country along the Rio Grande, above the sea. This elevation was accompanied by fold- ing and mountain-making, continuing into the Upper Cretaceous. It was at this time that the main masses of the Sierra Madre of the east and west coasts were ridged up. The Upper Cretaceous was laid down by a retreating sea and by the close of the Mesozoic, Mexico had its present general out- line. There were, however, certain differences. Although the country then as now formed a great triangle with the apex in Central America, the width of the landmass was much less than at present. The Pacific coast line was farther west and Lower California was not yet separated from the mainland. The Gulf of Mexico had a more irregular coast line and extended to the west and southwest, probably uniting with the Pacific south of Guatemala. Yucatan and Florida were as yet covered by the ocean. During the Eocene there was a series of vertical oscillations, but the total result was an increase of territory. In the Miocene there was along the east coast an invasion of the sea, though not so as to cover the whole of the Eocene area. At the same time the Pacific advanced inland and the first peninsula of the Repub- lic was cut off, forming Lower California. Toward the close of the Miocene a new movement of elevation in the Atlantic region caused the sea to abandon most of its former dominion, the ele- 390 H. FOSTER BAIN vation terminaiing at the beginning of the Pliocene with the emergence of the peninsula of Yucatan and all of the southern part of the country which at the beginning of the Cenozoic had been buried beneath the waters of the two oceans. Upon the Pacific coast the depression seems to have continued in the Pliocene, so that Lower California was for a time cut off from Upper California by a canal and thus formed an island. On the Atlantic coast the Pliocene included a period of elevation suc- ceeded by one of depression and in turn followed by an eleva- tion, the latter continuing into the Quaternary. It is not possible to fix absolutely the age of all the Tertiary beds. Those west of Laredo in the Rio Grande valley seem to belong to the Timber Belt beds of the Lower Claiborne of Harris. East of the same place are beds in part Eocene, and in part Miocene, possibly Lafayette. In the peninsula of Lower Cali- fornia, particularly along the Pacific coast, is a series of shallow water deposits resting upon the trachytes, andesites and dacites of the Eocene and containing pebbles of rocks erupted in the Miocene with fossils of Pliocene character. Along the Gulf of Mexico are Tertiary marine sediments made up of incoherent shell conglomerates and compact limestones. In the upper por- tion these contain fossils which in other parts of the continent are Miocene, mixed with recent and Pliocene forms. In the lower part the Miocene forms dominate. The Tertiary in Mexico, as in the western portion of the United States, was a period of great eruptive activity. The wide variety of rocks and the large masses extruded are equally aston- ishing. Syenites, hornblende-diorites, quartz-diorites, diabases, porphyritic andesites, mica-andesites, dacites, and basalts are all present. With the andesites are trachytes, rhyolites and obsidians, and many transition varieties are present throughout the series. The eruptions begun in the Tertiary have continued to the present and have had much to do in shaping the topogra- phy of the country. The whole of the eruptive rocks are treated separately by Sr. Ezequiel Ordofiez. H. Foster Bain. Editorial. The George Huntington Williams Memorial Lectures were fittingly inaugurated last month by the course of six lectures delivered by Sir Archibald Geikie at the Johns Hopkins Univer- sity. The establishment of this memorial is due to the generos- ity of Mrs. George Huntington Williams, who provides in this way for a series of lectures to be given by geologists from various parts of the world on topics of interest to advanced workers in geology. No more appropriate introduction to what we hope may become a permanent source of inspiration to living geologists, while it will be a perpetual memorial of Pro- fessor Williams, could have been chosen than the subjects selected by Sir Archibald Geikie for his lectures — The Found- ers of Geology. Reviewing the beginnings of the science and the ideas and conceptions of those early workers, he not only presented a worthy introduction to a course which will deal largely with modern views and advanced theories, but he placed within easy reach of many a knowledge of ideas which may have appeared as new to us, but which in reality presented themselves to the earliest investigators. And in selecting an historical subject he followed a line which would have been most attractive to Professor Williams himself, and one in which he was peculiarly successful. Those of us who were prevented from hearing the lectures look forward to their publication with great interest. Through the courtesy of the Governor of Maryland and of other state ofificials, and through that of the presidents of the railroads in the state, Professor Wm. B. Clark, the state geolo- gist, was able to conduct several excursions into various parts of 391 392 . EDITORIAL Maryland, which were attended by Sir Archibald Geikie and the visiting geologists. They included an examination of the Cre- taceous and Tertiary deposits along the Chesapeake Bay, the coalfields of Alleghany county, the Silurian and Devonian for- mations of the Appalachian district of the state, as well as the Cambrian strata of the Blue Ridge, and the pre-Cambrian vol- canic rocks of South Mountain. Opportunity was thus given to become acquainted with the geological features of the state, which were first brought into prominence by the investigations of Professor Williams and his associates, and are now being fur- ther developed by the workers on the state survey under the direction of Professor Clark. * , * Sir Archibald Geikie, as Director General of the Geologi- cal'Surveys of Great Britain and Ireland, was prevented by the urgency of his duties at home from spending as long a time in this country as his friends here would have wished. The hearti- ness of his reception at those places he was able to visit may be taken as a guarantee of the high esteem in which he is held throughout the country. The hospitality shown him in Balti- more was repeated in Washington, Philadelphia, New York, and Brooklyn, in several of which places he delivered addresses. The lectures at the Johns Hopkins University attracted geolo- gists from widely distant parts of the country and were well attended, and Mrs. Williams is to be congratulated upon so suc- cessful an inauguration of the memorial lectureship, J. P. I. Reviews. Sofne Queries on Rock Differe7itiatio7i. By G. F. Becker. Amer. Jour. Sci. (4), Vol. Ill, pp. 21-40, January 1897. In this discussion Dr. Becker views his subjects from the physico- chemical side. He attempts to apply some of the results of Van t'Hoff, Oswald, Nernst, and other investigators to the question, "How may rock segregate into portions of distinctly different, yet of allied composition ?" He thinks that differentiation, or segregation, as he prefers to call it, can occur only in two ways: (1) By the increasing or decreasing the concentration of certain components, caused by variations in temperature or pressure, which exist in different portions of the magma. These components are thought to be dissolved in the remainder of the liquid rock, and to obey the laws of dilute solutions. (2) By separation into two immiscible liquids due to changes in tem- perature and pressure. These take place by means of molecular flow, or diffusion, and he thinks convection would hinder this segregative process. He gives as instances of molecular flow the diffusion of salts in water, and explains the relation of osmotic pressure and diffusion to gas pressure and diffusion, and mentions that the three funda- mental laws of gases, viz., Boyle's, Gay-Lussac's and Avagadro's have their parallels in three similar laws of solution. Two cases of diffusion are next considered: (i) The diffusion caused by heating a solution at the top. (If the solution is heated at the bottom, convection commences and segregation is prevented.) (2) By a difference of pressure at the bottom from that at the top. Any diffusion caused by differential pressure would be too slight to have any appreciable effect. As to (i) it will appear from arguments yet to be presented that the diffusion will be so slow that in entire geo- logical periods no effective change of composition would be possible. In his paragraph on the character of diffusion he states the units by which diffusion is measured, and that the rate of diffusion is expressed by the same mathematical formula as is the conduction of heat. 393 394 REVIEWS However this law only holds for dilute solutions, and he attempts to show that magmas are such dilute solutions. Applying the foregoing to the rate of diffusion under the assump- tion that the solvent fluid is kept at constant composition and the dissolved substance is diffusing through it, he calculates the speed of the diffusion of various salts in water. He finds that common salt will diffuse fast enough to semi-saturate water at a distance of i"™ in one day and loo meters in 270,000 years. From certain observations he infers that lavas are at least fifty times as viscous as water, and there- fore diffusion in such a substance would be fifty times as slow as that in water. Taking copper sulphate as the salt whose speed of diffusion should be multipled by fifty to represent the rate of diffusion in an average magma, he finds that in a million years there would take place a diffusion sufficient to impregnate the magma for forty-nine meters, and semi-saturate it to a distance of twelve meters. Finally he dis- misses the hypothesis of the diffusion of miscible liquids because (i) diffusion is too slow a process ; (2) a higher temperature is postulated at the top than at the bottom, and (3) because more or less convec- tion is unavoidable, and he proceeds to consider diffusion of immisci- ble liquids. He gives some examples of liquids which, above certain tempera- tures peculiar to each, are perfectly miscible, and below these definite temperatures separate into two immiscible liquids. This separation is accompanied by a sudden contraction of volume, and therefore all such separation is aided by increase of pressure. His points against such separation appear to be as follows: (i) The temperature of separation is first reached along the walls containing the magma, and therefore the separating component will condense on the contain- ing walls much as frost and dew on good conductors of heat. Such a separation, however, involves molecular flow from the interior of the fluid to the walls, and therefore no large portion of even a moderately viscous magma could thus separate. (2) If the fluid does not con- dense on the sides, it must aggregate as a fog in the center of the fluid. This fog cannot condense farther, as a fog does in the atmos- phere, on account of the great viscosity that he assumes the lava to have. (3) Magmas are not heated much above the melting point, and therefore the range in their temperature is not great enough to present much likelihood of their crossing the critical temperature of separa- tion. This follows from the law of fusion that if a melted magma REVIEWS 395 receives an increase of temperature, this increment will be expended in melting more of the surrounding rock rather than in raising the temperature of the already molten mass. Concluding, therefore, that his theoretical inspection proves that differentiation can at most play only a minor part in the explanation of consanguineous rocks, he advances his theory of the original heterogeneity of the earth. The points advanced for such a heterogeneity are as follows : 1. The land and water hemispheres. 2. Anomalies in gravity. 3. Distribution of feldspars in the west. 4. Distribution of metallic ores. 5. The seeming permanence of continental plateaus show this heterogeneity to be original. 6. Condensation from nebulous ring would forbid perfect homo- geneity. 7. The viscosity of the fluid earth would permit only a rude approx- imation to uniformity of composition. 8. Hypogeal refusion results in a re-formation of a heterogeneous liquid. 9. The transitional series of erupted lavas may be explained by chance mixing of unlike magmas. No attempts will be made to criticise Dr. Becker's views on the het- erogeneous composition of the earth. Many of his points are strong, and have been advanced by others against the conception of a primitive homogeneous earth. Yet his application of physico-chemical princi- ples is open to investigation. Dr. Becker is doubtless correct in his conclusion that any change in relative concentration of miscible liquids by means of differential pres- sure or temperature is inadequate to explain any considerable diffusion of a magma. Prior to Dr. Becker, Dr. Backstrom pointed this out very clearly. Nevertheless we may be permitted to examine some of the facts he cites and inferences he draws to prove this point. He says there are onl}^ two conceivable ways in which differentia- tion may occur; either by segregation into miscible or into immiscible liquids, caused by differential pressure, or temperature or both. May not these two operations be assisted by the cr3'stallizing of different double salts at different temperatures and pressures peculiar to each ? 39^ REVIEWS The brilliant investigations of Vant' Hoff along this line do not seem to have been applied as yet to the problems of petrology. After showing that the law from which he calculates the rate of molecular diffusion is good only for dilute solution, Dr. Becker says: " Magmas must be regarded as solutions of a series of very similar sub- stances, and it is known that in such cases the solubility of each is diminished by the presence of the others." This law holds only for salts dissolved in a dielectric, and is due to equilibrium relations between free ions and the undissociated portion of the salts. By "a series of very similar substances" is evidently meant a series of salts the members of which have some common ion or ions. If the data in the paper on the change of electric conductivity observed in rocks of different composition, while passing from liquid to solid, by Carl Barnes and J. P. Iddings,' are sufficient to prove that there is electrolytic dissociation in igneous magmas, then Dr. Becker's method is correct, and we need only to examine his assumptions on the viscosity, etc., of the lavas. If these data are insufficient to close the subject then Dr. Becker is reasoning from premises not yet proved. Admitting, however, that the weight of evidence is with Dr. Becker on this point we may examine his views on the viscosity of lavas, for although of importance to his conclusions regarding segregation of miscible liquids, the value of his objections to segregation oi immiscible liquids, depends upon the correctness of his assumptions on this point. He states that "There is some reason to think that the viscosity of even the most fluid lavas is more than fifty times as great as that of water." This is drawn (i) from a comparison of the speed of the flow of lava from Kilauea, with that of a stream of water of the same cross section and flowing down a slope of the same grade. (2) Since banded rhyol- ites show no diffusion of their layers into each other, it follows that if they had viscosity fifty times that of water they would show appreciable diffusion. It does not seem probable that the viscosity of an extruded lava would be the same as that of one highly treated and supercharged with vapors. The erupted lava has crusts forming on both outer and inner surfaces. It has lost much of its superheated steam and other vapors, and more or less crystallization has already started throughout its mass. The unerupted lava is permeated with superheated gases, and is at a very high temperature. Of course pressure increases the viscosity and ' Amer. Jour. Sci. (3), XLIV, 242-249. 1892. REVIEWS 397 we do not know where the increasing pressure overtakes the effect of increasing temperature and superheated gases, and therefore we may expect some lavas to be very liquid, and to permit considerable molec- ular flow, while in others the viscosity might be so great as to prevent all such diffusion. Therefore is the assumption that the viscosity of unerupted lavas is the same as that of an erupted and cooling sheet warranted, and is there not a possibility of both far greater and far less viscosity than that assumed by Dr. Becker ? We have, moreover, positive evidence that lavas may be very fluid if the pegmatitic structure in the rocks of Christiania described by Brogger as igneous be so, as he affirms for his region at least. Also the intrusion of erupted matter for long distances along planes of fis- sibility, in the Archean rocks of the Lake Superior region, demands more or less fluidity. If the alternate layers of banded rhyolites have a different amount of absorbed water vapor, as Professor Iddings sug- gests, this phenomenon might better be considered, as the result of impregnation of the magma by vapor shortly before its eruption. Moreover the fact that in the banded rhyolites we do not have dif- fusion between the layers is no proof that under other conditions we should not have such diffusion. Some magmas have been erupted under such conditions that they are perfectly homogeneous, while others show marked differentiation. We reiterate that Dr. Becker is correct in concluding that differen- tiation into miscible liquids could play only a subordinate part in rock segregation. His attack on the theory of differentiation into immiscible liquids seems open to question; for the separation must instantly begin throughout the portion that has reached the critical temperature. This separation will start at many different centers, as in crystallization, and since molecular flow is rapid for short distances, and very slow for longer distances, this operation would be rapid. When these immiscible liquids are once formed, flow, stirring, gravity, etc., will help aggregate like to like. This segregation may be still farther assisted by the temperature sequence of crystallization of double salts. His point, that lavas are too viscous to permit such separation and aggregation, has been already answered. His third point, that we cannot have wide enough range of temperature for this mode of separation, following from "the law of fusion," may have force in some instances, but where we have considerable mass of magma there 398 REVIEWS could be a large increment of heat as we approach the center. To sum up the whole matter, his objections to the segregation of miscible liquids are qualitatively correct, but the quantitative accuracy is impaired by the nature of the assumptions on which the estimate is based. These assumptions are thought to be incorrect in some instances, and in all cases to lack the positive proof necessary for a scientific demonstration. It seems unnecessary to assume that when immiscible liquids are formed segregation can only take place along the walls, and therefore involve extended molecular flow. Instead of this we conceive that the process may resemble crystalli- zation and take place at many centers, and therefore involves molecular flow through short distances. All stirring, currents, etc., instead of hindering, would aid in this case. Finally, we have the fact of a regular rock gradiation established ; a gradiation which is too widespread and uniform to be explained by an original heterogeneity of earth, and a chance mixing of the lavas. Our present knowledge, and the data at hand, are probably too meager to exactly explain the processes by which such gradation was accom- plished. Yet, we may not affirm, with Dr. Becker, that rock differen- tiation is impossible under the known laws of physico-chemistry. The problem will doubtless yield when attacked by the methods of modern physics and chemistry, and for this reason Dr. Becker's paper is most timely. Cyrus Fisher Tolman, Jr. Afi Introductioji to Geology. By W. B. Scott. The Macmillan Company. 1897. 573 PP- ;Si-90. The preparation of a satisfactory text-book on any subject so large as geology must always be a difficult matter. There is so much differ- ence of opinion as to where emphasis should be laid, as to what should be said and what omitted, and as to the order in which the topics should be treated, that no book is likely to command universal approval in all respects. Yet in spite of all the difficulties Professor Scott has suc- ceeded in preparing a book which will command respect and probably very general approval. He has shown in its preparation a sense of proportion which the makers of text-books sometimes fail to exhibit. When to this is added that he has made use of the newer literature throughout, so that the book is up to date, it will readily be inferred REVIEWS 399 that the book is likely to prove a useful one in institutions where brief courses in geology are taken b}' somewhat mature students. The book perhaps departs as much and as satisfactorily from the text-books heretofore in use, in its treatment of the later parts of his- torical geology, as at any point. In his treatment of the Mesozoic and later periods, the author has brought together much data not hereto- fore incorporated in a text-book, and his handling of that difficult part of the subject is much more satisfactory than that found in most text-books of corresponding scope. The illustrations in the volume are mainly new and attractive, many of them being reproductions from photographs direct. The illustra- tions of fossils seem to have been selected with great care, but are, on the whole, fewer in number than could have been wished. The publishers have done their usual excellent work in the prepa- ration of the volume. R. D. S. Missouri Geological Survey, Vol. XI; Clay Deposits. By H. A. Wheeler. 622 pp., 39 pi. Jefferson City, 1896. The eleventh volume issued by the Missouri Survey is well up to the standard of the previous work. It is a report of much more than local interest, and will doubtless become the standard book of reference for clay workers, filling a position analagous to that of the Manganese Report of the Arkansas Survey. The Missouri clay report is the most comprehensive work treating this subject issued by any American state since the New Jersey report of 1878. It monographs the subject of clays and clay working as exemplified in the wide range of deposits and pro- cesses in Missouri. It is written from the point of view of the engineer and treats of the different clays as adapted to various uses. Never- theless there are many geological problems whose solution will be the easier for it. The large number of new analyses, as well as the careful tabulation of a wide range of older ones is alone a feature of great value. The physical tests, the studies of fusibility, plasticity, and shrinkage, aside from their immediate practical importance, may be used to advantage in studies of the origin of mountains and of moun- tain-making forces. Probably few portions of geology are less under- stood or more complex than that which relates to metamorphism, and in order to understand the nature of metamorphic rocks it is necessary to have somethinsr more than a ijeneral notion of the nature of the 400 REVIEWS metamorphosed material. This has been recognized so far as the pet- rology of igneous rocks is concerned but the petrology of the sedimen- taries, being less inviting, has been largely neglected. It seems prob- able however that the field will eventually yield important results and certainly until it is better understood dicta regarding the metamor- phism of sedimentary rocks must rest largely on assumption. Professor Wheeler's report was not undertaken with this point in view, and yet his results often have considerable bearing on the subject. His work will also, it is believed, prove helpful because of the methods of study and measurement which he has formulated. To one not already familiar with the trade, the wide variety of products and the extent of the clay industry in Missouri, will doubt- less come as a surprise, and yet among the most valuable portions of the report are the suggestions with regard to the expansion of the industry. If these be followed the state will soon receive returns many times in excess of the cost of the work. H. Foster Bain. The University Geological Survey of Ka?isas. By Erasmus Haworth and Assistants. Vol. II, 318 pp. 25 plates. Topeka, 1897. Under the direction of Professor Erasmus Haworth the University Geological Survey of Kansas has published a second volume upon the geology of that state. It is a companion and supplement to volume one, and covers the western half of the state as the former covered the eastern portion. It deals chiefly with the stratigraphy of the Upper Permian, Cretaceous and Tertiary formations, and while it affords much valuable information to the geologist it is written primarily for the citizens of the state. In the preparation of the volume Professor Haworth has enlisted the cooperation of Professor C. S. Prosser of Union College, Schenectady, N. Y., with his two assistants Mr. J. W. Beede and Mr. C. N. Gould, Professor S. Williston of the State Uni- versity, and Mr. W. N. Logan. The report is illustrated with numer- ous half-tone reproductions from photographs, geologic sections and maps. The first paper is on the "Physiography of Western Kansas" by E. Haworth. The drainage of the region as a whole is considered and the present drainage is compared with that of Tertiary time. Follow- REVIEWS 401 ing this general treatment each of the individual streams is discussed in greater detail. The next paper is by Professor C. S. Prosser upon "The Upper Permian and the Lower Cretaceous." Professor Prosser's work upon the Permian of Kansas is well known, and that portion of the present paper devoted to the Permian is but a supplement to what he has previously published. The "Red Beds', or Cimarron Series come in for consideration here. These beds have been placed by different workers in the Permian, Jura-Trias and Cretaceous. The entire absence of fossils in the series in Kansas makes their correlation at best uncertain. The presence of Permian fossils in the "Red Beds" of Texas, however, which have a similar position but differ lith- ologically, the seeming conformity with the true Permian below and the marked unconformity with the Cretaceous above, point to the Permian or Triassic age of the series, but Professor Prosser states that the cor- relation of the beds with either the Permian or the Triassic is as yet a matter of uncertainty. The second part of Professor Prosser's report is upon the Lower Cretaceous, which is represented in Kansas by the Washita division of the Comanche Series. The stratigraphy of the formation is discussed in detail, many sections and lists of fossils being given. A chapter on "The Upper Cretaceous of Kansas "is contributed by W. N. Logan. The series is well represented in western Kansas by the Dakota, Benton, Niobrara, and Fort Pierre groups. Each of these formations is discussed in more or less detail. The next paper is by Professor S. W. Williston upon " The Kansas Niobrara Cretaceous. This is followed by the " Physical Properties of the Tertiary" by Professor E. Haworth, in which the stratigraphy, lith- ology, origin and mode of formation of the Tertiary is considered. "The McPherson Equus Beds" are described by Professor Haworth and J. W. Beede. These beds have been considered as of glacial origin. This explanation of their origin is disputed by the authors and they state that no satisfactory interpretation of them can as yet be advanced. The last chapter in the volume is by Professor Williston upon "The Pleistocene of Kansas." The deposits of this age are treated briefly and mention is made of the various vertebrate remains which have been found in them. S. W. 402 REVIEWS Prelimi?iary Report on the Marquette Iron-Beari7ig District of Michigan. By Charles Richard Van Hise and William Shirley Bayley. With a Chapter 071 the Republic Trough. By Henry Lloyd Smyth. U. S. Geological Survey, 15th Ann. Rept., pp. 477-650, pis. 13-26, figs. 9-20; 1895. The Marquette district, the oldest important iron district in the Lake Superior region, is well known to students of pre-Cambrian geology, and while it has been studied in detail before, the present report represents the first systematic and successful attempt to unravel the structure of the region and to determine the sequence and relations of the various rock bodies. The key to the structure of this district, which is in general the key to the structure of much of the Lake Superior region, has been known for several years ; it is that below the Keweenawan are three unconformable rock series, frequently so closely folded together that their separation is very difficult. These three series are the Archean, the Lower Huronian, and the Upper Huronian, or, as termed in this report, the Basement Complex, the Lower Marquette, and the Upper Marquette. The district studied extends from the Lake Superior shore just south of Marquette westward to Michigamme, a distance of over thirty- five miles; it is nine miles in width, but in places much more than half of this distance is occupied by rocks of the Basement Complex, and has not been studied in detail. The Lower and Upper Marquette rocks occur in a basin flanked on either side (north and south) by the Basement Complex, which is composed of an intricate mixture of granites, gneisses, schists, and surface volcanics, all thoroughly crystal- line, and cut by basic and acid dikes. The Lower Marquette is separated into six conformable formations as follows, in ascending order : Mesnard quartzite, Kona dolomite, Wewe slate, Ajibik quartzite, Siamo slate, and Negaunee formation. During Lower Marquette time the transgression of the ocean was from the east, so that the lower formations are represented only in the east- ern part of the district, and on going westward higher and higher strata are found, resting directly on the underlying Basement Com- plex, between which and the Lower Marquette there is a marked uncon- formity. The Negaunee is the Lower Marquette iron-bearing forma- tion, and is composed of sideritic slates, griinerite-magnetite schists, ferruginous slates, ferruginous cherts and jaspilite, all of which are REVIEWS 403 genetically related. With the Negaunee are masses of basic igneous rock, both extrusive and intrusive, the latter being the more common. There are three classes of ore deposits, one at the bottom, one within, and one at the top of the iron-bearing formation. The first two are generally of soft hematite, while the third, which is associated with the uppermost part (jaspilite) of the iron -bearing formation and passes up into the lower part of the Upper Marquette, is of hard hematite. In origin the ore is the same as in the Penokee-Gogebic range, i. e., it is derived from an original sideritic chert, and has been concentrated in certain positions by percolating waters, and the silicious portions of the rock have been removed from these areas of ore concentration. The ore bodies lie in pitching troughs formed of comparatively impervious rock, as slate or basic dike rock, or both. The Upper Marquette, resting unconformably on the Lower Mar- quette, is divided into the Ishpeming, the Michigamme and the Clarks- burg formations. The first is composed of the Goodrich quartzite and the Bijiki schist, the latter of which is a griinerite-magnetite schist. Small ore bodies occur near the top of the schist and also in the Michigamme formation. The Clarksburg is composed of volcanic material, and the volcanic activity is thought to have begun during the time of deposition of the Ishpeming formation, and this igneous mate- rial is regarded as replacing, where it occurs, the upper part of that formation and the lower part of the Michigamme. The Republic trough is a syncline with the axis nearly horizontal and running in a northwest-southeast direction. The three uncon- formable series described in the Marquette area are here represented, but of the Lower Marquette only two upper members are present, and the Clarksburg formation of the Upper Marquette is lacking. The iron ore deposits, which are of magnetite and hematite, are in the upper part of the Lower and the lower part of the Upper Mar- quette. In structure the Marquette district is complicated, the rocks having been subjected to severe folding. The general structure is a broad complex synclinorium, in which the folds have an east and west direc- tion. At the edges the folds are sharply overturned, and on going toward the center higher and higher beds are encountered. The struc- ture is similar to the composed fan folds of the Alps, except that the whole has sagged down, forming a synclinorium rather than an anti- clinorium ; to this the name of Marquette type of fold is applied. In 404 REVIEWS addition the rocks have been compressed more or less closely by a force acting in an east-west direction. This report represents a vast mass of most detailed examination in the field and in the laboratory, the completed form of which is to be expected as a monograph of the U. S. Geological Survey. With this will be issued maps, on the scale of four inches to the mile, showing all outcrops and other details of structure and of topography. As already stated, this report is the first successful attempt to present the structure of the Marquette iron-bearing district, an area which has been much studied before, but most of whose problems have remained unsolved. While there are a few minor points yet unsettled, the major questions have been answered. ^\vt. whole is a practical and success- ful application of the advanced methods and principles of structural geology applied to nonfossiliferous rocks, which methods and principles have been so masterfully presented by the senior author. U. S. Grant. Abstracts. Geological Atlas of the United States. Folio 30, Yellowstone National Park, Wyoming, i8g6. The Yellowstone Park folio, recently issued, consists of six pages of descriptive text, three pages of illustrations, four topographic sheets (scale I : 125,000) and four sheets delineating the areal geology of the region. The general descriptive text, giving a succinct narrative of the geo- logical history and development of the park country from the time of the earliest continental land surfaces up to and including the hydro- thermal phenomena as seen today, was written by Arnold Hague, o-eologist in charge. It is followed by an account of the sedimentary rocks, from the earliest Cambrian deposits to the Tertiary conglomer- ates, by Walter Harvey Weed, and a brief petrographical descrip- tion of the igneous rocks, by Joseph Paxson Iddings. The area of country covered by the Yellowstone National Park folio lies between parallels 44° and 45'' and meridians 110° and 111°. It is situated in the extreme northwest corner of Wyoming. By far the greater part of the park is included within the area of the four atlas sheets, but a narrow strip lies to the northward in Montana, and a still nar- rower strip extends westward into Idaho and Montana. In the organic act establishing the park. Congress declared that the reservation was "dedicated and set apart as a public park and pleasure ground for the benefit and enjoyment of the people." Owing to the marvelous dis- play of geysers and hot springs, and such remarkable physical features as the Grand Canyon and Yellowstone Lake, this folio possesses more than ordinary interest to geologists. The central portion of the Yellowstone Park is a broad volcanic plateau, with an average elevation of 8000 feet, surrounded on nearly all sides by mountains rising from 2000 to 4000 feet above its general level. The continental watershed crosses the park, separating the w^aters of the Atlantic from those of the Pacific, the Missouri and the Columbia, by the way of the Yellowstone, and the Snake, finding their sources on this plateau. 405 406 ABSTRACTS The oldest rocks of this region are granites, gneisses, and schists regarded as of Archaean age. They occur in all the mountain uplifts that encircle the park, but are unknown in the central portion. Around these ancient continental land masses there was deposited a conform- able series of sandstones, limestones, and shales extending from the time of the Middle Cambrian, the lowest beds exposed, through the Upper Cambrian, Silurian, Devonian, Carboniferous, Juratrias, and Cretaceous, including the Laramie sandstone. Nearly all these great divisions of Palaeozoic and Mesozoic times are characterized by a typ- ical fauna. With the close of the deposition of the Laramie sandstone the con- formable series of sediments came to an end. The entire region was elevated above the sea, the movement being accompanied with plica- tion and folding of strata. This primary orographic uplift which blocked out the main ranges of the northern Rocky Mountains, has been designated the post-Laramie movement. Tertiary sedimentary rocks occupy only small areas in the park, the greater part of the region being covered by extensive flows of lava. A heavy mass of coarse conglomerate, designated the Pinyon con- glomerate, has been referred to the Eocene ; and Pliocene conglomerate and coarse sands are well exposed in the escarpments of the Grand Canyon. Volcanic energy, which has played so great a part in the geolog- ical development of the country, was connected with the post-Laramie movement and followed closely upon the elevation of the mountains, and the accompanying dislocation and compression of strata. The eruptive masses, in forcing their way upward, sought egress along lines of least resistance, or wherever strain has been greatest in the crumpled sediments. Volcanic outbursts continued on a grand scale throughout Tertiary time. During the Eocene and Miocene periods enormous volumes of fragmental ejectamenta, largely composed of andesitic and basaltic breccias, were thrown out. The Absaroka Range was almost wholly built up of volcanic material. Evidence of this long-continued action is shown in the well-preserved fossil floras of Eocene and both Lower and Upper Miocene age. The famous fossil forests of the Yellowstone are of Miocene age. After a period of great erosion the depressed basin lying between the encircling ranges was transformed into the present Park plateau by the extravasation of immense flows of rhyolite ABSTRACTS 40/ of Pliocene age. Still later the recent basalts, the last of the igneous extrusions, poured out over the rhyolite along the ridges of the plateau A generalized vertical section accompanies the text, showing the order of succession of the extrusive flows, from the earliest out- bursts to the final dying out of eruptive energy. It is shown that long- continued currents of heated waters and acid vapors have acted as powerful agents in decomposing the igneous rocks of the plateau, and date back to Pliocene time ; at least they were active before glacial ice covered the country. Hot springs, geysers, and solfataras are closely associated with the rhyolite, and in fact thermal activity is confined almost exclusivelv to areas of this rock. The illustrations relate mainly to the occurrence of both active and dormant geysers and hot springs or some phase of volcanic geology. The Grand Canyon, well shown in the illustration, is a profound gorge cut in the Pliocene rhyolite, the brilliant coloring being due to the action of thermal waters. Geologic Atlas of the United States. Folio 24, Three Forks, Montana, /( This folio by Dr. A. C. Peale, consists of five pages of text, a topo- graphic sheet (scale i : 250,000), a sheet of areal geology, one of economic geology, one of structure sections, and one giving a gener- alized columnar section for the district. The area covered comprises the square degree which lies between the meridians 1 11° and 112° and the parallels 45° and 46 , m the southwestern, mountainous portion of Montana, and includes 3354 square miles. In the extreme southeast corner the Yellowstone National Park barely falls within the area. The folio derives its name from the valley in which the Jefferson, Gallatin, and Madison rivers unite to form the Missouri. The "Three Forks" valley is important from an historic standpoint, as being the point which Lewis and Clark reached in July, 1805, when they named the three confluent branches of the Missouri. , The text begins with a general description of the geography and topography of the region, and then takes up the general geology. The oldest rocks in the region are the crystalline schists and gneisses, designated as of Archc^an age, which in pre-Cambrian time formed a land niass comprising nearly all the area included in the map. While 408 ABSTRACTS the Algonkian beds were being deposited to the extent of from 6000 to 12,000 feet, there was a gradual subsidence of the whole region, and shallow seas for the most part prevailed. During the Palaeozoic age there were many minor oscillations of the surface, which were more frequent during Cambrian time than during the deposition of the Devonian and Carboniferous limestones. Toward the close of the Cretaceous period a general elevation began,, which was accelerated after the deposition of the Laramie formation. The formation of the mountain ranges, together with the subsequent erosion, resulted in many valleys, which eventually were occupied by fresh-water lakes. These lakes attained their greatest extent in the Neocene period, last- ing in all probability until the Pleistocene period was well advanced, and during their earlier stages immense bodies of wind-carried volcanic dust were deposited in their waters, and are now seen as beds of pure white. At the same time the dust fell upon the surrounding country, from which it was afterward washed into the lakes, forming an upper series of yellowish and rusty colored beds. These dust showers destroyed both animal and vegetable life, and the remains carried into the lakes were buried in their deposits, where they are now found as fossil bones and opalized and silicified wood. Under the "Description of Rock Formations" are outlined all the formations from the Archean gneisses up through the Algonkian, Cambrian, Devonian, Carboniferous, Juratrias, Cretaceous, Eocene, Neocene, and Pleistocene. The rocks of more than half of the area are of sedimentary origin, while the crystalline rocks occupy approxi- mately 1000 square miles, the remaining third of the area being cov- ered with igneous materials. Prominent among the latter are the andesitic breccias which form the main part of the Gallatin range, the great porphyritic laccolite occupying the center of the Madison range, and the basaltic plateau which lies west of the Madison valley. Under the heading "Structural Geology," after a general consid- eration, the vertical and horizontal movements are discussed, and the development of the lake basins is described. The arrangement of the rock-mass is complex, the structure being complicated by laccolites, dikes, and surface flows of igneous material. Unconformities exist, showing that areas previously raised to land surfaces and worn down have subsided, have been crossed by an advancing shore, and later have passed beneath the sea. The great series of conformable strata is closely folded, and has been pushed up in arches, many of which have been ABSTRACTS 409 overturned from the effect of horizontal thrusts. The simple as well as the overturned synclines are marked by areas of Laramie Cretaceous beds, which, at the time of the folding, were the latest and highest of the formations. The arches between the troughs having been broken and exposed by the elevation, excessive erosion has worn them down to the older rocks, exposing the Archaean, which usually forms the axes of the uplifts. Unlike the Appalachian folds, which are strikingly parallel and continuous, these folds lie in various directions, due to several independent centers of uplift. Three great faults cross the Gallatin range, two of them extending across the Madison range to the extreme western part of the area. Following or accompanying the folding of the Cretaceous and pre-Cretaceous strata the detritus result- ing from the greatly facilitated erosion, together with volcanic material erupted during this epoch, was deposited unconformably on the eroded upturned edges of the earlier-formed strata. The lake basins are now the floors of extensive valleys separating the detached mountain ranges, which rise about 6000 feet above their bases. As the lake deposits are at least 2000 feet in thickness, the difference of elevation between the bottoms of the lake basins and the summits of the peaks must be at least 8000 feet. The region was a mo'untainous one before the development of the lakes, but in the evo- lution of the existing relief, movements and erosion have both oper- ated to accent the topographic differences. The principal economic resources of this region are gold, silver, iron ore, copper, limestone, and coal. The occurrence of coal in Devonian rocks on the north side of the Jefferson canyon is of geologic interest, although not of any great economic importance. The fine pumiceous volcanic dust found in the old lake basins has been utilized to a very limited extent as a polishing material. Brick clays occur, and are used to a small extent in a few localities, especially near Boze- man. In addition to the economic resources just referred to, the sheet of economic geology has indicated upon it the localities of building stone and mineral springs. Geologic Atlas of the United States. Folio 2g, Nevada City, special folio, California, i8q6. This folio, by Waldemar Lindgren, consists of seven pages of text, three special topographic maps (scale i : 14,400), the Grass Valley, 410 ABSTRACTS Nevada City, and Banner Hill; three corresponding maps showing the economic geology, and three others giving structure sections. These maps, on a scale of about four inches to the mile, have been prepared to illustrate the detailed structure of the gold-mining regions in the vicinity of Nevada City and Grass Valley. Each of them com- prises an area three miles wide by four miles long, the total area being nearly thirty-six square miles. The Nevada City and Grass Valley areas fall within the boundaries of the Smartsville atlas sheet, while the larger part of the Banner Hill area falls within those of the Colfax atlas sheet. The relief is that common to the middle foothill region of the Sierra Nevada — that is, the surface is a very irregular and undulating plateau deeply trenched by the canyons of the recent river systems. Sedimentary rocks, chiefly referred to the Calaveras formation, occupy small, usually long, narrow areas imbedded in the predomi- nating igneous masses. Granodiorite occupies a large part of the Nevada City and Banner Hill districts, while a small massif of the same rock is found in the Grass Valley district. Large areas of dia- base, porphyrite, and brecciated forms of these rocks surround and separate the granodiorite areas. In the southwestern part of the Nevada City district and the northeastern part of the Grass Valley, a large and complicated viassif \s found, consisting in part of diorite, in part of gabbro, pyroxenite, and serpentine. The slates of the Calaveras formation are the oldest rocks. Next younger are the diorites, gabbros, and serpentines. Still later are the diabases and porphyrites ; and the intrusion of granodiorite closed the succession of igneous rocks. The bed-rock series is, as usual, in part covered by several hundred feet of Neocene gravels, and rhyolitic and andesitic tuffs, the gently sloping top of the andesitic ridges forming a principal feature of the landscape. The Neocene auriferous gravels have been extensively worked in the Nevada City and Banner Hill districts, both by the drifting and the hydraulic processes, and considerable ground still remains which probably can be profitably worked. The gold-quartz veins are numer- ous and belong to several distinct systems. They are found in all of the formations represented on the sheet, and generally cross the con- tacts without change. In the Banner Hill district the veins are narrow but rich, and have a general east-west direction and a northerly or southerly dip. In the Nevada City district the quartz veins have a general north-south direction and an easterly dip of about 45°. Large ABSTRACTS 411 dislocations producing overthrust faults have occurred along several of the veins. In the Grass Valley district there is one system with a west-northwesterly direction and a steep northerly or southerly dip. On this system the celebrated Idaho mine is located. Most of the veins in the central and southerlv part of the district have a northerly direction and a flat easterly or westerly dip. The veins are often accompanied by strongly developed sheeting of the country rock. United States Geologic Atlas, Folio 28, Piedmont, West Virginia — Maryland, i8q6. This folio consists of six pages of text, signed by N. H. Darton and Joseph A. Taff, geologists, and closing with a series of vertical sections showing the positions and thickness of the coal beds ; a topographic map ; a sheet showing the areal geology of the district ; another showing the economic geology ; a third exhibiting structure sections; and a fourth containing a columnar section and a key to the synonymy of the various formation names. The maps are on a scale of 1 : 125,000. The area represented is about 925 square miles. In Maryland it comprises the southern portion of Garret county and a small area in the southwestern corner of Alleghany county. In West Virginia it includes nearly all of Grant county, the western portions of Hardy and Mineral counties, the northeastern portion of Tucker county, and a narrow area of Preston county adjacent to the Maryland boundary line. Its southeastern corner is in a region of Appalachian ridges, and it extends northwestward over the Alleghany Mountains and the Upper Potomac coal basin to the headwaters of the Youghiogheny River, a branch of the Monongahela River. The geologic formations comprise members ranging from the sandstones in the middle of the Silurian to the Upper Coal Measures of the Carboniferous. In the southeastern portion of the area there are two sharp anticlinal uplifts which bring up the Silurian rocks in two prominent mountains. New Creek Mountain and Patterson Creek Mountain. To the westward lies the coal basin which extends from the Alleghany front to the Backbone Mountain. Along its center is cut the deep gorge of the north branch of the Potomac River. The basin is a relatively shallow one, but it contains about 3000 feet of Carboniferous deposits.. To the westward is the anticlinal region of 412 ABSTRACTS Devonian rocks which underlie the characteristic glade country about Oakland, Mountain Lake Park, and Deer Park. West of Oakland is another synclinal basin containing about 2500 feet of Carboniferous beds. The geologic classification does not differ materially from that outlined by W. B. Rogers and others, but geographic names have been applied to all of the formations. The lowest members are a series of sandstones and quartzites, which have been referred to as " No. IV " and "Medina." This series has been subdivided into the Juniata forma- tion, consisting of brownish red sandstones and shales ; the Tuscarora quartzite ; and the Cacapon sandstones, consisting of thin-bedded red sandstones. Next there is the representative Clinton formation, which has been designated the Rockwood formation, as in other folios ; the Lewiston. limestones, including representatives of the Helderberg and associated limestones, and the Monterey sandstones, Romney shales, Jennings formation and Hampshire formation, representing the Devonian deposits. As the last three formations are not sharply separated from each other, the patterns by which they are represented on the map are merged in a narrow zone along their boundaries. The Carboniferous period is represented by the Pocono sandstone; the Greenbrier limestone; the Canaan formation, which in a general way is a representative of the Mauch Chunk shales ; the Blackwater forma- tion, which represents the Pottsville conglomerate in greater or less part; the Savage formation and Bayard formation, which are the Lower Coal Measures ; the Fairfax formation, or Lower Barren Measures, and the Elk Garden formation, a part of the Upper Coal Measures. The principal coal beds are in the Savage formation, containing the "six-foot" or Davis coal bed; the Bayard formation, containing the coal bed known as the "four-foot" or "three-foot," or "Bayard" or "Thomas" coal; and the Elk Garden formation, containing the "fourteen-foot" coal bed. On the economic sheet of this folio, the coal-bearing formations are strongly emphasized, and underground contours are introduced to show the lay of the "six-foot" coal bed in the Savage formation for each 100 feet. Other economic resources of the area are red hematite iron ores in thin beds in Rockwood shales and limestones at several horizons, of which the lower member in the Lewiston is locally availa- ble for cement. ABSTRACTS 413 United States Geologic Atlas, Folio 2j, Nomi/ii, Maryland- Virginia, i8g6. This folio consists of four pages of text signed by N. H. Darton, geologist, a topographic map of the district, a map showing the areal geology, and a map showing the distribution of underground waters and artesian wells. The scale of these maps is 1:125,000. The area represented in this folio is about 938 square miles, which lies partly in Virginia and partly in Maryland. In Virginia it com- prises nearly all of Westmoreland county, with parts of Essex, Nor- thumberland, and Richmond, and in Maryland it includes portions of St. Mary, Charles, and Calvert counties. It lies entirely within the Coastal Plain area. The Potomac River extends northwest and south- east across the middle of the area, the Patuxent River crosses its northeastern corner, and the Rappahannock River crosses its south- western corner. To the extreme northeastward it extends to the shore of Chesapeake Bay. These waters are all tidal estuaries. Along the river valleys there are wide, low terraces capped by the Columbia formation, of Pleistocene age. The intervening areas are plateau remnants capped by Lafayette deposits, of supposed Pliocene age. The under- lying formations are the Chesapeake and Pamunkey, the latter extend- ing from the westward only a few miles into the area, along the north side of the Potomac River. The Pamunkey formation, of which only the uppermost beds are exposed, consist in greater part of glauconitic marls of Eocene age. It is overlain unconformably by the Chesapeake formation, which is characterized by fine sands, marls, and clays, portions of which consist largely of diatomaceous remains. The formation is very fossiliferous at some localities. Its age is Miocene. The greatest thickness which it presents in the Nomini area is about 270 feet, but it continues to thicken gradually to the eastward. The Lafayette formation, which ranges from 25 to 40 feet in thickness, con=.ists of sandy loams of orange, brown and buff tints often variegated, containing irregularly disposed bands and sprink- lings of small quartzite pebbles and coarse sands. The pebbles and larger sand grains are orange tinted, mainly by superficial staining. The plateau surface, capped by this formation and deeply incised and dissected by the larger drainage depressions, inclines gently southeast- ward at an altitude ranging from about 190 feet along the northern and western border of the area to about 90 feet along its eastern 414 ABSTRACTS border. Its greatest altitude is 200 feet in a portion of Nomini cliffs. It has also in most cases a slight slope into each of the river valleys. The Columbia formation is a deposit of loam merging downward into coarser materials containing beds of quartzite, gravel, and bowl- ders. Its thickness averages 20 feet. Its surface extends from alti- tudes of 5 to 60 feet above tide level. The principal economic features are underground waters, which on the lower lands furnish flows for artesian wells. Three water-bearing horizons are known, one at the base of the Pamunkey, another 100 feet higher in the same formation, and a third in the lower sandy members of the Chesapeake formation. They all dip to the eastward at a very moderate rate. There are many artesian wells which obtain water sup- plies from 160 to 305 feet. On the artesian well sheet of the folio distinctive underground contours are given to show the depths below tide level of all of the water-bearing horizons. Other economic resources of the area are marls in the Pamunkey and Chesapeake formations, diatomaceous deposits in the Chesapeake formation which are often sufficiently pure for commercial use, brick clays, potter's clays, sand and gravel. Geologic Atlas of the U?iited States. Folio 26, Pocahontas, Virginia- West Virginia, i8g6. This folio, by Marius R. Campbell, consists of five pages of text, a topographic sheet (scale i : 125,000), a sheet of areal geology, one of economic geology, another of structure sections, and, finally, a sheet giving a generalized columnar section of the district. The territory mapped and described embraces an area of 950 square miles, the southern portion of which is in Virginia and the northern portion in West Virginia. It is located west of New (Kanawha) River at the place where the state line leaves East River Mountain, the last of the valley ridges toward the northwest, and follows the irregular crests of the ridges within the coal field. The southern portion of this territory is within the limits of the Appalachian valley, and its surface is marked by linear mountains and narrow valleys, which are the characteristic forms of this central division of the Appalachian prov- ince. The northern portion is within the Cumberland plateau region, and its surface is that of a tableland deeply dissected, so that it now ABSTRACTS 415 presents a confused mass of irregular ridges and hills, only the sum- mits of which reach the original level of the plateau. The geologic structure of this region varies as the topography varies. In the northern portion the rocks are nearly horizontal, their northwestward slope being rarely more than 200 feet per mile, whereas in the southern portion the rocks have been highly compressed in a horizontal direction, forming huge folds, which in many places have broken, allowing one portion of the fold to slip over the other. It is this tilted condition of the strata which gives rise to the regular topo- graphic forms of the Appalachian valley. The attitude of the rocks is shown on the structure-section sheet by four sections which cross various portions of the territory. The geologic history of this region is recorded in the rocks, which tell of prevailingly marine conditions from early Cambrian to late Carboniferous time. There were deposited during that time sediments to the extent of 17,000 or 18,000 feet in thickness, which have since been hardened into limestone, shale, and sandstone. Of this great mas-s the limestones form about 6700 feet; the shales 9500 feet; and thfe sandstones, about 1400 feet. On lithologic grounds these have been divided into twenty -three separate and distinct formations, which are shown on the general geologic map by various colors and patterns. There is little variety in the mineral resources of this region. Coal, iron ore, and marble constitute about all of the mineral wealth of the territory. A limited area of coarse gray marble occurs along the northern front of Big Walker Mountain, but no development has been undertaken. Iron ore occurs in two formations of the Upper Silurian rocks. It is of good quality, and probably in sufficient quantity to be of com- mercial importance, but its inaccessibility has prevented develop- ment. Coal is by far the most important mineral resource of this region. The territory represented by this sheet embraces almost the entire Flat Top or Pocahontas coal field at present developed. All operations are confined to the great No. Ill or Pocahontas seam of coal, which is semi-bituminous and ranges in thickness from four to ten feet. It is exposed along the valley of Bluestone River from Pocahontas to the edge of the territory ; along Tug Fork ; in the valley of Elkhorn Creek from Coaldale to Kimball, near the edge of the area; and at several 4l6 ABSTRACTS places on the head streams of Guyandotte River. Mining is restricted to the Bluestone region and the valley of Elkhorn Creek. In these two areas there are at present in operation thirty-seven distinct mines, which in 1894 produced 3^096,867 long tons of coal. Geologic Atlas of the United States. Folio 25, Loudon, Tennessee, i8g6. The Loudon folio represents that portion of the Appalachian province which is situated beween the parallels 35° 30' and 36° and the meridians 84° and 84° 30'. This area contains 968 square miles, divided among Blount, Monroe, Loudon, Knox, Roane and Morgan counties, Tennessee. The folio consists of a topographic map, a geologic map, structure sections, stratigraphic sections, a map of the economic resources, and descriptive text. The author is Arthur Keith. The text begins with a general description of the Appalachian province, and points out the relations of this area to the general region, with regard to its surface features. The local features of the drai.nage by the Tennesse River and its tributaries, Emory, Clinch, Tellico, and Little Tennessee, follow next in description. The various forms of surface, such as the great valley of Tennessee and the portions of the mountain district and the Cumberland plateau by which it is bounded, are pointed out, and the relation between these forms and the under- lying rocks is made clear. Under the heading " Stratigraphy," the geologic history of the Appalachian province is presented in outline, and the local rock groups are fully described in regard to composition, thickness, loca- tion, varieties, and mode of deposition. The formations, thirty-three in number, range in age from Cambrian, to Carboniferous ; being, for the greater part, Cambrian and Silurian. The mountain district is chiefly underlain by the Ocoee series, whose age is doubtful. Rocks of Carboniferous and Devonian age occupy two small belts on either side of the great valley, and Silurian and Cambrian strata are repeated in narrow belts along it. Limestones, shales, and interbedded sand- stones make up the Silurian and Cambrian strata; sandstones and shales with coal seams and a limestone near the base constitute the Carboniferous ; and the Ocoee rocks are conglomerates, sandstone, slate, and limestone. The details of the strata are graphically represented in the colum- ABSTRACTS 4^7 nar section. The different manners in which the formations deca}' is discussed, and the dependence of the residual soils and surface forms on the nature of the underlying rock is brought out. Great lithologic changes occur in the formations of this region, and the Knox dolomite is the only one which is uniform throughout. The direction of change was exactly reversed between Cambrian and Silur- ian times. In the discussion of "Structure," after a general statement of the broader features of the province, two processes by which the strata of this quadrangle were deformed are noted. Of these the extreme Appalachian folding accompanied by faulting and metamorphism is by far the more prominent and is about equally developed throughout the quadrangle. Faults, especially, are most strikingly exhibited here. Deformation by vertical uplift is also exhibited, but it is only noticeable in comparison with broad surrounding areas. In this quadrangle the great valley is at its narrowest, on account of the extreme shortening in deformation. The structure sections illustrate the sharp folds and ireqilent faults into which the strata were forced. Economic products otthis region are coal, variegated marble, red hematite, building stone, lime, clays, slate and timber. The outcrops of the formations containing these are indicated on the economic sheet, together with the locations of the mines and quarries. The iron ore and slate are at present of minor importance. The coal district is a part of the great coal basin of Tennessee, and the marble belts are a part of the principal productive region for that stone. Various conditions affecting the value of these deposits are pointed out, and the associations and availability of the buildina: materials and timbers are discussed. Geologic Atlas of the United States. Folio 27, Morristow)i, Tennessee, 1896. The Morristown folio by Arthur Keith, deals with that portion of the Appalachian province which is situated between the parallels 36° and 36° 30' and the meridians 83° and 83° 30'. This area contains 963 square miles, divided among the counties of Green, Cocke, Jefferson, Hamblen, Grainger, Claiborne, Hancock, and Hawkins, all in Tennes- see. Included in the folio are topographic, economic, and geologic maps, structure and stratigraphic sections, and five pages of descriptive text. 4l8 ABSTRACTS After a description ot the broader features of the Appalachian province, the local geography is analyzed. The various types of surface features are pointed out, and their relations to the underly- ing rocks are shown. Local phenomena such as elevations and the details of the drainage which is effected by the Nolichucky, French, Broad, Holston, and Clinch rivers, tributaries of the Tennessee, are detailed. Under the heading "Stratigraphy" the geologic history of the Appalachians is presented in outline. This is followed by a detailed account of the local rock groups, in regard to their location, composi- tion, thickness, variations, and mode of deposition. The soils and forms of surface produced by each formation are discussed with the formations. Twenty formations ranging from Cambrian to Carbon- iferous, are distinguished in this quadrangle ; the greater portion being Cambrian and Silurian. The rocks of Carboniferous and Devonian age are found only in two narrow belts in the ridge district, and are represented by only four formations. Over the rest of the area Cam- brian and Silurian strata are about equally represented. A gjeat variety of limestones, shales, and sandstones compose the -Cambrian^ and Silurian rocks. Shales and sandstones make up the Devonian, while only the limestone appears in the Carboniferous. Great changes take place in the Silurian strata; limestones on the northwest being represented by shales and sandstones at the southeast. The general character of the formations is graphically represented in the columnar sections, one being drawn for each of the two chief geologic districts. In the discussion of structure, after a general statement of the broader features of Appalachian structure, the two types of deforma- tion shown in this region are described, and illustrations are pointed out in the structure sections. In the ridge district the most prominent feature is the faulting, which has cut the strata into long narrow blocks and produced the characteristic ridge topography. Southeast of Hol- ston River the rocks were deformed by close folds. Deformation by vertical uplift is also present, but it can only be observed in compari- son with other and larger areas. In the structure sections, most of the details of the different structures are shown. Economic products of this region are marble, building stone, lead, zinc, cement, clays, and timber. The outcrops of the formations containing these are represented on the economic sheet as far as pos- sible, together with the locations of mines and quarries. The principle RECENT P UBLICA TIONS 4 1 9 industries are the production of zinc and marble; the timbers and water powers are also of general importance. The various conditions which affect the development of these resources are discussed. RECENT PUBLICATIONS. — Baker, James. Annual Report, British Columbia. Minister of Mines for 1896. — 95 pp., I pi.. 4 maps. Victoria, 1897. — Baley, W. S. Summary of Progress in Petrography in 1896. — From monthly notes in the American Naturalist, 25 pp. Waterville, 1897. — Barbour, Erwin Hinckley. Nature, Structure, and Phylogeny of Dasmonelix. — Bull. Geol. Soc. Am., VHI, 305-314, 9 pi., 1897. — Barrios, Charles. Sur les Phenom^nes Littoraux Actuels du Morbi- han. — Ann. de la Societe geol. du Nord., XXIV, 182. Lille, 1896. — Barton, George H. Lieutenant Peary's Expedition. — Science, N. S., V, 308-310. New York, 1897. — Baur, G., and E. C. Case. Morphology of the Skull of Pleycosauria and the origin of Mammals. — Anatomischer Anzieger, Bd. XIII, Nr. 4 u. 5, 109-120. Jena, 1897. — Brown, H. Y. L. Reports on Artunga Gold Field and Hart's Range Mica Field; Contributions to the Palaeontology of South Australia by R. Etheridge, Jr. — Pam., 16 pp., maps, section, plate. Adelaide, 1897. — California Academy of Science, Proc, (2), Vol. VI, 587 pp., 75 pi. Ibid. (3), Geol., Vol. I, No. i. The Geology of Santa Catalina Island, by William Sidney Tangier Smith, 71 pp., map. San Francisco, 1897. — Call, R. Ellsworth. Some Notes on the Flora and Fauna of Mam- moth Cave, Kentucky. — Am. Naturalist, 377-392, i pi., 1897. — Calvin, Samuel. Pleistocene Iowa — Annals of Iowa, (3), Vol. Ill, No. I, 1-22. Des Moines, 1897. — Cohen, E. Uber ein neues Meteoreissen von Locust Grove, Henry County, Nord-Carolina, U. S. — Sitzber. K. Preus. Akad. Wis. zu Ber- lin, VI, 76-81, 1897. — Columbia University. Contributions from the geological department, XXXVT : Geological Notes, Long Island and Block Island, by Arthur Hollick. Ibid., XXXVII ; Stratigraphic relationships of the Browns Park beds of Utah, by J. D. Irving; Glacial and post-glacial Diver- sion of the Bronx River from its old channel, by J. F. Kemp; Notes on the Eclogite of the Bavarian Fichtelgebirge, by D. H. Newland. — Crosby, W. O. Contributions to the geology of Newport Neck and Connecticut Island, — Am. Jour. Sci., (4), Vol. Ill, 230-236. New Haven, 1897. Englacial drift. — Tech. Quart., IX, 116-144. Boston, 1896. — Crosby, W. O., and M. L. Fuller. Origin of Pegmatite. — Tech. Quart., IX, 326-356. Boston, 1896. —Crosby, F. W. and W. O. Sea Mills of Cephalonia.— Tech. Quart., IX, 6-23. Boston, 1896. 420 RECENT P UBLICA TIONS — Darton, Nelson H. Preliminary Report on Artesian Waters of Por- tion of the Dakotas. — Seventeenth Ann. Rept. U. S. Geol. Surv., Pt. II, pp. 1-92, pi. 69-107. Washington, 1897. — Davis, Wm. Morris. State map of Massachusetts as an aid to the study of geography. Reprint. — Sixtieth Ann. Rept. State Board of Education, 18 pp. Boston, 1897. State Map of New York as an aid to the study of geography.— ^ Univ. New York, Exam. Bull. No. 11, 503-526. Albany. 1896. The Seine, the Meuse and the Moselle. — Nat. Geog. Mag., VII, 189-238. Washington, 1896. — Bashford, Dean. New Species of Edestus, E. Lecontei from Nevada. — Trans. N. Y. Acad., XVI, 61-69, 1897. Note on the ventral armor- ing of Dinichthys. — Trans. N. Y. Acad. Sci., XVI, 57-61. 1897. — DiLLER, J. S. Crater Lake, Oregon. — Am. Jour. Sci., (4), Vol. Ill, 165- 172, I pi. 1897. — Derby, Orville A. Estudo sobre o Meteorite de Bendego. — Archivos do Museu Nacional do Rio de Janeiro, IX, 89-184, 1896. — Doss, Bruno. Etymologisches iiber die Kanger ; Ueber einige Beson- derheiten bei Diinen und Rigas weiterer Umbebung ; Zur Kentnisse der lebenden und Dubfossilen Mollusken fauna in Rigas Umgebung. Kor- respondenzblatt des Nat. Ver zu Riga, XXXIX, 25-128, 1897. Ueber einen Mammuthfund im Diluvium von Jaraslawl a. d. Wolga. Zeit. d. Deutsch geolog. Gesell., 940-953, 1896. — Dryer, Charles R. Studies in Indiana Geography. — Inland Educa- tor, IV, 2, 63-69. Terre Haute, 1897. — Ells, R. W. Paleeozoic Outliers in the Ottawa River Basin. — Trans. Roy. Soc. Canada, (2), Vol. II, sec. iv, 137-149, 1896. — Fairchild, H. L. Lake Warren shore lines in western New York and the Geneva beach ; Old tracks of Erian drainage in western New York (Abs.), by G. K. Gilbert.— Bull. Geol. Soc. Am., VIII, 269-289, i pL, 1897. — Gannett, Henry. Magnetic Declination in the United States. — Seven- teenth Ann. Rept., U. S. Geol. Surv., pp. 203-440, map. Washing- ton, 1896. — Geological Survey of Canada. — Summary Report for 1896, 144 pp. Report on the Geology of a portion of the Laurentian area, by Frank D. Adams, 184 pp., 2 pi., maps. Part J. Ann. Rept. Vol. VIII (1896). Section on mineral statististics and mines. — Ann. Rept., 1895, 103 pp., Part S, Ann. Rept., Vol. VIII. Ottawa, 1897. — Grimsley, G. p. Gypsum in Kansas. — Kansas Univ. Quart., VI, 15- 27, 4 pi. 1897. — Harris, Wm. T. How to Teach Natural Science in Public Schools, 46 pp. Barden, Syracuse, 1895, — Harvard University. Report of the Department of Geology and Geog- raphy for 1895-6. — Extract from Ann. Hept. Curator Mus. Comp. Z06I., 37 pp. Boston, 1896. — Henderson, J. B. Water Supply. — Report of Hydraulic Engineer of Queensland, 23 pp., 16 pi., 2 maps. Brisbane, 1896. — Hoffman, G. Christian. Report on the Section' of Chemistry and Mineralogy. — Geol. Surv. of Canada, Ann. Rept., VIII, 59 pp., 1897. %- ^ v^ ::l,,,l,,nlNllil!lili.il:!:B>, ^=^^ .^ ,„, r\' ^fiiiy \^^ \f' ^x. "''V \ 4'