of all the cols, the terraces and water inscriptions upon the valley sides north of the cols, and particularly the many large deltas formed by the lateral streams debouching into the extinct lakes at the high levels.

Eighteen of these extinct lakes are recognized from the topography, the Attica lake (the flooded Tonawanda valley) being the most western, and the Tully valley lake (the flooded Onondaga valley) the most eastern. Ďata have been gathered by personal observation and measurements of several of the more important lakes. The Ithaca lake was the largest and deepest, being at its maximum over 1100 feet deep, five to ten miles wide and about thirty miles long, overflowing by the Six Mile Creek and Catatonk valleys to the Susquehanna at Owego. The Watkins lake with its outlet by Horseheads to the Chemung at Elmira was in dimensions but little under the Ithaca lake. A fine delta at Watkins gives the maximum level of the water and several lower levels. The Dansville, the Conesus, the Naples, and the Hammondsport lakes have been carefully studied. Some of these have a complex but interesting history.

With the removal of the ice barrier by the melting and recession of the ice, the local glacial lakes were lowered until they were merged into the great body of glacial water which buried all western New York, north of the dividing ridge, to a depth of several hundred feet. At first this was the "Warren water, having its outlet at the low divide south of the present Lake Michigan, near the site of Chicago. The beaches of this water have been traced by Mr. Gilbert, Mr. Spencer and Mr. Leverett to a point east of Buffalo, and at an elevation of 860 feet or more, with a differential rise toward the northeast. This indicates a depression of western-central New York following the ice-retreat. If Messrs. Upham and Spencer are correct in their estimates of the depression, then the Horseheads outlet of the glacial Watkins lake was certainly below the Chicago outlet of the Warren waters, and the uncovering of the Seneca Valley gave a new and lower outlet. There could have been no other escape for the waters unless there was a low uncovered pass into Hudson's Bay, which is not believed. In memory of a former distinguished worker in the field of glacial geology this lower lake is called Lake Newberry.

The further withdrawal of the ice sheet finally uncovered the valley of the Mohawk, and opened a still lower outlet, producing the episode of Lake Iroquois.

R. D. SALISBURY read a paper on ແ The Surface Formations of Southern New Jersey." The following abstract was prepared by him for this Journal. The series of deposits considered in this paper have heretofore been grouped under the name of "Yellow Gravel." It is found that the series of sands and gravels which have been referred to by this general name are divisible into several distinct formations. Of these formations there are certainly three, probably four, and possibly five. The oldest of these sev

eral formations finds its northernmost extension in the Sand Hills north of Monmouth Junction. It caps various high hills in Monmouth County, where it reaches its greatest elevation. Southward the formation becomes lower and at the same time much more continuous. In general it is essentially continuous south of the line where the Middle Marl bed comes to the surface. This formation has essentially the same dip as the Cretaceous beds beneath. That it is not strictly conformable with the Cretaceous is shown by the fact that it overlies different members of the Cretaceous series in different areas. The relations are such as to show that the main part of the deformation of the Cretaceous beds took place after this oldest "Yellow Gravel" formation was laid down. Professor W. B. Clark has reached the conclusion that this formation is Miocene. After it was deposited, there was a period of uplift accompanied by deformation. This was followed by a long period of erosion, during which a large part of the formation, especially to the north, was destroyed by subaërial agencies. During this time of erosion central Jersey was reduced to a peneplain. Subsequently the central and southern parts of Jersey were again submerged, and there was deposited on the peneplain, the second member (the Pensauken formation) of the 66 Yellow Gravel" series. Above the Pensauken sea stood a considerable number of hills capped by Miocene. The Pensauken is best developed along the trough-like depression which extends from the vicinity of Trenton northeastward to Raritan bay. The Pensauken formation is made up of constituents which came from very diverse sources. The Triassic shale, and most, if not all the formations of the highlands of northern New Jersey made their contributions, as well as the Cretaceous and Miocene. All decomposable constituents of the formation are completely decayed. This formation originally reached at least as far north as the Watchung Mountains. During the uplift which followed the deposition, there was slight deformation, the elevation being greatest in the region which suffered greatest uplift after the deposition of the Miocene beds. Erosion followed, and a well developed system of drainage was established on the Pensauken surface. Subsequently subsidence and a considerable degree of submergence followed, so that much of the Pensauken formation (its lower parts) and the valleys which were cut in it, received a new mantle of loam and gravel. This is known as the Jamesburg formation. Subsequent to the deposition of the Jamesburg, the Trenton gravels were deposited in the Delaware valley, during the last glacial epoch which affected New Jersey. About the eastern and southern coasts of the state there is a broad, ill defined terrace, having a general altitude of about forty-five feet, which may or may not correspond in age with the Trenton gravels. This broad terrace is certainly much younger than the body of the Jamesburg formation, though it may represent no more than a halt in the emergence which followed the deposition of the body of that formation. The correlation of these various subdivisions

with formations which have been heretofore recognized is not now attempted. There is some reason to believe that the Pensauken is considerably older than the oldest glacial drift. The two formations come together in but one area, and there the relations are not unequivocal. The general constitution of the Pensauken formation would ally it quite as closely with the Lafayette as with the Columbia of the regions farther south, but this is not regarded as a sufficient basis for so correlating it. It may be Pleistocene, antedating the earliest drift by a short interval only. The Jamesburg formation contains occasional glaciated bowlders. It is believed to be later in origin than the earliest glacial drift, and to be earlier than the latest glacial drift. Little erosion has taken place since it was laid down. Whether this is the result of a shortness of time, or of insufficient elevation, is not yet certain.

C. D. WALCOTT, in a paper on The Appalachian type of folding in the White Mountain Range, of Inyo County, California, gave an account of the discovery of the overturned synclinal-anticlinal structure in the White Mountain range, situated east of the Sierras, similar to that existing in the Appalachian Mountains in occupying a similar position to those on the opposite side of the continent. In a second paper (Lower Cambrian Rocks in Eastern California), he gave an account of the discovery of Lower Cambrian rocks and fauna in the same range. These two papers will appear in full elsewhere in this Journal. ARTHUR KEITH gave an interesting analysis of the structure of the Southern Appalachians in a paper on New structural features in the Appalachians. His paper reviewed the generalizations of long standing in regard to Appalachian structure, stated and analyzed the late generalizations, such as effect of bedding planes, of superincumbent lead, of initial basins of deposit, and of transmission of thrust, upon structure. Finally, statement was made of newly discovered structures, such as fan structure, cross folds, cross shear zones, secondary system of folding and the distribution of metamorphism; and a theory was advanced to account for these features that the crystalline rocks moved against the sediments, chiefly along the shearzones, and deformed them most in the southwestern Appalachians. F. B. ADAMS presented an excellent paper entitled A further contribution to our knowledge of the Laurentian, which was fully illustrated by lantern views with the novelty of projecting the enlarged rock sections directly upon the screen. Prof. Adams showed that essentially two types of rock structure are to be found in the metamorphosed Laurentian rocks, one a purely clastic one, the other one in which recrystallization has taken place and the importance of these in discriminating the origin of these rocks was pointed out.

G. K. GILBERT, in a paper on the formation of lake basins by wind, The Tepee Buttes, by G. K. GILBERT and F. P. GULLiver, described peculiar geological formations examined by him in Colorado. Shallow lakelets reaching a mile in greatest diameter, were observed in the arid regions in eastern Colorado along the

Arkansas River valley, often near the highest points in the surface topography, and were interpreted as hollows scooped out by winds blowing in one direction persistently, during dry weather and filled by rains, becoming lakes and finally drying up to mere hollows in the surface. The name "Tepee Buttes" is given to conical hills left after surface degradation of the approximately horizontal Cretaceous strata, composed of central cylindrical cones of limestone chiefly made up of fossil shells in the midst of the soft, thinly bedded shales. N. S. SHALER read a paper On certain features in the jointing and veining of the Lower Silurian limestone near Cumberland Gap, Tenn.,-gash veins, in dense dolomitic limestone, 1mm. in diameter and an inch to a few inches in length, filled by calcite, but the rock strata not faulted, folded or displaced. H. S. WILLIAMS, read a paper on Devonian fossils in Carboniferous strata which appears in full in the present number of this Journal. The large number of papers on Petrography and allied subjects determined the formation of a petrographical section which held a separate meeting on Friday.

In addition to the papers above referred to the following were presented, some of them, in the absence of the authors, being read by title only.

H. P. CUSHING-The faults of Chazy Township, Clinton Co., N. Y.

W. J. MCGEE-Remarks on the geology of Arizona and Sonora.

WALTER H. WEED and LOUIS V. PIRSSON-Geology of the Highwood Mountains, Montana.

CHARLES R. KEYES-Genesis and structure of the Ozark uplift.

J. W. SPENCER-The geographical evolution of Cuba.

T. C. CHAMBERLIN-Notes on the glaciation of Newfoundland.

C. W. HALL-The Pre-Cambrian floor in the Northwestern states.

J. F. KEMP-The crystalline limestones, ophiolites and associated schists of the eastern Adirondacks.

DAVID WHITE-The Pottsville series along New River, West Virginia. WILLIAM B. CLARK-The Cretaceous deposits of the Northern half of the Atlantic coastal plain. The marginal development of the Miocene in eastern New Jersey.

G. K. GILBERT-Stratigraphic measurement of Cretaceous time.

E. T. DUMBLE-Notes on the Cretaceous of Western Texas and Coahuila, Mexico.

N. H. DARTON-Sedimentary geology of the Baltimore region.

R. P. WHITFIELD-On new forms of marine algæ from the Trenton limestone, with observations on Buthograptus laxus, Hall.

W. S. BAYLEY-Spherulitic volcanics at North Haven, Maine. The peripheral phases of the great gabbro mass of northeastern Minnesota. The contact phenomena at Pigeon Point, Minnesota. (Simply exhibition of specimens.)

ALFRED C. LANE-The relation of grain to distance from margin in certain rocks. Crystallized slags from copper-smelting.

ROBERT BELL-On the honeycombed limestones in the bottom of Lake Huron. LEON S. GRISWOLD-On the nomenclature of the fine-grained siliceous rocks. ALFRED E. BARLOW-On some dykes containing Huronite."

WALDEMAR LINDGREN-The characteristic feature of the California gold quartz

veins.

SAMUEL WEIDMAN-On the quartz-keratophyre and its associated rocks of the Baraboo Bluffs, Wisconsin.

EDWARD B. MATHEWS-The granites of Pike's Peak, Colorado.

N. H. DARTON and J. F. KEMP-A new intrusive rock near Syracuse, New York.

G. P. MERRILL-On the decomposition of the granite rocks of the District of Columbia.

B. WILLIS-Ancient physiography as represented by sediments.

B. K. EMERSON-Serpentine pseudormorphs after olivine, formerly called quartzpseudomorphs, Middlefield, Mass. Skeleton crystals of salt which have been called chiastolite and later spinel, from the Trias, Westfield, Mass. Radiating puckering of corundum crystals around allanite, Pelham, Mass.

C. H. SMYTH, JR.-The crystalline limestones and associated rocks of the Northwest Adirondack region.

H. S. W.

3. Manual of Geology. Treating of the Principles of the science with special reference to American Geological History; by JAMES D. DANA. 4th edit., pp. 1088, with 1575 figures and two double-page maps. New York, 1895. (American Book Company).-Dana's Manual, in this new edition, is a wholly rewritten work, a total revision having become necessary in consequence of the great progress in the science during the last twenty years, and especially in the geology of the Continent of North America. In the Preface the author remarks that North American Geology is still its chief subject; and adds:

"The time-divisions in this history, based on the ascertained subdivisions of the formations, were first brought out in my Address before the meeting of the American Association at Providence, in 1855;* and in 1863, the 'continuous history' appeared in the first edition of this Manual, written up from the State Reports and other geological publications. The idea, long before recognized, that all observations on the rocks, however local, bore directly on the stages in the growth of the Continent derives universal importance from the recognition of North America as the world's type continent-the only continent that gives, in a full and simple way, the fundamental principles of continental development.'

The general arrangement of the work remains the same as in the earlier editions, except that Dynamical Geology is made to precede Historical. Moreover the Dynamical part commences with a chapter on "Chemical Work"; and the subject of Mountain-making and Earth-shaping, with which it closes, is treated under the general heading of "Hypogeic Work," the results, whatever the agency concerned, whether external or internal, being consequent on subterranean movements.

In the preparation of the Historical part of the Manual, the author, besides availing himself of all recent geological reports and other related publications, carried on an extensive correspondence with the chief geologists of the country; and further, in the paleontological parts, each of the chief divisions of the subject was prepared with the assistance of specialists familiar with the latest results of investigation.

The following are the principal changes in connection with the subdivisions of geological time.

Paleozoic time is divided at. the close of the Lower Silurian, into two sections:

* Republished. with added notes, in this Journal, II, xxii, 305, 1856. AM. JOUR. SCI.-THIRD SERIES, VOL. XLIX, No. 290.-FEB., 1895.