TRANSACTIONS OF SECTION C. arrows. On allowing water streaked with colouring matter to flow over it from two areas, supposed to represent the great gathering grounds of snow of the Glacial period, the water had taken paths, as shown by coloured streaks, corresponding to those taken by the ice, as shown by the arrows. 2. Notes on the Cromer Excursion. By CLEMENT REID, F.G.S. 3 On the Tertiary Lacustrine Formations of North America. The early French explorers in the western parts of North America discovered certain large areas of extraordinarily broken and difficult country which they -a term which has been translated and called 'mauvaises terres à traverser shortened into the modern phrase 'bad lands.' Geological examination soon showed that these areas were covered with fresh-water deposits of Tertiary age, and that they represented a series of great lakes in which were entombed a vast number of representatives of the vertebrate faunas which successively occupied the country. I. (1) The most ancient of these formations is the Puerco, which overlies with apparent conformity the Cretaceous. This rather small lake occupies parts of Southern Colorado and North-western New Mexico. (2) The Wasatch succeeds after an interval and was much larger: it extends from New Mexico into Northern Wyoming. At least two contemporaneous lakes represent this horizon, and at the south its strata lie upon the Puerco. (3) The Bridger is indicated by a much smaller series of lakes, in Wyoming, Utah, and Colorado, which seem to have been successive rather than contemporaneous, and has been divided into three stages: (a) Wind River, (b) Bridger, (c) Washakie. The Wasatch beds pass under the Bridger at very many points. (4) Extensive orographic movements followed the Bridger stage and led to the formation of a new lake basin in Northern Utah, overlying the Bridger-the Uinta. Osborn has shown that two distinctly marked horizons occur within the limits of the Uinta. (5) The Uinta was the last of the great bodies of fresh water in the region to the west of the main chain of the Rocky Mountains. The movements now affected the Great Plains region, and an enormous lake was developed which extended along the eastern front of the mountains from Nebraska into North Dakota, together with a second basin in Canada. This is the White River formation, which is plainly subdivisible into three horizons--the Titanotherium, Oreodon, and Protoceras beds. (6) This was followed by the John Day, which is mostly confined to Oregon, but a second (7) A considerable hiatus very small basin occurs also in Central Montana. separates the John Day from the overlying Loup Fork, which is by far the most extensive of all the Tertiary lakes, and covers nearly all the Plains region, from South Dakota far into Mexico. It is not yet certain whether this was one vast body of water or a connected series of lakes. Independent basins occur in Montana, Nevada, and Oregon. In these the Loup Fork overlies the John Day unconformably, as farther east it overlies the White River. The Loup Fork falls naturally into three horizons, separated both by their faunas and more or less marked unconformities: (a) Deep River, only in Montana; (b) Nebraska, the principal area of the Plains region; (c) Palo Duro, known in Kansas, but principally developed in Texas. (8) In the Indian territory and Texas, in places overlying the Palo Duro, occurs the Blanco, the boundaries of which have not been traced as yet. (9) Occupying the surface of most of the Great Plains region are the uncompacted and obscurely stratified Equus beds, which rest unconformably upon all other formations of the region, from the Cretaceous to the Pliocene. II. The Lacustrine beds have for the most part retained their horizontal position, only rarely being tilted. They are composed of felspathic muds, clays, sands, and conglomerates, generally cemented by some calcareous compound, and their hardness is, speaking broadly, proportionate to their antiquity. The removal of the soluble cement by rain-water causes the rock to crumble, and the bad lands are examples of atmospheric erosion on a very grand scale. This erosion is extremely slow, the soil shedding the rain almost completely, but land-slips and snow avalanches in the spring frequently expose fresh surfaces of rock. The John Day beds are largely composed of the glassy particles of volcanic ash and the whole is overlaid by late basaltic flows. In the lake basins the old shore lines and deltas may frequently be traced, and their fossil contents sometimes afford interesting hints as to the habitat of the land animals. Climatic changes are registered in the alteration of the floras, as in the gradual disappearance of the palms from the central latitudes and in the diminution in the abundance and variety of the reptiles. For example, large crocodiles are exceedingly common in all the Eocene formations, including the Uinta, but in the succeeding White River only a few dwarf forms have been found. III. The correlation of formations in different continents is a difficult matter, but least so, perhaps, in the case of lacustrine beds in connected areas. The Old and New Worlds were certainly so connected during much, if not all, of Tertiary time, and there is always a certain proportion of land mammals common to the two continents. The natural and sharply marked lines of division are, however, not the same, and were the American formations arranged without reference to those of any other continent, a system very different from the European would result. Thus the Puerco would form one group; the Wasatch a second; the Bridger, White River, and John Day a third; the Loup Fork and Blanco a fourth; and the Equus beds a fifth. The European system is, however, the standard, and must be employed, and even in this way some very close correspondences may be noted. Thus, the Puerco is somewhat older than the Ceruaysian, while the Wasatch is the exact equivalent of the Suessonian. The Bridger is Middle Eocene (Parisian or Lutetian), and the Uinta in a general way corresponds to the Paris gypsum. The White River is Oligocene (Ronzor), and much misunderstanding has come from calling it Miocene. The John Day may be placed in the Lower Miocene, though it is somewhat older than the beds at St. Gérand-le-Puy, and follows the White River epoch with hardly a break. None of the American lacustrines is referable to the Middle Miocene. The Loup Fork is Upper Miocene, the Deep River division corresponding almost exactly to the beds of Sausan and Steinheim, while the Palo Duro division is perhaps already basal Pliocene. The fauna of the Blanco series is not yet sufficiently well known for exact correlation, though there can be no doubt that it is Pliocene. The Equus beds are distinctly Pleistocene, though it still remains to trace their relations to the Drift and to determine whether they are pre-Glacial or Glacial. 4. The Glacial Age in Tropical America. By R. BLAKE WHITE. The deposits described by the author cover almost the whole of the Republic of Colombia, extending from 12° N. lat. nearly to the Equator, and from the summits and plateaux of the Andes at 10,000 and 12,000 feet down to the plains, valleys, and littorals of the Atlantic and Pacific Oceans. The Glacial age corre sponded, in his opinion, with that of greatest volcanic activity. He speaks of moraines from 2,000 to 3,000 feet thick, accumulation of boulders, erratics of enormous size, and a peculiar loess on the high lands, the last containing bipyramidal crystals of quartz supposed to have been formed in situ. Great denudation followed the melting of the ice, and the auriferous deposits of the country belong to this or to the Glacial age. The author considers that a decrease in temperature enough to bring the snow line from 2,000 to 3,000 feet lower than it is at present would account for the phenomena which he describes. This might be brought about by a diminution of the amount of carbonic acid in the atmosphere, TRANSACTIONS OF SECTION C. producing rarefaction. He concludes by urging geologists to investigate a country which offers such a promising field, and does not present to the traveller any difficulties of importance. 5. On pre-Glacial Valleys in Northamptonshire. By BEEBY THOMPSON, F.C.S., F.G.S. The paper refers to the pretty general belief that the larger physiographical features of this country were developed before the Glacial period, and that many of the present river valleys are of pre-Glacial age; and remarks that, if so, they were more or less completely choked with boulder clay during the Glacial period. Where a valley got completely filled up it would, perhaps, in many cases be easier for the ordinary drainage to cut out a new valley than to remove the infilling of an old one, where the initiative for a new valley had been given by superficial streams due to melting ice. So at the close of the Glacial period, although the main drainage of a district must take approximately the same direction that it did before it, the tributary streams would seldom accurately follow their old lines on having, as it were, a fresh start under somewhat different circumstances. Compromises would, no doubt, frequently result. Bearing these matters in mind, we should be prepared to find- 1. New valleys without drift, and filled-up old ones near at hand. 2. Valleys with one side drift and the other the normal rocks of the district. 3. Valleys still containing much drift, with the streams running over or through it. 4. Valleys in which only the coarser material of the drift is left in the form of river gravel. Illustrations of each of the four cases enumerated are given, all from Northamp tonshire. The author suggests that his explanation of some isolated patches of boulder clay near to Northampton may prove to be of more general application than previously suspected. 6. Notes on some Tarns near Snowdon. By W. W. WATTS, M.A., F.G.S. During a recent visit to Snowdon, the writer has taken the opportunity of examining a few of the tarns in its immediate vicinity. These include the two small lakes in Cwm Glas, Glaslyn, and Llyn Llydaw. In the hollow of Cwm Glas there are two tiny tarns named Ffynnon Frech and Ffynnon Felen; both lakes drain over a barrier of rock, but in a rainy season the upper one appears to find a second outlet over the long, low col to the East, so that, in this state, it has the two outlets depicted in the 6-inch map. There can be little doubt that this upper lake is a portion of a bending valley dammed at both ends by scree- and stream-débris, and thus compelled to find an escape over the rocky side. The lower lake is certainly confined in a rock basin, as rock occurs at its actual outlet and at every point where any former outlet might have been possible. The lake is, however, so shallow that its occurrence in a basin of rock is perhaps of little consequence. The neighbouring hollow of Cwm Dyli, as is well known, contains three lakes, the highest being Glaslyn, the next Llyn Llydaw, and the lowest Llyn Teyrn. Glaslyn is bounded on all sides by live rock except at and near its outlet. This exit is over moraine, which, however, is evidently not very deep, for rock makes its appearance just below, and in such a way as to almost compel belief in a complete rock bar. Beside the present course of the effluent stream is a parallel strip of moraine running down towards Llyn Llydaw, but living rock soon makes its appearance in this in such a way as to show that if there is any old channel in this direction it must be exceedingly narrow and tortuous. Thus, if this lake is not contained in a true rock basin, it must be very shallow or else must have found exit by a gorge quite as narrow as those found at the end of some of the Swiss glaciers. Immense quantities of moraine material occur on the south-east side of Llyn Llydaw, but a careful examination of the map and the ground shows that only two possible outlets exist-that now used for this purpose, and a second which is occupied by bog resting on moraine, and gives rise to a small stream which is joined lower down by the outlet of Llyn Teyrn. The moraine is, however, only a thin skin on the surface of rock. The present outlet shows live rock forty or fifty feet below the level of the lake, and the second possible exit at a rather less distance below the same level. If the moraine were stripped off, there is little doubt that this lake, like Glaslyn, would show a basin of rock which would hold water, unless it is very much shallower than is generally supposed to be the case. 7. Interim Report on the High-level Shell-bearing Deposits of Clava, &c. 8. Interim Report on the Calf Hole Care Exploration. 9. Report on the High-level Flint Drift of the Chalk. 10. Report on the Rate of Erosion of Sea Coasts. 11. Final Report on Underground Waters.-See Reports, p. 393. 12. On Modern Glacial Striæ. By PERCY F. KENDALL, F.G.S., and J. LOMAS, A.R.C.Sc. The authors have spent several weeks during the present summer among the glaciers of the Nicolaithal and the Val Tournanche, and have paid especial attention to the form and distribution of glacial striæ. The present communication deals with four principal sets of phenomena. 1. The Crossing of Two or more Sets of Striæ.-In the discussion of the glacial geology of Britain and other countries writers have ascribed the formation of two superposed sets of stria on one surface either to the action of floating ice or to a different period of glaciation. The authors have found that the phenomenon is of quite general occurrence, especially on the steeply inclined weather'-sides of roches moutonnées. They have observed an angular divergence of 89°. 2. The Forms of Strice as a Means of determining the Direction of Ic movement. It is often impossible to decide à priori whether a particular scratch or set of scratches was produced by ice moving, say, from south to north, or from north to south. The late Professor Carvill Lewis thought that striae having s broad and a narrow end would furnish reliable criteria, but the authors, afer careful examination of a large number, are unable to regard such characters as possessing the required degree of constancy. 3. The Phenomena known to American Glacialists as Semilunar Markings,' 'Pluck-marks,' and Chattered Stria.'-The authors found many examples of these features of glacial abrasion upon roches moutonnées. The pluck-marks' were found to be shallowest at their downstream' ends. Chattered striæ,' i.e., ragged striæ presenting somewhat the appearance of a succession of bruises, were very common. They were probably produced by boulders that were only partially embedded in the ice, and were thus dragged along with a jerking motion. It is satisfactory that these minor details of the glacial phenomena of the United States can be paralleled in the Alps. 4. The Occurrence of 'Screwed' or Curved Striæ.-Authors have ascribed the formation of sharply curved or screwed striæ to the swinging of floating ice when partly aground. The present writers have observed and recorded by heel-ball rubbings' and by photography many examples occurring on the roches moutonnées of the Gorner Glacier. 13. Notes on the Ancient Physiography of South Essex. The author refers to his paper, read before the Geological Society last year, entitled, Further Notes on some Sections on the New Railway from Romford to Upminster, and on the Relations of the Thames Valley Beds to the Boulder Clay.' İn that paper he mentioned the discovery, in a railway cutting at Romford, of part of an ancient silted-up stream course of considerable size, covered by gravel belonging to the highest, and presumably oldest, terrace of the Thames Valley system. In this communication he considers more fully the evidence bearing upon his view that the course taken by this ancient stream was between the high ground of Warley, Billericay, and Maldon, and that of Laindon, Rayleigh, and Althorne into the Blackwater, the basins of the Mardyke and Crouch having originated at a much later period. He also notes evidence tending to confirm Mr. Whitaker's view that the gravel and loam at and near Canewdon, Southminster, and Bradwell were deposited on the western flank of the old Thames Valley when there was a considerable breadth of land east of those places, which has been since removed by marine denudation. SATURDAY, SEPTEMBER 14. The following Papers and Report were read: 1. Restorations of some European Dinosaurs, with Suggestions as to their Place among the Reptilia. By Professor O. C. MARSH. For several years I have been engaged in investigating the Dinosaurs of North America, where these extinct reptiles were very abundant during the whole of Mesozoic time. The results of my study have been published from time to time, and I have already had the honour of presenting some of these to the British Association. In carrying out this investigation so as to include the whole group of Dinosaurs, wherever found, and bringing all under one system of classification, it has been necessary for me to study the remains discovered in Europe, and Í have made several visits to this country for that purpose. In comparing the forms known from the two continents, certain important differences as well as some marked resemblances between the two have been observed and placed on record. In concluding my investigations of the North |