The interval, No. 2, is filled with shale in the first two sections, but contains sandstone as well as shale in the last two. No. 4 is bony coal in the first two, but holds shale and sandstone in the last two. No. 6 is filled with clay at all of the localities.

The coal from the Laramie group in this field is soft. That from the Trinidad bed is excellent gas-coal and the slack is easily coked. The coking is done in bee-hive ovens similar to those used in western Pennsylvania. Little has been done, away from Trinidad, toward developing the mines. The whole country is cursed with old Mexican land grants, most of which are clouded, so that capitalists hesitate to make investments.

The Sandstones.

The sandstones are much alike and few of them are persistent. They change into shales and back again into sandstones in the most perverse manner. Still, some of these beds are constant and are serviceable locally as guides.

The Great Sandstone, closing the group within this field, is present at the summits of all divides and is readily recognized by its physical peculiarities. It is yellowish gray, compact, and for the most part comparatively fine-grained, though it occasionally contains a layer of not very coarse conglomerate. This rock is wholly non-fossiliferous at every locality where it was examined.

The sandstones above coal beds F, G and H, are usually present. These vary from light yellowish gray to decided buff. Ordinarily they are massive, but occasionally flaggy layers are found in which impressions of dicotyledonous leaves abound. No animal remains were observed except in the sandstone overlying coal bed F, in which obscure impressions of a Cardium were observed at one locality.

The Halymenites sandstone, at the base of the series, is comparatively fine-grained, usually gray, sometimes yellowish gray. It is invariably present and forms a distinct gray band on the bluffs from Cucharas Creek southward to Cimarron Creek, with the blossom of the Dillon coal bed almost immediately above it. I have given its name because the rock is loaded with Halymenites major Lesqx., which was not identified with certainty at any higher horizon within this field; though it is abundant at higher horizons in other fields farther north.

Almost without exception, the sandstones are fine-grained at

the eastern edge of the field, but they become coarser toward the west, until on the western border, some of them are conglomerates and the shales have almost wholly disappeared. This condition exists at the base of the mountains.

Limestone layers were seen within several of the sandstones. They are always present on the eastern side of the field in the intervals between coal beds F and G, H and I, and I and J. Similar layers are sometimes shown in other intervals, but they are not persistent. These beds are from two to eighteen inches thick. None was found above coal bed J, which is about midway in the section. The limestone is blue to flesh-colored, weathers yellow because of much iron, and contains no fossils. It is very similar to much of the limestone found in the Lower Barren coal group of the Appalachian coal field.

Impressions of leaves of dicotyledonous plants were found. in all the flaggy sandstones from the Dillon coal bed to the base of the Great Sandstone at the top of the section; but animal remains are rare. Some fish-teeth were obtained from the Halymenites Sandstone, associated with a Cardium, which is very similar to a shell of which imperfect impressions occur in the sandstone overlying the Lower Vermejo coal bed. But no impressions of leaves were found in the shales immediately overlying coal beds. The impressions occur only in sandstones; they are of isolated or fragmentary leaves and many of them were much softened by soaking before they were entombed. The plants belonged to upland vegetation and their leaves were evidently brought down to the shore by streams.

Relation of the Laramie to the Middle Cretaceous,

Throughout this southern coal-field the Lamarie rocks rest on the shales of the Fort Pierre sub-group, the Cretaceous, No. 4 of Mr. Meek's original section. These contain Ammonites placenta, Baculites ovatus, Inoceramus convexus, and other thoroughly characteristic species. The Fox Hills group, the No. 5 of Mr. Meek's section, appears to be wanting here. tainly is wanting if the Halymeniles sandstone is to be included in the Laramie group.

It cer

Lithologically, the transition from the Fort Pierre to the Laramie is so gradual that the line of separation between the groups must be assumed arbitrarily. The dark shales of the former pass upward into brownish shales with thin sandstones, which in turn shade away into the Halymenites sandstone above. The transition requires not far from two hundred feet of rock. But a great change took place at the close of the Fort Pierre group. For the most part, the shales of that group are rich in animal remains, but such remains cease to appear at forty or fifty feet below the line assumed as the summit. Animal remains are, to all intents, absent from the Laramie group in the Trinidad

coal-field. The newer conditions were unfavorable to animal life. Marine conditions, however, did not cease with the Fort Pierre. The sandstones of the Laramie are of marine origin, though perhaps only off-shore deposits. Halymenites major occurs profusely in the lowest sandstone within the Trinidad field as well as at much higher horizons in the Cañon City field. Huge knotted fucoids were found in a sandstone above coal bed J, in the former field. Other sandstones, showing no fucoids, contain many battered logs. Limestones, unmistakably of marine origin, occur up near to the middle of the Laramie. The change is not unlike that shown in the passage from the Lower to the Upper Carboniferous in the Appalachian coal-field.

ART. XXII.-On some points in Lithology; by JAMES D. DANA. II. ON THE COMPOSITION OF THE CAPILLARY VOLCANIC GLASS OF KILAUEA, HAWAII, CALLED PÉLÉ'S HAIR.

THE capillary volcanic glass of Kilauea collected by the writer at the volcano in the year 1840 was analyzed for the writer's Geological Report of the Exploring Expedition (1849) by Prof. B. Silliman (B. Silliman, Jr.), and the results are published in it on page 200. The large discrepancies between the two analyses there reported-one of a dark and the other of a pale variety—and especially the difference as to soda, one being stated to contain 21 62 per cent, and the other none, left the question of composition in great doubt.* I have now to report two new satisfactory analyses of the glass. For these, science is indebted to Mr. F. J. Allen of the Sheffield Scientific School of Yale College, excepting the determination of the state of oxidation of the iron, which is by Prof. O. D. Allen. The results were as follows:

*The analyses also of volcanic scoria and lava, given on the same page of my Report, are evidently too uncertain to be longer quoted, unless the results shall be confirmed by other analysts.

The composition obtained has great interest, since it shows that this most fusible part of the Kilauea lavas has almost precisely the composition of ordinary dolery te (=basalt=diabase, essentially). I cite for comparison an analysis by G. W. Hawes of the "trap" of West Rock (New Haven, Conn.), which agrees closely with the average composition of this basic rock.

Pélé's Hair

West Rock "trap"

Sio, Al0, Feo, FeO MnO MgO CaO Na,O K20 ign

50-75 16.54 2.10 7.88 tr. 7.65 11.96 2.13 0.56 0.35 99.92 51.80 14.21 3.55 8.26 0.42 7.63 10.68 2.15 0.39 0.63=99.72 [+phosphoric acid 0.14 The "trap" consists of labradorite and augite with some magnetite. It is hence identical with the most abundant kind of igneous rocks. The fusibility of such a compound is thus well indicated by the facts at Kilauea. Moreover it is not surprising, since the fusibility of both labradorite and ordinary black augite are each marked down as low as 3 by von Kobell. There is hence no question as to the complete fusion of such ingredients in a volcano, even where moisture is not present.

The analyses add another to the many examples already known, proving that there was no difference in constitution between a large part of the material in fusion and ejected in Mesozoic time and that thrown out by modern volcanos; and it illustrates the fact that Geology has no good basis for the distinction of "older" and "younger" among igneous rocks.

An important paper on the microscopic characters of Pélé's Hair has been published at Tubingen (in 1877) by C. Fr. W. Krukenberg, in a pamphlet giving also the results of the author's investigations on Tachylyte and Hyalomelan, Basalt glass, Porous and Spherulitic Basalt, and Obsidian. He states, and illustrates by figures, the following facts respecting Pélé's Hair. The fibers are sometimes bent and coalesced into loops; often are tubular; frequently contain air bubbles, and occasionally microlites. There is usually an enlargement of the diameter whenever a crystal (or microlite) exists within, and also about many of the air-cavities. The crystals are mostly rhombic, but as to their kinds the author makes no suggestion.

ART. XXIII.-On the size of Molecules; by N. D. C. HODGES.

If we consider unit mass of water, the expenditure on it of an amount of energy equivalent to 6367 units of heat will convert it from water at zero into steam at 100°. I am going to consider this conversion into steam as a breaking up of the water into a large number of small parts, the total surface of which will be larger than that of the water originally. To increase the surface of a mass of water by one square centimeter

requires the use of 000825 milligrams of work. The total superficial area of all the parts, supposing them spherical, will be 4πr N. The number of parts being N, the work done in dividing the water will be 4r2 N. For the volume of all the parts we have r3N. This volume is in accordance with the requirements of the kinetic theory of gases, about of the total volume of the steam. The volume of the steam is 1752 times the original unit volume of water.

Hence Nπr33000=1752
4N000825-636*7,423

One unit of heat equals 423 milligrams.

3000

Solving these equations for r and N, we get r equal to 000000005 centimeter, a quantity of the same order of magnitude as has already been obtained by Thomson, Maxwell and others, N equal 9000 (million)3n for the number in one cubic centimeter 5 to 6 (million)".

Around every body there is an atmosphere of more or less condensed gases. On the surface of platinum these must be nearly in the liquid condition, as shown by the power of platinum to bring the atoms of hydrogen and oxygen so near together that they combine. These vapors on the surface have a tendency at ordinary temperatures to expand; and part of them can do so, if the surface of the body is reduced. There is in these condensed atmospheres an explanation of all the phenomena of superficial tension. The energy in the unit of area ought to be equivalent to the amount of work done in compressing a quantity of the vapor from the gaseous to the liquid state sufficient to cover the surface a few molecules deep. The molecular attraction seems to be very slight in gases, when the molecules are ten or fifteen molecular diameters apart. To get some idea of the amount of work done in compressing one gram of oxygen to liquid form, we may consider that in the union of one gram of hydrogen with eight grams of oxygen 34,462 units of heat are produced. It matters not that the condensation is brought about by the energy of chemical separation rather than by the work done in pressing them together in a cylinder.

The superficial energy of platinum is 1694 milligrams per square meter or 01694 per square centimeter, equal to 00004 of a unit of heat. The proposition

9:34,462=x: '00004

gives the weight of water condensed on square centimeter of surface or the volume in cubic centimeters as '00000001, which agrees with the other result.

Physical Laboratory, Harvard College, May 14, 1879.