stances, and their persistence over so great an area, where physical characters have in a great degree failed, only serve to demonstrate the value of these means of studying and identifying the stratified deposits.

I have, also, had an opportunity of tracing this group, at intervals, across the country between Lake Michigan and the Mississippi river, and of recognizing it by its fossils, particularly its corals, even beyond that point, to the southwest. It exists in Ohio, Indiana, Kentucky, and Tennessee, and may be recognized, not only by its lithological characters, but by its numerous fossils, identical with those first described in the New York Geological Reports, as occurring in this group. Among these, we have, in addition to the corals, which are the more common fossils, several species of crinoids, the Caryocrinus, Eucalyptocrinus, Ichthyocrinus, and others which are identical with species known to occur in New York.

In a southwesterly direction, and without the limits of this district, this group has been examined by Mr. I. A. Lapham, of Milwaukee, who has had no difficulty in recognizing it by its characteristic fossils. His observations will be found incorporated in the subsequent pages of this report.

Thickness of the Niagara and Clinton Groups.--As already remarked in the commencement of this chapter, the lowest beds are seen at the lake-level on the eastern side of Drummond's island; while on the northern side, we find the higher beds of the Hudson-river group. Again, on the St. Mary's river, the lowest beds of the Clinton group come to the water-level just above Lime island. Passing westward, along the northern shore of Lake Michigan, although numerous undulations are visible, often enabling the observer to see a considerable thickness of strata, yet the lower beds are nowhere exposed, until we arrive at the bluffs on Big Bay des Noquets. I have already mentioned the exposure of the same beds here which we had seen on Drummond's island. The entire height of the cliffs does not exceed two hundred and fifty feet, and we have nowhere evidence of the existence of superior beds of more than one hundred feet in thickness belonging to these groups. The entire thickness, therefore, of the calcareous Beds of the Niagara and Clinton groups does not exceed three hundred and fifty feet. This is, I am aware, somewhat less than the estimate of Mr. Murray, in his section across the Grand Manitoulin islands, which, including both groups, amounts to five hundred and sixty feet.* Although there are, at intervals, exposures which appear along the coast, where the rate of dip is such as to give a greater estimated thickness than three hundred and fifty feet, yet we have no positive proof of a greater observed thickness.

* Geological Survey of Canada, Report of Progress, 1847-48.

In this estimate, it must be understood that the elevated portions of the island and peninsula of Mackinac are not included, for they are occupied only to a small extent, if at all, by strata of this age.

[The Onondaga Salt Group occurs at Mackinac and St. Ignace, but only in thin beds, having diminished in thickness to the west. North of St. Ignace Mr. Whittlesey found a marly bed about 50 feet thick, containing gypsum in isolated masses occurring under the same forms as in New York and Canada West. This marly bed with some higher and more calcareous ones, represent the Onondaga salt group. This bed is not recognized along Green Bay or at Milwaukee, and has probably entirely thinned out.

The Upper Helderberg series, including the Schoharie grit, the Onondaga and Corniferous limestone, is largely represented in the west, though those subordinate groups cannot always be made out. The series, like others, is marked by an increase of calcareous matter on passing westward. The limestones have been greatly worn or denuded, and upright or overhanging cliffs, tower rocks, and domes, are of frequent occurrence about the islands and peninsula of Mackinac. Mr. Hall closes with the following observations on the Fossils of the bed]:

So far as I observed, the island of Mackinac and the adjoining coast, furnish few of the larger corals, though Favosites and many species of Cyathophillidæ are common. Farther south, however, in the neighborhood of Presq'isle, on the western shore of Lake Huron, where the upper beds of this group come to the level of the water, the corals largely predominate, and the conditions of the ocean appear to have been highly favorable to their growth and development.

In collecting fossils about this island, one is liable to be deceived in regard to the character of the rocks, since it often happens that those of the Niagara group have found their way among the loose materials which have been transported from the northern shore. Thus, the Catenipora, Heliolites, and other corals of the Niagara group may be picked up at the base of these cliffs, associated with those which have fallen from the cliffs above. As a general remark, however, it may be stated that the fossils of the Niagara are much oftener silicified than those of the group under consideration. Owing to this fact, it might readily be inferred, without careful comparison, that the Niagara group existed on the island of Mackinac and on the peninsula to the northwest. Notwithstanding, however, the great similarity in aspect and color, they are very different in their chemical composition and their associated organic remains.

Most of the fossils collected at Mackinac prove to be undescribed, but are identical, however, with species which occur in the same rocks in New York. Some of the smaller bryozöoid SECOND SERIES, Vol. XVII, No. 50.-March, 1854.

25

corals belong to the genus Trematopora and Cladopora, while two or more species of Favosites were observed. A few shells only were collected, and these, with a single exception, belong to the Brachiopoda. Among the trilobites was a Phacops, resembling P. bufo and a species of Proetus, both apparently identical with species which occur in the upper Helderberg range. One of the most characteristic species, however, is the Phacops anchiops, characteristic of the Schoharie grit. The specimens of the latter fossil, though consisting only of portions of two bucklers, are so peculiarly characteristic, that I cannot doubt the identity in age of these widely-separated localities. Although it would have been desirable to identify a larger number of fossils with those occurring in the corresponding series elsewhere, yet the evidence is sufficient to remove this group from any below the Oriskany sandstone.

ART. XVIII-On the Theoretical Relations of Water and Hydrogen; by T. S. HUNT, Chemist to the Geological Commission of Canada.

IN carrying out his theory of types, M. Laurent proposed to consider water H2O2, having its equivalent represented by four volumes of vapor, as the type of the oxyds like M2O2, of the hydroxyds (MHO, and of the sulphurets corresponding to these two classes. By his system of compound radicals, Liebig had extended to organic chemistry the nomenclature of Lavoisier, and he looked upon spirit of wine CH O, as the hydrated oxyd of a radical ethyl (CAHs Et), while hydric ether CAH.O, was the simple oxyd of the same radical. But as ether-vapor contains in the same volume, twice as much carbon as the vapor of alcohol, Gerhardt had already proposed to double the formula of ether, and Laurent now showed that while alcohol is to be regarded as the hydroxyd of ethyl (Et HO, or water in which ethyl replaces an equivalent of hydrogen, ether is the anhydrous oxyd, in which the second equivalent of hydrogen is replaced, and should be written Et: 02. Hence while ether is neutral, alcohol is monobasic, having an equivalent of hydrogen replaceable by a metal, and is the type of monobasic vinic acids, while water is the type of bibasic acids: (Laurent, Récherches sur les Combinaisons azotées. Ann. de Chim. et de Phys., Nov. 1846.)

In a review of that remarkable essay, published in this Journal for Sept., 1848 (vol. vi, p. 173), I suggested that this view was "susceptible of still farther extension, and that we may include in the same type all those saline combinations (acids) which contain oxygen." I referred to the hypochlorites CIO, MO, as derivatives of the type H2O2, in which Cl replaces H; being (CIH)O:,

and (CIM)O2, while anhydrous hypochlorous acid is Cl2O2, the result of a complete substitution. "In the same manner nitric acid, N HO3, is a monobasic salt (i. e. acid), corresponding to water in which NO2 is substituted for H, as in many organic compounds; we have then (NO2, H)O and (NO2, M )O ;" or (NO1, H) O in the notation adopted above. "As an adaptation of this idea to bibasic compounds, sulphuric acid, SH2O4, is to be regarded as water in which S HO3 replaces H; thus (S HOз, HO. As the replacing elements contain an equivalent of hydrogen which is saline (i. e. replaceable by a metal), the acid is bibasic. When the hydrogen in S HO3 is replaced by a metal, we have a class of acid sulphates like (S KO3, H)O. The complete replacement of hydrogen in the original type yields (S HO3)20, which is the Nordhausen acid commonly represented' by 2SO3, H2O. This latter compound as Gerhardt has shown, corresponds to the anhydrous bisulphate of potash."

"The tribasic acids may equally be reduced to the same type, if we conceive the elements which replace one equivalent of hydrogen, to be bibasic instead of neutral or monobasic; phosphoric acid, PH3O, is (PH:03, H)O."

"The primitive saline type is then essentially, bibasic, and is presented in its most elemental form in water, while the simplest type of the monobasic salt, which is a derivative of the last, is found in hypochlorous acid." p. 174. This view of the derivation of polybasic acids is illustrated by the bibasic sulphacetic, and the tribasic sulphosuccinic acid.

On page 177 we further remark, that "the binary molecule of the metals, hydrogen, chlorine, bromine, etc. will be seen to be the type of an immense number of combinations, embracing the various alloys and amalgams, the hydracids like hydrochloric acid, with their corresponding salts, and such compounds as Cl Br and CII, while ICs is referable to a triple molecule of these elements, represented by H; to this type belong the perchlorids of antimony, arsenic and phosphorus, while the corresponding trichlorids form a double molecule."

In a subsequent Essay on Chemical Classification read before the American Association for the Advancement of Science, at Philadelphia, in September, 1848, and published in this Journal for May and July, 1849, (vols. vii and viii,) we observed that the relation between alcohol and acetene is that which subsists between the two types H2O2, and H2, acetene being hydrogen in which ethyle replaces H, thus CAH,H=C4H6, while hydrochloric ether is a chlorinized hydrocarbon corresponding to hydrochloric acid, so that having repeated what has been already cited as to the type H2, we add, "moreover it follows from the relations of HCl to the chlorinized hydrocarbons, that it (H2) is the type of all the hydrocarbons, as well as of the alkaloids

which may be described as amidized species of them, and are equally susceptible of substitutions by chlorine." It was also remarked that as many neutral oxygenized compounds, which do not possess the saline character, are still derivatives of acids which are referable to the type H20, we may regard all oxygenized bodies as belonging to this type." "While nitric acid is N HO3, or (NO,HO, the result of the complete replacement of H by NO: will be (NO).O, or the unknown dry nitric acid, homologue of the so-called anhydrous phosphoric and arsenic acids, which are equally (PO2)2O, etc." Vol. viii, p. 92.

One of the objects proposed in the essay just quoted, was a comparison of the views of Gerhardt and Liebig with regard to the formation of ethers, amids, and allied bodies. Gerhardt in accordance with the electro-chemical theory of Berzelius, had considered the acids in these reactions to be electro-negative by their oxygen, while the alcohols, ammonia, and the hydrocarbons were electro-positive by their hydrogen, so that these bodies minus H2, replaced O in the acid.* To this view we objected that it leaves unexplained that change in the basic relations of the acid, which Liebig rightly understood when he compared the ethers to salts, and represented the acid as losing H, which is replaced by the elements of the alcohol minus HO2. This theory, unlike that of Gerhardt, made the ethers of the hydracids enter into the same class with those of the oxacids; at the same time it did not include those bodies which are produced with the elimination of HO, by the action of oxygen acids upon ammonia and hydrocarbons, and which were recognized in Gerhardt's system, as completely analogous to the ethers in the mode of their formation. Here the compound radical theory is found to be defective, although the analogy which forms its point of departure In concluding this comparison we remarked that "we are led to recognize the view of Liebig, apart from his ideas of dualism, and his theory of compound radicals, as the one fundamentally true." (Vol. vii, p. 405).

In this Journal for March, 1848, (vol. v, p. 265.) we observed that the relation of wood-spirit to acetonitryl is the same as that of water to hydrocyanic acid, and that water differs from woodspirit, precisely as this last differs from spirit of wine, so that the relation of homology, recognized by Gerhardt in the compounds of carbon, is extended to water and hydrogen; for from the relations which we have asserted between H2O2, and H2, it follows that while water is the homologue of the alcohols, hydrogen H: is the homologue of acetene C1H, (Et H.) and of formene C2H4, (Me H) which Frankland calls hydrids of ethyl and methyl, as well as of his zinco-methyl C. H3, Zn. The bodies which he regards as the alcohol radicals are still homologues of

* Précis de Chimie Organique, tom. ii, p. 495.