but as the analysis made out and accepted as the composition of margarite did not accord with that of emerylite, I undertook to reëxamine margarite, when I found that its composition had been erroneously determined, and that it was, in fact, the same mineral with emerylite, which last name has had to yield to the priority of date of the other.

I have analyzed the margarite from Chester and find its composition as follows:

There is a little titanic acid with the oxyd of iron that I did not estimate.

Chlorite. This mineral as found with the emery is the socalled corundophilite of Shepard; on examination it proves to be, both chemically and physically, a chlorite of the variety ripidolite.

Biotite. In examining a specimen of dark green micaceous mineral which I took to be chlorite (the corundophilite of Shepard) and from its purity expected to get a very accurate idea of its composition, but in the very commencement of the examination it was discovered to be well characterized biotite.

This mineral occurs on the surface of a white rock that Prof. Shepard calls indianite, but which I have not had time to examine. It is in small thin micaceous crystals perpendicular to the surface of the indianite; in the mass, it is of a dark green color, so dark that at a little distance it looks like lamellar plumbago. A careful analysis gave the following composition:

This corresponds with the composition of the biotite from Monroe county, New York, as made out by Prof. Brush and myself in our reexamination of American minerals several years ago.

Corundophilite proved to be a chlorite.-About the time I published my memoirs on emery in 1850 and 1851, Prof. Shepard made the announcement of a new mineral (this Journal, 1851, xii, 211), stating that it "occurs with corundum near Asheville, in Buncombe Co., N. C., in imperfect stellate groups, and also spreading out in lamina between layers of corundum; color leek-green, etc." An analysis of it afforded silica 34-76, protox. iron 31.25, alumina 8:55, water 5·47, making a loss of nearly 20 per cent, a portion of which he attributes to alkalies; neither lime nor magnesia were detected. He operated on 140 milligrams; this mineral was considered a new one, and Prof. Shepard called it corundophilite. Supposing that I had observed the same mineral in certain specimens of emery and emerylite from Chester, Mass., I enclosed a fragment of the specimen to Prof. S. to ascertain if this was the mineral he called corundophilite; he returned the specimen, announcing that it was. I then analyzed the same and found it to be, both chemically and physically, a chlorite, identical no doubt with the chlorite I found associated with the emery of Asia Minor; both the Asia Minor and Chester varieties occur in compact mass, composed of an agglomeration of small crystalline plates-identical with the chlorites of Mont des Sept Lacs and of St. Christophe, and the ripidolites of Rauris and St. Gothard. In the following analysis I do not pretend to furnish that of the pure mineral, as from the thinness of the layers in the specimens at my disposal it cannot be separated in that state of purity I am in the habit of seeking for in all minerals that I examine:

The optical characters were not examined, there being no means at hand.

I may remark that the alumina and magnesia were separated by resolution and reprecipitation three times.

Tourmaline. This mineral is also found with the emery of Chester in the same manner as with the emery of Naxos.

Titaniferous iron (ilmenite).—This is found principally in flattened crystals in the margarite.

Oxyd of titanium (brookite or rutile).-With the diaspore we found some beautiful flattened hair-brown crystals; the specimen in my possession does not furnish the face of the crystals so as to enable me to make out what form of titanium oxyd it is. Prof. Shepard thinks he has sufficient evidence to pronounce it to be brookite.

Magnetic oxyd of iron. This ore of iron is found in great abundance associated with the emery, and is worked for the manufacture of iron; it contains a little oxyd of titanium.

The above, as well as some other associated minerals of less importance, justify the concluding remarks of my paper on emery fifteen years ago, viz: "I do not risk much in saying that the hydrate of alumina or diaspore, as well as the silicate or emerylite, chlorite or tourmaline, and the minerals of iron, as magnetic, titaniferous iron, &c., will be found almost everywhere with the emery and corundum."

ART. XV.-On some minerals associated with the Cryolite in Greenland; by G. HAGEMANN.

A NOTICE of the pachnolite, discovered by Prof. Knop in the Greenland cryolite, has already appeared in this Journal.' On examination of several cargoes of cryolite imported by the Pennsylvania Salt Manufacturing Company, I have not only found pachnolite, but also have observed some other minerals which may be of interest.

Dimetric Pachnolite.-Among these is a mineral first observed by Prof. Julius Thomsen of Copenhagen, the originator of the cryolite industry. As I am informed, he found a mineral, which on a preliminary examination he thought might prove a fluorid of silicon compound, but I have not heard of any further investigation of the substance. In looking over the cargoes of cryolite I have found a mineral which I believe to be the same with that noticed by Prof. Thomsen.

The mineral crystallizes in dimetric form, the dimetric pyramid and prisms being plainly seen, but no further crystallographic examination was made. It has a distinct basal cleavage. The color is white with a reddish tinge, the crystals have a bright luster, and are coated with a white earthy envelop (Si). Sp. gr., 2.74-2.76; hardness, about that of cryolite. Heated in the closed tube this mineral yields water with an acid reaction which etches the glass. At a higher temperature it melts to a clear glass, fusing even more readily than cryolite. When pulverized it is easily decomposed by sulphuric acid, and on qualitative analysis it proves to contain water and fluorine, aluminum, calcium, sodium, and some silica. In the quantitative examination it was found extremely difficult to separate the alumina from the lime when precipitated by ammonia from the solution in sulphuric acid. I redissolved and reprecipitated six times before obtaining a complete separation. The water was determined by heating the mineral with previously ignited

1 Vol. xli, p. 119.

quicklime. The fluorine was determined as fluorid of calcium by decomposing the mineral with a mixture of silica, and the carbonates of potash and soda. After the soluble fluorids were separated from the insoluble silicates, alumina and silica were separated by carbonate of ammonia, and fluorid of calcium with carbonate of lime were thrown down with chlorid of calcium; this precipitate was dried and ignited, and the carbonate of lime was removed by acetic acid. The silica was imperfectly determined, as I had not the means at my disposal to estimate it accurately. I treated the pulverized mineral with solution of soda and carbonate of soda, filtered, and decomposed the solution by chlorhydric acid; evaporating to dryness thus rendering the silica insoluble. Analysis gave,

2

The formula is very near Al, Fl,+2(Ca + Na)Fl+2HO, which corresponds closely with Knop's formula for pachnolite, Al, Fl,+3(3Ca+Na)Fl+2HO. I scarcely know how to place the silica, but I think it does not really belong to the compound. Arksutite. This is a white crystalline granular mineral with a high luster. No crystals were observed, but each grain shows at least one good cleavage. Sp. gr., 3'029-3 175, (variation probably caused by minute crystals of iron-pyrites). Hardness, the same as cryolite. Fuses at a red heat without giving off water. Analysis gave,

Hence the compound gives the formula Al Fl3+2(Ca,Na) Fl. I hope in time to find a series of these fluorids which will, perhaps, show how cryolite is decomposed into pachnolite, and this may be still further altered into what the Greenlanders call "natural soap," (a hydrate of alumina ?). Both these minerals are found associated with cryolite in the vicinity of Iviktant near Arksut-fiord, in South Greenland.

Natrona, Pa., May, 1866.

ART. XVI.-Evidence of Two distinct Geological Formations in the Burlington Limestone; by W. H. NILES and CHARLES WACHSMUTH.'

DR. CHARLES A. WHITE was the first to record any natural division of the Burlington limestone. In the Journal of the Boston Society of Natural History, vol. vii, No. 2, Dr. White has given a "Section of rocks exposed at Burlington." He there describes eight beds, which he numbers from the lowest upward. He refers the first six beds of his section to the Chemung group, and beds "No. 7" and "No. 8" to the Burlington limestone. In vol. ix of the Proceedings Bost. Soc. Nat. Hist., and in No. 4 of the Journal of the same society, Dr. White describes certain species of fossils from the Burlington rocks; and although he gives the beds or divisions in which the species occur, yet nowhere does he claim that the Burlington limestone comprises more than one geological formation.

Our own observations have led us to regard these two divisions of the Burlington limestone as two distinct geological formations. The lower division we call the Lower Burlington limestone, and the upper division, the Upper Burlington limestone. The reasons for ranking these divisions as distinct formations are as follow:

The Burlington limestones are eminently crinoidal in their composition, as well as in their better preserved fossils. While fragments of these remains form an important feature in the greater mass of these rocks, there are, likewise, some strata of considerable thickness, which are composed almost entirely of

While making a special study of a family of Crinoids, the Actinocrinidæ, the results of which were to have published in the Illustrated Catalogue of the Museum of Comparative Zoology, it became necessary to spend considerable time at Burlington, Iowa, for the purpose of studying the large and valuable collections of Mr. Charles Wachsmuth, Rev. W. H. Barris, and Dr. Otto Thieme. From the published observations of Dr. C. A. White, and from some notes made by Mr. Wachsmuth, I had been led to believe that a careful study of the distribution of the crinoidal remains at Burlington would be rewarded with interesting paleontological results. I found Mr. Wachsmuth acquainted with many important facts in this connection, which could be reached only by a long experience in collecting these fossils, by an intimate acquaintance with the species, and a series of most careful observations. Accordingly, with the consent of Prof. Agassiz, I associated myself with Mr. Wachsmuth, for a careful examination and comparison of all the specimens in the three collections, for the identification of the species, and for the determination of their stratigraphical position; with the intention of publishing, in some scientific journal, such results as. were of scientific interest.

We now give only a preliminary notice of some of the results of our study, designing to extend our investigations to more southern localities, where the Burlington limestone is exposed. In a future paper, we shall present a complete catalogue of all the described species of Crinoids occurring in these limestones, with notes upon their geographical distribution, their stratigraphical range, and the rarity or frequency of their occurrence.

W. H. N.