are but four of them yet known, viz: that which fell at Alais in 1806, that at Kold-Bokeveldt in 1838, that at Kaba in 1857, and that at Orgueil in 1864. They contain, respectively, about 3, 2, 06, and 6 per cent of carbonaceous matter.

I would here remark that the Alais, Kold-Bokeveldt, and Orgueil are more closely allied to each other than to the Kaba meteorite. The predominating mineral constituents are about as follows:

If we now contrast these mineral constituents with those predominating in well-known meteoric stones, a most striking fact presents itself-one not commonly realized by those engaged in the study of these bodies. It is seen on comparing the above with the following tables:

Silica..

Chassigny. Chateau Renard. Harrison City. Concord. Danville. Searsmont.

35.30

Magnesia.31.76
Iron protoxide 26.70

From these tabular statements, it will be seen that, deducting the small amount of carbon contained in the black meteorites, the mass of mineral matter constituting them is about the same, and corresponds thus with the so-called common type of meteoric stones; and hence the mineral matter to which these constituents belong must be the same in the two classes of meteorites, viz: olivines and pyroxenes, differing only in the more or less compact form of these minerals.

In the writings of some of the most astute observers of these bodies, we find little stress laid on these facts. Thus, M. Meunier, in a paper on the origin of meteorites, published in the Cosmos of December, 1869, expresses his amazement that I should speak of the circumscribed uniformity of the composition of meteorites as evidence of a circumscribed cosmical origin of these bodies, both with reference to the sphere or spheres whence they come, as well as their rock structure. He takes so opposite a view as to say (p. 9), "So far from the meteorites showing such a resemblance, we can establish between meteoric iron, olivine meteorites, aluminous meteorites, and carbonaceous meteorites, differences as great as between the most different terrestrial rocks." An assertion which would include all the ranges of rocks and sedimentary deposits from the basalt and granite to the cretaceous and tertiary deposits.

Let any one look at the above table, and say whether or not

he sees so vast a difference in the mineral constituents of the different meteorites there enumerated; and yet they represent the two extremes of these bodies so far as their external properties are concerned. It is well known that three or four minerals represent the great mass of the constituents of every meteorite in various proportions, viz: nickeliferous iron, olivine, pyroxene, and anorthite, especially the first three; and the purely iron meteorites must be recognized as magnified masses of the metallic particles to be found in every stony meteorite, not excepting even the carbonaceous meteorites.*

My object, however, in this paper is not to discuss at length the general internal resemblances of these bodies, as I may have occasion to do it more fully at another time. I wish simply to note, that black and pulverulent as are the carbonaceous meteorites, they are not removed by their mineral constituents from the so-called common meteorites. I now pass on to show that even in their carbonaceous constituent they are strongly linked even to the iron meteorites.

2. Graphite carbon in the Iron Meteorites.-Ever since the internal structure of this class of meteorites has been examined by sections through the center of these compact metallic masses, nodular concretions have been noted in their interior, the most common of which consist of troilite, a protosulphide of iron, and filling ovoidal cavities. Sometimes these troilite concretions have a thin coating of a lighter colored mineral known as schreibersite; and this last is also found alone in concretionary masses which are usually angular or lamellar.

Less frequent concretions than either of the above, and even more remarkable, consist of carbon of the character of graphite: these, like the troilite, usually fill irregular ovoidal cavities, and are more or less contaminated with the latter mineral.

The most important of the meteoric irons containing these nodules, that have come under my immediate observation, are the Toluca, the Cranbourne, the DeKalb, and the Sevier: the last two have received my special study, the latter furnishing much the larger part of the material in my hands.

Character of the graphite nodules. These concretions differ more or less in appearance, while their general character is the same. In this communication I call special attention to a large nodule taken from the very center of the Sevier iron, the largest that has come under my observation, and perhaps the largest known. It was detached from the iron entire and perfect in every respect. Its greatest length is 60 mm. ; its dimensions in

* At present, the Orgueil and Rhoda meteorites are the only two in which no positive evidence of the presence of nickeliferous iron has been traced; in the Orgueil, however, we find nearly three per cent of oxides, nickel and cobalt, and the Rhoda has not been very critically examined.

the other direction vary from 20 to 35 mm. The weight before it was cut was 92 grams. Its form is that of an irregular dumb-bell, flattened on one side, and slightly nodular on the surface. Its color is plumbago-black, except at small places on the surface, where there is a little bronze-colored troilite. Its texture is remarkably close and compact, and it is cut readily by the saw except when the tool encounters particles of enclosed troilite. Its structure and powder is not unlike that of the close-textured graphite of Borrowdale in Cumberland, England, and quite unlike the scaly graphite such as that from Ceylon, or that found in certain cast irons.

Examined from the circumference to the center this nodule presents the following appearances: About one-fifth of the circumference of the section is made up of troilite with a thickness of one millimeter. The remainder of the section has all the aspect of graphite, except in a few spots. In the nodule there is a small mass of troilite not unlike in form the entire nodule; it is 10 mm. long by about 5 mm. wide; it is not continuous from its circumference to its center, but the center portion is cut off completely from the exterior portion by a thin belt of graphite one-half to three-quarters of a millimeter in thickness. Again on other parts of the surface small particles of troilite are to be seen.

The specific gravity of this graphite is 2.26 mm., as determined on a piece in which no troili te was visible to the eye, and after it was immersed in water and placed under the receiver of an air pump to abstract the air from its pores.

Chemical character of the graphitic nodule.-When pulverized and heated in a short glass tube from 100° to 150° C., water is given off which is doubtless water absorbed from the air by the graphite. If heated a little higher and then brought close to the nose, a slight empyreumatic odor is apparent if heated still higher, there is a slight odor of sulphuretted hydrogen. If heated in the open air the carbon is burnt with difficulty, showing its true graphitic nature.

Treatment of the graphite by ether.-Very pure and concentrated ether was added to two grams of material in powder and rubbed up in a porcelain mortar; then poured into a small beaker; a little more ether was added and the two allowed to remain together for 12 or 18 hours, the vessel being covered to prevent evaporation. The ether was then filtered off from the graphite which was finally washed with a little ether. The ether was allowed to evaporate slowly in the uncovered beaker placed where the temperature was about 33° C. After the ether had evaporated, long colorless acicular crystals covered the sides of the vessel, and some shorter ones were in the bottom. There were also some rhomboidal crystals and rounded particles. The

solid residue exhaled a peculiar odor of an aromatic character, somewhat alliaceous. The quantity of these crystals was small, not exceeding 15 milligrams from two grams of the graphite. Heated on a piece of platinum foil they fuse at about 120° C. Heated in a small tube closed at one end, they first melt and then volatilize, condensing in yellow drops that soon solidify leaving a carbonaceous residue. They are not soluble in alcohol, but very soluble in sulphide of carbon. Fuming nitric acid oxidizes the material, and gives, as one of the products, sulphuric acid. The quantity was too small to admit an ultimate analysis, but it was very evident that sulphur was the predominating constituent, the remainder being carbon and hydrogen. These three elements may be combined, forming a peculiar sulph-hydrocarbon, which in a previous note I called celestialite, or it may be sulphur containing a minute quantity of a hydrocarbon that gives the peculiar odor and determines the somewhat singular form of crystallization of the sulphur; for these acicular crystals may be only elongated rhombohedrons.

Be the compound what it may, it is a matter of chemical and astronomical interest that a solid graphite nodule thus encased in iron should contain a sulph-hydrocarbon, or free sulphur and a hydrocarbon.

The graphite powder, after treatment with ether, was then treated with bi-sulphide of carbon (which was re distilled just before use) and after standing two or three hours was thrown on a filter; the filtrate was evaporated to dryness, and the residue was a yellow solid; in this instance, as in the last, the quantity was small. This, when heated in the open air on platinum foil to a red dull heat, first melts at about the temperature that sulphur melts, and finally the sulphur is burnt off, leaving a carbonaceous residue. When heated in a tube, it sublimes, leaving a black residue.

To all appearances this is the same substance, or mixture of substances, that was extracted by the ether, the ether not having exhausted the graphite in the first treatment.

The graphite nodules of the DeKalb and of the Cranbourne irons, on treatment with ether and sulphide of carbon, gave similar results. In the case of the Cranbourne graphite I had less than one hundred milligrams of the material to operate with, and I hardly hoped to obtain satisfactory results, but I did succeed, however, in obtaining such without the acicular crystals, for the whole residue was less than one milligram; but I had enough to recognize the peculiar odor, and also the minute quantity that could be scraped off the vessel in which the evaporation took place furnished the marked reaction by heat of volatilization in part and condensation of the same

with a carbon residue. The Cranborne graphite requires more trituration with the ether than that from the Sevier meteorite as it is more flaky on being rubbed up.

Further remarks about this peculiar substance will be made a little farther on, when I come to speak of the same compound as obtained from the black or carbonaceous meteorites.

[To be continued.]

ART. LI.-Contributions from the Sheffield Laboratory of Yale College. No. XXXVIII.-On the Oxidation product of Glycogen with Bromine, Silver Oxide and Water; by R. H. CHITTENDEN, Ph.B., Assistant in Physiological Chemistry.

WHILE submitting an aqueous solution of glycogen to the action of bromine in an open vessel with the aid of heat, it was observed that the strong opacity of the fluid gradually disappeared, and that, after the removal of the free bromine by partial evaporation, a perfectly clear fluid remained which contained considerable combined bromine.

This reaction, indicating union between the glycogen and bromine, pointed to the possibility of the formation of an acid from the glycogen by oxidation, in a manner analogous to the formation of "dextronsäure" from dextrin, and "lactonsäure" from lactose, as described by Habermann, Barth and Hlasiwetz. The following experiments were undertaken to form, if possible, a corresponding acid from glycogen. The glycogen employed was prepared from the muscular tissue of Pecten irradians, and was as pure as could be obtained. The process of oxidation was as follows: fifty grams of glycogen dried at 100° C., were dissolved in 300 c.c. of distilled water, and this solution transferred to a champagne flask fitted with a caoutchouc stopper, in which was a small stout glass tube drawn out to a point. Forty grams of bromine were then added and the stopper wired in. The flask was then heated in a water bath until the red vapors of bromine had entirely disappeared, which required about two hours' boiling. A heavy light yellow or white precipitate formed at first, which completely disappeared by the time the bromine had all been taken up. At the end of this first treatment the fluid was perfectly clear and of a pale yellow color. After cooling, the gases were allowed to escape, by breaking the end of the tube in the cork, and, being collected, were found to consist mainly of carbonic acid and bromoform. The stopper was then removed and 40 grams more bromine Ibid., cxxii, 96.

* Annalen der Ch. u. Pharm., clxii, 297.

This Journal, III, vol. x, p. 26. § Annalen der Ch. u. Pharm., clix, 315.