The Devonian formations, which on the map are colored in a broad belt through New York and thence narrowing westward to Cleveland and Sandusky, are there represented as cut off by the belt of lower carboniferous, before mentioned, which runs westward to the Illinois coal field. Now the fact is, that the formations traced through southern New York and thence to Cleveland, are absolutely and unmistakeably continuous along the western margin of the great Alleghany coal field, through Ohio and Kentucky, and even into Tennessee and Alabama.

Our author has recognized the formation about Richmond, Va. as Liasic, while the same formation in North Carolina is colored as New Red Sandstone. We do not discuss the question of the age of the New Red or Triassic of Connecticut Valley and further south we only say here that there is no reason whatever for regarding that of North Carolina as differing in age from that of the Richmond basin.

There are other errors with regard to regions less known, as, for example, that of extending the cretaceous area to the east side of the Missouri for several hundred miles above Council Bluffs; and of terminating the same formation on the northwest more than a hundred miles short of its known limits. But these are excusable, compared with many points we have passed in review.

.

ART. XX.-On the Chemical Composition of the minerals Algerite and Apatite; by J. D. WHITNEY.

1. Algerile.

In the Journal of the Boston Natural History Society (vol. vi, p. 118) an account of the analysis by T. S. Hunt of a supposed new mineral; to which he has given the name of Algerite, will be found. Another analysis of the same substance by Mr. R. Crossley has been published (see Am. Jour. Science, [2], x, 77).

Having been furnished by F. Alger, Esq. with specimens of this mineral, I have made an examination and analysis of it, the results of which appear to be of interest in their relation to an important branch of mineralogical chemistry, hitherto much neglected, but which has received a new impulse from the laborious and interesting researches of G. Bischof.

The first light thrown upon the real nature of this supposed new mineral species, was by J. D. Dana, who was struck by the evident altered appearance of the specimens examined by him, and published the following remarks (Am. Jour. Sci., [2], xv, 440): "I am satisfied that the form of the crystals is a square prism. In external appearance they would not be distinguished

from scapolite, and this naturally suggests some relation to this species." The results which I have obtained in my examination of Algerite seem to sustain the opinion of Dana, and it appears highly probable that this is an altered mineral, and also, as suggested above by Dana, a scapolite of the Wernerite variety.

On the first examination of this substance it seems evident that it is a mineral in the progress of decomposition. In some of the specimens the crystals have been wholly removed from the matrix, or only a trace of a brownish yellow powder left remaining. Their hardness varies from 2.5 or 3 to 3 5 or 4.

A comparison of the results of the analyses of Messrs. Hunt and Crossley will show that the discordances are too great to be explained on the ground of errors of analysis. My own analysis differs still more from theirs, than they do from each other, and though time and material were wanting to enable me to make as accurate a determination of some of the ingredients as I could have wished, yet it seems, at least, sufficient to settle the question of the claim of this substance to rank as a distinct species.

In noticing Hunt's analysis it will be seen at once that the ratio of the oxygen of the silica and bases cannot be expressed by any approximate simple numbers; so, also, in Mr. Crossley's, the ratio of the oxygen of Si, R, R & II is given as 7:3:1:1; while it is really 705: 3-22 : 1·22 : 1.

For convenience of comparison, the results of these two analyses are here given, side by side:

The results of my examinations are as follows:*

Before the blowpipe it blackens a little and soon fuses, intumescing considerably, giving a colorless glass, and glowing with a vivid light.

A small quantity of the mineral, selected with care as being the least altered in character, was ignited in small fragments, and lost 6.20 per cent. of water; another portion, which had the appearance of being more decomposed, lost, under the same treat-. ment 6-68 per cent. The pulverized mineral was found to be

It is proper to remark that these investigations were made two years since, and that the publication of them has been delayed in the hope of an opportunity to make still farther examinations.

very little acted on by chlorohydric acid, even when digested with it for a long time. The acid however took up a small portion of lime which was intimately disseminated through the mineral in the form of a carbonate. After the ignition it was noticed that portions of the ignited mineral remained nearly unaltered in appearance, while the larger part acquired a brick-red color, and on examination with the microscope was seen to contain silvery white scales, apparently of mica.

As only a small quantity could be used for analysis, the results can be relied on only as approximately correct. They are as follows:

The difficulty of reconciling this analysis with either of those cited above, on the hypothesis of the algerite being a mineral of a fixed composition, will be apparent. Several specimens were examined and all were found to contain phosphate of lime.

Scapolite is, of all others, a mineral which seems most liable to metamorphosis. Numerous examples of this will be found in Bischof's "Lehrbuch der chemischen und physikalischen Geologie" (ii, 403, 1433, &c.). It is proved by the analysis of a mica in the form of scapolite (from Arendal) that all the lime of this nineral may be removed and a part replaced by potash. It is proved by numerous analyses that a portion of the lime may remain in combination with carbonic acid. It is also demonstrated that the lime may be exchanged for magnesia so as to give rise to a magnesian mica; also that the entire mass may be converted into steatite, or a hydrous silicate of magnesia. My analysis shows in addition to these facts, that in the process of metamorphosis a part of the lime may enter into combination with phosphoric acid. It is interesting to notice, in this connection, the analyses by Messrs. Brewer and Garrett of a substance very similar in composition to the Algerite, and also evidently an altered scapolite, which are published in Dana's Mineralogy, third edition, p. 680.

So much is evident, that the Algerite is not a homogeneous mineral and cannot claim to rank as a distinct species; and it is highly probable that it adds one to the already numerous list of the products of the transformation of the scapolite family. Further analyses of specimens from this locality, if such can be obtained, will throw more light on this interesting subject.

2. Apatite.

The theoretical composition of apatite according to the latest atomic weights (Ca=250, Cl=4433, P=400, F=235·4) is as

follows:

There has been a good deal of discussion as to whether the formula proposed by G. Rose was borne out by the analyses. Rammelsberg analyzed the fine crystallized apatite from the Zillerthal, and endeavored to make a direct determination of the fluorine. He only succeeded in obtaining about one-fourth of the amount required by the formula, and remarks that he leaves it undecided whether this result is due to the imperfection of the analytical processes, or to the fact that the formula needs revision. At the request of G. Rose, R. Weber made several analyses of the Snarum apatite,† the results of which are regarded by Rose as decisive in favor of his formula. C. A. Joy‡ also analyzed a specimen from Faldig in Tyrol, which gave a pretty near accordance with the amount required by Rose's formula. No one has succeeded in obtaining the required amount of fluorine by direct determination, and the analyses have been made up from the amount of phosphoric acid, lime and chlorine by calculation. It will be noticed that all the analyses give an excess of phosphate of line. There is invariably a deficiency of chlorine or fluorine combined with calcium, and the analyses seem to indicate that the greater the degree of purity of the substance, the nearer the approach to the requirements of the formula. Almost all the analyses of the uncrystallized varieties show a very considerable deficiency in the fluorine and chlorine.

In order to throw some light on this question, I have analyzed a specimen of apatite from Hurdstown, N. J., furnished me by F. Alger, Esq.; the substance examined was broken from a large crystal of a greenish-yellow color, and was translucent on the edges. Two analyses were made, in one of which the separation of the phosphoric acid was effected by mercury, according to Rose's process; in the other the lime was separated by means of sulphuric acid and alcohol. The results were:

The analyses agree well together, but give too small a quantity of chlorid and fluorid of calcium.

ART. XXI.-Contributions to Chemical Mineralogy; by JAMES D. DANA.-Part II.

1. Relations of Anhydrous Carbonates and Sulphates. THE Carbonates that come under the general formula R C, and the Sulphates, RS, present each a case of trimorphism: and a third example of similar character is found in the combinations. of Sulphates and Carbonates (RS+n RC). These cases are as follows; the angles given are RR in the rhombodedral forms, and II (P) in the prismatic:

There is a remarkable parallelism between the sulphates and carbonates. The difference between the angle of Arragonite and Calcite is about 11 degrees; between Calcite and Barytocalcite 10 degrees. Nearly parallel with this, the angle in Anglesite or Anhydrite is about 10 degrees larger than that in Dreelite, and that of Dreelite is 10 degrees larger than that of Glauberite. The sulphatocarbonates correspond with the sulphates nearly in angle, and confirm this parallelism. Susannite and Dreelite are nearly the same in angle; so also Leadhillite and Anglesite, and Lanarkite and Glauberite.*

It hence follows that the homologous prisms in Arragonite and Anglesite are the prisms, above mentioned, of 116° 10' and

The angle of Susannite mentioned by Haidinger is that of 2R. Brooke and Miller give Leadhillite another position which makes the angle I: I=120° 20'. It is probable that this is not the correct position, from the fact that the angle instead of being between that of the carbonate and sulphate of lead, or identical with one or the other, is much larger than that of the carbonate. In another position, taking 1 as I, we have I: I=103° 16', or near Anglesite. This view is sustained by the relations of Susannite (which is identical with Leadhillite in composition) to Dreelite.

In Lanarkite, according to measurements by R. P. Greg, Esq., recently communicated to the writer, the occurring vertical prism has the front angle 49° 50'; and this being taken as the prism i2 (∞ P'2), makes I: I= 85° 48', as above stated