Thomsonite, Natrolite, 66 21 8 6 Si 6 Orthorhombic, [Na2,Ca], [Al2]OS2. 21H20 Hexagonal, Ca, 8 Na2, [Al JiOS ̧02. 2H2O 8 4. 2H2O Ca, [Al2OS 04. 6H20 Ba, [AllOvSi O 6. 5H2O Ca, AllOvSiOg. 5H20 [Al2OSO. 6H20 Ca, viii 8 8 8 6 8. And it will be seen that the atomic ratio of the basic radicals to the various complex acid radicals is the same for all these minerals, although the ratios of the basic radicals to the silicon alone are quite different. The last, however, vary according to a simple law and indicate a still further relation between the several species. These relations may be discovered by a simple inspection of the symbols, but they will be rendered much more striking by tabulating the atomic ratios deduced as above. 4 Lastly, the atomic ratios express the results of analysis in the simplest possible terms and in a form, in which they can be most readily compared in various combinations, and such combinations bring out unexpected results. Thus the atomic ratio of zircon ZrŎ, Si(4:4) is the same as that of garnet, when the sum of all the basic radicals is compared with the silicon, and the two minerals although belonging to different systems have nevertheless very nearly the same crystalline form. Again, zircon is isomorphous with tin-stone and rutile. Now the total quantivalence of all the radicals both basic and acid in each of these minerals (taking the symbols as they are usually written) is a multiple of four and there are reasons for believing that the molecules of the three, last named, are more condensed than the common symbols would indicate. It is therefore possible that all four may have a similar constitution as is shown in the following table, and the important class of minerals of which spinel is the type may be included in the same scheme. Such facts as these do not of course prove anything. They are however suggestive, and are adduced because they illustrate the use that may be made of this principle of atomic ratios. ART. XL.-Notice of some New Reptilian Remains from the Cretaceous of Brazil; by Prof. O. C. MARSH, of Yale College. THE Only account of vertebrate fossils from the fresh-water cretaceous deposit near Bahia, Brazil, which appears to have been published hitherto, is a short notice in a paper by Mr. S. Allport, in the Journal of the Geological Society of London for 1860. In this article the author gives a description of the locality, and figures several specimens of reptilian and fish remains, but with no explanation of them except a reference to the opinions of Prof. Owen, and Sir Philip Egerton, as to their general affinities. While engaged in a geological exploration of the coast of Brazil, in 1867, Prof. C. F. Hartt, of Cornell University, visited the same locality, and among the fossils obtained was a small collection of vertebrate remains, supposed to be mainly reptilian, which he has recently submitted to the writer for examination and description. Most of the specimens are too imperfect to admit of accurate determination, but some, however, are sufficiently well preserved to show clearly their main characters, and a number of them prove to be identical with those obtained by Mr. Allport. Several of the specimens were found on examination to be portions of large fishes, in part referable to the genus Lepidotus, and some of them indicating apparently a new type. These will be described, with other fossils from Brazil, in a work on the geology of that region, soon to be published by Prof. Hartt. The most interesting of the reptilian remains collected by Prof. Hartt in the Bahia deposit is the tooth of a large Crocodilian, from the arenaceous shale near Plantaforma station, on the Bahia railroad. This specimen is in an excellent state of preservation, and indicates a species new to science. It is farger, more slender, and more pointed, than the teeth of existing crocodiles, resembling most nearly those of some extinct American species. It is conical in form, round at the base, and slightly compressed at the apex. The crown is two inches and three lines in length, along the outer side, and ten lines in diameter at the base. One edge is somewhat more convex than the other, and this is also true of one of the sides, and hence the tooth appears slightly curved in two directions. On either edge of the crown there is a sharp ridge, most prominent near the apex, over which it passes, but gradually disappearing before reaching the base, resembling in this respect the teeth of Thoracosaurus, from which, however, this specimen differs in being longer, and less curved than the teeth of that genus usually are. The sides of the crown are covered with fine, interrupted, undulating striæ, which appear to be different from the dental sculpture of the Crocodilia hitherto described. These striæ are most distinct near the middle of the tooth, becoming much more delicate at the base, and nearly obliterated at the apex. In size and general appearance, this specimen resembles somewhat the teeth of Crocodilus antiquus Leidy, from the Miocene of Virginia, but differs from that species in being less tapering, and in having the ridge on the edges extend farther downward. It resembles still more closely the teeth of a new species of Crocodile discovered by the writer at Squankum, N. J., in the tertiary green-sand, which will soon be more fully described under the name Thecocampsa Squankensis Marsh. Both species have essentially the same proportions, and similar dental striæ, but the cutting ridge of the New Jersey specimens is more prominent, and extends nearly or quite to the base of the crown. The two species were apparently about the same size, both being considerably larger than existing Crocodilians. Other parts of the skeleton of the Brazilian species would perhaps show generic characters to distinguish it from the modern procœlian Crocodiles, but in the absence of these, it may for the present be placed in the same genus. Its form, cutting edges, and especially its peculiar striæ, readily distinguish it from any species with which it is liable to be confounded, and it may appropriately be named Crocodilus Hartti, in honor of its discoverer, whose recent researches have thrown so much light on the geology of Brazil. Several specimens of reptilian teeth collected by Mr. Allport at Montserrate, a locality in the same deposit about two miles southwest of Plantaforma station, evidently belong to this species, as the illustrations accompanying his paper (Plate XVI, figures 1, 2, 3, and 5) clearly indicate. The explanation of the plate refers to the specimens as, "Teeth of Crocodile with delicately wrinkled surface," but no further description is given. In the same paper Mr. Allport has given figures of several Crocodilian teeth from the localities at Plantaforma and Montserrate, which are quite different from those above described. These are represented in Plate xv, figure 5, and Plate XVI, figures 4, 6, 7, and 8, and are referred to on page 268 as, "Teeth of Crocodile with strong continuous striæ, and coarse riblets.” These specimens, taken in connection with some imperfect remains in the collection made by Prof. Hartt, indicate the existence in this deposit of a second, and smaller species of Crocodile, probably allied to the modern gavials. The teeth are not so large as those of Crocodilus Hartti, and are more tapering, and more curved. They also differ widely in the striæ and lateral folds. These specimens may provisionally be referred to the genus Thoracosaurus, and, as the species is evidently new, it may be called T. Bahiensis. An interesting fossil, found by Prof. Hartt at Plantaforma station, is a fragment of a bone, evidently reptilian, but the exact affinities of which it is difficult to determine from this specimen alone. It resembles in some respects the extremity of an ulna, but after a careful comparison the writer is inclined to consider it the proximal end of a rib. It is much flattened at the articular extremity, and tapers gradually to the broken end, which is somewhat triangular in outline. Its length is about four inches, the transverse diameter of the perfect end two and a half inches, and of the other, one and a quarter inches. The larger extremity is divided into two articular facets lying oblique to each other, the smaller one being elevated about half an inch above the other, and covering rather more than a third of the entire terminal surface. In form and general proportions this specimen is not unlike the upper end of a right dorsal rib of some of the amphicœlian Crocodiles, especially a rib in which the head and tubercle have so closely approached each other that their articular surfaces are nearly confluent. The size and other characters of the specimen, however, seem to exclude it from that order, and it probably belonged to a Dinosaurian reptile, possibly the same as a large vertebra from Montserrate, which Mr. Allport figured in his paper in Plate XVII, and which Prof. Owen suggested might prove to be allied to Megalosaurus. The only other specimen in this collection that need be particularly mentioned here is a small flat bone, about two inches in length, with one articular extremity partially preserved. This appears to resemble most nearly the fibula of a Tortoise, and probably should be referred to that group of reptiles. The other vertebrate remains from Brazil obtained by Prof. Hartt are, in general, of less interest, but will be fully described in his forthcoming work. Yale College, April 5th, 1869. ART. XLI.-Notices of papers in Physiological Chemistry- 5. On the formation of Sugar in the Liver. (57.) On the 19th of December, COLIN presented a memoir on the relation of animal glycogeny to the production and destruction of fats.* His researches were extended to include the influence of this function on the composition of the chyle, the lymph and the blood, as well as the liver-tissue. With regard to the latter, he concludes as follows: (1) The sugar of the liver, at least in part, is manifestly derived from the saccharine or starchy food taken by the herbivora, or from those elements in a mixed diet, being carried to this organ by the portal vein and the hepatic artery. (2) This sugar appears to be also a product of the transformation of the fatty matters which accumulate in the hepatic cellules and in the intra-cellular spaces. (3) It appears in much larger proportion in fat animals than in lean ones; though the quantity continues to increase only up to a certain limit, and even diminishes very notably in livers which have themselves undergone fatty degeneration. (4) In the case. of animals who have no fatty tissues to be absorbed, the amount of sugar diminishes from the first moments of abstinence, and soon entirely disappears. (5) When the animal is fat, on the contrary the sugar is constantly renewed and so maintained in considerable amount, even when food is withheld, if only the normal temperature of the body be preserved. (58.) On the 2d of February, 1860, G. HARLEY communicated to the Royal Society the results of some experiments made in connection with Dr. Sharpey at University College, London, to test the truth of Pavy's views. The first step was to ascertain the presence of sugar in the blood under favorable conditions of diet. Three-fourths of an ounce of the carotid blood of a terrier dog, killed three hours after a diet of bread, milk, and boiled liver, was added to four times as much water, to which a few drops of acetic acid had been added, and which was kept boiling in a capsule; when the albumin was completely coagulated, the liquid was filtered and tested with potassic hydrate, both with and without cupric sulphate; the reaction for sugar was obtained. A second portion of blood after standing 35 minutes gave a similar result. To obtain an animal whose condition was normal, a dog which had been running at large was taken for a second experiment, and to avoid the *C. R., xlix, 981. Proc. Roy. Soc., x, 289. AM. JOUR. SCI.-SECOND SERIES, VOL. XLVII, No. 141.—MAY, 1869. |