Young embryos present merely a fibrous rudiment of the dorsal layer. The ventral layer is formed by the successive deposit of calcareous laminm inwards. When the first lamina has been formed, a deposition of small cylindrical bodies takes place upon its inner surface. These increase, widen and become ramified at their extremities, forming ramified columns. A second calcareous lamina is now formed, connecting their ramified extremities, upon whose under surface the like process takes place, and this is repeated until the ven tral layer has attained its full thickness. The ramified columns are regularly transversely striated ; with the age of the shell additions are continually made to their lateral dimen sions until they coalesce and constitute the septa of the perfect shell, upon which the strim remain visible.
Biker was unable to find any cellular structure in the columns or laminm themselves, but describes a layer of nucleated cells under the shell, which he regards as the agents in its secretion. Some researches recently made by II. Miiller (Gegenbaur, KiiHiker and H.
/. c.) corroborate this view. He finds the shell of the Loligidm invested by an ex cessively vascular membrane, which is almost wholly covered by a layer of epithelial cells towards the shell. On the dorsal surface they are for the most part rounded ; on the abdo minal surface, and particularly towards the anterior point, they form narrow cylinders hich attain a length of as much as 0.07/". They appear to give rise to the structureless layers of the shell. The lateral styles of the Octopods present similar relations.
The structure thus described, though appa rently so widely different from that of ordinary mollusks, does not really differ very widely from the cancellated shell structure of Rudis tes, Sec., or still better of Pleurorhynchus as described by Dr. Carpenter. If we leave out the sides of the hollow prisms in the latter shell in fact, it will correspond exactly with one lamella of the ventral layer of the sepiostaire.
For a comparison of the shells of Spirula and Belemnites with those of Sepia and of Gasteropods, I must refer to Dr. Carpenter's Memoir so often cited.
2. Conversionary integument of the Ilfollusca containing cellulose.— This form of integument has hitherto been found in the Ascidians alone, in which the existence of cellulose was first detected by Schmidt in 1845. Schmidt's dis covery was confirmed, the fact of the existence of cellulose in all the genera of Ascidians deter mined, and the chief morphological characters of their test set forth in the memoirs by laliwig and Milker " Sur les Enveloppes des Tuni ciers," which appeared in 1846, since when further investigations have been made by Schacht and by my self. I must refer the
reader to these papers for an account of the various opinions which have been entertained with regard to the structure of the Ascidian test, as I can only lay before him what are in my belief the facts of the case.
The test of the Ascidians is never composed of pure cellulose, but consists of an animal membranous matrix, to which the cellulose has the same relation as the calcareous salts have to the membranous basis of bone or of shell. The cellulose is, in fact, diffused through the membranous matrix, thoroughly impreg nating it.
This membranous nitrogenous matrix in which the cellulose is deposited, presents great diversities of structure in the genera of Asci dians, reprcsen ting, in fact, almost every known tissue. Thus in one genus we have a test re sembling cartilage ; in another, like bone ; in a third, like connective tissue. It may either be without vessels or traversed by branched and ramified vascular processes of' the body. It is in all cases, however, a product of the metamorphosis of the ecderon of the outer tunic or mantle and, complicated as its struc ture may be, corresponds morphologically with the shell of other mollusks, or with the epi dermis of the higher animals. In fact, if a section be made through the outer tunic and test of an Ascidian, as in fig. 314. A, taking care not to disturb the natural relations of the parts, we observe at the line of contact between the outer tunic and the test (a) an arrange ment of the parts very closely resembling what exists at the junction of the derma with the rete Malpighii in the human skin. The outer tunic, like the former, is constituted by bundles of rudimentary connective tissue which run inwards to form sheaths around the muscles, leaving between them spaces, the sinuses of the blood vascular system, while externally they fuse together into a homogeneous substance containing endo plasts, which is thrown into processes and passes insensibly outwards into a layer of similar substance, with very close-set endo plasts almost perpendicular to its surface, which forms the commencement of the proper test. Externally to this rete Malpighii, the deposit of cellulose commences; the tissue un dergoing at the same time a fibrous metamor phosis. The line a is therefore a protomor phic line, and the test is the product of the growth and conversion, by deposit of cellulose within its elements, of a true ecderon.