\ 



E. H. L. Schivarz — Shell- Structure of Ammonites. 251 



the true cause is to be found in the ultimate structure of the prisms. 

 In offering a new theory, I will suppose that calcium carbonate 

 is deposited in the form of flat rhombohedra, since I have reason to 

 believe that the shells of Ammonites and Nautili consist of calcite ; 

 but if the calcium carbonate crystallizes in the rhombic system, 

 a similar form could be substituted without injuring the conclusion. 

 The shell substance as it is poured out from the mantle surface 

 consists of a gelatinous mass, holding calcium carbonate in solution 

 (mucoso-cretacee of Blainville ^), and may be seen in that condition 

 on the inside of a newly-opened oyster-shell. This homogeneous 

 secretion hardens by the separation of the calcium carbonate in the 

 crystalline state, while the conchiolin moulds itself around the 

 growing crystal. Eventually, then, there is formed a thin mem- 

 brane of conchiolin with innumerable flat rhombohedra lying 

 imbedded in it; the animal matter on their broad faces forming 

 the conchiolin laminas, while that on the short faces forms the 

 boundaries of the future prism. Now supposing the subsequent 

 layers place their positive rhombohedra on the positive ones of the 

 first, and the negative on the negative, we shall finally get two sets 

 of columns leaning in opposite directions (Woodcut 2), such as 



we find in the shells of Gasteropods.^ If, however, the positive 

 rhombohedra are placed on the negative ones of the layer before, 

 and vice versa, we shall get columns perpendicular to their base, but 

 with a jagged outline (Woodcut 1). That the crystals of one layer 

 have the power to influence those of the next is shown in the outer 

 layer of the Nautilus, where the proximal granules are in optic 

 continuity with the prisms below. 



The latter of the two cases is that which may very well be 

 imagined to obtain in the nacreous shells of the Cephalopoda ; for it 

 explains why the columns hold together so firmly, namely, by the 

 interlocking of the projecting angles of the rhombohedra ; and, also, 

 light entering such a tissue would be reflected and broken up into 

 its constituent colours by the myriad points, and would emerge 

 coloured in all the tints of the rainbow, as we actually see to be the 

 case. The conchiolin in the recent Nautilus is not very apparent 

 found the individual crystals, owing to its being caught between the 

 zigzag boundaries ; but in the extreme edge of the section, where it 



1 Manuel de Conchyliologie, 1825. 



2 Rose, Abh. Akad., Berlin, 1858, also Carpenter, loc. cit. 



