BRACHIOPODA. 305 



recently by Dr. Carpenter; according to this last observer, it 

 consists of flattened prisms of considerable lenoth, arranged 

 parallel to each other with great regularity, and obliquely to the 

 surfaces of the shell, the interior of which is imbricated by their 

 out-crop. This structure is found only in the Riiynchonellid;e ; 

 but in most — perhaps all the other Brachiopoda — the shell is 

 traversed by canals from one surface to the other, nearly verti- 

 call}', and regularly, the distance and size of the perforations 

 varying with the species (vol. i, t. 1, f. 5). Their external 

 orifices are trumpet-shaped, the inner often very small ; some- 

 times they bifurcate towards the exterior, and in Crania they 

 become arborescent. The canals are occupied b}^ cojcal processes 

 of the outer mantle-layer, and are covered externallj^ by a thick- 

 ening of the epidermis. Mr. Huxle3^ has suggested that these 

 coeca are analogous to the vascular processes by which in many 

 ascidians the tunic adheres to the test ; the extent of which 

 adhesion varies in closel}' allied genera. The large tubular 

 spines of the Productidtie must have been also lined by prolon- 

 gations of the mantle ; but their development was more probably 

 related to the maintenance of the shell in a fixed position, than 

 to the internal economy of the animal. — King. Dr. Carpenter 

 states that the shell of the Brachiopoda generally contains less 

 animal matter than other bivalves ; but that Discina and Lingula 

 consist almost entirely of a horny animal subtance, which is 

 laminar, and penetrated by oblique tubuli of extreme minute- 

 ness. He has also shown that there is not in these shells that 

 distinction between the outer and inner layers, either in struc- 

 ture or mode of growth, which prevails among the ordinary 

 bivalves ; the inner laj^ers only differ in the minute size of the 

 perforations, and the whole thickness corresponds with the outer 

 layer onl}^ in the Lamellibrancliiata, The loop, or brachial pro- 

 cesses, are alwa3'S impunctate. Mr. Hancock's researches would 

 tend to show that these conclusions are generally correct, but 

 not entirely so. " When the shell is dissolved in acid the free 

 border [of the mantle] which projects beyond the mai'ginal fold, 

 and which is applied to the extreme edge of the shell, can be 

 examined with advantage. The pallial coeca are then completely 

 exposed appended to the membrane in various stages of develop- 

 ment, and the spaces between them are found studded all over 

 with rather large, clear, oval, cell-like spots, which are arranged 

 with considerable regularity in rows, so that those in the approxi- 

 mate rows alternate. These spots apparently correspond to the 

 bases of the prismatic columns of the shell ; and if it be allowed 

 that thc}^ represent spaces in which calcareous granules had 

 been accumulated, it is easy to understand how the fibrous or 

 columnar structure is formed. A succession of layers of such 



