30 INTRODUCTION. 



to succeed in obtaining a clear elucidation of this point ; still, I cannot but think that 

 a near approximation may be made to it. I have already mentioned, that with the internal 

 surface of the shell is incorporated a membrane, of which the csecal tubuli are pro- 

 longations ; and I have found the free surface of this membrane to be almost covered with 

 cells, resembling both in size and appearance those contained in the cseca. I have found 

 similar cells upon that surface of the mantle which is applied to the interior of the shell, 

 and which (as long since remarked by Professor Owen) possesses a degree of adhesion to 

 it, that does not exist among the Lamellibranchiata. Hence I am much inclined to 

 believe tliat, as was first suggested to me by Mr. T. H. Huxley, the space between the 

 mantle and the proper lining membrane of the shell is, in reality, a great vascular sinus ; 

 and that the membranous caeca opening-ofF from that space are really to be considered as 

 vascular processes, analogous to those which present themselves in the " test" of many 

 Ascidians.^ The cells which are contained in these caeca, and which are found so abundantly 

 on the contiguous surfaces of the mantle and the interior shell-membrane, must then be 

 considered as blood-corpuscles. 



It can scarcely be supposed that these vascular prolongations can have any relation to 

 the nutrition of the shell ; since we find them absent in a considerable number of the 

 shells of Brachiopoda, which in every other respect are formed upon the same type with 

 those wherein these caeca abound. We can scarcely do otherwise, I think, than regard 

 their presence or absence as bearing a relation to the general economy of the animal ; and 

 they must consequently be admitted as furnishing a character of considerable importance 

 in a systematic point of view. We shall find, accordingly, that (so far as yet known) they 

 exist in all the true Terebratulida, both recent and fossil ; and that (with the same reser- 

 vation) they are absolutely and entirely wanting in all the true Mhynconellidce, both recent 

 and fossil ; so that by this character alone, any member of the former family might be 

 safely diagnosed from any member of the latter. 



As frequently happens, however, in other departments of Zoology, a character which 



^ The peculiar appearance of these caeca and their contained cells, led me at one time to surmise that 

 they might possibly be the extremities of the ramified ovaria, ■which extend themselves through the mantle ; 

 or that, like the glandular clusters in the dorsal papillae of the Eolidse, they might constitute a subdivided 

 liver. Neither of these views, however is consistent with the entire disconnection of the cseca with the 

 proper mantle, and the animal it encloses; and, moreover, as Professor Owen has remarked to me, "these 

 palleal processes are so much more minute than the ova discernible in the ramified ovaria, and their 

 presence is so equally manifested on the ovarian and non-ovarian parts of the mantle, that no ground is to 

 be got for associating them in speciality of function with the generative system ; whilst the palleal lobes 

 are so thoroughly distinct (in comparison with the skin of Nudibranchs) from the visceral mass, that they 

 cannot have any special relation to the liver." I mention these surmises and their refutation, merely to 

 assist in determining, par voie d' exclusion, what is the real character of these palleal processes ; it being 

 often easier in physiology to determine what a thing or function is not, than what it is. Professor Owen 

 permits me to state, that he considers the view given in the text to be not discordant with the results of 

 his researches on the vascular system of the Terebratulidse. 



