DEVELOPMENT OF THE OyVPE SPECIES OF PERIPATUS. 537 



Cycla s respectively, arising in the typically Annelidan manner; 

 but Ziegler, so far as I am aware, has alone succeeded in ascer- 

 taining what part these bands take in the formation of the peri- 

 cardium, generative organs and kidney. The generative cells are 

 derived from the mesoblastic bands. The pericardial cavity arises 

 as two cavities — one in each band — which subsequently unite. 

 The kidneys are hollowed out in certain masses of cells of the 

 bands. These results, if generally applicable, appear to con- 

 firm absolutely the anatomical proof of the coelomic nature of 

 the Molluscan pericardium. It is interesting to notice that in 

 Ziegler^s figure (fig. 27) the developing pericardial vesicles have 

 exactly the same relation to the primary body cavity or vas- 

 cular space, i. e. they lie within it, as the nephridial vesicles of 

 Peripatus have to the vascular cavity of the leg. 



There are certain animals to which the above general con- 

 siderations as to the distinctness of the ccelom and vascular 

 system do not apply. I refer more especially to the Nermer- 

 tinea and Hirudinea. In the Nermertinea, according to 

 Oudemans (No. 24), and in the Hirudinea, according to 

 Bourne (No. 4a), structures which it is difficult to believe are 

 not nephridia open into the vascular system. I do not intend to 

 discuss these cases now, because, on the one hand, this paper is 

 already too long, and because, on the other, I do not think our 

 present knowledge is sufficient to enable this to be done with 

 profit. But I venture to submit with regard to them that 

 it is not clear in either case that the vascular system into 

 ■which the nephridia open is homologous with that of other 

 types. The very fact that there is a communication with the 

 exterior is a strong point in favour of the space being coelomic ; 

 and it should be remembered that very little is known with 

 regard to its development in either group. 



In conclusion, I may point out, that whereas in most 

 animals, e.g. Annelida, Mollusc a, the vascular space or 

 pseudocoele appears before the ccelom, in Peripatus the coelom 

 appears first, and that in Arthropods, at least, the vascular 

 space is in the early stages very commonly occupied by yolk, 

 while the coelom is entirely free from yolk. This latter 



