1884.] 133 EHyatt. 



fusiform while others had the usual shape. The globular cells 

 when loaded with granules were always quiet, the central parts of 

 the outer sides swollen into projecting globes, and the periphery 

 of the upper sides only slightly convex. The outlines were fixed, 

 fusiform, and their bases were flat on the inner sides. 



The cells in the interior of the mesoderm were either coarsely 

 granulated, and globular or more or less flattened and clear. The 

 former when compressed showed a nucleus as a slightly clearer 

 central spot, but the clear cells had visible nuclei and nucleoli of 

 large size. No movements or changes of form were observed in 

 these cells, and they were never so regular in outline as were the 

 cells of the ectoderm and endoderm, from which also they were 

 easily distinguished by their large size. There were evidently but 

 one class of cells in the mesoderm, which might differ in shape 

 according to the quantity of granules they were carrying, or in 

 size in some parts. The specimens examined had invariably 

 passed through their reproductive periods, and all our observa- 

 tions were made with the view of studying the histology ; thus 

 no efforts were made to observe the effects of feeding by direct 

 experimentation. The distinctions which have been drawn 

 between the functions of the collared and flagellated cells in the 

 ectoderm of the embryo, and in the ampullae of adults, are purely 

 physiological, and histologically these membranes appear to be 

 very similar. 



The morphology of the central cavity and its membranes is 

 also opposed to the opinions of Von Lendenfeld and Polae- 

 jeff, that the ampullae are exclusively excretory. Most spongolo- 

 gists will probably admit that the collared and. flagellated cells 

 of the central cavity in the adult of Ascones and the ascon stage 

 of the larva of Sycones are feeding cells, and that in the Carneo- 

 spongiae and Leucones the ampullae are localized branches of 

 this cavity, lined with similar cells and standing in similar rela- 

 tion to the inflow of water and food through the supply canals. 

 However willing to grant, therefore, that these ampullae might 

 have functions of excretion additional to those they probably 

 possessed in Ascones, the facts do not justify the conclusion that 

 that their collared cells are not also true feeding cells. 



Whatever the result of future research, it is evident, from the 

 evidence above given, that the tissues in sponges are as primitive 



