1884.] 103 [Hyatt. 



tube of any sponge having a smooth lining membrane cannot be 

 separated from these cloacal vases. Thus the adult structure, 

 histology and color of the lining membranes of the perfect cloa- 

 cal tubes in Spongia do not permit us to imagine them as lined 

 by the endoderm, and the same must be said of the incurrent sys- 

 tem of superficial cavities and tubes. Vosmaer's results (Leucan- 

 dra aspera, Leiden, 1880, pi. n, f. 3-4) show that he con- 

 siders the supply system of tubes to be lined by the ectoderm in 

 Leucones and Carneospongiae. Though he regards the lining 

 layer of the excretory system as endodermic, his drawings, like 

 those of other authors, fail to show histological differences 

 between this layer and the ectoderm, both being composed of 

 flat epithelial cells. So far as can be seen both the histol- 

 ogy and the general aspect of these tubes give no support 

 to the idea that any of them are lined by membranes which 

 differ in any way from the ectoderm. This comparison of course 

 includes only massive sponges, Leucones and Carneospongiae, and 

 probably does not apply to any Sycones or to the primitive 

 Ascones or to the ascon stage of development, or to the ampullae 

 in any sponge. In all of these histology and morphology support 

 the idea that they are lined by the endoderm. In the Ascones the 

 ciliated cells of the endoderm come in contact with the cells of the 

 ectoderm on account of the extreme thinness of the mesenchyme ; 

 and in the thickening of the middle layer which takes place in the 

 more specialized forms, this border line of demarcation is perhaps 

 approximately maintained in the incurrent and excurrent canals. 

 That certain inner portions of the cloacal canals must be consid- 

 ered as lined by an endoderm seems probable from the structure 

 of Sycones and the results of Marshall's and Schultze's work on 

 the development of the central cavity in Reniera and Plakina, 

 which shows that the archenteron is formed by epithelial cells ; 

 but that the outer parts of the cloaca are formed by invagination 

 of the ectoderm as a result of peripheral growth seems also evi- 

 dent. The ampullinula, according to our view, shows that the 

 limit of the extension of the endoderm is the primitive cloaca, 

 as shown by Marshall, and that, contrary to what seems to be his 

 opinion, it may prove to be the limit of its extension outwardly 

 in older forms. There are no facts to show that the endoderm 

 extends into the incurrent tubes beyond the ampullae, or what 



