178 NATURE AND AFFINITIES OF THE SPONGES. 



sac-shaped structural type attributed by Haeckel to the so-called ciliated 

 embryos, the Haeckelian interpretation of the Mctazoic affinities of the 

 sponges has exerted so widespread an influence, and obtained such favour, 

 that every point has been strained on all sides to reduce these reproductive 

 structures, one way or another, to the Metazoic formula. 



Unfortunately for these authorities, however, the one dissentient party 

 to this seemingly plausible and, so far as the sponges are concerned, most 

 honourable correlation is encountered in the very object of their solicitous 

 attention. The ciliated sponge-embryo or, as shown later on, it may be 

 more appropriately denominated the " ciliated sponge-gemmule " stub- 

 bornly resists interpretation as the exact analogue of any one of the various 

 embryonic types prevalent among the Metazoa, and seems indeed to derive 

 a pleasurable satisfaction from the exhibition of a varying type of structure, 

 possessing by turns some shadowy semblance of all, but actually conforming 

 in no single instance to any one of them. Sometimes, in fact, not only in 

 the same order, or in the same family, but in the same genus, in the same 

 species, and even in the same individual sponge-colony, an entire series of 

 diversely constructed reproductive bodies may be met with. 



Abundant illustration is afforded of the more important variations of 

 form and structure found to obtain among the free-swimming sponge- 

 gemmules, by the figures numbered 22 to 36 of PI. IX., derived partly 

 from the author's personal investigation, and partly from the published 

 contributions of the various authorities previously quoted. Examined atten- 

 tively, it will become apparent that this entire series exhibits, with various 

 intermediate gradations, what may be denominated three fundamental 

 plans of structural differentiation. Thus, in the first of these, as shown 

 in Figs. 22, 23, and 36, such plan presents the simplest possible expression, 

 the so-called body-wall of the more or less ovate body consisting of a 

 simple and even layer of columnar cells, each giving origin peripherally 

 to a single elongate cilium or flagellum. In the second series, Figs. 27 

 and 29, an entirely distinct and more complex plan is exhibited. Here, the 

 cellular components of the anterior and posterior regions of the gemmule 

 differ in both size and structure ; those of the former being columnar, and 

 bearing flagella as in those of the first series, while in the latter they 

 are very much larger, usually more or less spheroidal, and entirely devoid 

 of flagellate appendages. The third and highest state of complexity is 

 arrived at in Fig. 30, where a new element is superadded in the form of 

 a central zone of smaller rounded cells, interposed between the anterior 

 columnar and posterior spheroidal series. It is scarcely to be wondered 

 that the energies of talented biologists have been taxed to their uttermost 

 to reconcile such entirely diverse structures with the typical developmental 

 formulae of the Metazoic embryo. By none of these, as yet, can such 

 identification be claimed to have been successfully established ; nor, on 

 examining more closely the very discordant interpretations that have been 

 suggested by different authorities with relation even to gemmules found 



