NATURE AND AFFINITIES OF THE SPONGES. 193 



extent, the developmental phenomena of the Spongida are productive of 

 composite structures, the swarm-gemmules or so-called ciliated larvae, 

 which bear the closest superficial resemblance to the segmented ovum and 

 primary form and disposition of the component blastomeres of the Metazoic 

 embryo. Penetrating beneath this superficial likeness, however, the points 

 of analogy are found to diverge and vanish altogether. No Metazoic em- 

 bryo, and no Metazoic structure, whatever, is distinguished by the posses- 

 sion of collared flagellate cells, with their attendant properties and functions, 

 as found to exist respectively in these ciliated reproductive bodies, in the 

 essential monadiform constituents of the adult sponge, and among the inde- 

 pendent Discostomatous Flagellata. It is this one important histologic 

 element, the collared cell with its attendant physiologic functions, that 

 so closely unites together the two sections of the Spongida and indepen- 

 dent Discostomata or Choano-Flagellata, while it isolates them at the same 

 time from the members of every other organic group. Were the interstitial 

 canals and chambers of the sponge-stock lined with cells possessing no 

 contractile vesicles, bearing simply flagella, or corresponding with ordinary 

 ciliated epithelium, and yet capable of ingesting solid food-matter, the 

 grounds for removing this organic section into the Metazoic series would 

 apparently be based on a surer foundation and some analogy would be 

 presented with the simpler Hydrozoa and many Turbellaria, where, as 

 shown by Kolliker and Kleinenberg, the endoderm cells lining the ali- 

 mentary canal develop long flagella or pseudopodic processes, and it 

 would appear, engulf food-substances after the manner of Amoeba. 

 Even here, however, the hypothetical analogy would be entirely delu- 

 sive, the matured sponge-stock being the sum total, not of the concourse 

 of sexual elements, but of the essentially Protozoic process of spore- 

 development. 



In the fashioning of the motile ciliated gemmules, or pseud-embryos, 

 upon the plan of the Metazoic morula and amphiblastula, and in the 

 peculiar arrangement and separation of the constituent flagellate and 

 amoeboid factors of the adult sponge, nature would certainly seem to have 

 marshalled her forces preparatory to crossing the border from the Protozoa 

 to the Metazoa, and so far as a transitional group between the two series 

 can be predicated, it is probably realized in the section of the Spongida. 

 The step, as a complete one, however, is by no means accomplished in this 

 group. As is at once made manifest by a closer insight, the sponges 

 remain in every detail of structure, function, and development, typical and 

 thoroughgoing Protozoa. Their position in, and affinities among, the 

 several groups of this sub-kingdom are evidently close to, and inseparable 

 from, that of the naked and independent Discostomata or Choano-Flagel- 

 lata, and which, having due regard to the clearly defined laws of organic 

 evolution, must be recognized as the ancestral progenitors or archetypes of 

 all sponge-forms. The phylogeny or backward passage, again, from these 

 independent collared types to the simpler monadiform Flagellata, is made 



O 



