358 Mr. H. J. Carter on the Reproductive 



By " ovibearing " liere is meant the " germs " which this 

 cell contains. I had never seen a genuine sponge ovum at 

 that time. 



" Fig. 42. Form of a sponge-cell which exists in a layer 

 around the young uncapsuled seed-like body." 



By a reference to the illustrations it will be seen that the 

 two cells, so far as their contents go, are remarkably different. 

 And it is worthy of remark that a similar cell to that of 

 "no. 42 " (charged with granules and a small nucleus), also 

 in plurality, forms a capsular layer around each ovum in some 

 if not all the marine sponges — at all events in Darivinella and 

 Dendrilla rosea. 



Now these observations, which were carefully made at a 

 time when I was much more able to pursue such investiga- 

 tions than I am at present, I believe to be correct, and there- 

 fore quite as good now as ever they were. But they do not 

 lead us a bit further back to the origin of the seed-like body 

 than the earliest state in which the ovum has been recognized 

 leads us back to its origin. And even if we could trace back 

 the seed-like body to a single cell no larger than the ovum at 

 this period, the resemblances probably would be so great that 

 it would be in)possible to say whether it would develop into 

 a gemmule or into an embryo. 



Up to this time we neither know what relation the first 

 cell of the seed-like body bears to the genuine ovum, nor 

 whence either ovum, spermatozoon, or gemmule respectively 

 are originally derived. 



If the spongozoon be identical with the free, solitary, flagel- 

 lated infusorium Codosiga^ as first supposed by James-Clark 

 (JVIem. Boston Soc. Nat. Hist. 1867, vol. i. pt. 3, pp. 19 and 

 20, pi. ix, figs. 40-44, pi. x. fig. 64), then, by assuming that 

 Codosiga possesses the organs which produce the elements of 

 sexual reproduction, the like may be conceived of the spongo- 

 zoon, in which case these elements in the Spongida might 

 originate in the same way. But all this remains to be proved, 

 and without proof assumptions are but vague aspirations. 



As regards the comparative size of the matured ovum 

 (that is, the embryo) and the gemmule respectively, I find 

 that this varies almost always with the species, as well as 

 in the same specimen. Thus the largest embryo that 1 have 

 seen in the marine sponges is that of Stelospongus jlabelli- 

 formis^ Cart., one of Mr. Wilson's ovigerous specimens from 

 " Port Phillip Heads," on the south coast of Australia, where 

 it is spherical and one sixth of an inch in diameter; whilst the 

 largest gemmule that I have seen is that of the freshwater 



