Elements of the Sjjongida. 359 



sponge Parmula (Sponffilia) Broionii, Bk., from the river 

 Amazon, which averages l-28th inch. Hence the former in 

 size looks like a small pea and the latter like a pin's head. But 

 the size generally of the matured ovum in the marine sponges 

 much exceeds that of the gemmule in the freshwater sponges, 

 while both are equally abundant when present in their respec- 

 tive species ; althougii it is seldom that an ovigerous specimen 

 is found among the former, while I hardly ever found one of 

 the latter without gemmules. 



Generally, too, the ova are largest in their matured forms in 

 the Psammonemata or so-called " Horny Sponges," and when 

 one of the size mentioned, after having been preserved in 

 spirit for some time, is cut in two, its contents present them- 

 selves in the form of a firm, transparent, jelly-like substance 

 charged with granuliferous cells. These cells, which are spheri- 

 cal, may be of two sizes, viz. one, the largest, 20-6000ths inch 

 in diameter, and the other, or small one, from 2 to 3-6000ths. 

 The former consists of a very delicate, transparent, spherical 

 membrane (? effete), filled with the latter, which in their turn are 

 nucleated, more or less opaque, and filled with light-refracting 

 granules. The capsule of the ovum in this condition and size 

 is colourless, translucent, and very tough ; and in the sponge 

 which I have described under the name of Qeelongia vasi- 

 formis it was found in many instances to be budding out from 

 the circumference in three or more places into short processes 

 which, viewed under the microscope, presented all the charac- 

 ters of amber-coloured, horny, laminar, genuine fibre ; but on 

 account of this development having taken place in the body 

 of the parent, I considered it " abnormal," and therefore have 

 described it under this heading (' Annals,' 1885, vol. xv. 

 p. 308). 



I find also a similar development of large and small cells 

 (? mother and daughter) in an advanced state of the ovum in 

 Ilalichondria sanguinea^ and therefore it is probably only the 

 result of yolk-cell division in all. 



But, again, it is worthy of remark that it is closely analogous 

 to the contents of the seed-like body or gemmule (' Annals,' 

 1849, vol. iv. pi. iii. fig. 6, a, i, h)^ wherein the large, trans- 

 parent, membranous, delicate spherical cells are filled with 

 small, compressed, transparent ones of different sizes mixed 

 with a great number of free granules, instead of small, 

 spherical, nucleated cells, rendered more or less opaque by their 

 granular contents, as in the advanced state of yolk-cell divi- 

 sion in the ovum. Indeed the similarity is so great and the 

 subsequent development of sponge-structure, both from the 

 embryo and the gemmule, so much alike, that one is inclined 



25* 



