No. 3.] NOTES ON DEVELOPMENT OF SOME SPONGES. 517 



flatten out over one pole, while over the rest of the body they 

 acquire cilia. In the metamorphosis, also, the gemmule develop- 

 ment resembles the egg development, in that the ectoderm of 

 the larva is flattened to form the ectoderm of the adult. As 

 regards the formation of the chambers, canals, etc., the egg 

 larvae differ too much among themselves to permit any such 

 comparison as I am carrying out. 



Such a resemblance between the sexual and asexual larvae as 

 I have indicated, can, I think, only be explained on the supposi- 

 tion of some essential likeness between the mesoderm cells, 

 which make up the gemmule, and ova. If for convenience' sake 

 we adopt the point of view of Weismann, and regard the egg 

 cell as characterized by the possession of a "germ plasm," the 

 gemmule cells too (unless the resemblances between the gem- 

 mule and egg embryo are accidental) have some claim to the 

 possession of this plasm. By making a little further use of 

 Weismann's theory, the nature of the gemmule cell (i.e. a meso- 

 derm cell, which, together with others, forms a gemmule), may 

 perhaps be stated with some precision. Of the two polar bodies 

 found in non-parthenogenetic eggs, Weismann believes that the 

 first carries off ovogenetic plasm, the second carries off one-half 

 the germ plasm. In the case of the sponge gemmule, a meso- 

 derm cell does not itself become differentiated into an ovum, 

 but the necessary amount of nutriment for the embryo is got 

 by the association of numerous such cells. Hence there is in 

 the gemmule cell no special histogenetic (ovogenetic) plasm, and 

 consequently no first polar body. There is, of course, no fertili- 

 zation, therefore no second polar body. The gemmule cell, 

 according to this view, must be regarded as a true germ cell, in 

 which all the germ plasm remains undifferentiated, viz. in which 

 none of it is transformed into ovogenetic plasm. Further, the 

 gemmule cell pursues the parthenogenetic course of develop- 

 ment — it keeps all its germ plasm. 



Gemmules apparently develop anywhere in the sponge mesen- 

 chyme. It must, therefore, be assumed that any mesenchyme 

 cell may become a gemmule cell, and consequently that it con- 

 tains germ plasm. The same conclusion is reached by the study 

 of the egg development, for it seems that any mesenchyme cell 

 may develop into an egg. 



The gemmule development in Tedania pursues much the same 



