262 DU. J. B. GATENBY ON THE GERM-CELLS AND 



Page 



27. Discussion 282 



(a) Organ-forming Substances ? 282 



(b) Chromidia and Mitochondria 283 



(c) Fertilization 283 



(d) The Sperm-carrying Cell considered as a Mesogleal Amcebocyte ? . . 284 



(e) Germ-cells and Sex in Sponges 285 



(/) Spermatogenesis in Grantia 286 



{(/) The Collar-cells as the Dominant and most Characteristic Tissue of the 



Sponge 287 



(h) Final Remarks 288 



28. Summary 288 



29. Bibliography 293 



Description op the Plates 294 



1. Introductory. 



The following investigation into the gametogenesis, fertilization, and early 

 development of a common British sponge, Grantia compressa, constitutes the 

 opening of a new field of research in sponge-embryology, for the new 

 methods of cytology have been used with fruitful results, and for the first 

 time. 



Quite recently Professor Arthur Dendy has published a paper in the 

 'Quarterly Journal of Microscopical Science', on "The Gametogenesis of 

 Grantia compressa," in which he has described many interesting facts. 

 Only such problems as those not treated by Professor Dendy have been dealt 

 with here : for a long description of the nuclear phenomena in the germ-cell 

 cycle of a Sycon, Jorgensen's paper (8 a) may be consulted. 



In this paper I am able to announce a remarkable discovery in the earl}' 

 stages of conjugation of the gametes in Grantia compressa and Sycon ciliatum. 



This work was partly carried out at Oxford in Professor Sherrington's 

 laboratory, but mainly at University College, London. 



Some time was also spent at King's College, Strand, London, in going 

 through some of Professor Dendy's preparations. It is a pleasure for me to 

 acknowledge the great help I have derived from various discussions with 

 Professor Dend\ r , who has carried out such valuable investigations on the 

 Porifera. Professor Dendy has followed with much interest this application 

 of the modern technique to sponge-embryology. 



2. General Statement of t]ie Problems investigated. 



It is a well-known fact that the amphiblastula larva? of the Syconidse are 

 formed mainly of two sorts of cells — flagellated and granular. According to 

 previous investigations, the cytoplasm in the two sorts of cells is of different 

 quality ; it was to examine this phenomenon that the present work was 

 undertaken. It was necessary to see whether, during cleavage, there was a 

 special segregation in any blastomeres of granules or organ-forming sub- 

 stances such as occurs in Ascidia (1). In addition, it was planned to study 

 the fate of the middle-piece of the spermatozoon during fertilization. As is 



