THE EMBRYOLOGY OF THE PORIFERA. 223 



nor the other divide at all actively, mitoses or dividing nuclei 

 being of exceptional occurrence. Finally, Delage admits, 

 and shows plainly in his figures, that the globules of the 

 groupes polynuclees are very different from the nuclei of the 

 free ciliees before and after being devoured ; and, further, 

 that when, as in Esperia, all the ciliees are not devoured, 

 their nuclei retain their normal character in the syncytium, 

 while they appear strongly modified in the amoeboid cells. 

 He declares, however, that he has found every transition 

 between the opaque, strongly stained globules, and the 

 much larger, oval nuclei of the ciliees, with their membrane 

 and granular contents. He also figures pseudopodial pro- 

 cesses of the amoeboid cells reaching to, and apparently 

 capturing, cells of the ciliated layer. 



In opposition to the arguments of Delage, Maas urges 

 that Delage's methods are not adequate to determine the 

 existence of such a process of phagocytosis, since he has 

 never observed it directly, but only inferred its occurrence 

 from a comparison of sections. In objects of such excessive 

 minuteness as the cells of the larvae of siliceous sponges 

 it is often impossible to make out clearly whether the nuclei 

 of the ciliees be in or on the amoeboid cells. Moreover, 

 Maas was able to repeat in Esperia the observation he 

 formerly made in Spongilla, namely, that after double stain- 

 ing with borax carmine and malachite green, the latter 

 stain colours the granules in the amoeboid cells, but not the 

 nuclei of the flagellated cells. Delage does not seem to 

 have succeeded with this reaction. Finally, Maas draws 

 attention to the difference in the degree to which this pro- 

 cess is described as occurring in different species, and the 

 fact that, according to Delage himself, a number of free 

 flagellated cells are to be met with in all stages of the 

 metamorphosis. Maas is willing to admit that flagellated 

 cells may occasionally be devoured by amoeboid cells, but 

 regards this as pathological. 



A decision on this question can, of course, only be 

 obtained by investigation of the objects. On the whole, 

 however, Delage's statements with regard to the phago- 

 cytosis and the formation of the syncytium do not seem 



