143 



It goos without saying, after the results obtained by Vosmaer 

 and Pekelharing with carmine-feeding to Spongillae, that I too 

 found the first questionQ,^vm?d\yG[j Q,ri.^yi2VQè.'. The particles fioating 

 hl the ivater are captm'ed in a mass hy the choanocytes of the 

 fayellated chambers; very often their layer is dyed quite red by 

 it (in case of carmine nutrition). It also stands to reason that, 

 since I had stated a mode of motion of the flagella and the water 

 within the flagellated chambers quite different from that described 

 by both the investigators mentioned, also the way in ivhich the 

 choanocytes capture the foocl particles was bound to prove wholly 

 diiferent: those particles are not captured inside the collars at all^ 

 as Vosmaer and Pekelharing thought, hut on the contrary out- 

 side-hetiveen the collars (especially at their base) or betiveen the 

 bodies of the choanocytes themselves. That this must necessarily 

 be the case, immediately appears from Fig. 55, 59 and from Fig. 63, 

 the diagrammatic representation of the water current in a flagel- 

 lated chamber as the resultant of the streamlets produced by 

 each flagellum separately. For the bodies and collars of the choano- 

 cytes must^ so to say, filter the water, circidating between them, 

 free from fioating particles. 



I shall now give a description of the capturing of carmine, as 

 I have been able to observe so many times in my living sponge 

 preparations. So the little sponge is in carmine suspension under 

 oil immersion in an Engelmann case ; one has selected a favourably 

 situated flagellated chamber. 



It is beautifully to be seen how the carmine is captured ! Con- 

 tinually grains run on rather rapidly to the flagellated chamber, 

 carried along by the water in the incurrent canal ; they slip into 

 the prosopyle, but then they are either immediately kept or first 

 they move quickly a little aside into the choanocytic layer, and 

 stick there. On more accurate observation, however, the grains, 

 after entering the prosopyles, prove in most cases to slip through 

 the choanocytic layer, but, when having got to the base of the 

 collars, to suddenly deviate aside and to be soon captured — still 

 at the bases of the collars. Only very seldom a grain penetrates 

 any farther, in the zone of the collars themselves ; as most of 



