155 



stands in a distant relation to the problom, wc are trcating just 

 now, tlio ingestion of food. 



Which of the two methods of capturing food, the one with 

 the choanocytes or the one with the plasmic layer, is the most 

 important one for the sponge, will depend, in my opinion, simply 

 on the size of the food-particles present. If the size is small the 

 l8t, if it is larger, then the 2"'! method preponderates, With 

 carmine nutrition the capturing by choanocytes was the chief one. 



Finally one might ask, if the sponge can capture food in still 

 more ways. I must answer that I think it quite possible. In the 

 first place I think of capturing particles in the incurrent canals 

 themselves, which show all sorts of irregular lumina, while fine 

 plasmic bridges, sieve-like membranes and what not, are extended 

 in them, so that they have plenty of opportunity of capturing. 

 Also ingestion of food at the outer-surface of the sponge seems 

 possible. I have, however, never observed these ways of capturing. 

 One thing would prove against them, viz. that with a carmine 

 nutrition of short duration there is hardly ever carmine to be 

 found anywhere else in the sponge than just in and at a short 

 distance from flagellated chambers, as I often stated. But I have 

 also made observations which are in favour of* them. 



In a living preparation there were numerous Flagellata moving 

 quickly within the canals and at the outside of the sponge. Such 

 organisms, now, some together and sometimes with carmine 

 grains, were also moving within small vacuoles in the canal-walls 

 or in the tissue at the outer-surface. It is quite impossible that 

 these living organisms have been captured by the choanocytes or 

 the plasmic layer; for then they would have been killed, at least 

 they would be motionless. They are more likely to have been 

 captured after having arrived in a blind ending part of the canal, 

 the latter partly narrowing — for the tissue is very plastic, the 

 canals arise and disappear while one observes them — and closing, 

 when lts size had been reduced to that of a vacuole ; while finally 



