136 



sider that the energy which is lost there for the sponge cannot 

 possibly be of much importance, as hardly any water is moved 

 and, besides, the water that is moved circulates for the greater 

 part closed in itself, as I mentioned. 



Let us return to our question. What does it mean when on 

 the one side we see in a Spongilla that only the tops of the 

 flagella protrude outside the collars, and on the other side that 

 sometimes these tops are beating outside the apopyle (Fig. 64)? 

 That means that in such a case 1st the centre of the flagellated 

 chamber with the apopyle has become zone 2, the zone of the 

 flagellar function, so of increasing pressure, and 2 'id that zone 3, 

 that of positive pressure, is removed from the chamber into the 

 excurrent canal. Consequently, it is absolutely excluded that any 

 water might flow from zone 3 into zone 1 ; and the more so, as 

 we know that the apopyle is always much narrower than the 

 chamber itself (I even think to be justified in supposing that it 

 may be able to change its lumen as a sphincter ; which Weltner 

 too mentions (65)). One can also understand, however, what a 

 powerful eff*ect the motion of so many flagella crowded together 

 in an apopyle, and all of them acting in one direction, must 

 have; what a very strong current of water it must cause! 



In the same way as I have described here for the flagellated 

 chambers of the Spongillidae, there will also be accessory arran- 

 gements to be discovered in the chambers of the other sponges, 

 helping to perfect the circulation of the water; of course always 

 by improving the system of the 3 zones of pressure, mentioned 

 on p. 134, and the rapidity of the removal of the water from 

 the 1st into the 2nd and into the 3i"d zone. 



Of course there occur also a number of accessory arrangements 

 outside the flagellated chambers — in other parts of the canal 

 system — for the same purpose. For shortness' sake I will not 

 enter into this question here; and the more so, because it has 

 been treated in extenso by Vosmaer and Pekelharing (62) and 

 my results do not give anything new. 



So we have seen in this chapter that the current of trater through 



