144 



them have been captured in advance, either between the bodies 

 of the choanocytes or between the bases of the collars. The 

 movement of the flagclla is always the rapid spiral- or undulating- 

 motion, which was mentioned above as the normal one; the col- 

 lars are normal, far extended. 



These different ways of capturing carmine grains are represented 

 in Fig. 65^ 66. In both figures the choanocytic layer is drawn 

 for convenience's sake as a broad circle ; and the way taken by 

 the grains is indicated by dots; while in the last figure — drawn 

 from nature — the choanocytic layer has been represented as 

 loaded with carmine. 



After all I have said and drawn about the structure and the 

 water-current of the flagellated chambers (p. 132 — 134, Fig. 55, 

 59, 63), the here described way of capturing carmine between 

 the choanocytes — by which the water circulating in a chamber 

 is so to say filtered — is quite intelligible. I only want to point 

 out that the phenomenon, that the carmine grains generally 

 immediately pass the choanocytic layer, but then suddenly 

 deviate aside along the base of the collars to be kept there, 

 can be explained by the fact that in a living chamber on the 

 one side the bodies of the choanocytes are shorter and bigger 

 — which makes the open spaces between the separate cell 

 bodies much smaller — than has been represented in Fig. 55 and 

 in the diagrammatic Fig. 63, while on the other side the 

 collars are approaching each other more and more in the direction 

 of the centre of the chamber. In other words: the flowing 

 water will find the widest passage in a chamber exactly at the 

 bases of the collars (compare Fig. 59). Therefore, however, 

 one should not think that zone of the bases to be extremely 

 unfit for capturing floating particles ; that only depends upon 

 the relative size of the latter. The carmine grains now are \ — 

 1 |Ct, the symbiotic algae 2 — 3 /^; thus so small that they can 

 just pass the prosopyles, while they will stick between the bases 

 of the collars. The more so, as we may suppose that the collar 

 cells, for the advancement of that purpose, will be provided with 

 a sticky, mucous surface, as is also accepted for protozoa. 



