IV] LEUCOSOLENIA 87 



a real living structure, not a cuticular tube, such as the hydro theca 

 of a calyptoblastic hydroid, and this is further illustrated by the 

 fact that it is withdrawn by the collar-cell under certain conditions. 



The water after being exhausted of its food is expelled through 

 the osculum, carrying with it all excreta. 



The outer layer of the body-wall consists, in the ordinary con- 

 dition of the sponge, of flattened cells. These however, especially 

 in the region of the osculum, have the power of changing their 

 shape so as to become shorter and thicker ; in a word they can 

 contract, although they show no trace of the fibrillae found in all 

 muscle and in the muscle tails of the contractile cells of Coelen- 

 terata. The contraction is slow, not quick, as in true muscle. 



FIG. 38. Section of flagellated chamber of Spongilla lacustris, showing the 

 flagella and collar cells x 1500. From Vosmaer. 



1. Nucleus. 2. Vacuole. 3. Opening into the inner space of the sponge. 



It has been proved that the pores are formed by large dermal 

 cells, the porocytes, which extend in from the outer layer and push 

 aside the choanocytes, and then become hollowed out. 



Between the two layers is found a certain amount of secretion 

 which may be termed jelly, in which in many sponges a large 

 number of cells is found. These form a portion of the dermal layer, 

 and are, for the most part, amoeboid. Some of these probably act 

 as carriers of food and possibly of excreta from one layer to the 

 other. Others at first very similar give rise to ova and sperma- 

 tozoa. A third class called scleroblasts derived in Leucosolenia 

 from the flat cells which cover the surface, but not so derived in all 

 sponges secrete the rods which form the skeleton and which are 



