170 



detritus within a vacuole — in simple to very irregularly branched 

 amoeboid cells, which lodged a nucleus but few or no symbiotic 

 algae and for which, in general, the same description counts, 

 which I gave on p. 153 for the there isolated amoeboid cells. 

 So these cells have, in short, the appearance of, what I called 

 in my living preparations, the apparently undifferentiated plasmic 

 substance. One also sees the striking conformity with what was 

 found on p. 161 — 164. Consequently this apparently undifferentiated 

 plasma, which, all over the wall of the (excurrent) canals proved 

 to bring about the defecation and excretion, belongs to amoeboid 

 cells. That this also counts for the plasma, which also performs 

 the function of defecation but then near an apopyle (p. 166 — 167), 

 was not yet quite decided by this, but I have been able to prove 

 it in the same way in another preparation. ' 



Perhaps one is inclined — just as on p. 154 — to look upon 

 these amoeboid cells in the canal walls as pinacocytes, But for 

 the same reason as I mentioned there, I believe to be justified 

 also here in taking these cells rather as belonging to the paren- 

 chyma. I will return to this subject afterwards. 



Next one could put the questions: 1-^*. If, in fact, there is in 

 principle any diiference between these two kinds of amoeboid 

 cells — situated in the living sponge as apparently undifferentiated 

 plasma in the walls of the excurrent canals, the one however 

 spread all over, the other only near to the flagellated chambers — , 

 both of which bring about the carmine defecation. 2^^. If, in fact, 

 there exists in principle any difference between these two kinds 

 of amoeboid cells (those of defecation) and the amoeboid cells, 

 mentioned on p. 153 — 154, which — situated in the living sponge 

 also as apparently undifferentiated plasma in the walls of the 

 incurrent canals — lodge the large grains of carmine (p. 149 — 

 150). I suppose we shall have to answer these questions nega- 

 tively ; we shall rather have to consider all these amoeboid cells 

 as identical, each sort only temporarily performing another function 

 (cnf. p. 149, Ic; 161 — 163). It would even be quite possible that 

 a cell, which has first taken a number of carmine grains in an 



