50 



The strongest advocate for the existence of Sollas's raerabrane 

 is Dandy. His statement that flagella and the membrane coexist 

 is quite new. This however he describes only in one species viz. 

 Halichondria panicea. 



On PI. V, fig. 1 of his paper on that sponge Dendy gives an 

 illustration , »which is a careful drawing of an actual preparation". 

 We see there the » flagella plainly, projecting from the bodies of 

 the collared cells through the collars and into the cavity of the 

 chamber". In looking closely at the figure we observe first that 

 not all the flagella communicate with the protoplasma oftheir(?) 

 cells, and secondly that although there are but ten choanocytes, 

 eleven flagella are drawn. In another part of the sarae figure 

 we observe eight choanocytes and only two flagella ; none of the 

 latter communicates with the protoplasma of its (?) cell. Our 

 supposition is therefore that those flagella , seen and drawn by 

 Dendy, are partly fragments , belonging to choanocytes not taken 

 in the section. If we omit this very doubtful case, all authors. 

 Dendy not excepted , agree in insisting on the fact that Sollas's 

 membrane and flagella do not coexist. So e. g. Sollas himself, 

 in speaking of Tetilla grandis (Chall. Rep. p. 12): — »Thefene- 

 strated membrane is preserved in some of the chambers only, and 

 when present the flagella are not visible." We think this rather 

 ominous indeed. As we are still convinced that Sollas's membrane 

 is an artificial product , we beheve that the lines drawn by Dendy 

 indicatiug Sollas's membrane are portions of flagella. With this 

 suggestion another statement of Dendy, fully agrees viz that 

 those lines appear » as a thicker line , than the outlines of the 

 collars is". The author explains this by supposing that it is »due 

 to the sections causing us to see a little more than the mere cut edge 

 of the membrane". We lay stress on the fact that in our prepa- 

 rations where the supposed membrane was visible, we saw the 

 line in question thicker, exactly as Dendy describes. But we can by 

 no means agree with the author's explanation. By careful focussing 

 it was , on the contrary, proved that we had not to do with a mem- 

 brane in optical section but with a more or less cylindrical body. 



