THE NEMATOCYSTS OF EOLIDS 
131 
solve the problem met with better success. A suggestion is all 
that I can offer. The wall of the cyst is tough, and has consider- 
able thickness. It is apparently formed during the period when 
the cnidophage is degenerating. Under the circumstances it is 
not unlikely that this cell has nothing whatever to do with the 
formation of the cyst, but that the membrane in question is trace- 
able to other cells. Grosvenor indeed suggested that the inter- 
stitial cells might aid in the secretion of the cyst; to me it seems 
equally plausible to assume that these cells alone are responsible. 
The best evidence which supports this view is fig. 10, in which are 
Fig. 9 Ingesting cnidophages showing differentiation and shrinkage from cell 
membranes. Also interstitial cells. 
shown degenerating cnidophages, enclosed within compartments 
formed by interstitial cells which themselves have lost practically 
all of their original characteristics. As the figure illustrates they 
are reduced to mere membranes. Since each nettle-sac is com- 
plete, and in the ripe state capable of isolation from its fellows 
we must suppose on this view of the matter that the interstitial 
cell which lies between two or even three cnidophages contributes 
an independent portion to as many cnidocysts. 
However they originate, the cnidocysts in their finished form 
are mere bags, thin-walled and transparent, filled with undis- 
charged nettles. The cysts ma}^ lie free in the lumen of the cnido- 
phore, or loosely attached to its walls. In either case they are 
easily extruded through the cnidophore by contractions of the 
cnidophoral musculature. 
