iv] i GREEN CELLS OF CONVOLUTA 117 



had not succeeded in making the animals biologically 

 clean. It was clear, also, that, if the infecting organism 

 came from the egg-capsule, it might be derived not 

 from the sea- water but from the body of the parent. 

 At the time of hatching, there might be liberated 

 from the body of the animal, colourless or green 

 cells which, though they could not live in sea- water 

 and were not to be cultivated artificially, might well 

 be capable of living in the walls of, or inside, the 

 egg-capsules. 



Fortunately, the elimination of this source of 

 error though laborious is not impossible. When 

 ready to hatch, a very little help suffices to enable 

 the young to escape, not only from the thin mem- 

 brane which encloses each, but also from the common, 

 mucilaginous egg-capsule. Thus, by drawing a clutch 

 into a small pipette and then ejecting it and the water 

 from the pipette, the capsule bursts and the young 

 escape. In this way, it was possible to separate 

 larvse from their capsule-remnants. By employing 

 this method, large numbers of larvse, white, minute 

 and active, were isolated in filtered sea-water in 

 which the only source of infection lay in such in- 

 visible shreds of the capsule as might have happened 

 to get themselves transferred with the larvse to the 

 filtered sea-water. 



The method proved successful. In one case, out of 

 forty-four larvae isolated from capsule-remnants and 



