THE BIONOMICS OF OONVOLUTA ROSCOFFENSIS. 385 



the '^giit/' the greeu cells iu the pei'ipheral parenchyma, 

 which is their definitive station. 



This change, however, does not immediately affect all the 

 colourless cells. A certain number remain unaltered, moving" 

 towards the hinder end of the body. Immature Convoluta, 

 •74 mm. in length, contain posteriorly placed colourless cells, 

 and some may be found in this position even in adult 

 specimens. New supplies of green cells may arise from these 

 colourless elements as well as by division of already 

 pigmented cells. Hence it follows that a knowledge of the 

 origin and development of these colourless cells is the key to 

 the problem of the green cells (see PI. 31, figs. 19, 20). 



The most feasible way of attaining a knowledge of the 

 real problem and its solution appeared to be to experiment 

 with Convoluta on the supposition that the colourless cells 

 must be either intrinsic or extrinsic. We have seen that, at 

 hatching, the larva is free from any trace of these cells or 

 their future contents. By allowing such larva3 to develop in 

 sterilised water, the problem of infection or non-infection 

 could, we thought, be solved. 



Accordingly we prepared vessels filled with ordinary sea 

 water, sea water sterilised at 60° C. and 80° C, and filtered 

 through a Pasteur-Chamberland filter. In some of the 

 vessels, ordinai'y water free from Convoluta, but rich in 

 diatoms, was given as food ; in others sterilised sand ; and, 

 again, in others no food was introduced. At first we employed 

 egg-capsules; afterwards just-hatched larvas were placed in 

 the vessels. During the experiment we subjected the animals 

 to various light conditions. Equivalent lots were kept iu 

 diffuse light, in darkness, and iu bright daylight. 



The results of this first series of experiments showed great 

 irregularity in the appearance of colourless cells; but coloui-- 

 less cells appeared, in some animals of all the cultures — whether 

 in unfiltered, filtered, or sterilised water; whether iu light or 

 darkness; and whether with sand or without. Not only did 

 colourless cells appear in isolated larvae or in larva3 of a batch 

 reared from the capsule, but when exposed to bright light 



VOL. 47, PART 3. — NEW SERIES. BB 



