148 PLANT-ANIMALS [OH. 



to appear in their bodies, the animals suffer. It may 

 be that the death of uninfected animals is not merely 

 the consequence of starvation, but is at all events 

 hastened by poisoning due to the accumulation in 

 the tissues of the products of nitrogenous metabolism. 

 According to this view, uninfected C. roscoifensis dies 

 as the consequence of an aggravated attack of " uric 

 acid trouble." 



Evidence is not lacking in support of this some- 

 what fantastic suggestion. Thus, if larval C. ros- 

 coffensis are protected from infection and kept without 

 food, as their large store of reserve food-material 

 derived from that contained in the eggs, disappears, 

 numerous vacuoles charged with long, acicular, 

 crystalline bodies make their appearance in the 

 tissues. The vacuoles and crystals increase in 

 numbers till they present a most striking appear- 

 ance. These crystals represent, in all probability, 

 the waste products of nitrogen-metabolism. 



Now, in infected animals, the crystalline bodies 

 do not occur, and if animals in which they are present 

 are caused to become infected by the green algal 

 cells, the crystals disappear as fast as the green cells 

 develop. Whence we may infer that the materials 

 of which the crystalline bodies consist are used for 

 the nutrition of the green cells. 



The evidence which C. roscoffensis provides in 

 favour of our hypothesis is, of course, but slender. 



