130 THE RADIOLARIA 



tion of this statement reference may be made to the Phaeodaria. 

 This group of Radiolaria possesses no zooxanthellae, and might 

 therefore be expected to show some accumulation of excretory 

 granules. This appears to be the case, for the complex phaeodium 

 is made up of refractory, insoluble vesicles which are generally 

 held to be excretory substances. The association of diatoms with 

 Radiolaria has probably a similar significance. 



Finally, when the endoplasm sporulates the dying ectoplasm 

 serves as a medium in which the zooxanthellae rapidly divide and 

 issue as naked biflagellated spores upon a new, free existence. In 

 the case of the Acantharia, which are also mainly epiplanktonic 

 or surface organisms, the zooxanthellae are naked cells, almost 

 exclusively confined to the central portion of the Radiolarian. 

 Whether they develop from antecedent zooxanthellae that occur 

 in the spores of Acantharia or infect it from sea-water, or whether 

 both modes of origin obtain, is at present unknown. The apparent 

 absence of xanthellae from young Acantharia makes the first sugges- 

 tion unlikely. Within the central capsule they divide, multiply, 

 and assimilate. Certain of them fragment into particles, the 

 process being initiated by nuclear fragmentation, so that the 

 zooxanthellae are no longer cells but mere chromatised, pigmented 

 corpuscles, associated with free granules of starch or amyloid 

 substances. There is no evidence to show whether in this or in 

 the earlier coherent stage the xanthellae are digested by the 

 Acantharian. They become in the last event mere assimilative 

 corpuscles, and when the endoplasm sporulates the whole of the 

 zooxanthellae, with their associated starch, pass into the bodies of 

 the flagellated spores, and are probably used up as food by the 

 developing zygote. Throughout this series we see that, in opposi- 

 tion to the idea of mutual benefit, the animal is the predominant 

 partner. The association is one beginning with myxophytism and 

 leading to a case of parasitism, in which the zooxanthellae are the 

 host and the Radiolarian the parasite. 



Skeleton. The skeleton of the Radiolaria has developed in each 

 of the great sub-classes into a complexity of form and variety of 

 detail that are found in no other group of animals. So characteristic 

 are the skeletal products that it is usually possible from them alone 

 to recognise broadly the systematic position of the organism that 

 produced them. So complex and diverse a tracery seems utterly 

 beyond the needs of simple Protozoa living under apparently similar 

 conditions of pelagic life ; and though attempts have been made to 

 explain this manifold skeletal development in terms of cytoplasmic 

 structure, its variety still evades a biological treatment. Recent 

 investigation has, however, done something to reduce this variety 

 to a few plans, and to attach a biological meaning to some of its 

 elaborations. These results justify the hope that, as we come to 



