no THE RADIOLARIA 



most complex Phaeodaria this shell acquires a bivalvular form and 

 carries many peculiar processes (Fig. 32). 



The nucleus is a large, usually single structure, and undergoes 

 a peculiar kind of mitosis accompanied by the formation of a great 

 number of chromosomes. The development and nature of the 

 spores is incompletely known. A characteristic feature of this 

 order is the absence of the yellow cells that occur almost constantly 

 in the other orders. This negative feature appears to be correlated 

 with the presence of that remarkable and still imperfectly analysed 

 complex, the phaeodium. The researches of Borgert (18) give 

 some ground for thinking that the phaeodellae (see p. 119) are 

 excreta, and if so, the retention of these substances in Eadiolaria 

 devoid of "yellow cells" lends support to the view, derived from 

 a study of the Turbellaria (Keeble and Gamble [41]), that these 

 symbiotic algae exert a depuratory function. 



Variation : Dimorphism. The Kadiolaria present three kinds 

 of structural modification. There is the divergent variation about 

 one or more centres that constitutes a " species." There is racial 

 somatic dimorphism in relation to pelagic or abyssal life. And 

 there is gametic dimorphism both in early and adult stages of life 

 in relation to reproduction. 



The conception of " species " in Radiolaria is only gradually 

 assuming a form similar to that held in the case of other Protozoa. 

 Hitherto skeletal characters have been mainly and rigidly employed 

 for the erection of a vast number of specific forms. The larger 

 collections made by Plankton expeditions of recent years have 

 shown that many of these earlier species, and even genera, are 

 either growth stages of one and the same form, fission products 

 common to several species, or divergent variations referable to a 

 central " type." The first kind of variation probably occurs in 

 every Radiolarian and has been recently worked out for several 

 Tripylaria (Immermann). In Auloldeptes flosculus, for example, 

 spicules of three kinds can be met with, each one of which was the 

 basis of a separate species in Haeckel's classification. It has been 

 shown, however, by Immermann that the spicules pass through two 

 or more forms before arriving at their definitive stage, and may be 

 arrested at an intermediate stage. Further knowledge of the 

 development of the skeleton will undoubtedly tend to diminish the 

 profusion of species that Haeckel has proposed. But it is not 

 skeletal characters only that are subject to change during growth. 

 Among the Collodaria, in which the spicules are a subordinate feature 

 and in some families entirely absent, the early stages of growth 

 differ so greatly from the later ones as to render their identification 

 a difficult matter and one particularly liable to misinterpretation. 

 Thus the genus Actissa, which Haeckel brought forward as the 

 most primitive of all Kadiolaria, has been shown by Brandt (25) to 



