604 Annals of the South African Museum. 



mined if so, then they disintegrate before or during the liberation 

 of the spermatozoids. As is the case with the daughter colony, so 

 with the sperm globoid, the whole structure when mature escapes 

 into the surrounding water and there assumes an active independent 

 life. But the end result is different, since, whereas the embryo 

 colony consists of both somatic and reproductive cells, the sperm 

 globoid is composed of reproductive cells (the male gametes) only, 

 and these, after escape from the globoid, lead for a brief period an 

 independent flagellate life which must, however, soon terminate in 

 fusion with a female gamete, or they perish. 



From the initial similarity many workers, notably Klein in his 

 later work (18896, p. 47) and Janet (1912, p. 78 ; 1923, p. 55) believe 

 the three types of cells to be homologous, and the sperm bundle 

 homologous to the coenobium, regarding it as a dwarf male colony 

 composed of cells each of which functions as an antheridium, pro- 

 ducing a single spermatozoid. Janet, as is only logical, extends the 

 homology to the oosphere, regarding it as a colony in which cell- 

 division has been partially or entirely suppressed. He considers 

 that either the content of the gynogonidium may round off without 

 division to form a single large oosphere or female gamete, and that 

 this is probably the case in V. aureus (1923, p. 58), or that a certain 

 number of divisions may take place as a result of which one nucleus 

 only persists, and, surrounded by the bulk of the cytoplasm, develops 

 into the oosphere, while the remaining nuclei abort. This he believes 

 to be the case in V. globator, where he found surrounding the oosphere 

 a " follicle " of about 63 chromatic bodies which he regarded as 

 abortive nuclei, the whole corresponding to a colony formed by six 

 successive nuclear divisions, i.e. he holds that in this species there is 

 a " blastea oospherienne " as well as a " blastea spermienne " (1923, 

 p. 57). As against this view it may be remarked that other considera- 

 tions point to V. globator being more highly evolved than V. aureus, 

 and that therefore one would expect to find traces of such a reduced 

 " female colony '" in the latter rather than in the former species. 

 Further, as yet no trace of such nuclei has been found in microtomed 

 sections of either of the two South African species in question, and 

 these are undoubtedly very close to V. globator. On the other 

 hand, however, on several occasions living oospheres of V. Rous- 

 seletii with two or four equal-sized nuclei have been observed as, 

 for example, in material from Rietfontein culture on 28th April 1931 

 (fig. 8, E, F), while in microtomed sections oospores with two nuclei 

 of equal size are not uncommon. But quite different interpretations 



