Volvox in South Africa. 585 



i.e. when denting was first appearing. Of these the largest measured 

 274 x 257 /x (nearly always before inversion the polar diameter is the 

 shorter). Another measured 164 x 164 fj. before inversion, 125 x 133 [JL 

 immediately after inversion. Yet another, 125 x 121 ^ in the dent- 

 ing stage, measured 116 x 125 p, on completion of inversion. 



B. Male Initial Cell, or Antheridium Mother-cell. 



In the case of the development of the male initial cell, F. Rous- 

 seletii is a far more favourable object for investigation than F. 

 globator or even than F. capensis. The very large number of sperm 

 bundles found in one colony, and the fact that they develop at very 

 diverse rates, makes it possible to find every stage of development, 

 from young, still ciliated mother-cells to escaping sperm globoids, 

 in one and the same colony. Further, the course of development 

 is extraordinarily constant it is very rare indeed in a healthy 

 vigorous strain to find sperm globoids that vary in either structure 

 or number of sperms. When such do occur it is almost invariably 

 in material which shows other signs of abnormality ; for example, 

 material which has been subjected to unfavourable conditions, has 

 been kept for some time in culture, or is nearing the end of its 

 period of sexual activity. 



In both V. Rousseletii and V. capensis the mature sperm globoids 

 are similar in appearance and structure, but there are several inter- 

 esting divergences in the course of their development. In both, the 

 life of the antheridium mother-cell and the sperm globoid which 

 develops within it can be divided into distinct periods, viz. : (1) 

 period of enlargement, (2) period of cell-division, (3) period of re- 

 orientation, including preparation for inversion and inversion proper, 

 (4) maturation of sperm globoid within the parent, (5) escape from 

 the parent, (6) life subsequent to escape culminating in (7) escape of 

 the sperms from the globoid. 



Much of the life-history is common to both species. The develop- 

 ment and structure will be described in detail in the case of V. Rous- 

 seletii, partly because it was in this form that it was first studied, 

 and partly because it is undoubtedly the most favourable for observa- 

 tion. Then the points in which F. capensis differs from it will be 

 considered. 



It was endeavoured to obtain as full a photographic record as 



possible of the successive stages. It was of course impossible to do 



this for one individual androgonidium, but photos have been taken 



showing most of the early stages in development (not the later 



VOL. xvi, PART 3. 38 



