STRUCTURE AND LIFE-HISTORY OF VOLVOX GLOBATOR. 39 



The history of those selected f,'ouidia, as it may be traced in a 

 daufjhter-sphere recently cast forth to seek its fortunes in the world of 

 waters around it, is as follows : — The enlarged gonidium is at first a flat 

 thin circular disk, appressed to the internal surface of the sphere, and 

 being surrounded by ei(jht of the ordinary zoospores, is derived from 

 the coalescence of the two central ones out of a group of ten. (Plate VII., 

 Fig. 1(1 a'.) Shortly, this disk assumes a more distinctly oval form, with 

 a slight constriction across its leaser diameter, in which stage it often 

 much resembles a young Cosmarlum. (Plate VII., Fig. ih b'.) It is soon 

 seen to be clearly sub-divided into four, and its thickness having grown 

 pari paifsu with its superlicies, the group now protrudes into the interna^ 

 cavity of the parent plant. (Plate VII., Fig. k c'.) Repeated sub-division 

 now goes on rapidly, (Plate VII., Fig. 4(7 (Z',) till the whole body assumes 

 a spherical form, a distinct cell wall being at the same time formed, 

 which is revealed by careful illumination, and still more clearly by the 

 use of re-agents, as a hyaline sphere concentric to and of larger 

 diameter than the gi-een one within it, so that there appears to be a 

 clear space or ring between the two when seen in section. (Plate VII., 

 Fig. ie e', and Fig. 6.) Finally, the young Volvox consists of a vast 

 number of deep green gi-anules, closely packed together, and by mutual 

 pressure driven to assume a more or less distinctly hexagonal form, and 

 corresponding in number to the gonidia which are to stud its surface 

 when its growth is completed. (Plate VII., Fig. fxi.) Shortly hereafter. 

 the whole organism continuing to increase in size, clear spaces appear 

 between the gonidia, showtiiri that the enlargement of the cell-wall and itit 

 interspaces is outstripjiinfi that of the gonidia, which are now approaching 

 maturity. The inteiiacing connecting threads are developed simul- 

 taneously. (Plate VII., Fig. la.) During the whole process, the centre 

 of the young Volvox spheres continually i-ecedes from the periphery of 

 the parent, so that when the group of young ones has attained the full 

 development of which it is capable in this stage, they are often pretty 

 closely packed in the internal space, and sometimes even slightly 

 deformed by mutual pressure ; each by this time closely resembling the 

 parent in miniatui-e, and already containing enlarged gonidia of the 

 third generation. (Plate YTL., Fig. 6.) By this time, the clear space 

 originally \isible between the gonidia and the cell-wall has been 

 obliterated, and the cilia may be seen protruding through the latter. 

 Some wi-iters state that the daughter-cells rotate at this j)eriod within 

 the parent cavity. I have frequently seen them oscillate so far in one 

 direction, and then back to their original position, but have never 

 observed a true rotatory motion. Finally, the young Volvoces are 

 liberated by the rupture of the parent sac, at a special point, clearly 

 marked out for this purpose in its structure. (Plate ^11., Fig. C.) I have not 

 met with any observations on this point, but have fully convinced myself 

 that it may always be predicted at what point this rupture will be 

 effected. 



The combined action of the pairs of cilia in whicli the gonidia 

 terminate is the actuating power whence proceed both the rotatory 



