286 



MICROSCOPIC FORMS OF VEGETABLE LIFE. 



tous Desmidiacece, 170, and usually but not invariably with the 

 Zygnemacece, 199), conjugate with each other; and this may 

 take place even before they have been completely separated by 

 self-division. Thus in Meloseira ( 188) and its allies, the endo- 

 chrome of particular frustules, after separating as if for the 

 formation of a pair of new cells, moves back from the extremi- 

 ties towards the centre, rapidly increasing in quantity and aggre- 

 gating into a sporangial mass (Fig. 83, 2 , , 5, c\ and around 



FIG. 83. 



Self-Conjugation of Aulacoseira crenulata: 1, simple filament; 2, filament developing sporangia: 

 o, b, c, successive stages in the formation of sporangia; 3. embryonic frustules, in successive stages; 

 a, 6, c, of multiplication. 



this a new envelope is developed, which may or may not resem- 

 ble that of the ordinary frustules, but which remains in conti- 

 nuity with them, giving rise to a strange inequality in the size of 

 the different parts of the filaments (Figs. 97, 98). Of the subse- 

 quent history of the sporangial frustule, much remains to be 

 learned ; and it is probably not the same in all cases. It has 

 been already shown that the sporangial frustule, even where it 

 precisely resembles its parent in form and marking, greatly 

 exceeds it in size ; and this excess seems to render it improbable 

 that it should reproduce the race by ordinary self-division. 

 Appearances have been seen, which make it probable that the 

 contents of each sporangial frustule break up into "gonidia;" 

 and that it is from these that the new generation originates. 

 These gonidia, if each be surrounded (as in many other cases) 

 by a distinct cyst, may remain undeveloped for a considerable 

 period ; and they must augment considerably in size, before they 

 attain the dimensions of the parent frustule. It is in this stage 

 of the process, that the modifying influence of external agencies 

 is most likely to exert its effects ; and it may be easily conceived 

 that (as in higher Plants and Animals) this influence may give 

 rise to various diversities among the respective individuals of 

 the same brood ; which diversities (as we have seen) are trans- 

 mitted to all the repetitions of each, that are produced by the 



