THE SCIENCE OF LIFE. 



are smooth. In the latter a circulation of fluid may be 

 seen between the cellulose coat and the " primordial 

 utricle." (Chap. IV, Sec. 6.) Some consider this circula- 

 tion to be caused by cilia, but it is rather doubtful. We 

 are inclined to regard it as an exhibition of the molec- 

 ular motion of bioplasm already described. Many of 

 the Desmids multiply by subdivision, but the plan is 

 modified so as to maintain the symmetry characteris- 

 tic of the tribe. At other times multiplication takes 

 place by the subdivision of the endochrome into gran- 

 ular particles, or " gonidia," set free by rupture of the 

 cell-wall. 



The process of conjugation differs from that of Palmo- 

 glaea, since each cell has a firm external envelope, which 

 cannot coalesce with another. In Cosmarium, (Fig. 30,) 



for example, the conjugating 

 cells become deeply cleft 

 and separate, so that the 

 contents pour out freely, at 

 first without a protecting 

 membrane. At length it 

 acquires an envelope, and 

 becomes a sporangium, or 

 spore-case, of reddish-brown 

 tint. It is covered with 

 spines, and greatly resem- 

 . bles certain fossil forms 



ric. 30. Cosmanum botrytts : A. Ma- 

 ture frond. B. Empty frond. C. Transverse found in flint called Xan- 

 view. D. Sporangium, with empty fronds. 



thidia. The Clostena con- 

 jugate after a somewhat similar manner, and it is 

 not uncommon to find a pair in this condition, but 



