Bacillariece 



269 



This is a true conjugation of aplanogametes with the formation of 

 a zygospore. (Fig. 125 B.) 



(4) Sometimes two frustules approximate and the cell-contents 

 throw off the old valves, but there is no conjugation. The two 

 rounded masses of cell-contents 



lie close together or separated 

 by some of the enveloping jelly, 

 and each develops "independently 

 into an auxospore. 



(5) A pair of Diatoms ap- 

 proximate, but before conjugation 

 the protoplasm of each cell divides 

 into two daughter-cells. Two 

 auxospores are then formed by 

 the fusion of a daughter-cell from 

 each mother-cell with a daughter- 

 cell from the opposite mother- 

 cell. This is known to occur in 

 Amphora ovalis Kiitz., Epithemia 

 Argus (Ehrenb.) Kiitz. 1 , and Na- 

 vicula limosa Kiitz. (Fig. 125 A.) 



A normal auxospore can be 

 regarded as one produced by the 

 conjugation of two cells (or ga- 

 metes), those produced without 

 conjugation being parthenoge- 

 netic. The first and the fourth 

 methods are the ones most fre- 

 quently observed. The most im- 

 portant feature in the formation 

 of an auxospore is the increase in 

 size of the cell. Karsten 2 considers that the majority of Diatoms 

 exhibit undoubted sexuality. 



Castracane and other observers have recorded another method 

 of reproduction by the formation of small spores within the 

 frustules. Kitton 3 and Lock wood 4 have each stated that Diatoms 

 may possess spores (" microspores ") so small as to pass through 



1 Klebahn in Jahrb. fur wissensch. Bot. xxix, 1896. 



- Karsten in Biol. Centralbl. xx, 1900 ; Flora, Ixxvii, 1900. 



3 Kitton in Journ. Quekett Micr. Club, ser. 2, ii, 1885, p. 178. 



4 Lockwood in Journ. New York Micr. Soc. 1886, ii, p. 153; 



Fig. 125. A, Navicula limosa 

 Kiitz., from the New Forest, Hants. 

 ( x 450). B,Achnanthesflexella (Kiitz.) 

 Breb. , from Craig-an-Lochan, Scot- 

 land (x450). C, Navicula Amphis- 

 b<ena Bory, from Barnes Common, 

 Surrey ( x 450). N. viridis Kiitz., 

 from Clougb, Antrim, Ireland ( x 350). 

 C and D illustrate the first method of 

 auxospore formation, B, the third 

 method, and A the fifth method. 



