16 JOUJINAL OF THE [January,, 



as those of the filamentous and flabellate diatoms, in which there 

 is no absolutely necessary connection between one cell and an- 

 other, and where there is no known reason why any single frus- 

 tule would not go on in the performance of all essential functions- 

 if it were entirely separated from its sister cells, we seem to be 

 quite justified in regarding the type as physiologically unicellular, 

 even though the anatomical units are not always physically soli- 

 tary or free. There is at least no evidence that those diatoms- 

 which are found within a thallus (as already mentioned) have 

 anything more than a mechanical connection with one another, 

 and vital independence is after all the best test we have of their 

 actual individuality. 



For some as yet unexplained reason, multiplication by fission 

 sooner or later results in an exhaustion of the vital powers, which 

 would lead to a rupture of the line of descent if it were not for 

 the introduction at this point of the phenomena of conjugation, 

 which, in some mysterious way, give fresh impetus to the repro- 

 ductive energy. This new departure is brought about by the 

 merging of two organisms into one and the formation from their 

 combined substance of an enlarged and reinvigorated mother- 

 cell, called a zygospore, which becomes the progenitor of a new- 

 family, descending from it by the original process of self-division. 

 But the exact nature and office of the zygospore are enveloped 

 in great uncertainty, which has been deepened rather than eluci- 

 dated by a good deal of hasty inference which has been put upon 

 record as established fact. Thus, it has been asserted that two 

 zygospores sometimes arise from a single union, and also that 

 conjugation results in the production of a " sporangial frustule," 

 which undergoes in itself a segmentation which ends in the for- 

 mation of true spores, which are set free by the rupture of the 

 containing envelope and which then establish large numbers of 

 new centres of diatom-life. These and other reputed phases of ' 

 the reproductive process rest upon observations which are more 

 than doubtful, but even if they shall be at last accepted as true, 

 the entire matter will still be found to be reducible to modes of 

 gemmation, either continuous or discontinuous. 



As has been indicated already, the medium through which the 

 vital forces work out their wonderful effects is the internal proto- 

 plasmic substance of the diatom, ot*herwise known as the endo- 



