462 C. CLIFFOKl) IXJEKLL, 



and uetworks (B. saccobraiichi) . There is evidence to 

 show that a nucleus in all these three forms may occur at 

 different times in the same organism (B. saccobranchi) . 

 There is also evidence that spherical nuclei, filamentar nuclei, 

 and chromidial nuclei may occur in the same organism at 

 different stages in its life-history (Bacilli of modified 

 flexilis form from Triton and Lacerta). 



(6) Spirillar forms which I have studied show three 

 different types of nucleus : tlie chromidial (Sp. monosporn, 

 etc.); the filamentar (Spirilla m from Lacerta); and the 

 spherical type (small Spirillum from Sty lopyga), which 

 divides by amitosis, and resembles the nucleus of Coccus 

 forms. 



(7) "Fusiform Bacteria" possess a single, usually 

 spherical, nucleus in each cell. 



(8) A number of large, parasitic, non-motile, rod-like 

 organisms, possessing a vesicular nucleus, which, appear at 

 first sight to be Bacteria, are really Fungi allied to the yeasts. 



General Discussiok. 



Now that I have briel3y reviewed the more important 

 literature bearing upon the cytology of the Bacteria, and 

 have given my own observations in some detail, I am in a 

 position to discuss my results. My miiin object, as I have 

 already pointed out, has been to decide the question, whether 

 or not the Bactei-ia are nucleate cells. The chief part of this 

 discussion will therefore be directed towards answering tliis 

 question. 



As I have already indicated, many of the observations 

 which have been made by others upon the cytology of the 

 Bacteria, are based upon material which has been so imper- 

 fectly fixed and stained that it is useless to consider them. 

 Of the researches reviewed in the " Historic " section (p. -^99), 

 therefore, only a part can be profitably considered here. 

 Furthermore, it is impossible to enter into a minute discus- 

 sion of many excellent contributions to the subject — extending, 



