412 C. CLIFFORD DOBELL. 



o-iven by Vay (1010). He finds large irregulai' masses of 

 darkly stained substance — which he calls chromatin — in these 

 organisms. He does not appear to be aware that the pro- 

 duction of these forms on coloured media had already been 

 described by AYalker and Murray (1904). 



Sucli, then-, is a very condensed account of tlie chief work 

 which lias hitherto been published concerning the problem of 

 the imcleus in Bacteria. 



In all work in which inadequate technique has been em- 

 ployed — for example, in all studies in which only dried and 

 flame-fixed organisms have been examined — -the conclusions 

 attained can have little value from a cytological point of view. 

 In many publications, moreover, the descriptions both of 

 results and of methods are so meagre as to render discussion 

 of them either unprofitable or impossible. Therefore I shall — ■ 

 on either or both of these grounds — eliminate the followiuo' 

 works from any further discussion : 



Kunstler and Busquet (1897, 1898), Kunstler (1900), 

 Kunstler and Gineste (1906, 1906a), Schottelius (1888), 

 Zettnow (1891, 1899), Protopopoff (1891), Wager (1891, 

 1895), Sjobring (1892), Trambusti and G-aleotti (1892), 

 Ilkewicz (1894), Lowit (1896), Wagner (1898), Ziemaim 

 (1898), Marpmann (1900), Feinberg (1900), Errera (1902), 

 Federowitsch (1902), West and Griffiths (1909), Yay (1910). 



I think most cytologists will agree with me that no profitable 

 discussion of these papers is possible. 



Material axd Methods. 



As I have already indicated above, I have made a special 

 point of working upon the largest forms of Bacteria which I 

 have been able to find; but I have studied in addition a. 

 number of small forms, when they have been suitable. 



Small Bacteria are not only very difficult to investigate on 

 account of the limitations imposed by the microscope, but 

 they are also in many cases unsuitable in other ways lor cyto- 

 logical study. The}' occur frequently in media which render 



