MITOSIS 



103 



klitzkei, which, according to him, takes place by typical mitosis with 

 extranuclear centrosomes, states that he has been able to demonstrate 

 simihxr centrosomes in four species of Hartmannella. The centrosome, 

 which does not originate from an intranuclear centriole, is easily overlooked, 

 and requires very special technique for its demonstration (Fig. 60). In 

 the resting nucleus it lies against the outer surface of the nuclear mem- 

 brane, and when division commences it divides into two daughter centro- 

 somes, which take up positions at the poles of the spindle. It is evident, 

 in the light of these observations, that the granule described as a centriole 



a ^. 



J I 



I ! 



Fig. 60. — Stages in the Nuclear Division of Hartmannella Iditzlcei to snow 

 THE Presence of the Extranuclear Centrosome as revealed by Mann's 

 Stain ( x 2,500). (After Arndt, 1924.) 



by Hartmann and Chagas cannot be a centrosome, and that the cases of 

 mitosis which have been recorded as taking place without centrosomes 

 require reinvestigation. 



Another type of nuclear division which is distinct from those described 

 above occurs in amoebae belonging to the genera Vahlkampfia and Dima- 

 stigamoeha (Fig. 61). The resting nucleus has a large central karyosome 

 and peripheral chromatin in the form of fine granules within the nuclear 

 membrane. The nuclear membrane persists throughout nuclear division, 

 during which the karyosome becomes elongate and then dumb-bell-shaped, 

 and finally constricted into two daughter karyosomes. These may 

 remain connected by a fibre or centrodesmose, which in some cases can 

 be seen to unite two granules which are embedded in the dense daughter 

 karyosomes. Between the two karyosomes and surrounding the centro- 



