KARYOKINESIS AND ITS RELATION TO FERTILIZATION. 193 
declared absent, too, by Overlach (152) in the epithelium of the 
cervix uteri, in spite of active nuclear increase ; and by Nissen 
(145) in the epithelial cells of mammary glands. Berggriin (25) 
found numerous amitotic nuclear divisions in the Tadpole’s tail 
and in the Frog’s cornea after mechanical irritation of the part 
concerned. Carnoy (47) finds direct as well as indirect divi- 
sion in various Arthropodan tissues. Division of an Infusorian 
(Euplotes harpa, Stn.), without the appearance of the mitotic 
figure in the nucleus, is described by K. Mobius (139), although 
we know true mitotic nuclear division in other Protozoa—see 
the examples given above. Especially remarkable are the 
numerous discoveries recently contributed on Spermatogenesis. 
A. Bolles Lee (82, 38) found amitotic division in the first 
generation of sperm-forming cells (Samenbildungszellen), the 
so-called “‘ spermatogonia” of La Valette St. George, whereas 
in the following generations regular mitotic division was 
recognised. Dostojewski showed the same state of things in 
his observations of the sperm-formation in Amphibia, carried 
on in this Anatomical Institute. Similar observations have 
been made by La Valette St. George (121, 122), by Gilson 
(79), by Sabatier (175), and others. Only Platner (161) estab- 
lished mitotic division exclusively in Pulmonata and Lepi- 
doptera ; of course deviations from the normal scheme would 
be expected. 
At the present time nothing precise can be said as to how 
we should explain these differences, or in what way generally 
we should regard the relation of mitotic to amitotic division. 
There is no lack of statements on this point. Thus Pfitzner 
(155) and EK. Zacharias are of opinion that amitotic division 
occurs only in those cells which must be regarded as gradually 
undergoing degeneration. The observations of Boveri (35), 
given in detail below, appear to be of the greatest interest in 
this respect. He finds that amongst the segmentation spheres 
of Ascaris megalocephala there is always a limited 
number which during division exhibit chromatin loops dis- 
tinctly, whilst in the majority of the segmentation spheres 
these loops are not easily visible, and the chromatin appears 
