258 BOTANICAL GAZETTE [ APRIL 
the fusion of the polar nuclei in Alisma was the fact that just 
before the contact of the nuclei the centrospheres appeared a 
little farther apart as though they were separating. The inter- 
pretation in such cases, however, necessarily must be merely an 
inference, and could not be decided absolutely except by observ- 
ing the phenomena in the living condition, which at present 
seems entirely out of the question in the case of the higher 
plants, unless a system of differential centrosphere stains can be 
developed. The whole subject must depend upon the question 
as to whether centrospheres are permanent bodies in the cell and 
deserve to rank with other cell organs. Although our knowl- 
edge in regard to these bodies is still too fragmentary to make 
any positive assertions, the permanent organ theory certainly 
seems to be receiving constantly new confirmation.. Recently 
Lauterborn+ has studied these bodies in certain diatoms, in which 
he was able to see centrosomes very readily, even in the living 
condition. This is a direct confirmation of the work done by 
Bitschli,5 who made the same observation in 1891. As far as 
my own observation goes, I can come to no other conclusion. 
These bodies appear so often that they can not be overlooked 
by a careful observer. They appear beside the resting nucleus, 
in the higher plants always two in number, and later at the poles 
of the spindle in the mother star stage. Ai little later, having 
divided into two, two centrospheres appear at each pole of 
the spindle, sometimes so prominent that they are as much in 
evidence as the nucleus itself (fig. 34). There is no reason 
why at times, especially during abnormal conditions, the centro- 
spheres should not fragment or break up into a number of pieces, 
just as is the case with the nucleus, but any objections raised as. 
_ to the individuality of centrospheres because of such action can 
have no more weight than a similar argument against the indi- 
: viduality ot the nucleus because it frequently fragments or dis- 
solves. It makes little difference what the function of the 
fiber Bau, Kernteilung und Bewegung der Diatomeen. Aus” 
BS elber; . 2 
2 ee Naturhist “Med. Vereins 2 za deers. N. F 6 535538 — 
o: 
