RA ae ee a eee ee 
1897 } THE LIFE HISTORY OF SAGITTARIA VARIABILIS 259 
centrosphere is finally discovered to be; the presence of the 
body must be explained. That its function may have been misin- 
terpreted is no argument against its existence. The centrosome 
may be a mere insertion point for spindle threads and cyto- 
plasmic radiations, as Heidenhain seems to suppose; it may be 
the special organ of division and a truly directive sphere; it may 
be even more, and have some function in transmitting hereditary 
characteristics ; but whatever its function may be, the point to 
be decided first is its existence. If this is established, questions 
as to its origin, purpose, and permanency naturally follow. 
The appearance of two centrospheres at the poles of the 
spindle (figs. 3, 34), it seems to me, cannot be explained by a 
crossing of cytoplasmic filaments; by an attraction from the 
periphery to a common center; or by the rather lately 
broached idea of a sort of whirlpool in the cytoplasm. So far 
as the writer is able to judge, no one has attempted to offer a 
satisfactory explanation of these double centers at the poles on 
the theory of their temporary nature, since he called attention 
to them in 1894.° At this stage these bodies can be identified 
readily, and there is no danger of mistaking other granules of 
the cytoplasm for centrospheres. 
In this connection I wish to refer to Humphrey’s’ implied 
criticism of my former work on centrospheres. He intimates 
that, having largely used Hermann’s method of staining centro- 
Spheres, I may have mistaken various proteid granules in the 
cytoplasm for true centrospheres. My only reply to this is that 
I was fully aware of the fact that many small bodies in plant cells 
often greatly interfere with the identification of centrosomes, 
and, therefore, a large number of stains and methods of killing 
were used in order to eliminate any faulty observations which =~ : 
might be possible in using only a single method of preparation. 
It must be recognized that methods of fixing and staining do 
not give the same results when used by different observers. : 
: au, The Baare st cnstaton of tection phar se centromere ee 
_ 70m some constituent of the cell. Ann. Bot. 9: 574 1895. 
