577 
Constituents of the Cell. 
permanent organs of the cell, increasing only by division. 
The question whether the centrosome or the whole centro- 
sphere is permanent may be left to be answered by future 
researches. It is easy to conceive of the production of 
a surplus of these organs in abnormal cases by one or more 
extra divisions. As to whether the supernumerary ones are 
disposed of by subsequent fusions or by degeneration we have 
no evidence. Strasburger states (’95) that spindles from 
nuclei originally having three or four poles in Pellia finally 
always become bipolar, and is inclined to regard fusion as 
more probable than degeneration. 
In all plant-cells in which it has been recognized, the 
centrosome is a tiny, homogeneous mass, so far as present 
methods of study show. Heidenhain has lately (’94) based 
on his observations upon leucocytes and giant-cells of marrow 
some new and radically different views of the centrosome 
from those above expressed. His discussion is long and 
very confusing to one whose acquaintance is chiefly with 
the definite and characteristic centrospheres of plant-cells. 
But Boveri (’95) has so clearly pointed out his misuse of 
terms and the entire agreement of the facts observed by 
him, when rightly understood, with the conception of this 
organ underlying the earlier work of van Beneden and Boveri, 
which is practically that above expressed, that no further 
discussion of his ideas is necessary, Heidenhain has shown, 
however, that in some leucocytes the centrosome may be so 
large as not to be uniformly stained, but to show deeply- 
staining granules within it. And in other animals similar 
structural features have been observed, as, for instance, in the 
centrosomes of the Sea-Urchin egg, by Boveri (’95). 
The question of the cytoplasmic or nuclear origin of the 
centrosomes, which has been considerably debated and to 
which a good deal of importance has been attached, becomes 
more and more secondary as we realize the extent to which 
interchange between nucleus and cytoplasm occurs, and if 
we regard, with Boveri, the nuclear cavity as merely a space 
set apart for the chromosomes. This writer has done good 
