78 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
Even in the resting stage, the centrosome is apparently double (fig. 
35, c). But in most cases this is not evident (pi. 13, fig. 36, c; pi. 16, 
fig. 90). No changes can, as yet, be discovered in the nucleus. 
Prophases of mitosis .— The two centrosomes separate. As yet, 
there is no evidence of an aster, and there is no spindle uniting the 
two centrosomes as they separate (pi. 13, fig. 35, d). Each centro¬ 
some moves around the nucleus through an angle of ninety degrees. 
The original position of the centrosome is usually in the long axis 
of the cell, the axis vertical to the tangent at the cell’s base or cyst 
membrane. Consequently the line joining the two centrosomes when 
they have reached their destination at opposite poles of the nucleus is 
at right angles to the vertical or long axis of the cell. 
Shortly before arriving at their respective poles (pi. 13, fig. 36, e , /) 
or immediately afterward, an aster is developed around each centro¬ 
some (pi. 16, fig. 92). 
The aster .— Usually both asters are developed at the same time, 
but a case has been seen where one aster was well developed while 
the other was not (pi. 13, fig. 37, a, c), the centrosome without an aster 
having apparently not yet reached its destination. 
The aster is developed entirely at the expense of the cytoreticulum, 
the entire cytoreticulum being involved (pi. 16, fig. 92). All fibers 
of the cytoreticulum converge at the two opposite poles of the nucleus, 
from which point they radiate in straight lines to the very periphery 
of the cell. 
Owing to the constant eccentricity of the nucleus, near the inner 
free pole of the cell, the asters are necessarily unsymmetrical, i. e., 
the rays on one side are very much longer than those on the opposite 
side (pi. 16, fig. 92; pi. 13, fig. 36, d, e, /). The archoplasm theory of 
the astral rays finds no support here. There is no evidence of a gradual 
outgrowth from the centrosome or spheres of the astral rays. The 
microsomes of which they are composed seem smaller than those of 
the former cytoreticulum. 
The study of these asters inclines one more and more to favor the 
view of an actual splitting of the original cytoreticulum and cytomi- 
crosomes. There is a very evident crossing of the astral rays in the 
larger proximal pole of the cell (pi. 16, figs. 92, 93, 94, 95, 96, 97). 
But there may be a connection at the periphery of the cell between 
the long fibers of one aster and a corresponding short fiber of the other. 
This might produce the crossed appearance, and would not necessa- 
