668 
Notes. 
though observation does not yet confirm this, and the regularly oval 
shape of the nucleus militates distinctly against it, it is not otherwise 
a priori improbable that a third centrosphere might be differentiated 
just outside the nucleolar mass referred to in par. 3, but in this case 
it must be excessively fleeting. 
(6) The spindle next becomes longer by the wider separation of 
the centrospheres, and at the same time it loses its ‘lop-sided’ 
appearance. The division of the chromosomes conforms to the type 
of the first mitosis in the pollen-mother-cell of the Lily. 
(7) After the separation of the elements of the daughter-nuclei, 
a cell-plate is formed, which does not extend to the peripheral walls. 
It subsequently shrinks somewhat, but remains suspended transversely 
within the cell. 
(8) The two daughter-nuclei, after almost going into a condition of 
rest, divide with the formation of wide-angled spindles. The ‘ lopsided ’ 
arrangement of the chromosomes is however seldom met with, but when 
present, radiations also extend from the aggregation of chromosomes. 
(9) The axis of these second division-spindles is variable, but from 
the axis of that pole which happens to lie nearest the primary cell- 
plate above alluded to, radiations are observed to branch off, and to 
connect with it. Thus the two spindle-poles which lie on opposite 
sides of the plate act as a ‘ couple,’ and they cause it to rotate through 
an angle of 90°, so that it now lies longitudinally in the cell. During the 
rotation, the spindle-poles shift somewhat, and the fibres connecting 
them with the plate form the fifth spindle referred to on page 493. 
This rotation is perfectly constant, as is the position taken up by the 
four daughter-nuclei (see PL XVIII, Fig. 65). The formation of the 
new cell-plates, and their mode of attachment to the first one, have 
already been described. The whole process is susceptible of an 
obvious mathematical interpretation ; soap-films introduced into a box, 
corresponding in shape to that of the spore-mother-cell, take up 
precisely similar positions. The arrangement of the spindle-fibres, 
and especially their mode of attachment to the cellulose framework, 
are in precise conformity with what, on mechanical considerations, 
might have been anticipated. 
(10) The cells which form the elaters separate from each other by 
means of a mucilaginous degeneration which affects the entire wall 
except the innermost layers around each cell-cavity. 
J. BRETLAND FARMER. 
