250 
C. F. U. MEEK. 
that chemistry will provide the solution of problems that so 
far we have failed to solve. That eventually we must hand 
over to others problems that are found to lie beyond the scope 
of our investigations cannot be denied ; but, until we have 
amassed data such that we can answer every question con- 
cerning the morphology and movements of structures that 
are visible in the cell, we are not justified in relying upon 
other branches of science for more than co-operation in our 
endeavour to explain these phenomena. 
In a paper upon chromosome dimensions published last 
year I was able to show that increasing somatic complexity 
of the organism is accompanied by increase of the volume of 
chromatin in its germ-cell, and that the diameter of chromo- 
somes becomes greater as we pass from low to higher phyla 
of the animal kingdom. I now propose to carry out similar 
investigations upon the phenomenon known as the mitotic 
spindle, and shall try to discover whether at a given moment 
in a given mitosis the length is constant or arbitrary ; if it is 
found to be constant, we must determine the relationships 
between lengths at corresponding stages of successive mitoses, 
and must ask if these relationships are connected with those 
of other phenomena. 
The stage that is most easy to identify in mitosis is the 
conclusion of the metaphase, when the chromosomes have 
undergone complete fission, and the daughter-rods, apposed to 
one another, are ready to move towards the two poles. I 
shall therefore make consideration of this stage the basis of 
my research, and shall determine the length of the spindle by 
measuring the distance between the centrosomes. I have 
chosen Forficula for these investigations, because the chromo- 
somes are either spheres or very short rods ; it is accordingly 
easy to recognise the moment at the conclusion of the 
metaphase when constriction is completed. Since the two 
centrosomes can seldom be brought into focus simultaneously at 
the highest magnifications, figures in which the spindle length 
is to be measured will be represented in the plates by two 
drawings : the first will show a lateral view of the equatorial 
