588 
C. F. U. MEEK. 
spermatogenesis of Salamandra ma.culosa (1896), we see 
in fig. 55, pi. 4, a typical example of this spindle and its 
asters ; and the arrangement of the mantle fibres closely 
resembles that of the cones drawn by Lawson. Now Lawson 
regards the cones as a visible expression of tension, and 
suggests that this tension “ decreases in proportion with the 
distance from the nuclear membrane.’^ And, since the 
primary spindle, as is shown in Meves’ figure, lies between 
centrosomes placed at the apices of the cones, it cannot have 
been formed entirely by the action of forces expressed by the 
cones; for the forces postulated by Lawson, however attenuated 
at these points, are like. But, if the spindle is not formed 
entirely by the action of forces expressed in the cones, to 
what is its formation due ? Lawson’s interpretation does 
not profess to tell us, and cannot tell us. It is therefore 
unnecessary for us to discuss it further. 
We will now consider the older theories of mitosis. These 
may be classified as centrosome theories and fibre theories. 
Let us deal first with centrosome theoides. Meves defined 
these by saying : “ Unter Centrentheorien verstehe ich solche, 
welche die bei der Mitose wirksamen Krafte in die an den 
Spindelpolen befindlichen Gebilde verlegen und die Strah- 
lungeu als die erscheinende Wirkung dieser Krafte ansehen.” 
He pointed out that all such interpretations of spindle forma- 
tion necessitate the assumption of either unlike or like poles ; 
and said in the former case that rays cannot cross, and in the 
latter that no spindle can be produced. With respect to his 
first objection, the adherents of these theories affirm that we 
are dealing here with material chains, and not theoretical 
lines of force, and that crossing of rays is therefore possible. 
We need not, however, consider this possibility; for we have 
found that the assumption of unlike poles is irreconcilable 
with the occurrence of certain mitotic figures, and this alone 
suffices to disprove the interpretation. With respect to his 
objection to the assumption of like poles, we have already 
sliown in our criticism of Gallardo’s theory that, if the forces 
concerned in spindle formation can be expressed by resultants 
