D.—ZOOLOGY 89 
After cell division has been in progress for a very short period the cells 
which are formed by an egg of a sea-urchin begin to show a marked 
difference in arrangement from those of a polychet worm. At the end of 
the third cleavage cycle, the cleavage pattern of a sea-urchin is seen to be 
orthoradial—the cleavage furrows between the upper quartet of cells lie 
immediately over the furrows of the lower quartet. In the polychet, 
however, the arrangement is spiral, not orthoradial, for the furrows of the 
first quartet of smaller cells lie between the furrows of the basal quadrant 
cells. By experimental means we can force the sea-urchin egg to 
divide in a way characteristic of the worm. ‘This is done by increasing 
the centripetal force which tends to press one cell against another, and we 
can show that the arrangement in the polychzt worm is that assumed by 
a system of spheres so arranged as to pack together within a minimum 
volume. The arrangement in the polychet is essentially the same as in 
the egg of the mollusc or polyclad turbellarian. What conclusions can 
we draw? The classical interpretation associates the similarity in the 
cleavage pattern with a common phylogenetic relationship. From an 
experimental point of view one is inclined to a totally different view—viz. 
that the similarity in form is due to a similarity in the intensity of the 
mechanical forces operating on the cells. In the worm, mollusc, or tur- 
bellarian the centripetal pressure acting on the cells is sufficient to force 
the cells to occupy a form in which a maximum volume is enclosed by a 
minimum area of surface. In the sea-urchin this is not the case. The 
pattern as such plays no essential réle in determining the fate of the egg. 
A spirally cleaving sea-urchin egg develops normally ; it does not develop 
into a worm or mollusc. The mechanical view is peculiarly attractive, 
but it has one serious objection. When the dividing cells of a molluscan 
egg rotate so as to reduce their centripetal pressure to a minimum, a rota- 
tion to the left is as effective as a rotation to the right—and on each occasion 
one would expect an equal number of rotations to the left as to the right. 
In a few cases this seems to occur, but in others the left-handed or right- 
handed pattern appears to be due to determinate and not to chance forces— 
for at any given stage of cleavage all the eggs show a rotation to the right 
or to the left. That this phenomenon is correlated with mechanical 
asymmetry is quite probable, and it may be that the nature of this 
asymmetry will eventually be observed. In the meantime, however, we 
seem to be faced with the fact that a mechanical condition which is satis- 
fiable in either one of two ways, is, in fact, only effected in one way. 
Does it not look as though a disturbance has occurred in the probability 
values of the system ? It is as though we were presented with a bag of 
black and white balls—and each time we pick out the black balls and reject 
‘the whites. Before we attribute a determinate behaviour to the cleaving 
egg-cell we must, of course, make certain that the chance of left- or right- 
handed cleavage is mechanically of equal probability. Up to the present 
we can only say that no mechanical difference is apparent—and in the 
absence of such definite evidence we are free to interpret the facts either as 
evidence of a deficiency in our knowledge of the mechanics of the system, 
or to the possibility that there exists in the egg a potentiality which makes 
certain events more probable than they could be in inanimate systems. 
One is tempted to suggest that the cells of a molluscan egg turn one way 
