166 
GEOFF KEY SMITH. 
and hybrid birds has shown that in point of size and histo- 
logical strncture they are indistinguishable ; the 6rst sign 
that anything is wrong appears in the mitotic spindle of the 
first maturation division, where the normal eight synaptic 
chromosomes are not formed. Instead of this, irregular 
Tuasses of chromatin are distributed to the opposite poles 
of the spindle, so that the secondary spermatocytes formed 
as the result of this division do not receive a normal 
heritage of chromosomes. It appears that the disturbance 
caused by this abnormal distribution of the chromosomes 
prevents the second division taking place, the secondary 
spermatocytes, instead of dividing, going straight on to 
form spermatids and spermatozoa. Many of the spermatozoa 
so produced are greatly deformed, as is not unnatural ; 
others, which have presumably received a more normal supply 
of chromatic material, give rise to spermatozoa which are 
normal in appearance, though double the normal size. But 
even those spermatozoa which are normal in appearance are 
impotent in fertilisation, and we must again ascribe this 
to the fact that they do not contain the right numbers or 
kinds of chromosomes. The observations detailed above 
afford strong support to Buyer’s view that the principle 
cause of the sterility of hybrids is the fact that the two sets 
of parental chromosomes, although they may co-operate 
together in the ordinary cells of the hybrid, yet when they 
have to go through the process of synapsis, repel one another or 
form abnormal fusions, with the result that the germ-cells so 
produced, though they may appear normal, are constitution- 
ally abnormal. This explanation of the sterility of hybrids 
can only apply to those hybrids which succeed in producing 
germ-cells of some sort, or at any rate get as far as the 
maturation divisions ; it may fail to explain the complete 
sterility of females, for instance, in which the eggs fail to 
grow or store yolk, since in these cases the abnormality of 
the reproductive cells manifests itself before synapsis occurs.' 
* Tliis difficulty, however, may he more apjiareiit than real, since 
von Jenkinson informs me that a iirecocious synapsis is known to 
