R. (\ PlTNNETT 
325 
$ adults.^ If we suppose the groups of embryos and adults with which we are 
dealing to be fair samples of the whole population, and there seems no valid 
reason for any assumption to the contrary*, then it is to some such process as the 
following that we must look for an explanation. We must suppose that higher 
variability is an attribute of the cT, but that, owing to some subsequent process of 
selection in this sex the variability of adult c^a is brought down to equality with 
that found among adult $ s. Tiiis, however, will not serve to explain the lessened 
variability of $ embryos as compared with $ adults. It is possible that these 
differences are due to unconscious selection of the embryonic material which is the 
product of not more than 40 mothers. On a priori grounds it is therefoie most 
unlikely that this relatively small number of mothers is truly representative of 
motliers as a whole. To test this point the variability of $ offspring of known 
mothers was calculated and compared with the variability of the weighted mothers, 
and the results given in the following table : — - 
TABLE 4. 
cr for ? embryos 
(N=115) 
(T for weighted mothers 
(Actual No. =25) 
Anterior spine 
•6242 
•6665 
Posterior .spine 
•7447 
•6715 
Whole vertebrae 
•9921 
•8192 
Half vertebrae 
1-5920 
r7358 
Total segments 
•990,3 
1 -0495 
Average 
•9987 
•9885 
Hence when there is identical selection of material the average variability 
of $ embryos and $ adults is also identical. It is therefore exceedingly likely 
that the comparatively low variability of $ embryos as a whole as compared 
with $ adults as a whole is due to the fact that the $ parents were here 
unconsciously selected. There is every reason to suppose that such a diminution 
of embryonic variability as the result of unconscious selection obtains equally 
among the cT embryos. The average variability of the embryos should therefore 
be appreciably greater tlian the figure (ri74) already given, thus accentuating the 
phenomenon of greater variability among cT embryos as compared with J' adults. 
In summing up briefly what is to be learnt from a comparison of the variability 
of (/s and $ s, embryos and adults, it may be tentatively stated that the average 
variability of $ embryos and of adults of both sexes is practically identical, whilst 
the variability of embryos is considerably greater. From which follows the 
[* The variability of the weighted mothers is in one case equal to that of the adult ? s. In the 
four other cases it is less, and in three of them — posterior spine, whole vertebrae and half vertebrae — 
very significantly less than that of adult ? s. K. P.] 
