12 A Third Cooperative Study of " Vespa Vulgaris" 
TABLE V. Correlations of Right and Left Wings in Nest and Population Queens. 
Spring 
Population 
On pon 
Autumn 
Population 
Nest Queen 
Batio 
Aut. P. Q./N.Q. 
Ratio 
Sp. P. Q./N, Q. 
Ratio 
Aut. P. Q./Sp. P.Q. 
A 
■954 ± -004 
"989 + '001 
■gy3 + .QJ2 
I'll 
i'0.7 
ro4 
B 
■911 + -008 
•824+ -022 
•831+ -018 
■99 
1-10 
•91 
C 
•964 + -004 
•9V7 + -003 
•858+ -016 
1-14 
1-12 
1-01 
D 
■862 + -013 
•883+ -015 
•600+ -038 
1-47 
1-44 
1-02 
E 
•694 + -026 
•762 ± -028 
•428+ "049 
1-78 
1-62 
1-10 
F 
•932 + -006 
•905+ -012 
•605 + -038 
1-50 
1-54 
•97 
G 
•856 ± -013 
•796 ± -025 
i787+ -023 
1-01 
1-09 
•93 
Mean 
■882 
•877 
■714 
1-29 
1-28 
1-00 
queens. In the case of correlations of parts of the same wing, the raising process 
is apparently more emphasised than in the case of homologous parts of the two 
wings. But the equality of correlations in these parts before and after selection, 
where we should expect a reduction, strongly suggests the same process as we have 
indicated in the correlation of parts of the same wing, i.e. the existence of an 
autumn to spring selection which tends to strengthen the proportionality of the 
parts of the wings, a feature as important as selection about type. That such a 
selection really exists is evidenced by the manner in which the variability of the 
indices of the wing parts has been reduced by the autumn to spring selection 
(see Table III). 
(8) Conclusions. 
An examination of autumn queens indicates that : 
(i) The chief result flowing from our second study, i.e. the remarkable 
reduction of variability within the nest, has no relation to the action of natural 
selection on the queens surviving to spring. 
(ii) There is some evidence to suggest that it is a result of heredity and not 
of nest environment. 
(iii) There is strong evidence to indicate that natural selection is actually at 
work reducing the deviations from type and increasing the proportionality of the 
parts of the wings of queen wasps between autumn and spring. 
We are fully aware that the importance of the latter conclusion is so great 
that it needs ampler treatment, but we have further material in hand and hope to 
publish within a reasonable time reductions for larger numbers. The chief difficulty 
is the collection in autumn of a sufficient population of queens from different 
nests within a reasonably circumscribed area*. 
* We are already in possession of many thousand spring queens. But Is. is not an unreasonable 
price to pay for the taking without damage of a complete nest in the autumn, and at least 500 nests 
have to be taken from one district and in one season to obtain a really adequate sample of autumn 
queens. 
