E. Y. Thomson, J. Bell and K. Pearson 
59 
(5) On the Relative Variability of the General Population and of the Nest 
Queens. Relative Values. 
In Table III the coefficients of variation of population and nest queens for the 
26 characters are placed side by side. We see at once that the variability ratios 
are as marked when we treat relative as when we treat absolute variation — the 
TABLE III. 
Coefficients of Variation of Characters for Nest and Population Queens. 
Character 
Population Queen 
Nest Queen 
Ratio P. 
q.\N. Q. 
EightWing 
Left Wing 
Uight Wing 
Left Wing 
Right Wing 
Left Wing 
to 
A 
2-36 + -08 
2-39 + -08 
1-05 + -04 
1-09 +-05 
2-25 
2-19 
2-70 +-09 
2-59 ± -09 
1.36 + -07 
1-46 + -06 
1 '99 
1 '77 
C 
2-66 + -09 
2-72 ±-09 
1-05 + -04 
1-10+ -05 
2-53 
2-47 
D 
3-41 + -12 
3-64 + -13 
1-31 + -06 
1-32 ±-06 
2-60 
2-76 
E 
3-65 ±-13 
3-80+ -13 
2-13 + -09 
1-68 + -07 
1-71 
2-26 
F 
3-27 + -11 
3-43 + 12 
1-45 + -06 
1-46 + -06 
2-26 
2-35 
G 
3-50 ±-12 
3-73 ±-12 
2-65 ±-11 
2-70±-ll 
1-32 
r.38 
Mean 
3-08 
3-19 
1-57 
1-54 
2-09 
2-17 
H 
3-07 + -11 
3-09+ -11 
1-12 + -05 
1-07 + -05 
2-74 
2-89 
01 
I 
2-78+ -10 
2-88 + -10 
1-72 + -07 
1-43 +-06 
1-62 
2-01 
Oj 
_o 
J 
2-83+ -10 
2-47 + -09 
2-01 + -08 
2-04 + -09 
1-41 
1-21 
C 
K 
1-04 +-04 
1-04+ 04 
0-72 + -03 
0-85 + -04 
J -44 
1-22 
L 
1-13 + -04 
1-15 + -04 
0-60 + -03 
0-61 + -02 
1-88 
1-88 
M 
2-12 ±-07 
2-10 ±-07 
1-54 ±-06 
1-47 ±-06 
1-38 
1-42 
Mean 
2-17 
2-12 
1-28 
1-25 
1-75 
1-77 
general results owing to the small changes in the average values of the characters 
as we pass from nest to general population being closely alike. Although the 
coefficients of variation for the general population are for absolute sizes double as 
large as for the single nest, we see that the conclusion of the first wasp paper, 
Biometrika, Vol. v. p. 415, is confirmed. The wing of the queen of Vespa vulgaris 
possesses extraordinarily small variabilit}'. If we deal with indices, even smaller 
values are found than Dr Warren found for his termites, and our population values 
for absolute sizes are comparable with his nest values. As far as our present 
experience reaches, the wiug of the queen wasp appears to be the least variable 
organ that has as yet been studied. We may note that the population queen 
is about as variable as the worker of an individual nest (Biometrika, Vol. v. p. 417). 
Looking at the results now obtained, it is not possible to assert any marked 
difference in variability between the two wings, thus the result of the earlier 
paper (p. 417) is confirmed. 
8—2 
