376 
NATURE 
[June 13. Tore 
was afterwards picked up by Mr. Neave and found 
to have lost one hind wing. This specimen will be 
preserved in the Hope Department of the Oxford 
University Museum. The bird ate four or five 
Lycenide after it had rejected the Acreea. It also 
missed many specimens. All the butterflies were 
swallowed whole. The observations were still being 
continued when the bird was unfortunately disturbed 
by a party of natives. 
The injury seen to have been inflicted on the 
Acreea is of special interest, because such mutilation 
is not uncommonly found in Lepidoptera with warn- 
ing colours. Thus on April 2 Mr. W. A. Lam- 
born, residing in the Lagos district of West Africa, 
found and sent to me a conspicuous Geometrid moth, 
Pitthea famula, from which both the wings on the 
left side had been shorn. Disabling injuries of this 
description are, in my experience, rarely seen in 
species with procryptic colouring, but are character- 
istic of those with warning colours. The facts 
suggest the reasonable inference that a disabled pro- 
cryptic species is devoured, and not rejected. Injuries 
that do not disable—chips out of a single wing or 
symmetrical notches out of both sides, injuries which 
leave the insect with undiminished powers of flight— 
are commonly found in butterflies with all kinds of 
patterns. The amount of this indirect evidence is 
even now large, and it could be obtained in almost 
any quantity if naturalists made a point of seeking 
for it, and did not discard the poorer specimens. I 
will refer to one more example only, and that because 
the species is mentioned by Mr. Banta. In 1897 I 
captured near Chicago a very fresh specimen of 
Limenitis (Basilarchia) archippus with a large piece 
torn out of one hind-wing, an injury that may 
reasonably be explained as the result of an attaclk. 
I do not agree with Mr. Banta’s inference that ‘‘to 
make a good case for mimicry in the sense in which 
that term is ordinarily used, the mimic Basilarchia 
archippus should be tested and found palatable” 
(NaturE, December 21, 1911, p- 243), and that we 
should expect birds ‘scarcely, if at all, [to] molest 
these two forms,” viz. the model and its mimic 
(NatuRE, May 9, 1912, p. 242). In the first place, 
the mimicry is probably Miillerian and not Batesian, 
for the mimic belongs to the genus Limenitis (s.1.), 
containing species which are themselves models for 
mimicry, and allied to the still more widely mimicked 
South American genus Adelpha. Secondly, it is not 
supposed, on any theory of mimicry, that the enemies 
instinctively know the qualities indicated by warning 
colours. This knowledge is believed to be gained by 
each individual enemy as the outcome of its own 
experience. Furthermore, it is probable that in times 
of special scarcity the dangers from the attacks of 
certain enemies would be increased by the presence 
of warning colours. 
It should be noticed that Mr. Banta’s criticism of | 
mimicry applies equally to protective resemblance. If 
his facts and ‘arguments be sound, the rock- or bark- 
like underside of a North American butterfly is as 
useless to its possessor as is, according to Mr. Banta, 
the resemblance of the mimic to its Danaine model. 
It is necessary to make a few remarks upon the 
negative evidence afforded by the examination of 
stomachs. Conclusions in harmony with those of 
Mr. Banta have been reached by Mr. G. L. Bates, 
who has found no traces of butterflies in the stomachs 
of insect-eating birds of the South Cameroon (Ibis, 
ser. g, vol. v., No. 20, October, 1911, pp. 630-1). In 
this case I know that Mr. Bates’s inferences are con- 
sidered erroneous by Mr. C. F. M’ Swynnerton, who 
has for some years been making a special study of 
the relationship between birds and butterflies at 
Chirinda, in south-east Rhodesia. It would not be 
NO. 2224, VOL. 89] 
right to anticipate the results which Mr. Swynnerton, 
in a truly scientific spirit, desires to establish on as 
solid a foundation as possible; but, as the question 
has been raised and dogmatic assertions have been 
made, I feel sure he will not object to the following 
brief statement, prepared in consultation with Mr. 
Guy Marshall :— 
(1) The results of Mr. Swynnerton’s earlier investi- 
gations, up to the end of 1908, were in accordance 
with those of Mr. Bates, and might well have 
justified the conclusions reached by him and by Mr. 
Banta. 
(2) From the time when, three and a half years 
ago, Mr. Swynnerton first saw his way into the 
details of the question and the methods by which to 
investigate it, he has obtained the records of nearly 
800 attacks made by thirty-five species of birds, 
belonging to thirty genera and eighteen families, 
upon seventy-nine species of butterflies, belonging to 
nine families or subfamilies. 
(3) Mr. Swynnerton is thus led to conclude, in 
opposition to Mr. Banta, that the negative evidence 
believed to be supplied by the examination of stomachs 
should not be too implicitly relied upon. The negative 
evidence itself, he considers, may be accounted for in 
various ways :— 
(a) The treatment of butterfly prey by birds. Some 
swallow the insect whole, but usually after masti- 
cating or beating it; some remove inconvenient por- 
tions by “‘worrying”’ like a dog or beating against 
perch or ground; some grasp the prey in one foot 
and tear off the rejected portions with the bill, eating 
the rest piecemeal. Except when the wings are 
swallowed the probability that butterflies will be recog- 
nised in the stomach-contents is extremely remote. 
(b) Insectivorous birds get rid of the chitin of their 
prey partly in a finely divided form in the excreta, 
partly in pellets ejected from time to time by the 
mouth. Mr. Swynnerton believes that he has noted 
the ejection of pellets by every purely insectivorous 
bird kept in captivity. The wings of butterflies that 
were swallowed whole appeared, for the most part, 
both in pellets and excreta, as minute fragments that 
could not be easily recognised except with the micro- 
scope. After a large meal of butterflies, the pellet 
cast up by a captive bird would often consist of fine 
débris, quite unrecognisable except after a thorough 
and minute examination. : 
(c) Other groups of insects, viz. the Diptera, 
Orthoptera, Coleoptera, and Hymenoptera, are each 
of them many times as numerous in individuals as 
the diurnal Lepidoptera, and we should therefore 
expect butterflies to be proportionately far less 
commonly found in the stomachs of insectivorous 
birds. 
Finally, Mr. Swynnerton has found that a recently 
captured adult bird shows by its behaviour that it 
possesses a very fair knowiedge of the main types 
of pattern and the relative edibility of the local butter- 
flies. That this knowledge is the outcome of in- 
dividual experience is proved by the fact that it is 
not possessed by a bird removed from the nest when 
young. Epwarp B. Povutrton. 
Oxford University Museum, June 3. 
The Weather of 1911 and the Ultra-violet Radiations 
of the Sun. 
In connection with an extremely interesting dis- 
cussion recently carried on in the correspondence 
columns of Nature I ventured to direct attention 
(NaturE, December 14, 1911) to the unusual diminu- 
tion of the ultra-violet radiation from the sun as a 
possible cause of the abnormal weather of the summer 
