654. 
many—is clearly explained. ‘This multiplicity of 
patterns is the very thing which assists the effici- 
ency of the leaf imitation, since the faded, dried, 
half-rotten leaf occurs in nature in a thousand 
forms and colours, with its transition from green 
to yellow and brown, its slits and jags, its traces 
of gnawing and of mould and fungi” (p. 37). 
And the objection, sometimes raised by those who 
have not sufficiently considered the subject, that 
a Kallima may be seen resting with expanded 
wings on green foliage is also effectively answered 
on p. 39. 
For Warning Colours and in other parts of the 
book ‘the author accepts the terminology intro- 
duced in this country in 1890. Haase’s- term, 
“Immunity” is wisely used only in a restricted 
sense. The unqualified word, carrying with it 
the assumption that the bearers of warning colours 
are exempt from all attack, even by parasitic foes 
—an assumption carefully guarded against on 
Pp. 52—gives an entirely mistaken impression. 
That such insects have their special enemies has 
now been shown by many observers. A good 
example is the highly conspicuous Acraea setes, 
of which Dr. G. D. H. Carpenter collected in 
Uganda seventy pupz and full-fed larva, but only 
reared sixteen butterflies. All the others—77 per 
cent. of the whole—were destroyed by parasitic 
insects. 
The theory of aposematic or warning colours 
is considered to stand on a much firmer foundation 
than that of cryptic colouring (p. 50); but the 
author, accepting the conclusions published in the 
_ Proceedings of the Zoological Society in 1887, 
recognises the intimate relationship between the 
two. ‘‘Aposematic species restrict the food avail- 
able for insect-eaters ” and must therefore “ pass 
on to other non-protected species the onus of 
satisfying the hunger of their foes. Now, if these, 
by any process of development, also attained im- 
munity, the foe would be compelled to overcome 
his disgust, and accept the disagreeable food, and 
thus the advantage of the warning colour as an 
advertisement would be reversed, for it would 
facilitate the discovery of the prey” (p. 52). 
The treatment of terrifying markings is incon- 
sistent. They are ridiculed on p. 23, but taken 
seriously on pp. 56-59. It must be freely admitted 
that markings which make so strong an appeal 
to the imagination require to be tested and re- 
tested by carefully observing their effect upon 
enemies, before the bionomic meaning can be 
accepted as proved. This can scarcely be claimed 
at present for any examples except the terrifying 
Sphinx larve, the objects of superstitious fear by 
man in different countries, and proved by four 
observers to excite fear in animals. This, the 
NO. 231I, VOL. 92] 
NATURE — 
[FEBRUARY 12, I914 
clearest example, is doubted by the author, 
although he accepts the far more problematical 
interpretation of the markings and attitude of the 
eyed hawkmoth (Smerinthus ocellatus) as terrify- 
ing. Such an interpretation is probably correct, — 
but before accepting it we require at least as 
much evidence as has been collected for the larve. 
In the historical account of mimicry a common 
error is repeated. H. W. Bates himself, im his 
classical memoir (Trans. Linn. Soc., vol. xxiii., 
1862, p. 495), grouped together the phenomena 
of mimicry and protective resemblance, and did 
not, as stated on pp. 60, 61, understand the 
former “as referring only to similarity in form 
and colour between creatures of different sys- 
tematic position.” . 
It is well-nigh impossible to get rid of an error 
of this kind when once it has been fairly started. 
However, we must do our best. Bates, on pp. 
508-10 of his paper, quotes numerous examples 
of procryptic resemblance to twigs, bark, lichen, 
the excrement of birds and caterpillars, dewdrops, 
&c., concluding with the words, on p. 510: “I 
think it will be conceded that all these various 
kinds of imitative resemblances belong to the same 
class of phenomena and are subject to the same 
explanation. The fact of one species mimicking 
an inanimate object, and another of an allied 
genus a living insect of another family, sufficiently 
proves this.” A footnote on pp. 508-9 is even 
more conclusive; for the actual term “mimicry ” 
is applied to the procryptic examples. Referring 
to Réssler’s interpretation of the buff-tip moth in — 
the resting attitude, Bates adds in a note: “In an 
article on resemblances between insects and 
vegetable substances (Wiener Entomol. Monat- 
schrift, 1861, p. 164), the author enumerates many 
very singular cases of mimicry; he also states his 
belief that the mimicry is intended to protect the 
insects from their enemies.” The convenient re- 
striction of the term mimicry to the resemblances 
to other specially defended animals—the models— 
came later, and is due to Wallace. 
Returning to the author’s section on mimicry, 
we notice a simple and convenient device for 
representing the mimetic association between two 
species, the names being connected by an arrow 
pointing in the direction of the model. 
In the sub-section on ‘‘Mimicry among 
Batrachians”” there is an interesting footnote on 
Pp. 75, Suggesting the specific identity of the con- 
spicuous, distasteful amphibian, which, as the 
author says, “hops about in all Darwinian litera- 
ture as ‘ Belt’s Frog.’” The species, he thinks, 
“can be nothing but Atelopus varius, which is ex- 
tremely common in Central America.” My friend, 
Mr. G. A. Boulenger, however, does not entirely 
. 
