56 
Society of London. The misfortune lay in the circum- 
stance that the entomologists of that time were un- 
prepared for new ideas, and the writer had accordingly 
to incur the opprobrium of an innovator. He has 
happily survived this treatment, but how far any advance- 
ment has been made by entomologists since the year 
1879 may be gathered from the discussion of the whole 
subject which was raised in the Entomological Society in 
1897, and of which a summary is given by Prof. E. B. 
Poulton in the paper now under consideration. So far as 
the writer of this notice is concerned, the first gleam of 
encouragement came from Dr. Alfred Russel Wallace, 
who, with his well-known power of mental penetration, 
had no sooner had the case submitted to his judgment 
than he accepted the new doctrine, and incorporated it 
in his book on “ Darwinism.” From the discussion of 
1897 it appears that the majority of our entomologists are 
still hostile to the Miillerian theory ; but conspicuous 
among those who have helped to support and develop it 
is the author of the paper now before us. We may claim 
also Dr. F. A. Dixey, of Oxford ; Mr. Roland Trimen, 
one of the early pioneers with Bates and Wallace in the 
subject of mimicry ; Colonel Swinhoe ; Dr. A. G. Mayer, 
of America; Mr. Gahan, of the British Museum, and 
some few others, as co-heretics in this later development 
of the theory of mimicry. 
The original theory propounded by Bates in 1861 
is so well known, and has been so frequently dis- 
cussed in these columns, that it is unnecessary to 
restate it. The fundamental condition is that the 
imitated form should be objectionable to insectivorous 
enemies, while the mimic should not be protected 
by any distasteful qualities. The Miillerian theory, 
briefly stated, is that two or more species belonging to 
distasteful groups will derive benefit from mimetic re- 
semblance because, although immune as compared with 
non-protected species, they are not altogether exempt 
from persecution, and the loss in individuals incurred 
by each mimetic species becomes proportionally more 
and more diminished the larger the number of indivduals 
over which the loss is distributed. Thus the resemblance 
being advantageous can be conceived to have been 
brought about by natural selection in the Miillerian 
mimicry in precisely the same way that it has been 
conceived to have been brought about in the Batesian 
mimicry. Whether it has actually been so brought 
about, is just the point about which there has been so 
much discussion ; but if natural selection plays any part 
at all in species formation—and the writer still finds 
himself in the position of being without any other 
adequate theory—then a perusal of Prof. Poulton’s 
paper, and the powerful arguments which he has mar- 
shalled therein, cannot fail to convince the unprejudiced 
naturalist that if natural selection was valid for Bates it 
is equally valid for Miller, and, further, that if natural 
selection is inadequate in either or both cases, then we 
have no theory of mimicry that will at all bear critical 
examination, and the whole body of facts remain as 
inexplicable as in pre-Darwinian times. 
The method adopted by Prof. Poulton in the present 
paper is that of exclusion. He discusses all the alter- 
native explanations which have been suggested, and finds 
them to be untenable when submitted to close analysis. 
There is thus left only the theory of natural selection 
The competing theories are all resolvable into three—viz. 
(1) external action of environment (2) independent 
development along similar lines by internal causes, 
and (3) psychical influence of predominating types of 
colour and pattern leading to the sexual selection of that 
type. The latter theoryis not verylikelyto survive, although 
Mr. Darwin in 1872 wrote to the writer of this notice: “I 
do not feel at all sure that this view is as incredible as it 
may at firstappear.” It should be added that the said 
NO. 1542, VOL. 60] 
NATURE 
[May 18, 1899 
suggestion also came from Fritz Miiller in a letter to 
Darwin. Rejecting for the present No. 3, the author 
deals at length with Nos. 1 and 2. Before marshalling 
the facts it is, however, considered necessary to insist 
that the resemblances herein dealt with are part and 
parcel of the general phenomenon of Protective Resem- 
blance. This point is strangely put into the background, 
or altogether ignored, by the upholders of non-Darwinian 
theories of mimicry, and Prof. Poulton has done good 
service in bringing it well to the front again. Itis sur- 
prising that the two sets of facts, viz. resemblance to 
environment and resemblance to other living species, 
should be dissociated, in spite of the circumstance that 
Bates and Wallace and most writers on the subject since 
have distinctly recognised the fundamental importance of 
grouping them together. It is, of course, inconvenient 
to the opponents of the Darwinian explanation to admit 
that resemblances to bark, leaves, twigs, &c., which are 
so well explained by natural selection, should be of the 
same order as a set of resemblances for which that 
explanation is regarded as indequate. And even if it is 
allowed that protective resemblance and the old 
(Batesian) mimicry are due to natural selection—as some 
of the speakers seemed to admit in the discussion of 
1897—the extension to the newer (Miillerian) mimicry is 
opposed by either ignoring or denying the facts, or by 
substituting untenable theories. 
The original theory of Miller was limited in its 
application to certain butterfles (/¢twma and Thyridia) 
which were not very remote in their kinship, but in which 
the superficial resemblance was too exact to admit of ex- 
planation by blood-relationship alone. In 1882 the 
writer of this notice, in a paper published in the Azmals 
and Magazine of Natural History, ventured upon an 
extension of the Miillerian principle to whole groups of 
related and “protected” species in which a_ general 
similarity in the type of pattern and colouring prevails. 
The idea was that the abstract type of marking became 
associated with a knowledge of inedibility in the mind of 
insectivorous enemies. Five years later (Proc. Zool. Soc., 
1887), the author of the paper now before us made a 
further advance by extending the Millerian principle to 
large groups of insects quite unrelated by affinity, and 
belonging, in fact, to different orders. Itis only necessary 
to bring together an assemblage of species belonging to 
different orders, and having a general superficial resem- 
blance among themselves, to constitute a presumptive case 
of “ Miullerian association.” If it can be shown that 
this group of species is for one reason or another more 
or less exempt from persecution as compared with non- 
protected species, the case would at once become 
Miillerian as distinguished from Batesian. It is, of 
course, doubtful in many cases to which class a particular 
example of mimicry may belong. The result of the 
recent work of Poulton, Dixey and Mayer is to make 
it appear probable that the Miillerian principle is of more 
widespread importance in nature than the older principle 
of Bates. 
Since the superficial resemblance of insects belonging 
to distinct orders, such as a moth to a wasp or beetle, a 
beetle to an ant, and so forth, cannot have been aided at 
the outset by blood-relationship, the result in all cases 
where the association is Miillerian, whether attributed to 
natural selection or to any other cause, can only have 
been brought about by a process of convergence. The 
essence of the Miillerian principle also is that the so- 
called protected species are subject to a certain per- 
centage of extinction, and the resemblance which we now 
find among them is accordingly advantageous, in the same 
sense that a distasteful caterpillar is gorgeously coloured 
according to Wallace’s well-known theory. For this 
reason Prof. Poulton prefers to limit the term mimicry to 
the Batesian principle ; the Miillerian cases are described 
