676 



NATURE 



[October 31, 1907 



each paying the whole — a consideration which, I think, 

 will go home to most of us. And what is true of a com- 

 bination of two species will be equally true of a larger 

 assemblage ; the greater number of forms that can be 

 got to share the tax, the better for all. Hence the form- 

 ation of these large " inedible associations," or, as they 

 might be called, Miillerian groups. I do not wish to be 

 understood as saying that the Batesian and Miillerian 

 theories are mutually incompatible. They are supple- 

 mentary to each other, and there is ample room for true 

 mimicry beside or within the ranks of the Miillerian 

 associations. 



Though the theory of which I have just given an 

 account is really quite simple, it has never been so 

 j<enerally understood and appreciated as that of Bates. 

 May I, at the risk of being tedious, try to illustrate the 

 relation between the two? 



Imagine a large box of sugar-plums, and a schoolboy 

 given carte hianche to help himself from it as he likes. 

 Imagine, further, that the sugar-plums are of different 

 colours and flavours, and that some of them are flavoured 

 with an essence which the boy does not like — we will say 

 aniseed. Further, let all the aniseed sugar-plums be 

 coloured pink. The boy will soon find out that the pink 

 sugar-plums are unpleasant to his taste, and after a trial 

 or two they will be left until all the others have been 

 disposed of, or, if sufficiently disagreeable, they will be 

 refused altoofether. The pink colour is here an aposeme, 

 to use Prof. Poulton's term, or the visible mark of a 

 distasteful character. 



Suppose a few pleasantly flavoured sugar-plums to be 

 coloured pink like the aniseed sugar-plums. These, if 

 there are not so manv of them as to destroy the impression 

 of nastiness associated with pink, will also be left. This 

 represents Batesian mimicry. The few pleasantly flavoured 

 sugar-plums share in the protection afforded by the pink 

 aposeme. 



Now for Miillerian mimicry. Let us suppose that there 

 are two flavours disliked by the boy, sav aniseed and 

 peppermint, and that the sugar-plums with these flavours 

 are coloured pink and green respectively. The bov would 

 'have to try both pink and green before he learned to 

 avoid them. Perhaps two of each, two of the pink and 

 two of the green, i.e. four in all, would bo sufficient to 

 complete his education in this respect ; biit if both kinds 

 •of disagreeable sugar-plums were coloured pink, a trial 

 ■of two only, instead of four, would be sufficient to protect 

 till the rest, of both flavours, aniseed and peppermint, 

 from the boy's depredations. In other words, the tax 

 paid by each would be halved, and so with larger numbers. 

 Hence the advantage of a common aposeme for dis- 

 tasteful objects, whether sugar-plums or buttei flies. 



This illustration refers only to the relation between the 

 two theories. It says nothing, of course, as to the means 

 "by which the sugar-plums originally became coloured and 

 flavoured ; but what we have done is to show the 

 advantage to be gained bv Miillerian association, and 

 therefore to supply the required motive power for natural 

 selection. 



Miiller's suggestion was brought to the notice of British 

 naturalists by Prof. Meldola in the year of its first publi- 

 cation, and in its further developments at the hands of 

 Meldola himself and of Poulton it was accepted both bv 

 Wallace and by Trimen, the two naturalists who had 

 (lono most bv their own observations to confirm the 

 validity of the supplementary (though earlier-devised) 

 theory of Bates. 



Fritz Miiller had spoken chiefly of the resemblance 

 between two butterflies, Ttuna and Thyridia, belonging 

 'n distinrt subfamilies, but it was soon pointed out bv 

 Prof. Meldola that the general likeness between members 

 ff the same distasteful family groups came easily under 

 The same principle. 



In order to ."npreciate this point fully, let us consider 

 Ih" common European Vanessas. the Peacock. Red 

 Admiral, large and small Tortoiseshells, Camberwell 

 Heauty, &r.. several of them familiar objects in our own 

 country'. We see at once that though there is certainlv 

 a family likeness between them, they are distinguishable 

 froiTi one another at a glance ; no one would think of 

 taking one of theip for another. Contrast this with a 

 NO. 1983, VOL. 76] 



similar group of closely allied species, known to be dis- 

 tasteful, from a part of the world where competition is 

 keen, for instance, the .\cr;eas of Africa. Of these, four 

 or five species may be taken on the same day, looking all 

 alike while on the wing, and practically indistinguishable 

 from one another without close examination. Or take a 

 group of Euplceas, another distasteful genus, from the 

 Oriental region. Here again we may have some five 

 separate species, all quite distinct, but so much like one 

 another that it needs much more than a casual glance to 

 distinguish between them. 



These and similar cases were shown by Meldola to be 

 easily explicable on the basis of the Miillerian theory of 

 mutual protection by the adoption of a common scheme 

 of warning colours on the part of inedible forms, and the 

 possibilities of the theory were still further expanded by 

 Poulton, who pointed out that in any given region the 

 fewer independent schemes of warning coloration there 

 were to learn, the better chance there was of the protec- 

 tion they afforded being effective ; so that the same simple 

 warning badge, such, for instance, as the alternate black 

 and yellow rings on the body of a wasp, might be 

 employed by insects, like the caterpillars of the Cinnabar 

 moth, which are widely separated from the wasp in point 

 of affinity. The aposeme, or signal to an enemy to keep 

 his distance, may be recognised and obeyed even when 

 hoisted by insects which have little else in common 

 between them. A great part of the, significance of the 

 facts that we have noticed depends, of course, on the 

 circumstance that the members of each of these closely 

 assimilated groups inhabit the same geographical areas. 

 We do not find an Eastern Euploea resembling an 

 American Heliconius, or an Ithomiine from Brazil re- 

 calling an .African .'\cr.-Ea. As a further illustration of 

 what Poulton has apth named " synaposematism," or the 

 adoption of a common warning badge on the part of dis- 

 tasteful forms, we may take the wonderfully diverse 

 assemblage that centres round the conspicuous and dis- 

 tasteful beetles belonging to the genus Lycus. This 

 assemblage, in South .-\frica, contains wasps, Braconids, 

 moths, a bug, and a two-winged fly. besides beetles 

 belonging to three or four different families. I have 

 myself seen several members of this group, heterogeneous 

 in aRinity though wonderfully similar in hue and pattern, 

 on or about one tree at East London, in South Africa. 

 Be it remarked that they were all conspicuous insects, and 

 exposed themselves freely, so that there could be no 

 question of a common cryptic coloration. The assemblage, 

 beyond doubt, is mainly if not entirely synaposematic. 



We have now reached what may at any rate rank as 

 a preliminary generalisation, that is to say, that the re- 

 semblance between distasteful forms is to their advantage, 

 .tnd is an adaptation brought about by natural selection. 

 Following the approved logical method of Mill and 

 Jevons, we ought next to see what consequences are 

 involved in the hypothesis we have formed, and then to 

 make a fresh appeal to the facts for verification or the 

 reverse. 



(i) It is obvious that in Batesian. or true mimicry, the 

 advantage is all on the side of the mimic. Experience 

 gained by tasting the mimic would be used to the injury 

 of the model. While, therefore, there is every induce- 

 ment for the mimic to seek safety by approaching nearer 

 and nearer to the aspect of the model, there is no reason 

 for the model to assimilate itself to the mimic, but rather 

 the contrary. 



In a Miillerian association, on the other hand, the 

 benefit is mutual. Each fresh accession to the group is 

 a source of strength, not of weakness. Everything is in 

 favour of the formation of such 'Jroups as rapidly and 

 on as large a scale as possible : hence there is nothing 

 to impede, and everything to promote, the free inter- 

 change of characters all round, each member being able 

 to act, so to speak, as hoth vtimie ami nieifc/. This, we 

 saw, could not happen in the case of Batesian mimicry. 



Now does this interchange of characters, as a matter 

 of fact, ever tak" place? If it does, it will be, of course, 

 a confirmation of our theory. 



One of the most characteristic features in the siibfamilv 

 of Pierlnes, or " white butterflies," is the possession of 

 red or yellow spots, streaks, or patches on the underside 



