^'"5-"' RECENT OBSERVATIONS UPON MIMICRY. ii 



in the background ; finally it appeals to another sense, for it has the 

 characteristic odour of hornets. Presumably, therefore, this perfectly 

 harmless insect deludes its foes into the belief that it is a veritable 

 hornet, sting and all complete. In this particular case it is necessary 

 to say " presumably," for, since the moth is so rare, it has naturally 

 escaped much experiment. Mr. Poulton offered one to a lizard ; the 

 reptile looked with marked suspicion at the moth, but after licking it 

 with its tongue, seized it by the thorax, which was at once crushed 

 so as to kill the insect. This experiment is not, therefore, very con- 

 clusive. From the moth's point of view, it is not more satisfactory to 

 be cautiously destroyed than to be bolted hastily. But there are 

 other instances where an insect, owing to its similarity to some other 

 insect, which is protected by a sting or a bitter flavour from much 

 persecution, shares the same immunity. It is, therefore, intelligible 

 how, as Mr. Bates first suggested, these facts of mimicry can be 

 explained by Natural Selection. Given the utility of the resemblance, 

 it is quite conceivable that minute variations in the required direction 

 were perpetuated and increased. 



The classical instance of mimicry is, of course, afforded by the 

 butterflies belonging to the families Pieridae and Heliconidai, which 

 were studied in South America by Mr. Bates. The Heliconidae possess 

 a rank flavour and odour, not only disgusting to the human nose, but 

 to monkeys, birds, and other creatures, which, as a rule, are partial 

 to butterflies. Two or three genera of the Pieridae have assumed the 

 livery of these unpopular Heliconidae, and are believed to trade upon 

 their reputation. It is supposed that the birds and other natural 

 enemies of insect life have gradually, through a series of unpleasant 

 surprises, learnt to distrust the gaudy colours of the Heliconidae, and 

 to assume that any butterfly coloured after the same plan would prove 

 to be also inedible. In all these cases there is, as Mr. Poulton has 

 admitted, " too little direct experimental proof of the unpalatability 

 of the specially protected groups which are the chief models of 

 mimicry." The burden of proving that the imitators as well as the 

 imitated enjoy a considerable immunity evidently lies with the 

 advocates of the theory. When Mr. Poulton wrote that "a peculiar 

 and frequently unpleasant smell has been noticed by all observers 

 who have studied " the three principal families of butterflies which 

 are models for mimicry, viz., the Acraeidae, Danaidae, and Heliconidae, 

 he was without doubt expressing the general opinion upon the subject. 

 Dr. Seitz, however, who has paid particular attention to this question, 

 was unable to recognise the least odour — disagreeable or otherwise — 

 in 50 distinct species of Danaids, both African and American. In 

 some, but not in all the Heliconids, Dr. Seitz found a decidedly 

 objectionable odour; and curiously enough in Heliconiiis beski, a species 

 with a particularly evil odour, it was discovered that only a small 

 percentage of individuals were odoriferous. This last fact is, of 

 course, not contrary to the theory of mimicry, because it may be 



