3.] 



SCIENCE. 



445 



investigation from both biological and 

 chemical standpoints. 



(c) The avoidance or rejection as food 

 by insectivorous animals of the insect pos- 

 sessing malodorous or distasteful juices no 

 longer rests merely on the negative evi- 

 dence given by Bates, Wallace, Belt and 

 other competent observers, to the effect 

 that in nature such distasteful forms are 

 habitually neglected and unmolested ; there 

 is now much positive experimental evidence 

 as to the manifest avoidance or disgust 

 with which such species are left untouched, 

 or thrown aside after tasting, when of- 

 fered to domesticated or captive verte- 

 brate animals that devour ordinary in- 

 sects with avidity. The numerous experi- 

 ments of this kind recorded by Butler, 

 Jenner Weir, Weismann, Poulton and 

 Lloyd-Morgan, as regards both larvse 

 and imagos pf European species, are sup- 

 ported by a few made by Belt with Heli- 

 coniinte in Central America, by D'Urban 

 and myself with Danainse and Acrseinse * 

 in South Africa, and by- Haase with Da- 

 nainse in Singapore. 



It is manifest, of course, that even the 

 most distasteful forms cannot enjoj' com- 

 plete immunity from persecution ; in or- 

 dinary circumstances they are doubtless 

 mainly kept down by parasitic insects, f 

 and during any scarcity of more palatable 

 prey it is certain that they will be devoured 

 faide de mieux by vertebrates and inverte- 

 brates alike. To the latter condition are 

 perhaps due such cases as Distant's J note 

 of the orthopterous Hemisaga devouring an 



* De Nic^ville (Butt. Ind., etc., I., p., 318) notes 

 that Acrsea violx was the only butterfly rejected by 

 all the species of Mantidx which he offered vari- 

 ous butterflies. 



fC. V. Riley (apud Haase, ?. c, II., p. 47) found 

 that a dipterous parasite was very prevalent in the 

 larvae of Danais archippus, often destroying a whole 

 brood. 



t Nat. in Transvaal, p. 65 (1889). 



imago of Danais chrysippus ; Col. Yerbury's * 

 observation that in Ceylon Euplcea core and 

 Delias eucharis were largely taken by a 

 Mantis, and Danais Umniaoe by two kinds 

 of Asilidse ; and Belt's remark that a flower- 

 frequenting spider captured Heliconiidae. 



(<i) As regards the important point 

 whether the protected forms have to suffer 

 a certain percentage of loss from the at- 

 tacks of young and inexperienced birds 

 and animals, it must be admitted that the 

 evidence at pi-esent forthcoming is exceed- 

 ingly scanty ; and I have long felt consid- 

 erable doubt as to the sufficiency of this 

 factor to account for the mimetic resem- 

 blances, often remarkably close, between 

 members of associated protective groups. 

 But on reviewing carefully the recorded 

 observations which appear to bear on the 

 question, I have found reason to think that 

 there is enough support to justify the pro- 

 visional acceptance of the Miillerian ex- 

 planation. We have, in the first place, 

 Fritz Miiller's own capture of Heliconii and 

 AcrjsinEe with a notched piece bitten 

 out of the wings, and Distant's (L c, 

 p. 65) of a Danais chrysippus whose 

 wings had been bitten unsymetrically, 

 apparently by a bird. Then there is 

 the significant record of Skertchley {I. c, p. 

 485) who, among twenty-three species of 

 Bornean butterflies taken with both hind 

 wings mutilated in the same manner, notes 

 no less than four Danainse, vid., Hestia 

 lynceus, H. leuconce, Ideopsis daos and Euplcea 

 midamus. Moreover, it is very remarkable 

 that several of those entomologists who 

 have specially emphasized the small part 

 played by birds in attacking butterflies 

 mention, among the few cases of such at- 

 tack as they witnessed, instances of pro- 

 tected forms being assailed. Sir G. Hamp- 

 son f remarking that in south India the 

 Euploese and Danaids were caught as often 



* Proc. Ent. Soc. Lond., 1897, p. xl. 

 t Proc. Ent. Soc. Lond., 1897, p. xxxvii. 



