194 MR. E. B. POULTON ON THE PROTECTIVE [Mar. 1, 



ose a great deal of blood, and although they may recover, are 

 generally stunted ; and often I am sure the loss of blood proves 

 fatal. If the wound be at all extensive, the fat-body and viscera 

 protrude, owing to pressure on the side distal to the wound (that 

 on the proximal side having been relieved by escape of blood)." 

 Therefore it is that throughout the varied means of defence possessed 

 by larvae " the object is always the same — to leave the larva 

 untouched, a touch being practically fatal." Wallace also originally 

 expressed this peculiar danger incurred by larvse in more general 

 terms, viz. " their soft and juicy bodies are so delicate that if seized 

 and afterwards rejected by a bird they would almost certainly be 

 killed" (see 'Contributions to the Theory of Natural Selection,' 

 187-5, p. 118). Other peculiar dangers of the larval stage will be 

 pointed out below. Although it will be seen that brilliantly coloured 

 and distasteful larvfe are often tasted by their Vertebrate foes when 

 impelled by hunger, all observers agree that a second trial is rendered 

 less likely because of the unusual appearance which accompanies the 

 unusual and unpleasant effect upon senses other than that of sight. 



Furthermore, I am now able to bring forward instances of very 

 distasteful species which have no warning colours, but, on the other 

 hand, are well disguised by protective tints and markings; and a . 

 comparison between the behaviour of Lizards towards these and the 

 conspicuous species respectively, affords strong confirmation of the 

 truth of Wallace's suggestion. It now remains to summarize the 

 whole of the evidence in favour of the prediction made in 1867 ; for, 

 after all, the question is purely one of evidence, and however con- 

 vincingly the a priori arguments may be put, they are chiefly 

 valuable as guides to practical investigation. And this is fully 

 recognized by Wallace, who strongly urged the practical test upon 

 the meeting at which his suggestions were first made. Experiments 

 have been made by Mr. J. Jenner Weir, Mr. A. Gr. Butler, Prof. 

 Weismann, and by myself. Jenner Weir (see Trans. Ent. Soc. Lond. 

 1869, part i., April) made use of the following birds in bis investi- 

 gations : — Erithaca rubecula (Robin), Emberiza cifrinella (Yellow- 

 hammer), Emheriza schaniclus (Reed-Bunting), Pyrrhula vulgaris 

 (Bullfinch), Fringilla coelebs (Chaffinch), Loxia curvirostra (Cross- 

 bill), Tardus musicus (Thrush), Anthus arboreus (Tree-Pipit). He 

 also used to a less extent Carduelis spinus (Siskin) and Linaria minor 

 (Redpoll). 



He obtained the following results : — 



" All hairy caterpillars (experimented upon) were uniformly 

 uneaten ; " viz. Arctia cnja, JEriogaster lanestris, Porthesia auriflua, 

 and Orgyia antiqua. " None of these species were even examined." 

 The writer believes that the hairs are not themselves disliked, but 

 that they " serve as a caution to the birds that the larvae so clothed 

 are uneatable." This suggestion is supported by the fact that the 

 young and comparatively hairless larvee of Spilosoma menthastri 

 were tasted by the Siskin, Redpoll, and by a West-African Finch 

 {Textor erythrorhynchus), but these three birds evidently found the 

 larvae disagreeable, and soon left them alone. On the other hand the 



