1887.] VALUE OF COLOUR AND MARKINGS IN INSECTS. 213 



in producing a conspicuous appearance, although it may be admitted 

 that the hairs do render subordinate assistance ; of the two remaining 

 larv8e, one is brightly coloured (O. anliqiia), although the hairy tufts 

 are in this case very important factors, while for S. menthastri no 

 other suggestion except that of Jenner Weir has yet been made. In 

 some of these larvae the effective colours are chiefly on the hairy 

 covering, and the latter practically makes up the vphole appearance. 

 Turthermore in the last species there is evidence (Jenner Weir) that 

 the insect is disliked for some quality other than the presence of 

 hairs. It seems probable that the hairs of larvae possess irritating 

 qualities to a much greater extent than is commonly supposed ; but 

 it is also likely that the hairy covering may be of direct value to the 

 organism in other ways, soujc of which doubtless remain to be dis- 

 covered. (Is it not likely that some tactile or other terminal organ 

 of the nervous system may be in relation with hairs or bristles?) 

 Of course it is well known that hairs are often exceedingly important 

 in defending the insects by the converse method of a protective 

 appearance (an extreme instance of this is afforded by the larva of 

 Acronycta leporina, see Trans. Ent. Soc. Lond. 1886, pt. ii. June, 

 p. 160). 



Prof. Meldola has kindly looked through the proof-sheets of this 

 paper, and has suggested to me that the probalde original significance 

 of the hairy covering was protection from injury after falling from 

 the food-plant. The suggestion is strongly supported by the habits 

 of tlie majority of hairy larvae. 



Any further considerations which arise out of the tables given 

 above are better deferred until after an examination of similar 

 instances in forms other than the larvae of Lepidoptera. 



II. Brightly coloured or conspicuous Insects {other than Larvce). 



Under this heading I have only included such species as have 

 been actually subjected to experiment. There are a very large num- 

 ber of additional species from many groups of insects which doubtless 

 belong to this category ; but as they have never been brought to the 

 experimental test, they are excluded from consideration on the present 

 occasion. An extended list will be found in Wallace's ' Contributions 

 to the Theory of Natural Selection ' already referred to (The essay 

 on " Mimicry and other Protective Resemblances among x\nimals"). 

 In most of the instances given by Wallace, we may feel confident 

 that the test would prove satisfactory, especially as the author shows 

 that in many cases the conspicuous form is mimicked by one or more 

 species belonging to totally different groups, which accompany the 

 former in its range and which, nearly always, keep in a small minority. 

 Such facts render it in the highest degree probable (in fact make it 

 nearly certain) that the mimicked species possesses some excep- 

 tional advantage in the way of inedibility or otherwise — some 

 unpleasant peculiarity which confers upon it a more or less com- 

 plete immunity from the attacks of the Vertebrate foes of its class. 

 But in the present paper I am rigidly limiting myself to instances 



