THE COLORATION PROBLEM. 155 



highest point, that no case occurs amongst them after the nestling 

 period is over, and downward migration, which starts at the end of 

 July, has got into full swing. The cases occur at just that time when 

 one would expect them to occur if the bases of the theories be correct. 

 To turn to another feature of this island fauna. It is an integral 

 part of the theories that warning colours lose their utility to a large 

 extent, where insect food is scarce. Let us take the insects which 

 exhibit warning colours, and which, by birds' behaviour, I judge to be 

 nauseous: — Antlironna /ilipt'iulxhie (June and July), Hi pocrita jarobacae 

 (May to August), Calliinorpha (lomiiuila and C (.luadnpnnctata (hera) 

 (June)*, Diacrisia sannio (May to June), Para^ewia idantafjinis (June), 

 {ParaAemia plantaf/inis is very oiiensive to my nose, it has a peculiarly 

 dank, corpsy smell about it, that is not easy to describe, but which is 

 very penetrating and persistent). Arctia raia (June and July), Aretia 

 villica (June and July), Spiloaouia iiientliaxtrl {Diacrisia /lava), S. 

 liibncepe.da, and <S. nrtirae (all June and July), Leiicoixa ftalieis 

 (June and July), Pdlura luonaeha (July). (Prom everything I can 

 see from the carelessness of this moth in its site of rest and general 

 behaviour and odour, I believe it to be protected). Satarnia pavonia 

 (April), Venilia iiuiciilaria (May to June), Abraxas i/rossidariata (June 

 to July), and probably the genus (Jabera (June), one and all appear 

 during the time insect life is most abundant, and when the birds 

 would not be driven to unpalatable food. How difterent is the 

 case when the downward migration of the hungry yearlings is in full 

 swing, during the winter months and early spring, when our insecti- 

 vorous winter residents are busy and very hungry, one and all of the 

 lepidopterous insects seek the retiring garb of russet-brown and dead 

 leaf colours, or else dress themselves in the soft greens and greys of 

 the lichen-covered trees. Professor Poulton has also pointed out that 

 Coccinellids, which pass the winter as imagines, hide when the insect 

 life is getting scarce ; so also Vespa. There also seems to be a ten- 

 dency, amounting in some cases to a bitterly prejudiced obsession, to 

 throw doubt upon the validity of the results attained by experiment on 

 captive birds, and to treat them as of no account in every case where 

 they show that the supposed protected insect is not palatable. In 

 some cases the writers seem to go to the absurd extremity of rejecting 

 everything in favour of the theories, whilst seizing upon and insisting 

 upon every little point which they think tells against the theories. 

 Then again, I think that those who reject such evidence are clearl}'^ 

 wrong, apart from the question of selection of species eaten. It must 

 now be admitted by all, in face of the abundant evidence forthcoming, 

 that birds of almost every species show a marked partiality for insect 

 food of some kind. It is notorious that birds in captivity do not get 

 a sufficiency of insect food either in quantity or variety, since it is so 

 difficult to procure. Accordingly one would expect that a bird in 

 captivity would accept greedily insect food presented to it, unless it 

 were too shy to feed in the presence of man, and that without being 

 too particular as to kind presented. When, therefore, one gets birds 

 under these conditions deliberately rejecting particular species time 

 after time, I believe one to be quite safe in drawing the inference that 

 the insect is unpalatable to that bird and inferentially to all birds of 



* I have never taken it till late July. — H. J. T. 



