PBOTECTIVE COLOKATION. 103 



Internal-feeding Larvae sometimes Striped. 

 But many of these larvte liave a far more efficient means of 

 protection than that afforded by longitudinal striping. They are 

 internal feeders, devouring the pith of reeds and other stems; 

 and yet they retain the longitudinal striping in a situation 

 which insures immunity from the attacks of all but entomo- 

 logists ! I am not aware whether birds ever do extract 

 such caterpillars from their galleries, but it is quite immaterial 

 to the present argument if they do. 



Striping sometimes occurs in certain Species only of a Genus. 



The caterpillars that have been hitherto dealt with belong to- 

 genera in which all the species are striped. There are other 

 genera, among British insects, in which some of the species 

 are striped and some not striped; though all feed upon low- 

 growing shrubs. Of the genus X// 1 ojj/iasia the species rure'/,. 

 Aepatica and scolopacimi, are striped, while lithoxylea and 

 pohjodon are not. X. pohjoclon is one of the commonest of 

 British moths; it is often quite a pest at " sugar," and turns ui> 

 everywhere; it is the commonest s])ecies of the genus, and does- 

 not therefore seem to have lost much by its (hypothetically) 

 less perfect adaptation in the larval stage. Still these instances- 

 suggest •' selection " more than the others. 



Striped Larvae do not always feed on or among Grasses. 

 It is perhaps not a very strong argument against the view 

 that longitudinal striping is protective in origin to point to- 

 caterpillars which frequent situations where the striping does 

 not appear to be of so much value; but in a discussion of the 

 subject these instances must be mentioned. Many of the 

 " Prominents " — all of them apparently that have not larv» 

 with excrescences and humps upon some of the pigments — 



