1887.] VALUE OF COLOUR AND MARKINGS IN INSECTS. 241 



chiefly relied upon this species for the evidence that hunger will force a 

 Vertebrate to eat an insect which is evidently distasteful to it. Further 

 the L. viridis being less used for this purpose, were not pressed by 

 hunger to the same extent as L. muralis. I think that the almost 

 certain explanation is that L. muralis was afraid to touch an insect 

 wliich was not far from its own size, while L. viridis was less 

 timid, the difference in size biding far greater. And such an 

 explanation throws light on the cases already discussed, in which a 

 large caterpillar is protected by gaining some marking which suggests 

 the appearance of a serpent. On this point Weisinann says (Mel- 

 dola's translation as above, p. 330) : — " It does not require much 

 imagination to see in such a caterpillar an alarming monster with 

 fiery eyes, especially if we consider the size which it must apjiear 

 to an enemy such as a lizard or small bird." The case of S. ligustri 

 enforces this last remark, and shows how size alone may be efficient 

 as a protection against the smaller insect-eating Vertebrates. So 

 far as I am aware, this important use of size, unaccompanied by 

 any accessory markings or any special attitude, is now brought for- 

 ward for the first time ; but it is an advantage which is probably far 

 from uncommon, considering the number of large species in the 

 same positiuu as S. ligustri, and it has doubtless been of special 

 importance as an initial stage in the development of the more 

 elaborated forms of terrifying appearance already described. 



Larva of Mania fi/pica. — In this case it was perfectly clear that 

 the larva possessed a very unpleasant taste, so that it was refused by 

 L. muralis even when very hungry. And yet the description given 

 in the table shows that the species is highly protected in the 

 larval state by protective colouring and habits which correspond. 

 As the exception is so important, I will add a few details to the 

 proofs given in the table. Newman gives the following facts about 

 the young larvte : when hatched from the eggs laid upon the leaves 

 of pear, phim, &c., the larvae " devour the upper cuticle and paren- 

 chyma of the leaf, leaving the lower cuticle entire, dry and brown ; 

 they lie closely packed side by side and apparently motionless," but 

 in reality gradually moving onwards, " leaving a larger brown space 

 behind." We see here a most interesting adaptation of the sur- 

 roundings to the brown colour of the laiva. A brown larva is 

 conspicuous on a green leaf, and a single larva could not eat away 

 the cuticle so as to surround itself with a brown area of sufficient 

 size until after the lapse of considerable time. Hence the subordina- 

 tion of gregarious habits to ends which are the exact opposite of those 

 usually sought, viz. the intensification of warning colours. We 

 have therefore a most elaborate and perfect mode of concealment ia 

 the younger stages of larval life. It is obvious, however, that such 

 ■ a method can only be successfully adopted while the larvae are very 

 small, so that a large number of them can rest for a long time on a 

 single leaf. Accordingly Newman tells us that "in a few davs, 

 perhaps from ten to fifteen, they abandon this arboreal life," and de- 

 scending, feed upon many kinds of low-growing plants. It hyberuates 

 in October, and again feeds greedily in the following spring ; but,^ 



