REMARKABLE INSTINCT. 199 



emerging, as the nervures are at first rilled with 

 fluid, and until that is dried up they do not 

 lay out nicely. It is not always easy to find 

 out when the fluid has completely disappeared, 

 and if the insect is left too long it will destroy 

 its beauty endeavouring to escape, but if care 

 be taken not to press strongly on the inflated 

 parts when laying out, the fluid in time com- 

 pletely dries up. 



Occasionally I have been so unfortunate as 

 to be unable to find the right food-plant for 

 caterpillars brought me ; and this is due in great 

 measure to their bad habit .of going to another 

 plant, sometimes a long way off, to change their 

 skins. I think the larvae of the Spingidce must 

 be more delicate than others at those times, as 

 I not only lose them in captivity, but have 

 often found their bodies with the skin partly 

 cast off being devoured by ants. 



The larvae of the death's-head moth vary so 

 much in colour, although the markings are 

 similar in two of them, that they appear to be 

 distinct species. One is pale straw-colour, with 



