192 STATES OF INSECTS. 



to nature and probability : but as we are now to enter 

 at large upon the Moults of insects, it will not be without 

 use if I add a few additional reasons which seem to me 

 still further to prove the correctness of Swammerdam's 

 system, as far as it relates to that subject. With regard 

 to the mere formation of the skin from the rele mucosum, 

 were this the whole question few would hesitate to adopt 

 the sentiments of M. Herold ; but when we come to con- 

 sider further — that the number of moults of individuals 

 of the same species is always the same, and that it varies 

 in different species, and takes place at certain periods, — 

 we begin to suspect that something more than the mere 

 formation of a new skin upon an old one being cast is to 

 be accounted for ; and that the law which prescribes its 

 own definite number of skins to each species, must begin 

 to act in the primordial formation of the larva. Again, 

 the hairs observable in the higher animals do not take 

 their origin from the epidermis solely, but are planted 

 below it in the rete mucosum, or deeper a ; so that the 

 change of skin does not affect them ; but in the larvas of 

 insects they are a continuation of that integument, since, 

 when the moult takes place, they always remain on the 

 rejected skin b : this is the case, also, even with spines. 

 If you shave a caterpillar ready to change its skin, either 

 partially or generally, you will find that the parts in the 

 new skin that correspond with those that are denuded, 

 are equally hairy with those that were not c ; and if 

 you pay attention to the new-clad animal, you will find 

 f irther, that the hairs never grow after a moult. From 



a Cuvier Anat. Comp. ii. 596. N. Diet. d'Hist. Nat. xxvi. 165. 

 b Cuvier Ibid. 624. <= Reaum. i. 183. 



