( xxiii ) 



No. II. 



On the Systems of Classification of MotJis. 



The natural arrangement of insects, and especially of Lepi- 

 doptera^ has always been regarded as presenting great difficulties. 

 The latter are covered with a thick coating of hair and 

 scales, which must be removed before we can even obtain more 

 than a superficial idea of their external structure ; the species 

 are numbered by tens of thousands, and without a careful com- 

 parison of characters it is impossible for us to ascertain either 

 their importance, or how far they hold good for genera, species, 

 or even for individuals ; and finally. Butterflies and Moths 

 pass through four very distinct phases of existence ; and, as 

 regards the larva state especially, many species differ extremely 

 in form and structure, when the larva newly emerged from 

 the egg is compared with the larva of the same species when 

 about to assume the pupa state. Add to this, that perfect 

 insects which greatly resemble each other, and would naturally 

 be classed near each other, sometimes prove, when reared, 

 to proceed from larv?e differing in the most fundamental 

 characters ; that the transformations of the greater number, 

 especially of the foreign species, are wholly unknown, and that 

 our observations, even of the few that are known, are often 

 confined to the full-grown larvae only ; and it will be sufficiently 

 obvious how much still remains to be done in this department 

 of Entomology. Here and there an Entomologist like Herold 

 or Lyonnet has devoted a life-time to the study of the trans- 

 formations of a single insect, such as the Cabbage Butterfly, or 



