2878 THE ZOOLOGIST—JANUARY, 1872. 
and the rules which he has laid down for his own guidance 
radically bad. 38. As to the accuracy of the synonymy, I know 
that it is correct in some genera with which I am more particularly 
acquainted, and I see little reason to doubt its general accuracy. 
4. The work appears scarcely complete. 
First, then, as to system—a subject in which I have always taken 
the deepest interest. It is rather amusing to find that the author 
should have thought it necessary to make anything like an apology 
for the systematic arrangement of this crowd of objects, but he 
certainly seems to do so in the following paragraph :—* Although 
full materials for a natural arrangement of all the species could not 
be obtained, it was thought better to attempt some kind of arrange- 
ment than to place the species alphabetically.” In this sentiment 
I fully concur, and I think Mr. Kirby somewhat needlessly distrusts 
his own unquestionable knowledge when he expresses any hesita- 
tion on the subject. He not only gives us “some kind of arrange- 
ment,” but an arrangement displaying deep study both of Nature 
and of books; an arrangement which some may fear to adopt 
because disclosing unlooked-for combinations, and because travel- 
ling out of the accustomed groove, but an arrangement which 
everyone will be inclined to study with attention and respect. 
Three ideas of arrangement prevail among those who have 
assiduously studied our pedunculate Lepidoptera. 
i. That the characters of the imago, more particularly size and 
colour, are available for systematic classification: such seems to 
have been the leading idea with the illustrious father of Natural 
History, John Ray, and he has been obsequiously followed by our 
own Haworth and many others. It is obvious that this arrange- 
ment is par excellence the arrangement of the museum, and results 
from a knowledge of the preserved specimen as acquired after the 
individual is prepared for the cabinet, and a careful and loving 
study of its peculiarities. 
ii. That the preparatory states of larva and pupa offer more 
natural, and therefore more reliable, characters for classification : 
this seems to have been the leading idea with Denis and Schiffer- 
muller, and to have been adopted very cordially by Horsfield and 
Swainson ; it is also the plan which Herrich-Scheeffer has followed, 
although not avowedly, in the Introduction to his beautiful and 
priceless ‘ Schmetterlinge von Europa,’ and by myself in the ‘ Ento- 
mologist’ and the ‘Illustrated Natural History of British Butterflies.’ 
