A BRIEF ESSAY ON CLASSIFICATION OF THE 

 HETEROCERA. 



By a. R. Grotk. 



A recent criticism upon tlie genera, which I have established 

 in the NoctiiidcB runs to the effect that I have regarded all modi- 

 fications of structure as of generic value. This criticism loses 

 force when we see that its author adopts seventy-five of my gen- 

 era in the family and rejects but very few of my generic names 

 for this reason. Since it appeared I have published many addi- 

 tional genera, so that in the Noctindce alone I have now described 

 upwards of one hundred. It is important that the structural 

 differences which have induced me to establish these genera 

 should be discussed, and I, therefore, state that I have studied 

 the following characters: 



1. The 77m^, the structure of the compound eyes, presence 

 of ocelli, form of tongue and palpi, width and appearance of the 

 front, structure of the antenna;. 



2. The Thorax, shape of collar, patagia, comparative shape 

 of wings, structure of the wings, neuration, structure of the feet, 

 especially of the fore tibiae. 



3. The Abdomen, comparative length and shape (consult my 

 Illustrated Essay, p. 31, et seq.). Added to these, the vestiture 

 of the whole insect has usually been studied and gives good char- 

 acters. The tuftings of the body and the nature of the squama- 

 tion, whether of scales like hairs, or flattened, or a mixture of 

 these, or, again, of what I call " flattened hairs," intermediate in 

 width, must be fully noted. 



However low a character the neuration may be regarded by 

 some writers on Lepidoptera, it is, at the last made the test of 

 family distinction by Herrick-Schaeffer and modern authors gen- 

 erally. Although, as I showed many years ago, in some forms, 

 e. g. Thyridopteryx, the neuration differs in opposite wings of 

 the same individual, or, again, separates two species, other- 

 wise closely related, such variation concerns merely the branch- 

 ing of certain nervules, and does not affect the general pattern, 

 the relative distance of the two principal veins from each other, 

 and the number of branches belonging to each. Thus, to some 

 extent, the neuration affords only a variational or specific char- 

 acter ; in another view it gives generic and even family criteria. 

 In other words it is not different from the balance of characters 

 used in classification ; the amount and extent of the peculiarity 

 gives the value. Every well-marked variation and modification 

 of structure, which can be clearly made out, by the microscope 

 or otherwise, is of generic value. The moment this rule is de- 

 parted from we are thrown upon " opinions." Certain struc- 



