148 THE CANADUN ENTOMOLOGIST. 



SO far the American examples may be picked out by experts. When 

 these conditions can no longer be fulfilled there would be no ground for 

 retaining a different name. The mere fact of their inhabiting different 

 continents is not sufficient, they must breed true to type and not produce 

 each other. Then we can be sure we have to do with separate cycles 

 of existence and we can catalogue the fact. As the genitalia are con- 

 cealed, their structure is not so apparent, and it is clear that repeated 

 observations are necessary to verify the statements drawn from solitary 

 dissections. But granting what has been published as substantially 

 reliable, there yet remains the test of breeding to be applied to the 

 genitalic species. We have an instance in the genitalic species of 

 Niso7iiades. These butterflies have not been bred to ascertain if they 

 remain true in their genitalic peculiarities, if one genitalic type does not 

 produce the other, if the caterpillars show no differences. Until all these 

 matters are cleared up we can arrive at no final conclusion as to the value 

 of genitalic characters, as to which single observations must be checked 

 by repeated experiments. Writers on the subject have apparently pro- 

 ceeded on the basis that the male genitalia are formed, not by deposits of 

 chitine but of cast iron, moulded so as to fit and give at last a stable and 

 firm reality to our artificial system of classification. Vain expectations ! 

 The characters, on which we are obHged to found all our categories, are 

 one in quality and only differ in quantity ; what is generic is specific 

 also, and what is specific is varietal. 



In my Buffalo lists, 1874-1876, I was at some trouble to give the 

 generic types of the Noctuidae, and my action, unless it can be shown 

 that I was in any one case in error, is binding from those dates. Prof. 

 Smith was, therefore, no longer free to retain Feridroma for occulta^ as I 

 accepted Envois for that species, without showing my action to have been 

 at the time unwarranted. To place my A. pelhicidalis in the same " genus " 

 with occulta., and on account of the genitalia, is not to be defended. 

 The variability of the genitalia cannot be made a basis for generic separa- 

 tion nor their agreement for generic grouping without other characters. 

 The two insects are strongly different in form and vestiture, the hind- 

 wings being in the Anicla group translucent, where I would refer my 

 species. The work of Prof. Smith bears proof, from internal evidence, 

 that the intention was at first to consider but one genus, Agrotis. Not 

 only are the "genera" called "groups" in the body of the text on 



