THE CANADIAN ENTOMOLOGIST. 67 



A CLASSIFICATION OF THE NORTH AMERICAN 



MYRMELEONID.'E. 



BY NATHAN HANKS, WASHINGTON, D. C. 



Some time ago in examining our species of ant lions I noticed that 

 our two large genera, Myrmeleon and Brachynemurus, could readily be 

 separated by the position of the origin of the radial sector ; in one much 

 nearer to the base of the wing than in the other. The characters previ- 

 ously used for Brachynemurus, as the forked costal veinlets, the length of 

 the tibial spurs, etc., had not been satisfactory : and the discovery of this 

 distinction led me to examine the other forms of the family with a view 

 to their better classification. 



It is evident that the double series of costals is not in itself of 

 generic importance, for in Maracanda one species has a double series, 

 and another species, closely allied, but one series ; and, more than this, 

 there are numerous gradations. The use of the tibial spurs, or, better, the 

 length of the first tarsal joint, differs so much in species that appear 

 otherwise closely allied that I cannot see how it can be of generic value. 

 Yet definite differences in this matter, taken in conjunction with other 

 important characters, may well serve to distinguish genera. 



In applying the characters I have put chief rank on the origin of the 

 radial sector. This point is best brought out by comparing the ending of 

 the anal vein with the origin of the first fork of the radial sector. This 

 divides the family, as represented in our fauna, into two groups, each of 

 four genera. By such a division I was surprised to see that Myrmeleon 

 was more closely allied to Acanthaclisis than to Brachynemurus, yet such 

 a relation is sustained by many other characters. 



Myrmeleon ingeniosus has long been recognized as differing con- 

 siderably from the other species of Myrmeleon, and for it I have erected 

 a new genus. Some might think that Brachyneimiriis longipalpis would 

 also form a new genus, but it differs in no important character from 

 other species of Brachynemurus, except the long palpi. There are 

 other important structural variations in this genus ; for example, the 

 origin of the radial sector in the hind wings, and the amount of elevation 

 of the vertex. Some specimens of B. abdominalis have an elevated 

 vertex, but many Eastern examples have a very low vertex and more 

 prominent eyes. But these variations, though very remarkable, do not 

 appear to indicate specific differences. 



