106 FAMILY EVANIIDiE. 



The combimition of veins and consequent nomenclature will 

 generally be different for each abscissa. In Fig. 73 it will be seen 

 that the medial cross- vein and the second branch of media form 

 apparently a single vein with no indication where they join. In 

 such cases where two different abscissas are joined end to end with- 

 out indication of the place of union, I have designated the entire 

 vein thus formed by the name of each abscissa connected by "and", 

 or if three or more are thus joined, use a comma between the 

 first two and "and" between the second, as m and M2, or, again, 

 I^s, R3+4 and R3. The -f- mark of course is only used where two 

 or more veins unite side by side, as R4+5 + Mi-[^2- 



On Plates XI-XV inclusive are arranged figures of all the types 

 of wings known to occur in the family, in what I believe to be an 

 order ascending from the most generalized to the most modified, 

 especially within each subfamily. The Aulacinse are very evidently 

 the most generalized. Take for example Aulacinus (Fig. 67), R4, 

 R5 and all of M2 are present. Omitting for the present the Fceninfe 

 (Figs. 72-74), we find in the wing of Evania (Fig. 76) only a 

 slight modification. R5 is lost and also the transverse part of M2. 

 M has migrated far forwards along the radius, carrying with it 

 m-cu. In this subfamily modification then proceeds by atrophy, 

 until in Evaniellus (Fig. 87) only C and Sc are left. The Faniinse 

 (Figs. 72-74) we may look upon as a side line. We cannot com- 

 pare the degree of their specialization with the others, because it 

 has been in an entirely different direction. It seems reasonable to 

 suppose that they may have arisen as an offshoot of Aulacime 

 (compare Figs. 71 and 72), or they may have arisen from some- 

 where among the ancestors of the Stephanidje. At any rate in 

 wing venation and other characteristics they are highly modified. 

 Of course I do not base my conclusions as to the relations of these 

 groups solely on the wings, but these are easy of illustration and 

 description, and in all more reliable for comparison than any other 

 single character, hence I emphasize them here. 



L'Abbe Kieffer* criticizes the table to the subfamilies of Evanii- 

 dte that I published in my former paper, on the grounds that the 

 characters given do not apply in all cases. But I think that his 

 criticism is unfair, since the table was intended only for the North 

 American forms, for which so far as I am aware it holds good. All 



* Spec. Hyiii. d' ICur. et d' Aljit-'r., vii, bis, p. 377. 



