22 Journal New York Entomological Society. [Vol. vl 



comes into near contact with the Cubitus, thus following the course of 

 specialization in the entire group, by which the middle branch of the 

 Media becomes radial in disintegration. But the pattern remains dis- 

 tinctly Aglian, the veins are stiff, tend, even in the most highly special- 

 ized forms, to remain equidistant, there is no effort to lead to the 

 Saturnian pattern, indeed there seems no possibility of a progression in 

 this direction, vein IVi having taken quite a contrary course, a course 

 entered upon already by the most generalized form, Eacles. But this 

 course is possible from the Aglian, not possible from the Saturnian types. 

 Vein VIII of the secondaries is retained, while it is shortening. Know- 

 ing, as we do from Dyar's studies, that the larva conforms to the Satur- 

 niades type, it becomes a matter of comparative less importance whether 

 we confer upon the Citheronians family rank. Under this general view 

 of the position of Citheronia, we consider the slighter correspondences 

 in venation with the Hawk Moths to stand in relation to the narrowing 

 of the wings and the habit of pupating in the ground to have been 

 separately acquired. The Citheronians have pursued a peculiar path in 

 evolution and one that stands in relation with their comparatively 

 limited geographical distribution. They seem confined, as long ago 

 pointed by me, to America, east of the rocky backbone of the two 

 continents. 



From the clear exposition of Dyar, Can. Ent., 28, 303, it seems im- 

 possible to reconcile a phylogeny based on the larval tubercles of the 

 Saturniades with the one proposed by me on the neuration. Taking the 

 latter as the final appeal we are obliged to suppose, that Attacus and 

 Saturnia on the one hand and Aglia on the other have independently 

 acquired the tubercles on anal plate. According to the value placed by 

 Dyar on these organs, I must agree that this seems impossible. On the 

 other hand, I cannot find it probable, indeed, it scarcely seems to me 

 possible, that Aglia (which, in the same wing pattern of venation, 

 clearly represents a more specialized type than Automeris} should belong 

 to the Saturnian branch and wing pattern, as a generalized type. Nor 

 does it seem to me within the range of probability, that Automeris or 

 Citheronia could have produced the wing pattern of Hemileuca. On 

 our respective trees, the groups represented by Hemileuca and Aglia 

 change places. The female antennae of Aglia, Automeris and Cither- 

 onia are of one type, so far as I can see ; also those of Attacus, Satur- 

 nia and Hemileuca hold together, both types appearing distinctive. 

 Hemileuca is just what one would expect of a generalized Saturnian ; 

 Aglia, just what one could agree that a specialized Automerid might 



