418 REVISION OF FAMILY SATURNIID^. 



in July — in Washington two to three weeks earlier. There are two 

 annual broods in these latitudes. 



Since the date of its introduction into this country (1861) this insect 

 has undergone a considerable change in color and wing form, quite 

 marked when compared with specimens from China. It is larger, 

 deeper in color, and the wings are much broader and more rounded, 

 much less excavated below the apex. 



ATTACUS Linn. 



This genus, to which I refer all the other American species belonging 

 to this subfamily, is an extensive one, and yet so characteristic in each 

 case that there is no difficulty in recognizing it at a glance. The defi- 

 nition of the family and subfamily contains all the characters of this 

 genus, and it differs from ISamia only in the untufted abdomen and in not 

 having the hind angle so distinctly produced. The genera based upon 

 wing shape and on the characters of the front appear to me invalid. 

 They are Gallosamia Pack, and Platysamia Grt. All our strictly North 

 American species agree in having the supra anal plate divided into two 

 divaricate hooks at tii). Sphndiila appears to lack the plate altogetlier, 

 if my specimen was normal — I had only one S that I could dissect — 

 while calleto shows an apparently abnormal structure, which I have 

 figured as I saw it, though I believe the figure incorrect. The venation 

 varies in the species and will be sep.arately described and figured. Two 

 rather sharply defined groups can be recognized. The first, containing 

 splendidus, orizaba, and cinctus, have on both wings a large, trigonate? 

 vitreous spot, and they are all subtro[)ical forms. Except cirutus, 

 indeed, the question of their occurrence in the United States is yet 

 uncertain. Splendidus makes the nearest approach to the type of the 

 genus in wing form and in general habits and maculation : orizaba 

 approaches it so closely in the male that they have been considered 

 identical, while cinctus, which is undoubtedly distinct from either, has 

 the typical maculation of orizaba and the wing form of cecropia. 



The other group lacks the vitreous spots, though their place is taken 

 by angulated or lunate spots or marks of a light color. Calleto ( poly- 

 ommata Teiiper), pro inetJiea, and aiigulifera differ from the others of the 

 group by having the sexes more or less dissimihir in color and wing 

 form. The males are very close in wing form to splendida, while the 

 females are like cecropia or Columbia. The pale mark in the primaries 

 is here angular, and the abdomen is not banded, but has a stigmatal 

 pale band, including black spots. In this feature these species agree 

 with the splendidus group, which has the abdomen maculate in the same 

 way. 



Columbia., gloreri, ceanotM, and cecropia have the sexes similar in 

 maculation and nearly alike in wing form. The abdomen lacks the 

 lateral spots and is furnished with white bands. 



