MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 95- 



middle of the baud is a dark line nearly parallel with tbe inner one, dilatiuy on the veins and all 

 the lines forming distinct dark costal spots. The outer third of the wing gray, with dark scales 

 and with an irregular blackish wavy line, much as in ridenda and yupla. Hind wings whitish, 

 with no lines, and with three dark dots on the Iringe of the internal angle. The outer third of 

 the wing faintly dusted more or less with line dark scales. Wings pale whitish beneath; two 

 blackish costal spots beyond the middle, and costa of both wings S])cckled with dark scales. A 

 faint diffuse band i)asses across the hind wings just beyond the middle, and tlie margin of both 

 wings is speckled with dark scales. Body beneath pale, and the legs (tarsi) ringed with dark 

 scales. 



After preparing the preceding descriptions I find that Mr. Edwards adds to his description 

 of G. formosa the fbllowing remark: 



It is jiossible that (I. riilotda aiiil C. rupta are forms of oue sjiecies, and that (!. nibofascia and G. formosa are 

 forms of anotbor, but I prefer to consider them as distinct until future investigation shall determine their true 

 position. 



Mr. Dyar tells me that he has a specimen of G.formosa from El Paso, Tex. Professor French 

 reports it from Utah. G. var. ridenda occurred at Fort Collins, Colo., Jlay -3, Juue 11, and at 

 Denver July 2o (C. P. Gillette). 



Section II (Eumelia Nenm.). 



This section corresponds to the genus (sic) Melia or Etimelia of JNIessrs. Neunioegeu and Dyar. 

 At first, with only a single specimen of var. dossonhv to judge by, I thought it was the type of a 

 distinct genus, as the head, antenna', palpi, and venation seemed so different, but after careful 

 and repeated examinations of specimens, labeled avimaciila, lintneri, icrighUi, and severa, and 

 observing the general identity of form of body, wings, and especially of markings, as well as 

 the larval characters. Dr. Dyar stating that the larva of the Califoruian severa does not 

 differ generally from that of E. triUneata, I think with our present knowledge it would be quite 

 unnecessary to recognize Eumelia as a distinct genus. 



The structural differences between G. tritineata aud G. severa, var. slossoniw, and which at 

 first led me to think them gcnerically distinct, are the following: 



A ? . The head is remarkably small, much more so than in G. trilincaUi, and is loosely scaled 

 in front. The antenaaj are pectinated, the branches a little longer than in 9 trilineata. The palpi 

 are short, small, depressed, with loose scales; and they are not quite so large and long as in 

 trilineata. The thorax differs from that of Glaphisia trilineata in having a median dorsal tuft. 

 The legs are hairy, and much as in trilineata, the tarsi being ringed with gray aud darker scales. 

 The fore wings are uariow, but with the costa unusually convex, much more so than in trilineata; 

 the apex is somewhat rounded, but much as in trilineata; the outer edge is very faintly excavated 

 below the apex. The hind wings are of the same shape as in trilineata. There are six branches 

 of the subcostal vein; bianch 1 is longer than in trilineata and ends half way between the end of 

 costal veiu and end of branch 2 of the subcostal; the costal area is wider toward apex thau in 

 trilineata. The fifth and sixth branches are nearly as in trilineata. The lower discal vein is not 

 so much bent as in the last-named species. The three cubital veins are nearly as in trilineata, 

 but the second median space is wider than in that genus. The submedian vein ( V ) is represented by 

 a simple fold. In the hind wings the two branches of the subcostal are much longer thau in 

 trilineata, the space between them long and narrow, in trilineata short and broad triangular. 

 The discal veins are, taken together, slightly curved, where in trilineata they make a decided 

 angle at the origin of the indei)endent vein; and there is a common origin of the lower discal and 

 of the two median veinlets. The second median interspace is much wider than in the species of 

 the other section of the genus. Vein VI is represented by a simple fold. 



