NORTH AMERICAN LEPIDOPTERA. 115 



The variations in this insect are in the color of the secondaries and 

 tlie width and color of the marginal band. The color varies fi-om 

 bright red to orange yellow, and the band varies from a narrow 

 black margin to quite a broad terminal band interru])ted at the anal 

 angle. The form with deep red secondaries and narrow black mar- 

 gin is typical ; the form with yellowish or orange secondaries and 

 more brownish rather than black margin, is juanita Strk. Interme- 

 diate forms occur. There are two broods, one in April the other in 

 July. 



The larva has been figured by Abbot and Smitli, and descj'ibed 

 from the figure by Clemens. A good description is still a desideratum. 



Chcerocampin.i^:. 



Usually robust, yet elegant species. Head usually distinct, palpi 

 rarely short and slender ; tongue corneous, usually as long as the 

 body and often exceeding the tip of abdomen. Antennae various, 

 usually fusiform with a rather long re-curved hook at the tip ; some- 

 times the hook is not well marked, at others it is very prominent ; 

 in De'dephlla there is a decided approach to a clavate tip and this 

 genus is most nearly related to the Macroglosdnw. The thorax is 

 usually smooth, crested only in Enijo, and the abdomen is usually 

 long and slender, tufted in some few genera. The venation offers 

 nothing abnormal. The wing form is variable, usually angulated in 

 those species in which the abdomen is tufted, while in all the others 

 the outer margin of primaries is sinuate; the apex distinct, a slight 

 excavation below, then outward curve of the middle and a somewhat 

 marked anal angle. There is a tendency to bright colors, and a 

 banded or strigose style of maculation very different from the simple 

 sober maculation of the Sphingince. 



In the Tropics this family is much more numerous in genera and 

 species, and there is an immense variety in form and color, yet they 

 are as a rule readily distinguishable from a peculiarity of habitus, 

 and partly also because almost everything which does not readily fit 

 elsewhere is placed in this subfamily. 



As enlarged by me, several of the genera usually placed in the 

 ' Macroglossince. find a place in this subfamily. It seemed to me that 

 the former group was so capable of definite limitation, and yet so 

 thoroughly indefinite as usually constituted that a decided advantage 

 would be o-aiued in removing all aberrant material from it and leav- 



