venation and head-characters to arrange the other higher moths 

 in the following order: * 



Family Agaristidie. 



" CastniidiT. 



" Hepialiiiic. 



" Cossidie. 



Thyrididu-. 



" ALgfiiiihc. 



" Sp/ii/igidr. 

 The Hepialidce and C'ossid?e appear to be more nearly allied 

 to the Sphingid^ than to the liombyces, though originally they 

 may have directly descended from the latter group. The two 

 families are closely related. 



The three families of Thyridcc, JvgeriicUv and Sphingidae are 

 closely related. 



Finally, it is not improl)i\ble that all the moths mentioned in 

 this paper, including also the Rhopalocera, have originated in 

 various ways and at different times from the Bombyces, and per- 

 haps all of them in the first place from the Notodontians; though 

 the Sphingidse may have evolved from the C'eratocampida;, or 

 F^ndromidae. At present this may be a fairly good working theory 

 to account for the relationship of these families, and at all events 

 the Bombyces are with little doubt the most ancestral and general- 

 ized forms of the higher moths, as the Tineina are of I-epidoptera 

 as a whole. 



Specimens of Dataiia iiitijoi, I), pa/niii, D. coutracta, D. angiisii 

 and D. intcgerrima have been sent to Mr. C I'alm. from .Arkansas, 

 which is a new locality for these species. 



A single example of Hannonia morisonii Hy. Edw. was taken 

 by Mrs. Slosson at Watkins (lien, N. V. Hitherto only known 

 from Montana and Missouri. 



*It is possible tliat the Agarstid.x and Castniid;^ form a side branch, standing 

 above the Sphinges, and ne.xt to the Hesperidiv, Megathynnis being the connect- 

 ing link. 



