— 9 — 



studies on the larvii;, especiall)' the freshly liatehed ones, as well 

 as the moths, I have been led to consider one of two alternatives. 

 I. Either the Cochliopodida; have originated from the Saturniida; 

 or from forms allied to them; or, 2. IJotli the Saturniichv and 

 CochliopocUdiV' have descended from a common stem-form, and 

 this perhaps some Notodontian. At all events the systematic 

 position (and in this connection I may say that the larval, pujial, 

 and imaginal characters bear me out) of the group represented bv 

 Limacodes and its allies, i^ very near Saturniida;, and not far from 

 the Notodontida^. 



I am inclined lo believe that the oldgst, most generalized, 

 living forms, though at the same time indeed the most highly 

 sjiecialized forms, are the tuberculated larvre of Kmpretia^ Eiiclea 

 and Adaiicta, as they resemble the larviie of Saturnians, and in some 

 respects those of the Notodontians. On the other hand the nearly 

 smooth slug-worms, when fully-grown without hairs or even tuber- 

 cles, such as the larval Limacodes and /ft'frroi^c/ica, which seem to be 

 the most aberrant and modified, viz.: have become the most adapted 

 to the peculiar mode of life emphasized by the term "slug-worm;" 

 these being cater]-)illars which have lost by disuse their abdominal 

 legs, the thoracic ones being greatly reducetl in size; while by their 

 sluggish disposition, their slug-like slow, gliding mode of progres- 

 sion, and by the peculiar coloration of the larva; (viz : Heierogenca 

 which mimics the red, swollen spots on the leaves of various 

 trees), we have, as the result of gradual modification brought about 

 by adaptation, perhaps the most strange and bizarre type of Lep- 

 idopterous larva in existence. 



The succession of genera we should propose is as follows: 

 Empretia, Euc/ea, Parasa, Adoncta, Phobctroii, Moiudeiica, Isa, Lima- 

 codes, Packardia, LJthacodia, ffctcroi:;c/ica (including Torfricidia and 

 perhaps Kroiica). 



F.\M!1,\ 10. I.ASIOC WIIM D.t:. 



(,)uite contrary to my former opinions and prejudices this 

 family instead of being ])laced at or near the bottom of the 

 Bombyces, belongs much higher up in the series and should be as- 

 sociated with the Liparidne rather than with the Ceratocampidae 

 and allied families; this view being based on a consideration of 

 both larval and imaginal characters. 



Beginning with Clisiocampa, Artacc, and Tohpc, Hctcnpacha 

 connects them with the most highly specialized genus, Gastnpacha. 



The tufted and very hairy larva; lead to the Liparidte. 



