280 



first discovered liabilat, across the l)road Xnuaini Valley, the 

 most important dividing valley of the range. 



nithcrto its food-plant has remained midetermined. It had 

 been originally taken only in small nnmbers sweeping over the 

 staghorn fern {(jleiclienia) and other vegetation along the old 

 native trail (in the crest of Kanmuohona ridge. ]\Ir. Timber- 

 lake and I had each taken a single male on Fiffosporion on thej 

 Lanihnli ridge at abont the same elevation as its original 

 hal)itat. 



On April 2-'^ while collecting with iSiv. Timl)erlake along 

 the Castle trail, in retnrning we separated and I proceeded 

 along the old ridge trail back to the Cooke trail descending 

 into Xnnaiin Valley, lio])ing to secure further material of this 

 species of which even as yet only a few specimens had been 

 taken, and perhaps to solve the prol)lem of its food-plant. 

 After a few strokes of the swee])-net over ferns and other low 

 vegetation, four adults were seen in the net. For some time 

 this ]n'ocess was continued, keeping in mind the ])lants swe])t 

 over, and it soon became evident that aside from the common 

 ferns and Mcfrosidcros. the only ])0ssible food-])laut among 

 those swept would be Eiiphorhia chisiaefoUa. The first plant 

 examined disclosed one adiilt, and repeated examinations others 

 both adubs and nym])lis. In one case a single leaf bore on its 

 under side alxmt a dozen individuals, including two or three 

 nym])hs. This leaf was placed in a glass tube and carried 

 home. During the succeeding week one of the nymphs com- 

 pleted its transformations and the greater part of tlie individ- 

 uals remained in a living condition until the leaf withered. 

 In all about ."iO individuals of the species were taken, being all 

 found in the I'egion where the host-]dant was ])i-esent. 



F.npliorhid cliislae folia thus a])pears to be the ju'oper host- 

 ])lant for the sjxries. This plant appears to be generally dis- 

 tributed along the lateral ridges back of Honolulu, though in 

 no other jdace have I happened to find it so abundant as on 

 this ])articular ridge. AVe may perha])s hope to extend our 

 knowledge of the distribution of the s])ecies by a careful exam- 

 ination of the ])]aut in othei- ])arts of its range. 



