Antennal Characters o/'Cosmophila erosa, Hiibn. 299 



antennas sometimes filiform and slightly pubescent, sometimes 

 very distinctly furnished with thick hairs, serrate and pubes- 

 cent. Of C. erosa he says u Antennes filiformes" thus giving 

 one the impression that in G. xanthindyma, indica, and 

 auragoides they are not so. Of C. xantliindyma he says, 

 " Les antennes du male sont nettement et fortement cilie*es." 

 The four supposed species separated by M. Guene*e have pre- 

 cisely the same variations in form, pattern, and coloration ; 

 but at one time I believed tha4 G. erosa and G. xantliindyma 

 could be separated by antennal characters. Mr. Hampson, 

 however, assured me that these also varied. 



In the Museum series we have fifty-seven examples of the 

 genus, twenty-five of which are males possessing antennas, 

 the result of an examination of which is as follows : — 



1. Antennae finely ciliated. 

 Two specimens from St. Domingo and one from Venezuela. 



la. Less finely ciliated. 

 One from Sao Paulo. 



2. Serrate-fasciculated. 

 Two from Alabama, one from Australia. 



3. Pectinated. 



One from Alabama, one from Aden, one from Kulu, one 

 from South India, one from Ceylon, four from Java. 



3 a. Strongly pectinated. 



One from Solun, one from the Nilgiris, one labelled simply 

 N. India, one from Java, one from Ceylon, one f rom Moreton 

 Bay, one from Fiji, and one from the Marquesas Islands. 



So far as our specimens show, the ciliated type of antenna 

 would appear to be confined to South America, but it is con- 

 nected with the pectinated type of the Old World by the 

 intermediate serrate-fasciculated type of the United States 

 and Australia. Indeed there is not the slightest doubt that 

 the variation in the structural characteristics of the antennas 

 in this widely distributed and abundant species are individual 

 and have no significance, so that the supposed four species of 

 M. Guenee, as well as Walker's Cirroedia variolosa and 

 G. edentata, represent nothing more than sports of one 

 variable species. 



