TERACOLVS. 13 



I quote Mr. Marshall's remarks on these female specimens : " In this species the male 

 is quite stable, as is the case throughout this group, but the female is very variable, 

 presenting four gradations: (1) like the male; (2) yellow, with the base of the 

 primaries suffused with pink ; (3) pure yellow ; (4) white. The latter might easily be 

 mistaken for the Indian T. puellaris, Butler, but may be distinguished by the curved 

 macular discal stripe on the underside of the secondaries." Besides these phases to which 

 Mr. Guy Marshall has drawn attention, tliere is apparently j-et a fifth form, represented 

 in the British Museum by four specimens from Aden. It is suffused throughout with 

 deep orange (see Plate 3, fig. le), and is evidently a dry-season form, corresponding 

 to the phase of T. rothscJdhli, which I have figured (Plate 4, fig. le). In addition to 

 these forms enumerated, there is still another form, which passes iu collections as 

 T. arne of Klug. Of this I have seen both males and females in the collections of the 

 Hon. Walter Eothschild, Mr. Philip Crowley, and in the British Museum. It is 

 smaller than the ordinary type of T. jjJnsadia and is apparently a dry-season form, 

 answering to the dry-season form of T. jjrotractus, but differing in the more broken 

 character of the apical portion of the primaries, where the light spots are more rounded 

 and not so linear as in T. protractus. The secondaries have the basal portion white 

 instead of salmon-colour, and a female in Mr. Crowley's collection has the hind margin 

 of the primaries light orange, so that the black is reduced to a zig-zag sub-terminal 

 band, while the secondaries have the black border interrupted with orange markings 

 near the posterior hind margin. 



Underside. — Compared with T. profrachis, the general colour is very much redder 

 in tint. Expanse 1'7 and 1'4 inch. I have seen specimens of the " T. arne" form 

 from Nubia, Somali-land, and from Senegambia. 



There is also a pale yellow female in Mr. Crowley's collection marked "Senegambia." 

 This is the specimen figured by me (Plate 4, fig. Id). Dr. Butler thinks that it is one of 

 the phases of T. rothscJdldi, and if this should turn out to be the case, some mistake 

 may have occurred with regard to the locality of the specimen. As I have, however, no 

 reason to doubt the correctness of the locality, I prefer to think that it is only a 

 strongly marked specimen of the yellow form of T. phisadia, which, as we already 

 know, occurs in Senegambia. 



Dr. Butler remarks as follows : — " The male of this species is a wet-season form 

 and the female (in all its varieties) invariably dry-season ; of course they all fly 

 together at the same time." 



Habitat. — Arabia: Aden, Jan., Feb., March, April, July, Aug., Sept., Oct., Dec. 

 {^Col. Ycrhury ] Mu>>. Brit.); Haithalhim, March [Col. Yerhurij ; Mm. Brit); Lahej, 

 IMarch [Col. Yerhurf/ ; Mus. Brit.); Lahej, Jan. {Capt. Xi/r-se; Mm. P. Crowkij ; Miis- 

 E.M.S.). Somali-Land. Gellokur, Peb., Dobar, Goolis Mts. Feb., Daras, March [Coll. 



