( 506 ) 



distinct ; the snbmarginal, oreiiate line heavy, sliglitly internipteil on the hind- 

 wins; at R-'. UiidtTxiili' rather nuiforiu in colonr, jmrplisli and ochraceous, the 

 admarninal line obsolete on liotli wings, the suhinarginal one faint ; bars of 

 hindwin" all ochraceons, not brown or black, jiostdiscal sjiots withont black or 

 brown centres ; ochraceons discal line heavy, eontinnons ; no whitish patches 

 outside this line. 



lldh. : Sokotra ; 2 J J in the Tring Museum, a series of specimens iu the 

 British Museum. 



Atella columbina. 



Papilio Ni/mjjlidll.i Phdtraliis cohimh'ma Cramer, Pup. K.r. in. p. 7G. t. 238. f.A.B. (1779) ("China, 



Coromandel," error luci). 

 Alella eiirytis Doubleday, in Doubl, Westw. & Hew., Geii. Ditini. Lcji. I. p. 107. t. 22. f. 3 (1848) 



(W. Africa). 

 Alella jihiilaidii var., Trimeii, S. Af'r. Bull. ed. 1. p. 115 (18(')2) ; id., Kirby, Cat. niimi. Lep. p. 154 



n. 1. 1871 (partim) ; id., Butler, Proc. Znol. Sui-.Loml. p. 53(1808) (wet form = cuhtmbhiu, dry 



form =phulaulha ! !) ; Pagenst., I.e. p. 137. D. 1 (1902) (partim). 

 Alella columbina, Trimen & Bowk., S. Afr. Bult. i. p. 193. n. 58 (1887). 



Butler was quite wrong in maintaining that ^1. phulaiitlia and columbina are 

 forms of one species. The fact that columbina does not occur in the Oriental 

 Region shonld have made liira hesitate to jiublish a statement which was 

 supjiorted by no evidence whatever. The two insects are constantly different iu 

 structure and pattern. Some distinctions in colour were well jwinted out by Trimen 

 in 1887, I.e. We add that the median spots jnst behind the cell on the 

 underside of the forewing are (as a rule) obsolete in columbina and distinct ia 

 phalantlia, that the hindwing is more obviously jirodnced at W m columbina, and 

 that in pkaluntha the cross-vein D' of the hindwing is jnst opposite M', while it is 

 more proximal in columbina. The copulatory organs are very different in pkaluntha 

 and columbina. 



In the male-i of both species the clasper bears at the dorsal margin, close to the 

 tenth (supra-anal) tergite, a very long, thin, tapering process, beset with bristles 

 and tubercles. This process, which is doubtless of a, sensory nature, being very 

 thinly chitinised at the tip, is strongly elbowed iu pkalantha, and feebly and 

 gradually curved in columbina. The apical margin of the clasper is obliquely 

 and very shallowly sinuate below the filamentous process in columbina, and deeply 

 sinuate in phalantlia. Above the sinus the clasper of pkalantha is produced 

 into a short obtuse lobe, and below the sinus into another, much longer, lobe. 

 In columbina the upper lobe ia barely indicated, and the portion of the clasper 

 below the sinus is only a little jjrodnced, being broad, obtuse, and slightly truncate. 

 This broad, short lobe is armed at the lower corner with an acute tooth. The penis 

 has a special armature within the sheath. The armature consists in columbina 

 of an elongate swelling on each side, beset with short teeth-like tubercles. Ia 

 phalantha the two swellings are enlarged, projecting from the sheath. The two 

 together resemble somewhat a pair of feet of an armadillo with the backs 

 turned towards each other, and beset all over the conve.K surface with long, 

 strong, curved, claw-like teeth. 



While the West African specimens of columbina are generally large, and bear 

 large black postdiscal sjiots on the underside of the hindwing, the individuals from 

 South and East Africa are small and have the postdiscal spots reduced in size. 

 The specimens from Somaliland and the western districts of the Abyssinian Empire, 

 for which we propose the name 



