84 TERACOLUS. 



Keiskamma, I saw this butterfly in the greatest profusion near tlie river ; but only a 

 few stragglers were seen a mile or so beyond, and none anywhere else ; though I have 

 just come off a ride of about 120 miles in four days, having visited the following 

 places, viz., Tamacha, Line Drift, Breakfast Vley, Alice, Middle Drift, and Fort 

 White." Mr. D'Urban also found the insect abundant in the Fish River Bush, and 

 informed Dr. Trimen that Captain CoUey, 2nd Queen's, reported its occurring 

 abundantly in a Kloof on the Tninika, a tributary of the Kei. 



Dr. Butler writes : — " The yellow form of this species appears to be strictly confined 

 to Kaffraria and Natal, but a somewhat paler race occurs in Matabele Land. The 

 extreme types T. auxo and T. keiskamma were proved by Mr. M ansel Weale, and recently 

 by Mr. Guy Marshall, to be ' wet-' and 'dry-' season forms of one species; T. top/ia, 

 which is usually regarded as identical with T. keid-amma, appears to be an intergrade 

 of which we possess six examples in the Museum." 



"Of the Matabele Land type, which only differs in its somewhat whiter coloration, 

 we only possess males of the ' wet ' and ' intermediate ' forms." Mr. Mansel Weale 

 (Transactions of the Entomological Society, 1&77, p. 273) writes as follows : — "In 

 Mr. Trimen's Rhop. Afr. Austr., vol. i., is described Anthocharis {Callosiaic) keis- 

 kamma, Trimen, a species there noticed as not improbably capable of being classed as 

 a sub-species of C. evarnc. Both of these butterflies I have often seen in the neigh- 

 bourhood of the Keiskamma River, the former being especially abundant ; in fact, the 

 whole valley of the Keiskamma is peculiarly prolific in Callosunes. During the last 

 three years I have been staying about three miles from King William's Town in tiie 

 valley of the Yellowwoods, and I was much struck by the abundance of C. keiskamma 

 near a small group of buslies from January to May 1876, especially so, as I had not 

 met with the insect in the neighbourhood before. Tfie spot in question is about 

 fifteen miles in a direct line from the Keiskamma with higli ground intervening. I 

 captured that season about twenty or thirty of both sexes, and wishing to discover 

 the food-plant I noticed that the butterflies especially frequented a bush which I had 

 not before noticed. This proved to be Cadaha natalensis {Cajjjjaridex), not hitherto, 

 [ believe, reported from the south of Natal. The females deposited their small, 

 lluted, orange-coloured eggs singly on the summit of the flower-buds. A chrysalis 

 was also found on one of the outer branches of a bright-green colour, and it proved 

 to be that of C. keiskamma. The larva, wlien first hatched, is of a bright-orange 

 colour, and penetrates the bud, where it passes its first stage. It afterwards assumes 

 a dull bluish-green colour with lateral stripes of a paler colour : these assimilate it to 

 the pellucid margins of the small leaves of the plant, whose general colour it resendiles. 

 The younger caterpillars of Eroiiia cleodora closely resemble the mature larva of 

 C. keiskamma, but are more brightly coloured, in harmony with the foliage of Capparis 



