194 SOUTH-AFRICAN BUTTERFLIES. 
unmarked upper-side basal field of the hind-wing, and the frequent 
failure of the lower spots of the elbowed row of the fore-wing, are 
also characters not presented by Plialantlia. A strict examination 
of Cramer's figures A and B on his Plate ccxxxviii. will show the 
lepidopterist that they represent the form now under notice ; while 
those marked D and E on his Plate cccxxxvii. thoroughly agree with 
the Fhalantlia of Drury. 
A. Columhina (as here restricted) is a purely African species, 
attaining its greatest size and distinctness on the North-Tropical West 
Coast. I have examined the specimen marked iLuri/tis D." in the 
British Museum, which is ticketed " West Africa," and appears to 
have been the type figured in the Genera of Dumial Zepidoj^tera, and 
found that it agreed with Natalian examples, except in the diminution 
of the cellular and discal markings, and in the absence of all but the 
uppermost of the spots of the lower part of the elbowed row in the 
fore-wings. 
? Larva. — Greenish-black, varied yellowish-green and light-green 
subdorsally ; a yellowish-white line above legs, slightly angulated 
upward on each segment. Top of head and of second segment yel- 
lowish-green ; face with black markings on each side. Spines long, 
shining, set with alternating hairs ; the second, third, and last segments 
having 4 (2 subdorsal and 2 lateral), and the remaining segments 6 
(the additional 2 being infra-spiracular) spines each. Length about 
1 1 lin. 
? Pupa. — Bright-green, with crimson streaks and spines and bur- 
nished golden spots. Inner- and hind-margin of wing-covers edged 
with raised golden streak, outwardly bounded by a crimson one. Five 
pairs of curved, tJiin, divergent, crimson spines on the back, springing 
from golden spots, viz., one pair on the head, one (shorter) on thorax, 
and three on abdomen. Also two pairs of dorsal small pointed crimson 
tubercles, one at hinder part of thorax, the other at base of abdomen. 
Pupal stage lasted eight days in the month of November. 
I give the above description of the larva and pupa with some little doubt, 
as Mr. Gooch refers the pencil outline and notes from which (with two pupa 
skins) they are drawn up to A. Plialantlia. He, however, was not at the time 
aware of the existence of two species of Atella in Natal, and would naturally 
include Columhina under Fhalantlia. I subsequently saw both species 
together in his collection. The differences presented by this caterpillar and 
chrysalis from those of Fhalantlia are considerable, especially as respects the 
pupa; and it may almost be assumed that they are the early stages of 
Columbi7ia. 
I did not take or notice this butterfly in Natal ; but Colonel Bowker early 
in April 1881 took several examples in the park at D'Urban, where they 
frequented a species of tree then in flower. In sending me the specimens, he 
wrote that while pursuing each other these butterflies did not have anything 
to do with A. Plialantlia, which was common at tlie same spot, but did not 
visit the particular tree. I examined three individuals ( $ and two $ s) 
ticketed "Port Natal" in the collection of the British Museum, and found 
them to present no differences from those received from Colonel Bowker. 
