rAPTLTONIN/E. 241 



of this pupa is seen to be very much more bent backward than it is 

 in Dcmolcus. 



Usually pale bluish-green, inclining to yellowish on under surface 

 of abdomen, but very variable in tint. Point of dorso-thoracic promi- 

 nence, two spots below it (at abdominal base), and edge of abdominal 

 lateral angles, creamy-reddish. A row of minute indistinct blackish 

 spots on each side of back of abdomen. 



This chrysalis, though usually pale-green, is variable in colouring, 

 specimens that I reared near Grahamstown pupating in the same 

 wooden box (on the sides) within a day or two of each other, varying 

 from that tint to a more or less ochreous-tinged, much duller hue. 

 Mrs. Barber was subsequently so fortunate as to observe the extreme 

 susceptibility of this pupa to colour influences, as pointed out in her 

 paper above quoted. From these most interesting observations it 

 appears that the colour of the object on which a larva pupated 

 was very closely reproduced in the pupa. Pupa among orange twigs 

 were of the ordinary green colour ; others, among half-dried leaves of 

 the " bottle-brush," were pale yellowish-green ; one attached to the 

 wooden frame of the case was of the yellowish tint of the wood ; and 

 another, attached to a part of the frame where wood and purplish- 

 brown brick joined each other, was coloured on the under surface like 

 the wood and on the upper surface like the brick. The experiment 

 of causing a larva to pupate on scarlet cloth had no effect except 

 that the ordinary small red s]3ots were brighter than usual; but this 

 is not to be wondered at, considering that the environment of these 

 insects could never, through endless generations in the past, have 

 rendered the assumption of a scarlet colour of any advantage in con- 

 cealment.^ 



Colonel Bowker, in 1874, sent me from King William's Town 

 four pupa3, three of the ordinary green colour, and the fourth (which 

 had been purposely placed w^ien changing on the mud-mortar of a 

 wall) of a dull greenish-yellow, much clouded dorsally with dull 

 creamy ferruginous-grey. These were winter pupae ; two became per- 

 fect insects in July (ist and 24th), one on October ist, and the last 

 on December 21st. I set three of the butterflies loose in the Museum 



^ A most remarkable instance of pseudo-mimicry recently came to my notice in connection 

 with the pupa of this Papilio. In September 18S7 I received from the Rev. N. Meeser, of 

 George, Cape Colony, a small box containing what I took at the first glance for three 

 ordinary green chrysalides of P. Lyaus. Only one of these objects, however, was a veri- 

 table chrysalis, the two others being the seed-capsules of a plant, stated by Mr. Meeser 

 to be a species of Hahca. The tint of green, the general lateral outline (especially the 

 bulging ventral convexity of the wing-covers), the projections of the bifid head, the attenu- 

 ated form of the posterior abdomen and anal extremity, and even the slight ferruginous 

 tips of the projections on the head, are all reproduced in the seed-capsules to a very decep- 

 tive extent. The chrysalis was found " in the neighbourhood of a hedge of the Ilakca ; " 

 and if this plant had been a native of South Africa, it can scarcely be qiiestioned that a 

 strong case of mimicry would readily have been admitted by observers. As a recent intro- 

 duction from Australia, however, it is clear that the Hahea cannot have been the model for 

 the pupa of a Papilio of a specially African group. 



VOL. Ill, Q 



