Explanation of Plates. 319 



the model. The specimen is unforturicately in too poor 

 a condition for figuring. 



Locality of specimen represented in Fig. 3a., N.E. 

 slopes, Mamba State, about 5000 feet, September 26, 1905 : 

 Rev. K. St. Aubyn Rogers. It will be observed that the 

 mimic was captured the day after that on which its model 

 was taken. 

 Fig. 4. Limnas chrysippus, L., form dorippus, Klug. ( = hlugii, 

 Butl.), $ . This is the dominant form of chrysippus in 

 British East Africa. Kilimanjaro, May 1905. 

 4rt. Acraeajohnstoni, iormfidvescenSjOheTth., ^. An obvious 

 and beautiful mimic of dorippus (Fig. 4). The ancestral 

 markings persist, faint but distinct ; and characteristic 

 in shape and position on both wings. On the under- 

 side they are more conspicuous. The basal spots on 

 the hind-wing under-side are distinct, but the most 

 prominent are in this form brown instead of black, and 

 therefore comparatively inconspicuous. December 15-31, 

 1905 : native collector. 



Explanation of Plate XXII. 



Forms of Acrsea johnstoni together with their Danaine and 

 Acrseine models. 



All the figures are of the natural size. 



Fig. 1. Amauris lobengula, E. M. Sharpe, ^, from the fore.^t, Mt. 

 Chirinda (about 3600 feet). Melsetter, Gazaland, S.E. 

 Rhodesia. Captured October 7, 1905, by Guy A. K. 

 Marshall. The model of Figs, la and 16. 

 la. Acrsea johnstoni, Godm., iorm proteina, Oberth., ^. From 

 the same locality as the last, and captured by Mr. 

 Marshall on the same day. The relatively large size of 

 the squarish discal patch of the hind-wing (compare Fig. 

 16 on Plate XXI) is an evident synaposematic approach 

 towards the Amaarris represented in Fig. 1, also charac- 

 terized by an especially large discal patch. 



