Tlic Bionomics of South African Insects. 315 



— Nym'phalinm, Picrinfe, Lijci&nidse, and Hes'pcridi&. The 

 only species with marked aposematic colouring and habits 

 was the Lycrenid Alxna aniawuhi, and of this only two 

 specimens were offered, both being eaten freely. After 

 fifteen days of this diet the Mantis escaped : she was then 

 as healthy and vigorous as when first captured. 



Mantis XL, female, species resembling Sphodromantis 

 lincolci, Salisbury. Fed solely upon Acr/va caldarcna and 

 A. axina, and Limnas chrysippus, with long periods of 

 starvation, two of them a month in duration. Seventeen 

 chrysip)pus, six caMarc?ia, and three a^jma were eaten without 

 any signs of distaste, while four cunnaweve discarded after 

 tasting several times. The Mantis was captured on April 

 3, 1898, and refused food on September 4, dying on Sep- 

 tember 8 without signs of ill-health or blindness. Mr. 

 Marshall suggests that the species may be a winter form 

 (possibly of S. lineola) specially adapted to eat Acrseas 

 when other butterflies are scarce. — E. B. P.] 



3. Conclusions from Experiments on Mantid.e. 

 (E. B. P.) 



Certain conclusions stand out very clearly, while others 

 are suggested as probable. These voracious insects did 

 not show any dislike of butterflies outside the DanainiB 

 and Acr/vincV. The undoubtedly aposematic Pierine genus 

 Mylothris was freely eaten, and so were the following 

 genera with probable warning colours, movements, and 

 attitudes — Ncptis, Ahvna, Fcniila, and the moth Egylolis 

 vaillantina. Even the Danainsn were generally eaten with- 

 out hesitation (II., IV., XI.), and never rejected alto- 

 gether. In marked contrast was the behaviour of 

 Mantidai towards Acnvhuv, which were constantly refused, 

 and often eaten only after one or more trials and long- 

 intervals of time. When the Acrseas wei'e eaten freely 

 and without hesitation there is reason for suspecting 

 exceptional hunger. The summary of experiments 

 shows very clearly that " Pardojjsis appears to be consider- 

 ably more distasteful. . . . than the general run of 

 Acrreas " (G. A. K. M., October 7, 1807, Malvern). There 

 were also less marked diff"erences in the defjree of dislike 

 shown towards other species ; thus axina was less freely 

 eaten than caldarcna (XL); caldarcna appeared to be 

 eaten more freely than halali, neohule, induna, and natalica 



