EXPLANATION OF FOEM AND COLOURING. 26J 



(jfl) 1. iY. agatha, Hesperia sp., M. lecla, P. cebrene, and probably 

 F. lyceus, A. phalantha, L. argia $ , and F. natalensis f. 



2. iV. saclava. r Probably C. dolorosa and possibly 



3. N. goochi, T. senegcdeiisis. \ L. hcetica. 



^^ ' ' ' \ L. argia § , F. natalensis Q f., A. phalantha. 



2. II. dcedalus. J 



3. Eurytela dry ope and probably hiarbas (latter, at any rate, 



below dcedalus). 



The bird's reconsiderations of JV. goochi and i\^. saclava show how risky it would 

 be in this experiment to include in the highest grades Henotesia and the other 

 insects the acceptance of which might have been the result of special stimulation.] 



Exp. 39. — June 9. Very hungry after a fast of several hours. Tasted and 

 rejected Danaida chrysippus, Amauris lobengiola, and Jcrcea areca, and refused 

 without tasting Acrcea caldarena, Acrcea serena, and Acrcea cabira. I next offered 

 Mylothris agathina (a gravid d ), with unusually strong menthol scent. It was 

 crushed and readily eaten, and Acrcea cabira was then crushed and readily 

 swallowed, as also a S Acrcea serena, rather rubbed, Acrcea caldarena S was 

 crushed and at once rejected, a white Acrcea esebria was thoroughly tasted and 

 swallowed, an Acrcea doubledccyi was tasted and rejected, as was also Acrcea 

 natalica. JVyctemera leuconoe $ was now readily eaten, a buff Acrcea esebria 

 tasted and swallowed after momentary hesitation, a freshly- emerged Acrcea serena 

 crushed and eaten, but Acrcea areca <$ rejected with intense disgust. Another 

 fresh-looking Acrcea serena was crushed and eaten, but Danaida chrysippus and 

 Amauris lobengula tasted and rejected. 



Five minutes later the bird was sitting with feathers ruiHed, possibly suffeiing 

 some discomfort from the Acrceas eaten, possibly not. She tasted and rejected a 

 bright-looking Acrcea serena and a black-and-red Acrcea esebria $ (gravid), crushed 

 and readily ate Mylothris agathina J (with a decided smell, less strong, however, 

 than that of the $ accepted earlier), tasted and rejected Acrcea serena and black- 

 and-red Acrcea esebria, refuxsed obstinately without tasting Nychitona medusa, but 

 tasted and readily ate Nyctemera leuconoe $ and a Terias {brigitta ?). 



After a short interval she refused without tasting Acrcea serena, crushed well and 

 ate Acrcea esebria, (red-and-black § ), refused without tasting Acrcea serena, tasted 

 and rejected emphatically Nychitona m,edusa, tasted very thoroughly and ended 

 by rejecting an Acrcea serena, disguised slightly (to secure its tasting) by being 

 offered with a forewing instead of the usual hindwing, ate a few grasshoppers, 

 then tasted and rejected Belenois meseiitina and Terias desjardinsii (?), ate a few 

 more grasshoppers, refused without tasting Neptis saclava, Neptis goochi, a Byblia, 

 and a Hamanumida dcedalus, but evidently recognized and at once stretched out 

 eagerly for a Gatacroptera cloantha, which she ate with apparent relish. She then 

 accepted, crushed, and readily ate Hamanumida dcedalus, Atella phalantha, and a 

 Byblia (dark hindwing, no trace of smell), twice tasted cautiously and rejected 

 a Teracolus, tasted and threw right away a Neptis saclava, and then refused 

 without tasting but shortly afterwards accepted and ate a Henotesia perspicua. 



[Summary, etc.: — 1. Note at the time. "The Catacroptera was evidently very 



