EXPLANATION OF FORM AND COLOURING. 239 



readily ate five, although five clays before that again she had with the greatest 

 readiness eaten seven in rapid succession and had doubtless then " sized it up 

 even if she had never met it before. She must then from April 10, with a small 

 rem"inder on May 12, have retained in her mind a sufiiciently clear impression of 

 the degree of unpleasantness possessed by Belenois^ ready to be put to use on the 

 first occasion on which she was sufiiciently replete to do without that butterfly.] 



Exp. 23. — May 22. The bird was fed on grasshoppers till she rafused to take any 

 more. She, nevertheless, then accepted and ate, though with apparent difl&culty as 

 the result of growing repletion, four Charaxes cithcey-on, wingless, refusing the next 

 without tasting. Ten minutes later she accepted and ate, though with obvious 

 lack of appetite, two more Charaxes cithceron, refusing the next. 



Five minutes later she again refused it. 



Five minutes later again (i. e. ten minutes after probable repletion) she crushed 

 and readily ate Charaxes achce7nenes, refused without tasting Precis cehrene and 

 Atella phalantha {each with one wing), took but threw away a wingless Precis 

 natalensis © f. She tested it well, however, and swallowed it without 

 apparent dislike on its being reoffered. The next was refused without tasting, 

 but a 5 Charaxes ethalion was accepted and eaten. The bird then again 

 tried and tossed aside the Precis natalensis © f., but swallowed readily, after 

 similarly crushing it, Charaxes candio])e. She then refused the Precis natalensis 

 without tasting, tasted and rejected a c? Leicceronia thalassina and a Terias, 

 refused a wingless Atella without tasting, quite likely however recognizing it from 

 its fulvous coloi'ation, also, once more, the Pi-ecis natalensis © f. (throughout 

 ofli"ered without wings). A wingless Melanitis was, however, crushed and readily 

 eaten, as was also a Precis cehrene. Charaxes neanthes with one wing was then 

 crushed slightly and thrown aside, Pinacopteryx pigea, wingless, was refused 

 persistently without tasting, Po^ecis natalensis © f. was refused and then accepted 

 and dropped, but three Melanitis in succession were crushed and readily eaten, as 

 were Papilio darclanus $ f. hippocoon (without perceptible smell), Hamanumida 

 dcedalus, and Precis ceryne. A second Precis ceryne was refused without tasting. 

 I have never yet seen the roller so obviously replete. 



All butterflies in this experiment were ofi^ered with one hindwing attached, 

 unless otherwise stated. 



[Conclusions : C. cithceron, judging from signs of repletion, was placed as high, 

 probably, as any buttei-fly yet. C. achcemenes is probably in Grade 1 too, or at any 

 rate no lower than P. cehrene ; the Charaxes ethalion $ and C. candiope were 

 preferred to P. natalensis © f., an individual of which, however, was eaten at, or 

 not far from, repletion-point ; M. leda and perhaps P. ceryne, P. cehrene, H. dce- 

 dalus, and P. dardanus 5 f- hijjpocoon were also all apparently preferred to a 

 P. natalensis © f. as well as to L. thalassina S , Terias, and, probably, Atella 

 phalantha. The Melanitis and perhaps the Papilio, the Hamanumida, and P. ceryne 

 were also ' preferred to P. pigea and Gh. neanthes, or else to whatever these two 

 hutterjlies m.ay have heen reminiscent of, and P. cehrene was at least not liked worse 

 than these latter.] 



Exp, 2^.— -May 23. Crushed and unhesitatingly rejected the larva of Pyrameis 



