240 MR. C. F. M. SWYNNERTON ON THE 



cardui. This was early in the morning, before the bird had been supplied with 

 food, and she was unmistakably hungry and eager. 



[At this time of day, before feeding, the bird, as will be seen later, was sometimes 

 hungry enough to eat even Acrceince, so that the unacceptability to her of the 

 Pyrameis larva is likely to have been very considerable.] 



Later in the day (the bird had just previously been feeding on grasshoppers) she 

 tasted and rejected Nepiis saclava, refused without tasting Phrissura isokani (quite 

 likely taken for Mylothi^is), readily ate, after crushing each, Precis clelia, Pyrameis 

 cardui, Precis cebrene, a Mycalesis safitza (with all its wings), and a Precis archesia 

 f,, but rejected, after tasting it, a Belenois mesentina. 



Three Charaxes brutus, two Gharaxes candiope, and a fourth Charaxes brtttus 

 were now readily eaten, the last two with difficulty ; the next was crushed and 

 dropped, and on being reoffered refused without tasting. Bird apparently 

 thoroughly replete. She now crushed slightly and threw away Gatopsilia fiorella c5' , 

 refused without tasting Pyrameis cardui and Precis cebrene, but then accepted, 

 crushed well, and swallowed Charaxes guderiana, afterwards crushing and dropping 

 Charaxes neanthes and refusing without tasting Charaxes cithceron. 



Fifteen minutes later she readily accepted and ate a Precis cebrene and a Cato- 

 psilia florella, but refused without tasting a Pyraineis, a Mycalesis safitza, a Leuce- 

 ronia, and a Charaxes neanthes, evidently still feeling replete. 



Five minutes later she ignored Precis cebrene and a Pyrameis cardui, but accepted 

 and swallowed, though with decided disinclination, Catopsilia florella and Charaxes 

 cithceron, refusing the next C. cithcBron. 



I afterwards at intervals (during 20 minutes) offered her Charaxes and other 

 butterflies and grasshoppers, but she was apparently thoroughly gorged and com- 

 pletely ignored them all as she basked in the sun with feathers ruffled out and bill 

 open — a common habit. 



[Apparent order of preference : — 



a. P. clelia and perhaps r 1. C. guderiana. 



M. safitza and P. \ 2. C . brittus Sim\ perhaps C. candiope, P. cebrene, and 

 archesia © f. I f. cardui. 



3. JV. saclava and P. isokani (or whatever the latter 

 was taken for) ; P. mesentina. 

 b.l. C. guderiana, 1 .^ ^_ cithceron. 



2. C. florella, J 



3. P. cardui, P. cebrene, C. neanthes, M. safitza, Leuceronia.'] 



The Pyrameis cardui larva was probably much disliked. 



Exp. 25. — Seeing that the bird was unlikely to eat more before his next pellet, I 

 now left scattered fairly closely and thoroughly mixed on the floor of its cage three 

 Charaxes hrutus, three Charaxes candiope, three Pyrameis card'td, three Precis 

 cebrene, one Eurijtela hiarbas, three NejMs saclava, and one Papilio lyceus. From 

 her experience since entering confinement the bird should have known well all these 

 butterflies except perhaps Eurytela, which she had hnd only once. I hoped in this 

 way to gain some confirmation by another method of her already displayed 

 preferences. 



