EXPLANATION OF FORM AND COLOURING. 259 



the insect each time longer in her bill and crushing it more thoroughly than the 

 Acrseas, being evidently tempted to swallow it. She ended each time in thro wing- 

 it right away. 



She then eagerly ate a Gharaxes candiope and thereafter obstinately refused to 

 touch a Nychitona, tasted and rejected a rubbed Acrcea serena S , an Acrcea 

 douhledayi $ , twice in succession a 5 Mylothris agathina, subjecting it, however, 

 to more prolonged tasting than any of the preceding insects, tasted and very 

 promptly rejected a $ Mylothris yulei (with stronger smell), tasted well and threw 

 away a second and different Mylothris agathiiia $ , and after it a c? , readily ate a 

 Belenois Tnesentina, refused without tasting a Terias senegcdensis with one wing, 

 tasted and rejected it when disguised by the removal of the wing, and once more 

 readily ate a Belenois mesentina, then tasted and rejected a rubbed Acrcea serena 

 and a 5 Acrcea areca, the latter with very marked signs of dislike. 



An interruption of fifteen minutes now occurred while I experimented with the 

 kingfisher. On my return the roller crushed and rejected Terias senegcdensis, 

 wingless, reofferod, but with the greatest apparent relish ate seven Neptis saclava 

 in rapid succession. 



I then gave ten grasshoppers, after which she readily ate a Ifeptis goochi and six 

 Neptis saclava but threw away the seventh, then very readily ate a Neptis agathci, 

 tasted and rejected Neptis goochi with very marked dislike, tasted and dropped but, 

 on its being at once reoffered, once more tasted and this time swallowed readily 

 Neptis saclava, again ta.sted and threw away with an emphatic shake of the head 

 Neptis goochi, and once again quite readily accepted and ate a Neptis saclava, 

 afterwards refusing without tasting a Neptis goochi. 



She then ate one large and two small grasshoppers, and readily ate two Neptis 

 saclava. 



I then gave her six small grasshoppers. After eating these she tasted and 

 rejected a Neptis saclava and then persistently refused it without tasting. Three 

 minutes later I again offered it and it was again obstinately ignored, but the bird 

 readily ate a Hesperia and a Baoris, tasted and threw away Lampides hcetica, 

 but on its being reoffejed tasted it well and swallowed it, then refused to touch 

 a Gatochrysops dolorosa, though she readily ate a 5 Pajjilio lycexis^ a 5 Leuceronia 

 argia, a Henotesia perspicua, a Byhlm, and a Neptis saclava. There had been some 

 slight delay between the last few offerings and the bird had once more become 

 somewhat hungry [or appetite had come in eating]. 



She now ate five small scraps of meat and then, quite readily. Precis cehrene and 

 Neptis saclava. On my now giving her two large pieces of meat (she refused the 

 next), she refused persistently and repeatedly to have anything to do with Neptis 

 saclava, but readily ate a Melanitis leda and a Pyrameis, also a Henotesia perspicxia. 



At this stage I gave her an enormous Acridium lineatum. This was battered 

 and crushed and finally with difiiculty swallowed. The bird now refused to touch 

 Neptis saclava, readily ate Precis cehrene, refused without tasting Eurytela dryope, 

 accepted Leuceronia argia $ , held it for a long time in her bill, evidently not 

 hungry, and finally swallowed it. She then refused, but changing her mind 

 tasted Precis natalensis 0f., and having done so, at once proceeded to crush and 

 swallow it, refused without tasting a Teracolus, crushed and ate with relish an 



