EXPLANATION OF FORM AND COLOURING. 32] 



tasted and rejected a Lytta designata, but eagerly accepted and ate a Eitrytela 

 hiarhas and a Mycalesis campina. 



[A good illustration of the value of the defences of gloss and toughness. 

 L. designata placed below E. hiarhas and perhaps M, campina.'] 



Exp. 111. — December!. Hungry; early morning ; no food yet. Crushed and 

 readily ate the Carabid beetle, Polyhirma cenigma, in spite of its strong, sour, 

 vinegary smell ; crushed and rejected the Chrysomelid, Hemiphracta ahyssinica, 

 readily ate a second Oarabid, Platynus sp., refused, then tasted with hesitation 

 and rejected a larva of Epilachna hirta, and refused to touch E. hirta, imago. 

 After a great deal of persuasion, however, she tasted it slightly and rejected it. 



[That is, the Polyhirma and the Platynus were preferred to the Hemiphracta, 

 and quite likely also to the larva of Epilachna hirta, and to E. hirta, imago, and 

 all these rejected insects were probably vei-y much disliked, as the roller was 

 very hungry.] 



Exp. Wlti.— Becem'ber 2. Tasted and at once rejected a Chrysomelid beetle, 

 Exosoma himaculata, and an Aletis monteironis, but ate with the sfreatest readiness 

 a Mycalesis campina. 



After a further interval without food I gave her a large black highly-polished 

 Cetoniid, Diplognatha gagates. Though very hungry she completely failed to 

 eat it. Time after time she accepted and attempted to crush it, or even got 

 so far as to commence to batter it, but the beetle every time slipped from between 

 her mandibles and shot across to the other side of the cage to the bird's intense 

 annoyance. She then ate readily not only a Myc. campina but an Aletis mon- 

 teironis, offered as before with all wings attached. She then once more went 

 through the same performance with the Cetoniid and completely failed to eat it. 



I then fed her up on grasshoppers till she persistently refused to eat another, 

 although several favourite species were offered. I then thrust through the bars 

 the Diplognatha gagates and used my utmost powers of persuasion to obtain its 

 acceptance. The roller would, however, have nothing whatever to do with it, 

 evidently not realizing — for the wings, too, are black — that it had been rendered 

 easier to crush, but she readily ate a Catacroptera cloantha with all its wings 

 attached, afterwards once more refusing to touch the Cetoniid. 



[Summary : — 



1. Myc. campina. ] ^ , , 

 „ r r. <7 • . • • r ^<^^- cloantha. 



M/xosoma J I. Aletis monteironis. J 



himaculata, I 3. Diplognatha gagates (complete failure)]. 



Exp. 113. — December 3. After a little hesitation tasted well and rejected an 

 Acrcea johnstoni with pale buff hindwing patch, and refused most persistently 

 even to taste the larva of a Coccinellid, owing no doubt to its general likeness in 

 form though not altogether in colour to that of Epilachna hirta with which she is 

 already acquainted. 



