INVESTIGATION OF THE PUSS MOTH 307 



success for this caterpillar, he came unexpectedly upon a 

 good-sized specimen feeding upon a willow. He was 

 about to seize it when the threatening attitude of the 

 larva and the protrusion of a pair of long filaments from 

 the hinder part of its body made him pause. He thought 

 for a moment of grasping the head-end, which looked 

 less formidable than the other, but as his hand came 

 near, the larva bent its filaments in that direction, and 

 Roesel shrank from provoking it. At last he found it 

 prudent to cut ofi" the twig on which the larva stood, 

 and let it drop into his collecting-box. Three days later 

 a number of parasitic grubs crept out of the unlucky 

 caterpillar, the sure token of an early death, but Roesel 

 soon got more specimens, and completed his study of the 

 life-history. Use made him bolder, and he came to 

 think that the threatening attitude was a mere feint ; 

 the larva had, he thought, no real power of injuring an 

 assailant ; modern inquiries have shown that this con- 

 clusion was premature. 



When the Puss-moth larva is alarmed, says Roesel, it 

 contracts its body, makes the hump on the third thoracic 

 segment more prominent, raises its fore part a little from 

 the ground, protrudes its filaments, and turns its head 

 towards the assailant. The head is set off" by a red hood, 

 formed out of the segment next behind, and a pair of 

 dark spots, which look like staring eyes. Neither Roesel 

 nor Reaumur dwells upon the resemblance to a face. 

 Roesel recognises that the tail-filaments or their sheaths 

 represent the last pair of false feet, found in other 

 Lepidopterous larvae ; he mentions their rose-colour, 

 their frequent retraction, and the curious fact that the 

 full-grown larva is very unwilling to brandish them. 



The tails of the Puss-moth larva reminded Roesel of 

 the projectile filaments behind the head of the swallow- 



