THE EARLY STAGES OF EUSTROMA RETICULATA. 87 



end of the fourth segment, afterwards being visible only at the 

 segmental divisions, where it appears as an elongated spot. At 

 the anal end two such spots join to form a conspicuous red blotch. 

 The amount of this coloured line varies greatly in different 

 individuals. There is not the slightest trace of red on 

 the sides of the larva. The fairly broad white subdorsal 

 lines join below the anal aperture. Two broad white lines, 

 continuations of the subdorsal lines, extend down the outside 

 of the wide-spread anal claspers. Spiracles dull yellow, con- 

 nected by a delicate hair-like white line. Body sparsely sprinkled 

 over with minute black warts, each emitting a short black hair. 

 These warts are arranged in circular series around the middle of 

 each segment. Down the middle of the ventral surface is a 

 narrow white line, broken at each segmental incision. In the 

 full-grown larva the white segmental divisions are not so con- 

 spicuous. Colour of body still dependent to a great degree on 

 the food, although the anterior portion of the body is now 

 almost always opaque white, due to the full-grown larva's un- 

 doubted preference for the seeds. 



Habits of the Larva. — The young larvae are found usually 

 resting in various positions on the under sides of the leaves of 

 Impatiens noli-me-tangere, and their presence on a plant is 

 denoted by numerous circular holes in the leaves. The majority 

 appear to affect the midrib as a resting-place in the daytime, 

 especially when " lying up " for a change of skin, and in this 

 position are extremely like the narrow unripe seed-pods, which 

 themselves lie along the midrib on the under side of the leaf. 

 Several young larvae, however, were found stretched at full 

 length along the upper stems and amongst the flower-buds. 



In confinement, where a variety of food was provided, the 

 young larvae fed indiscriminately on leaves, flower-petals, and 

 flower-buds, but seemed to like the latter, which they hollowed 

 out completely. The habit of drilling round holes in the food 

 appears to be the natural instinct of the larva, and points, I think, 

 to the fact of the seed being the proper food. I never observed a 

 leaf eaten at the edge in the usual manner of leaf-eating larvae. 



In the last stage there is no doubt that the unripe seeds 

 form the exclusive food of the larva, and I doubt much whether 

 in confinement they can be brought to normal size if this diet is 

 not provided for them. On several occasions, when my stock of 

 seed-pods got low, I noticed the larvae wandering about in a 

 restless manner in search of food, and although fresh leaves 

 were introduced, they refused to touch them, and consoled 

 themselves with nibbling the curled-up pods from which the 

 seeds had been expelled. The ripe pods burst at the slightest 

 touch, and the loose seeds lying on the box-bottom were eaten 

 readily. In a state of nature the full-grown larva has a very 

 characteristic habit of resting in the daytime with the claspers 



H 2 



