METAMORPHOSES OF BRAZILIAN LEPIDOPTEIU. 83 



I Sim informed by Mr. A. Libert, late of Trinidad, that 

 he has seen similar tubes in that island. 



'a, Tubes on large piece of bark, swollen out in 

 centre into bulb containing pupa. 

 Pupa. ■{ h, Bulb cut open, showing pupa. 

 Full-fed, 13th June, 1882. | 

 Imago, 2nd Sept., ,, ' •' ' 



When the caterpillar is full-fed, it draws up the tube in 

 the middle, and swells it out into a bulb five millimetres in 

 diameter, in which it changes to the pupa state. The tubes 

 are by this operation reduced from eleven to six centimetres 

 in length. 



On June 13th some of the caterpillars were full-fed, and 

 preparations for pupation began. By June 27th nearly all 

 the tubes had the bulbs formed, but pupation did not take 

 place till later, for the caterpillars could still be traced darting 

 into the bark if the tubes were disturbed. 



, Full-fed, 13th June, 1882. . 



Imago. \-r n i a i ^ 1 81 days, 



(imago, ind bept., et seq. ' "^ 



The tiny moths appeared at the beginning of September. 

 When they emerged they hid themselves in the cracks of the 

 bark, and did not seem at all inclined to fly away. When 

 touched they gave a kind of jump, and "shammed dead." By 

 holding my "kilhng bottle " below them, and touching them 

 on the head, I was able to kill them without any damage. 



Explanation of Plate : — 



Fig. 1. Amphonyx Tapayusa, n. sp., larva, p. 19. 



,, 2. Aneurocampa Mingens, larva, p. 21. 



,, 3. Automeris ophthalmica, n. sp., larva, p. 25. 



,, 4. Attacus JacobsDse, larva, p. 28. 



