THE CANADIAN ENTOMOLOGIST. 115 



close case, This is shifted frequently as the larva grows, and affords 

 food as well as shelter. Papilio Troiltis, when it issues from the egg, 

 proceeds to cut a slit at the edge of the leaf and folds down a bit thereof, 

 stitching it closely ; it lines the interior with silk and on this hes con- 

 cealed, going out when forced to feed. This feeding is done at the upper 

 end of the leaf, and when the shelter gets to be insufficient, another leaf is 

 sought and treated in same manner. P. Palamedes has exactly such a 

 habit. On the other hand, P. Turnus lies exposed to view on the upper 

 side of a leaf, on a bed of silk, which forms a sort of bridge as the leaf is 

 Somewhat drawn together, permitting water that falls on the leaf to pass 

 beneath it. The larva of P. Asterias rests on the stem of its plant entire- 

 ly unprotected. The larvae of Colias lie extended on the leaves, on the 

 upper sides, along the mid-ribs, well protected by the resemblance in 

 color to the leaf they feed on, clover or Astragalus, or Amorpha. 



And so each species has its larval peculiarities which repay study. 



I usually confine young larvae, particularly when careful examination of 

 them is desired, in glass tubes, and later transfer them to half-pint jelly 

 glasses (tin-topped). These tubes are corked tight, and the leaves will 

 keep fresh a long time in them. But they are usually examined once in 

 12 or 24 hours, and the larvse removed to fresh tubes. The date of 

 hatching is noted on a slip of paper, also the length of the larva, and 

 this paper is pinned to the cork. Every morning a measure of length is 

 taken and noted down. So the moults successively, and every moult is 

 described, and all the changes set down in a note-book. If the larvse are 

 in a glass, a slip of paper is held down by the cover with all particulars 

 written on it. Sometimes the individuals of a brood are kept singly 

 through all their stages, but most often it is only necessary to watch 

 for the successive moults, and to note the first and last larva to 

 moult in that particular stage. The approaching moult may always be 

 known by the swelling of second segment, and when any larva is observed 

 in this condition it is separated from its fellows. While moulting injuries 

 are apt to be received, which invariably result in the death of the larva, 

 and therefore it is best to separate the subject to prevent disturbance from 

 other larvae. Many larvae are cannibalistic, and if out of regular food 

 and hungry, will devour their next neighbor without compunction ; of this 

 sort is P. Philenor. As I make alcoholic examples of each species at 

 each moult, when possible, so I save the casts of faces at each moult for 

 future reference and study, using small homoepathic tubes for this purpose. 



