173 



glasses are thoroughly cleajied every morning. The advantage 

 of glasses over boxes or any description of breeding-cages is that 

 the larv<e can be inspected daily, and if need be hourly, and, by 

 having a slip of paper attached to each glass, the changes can 

 be recorded as they occur. If there be reason, any observation 

 can be verified again and again. In case of D. archippus the 

 facts were so verified in single individuals kept in separate 

 glasses, and this was done with several broods of larvt^e. So I 

 can state positively that there are four moults, neither less nor 

 more. 



The change to chrysalis I shall describe from observations 

 made last season, and it will be noticed that the account varies 

 from the usual version. My attention was drawn to a paper in 

 the Entomologist's Monthly Magazine for Aug. 1878, by Dr. 

 J. A. Osborne, in v.-hich he declared that the process of pupa- 

 tion of the Nymphalidae had been erroneously described by au- 

 thors, and that the safety of the chrysalis at the crisis of pupa- 

 tion was assured by a membrane which connected the chrysalis 

 with the old caterpillar skin, and which was ruptured after the 

 booklets of the chrysalis had grappled with the silk ; and not, 

 as had been asserted, by the seizing of the old skin between 

 the joints of the chrysalis, and the climbing up to the silk 

 thereby. I watched the pupation of quite a number of larvse 

 of Crrapta interrogationis soon after reading Dr. Osborne's pa- 

 per, and found that a membrane or ligament always bound the 

 last joint of the chrysalis to the old skin till the pupation was 

 accomplished. In G-. interrogationis this is white, 2.5 mm. 

 lono-, flat, in shape of a long triangle, the narrow base of which 

 is a little forked, and the forks fastened to the anterior ends of 

 the two anal ridges on the chrysalis, which ridges or their 

 equivalents (as in D. archippus, where there are two rows of 

 black knobs) may probably be seen in any of the Nymphalidae, 

 and also in the Papilionidae. That this ligament is the sole 

 support of the chrysalis at the crisis of pupation, may be made 

 evident bv cutting off the caterpillar skin with scissors just be- 

 low the ligament as the struggle to reach the button of silk be- 

 gins, and so exposing the whole body of the chrysalis. The 

 contortions of the chrysalis will not be in the least abated 



