42 BULLETIN OF THE 



W. H. Edwards, of Coalburgh, West Yirginia, corroborated Dr. 

 Osborne's statement by observations on some of our Amerieaa 

 species, recorded in the Canadian Enloviologist of last December. 

 Prof. Riley records the result of a number of observations on 

 this subject, and thus explains the philosophy of the act which 

 has so generally misled observers. His studies have been made 

 principally with the larvifi of Vanesna antiopa. The principal 

 means by which the chrysalis holds on and rises at the critical 

 moment, is a stout ligament, which is, virtually, the shed intes- 

 tinal canal; not alone the lining, but the whole organ, which, as we 

 know, becomes sub-obsolete in the imago state of so many Lepi- 

 doptera. It is the ilium and colon, more particularly, which are 

 serviceable, and the ligament holds with such force around the 

 anus of the cast larval skin that it cannot easily be severed. 

 The rectum of the nascent chrysalis draws this in, or lets it go, 

 by peristaltic action of the sphincter muscles, the whole liga- 

 ment being drawn out as soon as the hooks of the cremaster 

 reach the silk. In addition to this ligament, which is of a red- 

 dish color, there are two lateral ligaments, also quite long and 

 strong, and of the color of the skin, which serve as auxiliaries. 

 These are the shed linings of the tracheae issuing from the last 

 or ninth pair of spiracles, which in the chrysalis become closed 

 or blind. These ligaments may be called the tracheal ligaments, 

 and seem to be somewhat specialized to aid in this important 

 act. Lastly, there is the membrane proper, referred to by Dr. 

 Osborne, which is virtually but the anal portion of the skin it- 

 self, or corium, caught upon the knobs at the end of the ridges 

 which usually form the ventral part of the cremaster. It consists 

 chiefly of the skin that lines the region of the rectum and the 

 anal prologs, and takes on a more or less bifurcated form from 

 the pulling power of the knobs during the act of withdrawal 

 from the larval skin. These ligaments Prof. Riley considers 

 constant physiological factors in the problem, most necessary in 

 those species which have the knobs imperfectly developed, and 

 acting even during the larval molts, and so holding the shed tkin of 

 Lepidopterous larva that it is worked to the anus in a shrivelled 

 mass, as a stocking is pushed to the toes ; whei*eas, in most other 

 insects, and especially in those where the metamorphosis is in- 

 complete and the change in the intestinal canal but slight, the 

 exuviiB are crawled out of rather than worked oft", the anal parts 

 not being held within the end of the molted skin, but really be- 

 ing the first parts detached. The membrane is a purely me- 

 chanical factor, and may not always be properly caught and 

 drawn out. It may also be severed without necessarily causing 

 the chrysalis to drop. Yet that it is an important aid to the 

 rising of the chrysalis there cannot be much doubt, and we find, 

 in the chrysalis of Paphia glycerium for instance, a totally dif- 

 ferent mechanical provision for clutching the membrane, namely, 

 a notch between the ridges around the rectum and the base of 



