Phenomena of Autotomy. 473 



ever upon which to base such a decision, and therefore, in 

 spite of its atKiiity to the Angohm race of T. evippe, I am 

 constrained to consider it a distinct species. 



The argument on which Mr. Marsljall bases his synonymy 

 of T. evippe and T. omphale, which lie regards as one variable 

 species, is based upon the confusion which existed previous 

 to its rearrangement in the drawers of those species in the 

 Museum collection — a confusion largely due to the incorporu- 

 tiou of accessions since the date of the first arrangement of 

 the genus about the year 1876 or 1877. He says that in 

 Eastern Africa T, evippe is " an intermediate seasonal form 

 of theogone-omphale " ; but this is certainly not the case, for 

 T. evippe is essentially, in all its characters, a wet-season 

 phase ; nor does it a|>])ear to extend in the East farther north 

 than Natal. 



[To be continued.] 



LIV. — Phenomena of Autotomy observed in the Nymphs of 

 Monandroptera inuncans, Serv., and Rhaphiderus scabrosus, 

 Serv. By Edmond Bordage *. 



In the month of September last year I succeeded in obtaining 

 larvae and nymphs of Monandroptera inuncans and Rhaphi- 

 derus scabrosus, in which I studied the phenomena of autotomy, 

 in order to compare them with those that I had observed in 

 the case of the adult insects. 



These phenomena were exhibited very clearly by the very 

 young larva?. On pinching hard the distal extremity of the 

 femur, 1 generally produced the separation of the limb. The 

 interval that elapses between the stimulation of the nerve and 

 the rupture of the limb varies from a few tenths of a second 

 to three or four seconds. The same operation could be 

 attempted successfully upon all six limbs. 



In the case of the older larvae and nymphs autotomy is 

 sometimes produced still more easily ; but it may happen 

 that it becomes irregular and capricious — a feature that we 

 have already pointed out in the adult Phasmids. We must 

 also make mention of the increase in the time between the 

 stimulus and the rupture. Although this increase is not 

 manifested in an absolutely universal manner, it is of very 



* From the ' Coinptes Rendus, t. cxxiv. no. 4 (.Jan. 25, 1897), pp. 210- 

 212 : from a separate impression communicated by tl;e Author. 



Ann. & Mag. N. Hist. Ser. 6. Vol. xx. 32 



