1898.] APPENDAGES IS" THE ARTHROPODA. 9^9 



Collembola nothing is known concerning autotomj, and only 

 the reproduction of the antennm has been studied. Bourlet (14) 

 and in greater detail Lubbock (55) have made observations 

 in this case, and concluded that regeneration seemed to commence 

 from the actual point of amputation. 



Orilioptera-Saltatoria. — In many genera there is well-marked 

 autotomy of the posterior or jumping-legs, and, as shown by 

 Heineken {loc. cit.), Fredericq (loc. cit.), and Contejean (loc. cit.), 

 it occurs at the femoro-trochanteric suture. I am indebted to 

 Mr. J. Graham Kerr for the information that, as noticed by 

 him in the Paraguayan Chaco, administration of chloroform 

 to Tropinotus readily causes autotomy at the suture. It is 

 uncertain whether 2i point d'appui is necessary in Saltatoria. Till 

 recently it has been supposed that reproduction of the lost legs 

 does not occur in Saltatoria, on which point reference may be 

 made to the writings of Durieu (24), Fredericq, Peyei-imhoff (65), 

 and Werner (79) ; but Griffini (42, 43, 44) has lately described 

 some captured specimens which seem to indicate that, as in the 

 Cursoria, reproduction of the legs occurs in some at least of the 

 Saltatoria during post-embryonic development, and that it com- 

 mences from the femoro-trochanteric suture. In the nymphs of 

 certain forms regeneration of the antenna' has been observed by 

 Graber (39), after he had amputated them near the basal joint. 

 He also records that repair occurred in the wing-covers after 

 pieces had been snipped out with scissors. 



Orthopter a- Cursoria. — Many Phasmidse exhibit autotomy of the 

 le(/s during the later periods of immaturity and during the adult 

 state, but our information on the subject is not very extensive. 

 As he has described in an interesting series of papers, Bordage 

 (6, 6, 7, 8, 10) has found that autotomy at the femoro-trochanteric 

 suture was easily obtainable with several different kinds of stimuli, 

 though the latent period between stimulus and rupture was 

 considerably modified by such factors as age, sex, and mode of 

 stimulus. His experiments seem to show that autotomy may 

 occur either with or without a point d'appui being employed. 

 Scudder (72) did not observe autotomy in Diapheromera, but the 

 mutilated extremity of a leg eventually fell away up to the suture. 

 In these forms reproduction of the limb may occur not only at 

 the suture but from more distal points. 



As regards the lec/s of Blattidse, my previous paper gives the 

 facts which seem to demonstrate that their loss not infrequently 

 takes place by a feebly developed autotomy at the femoro- 

 trochanteric suture. Regeneration may commence at this point 

 or from the stump of either femur or tibia, but not from any part 

 of the tarsus. The stump of this latter region may be retained for 

 some time after mutilation, but it is invariably dry and shrivelled. 

 The antennce of Blattidse are certainly capable of reproduction. 

 The earliest observations seem to be those of Heineken (46) on 

 Leucoplicea. He cut oif the antennae " near the base," but did not 

 notice the effects of the injury or whether the reproduction 



