l02 bK. W. G. RlDfiWOOt) OX THE [Feb. 1 5, 



the preceding summer, and might have been of the same brood. 

 They wei-e collected at Maurenne, near Hastiere, in the province 

 of Namur, in May 1897, and were all treated in exactly the same 

 way, the left hind leg, tben between seven and twelve millimetres 

 in length, being amputated at the middle of the tibial segment 

 and left to heal. Five of the specimens completed their meta- 

 morphosis, but the sixth became arrested in its development, and 

 although kept under exactly the same conditions of life as the 

 other five, failed to make any progress. All six were killed in 

 October 1897, when Mr. Boulenger very kindly handed them over 

 to me, together with the above information as to their previous 

 history. 



While, thanks to Gotte (10), our knowledge of the normal and 

 regenerated limb-skeleton of Urodela is not deficient, the skeleton 

 of regenerated limbs of Anura does not appear to have hitherto 

 received any attention ; and it occurred to me that the best use to 

 which the material entrusted to me could be put w-as the pre- 

 paration and description of the regenerated cartilages. Bearing 

 in mind the close similarity found by Gutte to obtain between the 

 regenerated and the normal limb-skeleton of Urodela, a somewhat 

 similar correspondence was to be expected in Anura. But, having 

 regard to the greater specializatioia of the Anurau limb, it was 

 just possible that the restored skeleton might be simpler than the 

 normal. On surveying the results of the investigation one cannot 

 fail to be impressed by the closeness with which the skeletal parts 

 of the newly-developed limb approach those of the normal. 



While in animals other than Anura structural differences 

 between the regenerated and the normal Hmb may be explained as 

 phenomena of atavism [as claimed by Giard (7 and 8), Barfurth 

 (1. p. 113 (6)), and Bordage (3)], there is no evidence of such 

 phylogenetic reversion in the regenerated limb-skeleton of the 

 Anura under consideration. The astragalus and calcaueum are 

 elongated and are confluent with one another at both their proximal 

 and distal extremities. The remaining tarsalia are disposed exactly 

 as in the normal limb of the same age. Although in specimen C 

 four of the five digits have each one phalanx less than the normal, 

 there is abundant evidence in specimens A and B to show^ that 

 the typical number of phalanges for each digit can be reproduced. 

 And, lastly, the experiments throw no light whatever on the mor- 

 phology of the calcar, although one might fully have expected 

 atavism to be apparent here if anywhere. 



Mr. Boidenger has also communicated to me the very interesting 

 fact that, if the first appearance of the new limb be watched care- 

 fully, a single digit is first seen to grow out from the healed 

 stump, then another digit at its side, then a third, and so on in 

 succession. The tarsus and the distal half of the tibial segment 

 of the leg appear to be intercalated afterwards between the digits 

 and the 3.L.rap. The chief interest of the successive appearance 

 of the digits lies in the fact that this mode of development is 

 charade vistic of the Urodele limb. In the newts and salamanders, 



