932 MB. H. H. BEINDLET ON EEPEODUCED [DeC. 13, 



structures as a whole will be considered. A regenerated appendage, 

 or part of an appendage, is always smaller than its fellow, provided 

 that the latter is of congenital origin or is a reproduced structure 

 of earlier date. This natural state of things was first understood 

 rightly by Reaumur, the pioneer of the study of reproduction of 

 lost parts. He corrected the assumption of previous authors that 

 such instances were cases of congenital asymmetry. Nearly all 

 accounts of the reproduction of Arthropod limbs agree in stating 

 that if there are still several ecdyses to be accomplished, the 

 reproduced limb grows with special rapidity so as to approximate 

 or equal in size its congenital fellow. 



In Crustacea observations in this respect are recorded by certain 

 of the authors already mentioned (17, 18, 68), from which it 

 appears that some of the appendages of Decapods when reproduced 

 attain their normal size more rapidly than do others. There is, 

 however, considerable want of uniformity of result for the same 

 appendage, and Brook has recorded that temperature, the kind of 

 food, &c. are important factors in the matter. 



The special rapidity of growth of regenerated appendages in 

 Spiders, Myriapods, CoUemboIa, and Phasmidse has been recorded 

 in works already referred to, and also by Fortnum (25) in the last- 

 named group. 



There is evidence that in Crustaceans the regenerated appendage 

 more frequently attains equality with its congenital fellow than 

 is the case in Tracheates — a feature which perhaps has some 

 explanation in the fi'eer mode of growth seen in the former 

 group. 



In Blattidse my own observations show that the growth of 

 reproduced appendages is very rapid. Measurements were made 

 with a micrometer-eyepiece of a few nymphs of Stijlopyga orientalis 

 averaging "8 cm. in body-length and therefore quite young, the 

 body-length of an adult being about 2*0 cm., as opposed to a 

 length of -5 cm. in newly hatched young. I measured the total 

 length of the tarsus in these '8 cm. nymphs, and the total length 

 of the cast cuticles of the corresponding tarsi just after ecdysis 

 and apparently before any appreciable shrinkage had occurred. 

 In four instances of normal tarsi the average increase of length 

 after ecdysis was 13 per cent., while in four cases of reproduced 

 tarsi the increase was 29 per cent. 



An obvious result of the specially rapid growth of a reproduced 

 limb is that the disproportion in size between it and its normal 

 fellow is less in cases where regeneration has occurred early in the 

 life-history than in those in which it has taken place near maturity. 

 For instance, the tarsi of the third pair of legs in 20 adults of 

 PeripJaneta americana, taken haphazard from indinduals in which 

 one of the tarsi was normal and the other reproduced, showed by 

 measurement that if the length of the normal tarsus be taken as 

 100, the mean length of the reproduced tarsi was 96"o. On the 

 other hand, the mean length of the reproduced tarsi of seven 

 nymphs averaging -53 cm. in body -length was found to be 87'1, 



