1898.] APPBNDAGOiS IN THE ARTHBOPODA. 951 



of the normal. Grarbowski (33) has more recently described a case 

 in Hygrocardbm where a leg was similarly affected, and Bateson's 

 series of examples from the antennae of Forficula (2. p. 413) seem 

 for the most part to belong to the same category. In this genus 

 the number of antennal joints is usually 14, though specimens wiih 

 only 13 or 12 joints are not infrequent. Bateson found that in 13- 

 and 12-jointed examples j^ was markedly and j^ somewhat longer 

 than the corresponding joints in 14-jointed examples. In the case of 

 18 antennse from adults measured by myself, 12 had 14 joints, 3 

 had 13, and 3 had 12. In the 13- and 12-jointed specimens /., was 

 of about the same length as in the 14-jointed specimens, but in 5 

 out of the 6./^ was distinctly longer than in 14-jointed specimens. 

 Among the 6 cases of few-jointed antennae the more distal joints 

 were longer than the joints in the same positions in 14-jointed 

 specimens in 3 instances, and of practically the same lengths in the 

 other 3 instances. So that here again is manifest the tendency for 

 the appendage with abnormally few joints to approximate the total 

 length of the normal by increasing the lengths of its individual 

 joints. As Forficula is an orthopterous insect it is of course quite 

 possible that some of these cases of few-jointed antennae arose in 

 connection with reproduction. Bateson inclined to the belief that 

 the symmetrical condition of many such cases indicated a congenital 

 origin at least occasionally. But in the light of the evidence that 

 in Blattidae the mechanism of reproduction is able to bring about 

 symmetry in size between a normal and a reproduced tarsus on the 

 same pair of legs and between two reproduced tarsi on the same 

 pair, it seems possible that a similar compensating control may 

 exist over other cases of reproduction. In rheir general features 

 these exceptional antennse of Foificula approach the certainly 

 reproduced antennae of Myriapoda and CoUembola on the one hand, 

 and the abnormal and appnrently congenital antennae of certain 

 Hymenoptera and Coleoptera on the other. But much more 

 evidence regarding the reproduction of the antennte in a series of 

 selected forms must be forthcoming before we can say anything 

 as to the relationships between these peculiar appearances when 

 seen in genera with such different life-histories. 



In cases where the departures from the normal structural 

 arrangement are known to have arisen as reproductions, it is of 

 course permissible to regard them as in some sense analogous with 

 bud-variations in plants ; and as in their case, so also in that of 

 arthropod appendages, the idea has been advanced that the dis- 

 turba,nces seen are the results of insufficient and unequal nutrition. 

 Though no doubt the removal of an appendage does produce an 

 unusual demand on the nutritive channels directed to it, it would 

 appear that any failure to deal with the special circumstances of 

 the case is expressed rather in the small size of the reproduced 

 structui-e than in its morphological features. For it is character- 

 istic of the Tracheate groups at least that if any new growth at all 

 is revealed at the ecdysis succeeding injury, it is in a sense a 

 complete appendage and not an amorphous bud. The mechanism 



