942 ME. n. H. BEINDLEY ON EEPRODUCED [DeC. 13, 



normally shaped claws. Mr. Pocock informs me that he has 

 noticed other cases of a similar nature. The production of the 

 normal terminal structure, associated with deficiency of proximal 

 parts, as in the above cases, is one of the chief characters of the 

 kind of regeneration defined under (b). 



Myeiapoda. — Newport (61) removed the normal 7-jointed antenna 

 of Julus by cutting through j.-^ close to its base, but without 

 apparently injuring j,. At ecdysis a dwarf 6-jointed autenna 

 appeared, each joint being shorter and thicker than the normal, 

 while Jg closely resembled j^. [But section of j^ resulted at 

 ecdysis in the production of an antenna having Tjoiuts, an iustauce 

 of regeneration of the kind desci-ibed under (a), which is cited 

 in this place for the sake of comparison.] 



Insecta. ColUmhola. — Lubbock (55) amputated the normal 

 6-jointed antenna of Orchesella by cutting through j^. At ecdysis 

 the antenna was reproduced in a 3-jointed form, which was 

 perpetuated through all subsequent ecdyses observed. J^ was 

 longer than j.^, while J3 was slender and rt,sembled;;g. In Tonwcertis 

 the result of pulling out j^ of the normal 4-jointed antenna was 

 the production and perpetuation of a 3-jointed antenna with J^ 

 likej'j. Lubbock's own observations on AetJieoccrus, and those of 

 de Geer (34), Latreille (53), and Bourlet (14) quoted by him, all 

 tend to show that fewer joints than the normal combined with a 

 resemblance of the actual distal joint to the normal distal joiut is 

 the characteristic condition of an antenna when reproduced. 



Ortlioptera-Saltatoria. Acridiidce. — Griffiiii (42) describes a 

 Gonqjhocerus in which the antenna had almost certainly been 

 reproduced. In this genus the antenna has normally about 

 23 joints, the 7 terminal joints being clavate. The case described 

 had 9 joints in one antenna and 2 in the other clearly defined, 

 but in each case the terminal joint showed faint constrictions 

 suggestive of incomplete division into from 3 to 5 joints. These 

 terminal joints or series of fused joints were clavate, and so bore a 

 resemblance to the terminal joints of the normal antenna. 



Ortlioptera-Cursoria. Phasuiidfe. — In this family the legs bear 

 tarsi v\hich are 5-jointed as in Blattidse, and as in the latter there 

 is ample evidence that when reproduced the tarsus assumes a 

 4-jointed arrangement. The probability that the latter condition 

 arises in connection with reproduction was first pointed out by 

 Coquerel (20). Previously to this the 4-jointed tarsus in Phasmids 

 had perplexed several entomologists, especially on account of the 

 asymmetry involved by the presence of one or two reproduced legs 

 in otherwise normal individuals. Gi-ay (41) had established the 

 genus Heteronemiaiov specimens of Bacteria mexicana \\it\i "small 

 hind legs." Westwood (80) had figured a Gyplwcrania with 

 reproduced tarsi, and devoted a new subgenus (Craspedonia) 

 to cases of Monandroptera inuncans with 4-jointed tarsi on 

 the anterior pair of legs. The error involved was corrected 

 by Coquerel. Percheron (64) described an Acanthoderiis and 

 Newport (61) a Lopaphus with one or more tarsi 4-jointed. 



