108 Some Experiments with a Polymoiphic Phasmid 



monograph^ of the group express a distrust of the character, and it 

 is now evident that the genus Cuniculina as described by the second 

 author must be dropped. The hornless % has not been satisfactorily 

 determined, but the horned J appears identical with Bacillus cuniculus 

 Westw., and it will be sufficient for this paper if the subject of the 

 experiments is assigned to Clitumnus sp ?, probably Clitumnus cuniculus 

 Westw. 



The first two experiments consisted in isolating a horned green J 

 and a hornless yellow $ from the original brood and, as the ^/s and J s 

 had been left together after reaching maturity, it is assumed that both 

 these % s had paired. 



The following table shows the distribution of the two pairs of 

 characters among the descendants of these two % s and their progeny. 



Note. The number following an individual shows the brood from which it was taken. 

 Brood 4 died young and is omitted. In Brood 10 the two hornless green ? s died when 

 immature and there is some doubt as to their colour. 



These results are puzzling in man}^ ways, yet certain facts stand out 

 clearly. It is evident that the characters of the presence and absence 

 of horns are Mendelian, as also are those of colouration, and with respect 

 to both it is noteworthy that no intermediates occurred, so that every 

 individual could be assigned definitely to one of the four classes. The 

 connection between the possession of horns and the green colour, 

 suggested by the original brood, evidently does not exist, and the 

 two pairs of characters are probably quite independent of each other. 

 Finally, and most important, is the definite evidence in brood No. 6 

 of the segregation of Mendelian factors in parthenogenetic reproduction. 

 There appear to be no previous records of this phenomenon and con- 

 sequently confirmation is desirable ; at the same time the absence of 



1 Die Insektenfamilie der Phasmiden, p. 196. 



