( 490 ) 



The facts are briefly as follows : — 



(1) Japan : lload black. Harjio with dentate ridge, witlwnt free ajiical 

 process ; one specimon like (:?). 



(2) Loo Choc Islands : Head as in (3). liarpe as in (1). 



(3) China and Formosa : Head red. Harpe without dentate ridge, with free 

 apical process ; one specimen with dentate ridge (=1) and with free apical 

 process (= 3). 



Tlie only inference from these facts logically possible is that neither the red colonr 

 of part of the hairs of the head, nor the form of the harpe, is in the Papilios in 

 question of specific value. The present case reminds one strongly of that of 

 P. sarpedon as explained on p. 478. 



15. Papilio aristolochiae ; f. 86 to 95. 

 The gronji of Papilios to which the present species belongs has tlie valve much 

 reduced ; the harpe does not vary to any extent, while the uncux (dorsal ]iart of the 

 tenth segment, according to Peytoureau) is very variable. To show the amount of 

 variation in one of the species, we figure the uncus of three Sikkim individuals 

 of P. ariMolflchiae. aristolochiae (f 86 to 88), of three Sambawa specimens of 

 P. aristolnchiae austrosKndanus (f. 89 to 91), and of four Buuguran examples 

 of P. aristolorliiai' anfiphus (f. 92 to 95). The variability of the uncus within each 

 subspecies of arixtolochiae is so great that we fail to jjerceive any character in that 

 organ which could serve to distinguish the three (in external features easily recognis- 

 able) subspecies by. 



B. Female Gexital Armature. 

 Peytoureau,* in his researches on the morphology of the last segments of the 

 abdomen of i\i& female Lepidoptera, comes to the conclusion that there are, as in the 

 mule, ten segments, the anal segment of the femole imago, consisting of two lateral 

 pieces comparable in form to the valves of the male and liaving the function of a 

 protector of the anus and the orifice of the oviduct, being homologous to the ninth 

 and tenth segments of the pupa and larva ; the orifice of the oviduct has the same 

 position as the penis underneath the anus between the ninth and tenth segments. 

 The variation of the anal segment is of no importance for our present purposes. The 

 orifice of the vagina is separated from the orifice of the oviduct, and is situated 

 between the seventh and the eighth segments; in consequence of the development of 

 the copulatory apparatus the eighth, not the seventh, segment has undergone great 

 changes. The fin'in of the segment, as well as of the vaginal armature, is in a live 

 specimen easily perceivable when one gently presses the abdomen ; in a dried-up 

 individual the apparatus is concealed, being entirely removed into the vaginal cavity, 

 the opening of which is closed by the seventh and eighth segments being in contact. 

 As it was, in the iirst jilace, not the aim of onr researches to find characters in 

 the morphology of the abdomen by which the higher divisions in the system of 

 Lepidoptera, such as families and subfamilies, could be distinguished, but to try 

 wliether the morphological characters of the last segments could be made use of in 

 diagnostic work relating to the lower degrees of division, genera and species, and 

 particularly whether within the limits of a species there was in the copulatory 

 apparatus a variation similar in kind and extent to that of the wa/e organs of coi)ula- 

 tion, it was necessary to study the organs in question i?> situ, in order to be able to 



• See note, p. 461. 



