( 476 ) 



(;")) The iiiiliviilu;il variation nt" the genital armature in the Chinese sul)species 

 is entirely independent of external eharactors. 



(0) The seasonal dimorphism in external characters obviously niari<ed in the 

 Indian P. cloanthus cloanthus does not affect the genital armature in any way, as 

 far as we conld ascertain. 



HI Papilio sarpedon; f. 90 to \A^. 



Though some jieojile have treated I\ sarpedon and /■'. cloanthus as belonging 

 to two diiferent genera, the insects are, nevertheless, very closely allied to one 

 another, more closely than to any other species. This does not only follow from a 

 careful comparison of the wing-]iattern of the two Papilios, but also from the 

 structure. 



The range of /'. sarpedon eomjirises the whole of the Indo-Australian llegiou, 

 including Jajian ((^xce])t the north of it). The number of subspecies into which 

 the insect has develojied is very great. As it is one of the commonest species we 

 could examine a large number of specimens, and to this it is due that we here came 

 across an individual which stands in the genital armature far outside the usual 

 limits of variation of the subspecies to which it belongs. 



If we select some of the extreme forms here treated as subspecies of one sjjccies, 

 for example the Indian, Celebensian, and the Solomon Island forms, their external 

 differences are so very conspicuous that one might easily be misled to consider 

 these forms sjiecifically distinct. A comparison of the representatives from all the 

 various h'calities, however, conviuces us that the differences in colour, ])atteru, 

 shape, and size between every two nearest allied forms are very slight, and do in 

 some forms not even apply to every specimen ; and we observe further that, where 

 the differences are constantly met with, the characters amount- quantitatively not 

 to more than the differences between the seasonal forms of Jajianese sarpeilon, 

 or than the differences found between certain Indian exam])les. Hence we think 

 it to be quite correct to accept Mr. Rothschild's statement that mH the forms dealt 

 with in the fidlnwing lines are subspecies of one species. 



As we now are acquainted with the more simjile armature of the valve of 

 P. cloa/if/ius, that of P. sarpedon will ])e more easily understood. A comparison 

 of f. '.)6 and 149 will at a glance show the great similarity in tlie apparatus 

 of the two species. The valve, though differing in outline from that of i'. cldanthus, 

 has the same a])ical sinus, and the armature has nearly the same ]iosition. 



The ventral lobe of the valve (f. 90, a) is longer, mostly broader, than, the 

 dorsal one (i) ; its ventral edge is denticulate, as in cloanthus, but there is only 

 one row of teeth, and the toothed portion extends farther down towards the base. 

 There is a good deal of variation . in the shape of the lobes and the depth of the 

 sinus, both in respect to individuals and to geographical races. 



The fold ((') formed by the inner sheath of the valve begins ventiully at the 

 base of the valve, as in cloatit/ius, runs in an oblique direction to somewhat beyond 

 half-way to the apex, turns round here towards the dorsal side, forms, when having 

 arrived at a level with the ventral margin of the dorsal lobe, a subdorsal tooth (c), 

 and then is continued in a longitudinal direction to form a dorsal ridge ((/). The 

 homology of the organs is obvious. The fold e is homologous to tlie fold e in 

 cloanthus, but is here less raised, takes a somewhat different course, and is at the 

 jioint where it curves round towanls the dorsal side often feebly toothed. The 



