( 237 ) 



leiiresentcd by a low ridge : betweon the sternite and the penis-sheath there is 

 at each side a strongl\- chitinised, needle-like process (PI. XXV. f. 2, )i). The 

 clasper is divided apicall}' by a longitudinal slit (PI. XXV. f. 2) ; the ventral 

 )irocesses of the clasper are fused together, completely or partially, to form a single 

 ventral jiroccss (PI. XXV. f. 2, pr), while the submesial processes (pi/r and pf//) 

 are quite unef|ual. The details of the armature are ver}' different in the specimens 

 from diflerent countries, and there is also some individual variability. 



?. The eighth tergito i.s mesially sinnate ; on the upper surface of the lobes 

 there is a jinijectini;- cone or lobe (PI. XVIII. f. 5. ^), or there is no further 

 armature. The vaginal plate (PI. XVIII. f. 1. 4) agrees better with those of the 

 members of CIiihIh than with the plate of (kntafns : the antevaginal part is very 

 restricted, membranaseons, with a low, stronger chitinised, smooth ridge {nep) in 

 front of the vaginal orifice (1'); the post vaginal jilate (pi'p) is very large, this 

 jilate vi.sil)le in the specimens after removal of ]»art of the scaling of the seventh 

 sternite : this armature different, like the eighth tergite, in the different subspecies. 

 The antenna similar to that of c?, the middle and basal segments much higher 

 than broad ; the horizontal diameter of the ventral part of the transsection about 

 half as long as the diameter of the dorsal part. 



Larva green, granulose, with two yellowish green dorsal lines and whitish 

 oblique side-bands, granules white, head triangular, horn rather long. — Food-plant: 

 " Lnssorah " (N. India), " Tanteitai-lugai " (Philippines). 



Pupa not described. 



linb. Ceylon, N.W. India, N. India, ( 'hina, Philippines, probably also on the 

 Sunda Islands. 



The five subspecies are nearly all so very different in the structure of the sexual 

 nrmature (compare PI. XXV. f. '2, and PI. XXXIV. f 7. 8. 0. 10) that one might 

 be induced to consider these forms as being S[)ecitically distinct. But as they are 

 clearly geograidiical representatives of one another, and as, further, the material in 

 collections is quite insufficient to show us the range of variation of these individually 

 variable forms — only from X. India a series of specimens is available — we treat 

 tlieni as subspecies. 



ri. P. trif//h''it//fi hitrritiix snbsp. nov. 



cJ ? . Body and wings cinnamon-isabella colour, or cinereous-grey. ^Markings 

 aa in tnlinmtun (imlfiti/a, but the dentate line between tlie discal and postdiscal 

 lines of the forewing nearly as distinct as in (Iciitutim. The distal margin of the 

 forewiug not scalloj)ed, convex in middle. 



$. Tenth abdominal tergite (PI. XXV. f 4) much shorter than in trilineatiis 

 iiniliiiiin, the sternite higher, more strongly chitinised, the distal edge slightly 

 bisinnate, with a short, narrow, mesial lobe, t^lasper divided into a dorsal and 

 viMitral lobe (Pi. XXXIV. f. S), bnt the incision not so deep as in triUn. andntun, 

 and the lobes differently shiii)ed : the ventral one especially disagreeing in being 

 roinided and ])rovided before the end with a transverse fold which inclines apicad 

 and gives the lolje the appearance of being transversely cleft ; the ventral 

 impair jirocess common to the two harpes is apically bent towards the right side 

 (PI. XXXIV. f. 10) ; the figure gives the process in a dorsal view, and shows the 

 connection between it and the; processes of the iienis-fnnnel (ptlr and pdl) ; the 

 process jidr is either at'ute, as in figure, or obtuse. 



?. The eighth tergite has no sjiecial armature on the dorsal surface, the 



