CASTALIUS. By H. Fruhstorfer. 
885 
that specimens with a white disc of the forewing belong to the dry period. This will be also correct with respect 
to the continent. But according to the material of the Coll. Martin, both forms fly together in Sumatra 
in August, whilst all the other dark hamada of the Battak Mountains are labelled with the date ,,February- 
MarclT‘ and the lighter specimens are of October. It is certain, however, according to my material, that in 
Java only the light form occurs. Javanese specimens differ also from those from other habitats by the white 
discal brightening passing also over to the hindwings, for which reason the name: nivata Fruhst. (154 f) was nivata. 
introduced for them, nivata is so abundantly decorated with white, that it may be denoted as a white lepi- 
clopteron with a black margin, whereas the Sumatrans remain greyish-black with an occasional brightening 
only on the forewing. West Java from an altitude of about 1200 m, in the surroundings of Sukabumi. 
C. mahanetra Doh. Described from the Padang Ranges, Perak, discovered again in Sumatra by Dr. mahanetra. 
Martin, inhabiting the deepest forests and being very rare. Martin in 13 years found three specimens of which 
I owe him two in my collection. Particularly the $ is very ne&t, almost resembling a neotropical Dynamine. 
Forewing black with a purely white discal area. Hindwing snow-white with a small blackish-brown costal 
dot. The spotting beneath turns light brown, in its distribution recalling much rather Castalius than Taraka 
hamada. Antennae and formation of palpi, however, exactly as in Taraka. North East Sumatra, July, December 
Coll. Fruhstorfer). The clasping-organs of both Taraka- species were already examined by Doherty 
who quite correctly ascertained that owing to their structure Taraka is to be joined to the Lycaeninae. 
I illustrated them in the Tijdschr. voor Entomologie. In hamada the lower part of the uncus is horizontal, 
the upper part rounded, rising towards the apex. The valves with a rough tubercle at the apex. In mahanetra , 
however, the valve resembles a bird’s head with a high rounded crown and the beak turned down. Doherty 
also calls the ,,prehensores very simple", without lateral branches (apophysis) of the uncus, whilst Dr. Chapman 
wrote to Swinhoe that the genitals in no way approach any of the species hitherto examined by him. 
Group of species Castalius Hhn. 
Eyes bare. 
C. ananda is the species of the genus Castalius anatomically most closely approaching the genuine 
Lycaena, whilst in the colouring above it somewhat reminds us of the Tarucus-species *). By its marking beneath, 
however, it is closely allied with the Indomalayan Taraka hamada. Uncus short, clumsy, anteriorly rounded 
off, with a relatively short apophysis lateralis. Valve extraordinarily broad, bent up high, deeply indented, 
in the ventral part with two dehiscent, sharp points. Oedeagus slender and besides still more tapering off in 
the distal part. Three geographical forms: ananda Nic. (153 b, c). Indian Empire: Sikkim, Assam, Chin Hills, ananda. 
Nilgiris, North Kanara. The differs from the figured $ by whitish areas in the anterior part of the disc of 
the forewing, so that it resembles 2$ of eastern forms of Tarucus plinius. — vileja Fruhst. is a well separated vileja. 
race, above much darker violet than continental specimens, with a narrower black border of all the wings. 
Beneath, however, lighter, of a purer white, covered with smaller black maculae. In the 2 the whitish patches 
are still more developed than in the continental race, the black spotting beneath more transparent, but the 
magnificent steel-blue reflection is absent particularly on the forewing, whilst it embellishes so much the Sikkim- 
form. Discovered by L. Martin in North East Sumatra. — waterstradti Drc. (153 b) proclaimed by Druce waterstradti 
as a new ,,species*‘ is a locals race of ananda. The rare $ shows like the Sumatran form a purely white under 
surface, whilst it otherwise approaches more the continental form by the violet hue on the upper surface. Kina- 
Balu, type in Coll. Staudinger. 
C. rosimon. Phyletically surely one of the oldest species and at the same time the most persistent 
of the whole genus. Its range extends from India to the Lesser Sunda Islands, although there do not 
occur any particularly distinct geographical races. But as far as the physical conditions cause it, there are 
both on the continent and in some groups of islands temporal forms of which the products of the dry period 
are recognizable by very scanty black spotting and by their small size. In the hot plains of Pegu there 
are, according to Bingham, specimens with discoloured brown instead of black maculae beneath. Dr. Martin 
found similar specimens also near Calcutta in March 1895. Clasping-organs highly specialized, by the extremely 
narrow uncus allied to the Nacadziba; valve likewise entirely different from all the others, dorsally and ventrally 
homogeneous, ending into a long, sharp point. Oedeagus, as the only one in the genus, long subuliform, 
recalling that of Nacaduba dana Nic. and Orthomiella pontis Nic. rosimon F. (=maimonF., clyton Cr., coridon rosimon. 
Cr.) (153 a). — An extreme temporal form has been called chota Sivh. Type from Poona, but occurring also ctwta. 
in other parts of North West India. — rosimon is distributed on the whole Indian Peninsula, from the foothills 
of the Himalaya, and absent only in the deserts. Ceylon, Assam, Malayan Peninsula, Tonkin, Siam, Annam 
(H. Fruhstorfer leg.). — In Sikkim the species is found thoughout the whole year in the Terai and the foothills 
*) Bethune-Bakep. combines C. ananda and C. fasciatus with the ,,genus“ Tarucus. 
