890 
CASTALIUS. By H. Fruhstorfer. 
more extensive than in noliteia, though it does not attain the width of the Javanese race. The maculae beneath, 
particularly on the hindwing, are more prominent than in noliteia and elna, the spots of the Sumatran and 
Borneo-forms besides surrounded by a pale yellow periphery. North East Sumatra, Perak, Borneo (Kina Balu>. 
liilina. Banka. — hilitia Fruhst. Upper surface much more allied to noliteia than to elvira, in its size somewhat inferior 
to Sumatran elna. The under surface almost throughout hued yellowish, the spots, particularly on the hindwing. 
rhode. much more intense than in the Javanese race. Nias, rather common. — rhode Hpffr. Described by Hopffer 
without a more exact statement than from ,,Celebes“. I place the type, however, to the north of the island, 
because most of the specimens which Meyer brought along and handed over to Hopffer in order to describe 
them, originated from the north of Celebes. Besides Pagenstecher (Abhandlungen Senckenberg 1897, t. 18. 
fig. 1 c) has apparently figured a $ of the northern form from Donggala, so that the northern race was also 
pictorially produced, rhode is a highly specialized race, almost a distinct species; but if I subordinate it never¬ 
theless as an insular race of the collective species, it is owing to the under surface entirely agreeing with that 
of C. elna elna. The clasping-organs also show certain differences, though they are less developed than e. g. 
between Lycaenopsis albocoeruleus from the continent and the Sumatran race of this species. There are differences 
exhibited by the uncus being roundish instead of pointed, the valve being prolonged and more deeply indented, 
and the oedeagus being apparently more slender. Most probably we shall yet find in the satellite islands of 
Celebes other forms that will help to bridge over these contrasts. One thing, however, is certain that rhode 
begins to splinter off specifically, which fact is not to be proved of the colorially more modified form of C. ethion 
from Celebes, C. ethion Missus Fldr. North Celebes, Toli-Toli, November, December 1895 (H. Fruhstorfer 
rhodana. leg.), Donggala, Surroundings of Palu, January, February, very rare. — rhodana Fruhst. This Castalius cap¬ 
tured at first by Doherty, later on by myself in the South of Celebes, differs from C. elna elna by the width 
of the white median stripe on both wings. Beneath the white ground-colour is also predominant, and the black 
maculae are much reduced. South Celebes, Samanga, November 1895; Lompa, Battan, March 1896, ca. 1000 m. 
C. mindarus (153 a). It is not impossible that this species originally proceeded from C. elna or that 
it replaces it in New Guinea, mindarus at any rate represents the most easily recognizable species of Castalius, 
being of the largest habitus and the under surface being the most plainly marked. Only the Castaliine basal 
stripe has remained. Anatomically, C. mindarus is entirely isolated. The valve already reminds us of that 
of the group of species Tarucus, and as in Tarucus telicanus Lang (Vol. I, p. 293), the broad basal part of 
the valve sends forth one dagger-like appendage each, which, however, attains in mindarus nearly three 
times the length of the point of T. telicanus. The tegumen is extremely primitive, with a clumsy, rounded 
uncus. The oedeagus is moderately thick, with a cuneus sharply bent downward. Little is as yet known of 
the range of the species, and it has even escaped being inserted in Kirby’s catalogue, and this is also the reason 
why Ribbe has inserted Thysonotis (!!) mindarus in his List of diurnal lepidoptera of the Bismarck Archipelago 
as the only Lycaenid without mentioning its origin, and it may be that Pagenstecher has for this reason not 
at all mentioned it in his hastily finished Fauna of the Bismarck Archipelago. Thus of the more modern authors 
only van Eecke remains who conscientiously registered the New Guinean form. Two areal or insular races: 
mindarus. mindarus Fldr. (153 a). Described from Dorey where Wallace discovered it. Doherty has apparently not 
found it again in the Humboldt Bay, for Grose-Smith does not mention it in his list of Doherty’s collection. 
In my collection there is a $ from Dorey. A number of specimens from different parts of Kaiser Wilhelms 
Land correspond well with it. A 2 from the Dutch part (3rd Dutch New Guinea Expedition) is much larger. 
vocetius. North and South Dutch New Guinea, Kaiser Wilhelms Land. — vocetius Fruhst. $. The black distal margin of 
both wings above is nearly twice as broad as in mindarus, so that the median band being lactean, not yellowish 
as in mindarus, is reduced to half its width. Islands near New Guinea, patria not specified (Coll. Sciionberg). 
New Pomerania (Ribbe). 
Group of species Caleta. 
We now come to a singular group of Castaliinae, by the affinity of which no author was hitherto struck: nor is 
this astonishing since the specimens are very rare. The series of forms dealt with here interests us for two reasons, in 
the first place as to its relationship, in the second place as to its geographical range. 
According to the results of the examination of the clasping-organs, C. decidia and C. argola having hitherto been dealt 
with as separate species, are undoubtedly one and the same species. Geographically, however, they are remarkable, because 
these Castaliinae number among those rare species inhabiting India, the Philippines and Celebes in common, being, however, 
absent in Macro- and Micromalayana which is situate between. It is, moreover, very probable that the present gaps 
in the range will be closed by future explorations. The oldest name in question for our misjudged collective species is 
C. caleta, which had fallen into oblivion in the entomological literature since its description, except its insertion in Kirby’s 
catalogue. Lycaenids from Celebes had not yet been dealt with, and it seems that since Wallace it had remained reserved 
to me to rediscover the species in this island, for neither Doherty nor Ribbe brought them to Europe. 
There is no doubt as to whether C. caleta and C. argola belong together. More doubtful is whether C. decidia belongs 
to them, since the shape of the uncus is somewhat different. But we already know from Lycaenopsis albocoeruleus that 
the continental and Sumatran races differ to a remarkable extent in the configuration of the clasping-organs and that they 
nevertheless belong together. Here we have a parallel fact, and it is very easy to imagine the uncus being prolonged like a 
proboscis. The same is the case with the contours of the valve which in decidia exhibits only more sharply extended and 
divergent teeth than in the Chaeturian allies argola and caleta. 
