CHARAXES. By H. Fruhstoreer. 
731 
orchomenus subsp. nov. forms the transition from lampedo to lmnnibal from Celebes. Its median white band of orehomenus. 
the under surface of the hindwings is more distinctly pronounced than in the specimens from Mindoro and Min¬ 
danao of the Tring Museum. In the island of Palawan. Very rare. — zephyrus Btlr. is a deviation or insular race zephyrus. 
without the white band on the under surface of the hindwings and extremely narrow median area of the upper 
surface. Habitat the Philippines; the place, where it was found, not exactly ascertainable. — catulus -sub-sp. catulus. 
nov. was discovered by Doherty in Sangir. There is only one £ in the Tring Museum, with a broader discal 
band composed of reddish-brown spots, on the under surface of the hindwings. Flying time February and 
March. — mangolianus R. and J . is a considerably darkened satellite-insular race. Submarginal spots of the mangolia- 
forewings white instead of yellow, as in lampedo. Sula-Mangoli, one $ in the Tring Museum, one or two <$$ in 
the Coll. Staitdinger. — hannibal Btlr., an extremely rare butterfly from the north of Celebes, was founded hannibal. 
by its author upon 2 9$ from the north and south of the island, which were taken to be the two sexes of one 
species by Btjtler. from the Minahassa with hardly half as large preapical spots of the forewings, as in the 
southern race figured by Rothschild and Jordan (Nov. Zool. V, table 7, f. 1). The yellow band of the hind¬ 
wings narrower, shorter, distally not bordered by blue. Under surface brownish-grey. Forewing without a 
trace of red spots and with entirely darkened or vanished apical spots. Hindwings with a much narrower white 
longitudinal band and extraordinarily large brown spots in the submedian area. Very rare, occurring always 
but singly. The $ (Butlers type) in the British Museum, the only specimen known hitherto, hardly differs 
from the C, except the somewhat broader bands. The under surface blackish-grey, with a slight purple hue 
over it. — jordani subsp. nov., the race from South Celebes, for the first time depicted by Rothschild and jordani. 
Jordan, extraordinarily well differentiated. $ much larger than the $ of Ch. fabius hannibal, with a much 
broader yellow band than the northern form. Hindwing above with yellow anteterminal spotting being absent 
in the $ of hannibal. $ beneath white-grey the median band expanded. Across the whole upper surface a 
peculiar light waxy gloss. South Celebes, collected by Doherty* in both sexes near Maros in August-September. 
I saw a specimen sitting at the bank of a brook on the way to the Peak of Bonthain, at an altitude of about 
800 m, in the eastern part of South Celebes. The $ in the collection of the British Museum (my type) is of 
a more imposing shape and beneath much lighter than the $ figured in Nov. Zool. V. They are presu¬ 
mably temporal forms. Another form allied to hannibal is found in the Talaut Islands from where Doherty 
sent a $ to the Tring Museum. 
Ch. polyxena is the most common and at the same time most multiform species. Distinguished by 
the heteromorphism of the sexes, the polychroism of the $$ and in the continental races also of the In 
the Himalaya the areal races are, besides, subject to the influence of the seasons. The most interesting fact 
is the occurrence of specimens with a white band of the forewings, beside those of a plain brown ground-colour, 
which variability formerly caused the founding of many species, until Rothschild and Jordan united all 
the dubious forms into one collective name. Early stages figured at first by Moore, hardly distinguishable 
from those of the South Indian species Ch. imna. If, however, the original figure of 1857 is not recorded, the 
larva of polyxena hierax has longer horns on the head than that of Ch. imna, and the pupa is somewhat more 
slender. Larva snail-formed, the broadest in the middle, posteriorly somewhat tapering. Ground-colour beauti¬ 
fully green, beneath yellowish, with a pink-speckled, yellow lateral line. On the back a pink, large round spot 
with white ringlets, and on both sides three more, smaller pink spots being absent in imna. Pupa green with 
red ventral dots, polyxena is a butterfly of the lowlands inhabiting, according to Niceville, the low valleys 
of Sikkim. We little know what altitudes it reaches. The Javanese insular race does not go higher than 800 m, 
from Sumatra the Battak Mountains are mentioned as habitat, where the Sumatran race goes up as far as 
about 1200 m, while the Tenasserim form is reported to occur up to 6000 feet. An interesting fact is the 
increased darkening of the species towards the east, and the disappearance of the light-banded forms towards 
the south. I do not remember of having captured white-banded $<§ in Siam, and on the Malayan peninsula 
there exist only brown while very rarely there occur yet now and then white-striped $$. The maximum 
of the development is attained by polyxena in Sikkim and Assam. Further to the west and east, the specimens 
grow rather scarce, and crowds, such as were observed on the river-banks in Sikkim and Assam, have no more 
been noticed at the periphery of the range. — polyxena Cr. (Vol. I, p. 171, t. 61 a, b) represented in the white- polyxena. 
banded form, occurs also without the white median area being replaced in the $ by a dull yellow brightening 
(f. sinensis R. and J .). polyxena is generally darker than Indian specimens and the black apical area penetrates sinensis. 
as far as into the cell of the forewing, in the white-decorated qC- Rare in Hongkong and in the Pro¬ 
vince of Kwangtung. According to Rothschild also in the Yangtse Valley and according to Leech on the 
Omei-shan and near Mupin in West China *). — mahawedi subsp. nov. was discovered by myself in July near mahawedi. 
Than-Moi and in August-September near Chiem-Hoa in Tonkin, and by Pavie near Luang-Prabang. It forms 
a transition from the Chinese polyxena to Indian hierax; the black marginal area of both the wings, however, 
more extensive than in hierax and vicarious types, both in the white-banded and unicolorously brown forms. 
Rare and very local. Type in the Tring Museum. — agna Moore is the oldest (reliable) name of specimens from agna. 
*) Larva according to Kershaw on Acronycha laurifolia Bl. on the upper surface of the leaves. Imago flying in 
Hongkong all the year round. Egg hemispherical, smooth (?), yellowisch, deposited singly on the foodplant. 
