ELYMNIAS. By H. Frith storfer. 
377 
that it lives on the Rotang palm, but has hitherto evaded detection in consequence of its admirably protective- 
coloration and its retiring habits. Yet Martin frequently found the elegant green hanging pupa, decorated 
with fine red double streaks. The collective species itself is local, horodimorphic and subject to individual 
variations, which frequently mislead authors into setting up “new species”. It is often impossible to distinguish 
the cJc? from the undularis $<$, a proof of the close relationship of the two species. This makes the $$ all 
the more remarkable, since those of nigrescens must be described as euploeid, those of undularis as salaturoid. 
The discontinuous distribution of nigrescens is also remarkable, especially in the Macromalayan region, where 
Borneo, the Malay Peninsula and its Satelite Islands have nigrescens in common, while in Java the continental 
undularis reappears. This north-indian-Javan affinity is repeated within the genus in Elymnias casiphone, 
and probably owes its origin to the tertiary land connection between Java and the Continent.-—Larva in Su¬ 
matra on the Rotang palm, on Borneo according to Martin’s observations on the Coco palm. The white, 
spherical egg of nigrescens lies slightly flattened on the leaf, becomes gradually yellow, on the third day it 
is citron yellow, and on the last day (the sixth) before emergence of the larva it is dark and discoloured. A pupa 
found in the open is fastened on the under side of an old, hard coco leaf, on which it lies rather than hangs; 
it is a sappy yellow green with two processes on the head, containing the palpi, and there is a pointed nasute 
prominence on the thorax. This latter, a median line running backwards to the ventral segments, the processes 
on the head and a short, thick longitudinal mark right and left on the eight ventral segments are all yellow, 
bordered with red. This red bordering is on the inner side of the two first, on both sides of the third, which 
is lighter yellow, and only on the outer side of the remainder. The wing sheaths of the pupa become nigrescent 
by midday on the day before emergence, in the evening the entire pupa is deep black, the butterfly appears 
very early, towards 5 o'clock in the morning, Avhile it is still quite dark. The Elymnias pupae, of which 
Martin knows three species from the Celebes, panthera, nigrescens and heivitsoni, is of a peculiar type, and 
differs greatly from the, certainly closely allied pupae of Amathusia, Discophora and Thaumantis, which are very 
similar to each other. -— formosana Fruhst. (87 b). Comes very close to the dry form of tonkiniana Fruhst. — formosana. 
$ usually dark blue also on the hindwing, which has strikingly large white submarginal spots. Disc of both 
wings with a dark red gloss. Distal margin of the hindwing broadly black. Takau, Sept.-Oct., Taihanroku, 
July, August, and in other localities, nowhere scarce on Formosa. - — hainana Moore, judging from the hainana. 
which alone are at my disposal, differs, like formosana, in the very broad pale violet gloss on the forewing, which 
extends to the distal margin. Hainan. - — tonkiniana Fruhst. This and formosana differ in common from all tonkiniana. 
the other now known races of nigrescens in the much broader, dark claret distal margin on the upper surface 
of the hindwing. The majority of the have relatively large, blue submarginal spots, of which the upper 
ones are sometimes confluent, yet without ever forming complete bars, as in nigrescens from Borneo or Sumatra. 
In the dry season the blue is paler, and related to this is also the disappearance of the red distal borde¬ 
ring of the hindwings. My type is based on the wet season form. The $ has white, only peripherally blue 
submarginal marks on the fore wing and a rnfescent disc. Hindwing with red distal border and white submar¬ 
ginal spots. Tonkin, Haiphon, in November, and Central Annam, Phuc-Son, Nov.-Dec. — $-fa. depicta Fruhst. depicta. 
includes the dry form. Forewing uniform dark steel blue, with dull blue, not white centred submarginal spots. 
Hindwing unicolorous black. Tonkon, Chiem-Hoa, August—September. -— beatrice Fruhst. in the specific Beatrice. 
malayan region, therefore Malay Peninsula, Borneo and Sumatra, including perhaps the extreme south, §§ maris 
colore. “It appears to be necessary to treat briefly of my material of Sumatra nigrescens, as Fruhstorfer declares 
himself to be in doubt about the Sumatra race in consequence of insufficient material. On the island-continent 
of Sumatra there are at least three different forms of undularis, possibly more, as I have no material from 
the West coast and from Atjeh at my disposal. In Deli, Northeast Sumatra, the males taken were all very, 
slightly blue spotted on the forewing, with the very dark marginal bar on the hindwings but little developped, 
the also had only exceptionally the bluish-white spots on the forewing, such as are always well developped 
in the $$ from the Malay Peninsula; they are even frequently exactly like the AS, only the ground colour is duller 
and paler. In Central Sumatra (Indragiri) on the contrary have the AS a richer blue attire than is found elsewhere, 
the 3—4 apical spots coalesce into a broad, brilliant blue band and on the terminal margin there are in addition 
three other, very deep blue spots, the upper surface of the hindwings is almost unicolourous with only a very 
indistinct brown marginal zone. The commonest species of the region, which inhabits the gardens and lighter 
woods of the plains throughout the year. The A has a double tuft of hair, consisting of an inner and an outer 
section. The larva undoubtedly lives on the rotang palm, but must have either an excellent protective colouring 
or very retiring habits. On rotang palms set in tubs in front of my house in Bindjei I often found distinct 
traces of the ravages of caterpillars, but in spite of careful search, never the larva itself, but frequently the ele¬ 
gant green hanging pupa, with fine red double streak” (Martin). Valve decidedly smaller than in undularis 
from Sikkim, distally strongly attenuate, the lateral dentate npcurved margin considerably narrower, furnished 
with a number of delicate dentations. The following variations have been described; typica Fruhst., forewing typica. 
with the usual blue submarginal spots, which coalesce towards the costa.— decolorata Fruhst., forewing entirely decolorata. 
without blue spots or with only an obsolete costal spot. Both from Sumatra. -— - beatrice Fruhst. (87 b), fore- beatrice. 
IX 
48 
