EUMELEA. By L. B. Prout. 
3 
‘> 
•J 
Borneo and Palawan, though these ought to be separable. It is not certain that Stoll’s unsatisfactory figure 
may not rather be meant for a markingless aberration or rubbed specimen of obliquifascia Warr., which is com¬ 
moner on Amboina. -— sangirensis Warr. is a more yellowish form from Sangir. I do not know its $. sangirensis. 
— marginata Prout, from Celebes and Sula Mangoli is in the less yellow than sangirensis but of a less deep marginata. 
red than rosalia and is characterized by having the apex of the forewing and generally a narrow distal border 
yellow. The $ either corresponds to this or is yellow, in either case with the markings rather broad 
and rosy.-— aurigenaria Warr. (3 a). This name has been given to the forms which prevail on Java and the Lesser aurigenariu. 
Sunda Islands, but is perhaps not tenable. They are more mixed with yellow and more strongly marked than 
rosalia but the are not so yellow as in sangirensis and lack the yellow apex of the forewing which distingui¬ 
shes marginata. Second line commonly more proximally placed. The $, when yellow, has the markings narrower 
and greyer than in marginata. ■ — attenuata Warr. The is rather darker than in typical rosalia, the lines rather attenuuta. 
thin, but is scarcely worth naming. British New Guinea. — From the Solomon Islands I have seen too few 
examples to be able to discuss the variation. 
E. infulata Warr is a striking species. Both wings proximally pale yellowish with rosy irror at ion, distal infulata. 
half bright rose-colour, bearing a white-yellow transverse band, on the forewing straight, on the hindwing 
curved. Apex of hindwing also white-yellow. Choiseul Island. 
E. semirosea is very close to some forms of rosalia but I believe a good species, as the wings are appre- semirosea. 
ciably more rounded, especially the forewing apically. The subterminal band is generally wanting, never strong, 
the first line often also wanting (when present, straighter than in rosalia ); second line fairly strong, more 
gently and evenly curved than in rosalia. — semirosea Warr., from the North Moluccas, has the mixed with 
yellow (especially in proximal half), generally with some yellow scales at apex of hindwing. $ yellow, with dull 
rosy markings. — phoenissa Warr., from Obi and Buru, has the $ deeper and more unicolorous red, as in typical phoenissa. 
rosalia, the hindwing without yellow at apex. $ as in the preceding. 
E. choiseulensis sp. nov. seems to connect unipuncta with obliquifascia. The is bright red, blotched choiseuien- 
with yellow, nearly as in unipuncta or fulvida, but has the second line straight and parallel with the distal 
margin as in obliquifascia, though more proximally placed; the characteristic blackish distal spot of uni¬ 
puncta is entirely undeveloped. The $ is yellow, with the speckles and markings bright red, not greyish as in 
unipuncta, the second line placed as in the $ but more noticeably crenulate. Described from 5 $$ and 1 £ 
in the Tring Museum, collected by Meek on the North side of Choiseul Island, December 1903. 
E. obliquifascia Warr. (3 a) is generally easy to distinguish by the nearly straight, oblicpie second line, obliquifas- 
which, though often thin and indistinct in the $, is almost always discernible; in the $, which has the ground¬ 
colour yellow, it is always strong and nearly always thickened. Widely distributed, occurring on Buru, Amboina, 
New Guinea and most of its islands, the Bismarck Archipelago and North Queensland. 
E. obesata Feld. (2 i). A rather broad-winged species, yellow in both sexes and generally recognizable obesata. 
at a glance by the very extended brown blotch on the hindwing. Only known from the Philippines. — ab. 
corpulenta Th.-Mieg lacks the brown patch of the hindwing. corpulenta. 
E. praeusta Warr. (3 a). Also very easy to recognize in the by the dark apical patch (leaving free praeusta. 
the apex itself) and the broad band across the middle of the forewing. In the $ these markings are only 
indicated by irregular dark dusting and a dark distal spot is developed, recalling that of unipuncta (2 i). 
North Moluccas. 
E. apicata Warr. possibly represents a local form of the preceding. The central line of the forewing apicata. 
is incomplete, as in praeusta $, but thinner, and is also thinner on the hindwing; the apical patch of the fore¬ 
wing, on the other hand, is more extended than in praeusta <§, reaching the 2nd median vein, and is as dark 
as in genuina (3 a). Kapaur, Dutch New Guinea, only two examples known. 
E. genuina Kirsch (= gemina Kirb., craspedias Meyr.) (3 a). A very constant species, quite distinct in genuina. 
the black-brown borders of both wings. Distributed in New Guinea, with Jobi, Goodenough and Fergusson 
Islands. 
E. algidaria is an interesting species, of rather small size, characterized by the whitish band of the fore¬ 
wing (much more obliquely placed than in infulata), which is continued as an apical patch on the hindwing. The 
ground-colour is always dark, but there seem to be at least three geographical races. — algidaria Walk., algidaria. 
from Mysol and Waigeu, is more mixed with purple than the following (especially in Waigeu specimens) and 
has the whitish band on both wings broad. — lugens subsp. nov. is dull brown-grey and has the band slightly lugens. 
narrowed, sometimes lightly speckled with the ground-colour. Near Oetakwa River, Snow Mountains, Dutch 
New Guinea, October to December, 1910, a pair in the Tring Museum. — constricta subsp. nov. (5 a) has the ground- constricta. 
colour of algidaria, the band of the forewing greatly constricted, more or less interrupted, especially in the 
