590 
CYRESTIS. By H. Fruhstorfer. 
grows again narrower, the rows of ocelli are prominent and complete and the yellow anal area is greatly reduced, all of 
which is characteristic of the western species, and still they are entirely different and cannot fail either singly or in greater 
numbers to make a wholly heterogeneous impression upon the specialist. The only great difficulty is presented by the northern 
subspecies of laelia and the species treated hereafter, which on first sight bear such a striking resemblance to each other, that 
we cannot dispel the thought of Nature employing the same methods in the islands to the west as well as to the 
east of the mainland of New-Guinea. But notwithstanding the great general resemblance, a closer examination of the median 
band reveals some wide and constant differences. The Bismarck and Solomon Archipelagoes have one species each, and in 
each of the two southernmost islands of the latter archipelago, in Ugi and Ulawa, one subspecies has been discovered”. (Maktjx). 
fratercula. C. fratercula Godm. a. Salv. The types are from Neu Lauenburg and are in the British Museum. It 
is distinguished from acilia in the relatively narrow white band tapering to a, point both toward the costa 
of the f. w. and the anal angle of the h. w. The yellow anal area is limited to a narrow approach to the anal 
ocellus. Specimens from Neu Hannover, which exceed all others in the deep brown bordering of the anal area, 
have the ocelli but faintly margined with yellow, whereas in 3 specimens from Herbertshoehe the yellow border 
is quite distinct. Quite characteristic is the shape of the median band, which is broadest where it passes from 
the f. w. to the h. w. and displays at the costal margin of the f. w. a sharp spur pointing towards the apex, 
formed by a small white streak above the subcostal; the extremity of the costal margin above this terminal 
streak is distinctly brown. At the anal angle of the h. w. the median band is, below the third median nervule, 
marked with a white terminal eccentric spot, assuming thereby somewhat the shape of a spindle. On the 
upper surface of both wings the rows of ocelli are complete; on the f. v. the fourth ocellus is elongate, reniform. 
On the under surface the two middle ocelli especially in the $$ are frequently reduced in size or disappear 
altogether. Anal ocellus roundish, reversed ocellus very distinct. The $$ are lighter and generally larger. From 
Mr. Ribbe I received several very small specimens marked ‘var. minor ’ from Neu Pommern and ‘var. 
obscura' from Neu Hannover which, however, aside from their reduced size, do not deviate from fratercula ; 
probably they are products of a long-protracted dry-season, which may in those regions last from 3—4 months 
ribbei. (Martin). - — ribbei Mart. This considerably smaller subspecies (Expanse 42 mm as against 48 mm in fratercula) 
has the white median band broader and more convex toward the outer margin, perhaps also somewhat purer 
white than fratercula. On the upper surface of the f. w. all the ocelli are round and about of equal size, similar 
to biaka ; the whitish line accompanying them distally follows the curve of the ocelli, surrounding their outer 
half in such a way, that it forms a series of outwardly convex arches and crescents, the points of which are inter¬ 
posed between each pair of ocelli. In the majority of typical specimens of fratercula this line consists of 
six streaks which are either straight or may be on the veins a trifle concave towards the margin. The reversed 
ocellus is very plain. The types are in the possession of Mr. Ribbe. As we know specimens from Neu Pommern 
as well as from Neu Hannover, which in their small size and some other characteristics approach ribbei, 
it is difficult to decide, whether this new subspecies is a true local form or only a seasonal form. About Mioko 
I read that its lagoon is changed to a meadow and that with the exception of bananas and yams it supports 
but a scanty vegetation. Probably also the species of ficus, which serve as food to the larvae of cyrestis, are 
stunted, which together with a long protracted dry-period would, according to my experience, bring about 
a morphological change in the butterflies; for this reason I refer to what I mentioned under fratercula (Martin). 
nitida. C. nitida Math, from the Solomon Islands, Guadalcanal’, Ysabel, Kulambangra, Malaite, Florida, 
Alu, Shortland, Treasury New Georgia, Facaro. The author has based his most painstaking description 
on specimens from Treasury Island (type in the British Museum). The white median band is much narrower 
than in acilia and fratercula ; commencing with a little white spot located between the brown costal margin 
and the subcostal vein it reaches its greatest breadth, where it passes from the f. w. to the h. w., grows on 
the latter gradually narrower and ends in a rounded point at the anal angle. But the increase in breadth is so 
gentle, that on the whole it appears fairly straight and almost equally broad, whereas in acilia and fratercula 
it bulges considerably and in the following form salomonis is broadest atthecosta ofthe f.w. whence ittapers toward 
the anal angle. Whilst the outer edge of the band is sharp and fairly straight, the inner edge is interrupted, 
irregular on the veins and in the $$ frequently blurred. The changes in the band observed in specimens from 
tHe different islands are hardly worth mentioning; it is narrowest and most uniform in breadth in specimens 
from Treasury and Florida Islands, whereas those from Kulambangra, especially the $$, vary most. In speci¬ 
mens from Treasury Island the dark border of the costal margin is broadest above the starting point of the 
band, occasionally to the exclusion of the small terminal streak above the subcostal vein, whereas in spe¬ 
cimens from Florida this border is narrowest. The ocellate chains are complete, the two middle ocelli are 
never reduced in size or obsolete; the anal ocellus is roundish, the reversed ocellus but moderately distinct. 
The white lines accompanying the rows of ocelli are on either side exceedingly variable; they may have given 
rise to the name nitida ; being invariably lighter and broader in the $$ and the outer always surpassing 
