4 LYCANID. 
be observed in the wings. In this there is very great diversity ; many genera possess no 
tails whatever; two genera of Indian Lycenide (Nacaduba and Arhopala) possess some 
species which are tailed, others which are tailless ; two genera exist (Megtsba and Nacaduba) 
in which I believe the same species to be furnished with, or sometimes to lack, a single 
filamentous tail to each hindwing. Numerous genera have one short tail, others have two 
short tails, others again have two tails of different length, and sometimes it is the inner 
one, sometimes the outer one, which is much the longer; lastly, some genera have three tails, 
and in these the middle tail is usually the longest. The number of the tails is variable 
in the sexes also. Zracta and Surendra have one.tail in the male, two in the female, while 
Zesius has two tails in the male, and three in the female. In the genera 7hamala, Lehera, and 
Araotes, the males have one short tail, which is greatly elongated in the female. The 
Jength and breadth of the tail is as variable as its presence or absence, or the number 
found in one wing. In Zycenesthes the tails appear to be three in number, but these are 
very minute, and are nothing more than a few cilia which are rather more elongated than 
the rest, and are very liable to be broken off or otherwise destroyed. Ina very large number 
of genera the tail is only about an eighth of an inch in length, and filiform; in a few 
genera it is very long indeed, a full inch in length, comparatively broad, and highly ciliated ; 
while one genus (Ze/¢us) has two such tails to the hindwing, one of which is, however, 
much shorter than the other, There is almost every gradation amongst Indian ZLycenide 
between those species with the shortest and those with the longest tails. In the species 
with long tails, it will often be found that the tails are twisted at the ends. No other family 
of butterflies can shew such diversity of tails as can the Lycenide. 
The “anal lobe’’ of the hindwing is usually an important structural character ; some genera 
lack it altogether, some have it very small, while others again have it very large: it perhapS 
reaches its greatest development in the genus Bindahara ; lastly, the genus Arhopala shews 
no trace of it in some species, has it very aa in others, and moderately large in others 
again. 
It is thus manifest that, in the “caudal” and “lobular” appendages of the hindwing, 
the Lycenide offer very variable features, which might be expected to offer characters of 
primary importance ; and toa considerable extent this is so, the presence and absence of 
tails, their length, form, &c., serving to separate really natural groups. Here again, how- 
ever, we must allow for many deviations from a strict uniformity in genera and in the 
characters appertaining to groups, and such deviations must be admitted as occasional and 
really unimportant aberrations. In combination with these appendages, the general form 
and the outline of the hindwing must be studied, as these also, though not perhaps suscep- 
tible of very close and minute definition, appear to be generally constant in the main groups, 
and on these Mr. Distant has relied for the determination of two out of three of his 
divisions of the Zycenide in his magnificent “ Rhopalocera Malayana,” vz. :-— 
Posterior wings without filamentous tail-like appendages near the anal angle «» CURETARIA, 
Posterior wings with minute filamentous or prominent tail-like appendages near the 
anal angle PAC ose soe ose eee oe 
Posterior wings convex, Breit as broad as long ase eee ws» CASTALARIA. 
Posterior wings more or less elongate, distinctly longer than broad ... se APHNARIAs 
I should have been glad to have adopted these primary divisions, but am unable to do 
so, as they are mainly based on the presence or absence of tails to the hindwing, and, ‘as I 
have shewn above, three Indian genera, which also occur in the region treated in ‘Mr. 
Distant’s work, viz., Megisba, Nacaduba, and Arhopala, have both tailed and tailless species. 
In the divisions or groups adopted in this work, the form of the hindwings has been 
utilized, as correlated with the style of the caudal appendages and with other structural fea- 
tures, and has been found to help in marking natural groups, and forming a natural sequence, 
which is generally in accord with that adopted by Mr. Distant. The Gerydus group has 
