12 LYCANIDA. 
when the baffled pursuer almost invariably abandons the chase.” (* Rhopalocera Nihonica, ” 
p- 20.) b 
In forming a tabular index in linear order of the species of any family of the animal 
or vegetable kingdoms, it must be distinctly realized that any linear arrangement must be 
more or less artificial, and most naturalists will admit that the relationships between species 
can be most truly and naturally represented in the ramifications of a genealogical tree: or 
that allied groupstouch one another, so to speak, at several points, or show affinities on 
different sides and in different directions ; so that, when placed in a linear series, though in 
contact with genera and species to which they are manifestly related, they must be separated 
from others which appear also to be their allies: and that in attempting to form naturally 
continuous sequences there must be interpolations of more or less aberrant forms which mar 
the perfect continuity. 
In the selection of characters to serve as indices of primary or secondary importance 
in determining groups, no certain law or rule can be, or at any rate has yet been, adopted. 
Characters which serve in one order or family as of primary value are of only minor impor- 
tance in another, nor can any one organ or character be arbitrarily selected of more classi- 
ficatory valuethan another. The number and arrangement of pistils and stamens in the 
vegetable kingdom might naturally have seemed to Linnzus to be correlated with structural 
differentiations of primary value; but further botanical research has shown that this is 
not the case, and that a natural order seems to depend on combinations of numerous and 
varying characters. Inthe Coleoptera certainly it so happens that the articulations of the 
tarsus are so correlated with differences of habit and structure as to serve by themselves 
as characters of primary importance; but such pre-eminence of a single character is unusual 
in natural history. In studying the Indian Zycenide no single character seems to serve 
as a really satisfactory and constant guide for forming well-defined groups ; and it must also 
be borne in mind that no grouping, based on a study of the Zycenide of a restricted area, 
can be expected to serve for the classification of the whole of this family as spread over 
the earth’s surface; and that therefore any classification of Indian species only can be but 
provisional, though perhaps serving asa useful aid to future workers in a larger field. 
In seeking for some one structural feature whereby to define the primary groups, if not 
rigidly, yet sufficiently for the purposes of the Key, the subcostal nervules of the forewing 
may seem to offer suitable characters. The number and position of these nervules give much 
assistance in determining the relationship of genera and species in some families, but, as pointed 
out by Mr. Hewitson (‘* Illustrations of Diurnal Lepidoptera’’—Zycenide), these characters 
differ even in the sexes of the same species in some of the Lycenide. Such instances are 
however but rare, and if these be treated as abnormal deviations from general rules, the arrange- 
ment of these nervules can be utilized as characters of, at any rate,secondary or tertiary value 
in classification. Thus, if adopted as primary characters, they will be found to separate Zhecla 
far from Zephyrus, Euaspa, &c. ; Yasoda from Loxura ; Lycenesthes from the tailless blues” ; 
Amblypodia from Arhopala, &c. If however the arrangement of these nervules fails to 
define the limits of primary groups they may be usefully employed as minor characters. 
In discussing these nervules it seems necessary ev fassant to point out that in this work 
the view, advocated by Westwood and adopted in the introduction of this work (vide vol. i, 
pp. 11, and 17), as to the subcostal nervure being continued to the margin of the wing after throwing 
off 2, 3, or 4 nervules only towards the costa, is constantly maintained. Mr, Distant regards 
the terminal portion of the nervure as an additional and last nervule, and then he notes, 3, 4, 
and 5 nervules in the cases where only 2, 3, or 4 are specified in this work ; this is of course 
only an apparent discrepancy ; and it is really of no importance which view is maintained, so 
long as the student realizes the system adopted in the work whichhe is consulting. Mr, 
Moore in the “ Lepidoptera of Ceylon” often treats the upper discoidal nervule (as defined 
