90 Psyche [Sept.-Dee. 
ginal ornamentation 3 cannot be treated here; but a few words 
in regard to the disposition of the extradiscal series as I under- 
stand it may be of use. What we see as a transverse, more or 
less sinuous “line” or “row” of spots seems to me to be the 
outcome of two unrelated phylogenetic phenomena. The “up- 
per” part of the “row” (from the last radial interspace to the 
last median one) is formed by spots having radiated fanwise 
from the discoidal owing to an apicoid extension of the wing 
texture; the “lower” part (spot in Cux and the two Cu 2 spots, 
separated by the memory of an nervule) have been pulled 
out from a subcellular position (in the proximal corners of their 
respective interspaces) presumably by a cubitoid extension 
which did not necessarily occur at the same time as the other. 
Had not a third phenomenon taken place — namely the ap- 
pearance and expansion of subterminal ornamental markings 
(“caudae pavonis”) which held the advancing spots at bay — 
the latter might have gained the practically praeterminal posi- 
tion which they reach in some Glaucopsychince. This is why 
the classical conception of a row of ocelli as the result of a 
statically placed line or band having broken up into spots, 
seems to me absolutely irrelevant to the understanding of the 
Lycaenidae pattern. Insofar as spots have been evolved in this 
family, they occupy different positions in different species or 
genera, and what we see is not the remnants of a definite band 
in a definite place, but this or that stage of a more or less co- 
ordinated longitudinal movement of spots distad along the inter- 
spaces (certain comet-tail traces of this progress are sometimes 
caught and fixed aberrationally) . In a word it is not a row of 
squares on a chessboard, but a shifting line of attacking pawns. 
Lycaeides argyrognomon Bergstrasser 
As represented by my material, the nearctic argyrognomon 
forms, contrary to those of melissa , may be for convenience’s 
sake divided into groups A and B (“white underside” and 
“fawn underside”) and each may be subdivided again into 1 
and 2 (“weakly marked” and “strongly marked”). A1 to A2 is 
expressed by argyrognomon anna Edwards, 1861 (Proc. Ac. 
3 The latter shows a tendency towards obliteration throughout the Nearctic 
Lycceides — a feature unknown racially (except perhaps in the case of the 
Corsican argyrognomon bellieri Obthr) among western Palsearctic forms and 
paralleled only by certain Central Asiatic ones. 
