88 
Psyche 
[Sept.-Dee. 
Besides being dissimilar both in specific expression of generic 
external variation and in the basic shape of the male armature 
(e.g., the “retrousse” spur [processus superior of the valve] 
in argyrognomon , the spare tapering weakly hooked falx in 
melissa ) these three unities are separated throughout the nu- 
merous forms that cluster around the three peaks of speciation 
by constant relations between certain parts of that organ when 
its dorsum is viewed from below. If F (“forearm”) denotes 
the length of the falx from its distal point to its elbow; H 
(“humerulus”) the length of the falx from elbow to shoulder 
point; and U the length of the uncus lobe from its tip to the 
shoulder of the falx, then the following three categories can be 
formulated: 
(1) argyrognomon: H greater than U, F/H smaller than in (2) 
(2) scudderi: H equal to U, F/H smaller than in (3) 
(3) melissa: H smaller than U. 
The Palearctic and Nearctic forms of argyrognomon are not 
only absolutely conspecific, but in one or two cases are strik- 
ingly alike exteriorly. Argyrognomon was presumably derived 
from a form of which the Central Asiatic agnata Staudinger, 
1889, is the closest image to-day. Scudderi and the Asiatic 
cleobis Bremer, 1861 ( ?subsolanus Eversmann, 1851 2 ) are, 
except for a more robust build in the cleobis organ, practically 
identical in genitalic structure (and share at least two peculiar 
underside characters), but either have not been in touch for a 
longer time or are coincident species i.e., separately evolved 
from initial argyrognomon- like structures ( scudderi decreasing 
in the argyrognomon H while cleobis arrived at the same pro- 
portional result by an increase in the argyrognomon U). We 
find these two on parallel lines which after passing through two 
coincident stages have widely diverged to produce melissa on 
one hand and ismenias Meigen, 1830 (Heydemann, 1931; 
insularis Leech, Verity, 1921, nec Leech) on the other. This 
scheme of course is not a phylogenetic tree but merely its 
shadow on a plane surface, since a sequence in time is not really 
deducible from a synchronous series. What seems certain, 
however, is that scudderi in its actual structure stands about 
2 Eversmann’s type of subsolanus, if it still exists, should be examined: his lucid 
description (grotesquely mistranslated by Riihl and thus copied by Seitz) seems 
to me to fit quite exactly the species known as cleobis Bremer. 
