35 
Whether homology of crypton has connection with advances in 
evolution, or whether being an outcome of certain habits, through the 
lack of such habits, there has been no advancement to an heterologous 
crypton may well be considered open to discussion. But even in 
absence of any decision on this point, undoubtedly the slightest study 
of undersides, and perhaps especially of the crypton, must heighten 
one s impressions as to the importance of the characters of the under¬ 
sides. 
But one observation bearing on the point emphasising the 
possibility of the crypton having especial lessons has been partially 
referred to under the Vanessid remarks. In Aylais ( = urticae, etc.) 
the crypton is not only not nearly, in effect, of underside or upperside 
colour, but does not carry the same markings. Some portion of the 
band marking of polaris is much more common on underside than on 
upperside in this country, i.e., is more nearly fixed on underside, but 
the spots which ichnusa lacks are seldom, if ever, present in crypton, 
as compared with uppersides, even where they are enlarged and touch 
on uppersides. 
But before leaving the subject of the crypton, a passing note may 
be made here as to its greater liability to variation as compared with 
phaneron, at any rate in some groups. As in Polyowmatus phlaeas 
and in Lycaenids, for instance, if statistics were compiled of all possible 
observations of all variation of undersides year by year, aberrations of 
the crypton, I am confident, would be found to be much more frequent 
than those of phaneron on homologous undersides, and probably as in 
A. selene, A. aylaia, Melitaeids, and many others, notable variation 
would be found commoner in heterologous crypton than in accom¬ 
panying phaneron, with extended field observations. 
One firmly believes that in some reputedly rarer aberrations, viz., 
the ab. antico-obsoleta, of L. corydon and certainly of allies, as com¬ 
pared with ab. postico-obsoleta, e^en the reverse may be truth, but that 
to determine the presence or absence of variation of the crypton of 
Lycaenids is a matter of extreme care or patience, and practically 
demands the use of chloroform to be exhaustive, whereas the deter¬ 
mination of aberration in phaneron or its absence is a matter of the 
minimum amount of observation. And at all events some obsolescence 
of spots on upper wings as of basals in Agriades and Polyommatus as 
far commoner than obsolescence in hindwing is probably a generally 
recognised observation. 
In the consideration of undersides, such species as show such 
designs as in this country do E. hyperanthus and C. davits var. rothliebii 
require special consideration beyond the particular amount due to each 
species, and as a suggestion is offered the following series of observa¬ 
tions on rothliebii. 
The protective value of the underside of rothliebii is especially 
noticeable. In the case of the most southern forms the underside 
renders the identification of the settled imago extremely hard, usually 
impossible, when it is resting as is its (?invariable) habit in clumps of 
heather, Calluna, or sometimes E. tetralix. The brown colouration of 
phaneron, as well as the ringed eye-spots, are evidently better suited 
to match the heather than would be the plain gray wash of the 
northernmost forms. That on different mosses in W r estmoreland, 
xiv. 
