The latter oecur especially on the lower surface of the forewings 
and on both sides of the hindwings, while the former show them- 
selves in different aspects on the upper surface of the forewing. 
Now could these different types of modification be proved to 
correspond to the. distinctions between the genera, they certainly 
would be of important assistance in tracing the boundaries of the 
latter. As a matter of fact this seems to be the case to a certain 
extent, and undoubtedly the founders of the genera have been led 
amongst other features by the different character of the colour pattern. 
So for instance 1 do not doubt that the author of the genus Charagia, 
Waker, has been influenced by the conspicuous green shade, which 
in the majority of species colours the upper surface of the forewings. 
In the same way it cannot be denied that Leto venus, the single 
species of the South African genus Leto, is distinguished from all 
other Hepialids by the large silvery blotches, more or less regularly 
arranged in transverse rows all over the upper side of the forewings. 
Again we might be led to suppose that the members of the genus 
Dalaca might be easily recognized by the exceedingly curious 
character of their wing-markings, looking like the letters of an 
antique inscription standing out in high relief upon the brown- 
grey background, with which they contrast by their lighter shade 
as well as by the greater size and different shape of the wing-scales, 
which enter into their composition. Finally to cite one more instance 
the genus Pielus seems to differ from other groups of Hepialids not 
only by its special slender build ana elegantly pointed wings, 
showing a certain superficial similarity to Sphingidae, but moreover 
by the longitudinal white streak of fantastically broken contours, 
running from base to tip over the middle of the wing-surtace. 
Yet all these peculiarities as well as other special features of the 
colour-pattern can be shown to be the ultimate degrees of moditi- 
cation of the original design, occurring in a less grade in other 
genera or to be for other reasons lacking in importance for the 
determination of generic boundaries. 
The latter case is illustrated by the green shade in the majority 
of the Charagia-species, for not only do we find this shade replaced 
by a brown one in others, but this may be even the case with 
specimens of the same species (Ch. thermistis and virescens). More- 
over in Ch. lignivorus the forewings are partly green, partly brown. 
As to the relation between these two shades, | think it the most 
probable supposition to consider green as a modification of brown, 
instead of the reverse. For in a number of families and genera 
of Lepidoptera a certain percentage of green forms occurs beside 
