300 Dr. Frederick A. Dixey on the 
Catasticta and its allies, are not represented in D. bella- 
donna, but in all the members of the pasithoe group they 
are not only present, but have become a very conspicuous 
feature in the coloration. It is not difficult to bring the 
other markings of D. pasithoe, D. thisbe and their allies 
into relation with those of D. belladonna, though the 
dark series S and M are as a rule less clearly indicated. 
The pasithoe and belladonna groups of Delias may thus 
be considered as on the whole the most ancient.* Inthe 
well-known D. eucharis we have the series S and M 
complete; the female, as is usual, showing less departure 
than the male from the more generalised form. As in 
Catasticta, no sooner have the dark series emerged into 
distinct existence than they begin to disappear ; they are 
much less recognizable in D. hyparete and D. agostina 
than in D. eucharis, and in D. hierte they reach the 
vanishing point. D..eucharis and its immediate allies 
follow the early condition of Delias in having the inter- 
spaces on the underside of the hindwing more or less 
filled in with yellow, but they depart from it in having 
lost the red basal patches, and in possessing a conspicuous 
submarginal band of red spots (series I) between the 
dark series S and M; these are best seen in D, ewcharis 
itself. An early stage of this red series is probably 
present in D. ceneus and the closely allied D. philotis, 
which seem to be linked with the still more primitive 
butterflies of the pasithoe group through such forms as 
D. belisama and D. descombesi (see p. 284). Regarding 
then D. cceneus as a’ new starting-point, we may trace 
from it in one direction the Indian group headed by 
D. eucharis, and in another the Australian forms 
D. aganippe, D. harpalyce, and D. nigrina. It is to be 
observed that both D. ewcharis and D. aganippe, though 
probably more recent than D. ceeneus so far as regards 
the underside of the hindwing, nevertheless retain in 
other respects more of the primitive pattern ; from which 
we may probably conclude that some form now lost, 
rather than D. ceeneus itself, supplies the true link with 
the belisama and pasithoe groaps, while D. coeneus and 
D. philotis survive to show how “series 1” acquired in 
* T follow, in nearly every particular, Mr. Wallace’s division of 
this genus into groups. See Trans. Ent. Soc. Lond., 3rd series, iv., 
p. 344, ete, 
