mo) Er 
the coast region from Sierra Leone to the Niger. Here the female 
is always white-banded (like pl. XXI, fig. 9), the intermediates 
between this form and the tawny male being absent, or at any 
rate not known to me. In Abyssinia, on the contrary, the female 
is tawny like the male. On the island of Fernando Po, in the Bay 
of Benin, the male has a whitish subapical band like our female 
figure 6, andits female is like the male except for this band being 
pure white. It is evident that the environment plays an important 
role in this geographical variation, and if in various districts of 
West Africa we further find the changes in Planema epea more 
or less faithfully repeated in Zlymnias phegea, we might be 
inclined to draw the conclusion that the similarity between the 
two Insects is the effect of the direct action of the environment. 
However, if this agreement is due to the physico-chemical 
action of the surroundings, it follows that the two Insects must 
be similar to each other wherever they exist side by side. Is that 
the case? 
The systematics of Planema epea tell us that this species 
is represented in Uganda by the geographical race paragea 
(pl. XXIV, fig. 30). In this paragea the sexes are similar to each 
other, the bands being whitish in both and so much reduced that 
the Insect is very different in aspect from the other geographical 
forms of P. epea. If we now compare the Uganda form of 
Elymnias phegea, we see at once that the above conclusion was 
hasty. The bands in the Zlymnias are not reduced as in paragea, 
but on the contrary, at least on the forewing, much wider than 
in the West African specimens of £. phegea. There is hardly any 
resemblance between the Uganda Elymnias and Planema epæa 
paragea. This remarkable discrepancy is readily explained by 
natural selection. In Uganda occur several broad-banded P/a- 
nema which are much more frequent than P. epea paragea, 
which is apparently a rare Insect, and the Zlymnias has assumed 
the garb of these other Planemo. But if natural selection easily 
surmounts this difficulty, it is faced by another. According to 
the systematist the bands of Planema epwa are tawny in both 
sexes in Abyssinia, reduced, diffuse and whitish in Uganda, tawny 
in the male and white in the female in tropical North-West 
Africa, etc. Can selection by enemies be a possible cause of these 
geographical changes in a Planema which is protected to a great 
extent by its relative unpalatability? Why were these changes at 
all necessary in a « model »? 
