TRANSACTIONS OF SECTION D. 517 
destroyed by parasitic insects. Since all these had been collected when full- 
grown (i.e., they had escaped the dangers of infancy), the destruction at all 
stages, including ova, is probably greater. So that less than 23 per cent. is all 
that is left for destruction by both predaceous insects and vertebrates. 
The matter may be expressed as a formula. If X is the total number of 
offspring, then X-2= number destroyed by all influences. Putting P for those 
destroyed by predaceous insects, p for those succumbing to parasites, and V for 
those destroyed by vertebrates, then 
V=(x—2)—(P+p) 
V may thus be taken to indicate the degree of edibility to vertebrates, since 
the number destroyed by parasites, &c., must bear an inverse relation to the 
number destroyed by vertebrates. Could P (destruction by predaceous insects) 
be accurately estimated, a figure could be given for V, which might be called 
the edibility index; it would probably be found to be high for unprotected 
insects and the reverse for protected insects. 
On the other hand an eminently edible insect may escape vertebrate foes 
through the perfection of its procryptic resemblance to its surroundings, or, 
according to the Batesian theory, of its mimetic resemblance. In this case V 
would be low, not because the insect was distasteful, but because it escaped 
attack; so that in the case of an insect known to be edible, V would be the 
index, not of edibility, but of success in disguise. Such insects would also be 
highly parasitised. In the case of the larve of the mother-of-pearl moth I 
believe V to represent the index of agility in escaping vertebrates. The larva 
is highly edible. Some years ago I collected large numbers as food for green 
lizards, which eagerly devoured them. But I found a very large proportion 
indeed were parasitised. The larva lives in a tube of nettle-leaf and is 
extremely adroit at escaping through one end of the tube directly the other end 
is investigated, and it seems likely that it must escape birds in this way. 
5. Pseudacreas and their Acreine Models on Bugalla Island, Sesse, 
Lake Victoria. By Dr. G. D. H. CarpEnter. 
In Uganda are four forms of the Nymphaline genus Pseudacrea, which I 
have shown by breeding to be of one species—namely, Pseudacrea curytus, 
Linneus, as was suggested by Dr. Karl Jordan. This species occurs in other 
forms in West and Kast Africa and in Natal; but everywhere closely mimetic 
of a Planema in that locality, being either sexually dimorphic or polymorphic, 
as the Planema is also dimorphic or polymorphic. The four Uganda forms are : 
Hobleyi, mimicking Planema macarista (dimorphic). 
Terra, mimicking Planema tellus (monomorphic). 
Obscura, mimicking Planema epwa paragea (monomorphic). 
There is also a female form of hobleyi, with almost the colouring of the male, 
which mimics Planema poggei, in which both sexes are alike. This Planema is 
also mimicked by another Pseudacrea—namely, Ps. kiinowi. A collection of all 
these butterflies made without selection during 1912 and January-February 1913 
on Bugalla Island, Lake Victoria, showed that the models were extremely scarce, 
the mimics of the species’ eurytus extremely abundant and very variable, so 
that intermediates between the several types of hobleyi, terra, and obscura were 
very plentiful, thus :— 
Model Mimic 
1 Planema poggt . . 2 Pseudacrea kiimowt. . «© + «+ + 9 
Ps. eurytus, form hobleyi, 9 with ¢ 
colouring Aa oe tice grate 2 
2 Planema macarista . 17 Ps. ewrytus, form hobleyi PEIGA, cody et 68 
Intermediates between hobleyi terra : : 75 
3 Planema telus . . 33 Ps. ewrytus,form terra. wwe 103 
Intermediates between terraandobscura . 37 
4 Planemaepeaparagea 75 Ps. eurytus, form obscura 3 26 
Intermediates between obscura and hobleyi 45 
