selves away altogether during the months of drought; and although 

 several kinds of butterflies remain in evidence, they are at that time 

 usually much scarcer in individuals than they are during the rains. 

 The cousequence of this state of thmgs is that the struggle for life 

 becomes much keener in the dry than in the wet season. If a species 

 is to survive this period of stress, it has to employ some highly 

 efficient means of protection from its enemies, who from the scarcity 

 of their provender are very much alive to any chance of a meal that 

 presents itself. The wonderfully close resemblance to such an object 

 as a dead leaf affords in all probability an especially good means of 

 protection, and no doubt many individuals of these species of but- 

 terflies owe their survival through the period of exceptional danger 

 to this mode of defence. In the wet season, on the other hand, 

 insect life is abundant. Many butterflies produce several successive 

 broods in the course of the rains, and hosts of beetles, flies, and 

 other winged creatures make their appearance at the same time. 

 Hence the insect-eating birds and lizards find a large choice open to 

 them of easily-obtained food ; there is less occasion to search and 

 scrutinise, and the persecution of butterflies becomes so much less 

 strenuous, that the need for a protective disguise loses much of its 

 urgency. Hence we find that those species of Precis which show so 

 close a resemblance to dead leaves in the dry season, tend to lose 

 part or the whole of that resemblance when the rains begin. 



But although in the wet season the need for protection grows 

 less, it does not entirely disappear. We have just recalled the fact 

 that some species of Precis keep up their cryptic character during 

 the wet months, though to a diminished extent. But we have still 

 to account for the extraordinary change m such species as Precis 

 antilope and P. archexia, where we have the highly cryptic dry- 

 season phase replaced in the wet season by a form which is not 

 only not cryptic but is actually conspicuous in a high degree. Now 

 when we see an insect making no attempt at concealment, but 

 flaunting, as it were, a conspicuous colour-pattern in the face of all 

 the world, we frequently find that that insect is disliked and avoided 

 by some of the usual insect-eating animals. Whether the avoidance 

 is a matter of instinct, or, as is much more probable in the majority 

 of cases, the result of unpleasant experience, it is obviously in the 

 interest of the distasteful insect to advertise its undesirable qualities 

 as much as possible, as for instance by brilliantly contrasted colours 

 in conjunction with a slow mode of flight, or a chiMce of resting- 

 place well open to observation. This is the well-known " warning- 

 colour " theory of Wallace, a theory which has received much sup- 

 port not only from the studies of naturalists in the field, but also 

 from experiments actually conducted upon birds and other animals 

 in captivity. The suggestion has been made by Professor Poulton, 

 and there is certainly a great deal to be said in its favour, that the 

 bright colours of these wet-season forms of Precis are of the nature 

 of danger-signals, warning enemies that an attack on them would be 



