402 Some Haperiments in Seasonal Dimorphism. 
appearance of the wet-season forms. One of the most marked 
cases was in the summer of 1895-96, when I was camped at 
Gadzima, on the Umfuli River. In a normal season we 
usually get a few preliminary thunderstorms or showers in 
September, and always in October, and the first appearance 
of the wet-season broods may be looked for in the latter half 
of November. But in this case not a drop of rain fell in that 
locality until the 10th November, and even then there were 
only a few scattered showers during that month. When I 
left the district on the Ist January the dry-season forms of 
the Pieridz were still on the wing in full force, and I had 
seen no sign of the wet forms of the three common species of 
Pieris, or of the Teracoli except one wet-season male of 
T. pallene and one intermediate form of T. Anne. The 
Terias, however, were peculiar, for while in the case of 7. bri- 
gitta the winter form alone was on the wing at that date, 
the summer form of TT’. senegalensis appeared first on the 
17th November and increased in numbers thereafter. ‘This 
abnormal lateness of the wet-season broods could only be 
ascribed to the equally abnormal rainfall (it was under 
14 inches for the whole season), and at the same time it 
showed that heat alone is not the exciting cause which induces 
the appearance of the summer form, for the dryness in the 
early summer caused, as usual, a marked increase in the heat, 
so that the summer forms should have appeared earlier than 
usual instead of later had heat been the controlling factor. 
On the other hand, in the case of Precis sesamus I have 
noticed that the typical dry-season brood has appeared very 
regularly in the past four years during February, or even in 
the end of January, under the most contrary conditions, 
which leads to the inevitable conclusion that its emergence is 
regulated by some factor other than mere climatic conditions. 
Indeed, the behaviour of this species tends to strengthen a 
suggestion which I have already advanced, namely, that 
possibly we have here a case in which the dry form is 
gradually tending to eliminate and supplant the less adaptive 
wet form, as a result of natural selection. And the evidence 
is even stronger in the case of P. artaxia, Hew. During the 
same seasons, however, the winter forms of the less differen- 
tiated Pieridee made their appearance, as usual, at the end of 
the rains. 
The fact that one of the Precis archesia in the first series 
of experiments emerged as the var. Staudingert, Dew., in 
which the marbled markings on the underside are absent, is 
of special interest, for although I have had a fairly wide 
experience of the species in this country, I have never yet 
