Ixxvii 



many in Rhodesia after my return thither. I should say 

 that I note such attacks more readily than I used to, doubtless 

 as the result of practice. 



" Later. Going early one morning — 6.45 — I found them 

 all on the wing, scattered and feeding. At 7.45 they were 

 collecting under the bamboos again." 



Prof. Poulton said that, so far as he was aware, this obser- 

 vation had never before been made in Africa. Dr. G. A. K. 

 Marshall had observed the two E. African forms of Hypo- 

 limnas {Euralia) duhia, Beauv., — wahlbergi, Wallgr., and 

 mima, Trim. — collecting together, between 3 and 4 p.m., 

 for the nocturnal rest (Trans. Ent. Soc. Lond., 1902, pp. 491, 

 492), but he had never seen African butterflies congregating 

 in shady places to avoid the sun, in the manner described by 

 Mr. Swynnerton. Dr. Marshall had, however, found the 

 same thing happening in Trinidad and Jamaica, — so much 

 so, indeed, that, at the time of his visit, it was useless to 

 attempt to catch butterflies except in the cool morning and 

 evening hours. Dr. Longstaff's experience in many countries 

 agreeing with that of Dr. Marshall in Africa, was summed up 

 in the statement that " very few [butterflies] comparatively 

 are to be seen on the move before 9 a.m., and few after 

 3 p.m. . . ." (" Butterfly-hunting in Many Lands," London, 

 1912, p. 599). Mr. N. Annandale, too, had observed that 

 certain Siamese insects were active during the hottest hours 

 when birds did not hunt for food, but were rarely seen in 

 motion or indeed to be found at all in the cooler hours when 

 their enemies were at work (Proc. Roy. Phys. Soc, Edinb., 

 1900, No. XXIX, pp. 439-444). 



Dr. Marshall's experience in America and Mr. Swynner- 

 ton's at Dar es Salaam were therefore the precise opposite 

 of the other observations quoted above : in the first the 

 butterflies were active in the cool daylight hours and rested 

 in the heat ; in the second they rested in the cool hours and 

 were active in the heat. Extended investigations in various 

 parts of the tropics, and especially in the same locality at 

 different seasons, with due regard to the humidity as well 

 as the temperature of the air, would probably explain the 

 apparent inconsistency; and it was to be hoped that Mr. 



