Migralorij Birds at Koinatijwori. 21 



about the beginning of April there reappeared about 100 

 birds. I am inclined to think that those all represent 

 different migrations — for one reason, that the permanent 

 summer residents always perched on the same stretch of 

 telegraph-wire ; while the birds that came later favoured a 

 different place, and after each successive reappearance were 

 seen to take up different quarters. 



On the 14th April we had heavy rain and wind, and large 

 numbers of H. rmtica were driven indoors; many were 

 injured, and so I had no difficulty in clearly identifying them. 

 During the next few days the weather continued rather 

 stormy and several were killed by cats, so that up to April 

 20th I had plenty of chances of inspecting dead birds. After 

 this till the end of the month the Swallows, I am practically 

 certain H. riatica and H. alUgularis, continued to hawk 

 insects every evening round the cattle kraals, just as the two 

 species had done all the summer. The numbers seemed to 

 become gradually less ; on May 5th I noticed about ten 

 birds, but on the Gth there were none, nor have I seen any 

 sign of them since in their accustomed haunts. Very light 

 north-easterly breezes prevailed at the time, and there was a 

 marked fall in the night temperature. 



On the 8th I saw some Swallows hawking flies over the 

 river and bathing at sunset ; these appeared like H. alhigu- 

 laris, but on shooting one I found that it was H. smithii. 

 They were acting quite differently from the Swallows I had 

 been hitherto observing, however, and it was at some distance 

 from the place frequented by them. 



On the whole I am pretty sure that the last of the northern 

 Swallows did not leave here till the beginning of May, 

 though the exodus began perhaps two months earlier. 



Milvus (pgyptius arrived about the beginning of November, 

 was in great numbers throughout December and January, 

 when the afterbirths of the calving game seemed to attract 

 them, and disappeared somewhere about the end of February 

 (they had been increasing in numbers throughout the latter 

 month). From the 1st of March onwards I saw no more 



