6758 Birds. 



line, evidently intent only on coming tip with it. Now, if it be con- 

 sidered that we have here a swift of very extraordinary size which 

 does not, as might be supposed, secure a sufficient amount of prey by 

 hunting in small parties, but, notwithstanding its size, wonderful 

 activity, and the voracity which these imply, hunts after its prey in 

 very large numbers, it will be understood that the ordinary produce 

 of insects of one locality is not sufficient, especially in an island 

 where I have your authority for saying insects are by no means in 

 extraordinary abundance. It is absolutely indispensable to the flock 

 that they should seek for swarms ; but for this, it will be seen, extra- 

 ordinary obstacles are to be overcome. In a large flock of gregarious 

 birds, which have simply to rise to a certain height, survey the coun- 

 try beneath, and then, selecting some large geographical object, 

 descend and feed, their cries and ordinary powers of vision are quite 

 sufficient to keep the whole flock together ; but with a number of birds 

 seeking swarms of insects the case is widely different. No powers of 

 vision we have any cognizance of could possibly enable them to dis- 

 cover swarms half a mile off; indeed, there is abundant evidence 

 they do not see them till close upon them. The birds then would be 

 obliged to beat over immense districts, following a small number 

 of leaders if they are to keep together, — a mode of hunting evi- 

 dently involving great loss of time and great fatigue. If, on the 

 contrary, they availed themselves of their numbers in the obvious 

 mode of exploring in different directions at once, they would infal- 

 libly be separated, and then their numbers would be of no avail, for 

 when a party have found the swarm how are the rest to know it ? 

 They would cease to be swarm-destroyers at all. The probability is 

 that this mode of feeding requires the simultaneous action of great 

 numbers to explore large spaces at once. It is far from impossible 

 that without some such power the birds would perish. As you may 

 have gathered from the observations given above, this combination of 

 action is secured by the very remarkable habit of a varying proportion 

 of the flock rising and forming a gyrating column. Whilst thus 

 revolving a bird darts out, followed by others who form a screaming 

 party, and then rejoin the column. 1 have in other letters mentioned 

 that these screaming parties are so constant that I often, through 

 them, know of the passage of the column over head while in the 

 forest, or it gives notice of their approach when, from their great 

 height and distance, they would otherwise elude my observation. It 

 cannot be doubted that these perpetual screams, and the column 

 itself, so conspicuous as it floats, equally or in a far greater degree, 



