NATURAL HISTORY OF SELBORNE. 263 



all doubt j but though the 3rd November was a sweet day, and 

 in appearance exactly suited to my wishes, yet not a martin was 

 to be seen; and so I was forced, reluctantly, to give up the 

 pursuit. 



I have only to add that were the bushes, which cover some 

 acres, and are not my own property, to be grubbed and carefully 

 examined, probably those late broods, and perhaps the whole 

 aggregate body of the house-martins of this district, might be 

 found there, in different secret dormitories ; and that, so far from 

 withdrawing into warmer climes, it would appear that they never 

 depart three hundred yards from the village. 



LETTER LVI. 



THEY who write on natural history cannot too frequently advert 

 to instinct, that wonderful limited faculty, which in some instances, 

 raises the brute creation, as it were, above reason, and in others 

 leaves them so far below it. Philosophers have denned instinct 

 to be that secret influence by which every species is compelled 

 naturally to pursue, at all times, the same way or track, without 

 any teaching or example ; whereas reason, without instruction, 

 would often vary and do that by many methods which instinct 

 effects by one alone. Now this maxim must be taken in a quali- 

 fied sense ; for there are instances in which instinct does vary and 

 conform to the circumstances of place and convenience. 



It has been remarked that every species of bird has a mode of 

 nidification peculiar to itself, so that a schoolboy would at once 

 pronounce on the sort of nest before him. This is the case 

 among fields and woods, and wilds ; but, in the villages round 

 London, where mosses and gossamer, and cotton from vegetables, 

 are hardly to be found, the nest of the chaffinch has not that 

 elegant finished appearance, nor is it so beautifully studded with 

 lichens, as in a more rural district ; and the wren is obliged to 

 construct its house with straws and dry grasses, which do not 

 give it that rotundity and compactness so remarkable in the edi- 



