ORNITHOLOGICAL OBSERVATION IN ICELAND. 225 



constant one with them ; * and, as I have myself observed, they 

 often suffer severely for it, for no sooner are a pair seen to be 

 acting in this way, than as many of the rookery as happen to 

 be there assembled, taking the cue one from another, set 

 violently upon them, and the nest becomes shortly covered up 

 beneath a mass of tumbling, struggling birds. The scene, 

 indeed, is so animated and the main features of it — birds flying 

 from every direction to one particular nest, with a general 

 scrimmage upon it — so marked, that it has occurred to me it 

 may often have been misinterpreted as the destruction of the 

 nest of some couple convicted of pilfering. One would only 

 need not to be there at the beginning, and not to stay till the 

 end, to fall into this error quite easily. There is, however, no 

 doubt as to the true cause in these cases, however established 

 the other one (which I have not myself witnessed) may be. It 

 is, I think, remarkable that the Eooks should cling to a habit 

 attended with such severe penalties. If it has its roots deep 

 down in the sexual instinct, or its concomitant developments 

 arising out of the sexual frenzy, then here is an explanation ; 

 but otherwise it is not easy to understand how, in the face of 

 such difficulties and discouragement, it should ever have esta- 

 blished itself. To conclude, on the supposition that the nest 

 was originally the pairing-place, we have an explanation of the 

 facts to which I have here referred, but all of which I have not 

 yet fully recorded, nor need their being exceptional (though as 

 to the extent to which they are we are yet in ignorance) surprise 

 us, since the differentiating and specialising process, which is 

 at work everywhere, would be sufficient to account for this. On 

 the other hand, if these two instincts — nidification and the 

 sexual or pairing one — though thus sometimes, as I have shown, 

 combined and intermingled, have yet, so far as the origin and 

 growth of the former are concerned, had nothing to do with one 

 another, the fact of their being so is surprising, and has yet to 

 be accounted for. 



* Rooks are not included amongst the four species mentioned by me 

 as pairing habitually on the nest, nor do they belong to any of the families 

 represented by these. 



