REED WARBLER, 



sequent upon, one male crossing its boundary and entering 

 the area of its neighbour. And, lastly, in the case of two 

 males occupying adjoining territories, one paired and the other 

 unpaired, w T e find that the unpaired male frequently attacks 

 not only the neighbouring male but also the female when 

 either of them intrudes upon its territory. Any one who 

 will take the pains to keep under observation two pairs of 

 Chiff-chaffs, or of Blackcaps, living in proximity to one 

 another, or, better still, some large sheet of water inhabited 

 by a number of different pairs of Moorhens, can observe for 

 himself these struggles which are of almost daily occurrence. 

 And how great an influence they must have exercised upon 

 the history of any one given species we can well imagine, 

 when, on the one hand, we remember how incessant they 

 must in reality be in order that we may so readily observe 

 them, and, on the other, bear in mind the enormous period 

 of time through which species have been slowly evolving. 



The strength or weakness of the theory lies in the evidence 

 upon which it is based, that is to say, upon the observation 

 of one person, and herein, at least, lies a weakness. But 

 although it is probable that in the details there may be some 

 exaggeration — and it is doubtful whether the human powers 

 of observation are capable of always giving an accurate state- 

 ment of the facts — yet the essential facts I believe to be true ; 

 and I appeal to Nature herself for corroboration. Let these 

 principal points be borne in mind : that whatever species I 

 have hitherto studied I have found the question of territory 

 in some degree present ; that the problems with regard to 

 unpaired birds, rate of increase, extension of breeding range, 

 and, in the case of the migrants, the hurrying forward of the 

 males in spring, on this principle become capable of solution, 

 and once more, and finally, that although the evidence is 

 drawn from the lives of comparatively few species, yet some 

 of those species are not closely allied, but belong to widely 

 separate genera. 



39 



