198 DIRECTION OF MIGRATION 



The first of the two concerns the direction of migra- 

 tion; the second, the perturbations or irregularities 

 to which it is occasionally subject. Keeping still to 

 mv own observations on the autumnal behaviour of 

 birds previous to departure, I have asked myself: 

 When, or how soon, does this trouble in the bird, 

 which manifests itself as fear of an invisible danger 

 or enemy from which it seeks to escape, first incline 

 it to the north as the side where safety is to be found ? 

 I failed to detect any special inclination to fly to that 

 side in the swallows, even when the preliminary dis- 

 quiet and agitation lasted many days, during which 

 the birds would rise or rush away with cries of alarm 

 to this side or that and scatter and then return to 

 their perch and their brooding intervals, until the 

 very eve of their departure, for you could then see 

 that when they rose or sprang away into the air it 

 would be to the north side. It was different with 

 the upland plover: from the very beginning of its 

 period of unrest it invariably, when rising, rushed off 

 to the north side. 



And here again I would emphasise the difference 

 in the behaviour of different species when affected 

 by the same influence and impulse. It is, to my 

 mind, an extraneous influence — a *^ breath," as the 

 poet of I'he Seasons has called it, and he could not 

 have found a better metaphor. Touched by the 

 breath as by a coming wind, the migratory birds were 

 compared by me to globes of thistledown, resting in 

 still weather on the grass, trembling at the first faint 

 movement of the air, and finally lifted and carried 



