THE ROOK AND JACKDAW 35 



and may itself be a modified or limited form of the migratory 

 instinct. The evidence available points, however, to a marked lack 

 of uniformity in the sleeping arrangements of rooks, which leaves 

 any one explanation inadequate. 



The same applies to jackdaws. Those, or most of those at Wells 

 Cathedral, remain to roost there throughout the year. At Corfe a 

 certain number leave each evening during the winter, from October 

 on, to sleep in a neighbouring wood, the remainder stopping at the 

 castle all night. The daws of Furness, on the other hand, all roost 

 away from the abbey from September to March, quitting it at dusk 

 with their neighbours the rooks, to return at dawn. In this, and 

 probably in most cases when sleeping away from the nesting-place, 

 daws and rooks go to the same roost. 1 



Rooks and daws often travel by easy stages to their sleeping 

 quarters, having developed none of the modern passion for quick 

 transit. A Hock has been known to take from two to four hours to 

 do a twenty minutes' journey. - 



The daws fly sometimes with the rooks, sometimes in a separate 

 band, which appears generally composed of pairs. It has been 

 asserted that rooks fly in pairs within the flock. But, if so, they are 

 not easy to detect. Tennyson has also told us that there is a 



' many wintered crow 

 That leads the clanging rookery home.' 



Personally I have never noticed that any one rook acts as leader, 

 and the experience of others is the same. It is a matter, however, 

 about which each reader can find ample opportunities of forming 

 his own judgment. 



The flocks, on arriving at the common roost, do not go straight 

 into it, but, for some unknown reason, pass over or stop short of it, 

 to collect not far off at some common meeting-place. Here, as in the 



1 This information has been supplied me by the keepers at Corfe Castle and Furness Abbey, 

 and by Mr. Stanley Lewis and Mr. Thos. W. Phillips of Wells. 



- Zooloyiat, 1904, pp. 270-6. 



