102 BULLETIN 170, UNITED STATES NATIONAL MUSEUM 



where trees are absent or almost so, the eggs are generally laid in a 

 recess or on a ledge in a cliff face, sometimes sheltered by ivy. An 

 old nest of the hooded crow (Corms c. comix) among rocks is also 

 frequently appropriated. In the wooded districts of the midlands and 

 south, the most usual site is an old nest of a magpie {Pica p. pica) or 

 carrion crow (Corvus c. corone), generally somewhat flattened out, 

 while much less frequently old nests of rook (C. frugilegus), raven 

 (C. corax), buzzard (Buteo b. buteo), heron (Ardea cinerea), sparrow 

 hawk (Accipiter n. nisus), or even an ancient squirrel's "drey" may 

 be occupied. Where there is a scarcity of breeding sites, a pair of 

 kestrels will occasionally drive off a magpie from a new nest, after 

 some days of fighting. One case of this kind came under my own 

 notice in Derbyshire, and H. S. Davenport also describes a very similar 

 incident. A regular pitched battle took place between a pair of 

 kestrels and a pair of magpies, in which the latter were worsted and 

 had to abandon their newly built home. In both these cases the mag- 

 pies were evicted before eggs had been laid, but there is an even more 

 surprising case on record (Phillips, 1908, p. 139) in which a nest of 

 carrion crow, which contained young birds, was taken possession of 

 by kestrels, which killed and ate the young crows. When examined, 

 the nest contained not only a clutch of five kestrel's eggs, but many 

 feathers from the crows. 



Another site occasionally adopted by the kestrel is the broken top 

 of some old elm or the jagged hole formed by the breaking away of a 

 heavy branch. Old buildings, ruins, and church towers also provide 

 many pairs with breeding places, and where the birds are not disturbed 

 these are resorted to year after year. On the floor of a belfry in the 

 tower of a church in East Anglia, I once saw a clutch of fresh eggs of 

 this little hawk. Close by were several dirty, egg-shaped objects, 

 which proved on examination to be infertile eggs, faded and dried up, 

 from nests of previous years. Cases are also on record in which nest- 

 ing boxes put up in trees have been appropriated; and likely holes, 

 even in occupied houses and windmills, are sometimes taken possession 

 of. Perhaps, however, the most surprising site is on the ground! In 

 Wicken Fen, Cambridgeshire, nests have not infrequently been 

 recorded among the coarse grass and sedge; and similar cases have 

 occurred in other districts, such as Norfolk and the Outer Hebrides, 

 but usually in places where trees are scarce or absent. 



Eggs. — The number of eggs usually ranges from four to six, the last 

 number being fairly common. Sets of seven are decidedly scarce, 

 but I have known of 1 1 cases from various parts of the British Isles, 

 while instances of eight are even rarer, and only five have come under 

 my notice. There are cases on record in which two females have been 

 found in attendance at one nest. Thus M. C. H. Bird once found 

 nine eggs in a nest on the ground at Ruston, Norfolk, and this may have 



