98 BRITISH BIRDS. [vol. vii. 



mid-September (J. G. Jeffreys, Field, November 2nd, 1907). 

 Mr. J. Tomison considers mid-April as the usual date 

 for the first eggs, but once took an egg on April 5th, 

 1896 (Ann. Scott. Nat. Hist., 1904, p. 19). 



In the Shetlands, Saxby gives early May as the time 

 for the first eggs on Unst, but instances of much earher 

 breeding are on record. Thus Mr. H. Raeburn (Zoologist, 

 1895, p. 348) found some nests with young birds fully 

 a week old on May 6th, so that the first eggs must have 

 been laid by the end of March. It is a most extraordinary 

 fact that while in the case of the Cormorant the earHest 

 breeding-dates are from the south of Ireland, the Shag 

 on the other hand has been found nesting nearly a 

 month earher in the Orkneys than the earliest date in 

 the Channel Isles, while the earliest Irish date is a week 

 or so later still. 



At first sight it would seem that the fact of a bird's 

 breeding-season extending from February to September 

 was sufficient proof that more than one brood is reared 

 during the season. On the other hand, it is well ]cnown 

 that most of the earlier nests are washed away by heavy 

 seas. Thus Mr. J. Tomison (Ann. Scott. Nat. Hist., 

 1904, p. 18) states that whenever a few fine days occur 

 in February in the Orkneys, the Shags begin nest-building 

 at once, but that these early nests are washed away during 

 the first spell of rough weather. On June 2nd, 1902, 

 a very heavy sea washed away fully two hundred nests, 

 many of them containing young, which were, however, 

 too young to battle with the waves. During the foUowdng 

 week nest-building operations were re-commenced, and 

 most of the birds were again sitting by the end of June. 

 These late-layings were hatched in July and August, 

 and the yoiuig were able to accompany their parents in 

 September, but it is obvious that in this case the birds 

 could not strictly be called double-brooded. 



When we come to consider the number of eggs in the 

 clutch, we find the same curious discrepancies. In the 

 Shetlands the normal clutch consists of three, and four 



