ST. KILDA 145 



where they breed early in March, and by the end of that 

 month the rocks are fully tenanted. In April they set about 

 making and repairing their nests. They begin to lay spar- 

 ingly in favourable seasons at the end of April, but the eggs 

 are not obtained in any numbers before the first or second 

 week in May. Fully a fortnight will elapse from the time 

 of the first birds laying, before eggs are found in all the 

 nests. On the occasion of my visit, on June 10th [1902], 

 the sloping faces on the top of Stack Ijii were crammed 

 with nests as thick as they could be placed, every available 

 spot between the projecting angles and slabs of rock being 

 thus occupied. Almost every nest contained a single egg, 

 the great majority of which were stained a deep brown 

 colour, and these were all, so far as they were tested, con- 

 siderably incubated, the embryo being formed, but not of 

 large size. ... I endeavoured to form some sort of estimate 

 as to the number of Gannets resident in St. Kilda during the 

 summer by ascertaining the number of eggs taken. On May 

 14th there were taken from the top of Stack Lii (from the 

 summit only) 1,400 fresh eggs; the great majority of the birds 

 had then laid, but some nests were still unoccupied. Perhaps, 

 then, 1,500 nests might represent the number on the sloping 

 faces on the top of this Stack only. But the men say that 

 on the whole of the rest of this Stack there are more nests 

 than on the summit, and if this be so, the total number of 



