454 Mr. D. A. Bannerman on the Distribution of 



does not necessarily follow that we shall discover them 

 at their breeding-station in the Azores at the same 

 time. 

 2. That in a single island there may be two distinct 

 seasons in which birds lay their eggs. This is par- 

 ticularly the case on Porto Santo. 



From this we may deduce that the several nesting-colonies 

 of these birds breed entirely independently of one another, 

 although they each may, and probably do, keep to a definite 

 season, as well as to a particular island or part of an 

 island. 



As the Madeira Group has been better " worked " than any 

 of the other Atlantic islands, I have based my deductions 

 very largely on Padre Schmitz's Diaries, which have appeared 

 from time to time in the ^ Ornithologische Jahrbuch.'' 



Records connected with the Distribution and Breeding- 

 Season of Oceanodroma castro in the North Atlantic 

 Islands. 



Jan. 14. Two adults boarded yacht 50 miles S.E. of St. Vincent 

 on our way to Santiago (Cape Verde Is.) : P. Lowe. 

 29, Two down-covered young taken (Madeira Group) : 

 Scliinitz. 



Feb. 13. Down-covered young (Porto Santo) : Scbmitz. One egg 

 taken (Desertas) : Dalgleisb. 



24. Three young obtained, one just hatched (Porto Santo) : 



Schmitz. 

 March 16. Ten adults, four young-in-down, seven eggs (Rombos Is., 

 0. V.) : Alexander. 



25. One down-covered J'oung (Madeira Group) : Schmitz. 

 April 6. One young bird, almost full-fledged (Madeira Group) : 



Schmitz. 

 21. One adult obtained, birds not yet breeding (Great 



Salvage) : O.-Grant. 

 25. One adult obtained, birds not yet breeding (Praya Is., 

 Azores) : O.-Grant. 

 May 5. Adults, young, and eggs (Bianca, C. V.) : Alexander. 



1-29. None met with (Azores) : O.-Grant. 

 22 & 23. Numbers seen nearing S. Miguel (Azores) : P. Lowe. 



