304 Scott on Birds at the Dry Tortugas. [October 



larger. Near one end of it is a light-house of the first order and 

 near it are the keeper's quarters and buildings, surrounded by some 

 cocoanut palms, and a few scrubby button-wood trees. The usual 

 low bushes are abundant, but, so far as I am aware or could learn, 

 no birds breed at this point, though the superficial area of the 

 island must be at least twenty acres and it seems admirably 

 adapted to the breeding wants of Terns and the like. 



After a careful study of all of these islands, supplemented by 

 close questioning of the lighthouse-keepers and the sergeant who 

 who has been in charge of the government property at Garden 

 Key for the past five years, I have arrived at the general conclu- 

 sions : (1) That no land birds breed on any of the keys of the 

 group. (2) That the stay of any land birds is of very short 

 duration save on exceptional occasions which I shall have reason 

 to allude to in more detail presently. 



The migrations of the land species which I observed during my 

 stay always coincided with the approach or occurrence of some 

 pronounced aerial disturbance, and the advent of a strong north 

 or northwest breeze always meant a very perceptible, and gener- 

 ally a marked, accession of the bird visitors. This applies to land 

 birds in the main, and was not remarked regarding water birds, 

 except in the case of the Herons, which will be noticed in detail 

 further on. It seems probable that the flight of land birds was 

 at this time of year quite constant and that the northwest winds 

 arrested this flight and so increased the number of birds on the 

 keys. 



My stay at Garden Key was limited, and I left there on 

 April 10. The following list includes all the birds that were ob- 

 served up to that date and the records of several others, and notes 

 regarding their occurrence, made by Dr. F. S. Goodman, of the 

 U. S. Marine Hospital Service, stationed at the Quarantine Station 

 on Garden Key. This gentleman kindly consented to carry on 

 some work for me after my departure. In all cases he has for- 

 warded to me the birds secured, together with the date of cap- 

 ture, etc. My thanks are due to him for this aid and in all in- 

 stances I include his name in connection with information which 

 he has furnished. 



The list of birds observed at the Tortugas includes eighty 

 species and though it will undoubtedly be greatly added to, yet 

 enough land birds are present, fifty-seven species in all, to point 

 apparently to the following conclusions. 



