1922.] TRINIDAD BIRDS. 147 



It usually appears each year shortly after the beginning of the 

 wet season but the date of its first arrival does not seem to be 

 definitely connected with the date of the first rain?. The following 

 table sho ,vs what little information is available on this question. 



Year. First Rains. First Record of Bird. 



1918 .. May 25-27 .. Reported at Li Fortunee at 



middle of May, first seen 

 June 7. 



1919 .. First Week May First seen, June 10. 



1923 .. June 15 .. Already present, June 29, 



no previous observation. 



It increases rapidly in numbers and is usually in very great 

 abundance from July to September, often congregating in enormous 

 numbers in the sugar cane fields attacked by frognopper. 



By the end of September the numbers begin to lessen and by tne 

 end of October only a few stragglers remain. In 1918 none were 

 seen after October 27 ; in 1919 three males were seen on November 1 1 , 

 one on 14, and two on November 24. 



In 1918 there was an interesting departure from the normal 

 routine in the appearance of four individuals at La Fortunee Sugar 

 Estate about the middle of March. They were reported to me by 

 Mr. Creteau and were seen by myself on March 20 and 26. it is 

 further remarkable that the cane field in which they were seen was 

 suffering from an abnormal attack of froghoppers during the dry 

 season. 



It will be seen from the above that the birds usually reach the 

 Island about the beginning of June and leave about the middle of 

 October, during which time they do not nest. 



So far as I can find information, they breed during the remainder 

 of the year somewhere in Venezuela or Colombia. That is to say 

 their movements are entirely within the tropics. But the changes of 

 season in Trinidad are from dry season to wet about the middle of 

 May to the beginning of June, and from wet season to dry at the end 

 of December or during January. So that while the arrival of the birds 

 corresponds more or less to o.ie change, its departure does not 

 correspond to the other. Its movements therefore are apparently 

 not determined by the seasonal changes in Trinidad. It is possible 

 that they follow seasonal changes in the country where breedmg takes 

 place, but it is also worth pointing out that the migrations take place 

 almost at the same time as those of the birds of temperate climates 

 migrating North in Spring and South in Autumn, and we may have 

 here a relic of a much longer migration in past times. 



