Field Notes. 119 



Kentucky Warbler. Two found at the east end of the sand spit 

 May 13. Always scarce. 



Brewster's Warbler. One found near the Lake Laboratory. It was 

 in nearly typical plumage for this form. 



Caspian Tern. Two were ranging along the sand spit and over the 

 bay September 1.3 and 23, 1907. They passed within easy range sev- 

 eral times. Lyxds Joxes. 



A Purple Martin Roost. — On my coming from Ohio to Florida, 1 

 was impressed with the much greater abundance of Purple Martins 

 in the South than in the North. During the months of July and 

 August, 1907, I traveled over about all of west Florida, and was in 

 nearly all of the towns and cities west of the Suwannee river. In 

 all this territory I saw the Purple Martins nowhere nearly so 

 abundant as at Qirincy, Gadsden county, Florida. There were 

 very few to be seen here during the middle of the day, but towards 

 evening they would gather in. 



It was impossible to tell where they came from, but in a very 

 short time, and just before sunset, there would be thousands of 

 them in the air circling over the town. The sky was alive with 

 them until labout dusk, when they went to roost. The noise of 

 their chatter was continuous during this time. 



They gradually collected in a more compact body and swung 

 around over the town in large gyrations, until finally a little be- 

 fore dark, as if of one mind, they dropped into a small clump of 

 mulberry trees. With great noise of wing as well as of voice, 

 they fairly fell into these trees with a rush that was truly aston- 

 ishing. These trees are in the central part of the town, near the 

 Lorraine hotel. They had all settled in less time than it takes to 

 write it. I believe that it would be conservative to estimate their 

 number at more than five thousand. After getting settled, their 

 noise was intense; from a little distance it sounded much like 

 escaping steam, or like the patter of violent rain on the leaves of 

 the trees. This noise continued intermittently until late into the 

 night, and began again with energy before daylight in the morning. 



The birds began to leave just about dawn, — a few straggling 

 little bunches leading off, and then practically all of the rest 

 taking wing at once and swinging off in one grand departure. 

 Howevei-, there were a few that did not get started with the main 

 flight; but in a very few minutes they were all gone. The 

 branches of the trees sprang up very perceptibly as their load was 

 released. 



My observations extended over only four days. — August 6-0. 1907. — 

 and I have been informed that the INIartins have been roosting in 

 this clump of trees every night during the entire summer for a num- 

 ber of years. G, C. Fisher. 



