60 C. II. Merriam — Birds of Connecticut. 



brick chimneys, and to some extent since, tliese birds used to congre- 

 gate in multitudes about certain large hollow trees, where they bred 

 regidarly, and, according to some, spent the winter also. Williams, 

 in bis "Natural and Civil History of Vermont," published in ] 794, 

 relates the followino- concerning three "Swallow trees" which came 

 under his observation. " The species called the house or Chimney 

 Swallow, has been found during the winter, in hollow trees. At 

 Middlebiiry in this State, there was a large hollow elm, called by the 

 people in the vicinity the Swallow tree. From a man who, for several 

 years, lived within twenty rods of it, I procured this information: 

 He always thought the Swallows tarried in the tree through the 

 winter, and avoided cutting it down, on that account. i\bout the 

 first of May, the Swallows cauie out of it, in large numbers, about the 

 middle of the day; and soon returned. As the weather grew warmer, 

 they came out in the morning with a loud noise, or roar, and were 

 soon dispersed; about hall an hour before sun down, they returned 

 in millions, circulating two or three times round the tree, and then 

 descending like a stream, into a hole about sixty feet from the ground. 

 It was customary for persons in the vicinity, to visit this tree, to 

 observe the motions of these birds. And wlien any persons dis- 

 turbed their operations, by striking violently against the tree, with 

 their axes, the Swallows would rush out in millions, and with a great 

 noise. In November, 1V91, the top of this tree was blown down, 

 twenty feet below where the Swallows entered. There has been no 

 a})pearance of the Swallows since. Upon cutting down the remain- 

 der, an immense quantity of excrements, quills, and feathers, were 

 found ; but no appearance or relicks of any nests. Another of these 

 swallow trees was at Bridport. The man who lived the nearest to 

 it, gave this account : The Swallows were first observed to come out 

 of the tree, in the spring ; about the time that the leaves first began 

 to appear on the trees. From that season, they come out in the 

 morning, about half an hour after sunrise. They rushed out like a 

 stream, as big as the hole in the tree would admit, and ascended in a 

 perpendicular line, until they were about the height of the adjacent 

 trees; then assumed a circular motion, performing their revolutions 

 two or three times, but always in a larger circle, and then dispersed 

 in every direction, A little before sundown, they I'eturned in im- 

 mense numbers, forming several circular motions, and then descended 

 like a stream into the hole, from whence they came out in the morn- 

 ing. About the middle of September, they were seen entering the 

 tree, for the last time," " Neither of these accounts are attended 



