IIIRUNDINID^E. COTYLE SERRIPENNIS. 107 



to be confounded with it, though readily distinguishable from it by careful observa- 

 tions. It is difficult, if not impossible, now to say how much of the history of that 

 bird has had reference to this species. My own observations in regard to it have 

 been confined to the neighborhood of Carlisle, Pcnn., the only place in which I 

 have ever met with it, or where, to my certain knowledge, it has been ascertained 

 to breed. They have also been seen in Louisiana, where they were first discov- 

 ered by Audubon, Florida, Missouri, and Virginia, and were found by Dr. Heer- 

 mann in California. They probably breed in each of these States, especially the last 

 two, where they have been noticed in the breeding season. My friend, Professor 

 Baird, then of Carlisle, found them breeding in that vicinity in the summer of 1843. 

 Visiting that locality early in June, the following year, I had an opportunity to ob- 

 serve their habits in the breeding season, in company with that accurate ornithologist. 

 We found it quite common, and examined a number of nests. It is worthy of re- 

 mark, that none of those we found that season were in places excavated by the bird, 

 although Professor Baird informs me he has known them to nest in holes apparently 

 prepared in the same manner as those of the Bank Swallow. All the nests I have 

 ever seen were in situations accidentally suited to their wants, and all were directly 

 over running water. We found seven nests, all of which were in similar situations. 

 This may not be enough to be decisive of this reliance upon accidental aids, or 

 rather of its universality, but certainly shows it to be at least very common. Sev- 

 eral were constructed in crevices between the stones in the walls and arches of 

 bridges. In numerous instances the nests were but little above the surface of the 

 stream. One had been flooded by the rise of the water, and the eggs addled. In- 

 stead of forsaking the locality, the birds had constructed another nest directly over 

 the first. In this double nest we found twelve eggs, six of which were quite fresh, 

 and all had been deposited that season. Another nest was constructed between the 

 stones of a wall which formed one of the sides of the flume of a mill. Along the 

 top of this wall was a frequented footpath, while the water of the mill-stream was 

 not two feet below it. One nest was found in a somewhat higher situation, between 

 the boards covering a small building containing a water-wheel. The entrance to 

 the nest was through a knot-hole in the outer partition, and it was placed upon a 

 small rafter between the outer and inner partitions. 



The nest, very similar in its construction to that of the Bank Swallow, is com- 

 posed of dry grass, straw, and leaves, loosely put together, and lined with downy 

 feathers. Owing, perhaps, to the exposed situations in which we found them, and 

 their danger of being flooded by a rise of the stream, a much greater amount of 

 material was used in their construction than by the Bank Swallow. In other re- 

 spects there was no apparent difference. 



Nor are their eggs readily distinguishable from those of that species, being nearly 

 uniform with them in size, shape, and color. They are pure white, perhaps a little 

 more uniformly oblong, and usually quite pointed at the smaller end. They vary in 

 size from {i of an inch in length to ff, and in breadth from | to T 9 g of an inch. 

 The average length is {| and the average breadth i|- of an inch. 



