Coll. No.

ground, singing most heartily-"sliding down the scale of their own music," as some one has happily expressed it in the case of another species of similar habit. This song, I think, is superior to that of the Chestnut-collared Bunting, though of the same general character. When hovering in the manner just indicated, both birds resemble butterflies, there is something so light, wayward, and flickering in their motions.

It is difficult to understand how this bird eluded observation for thirty years-from the time of its original discovery by Audubon, on the Upper Missouri, nearly to the present day. If the species were really rare, this would not be remarkable; but it has lately been shown to be extremely abundant in different parts of the West. I did not meet with it along the Red River itself, but found it as soon as I passed from the Pembina Mountains to the boundless prairie beyond. In some particular spots, it outnumbered all the other birds together; and on an average through the country, from the Pembina Mountains to the Mouse River, it was one of the trio of commonest birds,—the Skylarks and Chestnut-collared Longspurs being the other two. The first specimens I procured were taken July 14. Some of them were newly fledged, but the great majority were adult males, showing that at that time the breeding-season was at its height. Out of thirty-one specimens secured July 14 and 15, only one was a female, the individuals of this sex being evidently occupied with the duty of incubating or brooding their young. The males at this time were very conspicuous, like Spizella pallida under the same circumstances, as they sat singing on the weeds or low bushes of the prairie. The song consists of two or three distinct syllables, followed by a trill uttered in a mellow, tinkling tone. The nest I never succeeded in finding, although I must have passed by many. The eggs were first discovered by Mr. Allen in the region just south of me. They were taken July 1, 1873, the date corresponding to that which I fixed as the laying season from consideration of the habits of the birds. The nest and eggs are described from his specimens in the "Birds of the Northwest". Whether or not two broods are reared, I cannot say; but some of my late summer specimens were so young that I judged they might belong to a second brood. Birds of apparently about the same age were shot six weeks apart.

The general habits of Baird's Bunting are much like those of Passerculus saranna, and the appearance of the two birds during life is so similar that it is difficult to tell them apart at any distance. The Centronyx is not truly gregarious, but, like many other prairie birds, affects particular spots, which are colonized by large numbers. When the young are all on wing, it associates in straggling troops, mixing freely with the Skylarks and Longspurs. During the summer, the plumage becomes extremely worn and faded; with the moult, which occurs in September, a much more richly colored dress is assumed. The bird remains in this country at least until October, though its numbers sensibly diminish during the preceding month.

The rediscovery of this little-known and extremely interesting species in Dakota was made in the season of 1873 by the Commission. On the march between Turtle Mountain and the first crossing of Mouse River, I came upon what seemed to be a small colony of the birds in a moist depression of the prairie, where the herbage was waist-high. By diligent search, after shooting the first specimen and perceiving what it was, I managed, not without difficulty, to secure five in all. This was on the 9th of August. I subsequently found the bird again, and secured a sixth specimen, amongst the reeds of a prairie slough near the headwaters of the river just mentioned. So far as I could determine from short observation, the birds are much like the Ammodromi in their general habits and appearance, and they inhabit similar situations. Their note was a chirring noise, like that of a grasshopper. They were started at random from the tall, waving grass, flitted in sight for a few seconds, and then dropped suddenly, so that the chances of shooting them were very poor. One was killed at very close range by a blow from the wad of my cartridge, the charge of shot having passed in lump close by. I have no doubt that the birds were breeding in this place, though no nests were found. Their retiring habits and the nature of their resorts have doubtless caused them to be overlooked for years. Audubon says that he found them common on the Upper Missouri. A specimen, in poor condition, from Texas, was the only one known to exist in any collection before these of mine were secured, Audubon's type having been lost or mislaid. A redescription of the species, in which it is shown that the characters originally assigned required modification, is given in the "Birds of the Northwest ".

Though not

Breeds in profusion throughout the region explored. exclusively a bird of the prairie, it seems to be as much at home in the open plains as anywhere, associating intimately with Centronyx and the two leading species of Plectrophanes. It is also found, however, in the brush along the streams and larger rivers, which are unfrequented by the species just named, in company with the Melospiza and Juncones. A large suite of specimens was taken, a part of it, however, unintentionally, for it is not an easy matter to always distinguish between the Savanna Sparrow and Baird's Bunting at gunshot range; and when I have killed a bird, I generally make a point of preserving it, even though it is not particularly wanted as a specimen, in order that its life may not have been taken in vain. The nest is placed on the ground, simply built of dried grasses, with a lining of horse-hair; the eggs are four or five in number, in this locality usually laid in the first half of June. Like nearly all the Fringilline birds of this region, the Savanna Sparrow is frequently the Cowbird's foster parent, and in one instance that came under my observation the nest contained two of the alien