b. Edge of wing smooth. Tarsus longer than in last; equal to middle toe and half the claw. Nostrils bordered along posterior half by membrane, but not overhung internally. Bill very small. Tail forked. Crissum dusky except in Neochelidon fucata. Various genera and subgenera, none North American, as Atticora, Notiochelidon, Neochelidon, and Pygochelidon.

c. Edge of wing armed with stiff recurved hooks. Tarsus as in preceding (tarsus and toes much as in Pygochelidon). Bill larger and more depressed. Tail emarginate only. Crissum white … Stelgidopteryx.

B. Nostrils lateral; bordered behind and inside, or overhung by membrane, the outer edge of which is straight, and directed either parallel with axis of bill or diverging from it.

a. Tarsus short; about equal to middle toe without claw. Tibial joint feathered; feathers extending along inside of upper end of tarsus.

Tarsus bare at lower end. Lateral claws reaching only to base of middle.

Tail very deeply forked, much longer than closed wings; lateral feathers linear and very narrow at end, twice the length of central. Upper parts and pectoral collar steel-blue; front and throat, sometimes under parts, rufous. Tail-feathers with large spots … Hirundo.

Tail with shallow fork, not exceeding half an inch, shorter than closed wings. Feathers broad. Color blue or green above, with or without white rump; white beneath. Tail-feathers without spots … Tachycineta.

Tarsus with a tuft of feathers at lower end. Lateral claws lengthened, reaching beyond base of middle claw.

Tail slightly forked. Color dull-brown above; beneath white, with brown pectoral collar … Cotyle.

Species and Varieties.

P. subis. Above lustrous blue-black; beneath lustrous blue-black or brownish-gray, uniform, or with the abdomen and crissum white, or whitish. Females always with the throat and jugulum gray.

A. Adult males entirely steel-blue.

a. Females and young males with the abdomen pure white.

Feathers about the anus smoky-gray beneath the surface.

Wing about 6.00; fork of tail, .80 deep. ♀ and Juv. Abdominal and crissal feathers always with dusky shafts, and with the concealed portion grayish. Forehead and nape hoary grayish. Hab. Continental North America, south into Northern Mexico … var. subis.

The young male of the second year is similar to the female, with the steel-blue appearing in patches.

Total length (of 1,561), 7.50; wing, 6.00; tail, 3.40; difference between inner and outer feather, .75; difference between first and ninth quills, 2.88; length of bill from forehead, .55; from nostril, .34; along gape, .94; width of gape, .74; tarsus, .61; middle toe and claw, .80; claw alone, .25; hind toe and claw, .54; claw alone, .27.

Hab. The whole of the United States and the Provinces; Saskatchewan; Cape St. Lucas and Northern Mexico (winter); Orizaba (Sumichrast); Bermuda. Accidental in England. South American and West Indian birds apparently belong to other races.

Many Western adult males are considerably less violaceous than any Eastern one; but there is so much variation in this respect among specimens from one locality, that this difference in lustre does not seem of much importance.

An adult female (No. 61,361, G. A. Boardman) from Lake Harney, Florida, is so unlike all other specimens in the collection as to almost warrant our considering it as representing a distinct local race. It differs from females and young males of all the other races (except elegans, from which it differs in other striking particulars) in the following respects: Above, the lustrous steel-blue is uninterrupted, the forehead and nape being uniform with the other portions; beneath, dark smoky-gray, inclining to whitish on the middle of the abdomen; the jugulum and crissum have a faint gloss of steel-blue, the feathers of the latter bordered with grayish-white. The chief difference from elegans is in lacking the conspicuous grayish-white border to the feathers of the whole lower part, the surface being uniform instead of conspicuously squamated. Wing, 5.60; tail, 3.00; fork of tail, .80 deep.

In a wild state the natural resort of this species, for nesting and shelter, was to hollow trees and crevasses in rocks. The introduction of civilized life, and with it of other safer and more convenient places, better adapted to their wants, has wrought an entire change in its habits. It is now very rarely known to resort to a hollow tree, though it will do so where better provision is not to be had. Comfortable and convenient boxes, of various devices, in our cities and large towns, attract them to build in small communities around the dwellings of man, where their social, familiar, and confiding disposition make them general favorites. There they find abundance of insect food, and repay their benefactors by the destruction of numerous injurious and noxious kinds, and there, too, they are also comparatively safe from their own enemies. These conveniences vary from the elegant martin-houses that adorn private grounds in our Eastern cities to the ruder gourds and calabashes which are said to be frequently placed near the humbler cabins of the Southern negroes. In Washington the columns of the public buildings, and the eaves and sheltered portions of the piazzas, afford a convenient protection to large numbers around the Patent Office and the Post-Office buildings.

The abundance of this species varies in different parts of the country, from causes not always apparent. In the vicinity of Boston it is quite unusual, though said to have been, forty years since, quite common. There their places are taken by the H. bicolor, who occupy almost exclusively the martin-houses, and very rarely build in hollow trees.

Sir John Richardson states that it arrives within the Arctic Circle earlier than any other of its family. It made its first appearance at Great Bear Lake as early as the 17th of May, when the ground was covered with snow, and the rivers and lakes were all icebound.

In the Southern States it is said to raise three broods in a season; in its more northern distribution it raises but one. Their early migrations expose the Martins to severe exposure and suffering from changes of weather, in which large numbers have been known to perish. An occurrence of this kind is said to have taken place in Eastern Massachusetts, where nearly all the birds of this species were destroyed, and where to this day their places have never been supplied.

The Martins do not winter in the United States, but enter the extreme Southern portions early in February. Audubon states that they arrive often in prodigious flocks. On the Ohio their advent is about the 15th of March, and in Missouri, Ohio, and Pennsylvania about the 10th of April. About Boston their appearance is from the 25th of April to the middle of May. Mr. Audubon states that they all return to the Southern States about the 20th of August, but this is hardly correct. Their departure varies very much with the season. In the fall of 1870 they were to be found in large flocks, slowly moving southward, but often remaining several days at a time at the same place, and then proceeding to their next halt. Their favorite places for such stops are usually a high and uninhabited hillside near the sea.

The Martin is a bold and courageous bird, prompt to meet and repel dangers, especially when threatened by winged enemies, never hesitating to attack and drive them away from its neighborhood. It is therefore a valuable protection to the barnyard. Its food is the larger kinds of insects, especially beetles, in destroying which it again does good service to the husbandman. The song of the Martin is a succession of twitters, which, without being musical, are far from being unpleasant; they begin with the earliest dawn, and during the earlier periods of incubation are almost incessantly repeated. The eggs of the Purple Martin measure .94 of an inch in length by .79 in breadth. They are of an oblong-oval shape, are pointed at one end, are of a uniform creamy-white, and are never spotted. They are quite uniform in size and shape. Eggs from Florida are proportionally smaller than those from the Northern States.

Progne subis, var. cryptoleuca, Baird.

CUBAN MARTIN.

Progne cryptoleuca, Baird, Rev. Am. Birds, 1864, 277. Hirundo purpurea, D’Orb. Sagra’s Cuba, Ois. 1840, 94 (excl. syn.). Progne purpurea, Cab. Jour. 1856, 3.—Gundlach, Cab. Jour. 1861.

Female (17,730, Monte Verde, Cuba, May 2; C. Wright). Above steel-blue, less glossy than in the male, and becoming lustreless dark smoky-brown on the forehead. Head, laterally and beneath, with jugulum and sides, uniform brownish-gray (without darker shafts or lighter borders to feathers, as in subis); whole abdomen, anal region, and crissum snowy-white, including the shafts. Wing, 5.40; tail, 2.80; fork of tail, .70 deep.

Young male (10,368, Cape Florida, May 18, 1858; G. Wurdemann). Similar to the female, but the steel-blue above more brilliant and continuous, the forehead and wings being nearly as lustrous as the back; throat and jugulum mixed with steel-blue feathers, and crissum with some feathers of steel-blue bordered with whitish. Wing. 5.40; tail, 2.90; fork of tail, .80 deep.

Hab. Cuba, and Florida Keys? (Perhaps Bahamas.)

This species has a close external resemblance to P. subis, for which it has usually been mistaken. It is of nearly the same size, but the feet are disproportionately smaller and weaker; while the wings are shorter, the tail is as long and more deeply forked; the feathers considerably narrower, and more attenuated (the outer .40 wide, instead of .46). The colors above are more brilliant, and extend more over the greater wing-coverts and lining of wings, while the quills and tail-feathers have a richer gloss of purplish, changing to greenish. An apparently good diagnostic feature is the concealed pure white of the feathers about the anal regions, replaced in subis by grayish, rarely approximating to whitish.

A Progne collected by Mr. Wright, at Monte Verde, is duller in color than that from Remedios, but has still more concealed white below, in the median portion, not only of the anal feathers, but of those of the entire crissum and of the belly. A female bird, which I presume to be the same species, can scarcely be distinguished from the female of dominicensis, except in the brownish shafts of the longer crissal feathers, and an almost imperceptible tinge of brownish in the webs of the same feathers. It is almost exactly like the P. leucogaster of Mexico and Central America.

This species is included in the North American fauna in consequence of the capture of a specimen (No. 10,368 ♂ juv., May 18, 1858) at Cape Florida, which is with scarcely a doubt referable to it. This specimen is a young male in its second year, so that it is difficult to ascertain positively its relationship to the two allied species; but as it agrees perfectly in its proportions with cryptoleuca, and its plumage differs from the corresponding one of subis in essential respects, we have little hesitation in referring it to the former.

Nothing distinctive is recorded as to the habits of this bird.

Petrochelidon, Cab. Mus. Hein. 1850, 1851, 47. (Type, Hirundo melanogaster, Swains. = P. swainsoni, Scl.)

Gen. Char. Bill stout and deep, somewhat as in Progne. Nostrils entirely superior, open, without overhanging membrane on the inner (or upper) side, but somewhat overhung by short bristles, seen also along base of inner mandible and in chin. Legs stout; the tarsi short, not exceeding the middle toe exclusive of its claw; feathered all round for basal third or fourth, though no feathers are inserted on the posterior face. Tail falling short of the closed wings, nearly square or slightly emarginate; the lateral feathers broad to near the ends, and not attenuated.

Of this genus as restricted we have but one species in North America, although several others occur in the West Indies and the southern parts of the continent. All have the back steel-blue, with concealed streaks of white; the rump, crissum, and a narrow nuchal band, and usually the forehead, chestnut.

Petrochelidon lunifrons, Baird.

CLIFF SWALLOW; EAVE SWALLOW.

Hirundo lunifrons, Say, Long’s Exp. II, 1823, 47 (Rocky Mts.).—Cassin; Brewer, N. A. Ool. I, 1857, 94, pl. v, No. 68.73 (eggs).—Baird, Birds N. Am. 1858, 309.—Lawrence, Ann. N. Y. Lyc. 1861, 317 (Panama R. R.; winter).—Verrill, Pr. Bost. N. H. Soc. 1864, 276 (migration and history).—Lord, Pr. R. A. Inst. Woolwich, IV, 1864, 16 (Br. Col.; nesting).—Cooper & Suckley, P. R. R. XII, II, 184 (Wash. Terr.).—Dall & Bannister, 279 (Alaska).—Cooper, Orn. Cal. 1, 1870, 104.—Samuels, 256. Petrochelidon l. Baird, Review, 1864, 288. H. opifex, Clinton, 1824. H. respublicana, Aud. 1824. H. fulva, Bon. (not of Vieillot).—Aud. Orn. Biog. I, pl. lviii.—Ib. Birds Am. I, pl. xlvii.—Maxim. Cab. Jour. VI, 1858, 100.

Total length, 5.10; wing, 4.50; tail, 2.40, nearly even; difference of primary quills, 2.10; length of bill from forehead, .38, from nostril, .25, along gape, .60, width, .50; tarsus, .48; middle toe and claw, .72; claw alone, .22; hind toe and claw, .44; claw alone, .20.

Hab. Entire United States from Atlantic to Pacific, and along central region to Arctic Ocean and Fort Yukon; Panama in winter. Not noted at Cape St. Lucas, in Mexico, or in West Indies.

There is no difference between the sexes, but the young bird is very different from the adult in the following particulars: the steel-blue above is replaced by a lustreless dusky-brown, the feathers (except on head) being margined with a creamy tint; the neck merely tinged with rufous; the throat has only a dusky suffusion, and the chin is much mixed with white; the frontal patch is obsolete.

A closely allied species from Mexico, P. swainsoni (see Baird, Rev. Am. Birds, 1865, 290), possibly yet to be found near our southern border, differs as follows:—

Frontlet reddish-white, with narrow band of black along upper mandible lunifrons.

Frontlet chestnut-brown, without black at base of upper mandible. Size smaller swainsoni.

Sometimes (as in 11,027 ♀ and 11,025 ♂, Fort Bridger) the black patch extends upward, somewhat broken, however, to the bill.

Habits. The early history of the Cliff Swallow must always remain involved in some obscurity, so far as concerns its numbers and distribution before the first settlement of the country, and even down to the early portion of the present century. Its existence was unknown to Mr. Wilson, and it was unknown to other naturalists until obtained by Say, in Long’s expedition to the Rocky Mountains in 1820. It is now known to occur nearly throughout North America, and to breed from Pennsylvania to the Arctic regions, and from the Atlantic to the Pacific. Yet to many parts of the country it is a new-comer, where, a few years since, it was entirely unknown. It seems to be probable that at first this species was to be found only in certain localities that offered favorable places whereon to construct their nests. Where high limestone cliffs abound, these birds may have always occurred, although escaping observation.

In 1824, De Witt Clinton read a paper to the New York Lyceum, stating that he had met with these birds at Whitehall, N. Y., at the southern end of Lake Champlain, in 1817, about the time of their first appearance on the Ohio; and Rev. Zadock Thompson met with them in Randolph, Vt., at about the same period. General Dearborn noticed them for the first time in Winthrop, Me., in 1830. They first appeared at Carlisle, Penn., in 1841.

Professor Verrill discovered, in 1861, a large colony of these birds breeding on the high limestone cliffs of Anticosti, apparently in their original condition, and entirely removed from the influences of man. This suggested an inquiry as to their early presence in Northeastern America. From the information he received, he was led to conclude that this Swallow was known to certain parts of Maine earlier than its first discovery anywhere in the West. Whether these birds were indigenous to the West or not cannot now be determined. That they were discovered there only so recently as 1820 proves nothing. We only know that in certain localities—such as Rock River on the Mississippi, and at Anticosti on the St. Lawrence—their occurrence in large numbers in their former normal condition of independence suggests in either locality an equally remote beginning. It is possible, and even probable, that in favorable localities in various parts of the country they existed in isolated colonies. The settlement of the country, and the multiplication of convenient, sheltered, and safe places for their nests, gradually wrought a change in their habits, and greatly multiplied their numbers. At St. Stephen, N. B., and in that neighborhood, Mr. Boardman found this species as abundant in 1828 as they have been at any time since. They were then very plentiful under the eaves of several old barns in that part of the country. Yet twelve years afterward they were entirely unknown on the lower Kennebeck.

The nests of this Swallow, when built on the side of a cliff or in any exposed position, are constructed in the shape of a retort, the larger portion adhering to the wall, arched over at the top and projecting in front, with a covered passage-way opening at the bottom. The normal original nest, in a state of nature, is an elaborate and remarkably ingenious structure, sheltering its inmates from the weather and from their many enemies. Since they have sought the shelter of man and built under the eaves of barns and houses, the old style of their nests has been greatly changed, and the retort-like shape has nearly disappeared.

In building and in repairing their nests they work with great industry, and often complete their task with wonderful celerity. Where they exist in a large colony, it is not an uncommon thing to see several birds at work upon the same nest,—one bird, apparently the female owner, always assisting and directing the whole. After the work of construction has gone so far as to permit the occupation of the nest, it is often to be observed that the task of completing and improving the structure is kept up by the male. In a large colony of these Swallows, whose nests were built under the projecting roof of a barn in a small island in the Bay of Fundy, every nest was as open as are those of the Barn Swallow. These birds had been encouraged to build by the owner, and boards had been placed above and below their nests, of which they at once took advantage to build an unusual nest. These nests are made of various kinds of adhesive earth and mud. They are neatly and warmly lined with fine dry grasses and leaves, intermingled with feathers, wool, and other soft, warm substances. It has been thought that the mud of which these nests are composed is agglutinated by the saliva of the birds; but of this I have never been able to detect any evidence in the nests themselves, the crumbling nature of which when dry is against this supposition; and the birds themselves are often to be seen about puddles of water, apparently gathering materials.

When the nests of a large colony are invaded, the birds manifest great uneasiness, collecting in a swarm over the head of the intruder, wheeling around in circles, uttering loud outcries, and even flying close to his head, as if to attack him, with loud snapping of the bills.

The song of this Swallow is an unmusical creak, rather than a twitter, frequent rather than loud, and occasionally harsh, yet so earnest and genial in its expression that its effect is far from being unpleasant.

Genus HIRUNDO, Linn.

Hirundo, Linn. Syst. Nat. I, 10th ed. 1758, 191. (Type, H. rustica, Linn.)