From its unknown winter quarters, somewhere in Central America or on the South American Continent, the chimney swift comes northward in spring and spreads out over a wide area, which includes a large part of the United States and southern Canada.

Individually the swift is an obscure little bird, with a stumpy, dull-colored body, short bristly tail, and stiff, sharp wings, but it is such a common bird over the greater part of its breeding range and collects in such enormous flocks, notably when it gathers for its autumnal migration, that as a species it is widely known.

The birds also have the habit of continual flight during the hours of daylight throughout the summer, and therefore keep always before our eyes when we look up at the sky. They exemplify speed and tireless energy; they sail and circle on set wings, then with flickering wing beats they are off in a burst of speed, shooting like an arrow through the air, chattering their bright notes as they race along--little arrows "cutting the clouds" over country, town, and woodland.

Spring.--Swifts move up into the northern latitudes only when spring is rather far advanced, not until their aerial insect food is plentiful well above the ground. Therefore their arrival varies a good deal from year to year.

Kopman (1915) reports that the average date of appearance in New Orleans is about March 18. In New England, in an average year, we do not expect the birds for fully 30 days after this date; hence we may infer that they spend a month moving across a dozen degrees of latitude.

A daylight migrant, solely, so far as is known, we see the first arrivals of this swift commonly in the afternoon, sailing in small companies--perhaps only a single bird--often high in the air. As they fly along, they give an occasional chatter, or a few rather feeble chips, but with none of the energy and volubility characteristic of the breeding season. On cloudy days in spring, when the swifts dip down over the surface of a pond and feed among the twittering swallows--a common habit of theirs--they are apt to be silent.

When the birds appear, leisurely drifting up from the south, they often fly in great loops. They turn slowly aside from their northerly course, swinging farther and farther around until they are moving for a time toward the south, then, veering gradually, they resume their journey, but soon turn again and make another sweeping curve, each loop carrying them nearer their destination.

An hour before dark, in the lengthening evenings of early May, we often see a little gathering of New England swifts that have settled on their nesting grounds but are not occupied as yet with breeding activities, flying about in company, high over their chosen chimney, chattering together. The birds may be so high in the air that the sound of their voices barely reaches our ears. These newly arrived birds pay little attention to each other and do not approach near or chase one another as they will in June, yet they keep in a loose flock, sailing and flickering in a somewhat circular path and sometimes coast down from their high elevation, and climb up to it again. Then, as dusk deepens, at about the time the bat appears, they gather around their chimney and drop into it.

Although swifts, during their spring migration, often collect, before going to roost, in flocks of considerable numbers, they are less conspicuous at this season than during their impressive gatherings in the autumn. These are described under "Fall."

Courtship.--In June, here in New England, the swifts become very noisy. Even from within doors we hear their voices as the birds hurry past not far from our roof. As we listen their chips appear sharper and faster than they did the week before, more clearly enunciated, and they run in a long series that seems to grow in intensity as the birds come nearer, reaches a maximum when they pass overhead, and dies away as they rush on.

When we watch the birds at this season we notice also a difference in their behavior. There is little of the slow, apparently aimless circling of early spring, when, although the birds gather in small companies and follow similar paths in the air, they are seemingly indifferent to one another's presence.

The breeding season is here. Purpose has come into the swift's brain, and purpose has brought intensity and speed, and concentration on a mate. Now they fly close together, two birds, three birds, sometimes four in a little bunch. The length of a swift's body scarcely separates them as they tear along, ripping through space, following the twists and turns of the bird in the lead.

Soon two birds are left alone, the others circling off for a time. Both of these birds are chipping sharply, flying fast, close together, and during their mad dash, one, if not both, uses a peculiar note--a line of chips, a chatter, then the chips again. This combination of notes accompanies the height of the pursuit, and the swifter and closer the chase the sharper and quicker the notes. It seems also that the nearer the birds are to one another, the faster they fly. They may fly sometimes, their wings almost touching, at a pace that seems reckless; then the notes spatter out as if self control were lost, and at last, as the pursuer overtakes his goal, he rises a little above her and lifts up his wings, and there appears to be a moment of contact.

The probability that the nuptial flight leads, at least sometimes, to sexual contact in the air is increased by Sutton's (1928) careful study of the swift. He says: "In this courtship flight, the pair of birds may fly rapidly about, twittering loudly; suddenly the upper bird will lift its wings very high above the back and coast through the air, sometimes for several seconds, while the bird beneath may soar with its wings held in a fixed position below the plane of the body. It may be that this graceful and interesting display is at the culmination of courtship activity."

From the fact that swifts in the courting season so often fly three together when engaged in their pursuits--in the initial part at least, for at the culmination the pair find themselves alone--a surmise has arisen that one male and two females make up the trio and that the swift is polygamous. This surmise, however, is not yet attested by any conclusive evidence.

Nesting.--Of the few North American birds--and they are very few--that were influenced favorably by civilized man when he settled on this continent, the chimney swift received the greatest benefit. Before the coming of man, the swifts had been building their nests for thousands of years in hollow trees, here and there in the American wilderness. Then man came, and unwittingly supplied, within his chimneys, exactly the situation the swift required for nesting, an upright surface inside a cavity, protected from the weather--the equivalent of a hollow tree. Thus the birds' nesting sites were increased a millionfold.

Nowadays the typical site is in a chimney, "from near top to 22 feet below it," Forbush (1927) says. Yet, as the following quotations show, the swift occasionally avails itself for nesting purposes of some other of man's works; also from time to time it is found breeding in its ancestral manner.

The nest itself is a little hammock--half-saucer-shaped--composed solely of dead twigs, which the bird breaks off as it flies past a tree. The twigs are attached to the wall, and the twigs themselves are fastened to one another by the glutinous saliva of the bird, which hardens and fixes the structure so firmly to its support that it withstands, as a rule, the rain of summer storms.

Lewis (1927) reports that "a pair of Chimney Swifts built a nest and hatched a brood of young in an open well near an old deserted farm-house in the southern part of the county [Lawrenceville, Va.]. The nest was typical for the species and was stuck just above a bulge in a rock in the well wall, just as they are stuck to the rocks in a chimney. It was located about 7 feet below the surface of the ground, and 10 feet above the water."

Hyde (1924) found an occupied nest "in an abandoned cistern about one mile east of the town of Magnolia, Putnam County, Ill." He says: "The cistern was half hidden by vegetation. The diameter at the aperture was three feet and at the bottom nine feet. There was water nine feet below the aperture. The nest was in an entirely sheltered position four feet above the water. All these figures being approximate."

Kennard (1895), speaking of a nest in New York State, says: "I found a Chimney Swift's nest placed just under the ridge pole of an old log barn and against the side of one of the logs of which it was constructed. . . . It was within a foot of an enormous hornet's nest. The five young birds which were nearly fledged were clinging to the bark of the logs in the immediate vicinity and seemed to get on much better with the hornets than I did."

Evermann (1889) describes a "peculiar nidification of this species" as follows: "A pair fastened their nest in 1884 upon the inside of the door of an out-house at the Vandalia depot in Camden [Indiana]. The birds entered the building through small holes made in the gables. This building was in daily use, but those who visited it were cautioned by the railroad agent to open the door with care so as not to jar the eggs from the nest. Four eggs were laid, one of which was jostled from the nest, the other three hatched, and the young were reared in safety. The nest was repaired and used again in 1885, and again in 1886, a brood being reared each season."

Most astonishing records of nesting are reported by Moore (1902b) thus: "In this locality [Scotch Lake, New Brunswick], more nests are built inside buildings than there are inside chimneys. The nests are usually glued to the gable end of the building--sometimes barns, sometimes old uninhabited houses are chosen--and one nest, the past summer, was built in a blacksmith shop within fifteen feet of the forge. A number of years ago a pair nested in the upper part of a house in which a family lived, and near to a bed in which children slept every night. In this case the birds entered through a broken window."

Daniel (1902) gives an instance of the nesting of swifts in hollow cypress trees in the Great Dismal Swamp. He says:

Along the southeastern shore, growing in the lake some distance out from the shore line, are a number of large hollow cypresses. The roots or "knees" of these trees extend upward and outward from the surface of the water, curving inward some distance up, and in most of them, between the water and the base of the tree proper, there are openings large enough for a canoe to enter. By pushing our canoe in these intervals between the roots, we were able to examine the interiors of the hollow trees. In these we found the swifts nesting in their primitive fashion, the nests being fastened to the interior walls about midway down.

T. E. Musselman wrote me in 1935 that he has noted that swifts are beginning to use silos as nesting sites in the Middle West.

Eggs.--[AUTHOR'S NOTE: The chimney swift lays three to six eggs, more commonly four or five. These are pure white and only moderately glossy. In shape they vary from elliptical-ovate to cylindrical-ovate. The measurements of 56 eggs average 20.10 by 13.24 millimeters; the eggs showing the four extremes measure 21.59 by 13.46, 21.34 by 13.72, 17.53 by 13.72, and 18.29 by 12.70 millimeters.]

Young.--The young swift starts life in a world of danger. It comes from the egg a blind little naked thing, no bigger than your fingernail, lying in a frail cradle of sticks that overhangs a black "drop into nothing." The little swift, however, is equipped to deal with the dangers of its birthplace. Very early in its life it can cling and crawl; it can hide under its nest; it can move about over the walls of the abyss in which it lives; and, when the time for flying comes, it can clamber toward the free air, taking, perhaps, the longest and last walk of its career.

Frederic H. Kennard illustrates in his notes the hardiness of the young swift when it comes from the egg. He says: "On July 15, 1918, somewhere between 9 and 10 a.m. at Duck Lake, Maine, I found among the ashes of the fireplace, in a friend's unoccupied camp, a chimney swift's nest which I had been watching and which, when I had last seen it, the previous noon, in the chimney, had contained two eggs. It had evidently been dislodged from its proper place by a thunderstorm and torrential rain of the night before.

"The nest and both eggshells lay among the ashes, close together, at the back of the large fireplace. Both eggs were broken, and the shells lay just where they had fallen. One of them was evidently addled, while the content of the other was apparently missing.

"Imagine my surprise, when after hunting for some time, I discovered that the content of the other egg was a tiny swiftlet, which, blind and with the back of its skull badly bruised and suffused with blood from its fall down the chimney, had nevertheless made its way out through the ashes, dropping down the thickness of a brick from the fireplace proper to the hearth beyond, then across the hearth; and had climbed, in a style worthy of a young hoatzin, and was still clinging in an upright position to the finely woven wire fender that had enclosed the fireplace, but which I had moved aside in order to facilitate inspection.

"Of course this bird might have been hatched the day before, sometime between noon and the time of the storm, which occurred about 10 p.m., but my impression was, from the position of the eggshell as it lay broken in halves among the ashes, that the little fellow ready to hatch had come down either in the shell or in the act of emerging from it. There were no signs of any yolk to be seen. Judged from his size and development, he must have been less than a day old. There had been but one nest in the chimney; and there was no possibility of any outside intermeddling, as the camp was kept locked and I had been the only one to enter it in weeks."

Of the young nearly ready to fly he says: "Found a swift's nest down in Charlie Boyce's boat house on the wall about 5' 9" above the floor, and the four nearly fledged young clinging to the pretty smoothly sawed board wall from 18" to 24" away from the nest. Upon investigating I found that their toenails were long and sharp and that they could flutter up or across the wall at will, though when undisturbed they kept well together in a compact little group, propped up on their tails. When disturbed the young birds squealed loudly something like an exaggerated rattlesnake."

Burns (1921) states that the eyes of the young swift become wide open on the fourteenth day. This accords with the observation of May F. Day (1899), who watched at close range the development of a brood of swifts and noted that the incubation period was 19 days; that "even at the tender age that must be reckoned by minutes, these young birds were fed, seemingly by regurgitation"; that the "two first ventured from home when nineteen days old" and that they flew from the chimney four weeks after hatching. Speaking of the exercising of the nestlings, she says: "The young aspirants would stand in the nest and for a time vibrate the wings rapidly, so rapidly that the identity of wing was lost." And of the fledglings 26 days after leaving the egg she says: "They take flying exercises up and down the chimney, but I believe have not yet left it."

Carter (1924) studied the feeding of five fledgling swifts at a nest built on the wall of an abandoned cabin in Ontario. He says:

The old birds gained access to the interior of the building through a broken window and were remarkably tame, feeding the young within three feet of the observers, thus giving an excellent opportunity to observe the process of feeding. The parent, with greatly extended cheeks and throat, alighted upon the wall among the young. Immediately there was a great commotion. After a short hesitation a young bird would be fed by forcing some of the food from the mouth of the parent into that of the offspring. After a moment's feeding there was a pause and then the process was repeated, either to the same young or another. As many as three were served at a single visit.

Lewis (1929) describes thus the feeding of a brood of young birds that had fallen when their nest had been dislodged by rain but were clinging to the wall of the chimney:

From the start the old birds did not see me sitting on the hearth, or, seeing they paid little attention. I was much surprised to see that they always fluttered down and lit on the wall a little below the young birds, bracing themselves in the same manner as the young and reaching up to feed them. The young would turn their necks down as far as possible without changing the position of their bodies. The old birds would stretch up, putting the bill inside the gaping mouths of the young, and seemingly feed by regurgitation. This was invariable during the time I spent watching them, which amounted to a number of hours.

The four young clung to the wall without moving noticeably, always side by side, and were fed from daylight until dark at intervals of from 1 to 28 minutes until July 31, when I was obliged to leave home. [The nest had been dislodged on June 25.]

After the nest had fallen, but before the parents came down into the lower part of the chimney to feed their young, the little birds gave a note that Mr. Lewis describes as "a loud, harsh squeal, quite unlike the chattering they always make when being fed."

Townsend (1906) comments on the noisiness of a nestful of chattering swifts he found inside "a small hay barn" at Cape Breton. He says: "The shrill twittering of the young was almost deafening in the small hay loft."

Guy A. Bailey (1905), in a study of a swift's nest built inside a barn near Syracuse, N.Y., shows that the parent bird urges the young to leave the nest even before (according to his photographs) the flight feathers are more than half released from their sheaths. He says:

Generally, after feeding the young, the old bird crawled over to one side of the nest and cautiously insinuated its body behind the young birds. The adult bird kept crowding until all but one or two of the brood of five were forced out of the nest and took up positions on the vertical roost. The remaining birds would sometimes leave the nest of their own accord and follow their mates. This was noticed especially after those clinging to the boards had been fed.

It often happened that the adult birds would remain away from the young as long as twenty minutes, during which time the little ones would return to the nest. Usually, however, one parent would remain with the brood until relieved by the mate. On such occasions there was a period of several minutes when both parents were present.

Plumages.--[AUTHOR'S NOTE: The young swift is hatched naked and blind, but the spinelike quills soon begin to appear, and these develop into a juvenal, or first-winter, plumage, which is much like that of the adult; there are some light edgings on the scapulars and rump, which soon wear away, and the under parts are somewhat darker than in adults, especially on the throat. I have seen young birds acquiring the first winter plumage as early as August 10 and others still in the postnatal molt as late as September 25. This plumage is probably worn through the winter, though no winter specimens have been available for study. Forbush (1927) says that a "complete prenuptial molt beginning late winter or early spring is followed by a plumage as adult; adults apparently molt twice a year, a complete postnuptial molt in autumn and a partial (possibly complete) molt in spring."]

Food.--Pearson (1911) quotes a letter from W. L. McAtee of the United States Biological Survey: "The bird's food consists almost wholly of insects, and beetles, flies and ants are the principal items. It gets many beetles (Scolytidae), the most serious enemies of our forests, when they are swarming, and takes also the old-fashioned potato beetle (Lema trilineata), the tarnished plant-bug (Lygus pratensis), and other injurious insects. The bird is, of course, largely beneficial to the agricultural interests of the country."

Knight (1908) says: "The food of the Chimney Swallow consists of almost any of the smaller insects which fill the air of a summer's day."

Behavior.--The relationship between man and the chimney swift is a rather curious one. Although the species spends the summer scattered over a large part of the North American Continent, it never, except by accident, sets foot upon one inch of this vast land. The birds build their "procreant cradle" in the chimneys of thousands of our homes and crisscross for weeks above our gardens and over the streets of our towns and cities, yet, wholly engrossed in their own activities far overhead, they do not appear to notice man at all. Indeed, it is easy to believe that the swift is no more aware of man during the summer, even when it is a denizen of our largest cities, than when in winter it is soaring over the impenetrable jungles of Central America.

How do we regard this bird that does not know we are on earth? We are glad to have swifts breed in our chimney; we like to see them shooting about over our heads, and we enjoy their bright voices; yet, do we feel such friendship for them as we feel for a chipping sparrow, for example, which builds sociably in the vines of our piazza? The little sparrow may be wary, and may fly away if we come too near, but at least it pays us the compliment of recognizing our existence. The swift, however, is not even a semitrustful neighbor; it is a guest that does not know we are its host. We may almost think of it as a machine for catching insects, a mechanical toy, clicking out its sharp notes.

But let us note this fact. Every ten years or so the swifts do not appear about our house in the spring. Something has gone wrong on their journey northward. Our chimney will be empty this year; there will be no dark bows and arrows dashing back and forth above our roof, no quick pursuits and chattering in the evening. All summer something is lacking because there are no swifts to enliven the season. We realize, now that they are gone, how we should miss their active, cheerful presence, if they never came back again. But we may be sure they will come back--next year perhaps--to visit us again, this most welcome "guest of summer."

Bird banding has brought man and the chimney swift for the first time into close association. During the past few years, swifts have been banded in very large numbers. At daybreak, as the birds pour out of the chimneys where they have roosted during their autumnal migration, they are captured in traps placed over the chimney and so ingeniously devised that the outward flow of hundreds of birds is not interrupted. The banders who have handled the birds report that they show little or no fear (or consciousness) of man and appear tame to an extraordinary degree.

The following quotations from Constance and E. A. Everett (1927) illustrate their behavior after being caught. These authors state that: "In less than five minutes, with but one casualty, one hundred sixty-four Chimney Swifts were inside of that cage ['a six-foot house trap'], clinging to its walls of wire mesh like a swarm of bees, except that though densely massed, they were clinging to the wire and not to each other. A few were at all times on the wing, as they changed from one group to another, bewildered, perhaps, but not in the least frightened. Most of them, however, promptly alighted and tucked their heads under the wings and tails of those birds above them, until the inner walls of the cage took on the appearance of being shingled with birds."

When removed from the cage, "these swifts were very quiet, and apparently comfortable at all stages of the game. When held in the hands they would snuggle between the fingers confidingly; and when held against the clothes they would wriggle under the folds of the garments and contentedly go to sleep."

Of the next morning's work they say: "Since there were so few birds, we took the time to enjoy playing with them. Miss Constance and the boys tried wearing them either singly as a brooch, or collectively as a breast plate; and always the birds snuggled down as though perfectly willing to join the game, provided their naps were not interfered with. Finally some passing school girls were adorned with live breast pins to take home for show, while several birds, clinging to Constance's coat rode many blocks in the car, and, scolding, had to be dragged off to their liberty."

These observations were made at Waseca, Minn., on September 8 and 9, 1926.

In flight, the swift, perhaps from necessity because the bird spends so much of the day in the air, relieves its wings from time to time from their quick flickering and sails--the wings held motionless, fully extended from the body. While beating its wings, the bird appears always in a hurry; it seems to be moving them up and down as fast as it can; it often rocks from side to side, as it turns this way and that, and ever seems to be trying to fly a little faster.

Sutton (1928), in an able study of the swift's flight, aided by examination of captive birds, states that "no intermediate, half-spread position [of the wing] was ever maintained in healthy individuals. In fact such an intermediate position seemed impossible [on account of anatomical structure]. . . .

"It may be stated broadly, therefore, that the Chimney Swift wing, so far as its spreading is concerned, has but two normal positions; one, folded at rest, the other, open for flight, whether that flight be rapid forward flapping, soaring, coasting, or even sudden descent."

One evening Dr. Sutton, standing at the mouth of a chimney while swifts were going to roost, watched the birds enter, within arm's reach. Describing this experience, he says:

I was amazed at their precision and speed. As a rule, they slowed up abruptly just before making the final plunge, this being accomplished by a spreading and lowering of the tail, and by rapid, vigorous, downward and forward strokes of the wings, during which the loosely and widely spread primaries seemed to aid in checking the speed. When a proper point above the mouth of the chimney was reached the birds suddenly pressed the spread tail downward as far as possible, and with outstretched wings high above the back, still loosely fluttering, through an arc of about forty-five degrees, either dropped directly, turned jerkily from side to side, or twirled gracefully downward into the chimney.

Again, in the morning, peering down the chimney as the birds emerged, he says: "I was surprised to see that the birds were flying almost directly forward, but in an upward direction. Their bodies were not in a horizontal position; they were almost vertical, and the whole spectacle gave the impression that the birds were crawling up invisible wires."

For years there has been a controversy concerning the swift's flight. Some observers held that the swift moved its wings simultaneously, like other birds; others believed that the wing beats were alternate, like the strokes of a double-bladed paddle. It is easy to see how confusion between fact and appearance might arise. Swifts do appear to fly with alternate wing beats, but chiefly, if not wholly, when the birds tilt to one side in making their quick turns. Then one wing appears to be up and the other down, and as a matter of fact such is the case in reference to an imaginary line drawn across the swift parallel to the ground--one wing is above the line, and the other is below it. But the bird being tilted to one side, in order to show the relative position of one wing to the other, we must allow for the tilting, and we must draw the imaginary line, not parallel to the ground, but through the short axis of the bird's body. The observer, standing on the ground, does not make this adjustment, for he does not take into account the instantaneous tilting of the bird.

The question was definitely settled by Myron F. Westover (1932), who demonstrated by motion photography that "there was no instance where there was any alteration of wing-movement; the wings move in unison as do those of other species of birds." Dr. Chapman appends an editorial note to the article: "Mr. Westover's film was shown in the American Museum to members of the Bird Department who agree that it demonstrates beyond question the truth of his conclusions."

There is a difference of opinion also among observers as to whether the swift, when collecting nesting material, breaks off dead twigs with its feet or with its beak. Coues (1897), questioning the correctness of a drawing by Fuertes representing the bird snapping off a twig with its feet, says: "We have always supposed the bird secured the object with its beak, as it dashed past on wing at full speed; or at any rate that has been my own belief for more years than I can remember. But Mr. Fuertes vouched for the correctness of his representation from actual observation. The question being thus raised, I set it forth recently in a query inserted in one of our popular periodicals, asking for information."

There are six replies to Coues' query printed in The Nidologist (vol. 4, pp. 80, 81), five of which are in accord with his opinion, while one is against it, as is one more reply published in The Osprey (vol. 1, p. 122). Dr. Coues declares that "these leave the case still open!"

More recently Shelley (1929), from an ample experience of 13 uninterrupted years of observation of swifts at close range, states unequivocally that they "gather their nesting material. . .with their feet." He adds: "I never yet observed a Swift grasp or carry a twig in its beak."

Mr. Shelley's well-weighed opinion added to that of Mr. Fuertes, whose accuracy and skill in observing birds have never been surpassed, should be accepted with confidence until motion photography shall prove or disprove the correctness of their view, although the swift may adopt both methods of collecting nesting material.

We may recall that Audubon (1840) appears to have had no doubt upon the question, for as he says: "They throw their body suddenly against the twig, grapple it with their feet, and by an instantaneous jerk, snap it off short, and proceed with it to the place intended for the nest."

Although without much doubt swifts pluck off twigs with their feet, they may find it convenient to arrive at the nest site with their feet free to grasp the wall of the chimney. To gain this end, it is possible that on the way to the nest, the birds may transfer the twig to their beak, for William Brewster (1937b) in his Concord journal says that on June 15, 1905, he saw one "drop into the chimney this evening carrying a short twig held crossways in its bill."

An entry in Brewster's Concord journal also indicates that the swift may be more nocturnal in its habits than is commonly believed. He writes under the date August 5, 1893: "At about 2 a.m. I was surprised to hear Chimney Swifts twittering outside the window. There seemed to be a good many of them and the sound of their voices indicated that they first circled about the house several times and then went off towards the South. When I first heard the twittering, there were also several birds making their peculiar rumbling in the chimney, but this soon ceased and was not again repeated. The night was dark and still at the time, with rain falling gently and steadily."

We must remember also that Wilson (1832) states that "the young are fed at intervals during the greater part of the night," and Henry C. Denslow writes to Mr. Bent of "the vivid memory" of an observation of the birds' nocturnal activities. He says that "the chimney swift feeds its young in the middle of the night, going out and in the chimney several times with the usual rumbling of wing beats and the usual chirring sound of the young birds while being fed. I chanced to sleep in a small room with a chimney, near Rochester, N.Y., for several years, and so became familiar with this habit of the bird."

Frederick H. Kennard, in his notes, describes an interesting habit that he observed at very close range at a nest containing young built on the inner wall of a boat house. He says: "She (?) [a parent bird] sat very close, moving her head only occasionally, panting with the heat, and did not appear to mind me much until her mate flew in, lit on the wall nearby, when she got off the nest and fluttered up and down the wall beside or below the nest, snapping her wings together (apparently behind her), a note of warning or anger, or something of the sort, perhaps to scare me away or to show her displeasure. She would raise her wings slowly until they stood out straight behind her back, parallel and almost touching." And later he adds: "They do not seem to snap their wings except when disturbed by me. There is no snapping when they ordinarily leave their nests. When they do snap they slowly raise their wings until they are straight up from their backs and then snap them a couple of times."

At the same boat house he "saw one of the swifts fly through a crack in the door just after I had come out and closed it. He didn't slow down at all, never missed a beat, but merely turned on one side and went through, full speed."

Voice.--The notes of the swift remind us of the bird itself--energetic and quick; sharp and hard like the bird's stiff wings. The note most commonly heard as the birds shoot about over our heads is a bright clear, staccato chip or tsik--whichever suggests the sharper sound--often repeated in a series and sometimes running off into a rapid chatter. The chip note varies little, if at all, except for the quickness of the notes, and seems to punctuate the bird's ceaseless rush through the air. Sometimes, when the birds are very high in the air, the chattering call comes down to our ears, softened by distance--like sparks slowly falling to the earth after a rocket has burst.

Simple as these notes are, the birds introduce a good deal of variety into them by modifying the interval between them, thereby changing the expression of their lively theme.

One modification, which I have mentioned under "Courtship," having heard it only in the breeding season and only when the birds were under stress of excitement, serves to illustrate this ability and may be regarded as representing the song of the swift. It is made up of a long series of notes in which the birds, after giving several isolated chips, change abruptly to a series of very rapid notes, a sort of chatter, then with no pause between, change back to the chips, then back again--chips- chatter-chips, and so on. We may term it the "chips and chatter call."

Another modification of the chip note, often heard in summer when the birds are in a comparatively quiet mood, is a long chatter in which the volume increases and lessens, suggesting the sounding of a minute watchman's rattle.

There is one note quite different in quality from the above notes and less frequently heard than any of the variations of the chip. This is a musical monosyllable--sometimes divided into two syllables--a squeal, almost a high whistle with a slight upward inflection, like eeip, sometimes repeated once or twice. I have heard it both in spring and fall; hence it cannot be, as I once thought it was, a note of immature birds.

Field marks.--The swift may be distinguished readily from any of the swallows by the shape of its wings and the manner in which it moves them. Swallows' wings are roughly triangular, the triangle seeming to join the bird's body by a fairly wide base, whereas the swift's wings are narrow at the base--they are pointed, and slightly curved like the terminal part of a sickle's blade--and appear to be set on well forward.

The stroke of the swift's wing gives a jerky, hurried effect compared with the more leisurely movement of a swallow, and the tips of the wings are not swept backward, even when the birds are sailing, as they are, in varying degree, in the flight of all the swallows.

The swift has been likened to a winged cigar, tapered at both ends, flying through the air. The resemblance is very close, except when the bird fans out its stumpy tail, as it does from time to time.

The nearly uniform dull color of the swift's under parts and its very short, square tale, combined with its characteristic flight, serve to identify the bird even at a considerable distance.

Enemies.--Because the swift spends a large part of its life moving rapidly through the air, almost never coming to rest except at its nest or when roosting in a chimney or a hollow tree, it is practically out of reach of any mammal that otherwise might prey upon it. And while flying its speed and its erratic course render it almost immune to attack by hawks.

In his notes T. E. Musselman cites an exception to this immunity. He says: "I was watching a flock of about 1,500 swifts circling about a chimney in Quincy, Ill., forming an avian funnel which was dropping into its black depths. It was almost dusk when a sharp-shinned hawk flew from a neighboring sycamore tree to the top of the chimney and seized one of the swifts as it was poised with upturned wings and was just about to drop into its night's sanctuary. The swift squealed as it was being carried away, so I was able to follow the course of the tiny hawk as it flew through the semidarkness back to the tree."

Musselman (1931) also reports an occasion when swifts were overcome by gas while roosting in the chimney of a church in Quincy, Ill. He says: "One cold October night it was necessary to turn on the fire, which resulted in the killing of between 3,000 and 5,000 Chimney Swifts that had harbored there. Three bushel baskets of dead birds were taken from the flue."

Julian Burroughs (1922) tells of an instance in which a large number of swifts, taking refuge in a chimney, dislodged the soot. Many were smothered in the chimney, while others, several hundred evidently confused by the soot, continued down into the house where they were found "on all the mouldings and pictures." These were released apparently unharmed. "There were about fifty live ones and fifty dead in the furnace--also ten water-pails full of dead ones in the pipes and bottom of chimney."

The greatest hazard of the swift's life, perhaps comes in the spring or early summer when, once in a dozen years or so, a prolonged, drenching downpour of rain clears the air of insects, and threatens the local birds with starvation. Brewster (1906), referring to such a storm, says: "The Swifts. . .were seriously reduced in numbers, throughout eastern Massachusetts, during the cold, rainy weather of June, 1903, and the losses which they suffered that season have not as yet been made good." Since 1903 the birds here in New England have been decimated by several minor storms but have quickly recovered their loss.

Fall.--Fall comes early in the yearly cycle of the swift's life. At the end of the summer there is a long journey before the birds, old and young, to the warm air of the Tropics where they can find food throughout the winter months.

Late in July and early in August we often see small groups of swifts in the air, evidently preparing for migration. These flocks are doubtless made up of our local birds, those that have spent the summer in our vicinity, and they are accompanied, presumably, by as many of their young as are on the wing. They travel such long distances through the air, often curving round and round a chimney or church tower, that they derive a good deal of exercise from the flights--exercise that must serve to strengthen the wings of the young birds.

Under date of August 7, 1917, my notes mention this habit. "In these exercising flights, as I take them to be, the birds fly mostly in long curves; they are really circling, although they may turn at any time to either side. The birds, a dozen or more of them, are sailing in a great ring; they suggest bits of wood floating in an eddy of a slow-moving stream. They are far from one another, flying silently, mainly on set wings. One veers toward another, which quickens its pace by rapid, flickering wing beats. A chase is on. One or more birds join in, giving the long chatter. Now they hurry through the air, close together. When one comes near another, it may raise its wings in a V above its back, soaring for a moment. The chases are soon over, however; the birds seem to lose interest in speed and resume their circular, soaring flight. They often turn out from the circle, tilting to one side, the outer wing uppermost.

"During the middle of the day I do not see the swifts gathered about the house; it is chiefly in the morning and evening that they are most active. This evening two birds, close together, flew slowly over my head at a low elevation. One gave the long chatter and chips alternately, but in a quiet way with little staccato quality."

It is at about this time, the first two weeks of August, that we see evidence of molting in our local swifts. As they fly overhead we notice a narrow gap in their wings where the flight feather or two is missing, and in every case the little gap in one wing corresponds almost exactly with the gap in the other, but this slight bilateral loss of wing surface seems to hinder the birds' flight little, if at all. Apparently molting does not cripple them, as it does many woodland birds; indeed, the swift, spending the hours of daylight in the sky, must not be disabled for even a single day.

The most spectacular event in the swift's life, from our point of view, occurs during the autumnal migration when the birds, late in the afternoon, congregate in a large, wheel-shaped flock and circle about the chimney they have selected as their roosting place for the night. The following quotations describe in detail such gatherings.

Townsend (1912), writing of the bird in the St. John Valley, New Brunswick, says:

At Frederickton, on July 25, I watched a large flock of Swifts enter for the night a chimney on the southwest corner of the Parliament Building. Sun set at about 8 p.m. At 8:24 p.m. one bird set its wings and dropped into the chimney and soon they began dropping in fast, while the flock circled first one way then another or crowded together in a confused mass, twittering loudly all the time. Owing to the proximity of the dome, regular circling was somewhat interfered with, but as a rule the birds circled in the direction of the hands of a clock, and individuals would drop out and into the chimney in dozens when the circle passed over it. Occasionally they would all swoop off to the other side of the building, soon to return. At 8:45 p.m. practically all the birds had entered the chimney and I had counted roughly--at first singly and later by tens--2,200 birds.

The setting of the wings, which Dr. Townsend speaks of, takes place just over the mouth of the chimney. The bird raises its wings above its back and drops into the chimney or very often shies off, like a horse refusing to take a fence, and, after making another circuit, tries again.

Linton (1924) gives us a vivid picture of the Swifts "at bedtime," showing a spirit of play among the birds. He writes from Augusta, Ga.:

October 5, 6:05 p.m.: Sky overcast; large numbers of Swifts in the upper air; look like swarm of bees; general direction of flight in circle, counter-clockwise. 6:07: A few began to enter the chimney, when a passing auto frightened them for a short time. 6:08: Entering again, average probably not far from 15 per second, at times many more than this [the flue of this chimney was said to be 3 feet square]; circling continuously counter-clockwise. As the circle approaches the chimney, a column of Swifts, from a point 20 feet above the level of the top of the chimney descends to the chimney. The Swifts in this column which fail to enter continue the circle at a lower level, joining the higher level at the opposite side of the circle, and in a position which makes them contributing parts of the descending column, when they again come to that point. Great swarms of Swifts could be seen in the upper air, their paths apparently crossing and recrossing, but really all flying in circular paths at different levels. Many appeared as minute specks in the upper air. At 6:23 all were in, stopping abruptly; probably no more than a dozen stragglers in the last 5 seconds. It thus took the flock a little over 15 minutes to enter the chimney.

October 8, 6:12 p.m.: Sky clear; 3 or 4 Swifts seen from window. 6:13: 12 or more Swifts in sight. 6:15: 100, more or less, in sight. 6:15:20: 500, more or less, in sight. 6:16: increasing in numbers rapidly; general course in wide circles, counter-clockwise. 6:17: Seem to be enjoying themselves too much to go to bed; immense numbers; upper air full of them. 6:19:20: Getting closer to chimney; some of them dipping down to within a foot or two of the top. 6:20: Changed their minds for a few seconds; again enjoying themselves in the air. 6:21: Getting closer again. 6:21:30: Changed minds again. 6:22:30: Look as if they were getting ready to go to bed. 6:23: Getting closer; circles variable, 150 to 200 feet in diameter nearest level of top of chimney, lower portion, at times possibly no more than 50 feet in diameter. 6:23:30: Passing near top of chimney. 6:24: Passing very close to top of chimney. 6:24:30: A few going in. 6:24:40: Entering at a rate of 15 or more per second. Same maneuvers as on the previous evenings. 6:25:30: Going in very rapidly; 15 per second a very conservative estimate. 6:28: A second or two when they did not go in so rapidly, being disturbed by the puffing of a locomotive on the Georgia Railroad near by. 6:28:30: Going in as rapidly as ever. 6:30: All in; stopped suddenly.

Audubon (1840) gives an interesting account of a large number of swifts he found roosting in a hollow tree in Louisville, Ky. He says:

I found it to be a sycamore, nearly destitute of branches, sixty or seventy feet high, between seven and eight feet in diameter at the base, and about five for the distance of forty feet up, where the stump of a broken hollowed branch, about two feet in diameter, made out from the main stem. . . . Next morning I rose early enough to reach the place long before the least appearance of daylight, and placed my head against the tree. All was silent within. I remained in that posture probably twenty minutes, when suddenly I thought the great tree was giving way, and coming down upon me. Instinctively I sprung from it, but when I looked up to it again, what was my astonishment to see it standing as firm as ever. The Swallows were now pouring out in a black continued stream. I ran back to my post, and listened in amazement to the noise within, which I could compare to nothing else than the sound of a large wheel revolving under a powerful stream. It was yet dusky, so I could hardly see the hour on my watch, but I estimated the time which they took in getting out at more than thirty minutes. . . .

The next day I hired a man, who cut a hole at the base of the tree. . . . Knowing by experience that if the birds should notice the hole below, they would abandon the tree, I had it carefully closed. The Swallows came as usual that night, and I did not disturb them for several days. At last, provided with a dark lantern, I went with my companion about nine in the evening, determined to have a full view of the interior of the tree. The hole was opened with caution. I scrambled up the sides of the mass of exuviae, and my friend followed. All was perfectly silent. Slowly and gradually I brought the light of the lantern to bear on the sides of the hole above us, when we saw Swallows clinging side by side, covering the whole surface of the excavation. In no instance did I see one above another. Satisfied with the sight, I closed the lantern. We then caught and killed with as much care as possible more than a hundred, stowing them away in our pockets and bosoms, and slid down into the open air. We observed that, while on this visit, not a bird had dropped its dung upon us. Closing the entrance, we marched towards Louisville perfectly elated. On examining the birds which we had procured, a hundred and fifteen in number, we found only six females. Eighty-seven were adult males; of the remaining twenty-two the sex could not be ascertained, and I had no doubt that they were the young of that year's first brood, the flesh and quill-feathers being tender and soft.

Audubon estimates that the number of birds "that roosted in this single tree was 9,000." This investigation took place "in the month of July." He visited the tree again on August 2, after the local young birds "had left their native recesses." Of this visit he says: "I concluded that the numbers resorting to it had not increased; but I found many more females and young than males, among upwards of fifty, which were caught and opened."

Musselman (1926), writing of swifts overtaken by wintry conditions with snow in Quincy, Ill., says: "I discovered that on days when the thermometer indicated an approach to the freezing point the birds remained in the chimneys until about nine o'clock in the morning. During the daytime the birds quickly returned from their feeding over the river, circled but a time or two, and dropped into the chimney until warm. . . .

"The most popular chimneys were those which connected below with the basement, and served, therefore, as warm air flues. In such chimneys the temperature reached 70^o. Little wonder that the birds preferred these chimneys on damp and cold nights!"

The two following quotations describe very unusual departures from the swift's regular habit of roosting. Latham (1920), writing from Orient, Long Island, N.Y., states:

About one p.m. August 17, 1919, while collecting insects near the eastern border of a broad brackish meadow, my attention was attracted to Chimney Swifts (Chactura pelagica) frequently flying slowly in from the west and disappearing in the fringe of vines and shrubs that separated me from the extreme east boundary of the marsh. In this heavy growth, from waist to head high, were elderberry bushes (Sambucus canadensis) heavily hung with ripe fruit. I selected a bird for special study. It advanced on descending, hovering flight. About four feet above the tangle, near the farther side, it paused and dropped abruptly into a clump of elderberries. Carefully marking the locality, I worked my passage to a few feet of the spot. The swift was clinging to the cymoid head of the elder eating the fruit. The ease with which the bird took flight from its slender perch, rising directly upward several feet above the cover and dropping rail-like back into it, was interesting and worthy of note.

The cover harbored at the time not less than fifty swifts. Most of them were flushed with more or less difficulty, but some individuals took wing within arm-reach of the observer. No others were noted eating fruit. . . .

It is evident that the birds had established a roosting, or resting place out of the ordinary. It is not satisfactorily settled whether the birds sought the brush to feed on the elderberries or for shelter. The writer is of the opinion that the bird seen eating berries was only an exceptional case where the bird took a berry after alighting within reach of it.

E. K. and D. Campbell (1926) report from Cold Spring, N.Y., an astonishing roosting place for swifts, the bark of an oak tree. They state:

At 2:30 p.m. September 5, 1926, we observed an excited flock circling between the house-front and the adjacent oak trees, and above the house-top and back. Their flight seemed to focus at a point 25 feet up on the trunk of a tall oak. The day was dull and we judged there was some sort of food there. Really, however, they were gradually alighting on the bark, as we discovered at 4:30 p.m. when most of the flock was found to have grouped itself in close formation, as shown in the rough sketch. . . .

The birds seemed two or three deep, and several of us estimated well over a hundred of them. They were snuggled together, seemingly to keep warm, and the heads all concealed beneath the wings of those above. This patch of birds was of irregular shape, nearly 5 feet high and 7 to 8 inches wide at the widest part. It was constantly changing, as some birds seemed to lose their grip and fly off and return, so that a dozen or two were on the wing and seeking a place to work into the group. We saw some alight at the edge and work up close, while others lit in the middle of the group and must have reached through with claws to grasp bird or bark, those failing falling back and taking wing. All had their heads concealed but the few upper ones. Toward dusk the birds, matching the moist bark, were invisible, but we examined them again by flashlight after dark, and all was quiet.

Next morning, to our surprise, they were still there, in broad daylight, and some remained through the afternoon.

Cottam (1932) describes in detail some remarkable gatherings of swifts "at night circling the great dome of our national capitol, feeding on the small insects attracted there by the powerful flood lights." He observed the birds on many evenings, both in spring and fall, once in a flock of "approximately 2,000," circling "the dome--the area of greatest light concentration--where they remained until the lights were turned off shortly after midnight." Of the bird's evolutions, he says:

On the nights when flocking occurred at the capitol, the birds began to arrive in small groups from all directions about sundown, and by the time they normally would have been going to roost they had formed into one great swarm. For the first fifteen or twenty minutes after sundown the birds foraged over the tree tops and flew in all directions without any apparent system to their movements, except that they remained in a rather restricted area. Gradually, as it grew darker, a greater number were seen to fly more or less in the same general circular direction; in other words, there was a distinct impression of group movement. About the time the lights came on shortly thereafter, all were following a definite course. Each time flocks of incoming birds disrupted the rhythm and unison of the concentric flight there was a momentary disbanding. When they reformed, however, all seemed instinctively to fly in the same direction. Most often the flight was uniformly circular, but occasionally it took the form of a conical cloud somewhat resembling a cyclone funnel. On one occasion it was seen to form a great figure "8" with one loop at a lower elevation than the other.

Frederic H. Kennard makes this note of an unusual roosting place at Duck, Maine: "In the evening [August 8, 1924] I was treated to a performance of flocking, roosting chimney swifts, which at sundown flocked up and, flying in circles about one end of the smaller of Charlie Boyce's barns, gradually dribbled into a little window up under the ridgepole of the gable end and were clinging by the hundreds in an almost solid sheet against the gable end of the barn. We climbed up into the hayloft and flashed a light onto them, and they gradually flew out until only perhaps 75 or 100 were left. There must have been 500 or 600 in all, though, of course, difficult to estimate."

Winter.--In 1886 all that was known of the chimney swift in winter was that it passed south of the United States (A. O. U. Check-list of North American Birds, ed. 1, 1886). The third edition, published in 1910, adds that the bird winters "at least to Vera Cruz and Cozumel Island [Yucatan] and probably in Central America." The fourth edition, 1931, extends the probable winter range to Amazonia.

Chapman (1931) cites two specimens of the chimney swift, taken late in autumn in West Panama at a time when many South American bound migrants were passing through this region. He says:

If we may assume that they [the chimney swifts] winter in a forested, rather than an arid region it is not improbable that they were bound for Amazonia, where the presence as permanent residents of five species of Chaetura shows that the region offers a favorable habitat for birds of this genus. From at least two of the Brazilian species, pelagica could not certainly be distinguished in the air. Sight identification, therefore, is out of the question, and until a specimen is secured we shall not know where the Chimney Swift winters. But, as every collector of birds in tropical America knows, to see a Swift is one thing, to get it quite another. Native collectors are not willing to expend the ammunition required to capture Swifts, and even visiting naturalists secure comparatively few. With our attention directed toward Amazonia as the possible winter quarters of the North American species it may be long, therefore before our theory is confirmed by specimens.