Spring.--The tree swallow is the earliest of the six swallows that move up the Atlantic coast in spring on the way to their breeding grounds in the Northern States. It is a hardy species; many individuals spend the winter in the southern part of the United States and hence have a comparatively short migration to their summer homes. As they press northward, arriving in New England often late in March, they may encounter frost and snow, for at this season the advance of spring is very variable here. Under such conditions they often disappear for a time, and it is supposed that they retreat until spring steps northward again.

It is interesting to watch these early arrivals when caught by unfavorable weather. As an example of their resourcefulness, Richard J. Eaton relates the following account: "On the morning of March 19, 1939, G. W. Cotterell and I closely observed three tree swallows feeding on the ice on Heard's Pond, Wayland, Mass. The fields were blanketed with hard, crusty snow. The day was sunny and calm, the temperature slightly below freezing in the shade. We first saw the birds from a distance, flying back and forth over a restricted portion of the old white ice. The entire pond and the flooded meadow between it and the Sudbury River were frozen tight. We hitched ourselves (in a sitting position) to within 10 feet of the feeding swallows. Watching at this close range, we could see that they were feeding on what looked like seeds frozen to the surface of the ice in long windrows or irregular bands about 20 yards from shore. Maneuvering within a few inches of the ice in a zone about 30 feet long, they picked up the food while they were on the wing, making an audible clicking sound with their bills when pecking at the seeds. The swallows had discovered that a vigorous dig with the bill was necessary to dislodge the seeds. They accomplished this with an emphatic down thrust of their heads without interrupting their somewhat deliberate flight. Frequently the birds rested on the ice for a few seconds and occasionally pecked at the food in this position, like very young chickens.

"These swallows seemed to be plump, vigorous, and in good condition. After they left we carefully inspected the ice where they had been feeding and found an abundance of seeds, chiefly Scirpus and Carex. We discovered no living animal matter, not even snow fleas."

Walter Faxon and Ralph Hoffmann (1900) speak of conspicuous flights of tree swallows in Berkshire County, Mass. They say: "Vast numbers collect at the head of Pontoosuc Lake during the vernal migration, where we have seen them take their departure for the north as late as the 22d of May. After sitting toward the close of day upon the low bushes that protrude above the surface of the lake, at half past seven o'clock, myriads at once soared into the air, parting to form two flocks, one of which took a course due north, while the second struck off to the W.N.W."

As we see tree swallows arriving normally in "spongy April," when the weather is mild and insects are in the air, they often follow the course of a river, either low down over the water, feeding as they go, flying in long curves, often sailing on set wings over the greater part of a wide circle, or well up in the air in loose flocks, 20 to 30 perhaps, exchanging in their progress high, delicate salutations.

In Florida, late in February, there is a conspicuous migration from farther south. I have often seen large numbers passing over the Kissimmee marshes, moving steadily toward the north in wide array on a broad front, hundreds flying past within a few minutes. At such times they make use of a monosyllabic call that I have rarely heard on the breeding grounds. At a little distance it resembles the rough note of the bank swallow.

Courtship.--The tree swallow's courtship apparently consists in a pursuit that enables the male to display his proficiency and expertness in flying. Dr. Samuel S. Dickey describes thus "the maneuver of mating" of a pair just before the eggs were laid: "They gyrated rapidly to and fro, up and down; then mounting still higher than their previous level they dallied in midair. The male drew up to the female and grasped her breast feathers with his feet, and the two birds tumbled downward together, not parting until they were near the ground. The female then flew to the vicinity of the nest and settled down on a perch, lifted her wings slightly, and expanded her tail. The male glided above her and, dropping his wings, alighted on her back. I have never seen these birds have sexual contact in midair."

Francis H. Allen states that in the courtship flight the wings are never raised above the horizontal.

Observations on banded tree swallows show that a pair may breed together in successive seasons. Mrs. Kenneth B. Wetherbee (1932), Oscar McKinley Bryens (1932), and Lewis O. Shelley (1934a) speak of this habit, but Laurence B. Fletcher (1926) reports a case in which two birds, mated in 1925, returned the next year and paired off with different mates.

Shelley (1935) has shown that male birds not only may change mates from year to year but may have two mates at the same time. He explains irregular matings thus: "The first Tree Swallows of the season arrived at the station on April 3d, when the migration of the species was nearly over. About May 1st brown females commenced to appear, and they were common throughout their migration, which lasted to June 16th. . . . Since these unmated birds arrive to such an extent after nesting by older pairs is well under way, they are susceptible to mating, and they do mate with paired males, who often desert their former mates for the newcomers."

Observers have sometimes seen more than a single pair of adult birds feeding young in a nest, and Forbush (1929) says "occasionally three birds, usually two males and one female, engage in preparing a nest, incubating the eggs and feeding the young."

Harrison F. Lewis writes to Mr. Bent: "A pair of tree swallows nested on the grounds about my residence near Quebec, Canada, in 1920. The young birds left the nest on July 13, a few days later than did many other broods in the general vicinity. On July 11 I noticed that the young in my box were being fed by two adult males and one adult female. Later in the day I saw one adult tree swallow on the box, one on the wire nearby, and at the same time four more circling close overhead. It would appear that the young swallows in the box were being cared for by four to six adults, of which at least three were males and at least one a female. The males seemed to work together very harmoniously, but the female sometimes acted as if she objected to having so many males about her home. Whether the assistants to the actual parents of the brood were birds that had not bred successfully or were the parents of other broods already able to feed themselves is a matter of speculation."

Seth H. Low (1934) accounts for some instances of such behavior. He studied a colony of tree swallows whose broods had been decimated by a summer hail-storm, after which he says: "Three and even four adults were caught actually bringing food to the same brood. In each of three different boxes a female whose young had previously died was caught along with the rightful parents feeding the brood. Apparently this is a carrying over of the maternal instinct."

Nesting.--Before North America was settled the tree swallow built its nest in hollow trees, but with the advent of civilized man the bird quickly appropriated as nesting sites the houses that our ancestors put up to accommodate the purple martins--always a popular bird with the early settlers. Both Wilson and Audubon mention this habit.

At the present time tree swallows still build in old apple orchards and in holes in trees, especially when they stand in open ground near meadows or bodies of water, but they seem to prefer wooden boxes, even ramshackle affairs affording incomplete shelter.

The Austin Ornithological Research Station recently made an interesting experiment with tree swallows on Cape Cod, Mass. They put up a large number of breeding boxes and obtained astonishing results that indicate that the birds sometimes have difficulty in finding places to nest. Seth H. Low (1933), summarizing the results, says: "With 98 wooden boxes, mostly in the open, the population jumped in 1931 from 4 pairs to 60 pairs. In 1932, with over four hundred boxes available in favorable sites, there were 113 pairs of breeding birds." Austin and Low state (1932) "They had a choice of boxes in open fields, in partial cover, on the salt marsh, and in dense woods. They showed a preference for those in the open fields, most of which were occupied, and used none of the ones that were sheltered by vegetation."

Forced, perhaps, by a scarcity of breeding sites, tree swallows have been found nesting in several unusual places. For example, Dr. Dickey reports that they build their nests in the "eaves and cracks of log shelters erected by campers and foresters in lumber camps in Ontario and in old excavations of woodpeckers, notably those of the yellow-bellied sapsucker and arctic woodpecker." Milton S. Ray (1903) found a nest "in a hole of a pile of an old wharf, over the water." Henry Mousley (1916) says: "In my experience the nesting site here [Hatley, Quebec] is generally some small cavity in the eaves or cornices of farm buildings." Hartley H. T. Jackson (1923) says: "A nest containing five eggs was found in a fence post by the roadside, June 7, the entrance to the cavity being in the top of the post." John Treadwell Nichols (1920) speaks of a nest in an abandoned hydrant, the opening to the nest only a couple of feet from the ground.

The nest of the tree swallow consists of an accumulation of dry grass and straw, hollowed out, and lined with feathers. The birds show a marked preference for white feathers and often arrange them so that the tips curl upward over the eggs.

Oliver L. Austin, Jr., and Seth H. Low (1932), speaking of nests on Cape Cod, say:

The time occupied by the nest-building varied individually from a few days to two weeks. The peak of these activities occurred during the last week in April and the first week in May.

In general the foundation and bulk of the nests were of upland or marsh grasses, pine needles, or a combination of these materials woven together. A hollow was formed in this foundation, sometimes in the center, but as often in one side or corner, and profusely lined with gray and white feathers packed in tightly with the quills buried in the grass or pointing away from the central hollow. In forty-six boxes were found over 3,300 feathers, ninety-nine per cent of which were those of the Herring Gull, though a few feathers of domestic fowl, Black Duck, Scaup Duck, Wood Duck, Canada Goose, Great Horned Owl, and Red-tailed Hawk were mixed in with them. There was an average of 72 feathers to a nest, but over a hundred were found in each of ten boxes; one contained 182, one 184, and another a maximum of 147.

A. Dawes DuBois describes a nest "profusely lined with soft, pure white chicken feathers--a beautiful nest. After three eggs had been laid, I saw the male bird catch a large downy white feather that was floating on the breeze and carry it into the box." Thomas D. Burleigh (1930) found a nest "ten feet from the ground in a cavity in an old rotten willow stub in underbrush bordering a stream. . .built entirely of large chicken feathers."

The tree swallow commonly nests in isolated pairs, showing none of the strictly communal habits of the purple martin and the cliff swallow, but the birds are frequently found nesting in groups, their nests scattered about, not far apart, in favorable feeding localities. Ralph Works Chaney (1910) speaks of such a case in Michigan. He says: "Large colonies of these swallows nested in cavities of dead stumps which projected out of the lake." Charles L. Whittle (1926) sums up an investigation on the distribution of nests by concluding that "the determining factor is the adequacy of a nearby feeding area, or areas, be they meadows (old, filled lakes), marshes or water, to furnish the necessary quantity of food for the young at the requisite period."

The female bird builds the nest, aided little or none at all by her mate. Winton Weydemeyer (1934b) says: "Although the male often makes a pretense at gathering straws, and occasionally carries feathers to the nest, his principal job is that of overseer. . . . Only occasionally do the male birds share the task of incubating the eggs; frequently: especially during days when few insects are in the air: they carry food to their mates in the nests. Generally, however, the females during the day leave their eggs long enough to secure food for themselves." Both parents feed the young and remove excreta from the nest, dropping the sacs from the air a few yards away.

Winton Weydemeyer (1934b) states: "As a rule, at no time during the entire season do the males share the houses at night. . . .  Frequently the males perch on the houses for an hour or more after their mates have retired, not leaving for their own sleeping places until after darkness has settled." The same observer (1935) reports from Montana: "In the case of sixty nests a full record has been obtained of the percentage of hatch and survival of the nestlings." The latter shows 94.7 per cent for the first brood and 84.6 for the second brood. Austin and Low (1932) calculate a reproductive efficiency of 56.5 percent from 278 eggs.

Eggs.--[AUTHOR'S NOTE: The tree swallow usually lays four to six eggs; as many as 10 have been found in a nest, but any over five or six eggs may be the product of two females. The eggs vary from ovate, the usual shape, to elliptical-ovate or, rarely, elongate-ovate. They are pure white, unmarked, and usually without gloss.

The measurements of 50 eggs average 18.7 by 13.2 millimeters; the eggs showing the four extremes measure 20.3 by 13.5, 19.3 by 13.7, 16.8 by 13.2, and 17.8 by 12.7 millimeters.]

Young.--Austin and Low (1932) from a study of a large number of nests "found the length of the [incubation] period, estimated from the day the last egg was laid to the day the first one hatched, to vary from a minimum of 13 days to a maximum of 16." According to their records the young birds remained in the nest 16 to 24 days, and to account for this variation they point out "that the most food per young will be delivered in those nests containing the fewest nestlings, and hence the rate of growth will be slowest where the broods are largest." They state that "at no time were the young birds observed to return to the boxes once they had flown." Winton Weydemeyer (1934b), however, says: "For a few days after taking to the air, the young birds enter and leave the houses frequently, and remain in them all night."

The nestling tree swallow is an attractive little bird when, well grown, it comes to the doorway and peers about, watching for its parents to come through the air with food. As it waits at the entrance its low forehead and immaculate throat call to mind a little frog sitting there in the box. Its eyes shine eagerly, and when the parents come near it stretches out toward them, its throat gleaming white against the dark interior.

A. Dawes DuBois says: "The young are strong of wing when they leave the nest." He speaks of one young bird which "took to the air like a veteran, both parents accompanying it." Austin and Low (1932) state that "usually they showed remarkable ability on their first flight, often remaining in the air well over a minute, and flying a quarter of a mile." George Nelson tells me that he has often watched the young birds leave the boxes in his garden. They launch out, then fall, fluttering, nearly to the ground sometimes, when, of a sudden, the power of flight comes to them, and they rise into the air and fly off, seemingly as ably as their parents.

Plumages.--[AUTHOR'S NOTE: The juvenal plumage of the tree swallow is quite unlike the plumage of either adult. The upper parts, including the sides of the head and neck, are dark sooty brown or very dark brownish gray, "dark mouse gray", without any trace of the iridescent bluish green of the adult plumage but with a fine silky gloss. The feathers of the interscapular region are at first faintly edged with pale fawn color, but these edgings soon wear away. The wings and tail are slaty brown, with slight greenish reflections, the secondaries and tertials with faint grayish edges and tips. The under parts are duller white, less silky, than in the adult; and there is a very faint and incomplete pectoral band of ashy brown.

A complete postjuvenal molt takes place, beginning late in August and continuing into October. This produces a first winter plumage, which is practically indistinguishable from the winter plumage of the adult. Dr. Dwight (1900) says of this plumage: "Above iridescent green, sometimes with steely blue reflections. Wings and tail deep bottle-green slightly iridescent, the tertiaries broadly tipped with white. Below, pure white slightly smoky gray on the sides."

The white tips on the tertials are characteristic of the winter plumages of both adults and young birds; but these tips wear away before spring. There appears to be no spring molt, but a complete postnuptial molt begins about the middle of August and is usually completed before the birds go south. This is the only one of our swallows that completes its molt before migrating; it breeds early, molts early, and migrates late.

The sexes are much alike, except in the breeding plumage, when the female is duller, the upper parts often being largely dark grayish brown with only the tips of the feathers glossy greenish.]

Food.--F. E. L. Beal (1918) points out that "in its food habits this species differs somewhat from other American swallows in that it eats an appreciable quantity of vegetable food, frequently filling its stomach completely with berries or seeds." In an examination of 343 stomachs, collected in every month of the year over a wide range, Professor Beal found that "the food divided into 80.54 per cent animal matter to 19.46 per cent vegetable," and he states: "The vegetable food is made up of a few varieties of seeds and berries, but more than nine-tenths of it consists of the fruit of a single shrub, the bayberry, or waxberry (Myrica carolinensis)."

Of the animal food he says beetles make up 14.39 percent, ants 6.37 percent, and that Diptera form the largest item of the tree swallow's food (40.54 percent). Minor items are grasshoppers, dragonflies, and spiders. He summarizes his findings thus: "In the food of the tree, or white-bellied, swallow one point is prominent--in its vegetable food it has no relation to man. Every item is wild and of no use. In its insect diet it destroys some parasitic Hymenoptera, some carnivorous Diptera, and a few other useful insects, but this fault it has in common with most other insectivorous birds, and in common with them it is engaged in reducing the great flood of insect life to a lower level. Let it be protected and encouraged."

When we watch feeding tree swallows we see them chiefly in the role of flycatchers. They tour over meadows, ponds, and rivers, veering from side to side, doubling back with marvelous quickness, snatching up insects as they overtake them or meet them in the air, coursing low down over the meadow grass where flies abound, or, mounting, crisscross through the swarms of higher flying insects, gorging their throats with the tiny bodies. Ever on the move, they pass back and forth across their feeding grounds, their quick turns evincing success in capture after capture.

Arthur C. Bent speaks in his notes of seeing a flock of tree swallows alight on a marshy shore and feed from the ground on what appeared to be minute insects almost too small to be seen; also of their catching insects on the wing under the lea of a hillside.

John J. Elliott (1939) during his study of a company of tree swallows that wintered on Long Island, N. Y., collected their excrement. He reports that "analysis of the fecal material by the U. S. Biological Survey showed that crustacean material (Orchestia platensis) formed the bulk of the food, along with water boatmen, spiders, bulrush, sedge, bayberry and smartweed seeds, and fragments of rose thorns."

George Nelson had an interesting experience with tree swallows in Florida on November 24, 1938. He was crossing a hammock about 10 o'clock in the morning when he noticed, rising straight up from the path in front of him, a thin pillar of what he took to be a wisp of smoke no bigger round than a pencil. Following the column upward with his eyes, he noted that, at a height of 10 to 12 feet above ground, it swayed in the light breeze, broke up, and became dispersed. When he stepped up close to the wisp of smoke, he saw that it was not smoke at all, but a closely packed column of winged red ants issuing in countless numbers from a board walk. They mounted straight toward the zenith and then spread out like a funnel and scattered over the space of an acre. Soon a tree swallow flew past, up in the air where the insects had separated. It turned back and flew over and over again through the swarm, snapping up the insects. Before long more tree swallows appeared, until within 15 minutes many hundreds had collected, all swooping back and forth at great speed where the insects were flying. In less than half an hour the insects had disappeared and not a swallow was in sight.

Mr. Nelson was surprised to see so many tree swallows together, for at this season of the year the bird is not abundant in Sebastian where the incident occurred. They had assembled, apparently, in the way sea birds collect at points in the ocean where food is plentiful, bird after bird being drawn to the spot by seeing from a distance others feeding.

On another occasion, when Mayflies were rising from an extensive marsh, Mr. Nelson saw tree swallows gathered in a great cloud, so thick and dense that in the bright sunlight the flock cast a dark shadow on the marsh.

Behavior.--As we watch swallows in flight we notice that they do not all fly in the same fashion, and after long watching we become able to tell them apart when they are far away, or at least to suspect which is which. For example, perhaps the most distinctive in its manner of flying is the barn swallow. It is characteristic of this species to drive along through the air, seemingly with a strong push.

At the end of each stroke, the tips of the wings are brought backward until the primaries are nearly parallel with the long axis of the body. A robin also shows this peculiarity but to a less degree. The bird swings to the right and left, to be sure, but there are periods of straight flying or sailing, and always there is the impression of a steady drive through the air, with a good deal of power for so small a bird.

The tree swallow, compared with the barn swallow, appears to be less steady in the air, although doubtless it possesses complete mastery over it. There is a suggestion of flickering in its flight, due perhaps to the quicker, less forceful motions of its wings. Flying at a distance, it sometimes resembles a starling--another quick-moving bird--but most characteristic is the habit of hunching up its back, or seeming to do so, and lowering its wing tips as it sails, like an inverted saucer in the sky. Francis H. Allen speaks of their flight as "largely a succession of reaches and runs with periods between them when the bird seems to hang in stays for a while--to speak in nautical terms."

In the air the tree swallow resembles somewhat the purple martin, the similarity being due probably to the triangular shape of the wing in both birds--a triangle with a sharp apex and a fairly broad base. The bank swallow is readily distinguished from the tree swallow by its habit of hugging its wings close to the side of its body when it sails and by the suggestion of soft fluttering in the motion of its wings.

Tree swallows do not linger long about their nests once the young are on the wing. Both the young and the adult birds apparently retire to broader feeding grounds--the meadows bordering river valleys or marshes near the seacoast--where, gathering in increasing numbers, they form the nucleus of the autumn flocking.

Often because of a scarcity of nesting-holes or boxes, but sometimes because of a preference for a certain site, tree swallows come into conflict with other species of hole-nesting birds as well as individuals of their own species. F. Seymour Hersey (1933) relates a remarkable example of such an encounter. A pair of bluebirds were breeding in a box in his garden, and the female was incubating a set of eggs. He continues the story:

Then, one day, a pair of Tree Swallows arrived and decided they wanted that particular box. I hurriedly put up boxes No. 2 and No. 3 but the Swallows paid no attention to these new nests and after a day of constant bullying the Bluebirds surrendered their nest and eggs and retired to box No. 2. The Swallows remodeled the Bluebird's nest, incidentally disposing of the eggs in some way, and the Bluebirds started another nest in box No. 2. For awhile peace and quiet reigned and both pairs of birds had young a few days old when a second pair of Swallows put in an appearance. Once more there was fighting of a rather general nature among all three pairs of birds, but soon the Swallows in box No. 1 managed in some way to make good their title and the scene of battle centered about box No. 2. For two or three days this second pair of Swallows constantly harassed the Bluebirds, so that it was difficult for them to bring any food to their young, and at the end of this time they abandoned their nest and left the garden and vicinity.

The following year the swallows had a pair of starlings to deal with. Mr. Hersey says: "Their method of getting rid of the Starlings was interesting. While either of the Swallows was away from the nest the other was on guard, perched on the roof of the box. When the Starling appeared she was either attacked and driven away, or the Swallow immediately entered the box and sat looking out the hole, effectively blocking the entrance. For several days I did not see the nest left unguarded for a moment and the Starlings soon went elsewhere."

The swallows are not always able to oust a bluebird; a bluebird may even drive them away from the vicinity of its nest. Helen J. Robinson (1927) tells of such a case. She says: "The Bluebirds kept a vigilant lookout from their own tree, a young oak, and watched the Swallows circle about the other boxes. When the Swallows seemed about to enter, one or both Bluebirds charged them, straight as an arrow. The mere sight of the enemy was usually enough to put the Swallows to flight, circling and screaming as they retreated a short distance, but returning as soon as the coast seemed clear. Sometimes Swallows flew bravely to attack the Bluebird, but such birds were always borne to earth by the larger bird, which then tweaked the victim's crown feathers without mercy." The same author speaks also of the discord that arises among the swallows. She says: "The usual internal warfare among the Swallows themselves proceeded briskly, beginning the day of arrival and continuing a full calendar month."

There are several reports in the literature of strange behavior on the part of the tree swallow. Lewis O. Shelley (1934b) describes the actions of a female bird with a lust for killing the nestlings of other pairs of tree swallows; he (1936b) tells of another female that reared a nestling cliff swallow, which he introduced into her nest just as her own brood was about to fly, and J. A. Munro (1929) gives an account of a male that fed the 8-day-old nestlings of a pair of western robins whose nest was built on the top of the swallow's bird-house.

Voice.--The voice of the tree swallow has a pleasing, gentle quality. Silip, he seems to say, a quick, rapidly pronounced note, sometimes rippled into three or more syllables. It may run into a chatter, but it is never jarring; it always retains its delicacy. A little excitement brings more emphasis to the voice and introduces a long doubled e, so that the note suggests our word, cheery. Real alarm calls out greater power; the tone now rises to a squeal, almost a shrill whistle.

Francis H. Allen (1913) speaks of the tree swallow as "one of the very earliest singers in the morning concert." Of the song he says: "It is really a remarkable performance regarded as an exhibition of endurance." He describes thus the song of a bird singing at 2:53 a.m.: "He sang continuously, apparently without interruption, from the time I first heard him till 3:40. The song came and went, as the Swallow flew about over the pond, now nearer, now farther away, now to the right, now to the left, but never stopping,: a constant tsip-prrup, tsip-prrup-prrup, tsip-prrup, tsip-prrup-prrup-prrup, tsip-prrup-prrup, tsip-prrup-pruup-prrup-pruup, varied only by the varying number of bubbling notes following each tsip. The ending of the performance seemed to come gradually."

Winton Weydemeyer (1934a) says: "Singing is done both in flight and from perches near the nests. A series of phrases, repeated over and over in slightly varying order, at the rate of 125 to 140 a minute, is given for several minutes or as much as an hour without pause."

Field marks.--The under parts of the tree swallow are pure white. It is the only swallow that shows this character except the violet green, which may be distinguished by an opalescent coloring of the back. Jonathan Dwight (1900) speaks of "a very faint incomplete sooty collar on the jugulum" in the juvenal plumage of the tree swallow, but this mark is lost by the time the first winter plumage is acquired in October and should never be confused with the broad pectoral band that the bank swallow possesses in all plumages.

John Treadwell Nichols (1920) points out two excellent diagnostic points. He says: "When one gets a good view of them, our different Swallows are well marked and easy to identify. They also present differences in size, flight and call-notes which one learns to recognize. However, it may aid in the determination of a bird darting by at a difficult angle, to call attention to the white on the Tree Swallows' flanks, which encroaches on the dark upper parts in front of the tail so as to be conspicuous. The Tree Swallow also has an angle in the posterior outline of the wing unlike the other species, as though the primary feathers projected more abruptly beyond the secondaries."

Enemies.--The speed and agility of the tree swallow render it comparatively safe from attacks by birds of prey. Tree swallows come into competition for nesting sites with other hole-nesting birds, but as shown under "behavior," often hold their own.

A nest sent to the Bureau of Entomology, Washington, D. C., by Mrs. Kenneth B. Wetherbee (1932) "contained fifty-two Protocalliphora splendida, var. sialia, thirty Mormoniella vitripennis, and one hundred and sixty fleas."

The chief danger in the life of the bird is the inclement weather that it may be subjected to during the winter in the Southern States or that it may meet after its arrival on its breeding grounds, as noted under "Spring."

Fall.--Fall is a season of drama in the tree swallows' yearly cycle. A single idea, or an urge, seems to grip every swallow in the land. The nesting season with its quarrels over, the swallows draw together with a common interest in preparation for their next step, the long migration they will take in companies of hundreds or thousands. In August and September we see them gathering in the great marshes by the sea, where they linger for many days in ever-increasing numbers, young and old, sometimes associated with other species of swallows, notably the barn swallow.

The following quotations show the tree swallows gathering in autumn: John Lewis Childs (1900) gives an idea of the great number that collected at Barnegat Bay, N. J., in September. He describes the birds as he saw them flying overhead between their feeding grounds and their night quarters. He says: "Not a Swallow was seen until the solid column of the flight appeared, and it was at once apparent that where there were hundreds two weeks previous there were now thousands. The flight was compact like a swarm of bees and at times almost darkened the sky. Most of the time there were two distinct columns, one flying low just over the water, and the other high up in the air. I watched the flight for hours, and the air in both directions seemed alive with them as far as the eye could reach." Of his observation on the following day, he adds: "After watching the birds nearly all of the forenoon we made a careful estimate of the number that had passed and we calculated that it was not to be reckoned by tens of thousands or hundreds of thousands, but by millions."

Lynds Jones (1910), speaking of tree swallows at Cedar Point on the shore of Lake Erie, in Ohio, says: "After the breeding season, during late July, I have seen great companies gathering to roost in the swamp vegetation east of the mouth of Black Channel. They formed the characteristic funnel group before finally settling into the vegetation for the night."

Bradford Torrey (1893) describes tree swallows as seen at Ipswich, Mass., in fall. "At eight o'clock," he says, "when we took the straggling road out of the hills, a good many--there might be a thousand, I guessed--sat upon the fence wires, as if resting. We walked inland, and on our return, at noon, found, as my notes of the day express it, 'an innumerable host, thousands upon thousands,' about the landward side of the dunes. Fences and haycocks were covered. Multitudes were on the ground-- in the bed of the road, about the bare spots in the marsh, and on the gray faces of the hills. Other multitudes were in the bushes and low trees, literally loading them. Every few minutes a detachment would rise into the air like a cloud, and anon settle down again."

We often see tree swallows using the sea beach as a path of migration, flying either over the sand itself or over the ocean a short distance from the shore. As they course along they sometimes swoop playfully at a shorebird, put it to flight, and chase it, twisting and dodging, rising and darting down in unison with it, at the same time continuing their southerly course.

Years ago Walter Faxon and I watched about 500 tree swallows circling around the steeple of a church at Ipswich, Mass. It was in the latter part of an afternoon in mid-September at the height of the swallow migration. Suddenly their haphazard flight changed to an orderly procession in which about half the birds wheeled in a great spiral and, mounting high in the air, sailed away due south.

Winter.--Alfred M. Bailey (1928) says of the tree swallow on the gulf coast of Louisiana in the winter: "Very common over the marsh, where thousands were seen at once."

John J. Elliott (1939), who watched a little company of 28 tree swallows through the fall and winter of 1937 - 38 on Long Island, N. Y., found that the majority of the birds survived, although the locality is far to the north of the normal winter range. He says:

I decided to give this flock, which to all appearances was going to winter, as much of my time as possible, and to learn of their winter habits and peculiarities. To this end I made 78 trips and spent 131 1/2 hours. The frequent trips had the advantage of permitting me to observe the actions of the birds in many types of weather. I found cold rains, windy and snowy weather, extremely discouraging for flying birds, and after hours of watching, saw not a bird in the air. During these periods they remained in their sheltered situations, chiefly north of the pond, subsisting principally on bayberries. Cloudy days, especially if cold, also discouraged them to a certain extent from taking the air. In fact, I found that, generally, the brighter the sun, the more I saw of flying birds, and the warmer the sun and surrounding air, the higher their flight, both proportionately, except during windy days. At no time did they use telephone wires or any high exposed perch, as in summer; no long flights high in the upper air. . . .

They are given to much resting in sunny sheltered places, in little huddled groups of four or five birds each, and resort to low broken stubs of bayberry bushes, strong blackberry canes, squatting on the sand, or on boards imbedded in the sand. They are usually silent but on occasion utter a cheery, reedy, double-noted twitter when sunning themselves and in moderate weather.

[AUTHOR'S NOTE: Since the above was written, accounts have been published of the great mortality among tree swallows and other birds during the "big freeze" in southern Florida during January 1940. Bayard H. Christy (1940) writes:

On the morning of January 28, 1940, after ten days of north wind and continued cold, the temperature at Coconut Grove, Florida, fell below the freezing point. There was widespread destruction of cultivated plants, and small fishes in countless numbers lay dead on the tidal flats. . . . .  As in the preceding days the cold grew more intense, these birds were seen to hover more closely above the water surfaces, and to leeward of the walls of houses. On the morning of January 28, about fifty of them were found, densely packed in the cavity of a Pileated Woodpecker, formed long since in the stub of a palmetto, standing among mangroves a few yards from the margin of the bay. Twenty-eight were already dead or died soon after removal. A few flew away at once; others revived in the sunshine. On a sea-wall near by, about fifty more were resting in the sun, clustered like bees, some on the level top of the wall, others on its rough and sheltered face. Later in the day a half-dozen more were picked up dead on the lawn of the adjacent property.

A few of the bodies of the dead swallows found in the woodpecker cavity were opened and their stomachs were found to be empty. "No doubt there had been considerable shortage of food; no doubt the massing within the cavity had increased the destruction, but the fundamental cause of death was the cold."

The next morning he found only about a dozen swallows in the cavity, three of which were dead. Driving across the Tamiami Trail he found further evidence of disaster to the swallows, many of which had been run over by passing cars. "Even on wing the living birds seemed to have lost their usual agility, and twice, to my regret, my moving car struck birds in the air. . . . I found them thronged in an outhouse on the brink of the canal--one dead, two or three others fluttering to lie widespread upon the ground. At another place the graveled parking area was strewn with bodies--perhaps fifty of them."]