***
Spring.--Crow blackbirds, as they are often called, start migrating northward from their not far distant winter range during the latter part of February and reach their breeding grounds in southern New England around the middle of March. St. Patrick's Day, March 17, has always been associated in my mind with the arrival of the grackles about my home; then we may expect to hear the creaking notes of the males and see the glossy black birds posturing in the leafless treetops or exploring the tops of the tallest pines and spruces for possible nesting sites, preparatory for the coming of the females a week or two later. If weather conditions are favorable, they may remain, but a late snow storm or severe cold spell may cause them to retreat.

Courtship.--On April 8, 1946, two grackles, apparently both males, were moving about in the branches of a big ash tree close to my study window. One was evidently following the other as he traveled along the branches or hopped from one branch to another. Every few seconds one would stop, crouch down on the branch, lower his head, puff out his body plumage, spread his wings downward, and lower and spread his tail, at the same time giving voice to his unmusical notes. The other male went through the same motions at intervals, alternating with the first one. Eventually they separated and flew away in different directions. Apparently, it was a competitive display for the benefit of some hidden female, of which there were several in the yard.

Mating is evidently earlier at Cape May, N.J., for Witmer Stone (1937) writes:

As early as March 13, many of the Grackles are flying in pairs, the male just behind the female and at a slightly lower level. They are noisy, too, about the nest trees and there is a constant chorus of harsh alarm calls, 'chuck, chuck, chuck,' like the sound produced by drawing the side of the tongue away from the teeth, interspersed with an occasional long drawn 'seeek,' these calls being uttered by birds on the wing as well as those that are perching. Then at intervals from a perching male comes the explosive rasping "song" 'chu-seeeek' accompanied by the characteristic lifting of the shoulders, spreading of the wings and tail, and swelling up of the entire plumage.

As early as March 5 I have seen evidence of mating and sometimes two males have been in pursuit of a single female, resting near her in the tree tops, where they adopted a curious posture with neck stretched up and bill held vertically.

Nesting.--At the extreme northeastern end of their breeding range, near my home, we have found purple grackles nesting in a variety of situations. Many years ago, in eastern Rhode Island, a colony of a dozen or more pairs nested for several years in a hillside grove of red cedars (Juniperus virginiana). The nests were placed in the cedars, 10 or 12 feet from the ground, and were made of dried grasses and weed stems, lined with fine dry grass. In the extensive cattail marshes surrounding Squibnocket Pond on Martha's Vineyard Island, we found two well-hidden grackles' nests in the tall, dense green flags, firmly attached to these cattails, and placed from 2 to 3 feet above the water. In that same vicinity there was a colony of eight or ten nests of these birds, 7 or 8 feet up, in a swampy thicket of large bushes.

On May 29, 1904, at Chatham, Mass., while passing through an apple orchard in full bloom, we noticed a pair of grackles making quite a fuss; their nest was soon located in an upright crotch near the top of one of the apple trees, about 12 feet from the ground; the nest, made of seaweed and coarse grasses and lined with fine grass and horsehair, contained five fresh eggs.

By contrast, our local purple grackles sometimes select much more inaccessible nesting sites. Within sight of my former residence is a row of tall white pines (Pinus strobus), along the banks of the Taunton River; every year several pairs of grackles have nested near the tops of these trees, where the nests must have been between 50 and 60 feet from the ground; the nests were never disturbed by egg-collecting boys. We found another safe nesting site in a cedar swamp on Cape Cod. The swamp had been flooded as a reservoir and the white cedars (Chamaecyparis thyoides) were standing in water from 4 to 5 feet deep; it was a very large colony and there were evidently many nests in the cedars, but we did not care to make any accurate count of the nests, nor could we even estimate the number of the birds that were flying about over the swamp.

Bendire (1895) gives the following description of the nests: "The nests are rather loosely constructed and bulky. The materials used vary greatly according to locality; the outer walls are usually composed of coarse grass, weed stalks, eelgrass or seaweed, sometimes with a foundation of mud, and again without it. The inner cup of the nest is composed of similar but finer materials, and is generally lined with dry grass, among which occasionally a few feathers, bits of paper, strings, and rags may be scattered; in fact anything suitable and readily obtained is liable to be utilized. Exteriorly the nests vary from 5 to 8 inches in height, and from 7 to 9 inches in diameter, according to location. They are ordinarily about 3 inches deep by 4 inches wide inside." After describing nesting sites, similar to those mentioned above, he adds:

Sometimes natural cavities in trees or hollow stubs, as well as the excavations of the larger Woodpeckers, are also used, and along the seashore, where the Fishhawk is common, they often place their nests in the interstices of these bulky structures, notably so on Plum Island, New York. Speaking of this locality, the late Dr. Charles S. Allen [1892] says: "In every Fishhawk's nest, except those on the ground, I always found from two to eight or ten nests of the Purple Grackle. They were situated in crevices among the sticks under the edges of the nest, or even beneath the nest itself, so as to secure protection from rain and bad weather. They were very bold in collecting fragments from the table of their powerful neighbors."

Mr. J. H. Pleasant, Jr., of Baltimore, Maryland, writes as follows: "On May 19, 1888, I discovered a colony of Purple Grackles nesting under the eaves and rafters of a hay barn. In some instances the entrance to the nest was so small that it was extremely difficult to obtain the eggs. The crevices in which the nests were built were very much of the same character as those frequently chosen by the English Sparrow, and were situated at an average height of 25 feet from the ground; over a dozen nests were observed."

T. E. McMullen has sent me the data for 20 New Jersey nests; 9 of these were in grapevines or ivy vines climbing over various deciduous trees; 9 others were in red cedars; one was 20 feet up in a gum tree, the highest one was 45 feet from the ground in a large pine, and the lowest nests were 6 or 8 feet up in vines.

Eggs.--The purple grackle lays ordinarily four of five eggs to a set, very rarely seven; sets of six are not especially rare; the only set of seven that I have found contained two eggs that were quite different from the other five. The eggs are generally ovate in shape and are slightly glossy. Bendire (1895) describes them as follows:

The ground color of the Purple Grackle's eggs varies from a pale greenish white to a light rusty brown; they are generally blotched or streaked with irregular lines and dashes of various shades of dark brown, and in an occasional set different tints of lavender markings are also noticeable. Only in rare instances are these markings so profuse and evenly distributed over the entire egg as to hide the ground color. They vary greatly in style and character in different sets.

The average measurement of 85 eggs is 28.53 by 20.89 millimeters, or about 1.12 by 0.82 inches. The largest egg in the series measures 32.76 by 23.11 millimeters, or 1.29 by 0.91 inches; the smallest 25.65 by 20.57 millimeters, or 1.01 by 0.81 inches.

Young.--Of the young, Bendire (1895) says: "Incubation, in which both parents assist, lasts about two weeks, and they are equally solicitous in the defense of their eggs or young; the latter are able to leave the nest in about eighteen days, and sometimes a second brood is raised. They are fed almost entirely on insects while in the nest." Eighteen days seems a long time for the young to remain in the nest; 12 or 14 days would seem to be the usual time. It seems strange that so little has been published on the care and development of the young of such a common bird as the purple grackle.

Plumages.--The plumage changes of the purple grackle are very simple and hardly noticeable after the young bird's first summer. Dwight (1900) calls the color of the natal down pale sepia-brown. The whole juvenal plumage is "dull clove-brown, the body feathers often very faintly edged with paler brown. Tail darker with purplish tints." A complete postjuvenal molt takes place early in August, at which the iridescent black plumage of the male is acquired, and old and young birds become indistinguishable. The nuptial plumage is "acquired by wear which produces no noticeable effect as is regularly the case with iridescent plumages." Adults have one complete annual molt, the postnuptial, beginning early in August.

Of the plumages of the female, he says: "In juvenal dress the female is perhaps paler below than is the male and usually indistinctly streaked. There is a complete postjuvenal moult and later plumages differ from the male only in being much duller and browner with few metallic reflections. They also show more wear."

Witmer Stone (1937) makes the following interesting observation: "The progress of the molt in Grackles can easily be noted by the appearance of the wings and tail as the birds fly overhead, although the new and old body plumage of the adults are the same. They show gaps in the flight feathers as early as July 18 and some are still molting as late as September 8, 11, and 16 in different years. When the tail molt begins the long central feathers drop out first so that the tail appears split or forked; this gap becomes wider as successive pairs of feathers are lost, but by the time the outer pair is dropped the new central feathers have grown out and the outline of the tail is pointed or wedge-shaped."

Harold B. Wood has sent me the following notes on the colors of the iris in the purple grackle: "The young have brown irides, which by the absorption of the pigment, change to gray and lemon, ivory or white. The young of the year have uniformly dark brown irides until fall. Early spring birds have gray, lemon, ivory, or white irides. No bird which I trapped and banded with brown or gray eyes ever returned to the traps." As the iris in the adult is pale lemon color, or almost white, it appears that the brown iris is confined to the youngest birds and that the gray iris marks a transition stage of adolescence.

Food.-- *** In his report on the birds of Pennsylvania, B. H. Warren (1890) gives the following list of the contents of several series of stomachs collected in various months:

March--Twenty-nine examined. They showed chiefly insects and seed; in five corn was present, and in four wheat and oats were found. All of these grains, however, were in connection with an excess of insect food.

April--Thirty-three examined. They revealed chiefly insects, but a small amount of vegetable matter.

May--Eighty-two examined. Almost entirely insects, cut-worms being especially frequent.

June--Forty-three examined. Showed generally insects, cut-worms in abundance; fruits and berries present, but to very small extent.

July--Twenty-four examined. Showed mainly insects; berries present in limited amount.

August--Twenty-three examined. Showed chiefly insects, berries, and corn.

September--Eighteen examined. Showed insects, berries, corn and seeds.

October--During this month (1882), the writer made repeated visits to roosting resorts, where these birds were collected in great numbers, and shot 378, which were examined. Of this number the following is the result of examinations, in detail, of 111 stomachs:

Thirty, corn and coleoptera (beetles); twenty-seven, corn only; fifteen, orthoptera (grasshoppers); eleven, corn and seeds; eleven, corn and orthoptera; seven, coleoptera; three, coleoptera and orthoptera; three, wheat and coleoptera; two, wheat and corn; one, diptera (flies).

The remaining 267 birds were taken from the 10th to the 31st of the month, and their food was found to consist almost entirely of corn.

These examinations show that late in the fall, when insect food is scarce, corn is especially preyed upon by these birds, but during the previous periods of their residence with us, insects form a large portion of their diet.

Bendire (1895) makes the general statement that--

Their food consists largely of animal matter, such as grasshoppers, caterpillars, spiders, beetles, cutworms, larvae of different insects, remains of small mammals, frogs, newts, crawfish, small mollusks and fish. While it must be admitted that Indian corn, oats, and wheat are also eaten to some extent, much of the vegetable matter found in their stomachs consists of the seeds of noxious weeds, such as ragweed (Ambrosia), smartweed (Polygonum), and others. Fruit is used but sparingly, and consists usually of mulberries, blackberries, and occasionally cherries. One of the gravest charges against them is the destruction of the young and eggs of smaller birds, especially those of the Robin. . . .

They spend much of their time on the ground, being essentially ground feeders; they walk along close to the heels of the farmer while plowing, picking up beetles, grubs, etc., as they are turned up by the plow, or search the meadows and pastures for worms, grasshoppers, and other insects suitable for food.

The purple grackle eats the Japanese beetle, that imported pest that does so much damage to lawns, fruit trees, and flower gardens. I constantly see grackles and starlings feeding on my lawns, and like to think that they are probing for the grubs of this beetle; but I have never seen them feeding on the adult beetles in my rose garden. However, Japanese beetles were found in all the stomachs of purple grackles, meadowlarks, starlings, cardinals, English sparrows, wood thrushes, catbirds and robins that were taken in the heavily infested areas in New Jersey and eastern Pennsylvania. Smith and Hadley (1926) say: "The purple grackle accounts for more of the beetles than any other bird. . . The percentage of beetles eaten by the more important birds is as follows: purple grackle, 66.3; meadowlark, 50.7; starling, 42.3; cardinal, 38.6; catbird, 14.8."

About our city parks these grackles are scavengers, picking anything edible from the rubbish cans, or eating any crumbs or bits of food dropped from the lunch baskets of visitors. Frank R. Smith sends me a story illustrating the sagacity of the bird: "This morning, as I passed through the park back of the National Museum, I noticed a grackle that had found a dry, hard crust, left from a lunch. The bird made several attempts to eat the crust, but its hardness resisted his efforts. Picking it up, he flew across the walk and alighted near a hydrant, beneath which a bird-bath was sunk to the level of the ground. Soaking in the water sat a pigeon; and the grackle, while evidently wanting to enter, feared to trust his prize so near the larger bird. After several false starts, he waded boldly into the water and turned his back on the pigeon, so that his own body was between the bread and the bird he feared. He dropped the bread into the water, waited a few seconds, picked it up and walked out to the grass, where he ate the softened bread. During this time the pigeon sat watching him curiously."

Hervey Brackbill writes to me: "Acorns are a prominent fall food. Flocks as large as a couple of hundred birds come into the oak-wooded suburbs of Baltimore in late September and October, and feed both in the trees and on the ground beneath. The grackles, incidentally, do not open the acorns as blue jays do, by holding them down with their feet and hammering them with their bills; they grip them back in the angle of their mandibles and crack them by direct pressure."

Clarence Cottam (1943) observed an unusual feeding habit of grackles and crows at the outlet of a reservoir where--

About 12,000 cubic feet of water per second was passing through the electric turbines, "boiling up" to form the headwater of the Cooper River. Apparently the turbines were cutting up or otherwise killing large numbers of gizzard shad and other small fishes. These, brought to the surface by the churning water, attracted Ring-billed, Herring, Laughing, and Bonaparte's Gulls, as well as crows, Purple Grackles, and even a solitary Red-wing. . . . The grackles and crows fed over the turbulent water, picking up morsels of food with the skill and dexterity of the typical water birds. The feet and even the breast feathers of many of the crows and grackles were seen to touch the surface of the water momentarily as the birds hovered over this (for them) uncharacteristic feeding place. . . . Purple Grackles. . . use a wide variety of foods, and we have occasionally observed them feeding in shallow water on stranded insects and even small fishes. To see several dozens of these birds feeding in deep and turbulent water after the manner of gulls and terns, however, was indeed a surprise.

Economic Status.--The grackle's reputation among farmers is almost as black as its plumage, for its faults, and it has plenty, are more conspicuous than its good deeds. Nor is it any more popular among its bird neighbors, as can be seen by the hostility they show toward it, for many a robin's or other small bird's nest has been robbed of its eggs or callow young to satisfy the appetites of young grackles. Analysis of stomach contents does not show any large percentage of such food, but it must be remembered that the yolks of eggs and the soft parts of small young are quickly digested and thus not easily detected; and the egg shells are not always swallowed.

The grackles are condemned by farmers on account of the considerable damage done by them to the grain crops during the planting season and until after harvesting has been completed. They are accused of pulling up the sprouting corn and wheat in the spring, but much of this is done to obtain the cutworms that are attacking the seedlings. Warren (1890) says on this point: "Some four years ago I was visiting a friend who had thirty odd acres of corn (maize) planted. Quite a number of the 'blackies,' as he styled them, were plying themselves with great activity about the growing cereal. We shot thirty-one of these birds feeding in the cornfield. Of this number nineteen showed only cut worms in their stomachs. The number of cut worms in each, of course, varied, but as many as twenty-two were taken from one stomach. In seven some corn was found, in connection with a very large excess of insects, to wit: beetles, earthworms, and cut worms. The remaining five showed chiefly beetles."

Perhaps the chief damage to the corn crop is done when the grain is in the milky stage in the summer; the grackles are flocking at that season and, where they are abundant, they swoop down in great black clouds into the standing corn; they strip the husks off the ears and eat the tender kernels, taking perhaps only a few from each ear, but rendering many unfit for the market. Sometimes as much as a quarter of the crop is thus damaged. The farmer is nearly helpless to protect a large field, for shooting only drives the birds from one portion of the field to another. All that can be said in favor of the grackle here is that it is a persistent enemy of the destructive corn borer. Later in the season, after the corn is harvested and shocked, the grackles do some damage to the ripened ears by extracting the hard kernels; and Nuttall (1832) says that "in the southern states, in winter, they hover round the corn-cribs in swarms, and boldly peck the hard grain from the cob through the air openings in the magazine."

Referring to the attacks on sprouting winter wheat, Judd (1902) writes: "During November 1900, a flock of from 2,000 to 3,000 pulled wheat on the Bryan farm, and only continual use of the shotgun saved the crop. At each report they would fly to the oak woods bordering lot 5, where they fed on acorns. Nine birds collected had eaten acorns and wheat in about equal proportions. The flock must have taken at least half an ounce of food apiece, and therefore, if the specimens examined were representative, must in a week have made away with 217 pounds of sprouting wheat, a loss that would entail at harvest time a shortage of at least ten times as much."

Although grain forms nearly half (47 percent) of the food for the year it is not all a loss to the farmer, as much of it is waste grain dropped during harvesting or left on the ground after that. Some slight damage is done to green peas, cherries, strawberries, blackberries, and other small fruits, but less than is done by some other birds.

All this damage may seem considerable, but it is largely offset by the good done in the destruction of those insects, harmful to the interests of the farmer, which make up over 50 percent of the food for the year. Consequently, where grackles are overabundant, they should be controlled or the crops be protected; otherwise they are fully as useful as harmful.

Behavior.--While feeding on my lawn the grackle walks with a slow, dignified gait, head held high and tail somewhat elevated, or runs nimbly over the ground, nervously flitting its long tail up and down and occasionally making long, high hops in pursuit of some insect. Occasionally it jumps or flies up a foot or two to catch a flying insect in the air. It forages also in the shrubbery of trees, evidently after insects, but for the most part finds its food on the ground, picking something off the grass or probing in the earth for grubs or worms. When robins are feeding on the lawn at the same time, the grackles watch them and follow them about; as soon as a robin is seen pulling up a fat worm, the grackle rushes in and seizes the worm, driving away the gentler bird; the robin seems to be unable to defend itself and must yield its prize to the more aggressive robber. I have often seen a grackle, while foraging on my lawn on a warm sunny day in spring, stop and squat close down on the ground, remaining there for several minutes with its body pressed close to the warm earth, as if it enjoyed the warmth, or perhaps just taking a sunbath. It may have been "anting," as other birds do in order to anoint their plumage with formic acid.

In this connection, the following observation by Mary Emma Groff and Hervey Brackbill (1946) is of interest:

The recent discussions of anting and supposedly substitute activities by birds makes it seem worth while to describe the behavior of Purple Grackles in anointing themselves with a juice, apparently an acid, from the hulls of English walnuts (Juglans regia). . . . The walnuts grow in clusters of as many as five or six, at the ends of branches. The grackles would alight upon these clusters--just one bird to each--and begin pecking a hole in the sticky hull of one of the nuts, usually throwing away the pieces of hull they gouged out, occasionally seeming to swallow a piece. When a good-sized hole had been made, the birds would dip their bills into it, undoubtedly wetting them against the pulpy interior, and then thrust their bills over and into their plumage. A great part of the body was thus anointed--the breast, the under and upper surfaces of the wings, the back, and very often apparently the rump at the base of the tail. . . . Particularly birds that were watched worked as long as 10 to 15 minutes at a stretch. Many males sang at intervals, with display, and there was also much noise because of commotion among the birds, two or three of which would often contest for the same cluster of nuts. . . . The indication that it was an acid the birds were using was obtained when one of the English walnut hulls was cut open and litmus paper quickly placed against it; the paper instantly gave a strong acid reaction.

In the air the purple grackle flies in a direct line, not undulating like redwings, and generally at a considerable height, with strong steady beats; its flight is well sustained but not especially rapid. Witmer Stone (1937) says that when they descend from a height to alight in the trees "they sail down on set wings which form a triangular, kite-like outline, with the long tails of the males deeply depressed into the characteristic boat or keel." As fly-catchers the grackles are not experts. Stone saw one "pursuing a flying beetle on the street, an unusual performance; the bird was exceedingly clumsy in turning on the wing and after following its erratic prey for several minutes without result it gave up the chase. On August 31, several Grackles were observed darting up into the air from the tree tops in pursuit of flying ants in which activity they also proved very clumsy."

In its relations with other species the grackle not only indulges in the well-known habit of stealing eggs or young birds from the nests of its neighbors, but sometimes attacks and kills other birds in open places. In the National Zoological Park, in Washington, Malcolm Davis (1944) saw a purple grackle kill an English sparrow, which it "had been stalking in almost catlike manner. . . . The grackle approached the sparrow and as the smaller bird flew away, the attacker seized its prey in its beak and gave it several hard shakes, with the body of the sparrow hitting the hard concrete pavement. At this moment passersby frightened the grackle away, but later the bird returned to eat the viscera of the sparrow."

Frank B. Foster (1927) reports: "At my Game Farm on the Pickering Creek, in Chester County, Pa., we lost in the Pheasant field, almost three hundred little Pheasants (Phasianus), a few days old, which were destroyed by Purple Grackles. The male Grackles were the ones that did the damage. They came into the enclosure and simply took the heads off the little birds, leaving the bodies."

The purple grackle is highly gregarious at all seasons; even during the nesting season the birds often breed in sizable communities; and those that are not incubating resort to communal roosts at night. In the larger roosts they are often associated with starlings, redwings, or cowbirds.

Several roosts in eastern Pennsylvania have been studied, of which the Overbrook roost, described by C. J. Peck (1905), is typical: "The Overbrook Grackle Roost is situated upon the property of Mr. David L. Hess at the corner of Sixty-third street and Landsdowne Avenue, Philadelphia. The estate comprises about ten acres, is rolling and wooded and has an artificial lake of about an acre in extent. The trees are deciduous with a goodly sprinkling of conifers and are of fair size. The roost has been in constant use for more than twenty years--how much more I have been unable to ascertain." This roost was used by varying numbers of birds during every month in the year, the smallest numbers being found in December and January. He gives a short account month by month showing the fluctuations in the population of the roost. In January, fewer birds use the roost than at any other time of the year. "On a few very severe nights the roost may be deserted, but such nights are rare and usually four or five hundred birds remain throughout the month." Conditions are about the same until the last week in February, when the migration begins. "Probably five thousand birds use the roost during the last few days in February." In March the "number of birds rapidly increases throughout the month until from twenty to twenty-five thousand are using the roost nightly." In April and May, the nesting months, the numbers fall off, "but the number never seems to fall below two or three thousand--birds which have not mated as yet or else males which have nests near by, probably both." June is very much like May, except that a very few females and the first of the early young begin to come in. But all this is changed after August first.

The birds have for the most part completed their domestic cares and family groups are rapidly consolidated into large flocks which come to the roost from considerable distances. The numbers are very greatly increased and the birds in flying to and from the roost follow much more closely a regular well-defined route.

During September and October the greatest numbers are reached and the birds come in at night in great flights, one flock following another so closely as to give the impression of a single long-drawn-out flock. The flight begins about 5:30 p.m. and lasts for about twenty or twenty-five minutes, but scattered birds and small flocks continue to come in until dark. I believe that from fifty to seventy-five thousand birds visit the roost every night during these two months. . . . Robins use the roost to the number of one thousand or more, their numbers being hard to judge with any degree of accuracy on account of the way they mix with the Grackles.

By 6:30, on September 17, the noise from the birds had begun to subside; and by 6:45 darkness and silence had come.

When grackles and starlings select a roost in a thickly settled community, or in the trees of a city street, as they sometimes do, they create a decided nuisance. Lewis W. Ripley (1914) tells how such a roost was established in one of the finest residential streets in Hartford, Conn., and what was done about it: "The birds, numbering probably several thousand, began to come in just before dark, and by seven o'clock all had arrived, and from this time until about six in the morning constituted a first-class nuisance, whistling and chattering until about 8 p.m., and beginning about 4 a.m., making a tremendous racket so that it was difficult to sleep. Not less annoying was the filthy condition of the walks and lawns, and the damage to the clothing of those passing along the street was not inconsiderable."

Several plans were discussed for getting rid of them and some were tried without much success; ordinary roman candles had no permanent effect, even when fired by men in the trees; but finally it was learned that the persistent use of high-powered, 10-ball candles, weighing 56 pounds to the gross, would produce the desired result. "As a net final result, about eight dozen candles were used at a total expense of about $10 and, at the end of a week, only a couple of dozen birds are to be found where there were thousands."

Voice.--The unattractive voice of the purple grackle is described in the following notes sent to me by Aretas A. Saunders: "While the sounds produced by grackles are far from musical, nevertheless some of them are largely confined to a definite season, including the period of nesting, and therefore may be considered to be songs. The commonest of these is a grating, metallic sound that might be written kuwaaaa. The main note is pitched about F'', and a short note at the beginning is a tone to a tone and a half lower. The matter of pitch, however, is more difficult to determine definitely in sounds that are not of musical quality. This is particularly true in determining the octave. The pitch of this note is near F, but whether F', F'', or F''' I do not feel entirely sure. This particular sound is to be heard from the first arrival of the birds in March to the end of the breeding season in late June. It is sometimes also heard in late September and October from individuals in the flocks that congregate at that season.

"In the time of courtship in late April or early May, grackles produce another songlike sound that is accompanied by spreading of wings and tail. This is a series of four or five notes, each higher in pitch than the former one. The lower notes are rather harsh, while the higher ones are squeaky. These sounds are something like kogubaleek or koochokaweekee. The pitch begins on C'' or D'' and rises to B flat'' or C''' at the end. The common call-note of the grackle is a loud chak, very similar to that of the redwings, but louder and somewhat lower in pitch."

To the nonmusical ear the squeaky notes of the grackles sound like the creaking of a rusty hinge and are decidedly unpleasant, but when heard in chorus from a migrating flock the effect is rather pleasing. During the courtship display the contortions of body, wings, and tail seem to indicate that the notes are produced with considerable effort.

Field marks.--The grackles are the largest of our northern blackbirds and have the longest tails; these are wedge-shaped and rounded or graduated at the end; and the male often carries his tail keeled, the middle feathers lower than the others. Grackles differ from redwings in having a straighter, more level, less undulating flight. They can be distinguished from rusty blackbirds by the longer tails. ***

Fall.--The migrations of purple grackles are not long ones. They leave the northern portions of their breeding range in November, but even here a few remain occasionally in mild winters, though they are rare north of Washington, D.C., in winter.

As soon as the breeding season is over and the young birds are well grown, they begin to gather in the summer roosts, the family parties joining to form immense flocks. During October and November, these great flocks wander about over the country, often joined by starlings, cowbirds, and other blackbirds, seeking suitable feeding places in the grain fields, grasslands, and swamps. Stone (1937) describes one of these large feeding flocks "which contained many thousand birds. They covered the ground in great black sheets, the rear ranks constantly arising and flying over to take their place in the van which gave the impression of rolling over the ground. When they took wing in force the long procession streamed past shutting off from view all that lay beyond and when they alighted in the trees the bare branches appeared to be clothed with a dense black foliage."

Winter.--The main winter range of the purple grackle seems to extend from the Carolinas southward to the Gulf coast, though Skinner (1928) says that it occurs mainly as a migrant in the sandhill region of North Carolina, and Wayne (1910) considers it rare in coastal South Carolina. Probably most of these grackles spend the winter farther south in the Gulf states.

Wilson (1832) gives the following graphic account of a large wintering flock:

A few miles from the banks of the Roanoke, on the 20th of January, I met with one of these prodigious armies of Grackles. They rose from the surrounding fields with a noise like thunder, and, descending on the length of road before me, covered it and the fences completely with black; and when they again rose, and, after a few evolutions, descended on the skirts of the high timbered woods, at that time destitute of leaves, they produced a most singular and striking effect; the whole trees for a considerable extent, from the top to the lowest branches, seeming as if hung in mourning; their notes and screaming the meanwhile resembling the distant sound of a great cataract, but in more musical cadence, swelling and dying away on the ear, according to the fluctuation of the breeze.