Although it was thought for a long time that migratory flights were made at normal cruising speeds, Harrisson (1931) and Meinertzhagen (1955) showed that migration speeds were in between cruising speeds and escape speeds. The theory that migrating birds attain high speeds received encouragement from the German ornithologist Gatke (1895) who, for many years, observed birds at the island of Heligoland. He postulated that the bluethroat, a species of thrush smaller than the American hermit thrush, could leave African winter quarters at dusk and reach Heligoland at dawn; this flight would mean a sustained speed of 200 miles per hour! He also thought the American golden plover flew from the coast of Labrador to Brazil in 15 hours at the tremendous speed of 250 miles per hour. Most ornithologists now consider these conclusions to be unwarranted.
Reliable data on the speed of birds are accumulating slowly. Accurate measurements are difficult to obtain unless the bird travels over a measured course and wind conditions at the level of flight are known. Several subtle factors, besides wind and pursuit, can influence the speed of a flying bird. For instance, species that have a courtship flight often reach their maximum speeds then. Small woodland birds often fly faster across an open area where they might be attacked by a bird of prey than under cover where there is less danger. Birds in flocks generally fly faster than when flying alone. A thermal draft may induce an almost imperceptible air movement at the Earth's surface, but a good glider with motionless wings may make 35 miles per hour on a current of air that is rising vertically at less than 2 miles per hour. If the bird coasts downhill at a slight angle in still air, it can attain a similar speed.
For sustained flight, it may be generally concluded that larger birds fly faster than smaller birds. A common flying speed of ducks and geese is between 40 and 50 miles per hour, but among the smaller birds it is much less. Herons, hawks, horned larks, ravens, and shrikes, timed with the speedometer of an automobile, have been found to fly 22 to 28 miles per hour, whereas some of the flycatchers fly at only 10 to 17 miles per hour. Even such fast-flying birds as the mourning dove rarely exceed 35 miles per hour. A peregrine falcon will have difficulty catching a pigeon during a level chase at 60 miles per hour, but this predator can probably exceed 200 miles per hour during a swoop from a greater height onto its prey.
The speed of migration is quite different from that attained in forced flights for short distances. A sustained flight of 10 hours per day would carry herons, hawks, crows, and smaller birds from 100 to 250 miles, while ducks and geese might travel as much as 400 to 500 miles in the same period (without the aid of a tail wind). Measured as straight line distances, these journeys are impressive and indicate birds could travel from the northern United States or even from northern Canada to winter quarters in the West Indies, Central, or South America in a relatively short time. It is probable that individual birds do make flights of the length indicated and that barn swallows seen in May on Beata Island, off the southern coast of the Dominican Republic, may have reached that point after a nonstop flight of 350 miles across the Caribbean Sea from the coast of Venezuela.
Radar has given us some of our best estimates of ground speeds for migrating flocks, especially at night. Radar echoes, identified as shorebirds migrating off the New England coast, moved steadily about 45 miles per hour for several hours; songbird echoes typically traveled around 30 miles per hour (Drury 1960). Some birds appear to reduce flight speed in proportion to the degree of assistance or resistance. The literature is in some disagreement on the flight speed of birds and the influence of wind, but good radar observations coupled with accurate measurements of winds aloft will help give us a more accurate estimate of migrating speeds for different species under varying wind conditions.
The intensity of migration depends on circumstances including the need for haste. In fall the flights are more likely to be performed in a leisurely manner, so that after a flight of a few hours the birds often pause to feed and rest for one or several days, particularly if they find themselves in congenial surroundings. Some indication of this is found in the recoveries of banded birds, particularly waterfowl. If we consider only the shortest intervals between banding in the North and subsequent recovery in the South, it is found that usually a month or more is taken to cover straight-line distance of a thousand miles. For example, a black duck banded at Lake Scugog, Ontario, was killed 12 days later at Vicksburg, Mississippi. If the bird was taken shortly after its arrival, the record would indicate an average daily flight of 83 miles, a distance that could have been covered in about 2 hours' flying time. Among the thousands of banding records of ducks and geese, evidences of rapid migrations are decidedly scarce, for with few exceptions, all thousand-mile flights have required 2 to 4 weeks or more. Among sportsmen, the blue-winged teal is well known as a fast-flying duck and quite a few of these banded on Canadian breeding grounds have covered 2,300 to 3,000 miles in a 30-day period. Nevertheless, the majority of those that have traveled to South America were not recovered in that region until 2 or 3 months after they were banded. Probably the fastest flight over a long distance for one of these little ducks was one made by a young male that traveled 3,800 miles from the delta of the Athabaska River, northern Alberta, Canada, to Maracaibo, Venezuela, in exactly 1 month. This flight was at an average speed of 125 miles per day. A very rapid migration speed was maintained by a lesser yellowlegs banded at North Eastham, Cape Cod, Massachusetts, on 28 August 1935 and killed 6 days later, 1,900 miles away, at Lamentin, Martinique, French West Indies. This bird traveled an average daily distance of more than 316 miles.
It seems probable that most migratory journeys are performed at little more than the normal, unforced rate of flight, as this would best conserve the strength of the birds. Migrating birds passing lightships and lighthouses or crossing the face of the moon have been observed to fly without hurry or evidence of straining to attain high speed. The speed or rate of migration would therefore depend chiefly on the duration of flights and tail wind velocity.
The speed of migration is demonstrated by the dates of arrival, particularly during the spring movement. The Canada goose affords a typical example of regular but slow migration. Its advance northward is at the same rate as the advance of the season ([Fig. 5.]). In fact, the isotherm of 35° F appears to be a governing factor in the speed at which these geese move north. (An isotherm is a line that connects points that have the same temperature at the same time.) From an evolutionary viewpoint we might expect this. If the geese continually advanced ahead of the 32° F isotherm, they would always find food and water frozen and unavailable. By migrating north just behind the advance of this isotherm, birds that breed in the far north will find food and open water available and have as long a breeding season as the climate will allow.
Few species perform such leisurely migrations; many wait in their winter homes until spring is well advanced, then move rapidly to their breeding grounds. Sometimes this advance is so rapid, late migrants actually catch up with species that may have been pressing slowly but steadily northward for a month or more. The following several examples of well-known migrants illustrate this.
The grey-cheeked thrush, which winters in the Colombia-Ecuador-Peru-Venezuela-British Guiana area, does not start its northward journey until many other species are well on their way. It does not appear in the United States until the last of April—25 April near the mouth of the Mississippi and 30 April in northern Florida ([Fig. 6.]). A month later, or by the last week in May, the bird is seen in northwestern Alaska. Therefore, the 4,000-mile trip from Louisiana was made at an average distance of about 130 miles per day.