Correction: Figures 41 and 45 were inadvertently transposed.

[Ill.u.s.tration: FIG. 46. Comparison of flight trends and surface weather conditions on May 21-22, 1948. The meteorological data were taken from the U. S. Weather Bureau Daily Weather Map for 12:30 A. M. (CST) on May 22. The nightly station densities and the average hourly station density (shown in parentheses) are as follows:

2. Ottumwa: 6,900 (1,400) 5. Louisville: 1,500 (200) 9. Knoxville: 3,200 (500) 10. Memphis: 7,000 (1,200) 13. Oak Grove: 5,800 (800) 14. Mansfield: 2,500 (800) 18. Pensacola: migration negligible 21. Winter Park: 1,200 (200)]

[Ill.u.s.tration: FIG. 47. Winds aloft at 10:00 P. M. on May 21 (CST). Winds at 2,000 feet above mean sea level are shown.

Velocities are indicated by standard Beaufort Scale of Wind Force. The numbers in circles refer to the stations shown in Figure 46.]

On the first night, following the pa.s.sage of a cold front, migration at Ottumwa was comparatively low (6,900 birds in five hours). On the following night, when the trajectory of the winds was toward the north, the volume of migration was roughly twice as high (22,300 birds in eight hours). At Louisville, on May 21-22, the nightly station density was only 1,500 birds in seven hours, whereas on the following night, it was 8,400 birds in the same length of time, or about six times greater.

The evidence adduced from the present study gives support to the hypothesis that the continental pattern of spring migration in eastern North America is regulated by the movement of air ma.s.ses. The clockwise circulation of warm air around an area of high pressure provides, on its western edge, tail winds which are apparently favorable to northward migration. High pressure areas exhibit a centrifugal force outward from the center, which may tend to disperse the migratory flight originating at any given point. In contrast, the circulation of air in the vicinity of a low pressure area is counterclockwise with the force tending to be directed inward toward the center. Since the general movement of the air is from the high pressure area toward a low pressure area, birds starting their migrations with favorable tail winds, are often ultimately carried to a region where conditions are decidedly less favorable. In the vicinity of an area of low pressure the greater turbulence and high wind velocities, combined with the possibly slightly less buoyant property of the air, cause birds to descend. Since low pressure areas in spring generally precede cold fronts, with an attending shift of the wind to the north, an additional barrier to the northward migration of birds is imposed. The extreme manifestation of low pressure conditions and the manner in which they operate against bird flight, are a.s.sociated with tropical hurricanes. There, the centripetal force of the wind is so great that it appears to draw birds into the "eye" of the hurricane. A cla.s.sic example of this effect is seen in the case of the birds that came aboard the "West Quechee" when this vessel pa.s.sed through the "eye" of a hurricane in the Gulf of Mexico in August, 1927. I have already discussed the details of this incident in a previous paper (1946:192). There is also the interesting observation of Mayhew (1949), in which a similar observation was made of large numbers of birds aboard a ship pa.s.sing through one of these intense low-pressure areas.

Although the forces a.s.sociated with an ordinary low-pressure area are by no means as intense as those a.s.sociated with a tropical hurricane, the forces operating are much the same. Consequently birds conceivably might tend to be drawn toward a focal point near the center of the low, where the other factors already mentioned would tend to precipitate the entire overhead flight. Visible evidence of migration would then manifest itself to the field ornithologists.

CONCLUSIONS

1. Telescopic counts of birds pa.s.sing before the moon may be used to determine reliable statistical expressions of the volume of migration in terms of direction and of definite units of time and s.p.a.ce.

2. Night migrants fly singly more often than in flocks, creating a remarkably uniform dispersion on a local scale throughout the sky, quite unlike the scattered distributions observable in the daytime.

3. The nocturnal migration of birds is apparently preceded by a resting or feeding pause during which there are few migrants in the air. It is not to an important degree a non-stop continuation of flights begun in the daylight.

4. Nightly migrational activity in North America varies from hour to hour according to a definite temporal pattern, corresponding to the _Zugunruhe_ of caged European birds, and expressed by increasingly heavy flights up until the hour before midnight, followed by a p.r.o.nounced decline.

5. The visible effects of the time pattern are subject to modification at a particular station by its location with respect to the resting areas from which the night's flight originates.

6. Quant.i.tative and directional studies have so far failed to prove that nocturnal migrants favor narrow, topographically-determined flight lanes to an important degree.

7. Flight densities on the east coast of Mexico, though of first magnitude, have not yet been demonstrated in the volume demanded by the premise that almost all migrants returning to the United States from regions to the south do so by coastal routes.

8. Heavy flights have been recorded from the northern coast of Yucatan under circ.u.mstances leading inevitably to the conclusion that birds migrate across the Gulf of Mexico in considerable numbers.

9. There is reason to believe that the importance of the Florida Peninsula as an April and May flyway has been over-estimated, as regards the numbers of birds using it in comparison with the numbers of birds using the Mexican and Gulf routes.

10. The amount of migration is apparently seldom sufficient to produce heavy densities of transient species on the ground without the operation of concentrative factors such as ecological patterns and meteorological forces.

11. The absence or scarcity of transients in some areas in fine weather may be explained by this consideration.

12. A striking correlation exists between air currents and the directional flight trends of birds, suggesting that most night migrants travel by a system of pressure-pattern flying.

LITERATURE CITED

ALLEN, R. P., AND R. T. PETERSON

1936. The hawk migrations at Cape May Point, New Jersey. Auk, 53:393-404.

ANONYMOUS 1936-1941. Tables of computed alt.i.tude and azimuth. U. S. Navy Department Hydrographic Office. U. S. Govt. Printing Office, Washington, D. C., vols. 3-5.

1941. Airway meteorological atlas for the United States.

Weather Bureau Publ. 1314. U. S. Dept. Commerce, Washington, D. C.

1945-1948. The American air almanac. U. S. Naval Observatory.

U. S. Govt. Printing Office, Washington, D. C., 3 vols., issued annually.

1948_a_. Meteorological and climatological data for April 1948.

Monthly Weather Review, April 1948, 76:65-84, 10 charts.

1948_b_. Meteorological and climatological data for May 1948.

Monthly Weather Review, May 1948, 76:85-103, 11 charts.

BAGG, A. M.

1948. Barometric pressure-patterns and spring migration.

Auk, 65:147.

BERGMAN, G.

1941. Der Fruhlingszug von _Clangula hyemalis_ (L.) und _Oidemia nigra_ (L.) bei Helsingfors. Eine Studie uber Zugverlauf und Witterung sowie Tagesrhythmus und Flughohe.

Ornis Fennica, 18:1-26.

BRAY, R. A.

1895. A remarkable flight of birds. Nature (London), 52:415.

CARPENTER, F. W.

1906. An astronomical determination of the height of birds during nocturnal migration. Auk, 23:210-217.

CHAPMAN, F. M.

1888. Observations on the nocturnal migration of birds.

Auk, 5:37-39.

DAVIS, L. I.