While conservation-minded individuals have been concerned with the tremendous mortality involved in these various events, the ill wind blows some good in that, properly used, the data provided by such accidents can shed light on many obscure aspects of bird migration. Each accidental kill of birds affords a cross-section, approaching in variable degree a random sample, of the migrants passing a given point on a given date. The types of information provided by such kills are numerous, for example: (1) information on the presence of various species and the dates of their occurrence; (2) information on the relative abundance of species; (3) quantitative data on the relative sizes of males and females, and immatures and adults (of importance to taxonomic ornithology); (4) information on the relative times of migration of males, females, adults, and young; (5) information on molts and plumages; (6) quantitative information on composition by subspecies of migrants of the same species; (7) physiological data (fat condition, etc.) pertinent to the study of migration; and probably others.
In spite of the great potential of this kind of material, the majority of ornithologists with access to such data have contented themselves with listing the species and sometimes the numbers of birds killed. A few have gone further. James T. Tanner (unpublished) attempted to compute the longevity of the Ovenbird (Seiurus aurocapillus) by analysis of ceilometer-killed birds at Knoxville, Tennessee (see below). Mention should be made of the reports of Rintoul and Baxter (1914) supplemented by Ticehurst (1916) who used rather small numbers of birds killed at Scottish lighthouses in studies of molt. However, the only effort to utilize the results of accidental kills on a large scale over a period of years appears to have been that, already mentioned, of Barrington (1900) and his co-workers in Ireland. An idea of the potentialities of the large recent kills in the United States may be obtained when it is recalled that in the 18 years of Barrington's work, which embodied some 1000 reports from lighthouse keepers, Barrington obtained for study only about 2000 specimens, many of these consisting of wings and feet only (Barrington's paper not seen in original; see J. A. Allen, 1901:205). More recently Dobben and Bruyns (1939) have analyzed the age and sex classes of some birds killed at lighthouses in Holland.
As far as we have learned, there is no previous thorough analysis in the literature of large, accidentally-killed samples of birds. On the following pages we emphasize some of the uses which can be made of such material. We think that intensive analyses of such events, whenever they occur, should become a regular part of ornithological investigation and that integration of numerous studies of such incidents will provide an unprecedented mass of information on migration.
Accidents to Migrating Birds in early October, 1954
General.—The few days around the end of the first week of October, 1954, were notable for a series of accidents which occurred to migrating birds over much of eastern United States. So far as we know, these were all associated with an extensive belt of bad weather (cold fronts and stationary fronts) which covered much of the country during that period, and the accidents involved ceilometers and solid structures alike. Accidents known to us occurred as far south as Macon, Georgia (David W. Johnston, letter: Nov. 1, 1954), as far north as New York City, where many migrants were killed at the Empire State Building (New York Times, Thursday, October 7, 1954, p. 1) and elsewhere, and as far west as Smoky Hill Air Force Base at Salina, Kansas (ceilometer, October 7, some birds received at the University of Kansas). Some of the above, and incidents from a number of other localities, were mentioned in varying detail in Audubon Field Notes (vol. 9, no. 1, pp. 6, 10, 15, 17, 18, 32, February, 1955). Still other accidents occurred at Columbia, Missouri (Richard P. Grossenheider, verbal communication), and Topeka, Kansas (present paper). Some probably have escaped our notice; summaries of some of these will probably appear in ornithological journals for some time to come. At Robins Air Force Base near Macon, Georgia, at least 50,000 birds were killed, of which about 2500, representing 54 species, were picked up (Johnston, loc. cit.).
Accidents at Topeka, Kansas.—At Topeka, Shawnee County, Kansas, all birds were killed by collision with the newly-erected (1954) television transmitting tower of station WIBW-TV. This tower is one mile west of the city.
The first casualties (see Table 1 for all others) were a Sora (Porzana carolina) and a Yellow-bellied Flycatcher (Empidonax flaviventris) found on September 7. The major accidents, however, occurred on the nights of September 24-25, September 30-October 1, October 5-6, and October 6-7. Totals of birds picked up (probably over 95 per cent of birds killed) are given in Table 1, in which each date given is that of the day after the kill, i. e., the date on which the birds were collected.