HOW BIRDS MIGRATE 51 
the vast numbers of birds which throng the upper air during nights of 
active migration. 
At Madison, Wisconsin, on the nights of September 11-13, 1897, 
Libby (1. c.) saw 583 birds through a 6-inch glass, the largest number 
counted during a fifteen minute period being forty-five. Additional 
data of this nature are included in a paper by H. A. Winkenwerder 
(02), while Stone’s (’06) observations on birds seen migrating at night 
by the light of a great conflagration in Philadelphia have exceptional 
interest. 
The height at which migrating birds fly has been variously esti- 
mated. Many diurnal migrants may be seen traveling from a few 
yards (e. g. Shearwaters) to several hundred yards (e. g. Geese) above 
the earth, but Shore-birds evidently seek a greater altitude, and the 
experience of Bray, referred to above, hints at a diurnal flight of which 
we know practically nothing. 
Our knowledge of the height at which nocturnal migrants journey 
is based on the telescopic observations already mentioned. As intimated 
in the article itself (7. c.), the conclusions presented in my paper on 
birds seen flying over Tenafly, were not satisfactory. More recently, 
the problem has been attacked by F. W. Carpenter (’06) and Stebbins* 
with results which seem far more worthy of acceptance. Their calcu- 
lations show that while an altitude of 4,000 to 5,000 yards may occas- 
ionally be reached, the greater number of birds observed were not 
over 1,600 yards above the earth, while many passed considerably 
below this elevation. The ease with which the calls of night-flying 
birds may be heard also argues for a lower altitude than has been com- 
monly accredited to them. 
The speed at which migrating birds fly has also been greatly over- 
rated. Two observations with theodolites give to migrating Ducks 
a speed of 47.8, and to migrating Geese a speed of 44.3 miles an hourt. 
Homing Pigeons do not often exceed forty to forty-five miles an hour. 
It is a common experience, when traveling in a train at a rate of thirty- 
five to forty-five miles an hour, to pass birds which are flying parallel 
to the track. I have had this occur repeatedly with such compara- 
tively large and swift birds as the Mourning Dove. It seems probable, 
therefore, that our smaller birds do not average more than thirty miles 
an hour when migrating. 
Gitke’s estimate of 212 miles per hour as the speed of the Golden 
Plover when migrating, is based wholly on the assumption that fifteen 
hours is ‘the longest spell during which a bird is able to remain on 
the wing without taking sustenance of any kind.’ The Plover, how- 
ever, does not fly from “Labrador to northern Brazil,” as he stated, but 
from Nova Scotia to the Lesser Antilles; as we have seen, it makes 
special preparation for the journey, is extremely fat when it starts and 
thin when it arrives. 
*Popular Astronomy,” XIV, 1906, pp. 65-70. 
+ Clayton, Science, 1897, pp. 26, 585. 
