
738 . THE AMERICAN NATURALIST.  [Vou. XXXVI. 
southeast until they reach the Atlantic coast line. They then 
turn westward and follow the shore until they have reached 
New York and New Jersey, where they gradually separate and 
pass on southward. Thus the hawks migrating from a large area 
of country — eastern Canada, New England, and perhaps north- 
eastern New York — are forced to travel along a narrow coast- 
line path at right angles to the main direction of their migration. 
A condensed table of flights which occurred from 1885 to 
1895 is given below. It has already been published in part, 
but it is necessary in the present paper for the purpose of 
showing certain new facts that will be presented. 
The meteorological data in the table as formerly given was 
that for New Haven, Conn., where the observations on hawk 
flights were made. In the table as given at present the obser- 
vations of the U. S. Weather Bureau taken at Boston are sub- 
stituted for those taken at New Haven. This change is made 
because the former station is nearer the locality where the 
hawks start on their migratory journey than the latter place. 
Another column has been added to the table for the purpose 
of showing the effect of temperature changes on the migratory 
movements of the hawks. The meteorological data in the 
tables have been kindly furnished by the Boston office of the 
U. S. Weather Bureau. 
It is difficult to determine the number of hawks that pass 
a given point on the Connecticut shore, during a moderately 
large flight in September. An estimate, which appears to the 
author to be conservative, is that over 15,000 of them pass 
New Haven in one day. The sharp-shinned hawks outnumber 
the other species several times over. The broad-winged hawks 
are next in abundance, then the ospreys and marsh hawks which 
are about equally numerous. 
Observations show that many ospreys and marsh hawks cross 
to Long Island, but it is evident that the main flight of hawks 
is along the Connecticut shore. 
Since 1894 observations have been continued on flights of 
hawks in Connecticut in autumn, and in New Jersey in spring, 
Which have resulted in bringing to light additional facts relat- 
ing to the effect of wind on the migration of birds. 
