744 THE AMERICAN NATURALIST. [VoL. XXXVI. 





TABLE IV. 
WIND IND 
DATE, 1895. chess’ qn DIRECTION VELOCITY Noen — 
(New Haven). (N.H.) OBSERVED. 
Sept. 13 | Clear. N.N.W. 22 Large flight. 
"o ner] Clear. N.W. to N. 9 Very large flight. 
IIS Clear. N. I3 Small flight. 
S SE Cloudy. N. 12 " " 


The observations made of flights of hawks that have occurred 
during a period of fifteen years show that they have taken place 
on nearly every date during September, after the sth of that 
month. This is readily seen from Table I, which gives the 
flights of hawks which occurred during the years 1885 to 1895. 
It is evident that there is no special date on which the hawks 
migrate during the month of September. Their appearance in 
large numbers must be due therefore to some special influence. 
This has already been shown to be the wind from the north- 
west quarter, in which the birds drift southeastward to the 
coast line, making necessary their flight along the shore of 
Connecticut. 
The observations made during September, 1895, show an 
additional fact; if hawks migrated regularly in New .England 
throughout September without regard to the direction of the 
wind, then there would be a small continual flight of hawks 
in southern Connecticut during the entire month, due to the 
peculiar shape of the coast line. Such would be the case, 
because hawks migrating in a southward direction through 
the New England States would eventually reach the south- 
ern border of Connecticut, where it would be necessary for 
them to fly westward along the coast, producing the small 
daily flights referred to. 
It has been proved that no daily flight takes place, only 
a few stray hawks being observed. Moreover, at any time 
1In Table IV the velocity of the wind at 8 A.M., September 14, is given as 
9 miles at New Haven. At Boston, however, at the same time the wind was 
N.W. 14 miles, and at New York 12 miles 
On September 24 the wind at Boston was N. 18 miles, and at New York 
N.W. 14 miles, 
