77 
1891-92.] Mr R W. Western on the Tactics of Birds. 
This relation is not exact. When a ship lies at anchor, and 
swings to the tide, the pressure produced by the stream dividing 
at the hows is balanced by the waters closing again at the stern 
(provided her lines are sufficiently tine), so that the only force 
straining the cable is due to skin friction. If this is true for a 
dense fluid like water, how much more must it hold in thin air 
whose mobility enables it to get round the sharp corner of the most 
angular bodies with very little eddying? Still, by treating the 
area which a bird presents, as a flat surface, we introduce an error 
in the opposite direction, for the form tends to reproduce the 
velocity of impinging air in an opposite direction, thereby increasing 
the pressure sustained. 
In the absence of better information on this point, we must 
suppose the two errors to cancel one another more or less, awaiting 
the time when the results of a number of delicate experiments will 
enable us to treat the matter with greater exactitude. 
It may be noted that the distribution of feathers at the edge of 
the wing is such as to promote the formation of eddies. 
To proceed, then, on the hypothesis explained above, the 
resistance of the air to the bird, when falling with out- 
GA?; 
stretched wings at v feet per second, will be where A square 
feet is the area presented by the bird, and G lbs. is the weight of a 
cubic foot of air; p' = 33’2. It is obvious that this quantity cannot 
be greater than the weight of the bird. Thus from the equation 
GA 
W = — we obtain v = 
g 
as the greatest velocity which the 
falling bird can attain. 
From a number of measurements of the particular birds to which 
the observations of the author relate, W may be taken at lbs. 
and A at 2^ square feet. So that the greatest rate of falling will 
be 
fix 32-2 
•0807 X 2*5 
= 14*5 feet per second. 
The falling bird will never exactly attain this velocity, but it 
very quickly approaches it. 
To find the velocity attained after falling any number of seconds 
t ; we must equate the increase of motion, to the excess that the 
