1913 ' Forbes, Concerning the Flight of Gulls. 365 



maximum ;is the air rounded the corner, to he followed by expan- 

 sion as it passed over the horizontal surface of the deck. The 

 result of such expansion wonld be a divergent upward slant in that 

 portion of the air which was at a slight distance from the deck. 



On the leeward side of the ship the wind wonld probably cant 

 downward to form an eddy. If this were so it wonld readily ex- 

 plain the avoidance of 1 he lee side of the ship by the gulls. 



Over the smoke-stack the ascending current must have been 

 powerful, because of the heat. Its presence in the wake of the ship 

 may be explained l>\ the fact, that the wind was not dead ahead 

 but from a point and a half to two points off the port bow. As long 

 as the volume of smoke and air poured from the funnel remained 

 at a higher temperature than the surrounding air it must have 

 continued to rise. And since such upward movement tends to 

 be imparted to adjacent air it is probable that for a short distance 

 to windward of the smoke an upward diversion of the wind oc- 

 curred. With the wind blowing obliquely on the bow any given 

 point in the wake must have been directly to windward of some 

 point in the trail of smoke, namely, that portion of the smoke 

 which was discharged when the boat was at the given point. The 

 birds, although directly in the wake of the steamer, were directly 

 to windward of a large mass of air and smoke, probably still warm 

 and rising vigorously. They may well have been aided in this 

 manner by rising currents for several hundred yards astern of the 

 boat. 



I do not claim that these suggestions cover all the factors in- 

 volved in the explanation of the gliding Mr. Brewster describes. 

 But I contend that the wind must have presented other than 

 uniform horizontal motion to render the feat possible. Ascending 

 currents caused in some of the ways I have suggested seem to me 

 to present the easiest explanation that has occurred to me. 



Addendum. 



Since writing this discussion my attention has been called to 

 the papers on the subject by Lord Rayleigh, in w Inch he discusses 

 these points so clearly and concisely that my remarks seem almost 



