C. GENERAL PHYSICS. 133 



ments establishing this explanation were made over a nearly 

 flat meadow when covered with snow, and again when cov- 

 ered with grass. He finds that the velocity of wind over 

 grass differs by one half at elevations of one to eight feet, and 

 by somewhat less over snow ; that the same sound is heard 

 at more than double the distance over snow at which it can 

 be heard over grass; that sounds proceeding with the wind 

 are brought down to the ground, so that the range of sound 

 over rough ground is greater with the wind than at right 

 angles to it ; that sounds proceeding against the wind are 

 lifted off the ground, and hence the range is diminished. A 

 second class of phenomena, namely, the fact which has been 

 often observed, that distant sounds can be heard much better 

 during the night than during the day, and on cloudy days 

 better than on clear days, Professor Reynolds explains as 

 the result, not of the heterogeneous condition of the air, as 

 maintained by Professor Tyndall, but of the refraction of the 

 sound. Since the velocity of sound through air increases 

 with the temperature, every degree from 32 to 70 adding one 

 foot per second to the velocity, therefore an upward diminu- 

 tion in the temperature of the air must produce a similar ef- 

 fect to that of wind lifting the sound. Pie shows that with 

 a clear sky, and a diminution of temperature, such as results 

 from Mr. Glaisher's observations, the rays of sound, other- 

 wise horizontal, would be bent upward on arcs of circles, the 

 radii of which are about twenty miles, while with a cloudy 

 sky the radii are forty miles; so that the upward refraction 

 is twice as great on bright hot days as it is when the sky is 

 cloudy, and, of course, still more under exceptional circum- 

 stances. This theoretical result is then applied to one of 

 Professor Tyndall's observations made on the 3d of July, on 

 which occasion he observed that the guns fired on the cliffs 

 of South Foreland, 235 feet high, were inaudible at a distance 

 of two miles. Professor Reynolds shows that, in this case, 

 the sound-wave would be refracted, so as to pass completely 

 over the head of the observer standing on a ship's deck 

 twenty feet high ; but that, had Professor Tyndall been able 

 to go thirty feet up the mast, he might have heard the sound 

 distinctly. Instruments placed at the bottom of the cliff 

 would, for the same reason, have a still shorter range of au- 

 dibility; and it is singular that, throughout his report, Profess- 



