PHILOSOPHICAL SOCIETY OF WASHINGTON. 169 



suits in previous investigations of the same subject. To test 

 this, and to ascertain the dependence of the phenomenon on the 

 wind, the position of the focus or the origin of the sound was 

 changed. For this purpose, the whistle of the steamer was 

 sounded, while a portion of the observing force was placed at 

 the station; by this arrangement it was found that while the 

 vessel, in reference to the sound of the signal at the station, 

 passed through a region of silence, the observers at the station, 

 who gave attention to the sound from the steamer, heard no in- 

 terruption of the signal. This experiment was repeated each 

 way, going to and coming from the station. 



From this result it appears that the sound going luith the wind 

 was heard from every point on its course, while the sound moving 

 against the wind was suddenly lost at a given point and not re- 

 covered again until a distance had been traversed by the vessel 

 of more than a mile. This result was in strict conformity with 

 the theory of refraction;— in the case of the sounds travelling 

 against the wind, the upper part of the wave would usually be 

 more retarded than the lower, and consequently the sound wave 

 would be thrown upward, above the head of the observer. At a 

 given altitude this difference of velocity would cease and by the 

 tendency of sound constantly to spread, the sound wave would 

 again reach the earth. 



But, to test this still farther and to show that the locality was 

 not an essential condition of the existence of the interval of 

 silence, the experiment was repeated on the opposite side of the 

 station, so that the sound from the fog-signal would move in the 

 direction of the wand. Part of the observers were placed at the 

 station and the other part remained on board the vessel ; both 

 instruments were sounded, the one in the intervals of the sound- 

 ing of the other. 



In this case the sound from the fog-signal was continuous to 

 those on board of the vessel through a distance of over four miles 

 and could probably have been heard many miles farther, but the 

 progress of the steamer in that direction was stopped by the land. 

 From the report of the observers at the station it appeared that 

 as the vessel passed into the distance but one blast was heard 

 during its whole course. In this case (as in the preceding 

 experiment of sailing to the southwest) the sound moving against 

 the wind was refracted upward, aad as the whistle was but sis 



