PHILOSOPHICAL SOCIETY OF WASHINGTON. 35 



tances through the densest fogs, which may be totally inaudible in the same 

 directions and at the same distances in the clearest atmosphere. It is not meant 

 by this last statement that the fog may assist the sound ; as at another time the 

 signal may be absolutely inaudible in a fog of like density, where it had before 

 been clearly heard. That fog has no great effect can easily be understood when 

 it is known, (as it certainly is known by observers,) that even snow does not 

 deaden sound — there being no condition of the atmosphere so favorable for the 

 far reaching of sound signals as is that of a heavy N. E. snow storm, due sup- 

 posably to the homogeneity produced by the falling snow. 



" It seems to be well established by numerous observations that on our own north- 

 ern Atlantic coasts the best possible circumstances for hearing a fog-signal are 

 in a northeast snow storm, and, so far as these observations have extended, they 

 seem to point to the extraordinary conclusion that they are best heard with the 

 observer to windward of the signal ; and that in light winds the signal is best 

 heard down the wind, or at right angles with the wind. 



" The worst conditions for hearing sound seem to be found in the atmosphere 

 of a clear, frosty morning on which a warm sun has risen and has been shining 

 for two or three hours. 



" The curve of audibility in a light or moderate breeze, in general, is similar 

 to that plotted by Prof. Henry, as in the accompanying diagram. 



" I think it is established that there are two great causes for these phenomena, 

 non homogeneity of the atmosphere, and the movement of the wind ; how this 

 latter acts no one can say. The theory of retardation of the lower strata of the 

 atmosphere near the earth's surface, as advanced by Prof. Stokes, of England,* 

 seems good for moderate winds, but it hardly holds in cases where the siren is 

 heard from eighteen to twenty miles to windward during N. E. gales." 



While the mariner may usually expect to hear the sound of the 

 average fog-signal normally as to force and place, he should be pre- 

 pared for occasional aberrations in audition. It is impossible at 

 this point in the investigations which are still in progress, to say 

 when, where or how the phenomena will occur. But certain sug- 

 gestions present themselves even now as worthy of consideration. 



It seems that the mariner should, in order to pick up the sound of 

 the fog-signal most quickly when approaching it from the wind- 



*See Henry on Sound, p. 533; or, Sm. Rept., 1878, p. 533; or, L.-H. B. Rept. for 1875, p 

 120. See Henry on Sound, p. 512, and Taylor in Am. Jour. Sci., 3d series, XI, p. 100,' 

 also, Rept. Brit. Assoc, XXIV, 2d part, p. 27. 



