1877] WRITINGS OF JOSEPH HENRY. 525 



echo was returned from the horizon around its whole cir- 

 cumference. 



In another experiment with a vertical trumpet at Little 

 Gull Island, a small cloud, from which a few drops of rain 

 fell on the area of the base of the light-house, passed directly 

 across the zenith, and during this passage no echo was ob- 

 served from the cloud, although the trumpet directed toward 

 it was sounded several times in succession. 



Again, in order to obtain additional facts in regard to the 

 nature of this echo, observation was made from a vessel, by 

 steaming out directly as if into the region of the echo — i. e., 

 in the direction of that point in the horizon from which the 

 echo appeared to emanate. 



In this case the loudness of the echo appeared to grad- 

 ually diminish as we advanced, and to spread itself through 

 a much longer arc of the horizon, while the duration of the 

 echo increased in time. 



It would follow from this experiment that the echo is not 

 a reflection from a definite surface, since it would then in- 

 crease in loudness as the surface is approached, but a series 

 of rebounds from points at various distances. 



Another fact of great importance in determining the nature 

 of the echo is that derived from the observations of the keeper 

 at Block Island. He has recorded every Monda}' during a 

 year, the observations of the length of the continuance of 

 the echo, the state of the weather, the direction of the wind, 

 and the other meteorological data. Whence it is found that 

 the echo from the sound of the siren is always heard during 

 a wind in any direction, and of all intensities, but during 

 the occurrence of a very high wind, with less duration after 

 the original blast, than in calmer weather; and above all, 

 that it is heard equally well during a dense fog, when evi- 

 dently the air must be homogeneous and saturated in every 

 part with vapor. 



From these facts it appears to me conclusive that the 

 reverberation, constituting the oceaoi echo, cannot be due to 

 invisible clouds. The only hypothesis suggesting itself to 

 my mind as a basis for further investigation of this subject 



