PHILOSOPHICAL SOCIETY OF WASHINGTON. 31 



another direction. All this took place within half an hour in open 

 water, always in full view of the signal station, and without any- 

 visible obstacle being interposed or removed. 



While on the island we learned that one of the light-house 

 keepers, who had been on leave, had just returned from Sag Har- 

 bor, twenty miles away to the southeast. He had failed to hear the 

 signal at all, until opposite the eastern end of Great Gull Island, 

 and until he was within half a mile of the siren which Was in full 

 operation. 



On the next morning our steamer anchored about a mile north 

 of Little Gull ; the wind was light, the air was clear, and the day 

 was warm and beautiful. As it had been preceded by a warm 

 night the atmosphere was homogeneous, and it was expected that 

 we should have a day of normal audition and barren of curious 

 phenomena. After the siren had commenced its noise we ran down 

 to a point within half a mile of the light-house, and then steamed 

 for Plum Island, running a little south of east for six miles, when 

 we returned as nearly as might be on our own track. The results 

 were curious. We lost half the force of the sound when within a 

 quarter of a mile of the siren ; a moment later we had lost four- 

 fifths of it. Running another half mile we were off the middle of 

 Great Gull Island, and the sound had increased to a foroe of four ; 

 in five minutes more it had dropped to three ; from that on, until 

 we reached the end of our six mile run, it gradually weakened, 

 and it had dropped to a force of two when we turned and ran 

 back to our anchorage. It is particularly curious that the sound 

 had the same intensity at three-sixteenths of a mile from its 

 source, and at six whole miles from that point, while it varied 

 from two to ten in a scale of ten between those points. The results 

 of the trip are more fully and exactly given in Table E. 



Thinking that possibly this peculiarity might have been induced 

 by those differences of temperature in the strata of the atmosphere 

 suggested by Dr. Tyndall as probable cause for such phenomena, 

 effort was made to ascertain something of these differences by send- 

 ing a thermometer to the upper air. In the course of the afternoon 

 we made a kite some six feet high, attached to it a self-registering 

 thermometer, and after a number of trials succeeded in getting it 

 up about five hundred feet, and in hauling it safely in again after 

 it had been up over an hour. The thermometer had a wet bulb, 

 and beside was protected from the direct rays of the sun ; but it 



