74 



marble, with a circular opening in it eighteen inches in diameter, 

 with a curb windlass and a cover to the latter, to preserve the rope 

 from rain. 



A boy, son of Mr. Twombley, who had been in the habit of de- 

 scending into the well daily to cut open the ice in the winter months, 

 was sent down the well, by means of the bucket and rope, a candle 

 being lowered down also to light the weU, and to show where the 

 ice rim existed. 



It was found to be a few inches above the surface of the water, 

 and to extend up the sides of the well about five feet, while a well- 

 marked rim of ice projected out over the water, and was about eight 

 inches thick. On this the boy stood, and with a hatchet cut away 

 masses of the ice, and sent them up in the bucket for our examina- 

 tion. 



The temperature of the water in the well was 0.5° centigrade, or 

 32^*'^° Fah.* The temperature of the air in the well near its bot- 

 tom was 2° centigrade, or 35^5^.° Fah. That of the air on the sur- 

 face was 9|-° centigrade, or 49-jL° Fah. 



On measuring the temperature of a spring just outside of the 

 gravel bed, we found it to be 11° centigrade, or 51^^^^° Fah. 



A well belonging to Mr. Strong, a few hundred yards east of 

 Twombley's, was found to be fifteen feet deep, and the temperature 

 of the water was 8° centigrade, or 46^*^° Fah. 



June 11th. — This morning we renewed our labors, re-examining 

 the wells in the neighborhood on all sides of the frozen well. 



Twombley's well had this morning precisely the same temperature 

 it had yesterday, viz., 0.5° centigrade, while that of the air was at 10° 

 centigrade, or 60° Fah. 



Mr. Clarke's well, in a field north-west of the boulder bed, and not 

 far from it, had a temperature of 6° centigrade, or 42^^^° Fah. 



A spring south-west of the frozen well, and not far distant, had a 

 temperature of 9° centigrade, or 48^2_o Fah. 



None of the other wells in the town freeze in the winter, or are 

 remarkably cold. It is evident, therefore, that the geological forma- 

 tion around the frozen well determines its freezing character, and that 

 the gravel bed, in some way, causes the water in that well to freeze, 

 and to continue frozen through the summer months. 



It was therefore decided that a shaft or pit should be sunken the gravel 

 or boulder bed, and a point half way between the frozen weU and the 

 top of " Hogback " was selected as the location ; but Mr. Twombley 



*A11 the measurements of temperature were made with centigrade thermometers, 

 which had recently been carefully veritied as to their zero points. We give the 

 temperature also reduced to Fah. scale. 



