FKOZEN WELL. 197 



north of the frozen well. The average depth of the water was found to be between 50 

 and 60 feet. In one place at 75 feet in depth, Mr. A. found 



Temperature of the air, .... 73 

 do of the surface water, . . 70 

 do of the bottom water, . .41 



In another place, 65 feet deep, the water was 46 

 We are indebted to David A. Wells, Esq. for referring us to these facts. 



We have been led to inquire for cases elsewhere, analogous to that at Brandon, and 

 have found a few, which we will briefly describe, because we think they cast some light 

 upon the example in Vermont. 



Rev. Ariel E. P. Perkins, of Ware, in Mass., dug a well in that place, in 1858, 35 feet 

 deep through gravel and sand, which we know from personal recollection to be very much 

 like that in which the Brandon frozen well occurs, except that there is less clay at Ware, and 

 none of the pebbles are limestone. Last winter the water in Mr. Perkins' well froze over 

 one night at least, though the ice was not much thicker than window glass. July 26th, 

 1859, the temperature of the water was 46, which differs but little from the mean tem- 

 perature of the region ; and in the winter of 1859 it froze over again. Mr. Perkins' well 

 is within his house and of course protected above, so that we must look to some other 

 source for the freezing than cold air descending the well. 



In the 36th volume, first series, of the American Journal of Science, p. 104, D. 0. Ma- 

 comber has described a frozen well 77 feet deep, in Owego, JX". Y., which "for four or five 

 months in the year is frozen so solid as to be entirely useless to the inhabitants." The sides 

 of the well in winter are so coated with ice as to leave at the bottom a space of only a foot in 

 diameter ; and the ice on the surface of the water could not be broken by a heavy iron 

 weight attached to a rope. It continues certainly as late as July, but disappears (as we 

 understand the account) in the latter part of summer. 



Another fact it may be important to state as to this well. "A lighted candle being let 

 doAvn, the flame became agitated and thrown in one direction at the depth of 30 feet, but 

 was quite still and soon extinguished at the bottom." 



This well "is excavated on a table land elevated about 30 feet above the bed of 

 Susquchanna River, and distant from it three-fourths of a mile." In the excavation no 

 rock or slate was thrown up. This statement makes it almost certain to our mind that 

 the Owego well occurs in one of the sandy or gravelly terraces of the Susquehanna, and 

 we have ascertained that such is the fact. 



One other notice has been kindly sent to us by Dr. Samuel L. Dana, of Lowell, since we 

 presented the subject to the American Scientific Association, and in consequence of it. 

 "The frozen well discussion," says Dr. Dana, "has called to my recollection an account 

 of a frozen well, which I saw some thirty six years ago in a Miscellaneous and Literary 

 Journal published at Concord, N. H. 



"I have hunted up the paper published in 1823, and signed Caleb Emery of Lyman, 

 Grafton Co., N. H. He states that in 1816 that memorable cold summer he visited in 

 June a well in that town, which had become frozen over solid, eight feet from the surface 

 of the ground, so that a hole had to be cut through the ice so as to get water. In July 



