1886.] 



Observations on Pare lee and Snow. 



547 



cylindrical mass of snow was at the uniform temperature of 0° F. The 

 freezing mixture was then removed and replaced by about 15 cwt. of 

 snow, and the temperature readings regularly taken until the snow 

 mass in the tank arrived at one even temperature of 32° F. through- 

 out. The readings are delineated on Table I. 165^ hours were occu- 

 pied in arriving at this result. 



The relative conductivity of the ice was thus found to be about 

 122 per cent, greater than the snow under the conditions of experi- 

 mentation. 



Set II.- — Dilatation of Ice between the Temperatures of — 35° F. 

 and +32° F. 



In conducting these experiments, the dilatation of the ice between 

 the temperatures of zero (0° F.) and 32° F. was ascertained by the 

 expansion of the ice measured between two iron bars 1 inch square 

 by 19-J inches long, securely embedded to a depth of 13 inches in 

 the ice, and perpendicular to its surface, at a distance of 16| inches 

 apart. Two other square iron bars of the same dimensions were also 

 securely frozen horizontally into the side of the ice cylinder. The 

 measurements between the respective sets of iron bars were accurately 

 taken by a delicate micro-vernier gauge, the limit of error in reading 

 which by the aid of a telescope did not exceed one two -thousandth of 

 an inch. The gauge was securely placed on a suitable wooden frame, 

 and thus removed from the influence of the low temperature, so that 

 all the measurements were taken throughout at exactly the same spot 

 on the bars. The average of 100 measurements in each case was re- 

 garded as the correct reading, and the results are recorded on Table II. 

 After the measuroments had been taken with the ice at zero (0° F.), 

 the freezing mixture surrounding the ice tank was removed and re- 

 placed by about 15 cwts. of snow, until the whole mass of ice again 

 reached the temperature of 32° F. The relative distances between 

 the respective sets of bars at the higher temperature of 32° F. were 

 again taken, and the average of 100 readings regarded as correct. 

 The dilatation of the ice cylinder, both longitudinal and transversely, 

 thus obtained between the temperatures of zero (0° F.) and 32° F. is 

 given on Table II. A difference was observed between the longi- 

 tudinal and transverse dilatation. This was not owing to error in 

 observation, as the deviation was constantly noticeable during the 

 course of the measurements. Certain crystalline bodies dilate un- 

 equally, and the ice also appeared to behave in a similar manner ; the 

 difference noticed may, therefore, possibly have been due to the mode 

 of crystallisation of the cylindrical mass of ice. 



The observations relating to the expansion of ice between — 35° F. 

 and +32° F. were conducted in a similar manner: but to obtain the 

 lower temperature using a freezing mixture of three parts by weight 



2 o - 2 



