110 



Mr. T. Andrews. 



[June 5, 



tionate preponderance of organic matter in the ice compared with the 

 water. The experiments were made as follows : Portions of the ice 

 were taken from numerous places on the pond, and very thoroughly 

 washed with distilled water, the ice was then melted in large glass- 

 beakers, and the melted ice was found to be perfectly clear and 

 translucent, there being no deposit after the water had stood for a 

 considerable time. The melted ice, in quantities of half a gallon, was 

 evaporated to dryness in a platinum basin, the residue dried at 212 

 and weighed, being subsequently ignited and again re-weighed. The 

 results are given on Table II. The results obtained confirm the 

 observations made by Buchanan on the composition of ice in Arctic 

 seas ; see remarks hereon towards the close of this Memoir. 



Table II. Analyses of Pond Ice. 

 Results in Grains per Gallon of the Melted Ice. 



The penetration of the steel rod into pure ice at constant tempera- 

 ture of 35 F. is shown on Diagram I by curve I; at F. by 

 curve II ; at 28 F. by curve III ; and at 32 F. by curve IV. 



In the experiment, curve II, the needle was allowed to remain on 

 the ice for a total period of 42 hours, the penetration at 18 hours 

 from commencement was 0'906 inch, at 30 hours 1*031 inch, and at 

 42 hours fioni commencement T187 inch. 



The penetration of the steel rod into pond ice at 28 F. is shown on 

 Diagram II by curve I, and at 32 F. by curve II. 



The pond ice, curve I, was 6 inches thick, and curve II, 5^ inches 

 thick. 



