﻿H. Carmichael — Determination of Arsenic. 129 



stantly at the temperature of maximum density or 38*8° F. ; 

 and when we recall the excellent conducting power of water as 

 regards heat, the influence of this great enveloping body of 

 cold water can be at once appreciated. And it is primarily to 

 it acting as a great cold blanket, that the writer attributes the 

 general coolness of the rocks in this region, and secondarily in 

 accounting for the local variations in the thermic gradient of 

 the different mines. The nearer the mines are to this great 

 cooling influence, the lower the thermic gradient will be found 

 to be, while the farther one goes away the less will its influence 

 be felt, until, should we pass entirely beyond such influence, 

 we should probably get the very general and usual gradient of 

 50 to 55 feet. That such is the case seems to be indicated by 

 the observations cited, and this is the theory offered by the writer. 

 None of the mines mentioned are so distant from the cold 

 body of lake water but that, notwithstanding the much smaller 

 efficiency of rock-formations as heat conductors, they still show 

 the effect of its strong influence. While the Osceola mine, with 

 nearly 5 miles intervening between it and the main lake, shows 

 that this thickness of rock-mass moderates very appreciably the 

 refrigerating action of the lake it also indicates that, perhaps 

 at a distance of 7 or 8 miles from the lake, such a thickness 

 of rock-mass would render its effect inappreciable, and we should 

 again have the normal gradient. 



Art. XIY.— An Application of the Copper Reduction Test to 

 the Quantitative Determination of Arsenic ; by Henry Car- 

 MICHAEL, Ph.D. 



One milligram of arsenious oxide added to strong chlorhydric 

 acid and digested at the boiling temperature with a clean pol- 

 ished piece of copper having a surface of forty square milli- 

 meters imparts to the copper a grayish tint without fully 

 extinguishing the red color of that metah The color darkens 

 rapidly if a larger quantity of arsenious oxide is supplied, and 

 if the copper is removed as soon as it has assumed a steel gray 

 tint, it will be found that a milligram and a quarter of the 

 oxide has been removed from solution. The transition from 

 reddish gray to steel gray is so marked that the experimenter 

 can readily match a specimen which has been selected as the 

 standard of color. 



As the arsenic compounds react more slowly and with less 

 certainty, they should always be reduced to the -ous equivalent 

 by means of sulphurous oxide before the trial is made. Using 

 a standard square of copper as an indicator, the writer has been 

 led to the adoption of a method which is believed to be, for the 



