26 



PROCEEDINGS OF THE AMERICAN ACADEMY 



[^ 



Fig. 2. 



point of antimonious iodide at the normal pressure of the 

 air within half a degree on either side. 



The method we have here described we can most con- 

 fidently recommend as a most efficient and accurate means 

 of determining high temperatures in chemical laboratories. 

 It requires no expensive apparatus, and no more delicate 

 manipulation than most processes of gas analysis. Indeed, 

 this method is most readily associated with Bunsen's meth- 

 ods of gas analysis ; and, in a laboratory provided with a 

 room fitted for that work, the observation of temperature 

 we have described can be made in a very short time. 



In connection with these experiments, we were led to 

 devise a very simple and inexpensive form of differential air 

 thermometer, that can be used almost as readily as a 

 mercury thermometer, and which will measure either high 

 or low temjjeratures with all the accuracy that is usually 

 required. The instrument is represented by Fig. 2 of one 

 half its linear dimensions. The long stem is made of 

 "barometer tubing," a little over a millimetre in diameter, 

 and by careful calibration is arbitrarily divided into parts of 

 equal capacity, making, we will say, two hundred divisions 

 on the length of the stem. While the instrument is still 

 open at both ends it is easy to determine, first, the weight 

 of mercury which fills the bulb up to the first division of 

 the stem ; and secondly, the weight of a column of mercury 

 covering an observed number of divisions of the stem. 

 These constants being known, and the interior of the instru- 

 ment having been most carefully dried, for which the two 

 openings offer great facilities, a short column of very pure 

 mercury is introduced, and brought into the position repre- 

 sented in the figure. The two ends are now hermetically 

 closed with a blow-pipe, and the instrument is made. It can 

 be used either in a vertical or horizontal position, although 

 the zero point of the scale is slightly different in the two 

 cases, owing to the weight of the short mercury column. 

 Of course this column remains immovable so long as the 

 temperature of the two ends remains the same ; but when 

 the bulb is heated, the column, which we will call the index, 

 moves up the stem, which becomes a closed monometer. If 

 the instrument is to be used for measuring low tempera- 

 tures, the index should be placed about one third way up 



