Calorimetric Thermometers. 331 



mometers employed had a range of 15° C, and a total length 

 of (300 millim.; it was therefore impossible to increase their 

 length sufficiently to make one instrument embrace the whole 

 range of 25° C, and any increase in the delicacy (by increasing 

 the size of the bulb) would have involved a diminution in 

 their range, so that more than two instruments would have 

 been required, while in such experiments it is obviously of 

 the greatest importance that as few instruments as possible 

 should be used. 



With a view to overcoming these difficulties the following 

 expedient was adopted, whereby the delicacy of the instru- 

 ment was increased six- or sevenfold, its range was made to 

 embrace that of all temperatures at which a mercurial ther- 

 mometer can be used, and its registration was rendered (for 

 these purposes) practically independent of calibration and 

 standardization error. 



The principle on which it was worked consisted in removing 

 a portion of the mercury, by the application of heat, into the 

 upper chamber, and thus adjusting any particular point in the 

 instrument so as to register any required temperature. Ex- 

 periments on the same substance, at different temperatures 

 were thus made absolutely comparable, exactly the same por- 

 tion of the scale being used in each case. 



The zero point of the thermometer being thus altered for 

 each experiment, the actual temperature in degrees Centigrade 

 which it registers will not be known with any great degree 

 of accuracy; but this is not necessary in calorimetric work: 

 the heat of chemical combination is affected to such a small 

 extent by temperature, that an error of -05° or -1° C. in this 

 will be unimportant; and by a rough comparison of the 

 thermometer with some standard instrument after each ad- 

 justment of the mercury, the temperature which it registers 

 can be easily ascertained w T ithin "01° or "02° C. 



This thermometer (No. 62839) possessed a total length of 

 746 millim., of w 7 hich 565 millim. constituted the divided por- 

 tion of the stem, each division being 1 millim., and ten such 

 divisions being termed 1 degree. The bulb measured 72 millim. 

 in length, and was made out of glass cylinder instead of being 

 blown, a detail which is rarely attended to in the construction 

 of delicate thermometers, but one which is most important in 

 giving the bulb greater stability and greater uniformity of 

 action. The weight of mercury in the bulb was 36*23* grams, 



* In the other instruments, with a range of 15° C, the weight of 

 mercury varied between 13*6 and 15*6 grams ; the bore of the tube of 

 No. 62839 would be considerably more than twice as fine as in these. 

 Fuller details respecting these instruments will be found in the Phil. Mag. 

 March 1886. 



