1869.] 



of Thermometers in a Vacuum, 



323 



The thermometers were tested before putting them into the glass cases 

 by comparing them from three to three degrees with the Kew standard, 

 taking a great number of readings by two independent observers for this 

 purpose. From this comparison and by interpolation, the following Table 

 of corrections for every degree over the range of temperature during the 

 experiments was constructed. It will not only prove that the utmost 

 precaution was taken to ensure the experiments against errors inherent in 

 the instruments employed, but will also show the excellency of the ther- 

 mometers and the degree of accuracy now obtained by eminent makers in 

 the construction of these instruments. 



Table I. Corrections to be applied to the Readings of the Thermometers. 

 N.B. The corrections are in all cases subtractive. 



Thermo- 

 meters. 



4o° 





42° 



43° 



44° 



45" 



46° 



47° 



4 8° 



49° 



5°° 



5i° 



52° 



53° 



54° 



55° 



56° 



57° 



58° 



59° 



6o° 



No. 6700. 























































































•12 



•12 



•12 



•12 



•11 



•13 



•16 



•19 



•18 



•17 



•16 



•16 



'17 



•17 



•16 



•16 



-I 7 



•19 



•20 



-I 7 





No. 6701. 



,][ 3 



•13 



•13 



•12 



•12 



•15 



•19 



"22 



•21 



•19 



•18 



"*7 



•15 



-I 4 



• r 5 



-I 5 



,J 7 



•19 



•20 



•19 



•18 



No; 6702. 



-I 3 



•13 



"I I 



09 



•08 



•09 



•10 



•I I 



II 



•11 



•11 



11 



•11 



•11 



•11 



•12 



-I 3 



•14 



•*5 



-I 3 



11 



No. 6703. 



•09 



•09 



•IO 



•12 



•13 



•16 



•18 



•20 



•I 9 



•18 



•18 



•16 



•13 



•10 



•08 



■07 



•08 



•09 



■10 



•10 



•09 



No. 6704. 



•09 



•09 



•IO 



•10 



•11 



•13 



•16 



•I 9 



•17 



■15 



-I 3 



•13 



•12 



12 



12 



•12 



•12 



•12 







No. 6982. 



•24 



•24 



•23 



•22 



•21 



•22 



•24 



•25 



•25 



•25 



" 2 5 



•25 



•26 



•27 



•28 



•28 



•30 



•32 









7. The thermometers were placed in the receiver, arranged close to each 

 other on a board fixed to a support, the four smaller thermometers on one 

 side, the two larger ones on the other ; and the manner of proceeding with 

 each experiment was the following. Before pumping, all the thermometers 

 were twice read in rapid succession ; after exhausting the receiver to be- 

 tween one and two inches of pressure (a manipulation which generally 

 lasted about ten minutes), two or more readings were again taken to deter- 

 mine the " immediate effect of exhaustion " on each thermometer. After 

 an interval of several hours the thermometers were supposed to have 

 assumed the surrounding temperature, and two readings were now taken 

 for the " residual effect of exhaustion." The whole apparatus was then 

 left undisturbed for nearly a whole day, when another set of readings were 

 taken, and the apparatus was refilled. After readmission of the air the 

 temperature shown by the instruments was immediately registered to find 

 the heating-effect upon them of the inrush of air. 



The readings, both for the comparison of the instruments and during 

 the experiments themselves, were taken alternately by Mr. Thomas Baker, 

 Assistant at the Kew Observatory, and myself. By the kind permission 

 of Mr. Balfour Stewart, Superintendent of the Kew Observatory, I was 

 enabled to avail myself of the obliging assistance of Mr. Baker and his 

 great experience in thermometric experiments. I take this opportunity of 

 expressing to both these gentlemen my gratitude for the aid given to me 

 in the pursuit of this inquiry. 



