326 Mr. B. Loewy on the Behaviour [Mar. 4 



Immediate effect of refilling the 'Receiver. 



Thermometers rising. 



Experiment I. 



II. 

 HI. 

 IV. 



v. 



VI. 



Means 















A. 



A'. 



B. 



B'. 



c. 



c. 

























r 97 



1 '24 



1-69 



1-28 



1 '47 



I'OI 



1-58 



0*62 



1 '74 



o-86 



I"22 



o-68 



1*64 



0S2 



i-8i 



o-88 



I-30 



o-68 



1-47 



076 



177 



0*76 



I'20 



o-68 



1-41 



0-51 



1*69 



o-68 





c> 55 



1*64 



o'6j 



2-08 



0-94 



178 



075 



i*6a 



O'JO 



1-63 



0*90 



"V* ' ' 



1-29 



073 



Differences immediately 1 - 

 observable . J °'9 2 °'73 0-56 



10. Now if this immediate difference would entirely disappear after some 

 time (say, after a number of hours, or a whole day), or would become so small 

 as to be within the limits of experimental errors, the question whether a 

 vacuum-correction is necessary would have to be negatived. "We may 

 presume that after some time the refrigeration caused by the exhaustion 

 disappears, and that tire thermometers are then solely or chiefly influenced 

 by the temperature of the surrounding air ; if then a difference still appears 

 in the behaviour of the thermometers, this permanent difference must 

 obviously be caused by something independent of the temperature, and its 

 source must be looked for in a change of the instruments themselves. 



The thermometers C, C (that is, those with unusually large spherical 

 bulbs and long stems) differed in their behaviour entirely from the others 

 of common form and size ; they will be spoken of afterwards. 



The thermometers A, A', B, B'' exhibited, on the contrary, as the follow- 

 ing Table shows, constant differences when read from three hours to as 

 long as two days after exhaustion. A, A', B, B' signify in this Table the 

 readings of the corresponding thermometers taken so long a time after 

 exhaustion as to exclude all possibility of introducing the effect of it. 

 The Table gives only the differences, the readings themselves are given in 

 Table II., with the times at which they were taken. 



*A'-A. B'-B. 



_ o o 



Experiment 1 078 0*47 



072 072 



„ II 0*40 0*48 



0*48 076 



„ III 071 , 0-48 



070 075 



IV 071 0-44 



0-29 078 



V 072 0-46 



070 070 



„ VI. , 21 .. .-. . 073 



070 0*46 



0-25 076 



Mean difference , 077 C40 



