140 On the Contraction which takes place in Mercury 



at 65 the mercury loses 318'5 less by 0-5 



62 — — 3iy _ 



50 — — 320 more by 1 



45 — ., — 320-5 — 1-5 



40 — — 321 — 2 



36 — — 321-5 — 2-5 



^4 — — 322 — 3 



27 — — 323-5 — 4-5 



14 — — 324 — 5 



5 ^ — 324-5 -— 5-5 



O — — 325-5 — 6-5 



5 below zero 326*5 — 7*5 



J8 — — 327-5 — 8-5 



21 — — 328 — 9 



We may observe in this experiment that some difference is 

 cJccasioned by the direction in which the heat passes, 

 whether from without to the alcoliol, and thence to the 

 mercury, as in the first part of these observations from 

 45® to 135^; or from within from the mercury to the al- 

 cohol, as from 135^ lo zero and 21'' below : and it appears 

 that the temperature was not the same in these two fluids 

 when the observations were made j for when I weighed the 

 same mercury with a hole in the bladder to introduce the 

 bulb of a thermometer, that I might observe the variation 

 of loss of weight when the alcohol and mercury were at the 

 same temperature exactly, it showed the mean of these va- 

 riations to be correct. So that in comparing 52/6 parts of 

 mercury with alcohol, there is a variation in the contraction 

 of these two fluids expressed by 22 of loss of weight in 

 passing through 156 degrees of temperature. In comparing 

 1000 parts of mercury with alcohol, it follow:^ that there is 

 a variation of loss of weight in 156 degrees of temperature 

 expressed by 4' 17, which shows -0267 is the loss of alcohol 

 in each degree greater than that of mercury. 



If Uiese two fluids were to contract by deprivation of heat 

 in the ratio of their specific gravities, it is obvious tliat the 

 Joss of weight shown by the hydrostatic balance would be the 

 same throujih all the changes of temperature, the changes 

 Ibting the &amc in both fluids at the lime of obscrvatfon. 



It 



