400 M. Descloizeaux's Physical and Geological Observations 



Thus, notwithstanding the relative lightness of the most 

 heated water, which tends to rise to the smface, we see that, 

 even after the eruptions, the temperature of the different strata 

 does not become equalized ; still further, in proportion as the 

 inferior strata become heated, to reach their maximum, the 

 upper strata become cool, uiidoubtedlj' in consequence of 

 evaporation, and of the heat communicated to the sides of the 

 well and the basin ; so that the greatest difference of tempera- 

 ture at the two extreme levels corresponds precisely to the 

 moment of the eruptions. 



We can now ascertain what would be the boiling-point of 

 water submitted to a pressure equal to that of the atmosphere, 

 augmented by the liquid column the height and mean tem- 

 perature of which we have determined in our five experiments. 

 For this purpose, we must first express the pressure of this 

 liquid column in millimetres of mercury at (f. The density 

 of the water of the Geysir, found at from TOOIO to 10° cen- 

 tigrade; the curve which represents the results obtained by 

 M. Isidore Pierre for the dilatation of distilled water, and 

 which I admit as being applicable to the water of the Geysir, 

 extending it up to 113°; lastly, the density of the mercury 

 at 0°, compared to that of distilled water at 4°, and found at 

 13*596 by M. Regnault, enables us to make this reduction. 

 We thus find that the heights of the columns of water in our 

 different experiments correspond, in millimetres of mercury at 

 0°, to the following numbers : — 



m 



Third and fourth experiment . . 1*645 



Fifth experiment 1*593 



First experiment 1*587 



Second experiment 1*654 



From the equations 



loge=a — ^a% x=t—\QO°i 



and from the data 



log a=T-997 7641, log i = 0*4692291, « = 5*8267890, 



laid down by M. Regnault, in which e represents the tension 

 of steam at saturation, expressed in millimetres of mercury, 

 and X the temperature to which that pressure corresponds, 

 we can derive the value of x and of /; and this last quantity 

 will precisely represent the boiling-point of water submitted 

 to the pressures 



0*760 + 1™-G45, 0*760+ r"-593, 



0*760 + l'»*587, 0*760 -f- 1™'654. , 



