430 



THE FIGURES OF EQUILIBRIUM OF A LIQUID MASS 



diameters. During the continuance of tlae operations the temperature varied 

 from IS^.5 to 20=. 



The general mean of the products is 22.75, and it will be seen that, except 

 in the last two, the departures from this general mean are inconsiderable, and 

 that, moreover, they are irregularly distributed. Further, as the first diameter 

 is to those of the last group very nearly as 1 to 6, these results suffice, 1 think, 

 to establish distinctly the constancy of the product pd, and consequently the 

 law according to which the pressure is in the inverse ratio of the diameter. In 

 the next series we shall see this law verified by experiments of a wholly 

 different kind. 



I should say here that, in the measurements relating to the smallest bubble, 

 that, namely, of 7™'". 55 diameter, I have been constrained to make a slight excep- 

 tion to the procedure indicated at the end of the preceding paragraph : the second 

 measurement of the pressure exceeded the first by 0'"'".02 ; it was decided 

 therefore to take a third measurement after a new interval of five minutes, but 

 during this time the bubble burst. The attempt was several times made to 

 renew the experiment, and in each instance one or other of the causes which I 

 have noticed in regard to very small bubbles interfered with the success. As 

 the difference 0'""\02 was so minute that it might be attributed to an error of 

 observation ; as, moreover, by reason of this smallness, it was very improbable 

 that a new excess Avould be manifested in a third measurement ; as lastly, with 

 a diameter of that order, such faint differences have no influence except on the 

 decimal part of the product, I felt authorized to consider the second measure- 

 ment as giving the value of the pressure, and to retain the result of the experi- 

 ment. As to the general mean 22.75 of the results of the table, its decimal 

 part is necessarily a little too high, on account of the excessive value 26.45 of 

 the last product. As this product and that which precedes it are, as has been 

 already remarked, those which alone deviate materially from 22 in their integral 

 part, it will be admitted, I think, that a nearer approach to the true value will 

 be made by neglecting these two products and taking the mean of the others, 

 a mean which is 22 5C, or more simply 22.6 ; we shall adopt, then, this last num- 

 ber for the value of the product prZ in regard to the glyceric liquid. 



§ 29. It remained to be verified whether this value satisfied our formula, 

 according to which we have 2^d=2hp, the quantities p and k being respect- 

 ively, as has been seen, the density of the liquid and the height which this 

 liquid would attain in a capillary tube one millimetre in diameter. "With tjus 

 view, therefore, it was necessary to seek the values of these two quantities in 

 reference to the glyceric liquid. The density was determined by means of the 

 areometer of Fahrenheit, at the temperature of 17°, a temperature little inferior 

 to that of the preceding experiments, and the result was /)=1.1065. To 



