434 



THE FIGURES OF EQUILIBRIUM OF A LIQUID MASS 



absorbed aqueous vapor, and this absorption, which is explained by the 

 hygrometric propei-ty of the glycerine, had continually repaired the diminution 

 of thickness due to the descent of the liquid. Now, in the present experiment, 

 as there was no water in the jar, and hence the atmosphere which surrounded the 

 bubble was less humid, it might be supposed that the effect of the descent of. 

 the liquid had at first prevailed over that of the absorption, and that afterwards 

 it was the contrary which took place. Upon this assumption, I deposited in 

 the bottom of the dry jar morsels of caustic potash, and contrived by the appli- 

 cation of a little lard ai'ound the orifice of the jar and of the aperture through 

 which passed the copper tube, that, after the introduction of the bubble, the 

 pasteboard disk should close the opening hermetically. Further, as the small 

 quantity of liquid which always accumidates by degrees at the base of the 

 bubble must contribute by its weight to cause a rupture of the latter, a space 

 of ten minutes was allowed to elapse before the removal of the drop ; the 

 film had thus already become thinner when it was introduced into the jar, and 

 the accumulation resulting from the further descent of the liquid would be much 

 less. Now, under these conditions, the diminution of thickness of the film was 

 continuous, the bubble lasted for nearly three days, and when it burst, it had 

 arrived at the transition from the yellow to the white of the first order; it then 

 presented a central space of a pale yellow tint, surrounded by a white ring. The 

 level of the water in the observed branch of the manometer underwent small 

 oscillations, sometimes in one direction, sometimes in the other, but the last of 

 them was in the direction of an augmentation of pressiu-e. Although, dm-ing 

 the long duration of this bubble, the temperature of the apartment had neces- 

 sarily undergone slight changes, the oscillations just mentioned could not be en- 

 tirely attributed to these, for, if that had been so, there would have been seen, 

 after each of the three nights, a movement of the manometer in the direction of 

 an augmentation of pressure ; but the contrary was observed after the first 

 two nights, and it was only after the third that there was a movement in that 

 direction. From the progression of these movements it results that if the pres- 

 sure varied, it was in an irregular manner, in both directions, and terminating 

 not in a diminution, but an augmentation at least relative ; we may therefore 

 admit, I think, that the final thickness of the film was still superior to double 

 the radius of sensible activity of the molecular attraction. 



§ 33. Let us now see what we may deduce from this last experiment. Ac- 

 cording to the table given by Newton, the thickness of a film of pure water 

 which reflects the yellow of the first order is, in milliouths of an English inch, 

 f>J, or 5.333, and for the white of the same order 3|, or 3. 875. We may 

 therefore take the mean, namely 4.064, as the closely approximative value of 

 the thickness corresponding, at least in the case of pure water, to the transition 

 between those colors and the English inch being equal to 25™™. 4, this thick- 

 ness is equivalent to R-Jg-j of a millimetre. Now we know that, for two dif- 

 ferent substances, the thickness of the • films which reflect the same tint is in 

 the inverse ratio of the indices of refraction of those substances. In order 

 therefore to obtain the real thickness of our film of glyceric liquid, it sufiices 

 to multiply the denominator of the preceding fraction by the ratio of the index 

 of the glyceric liquid to that of water. I have measured the former approx- 

 imately by means of a hollow prism a liquide's, and have found it equal to 

 1.377. That of water being 1.336, there results, for the thickness of the gly- 

 ceric film, g-g'yi of a millimetre. The half of this quantity, or yyIts ^^ ^ ^^^' 

 limetre, constitutes therefore the limit furnished by the experiment in question. 

 Hence we arrive at the very probable conclusion, that in the glyceric liquid the 

 radius of sensible activity of the molecular attraction is less than lYhoo ^^ ^ 

 millimetre. 



I had proposed to continue this investigation with a view to reach, if possible, 

 the black tint, and to elucidate the valuations of the manometer; but the cold 



