72 French National Institute. 



probable, was well worthy of being established by direct cxt 

 periments. This is what M. Biot undertook last siimnjer, 

 with the most delicate accuracy. He had in the first place 

 to determine the efTects of vapour by itself: he dried, by 

 means of potash, the warm air continued in his prism; 

 outside of it he had air charged with the natural moisture of 

 the atmosphere. The pressure of these two airs, indicated 

 by the interior and exterior barometers, was not the samej 

 the difference was equal to the tension of the aqueous va- 

 pour of the atmosphere. The deviation of the ium^inous ray 

 rn the prisni then gave the refraction produced by the vapour; 

 and we might see whether this refraction differed from that 

 which would have been produced by the air alone at a similar 

 temperature. The difl'eieiices never rose above some tenths of 

 a second, and the medium was only o"\b, a quantity truly 

 insensible, since it only produces a sixtieth of a second at the 

 height of 45 degrees. M. Biot hence concludes '^ that the 

 vapour of the air sensibly refracts like the atmospheric air j 

 and thus, in astronomical observations, we should be con^ 

 tented with having in view the height of the barometer and 

 thermometer, and neglect the vapours more or less witb 

 which the atmosphere may be charged." 



The first experiments of M. Biot were made in winter and 

 jn low temperatures: the last took place in the greatest heat 

 of summer, and yet the difference upon the mean refraction 

 only differed by an extremely small quantity, which is still 

 jnade less by M. Delambre's result. To concUide: all aslro- 

 jiomers will easily agree, that the direct observalic ns of the 

 refractions could not give, notwithstanding all possible care, 

 cither the same agreement in the particular results, or 

 the same precision in the absolute quantity ; because, by 

 jhe astronomical methods, this value, or the constant part 

 of the refraction, is always dependent upon the altitude of 

 the pole, because we can only determine simultaneously the 

 two unknown quantities ; and we niay always, by causing 

 small changes in a contrary direction, represent the observa- 

 tions equally well. M. Deiambre has declared that he can- 

 not account for the small difference which exists between his 

 jefraclions and those of M. Biot. Wc may therefore adopt 



in 



i 



