Oscillation of the Barometer. 



319 



Changes of the Meteorological Elements during the daily 

 barometric periods. 



10. We have here a clear comparison of the theory of the 

 semidiurnal oscillation as generally received, with that condition 

 of it to which my arguments would reduce it. If it were true, 

 as it probably is not, that the temperature of the air acts in- 

 versehj, and the elastic force of vapour at the surface of the 

 earth acts directly upon the barometer, the motions of the latter 

 would be to some extent at least explained. But it is the num- 

 bers in the last column of the Table which truly express the 

 whole eflfect of the daily variation of humidity upon the baro- 

 meter. They represent the height of a column of mercury, of 

 which the pressure is equivalent to that of the water which eva- 

 porates during each period of the day. It is evident that not 

 only are these numbers quite insignificant in amount, but that they 

 also bear no apparent relation whatever to the barometric oscil- 

 lations. 



The humidity of the air is, as a general rule, always in- 

 creasing by evaporation from the surface of the earth, and the 

 barometer therefore has a tendency to be continually rising. But 

 by far the largest amount of evaporation is observed to take jjlace 

 between 9 a.m. and 4 p.m., during which time the barometer 

 falls as much as eight-hundredths of an inch. Neither can the 

 upward motion of the barometer from 3 a.m. to 9 a.m. and from 

 4 P.M. to 10 P.M. be explained by an increase of humidity, be- 

 cause the real amount of this increase is so minute. It is true, 

 indeed, that during the night dew is often deposited, the ten- 

 dency of which would be to produce a minimum of the baro- 

 meter ; but the average amount of dew, I consider, is altogether 

 inadequate to this effect in England or any other country. 



11. The prevalent theory, that the semidiurnal barometric 

 oscillations are partially owing to the variations of humidity, 

 seems to have been derived from the practice of separating the 



* Dew was often detected during this period to the extent of a few 

 thousandths of an iiicli ; but as tiie opjjosite process of evaporation some- 

 times proceeded shnvly throuj.;hout the night, tlie dew does not appear in 

 the average resuh. 



