364 On Storms and Meteorological Observations. 



results. It is not meant to be asserted that mean results are of no 

 value, but we must not hope to deduce many of the great laws of na- 

 ture from them. This were much as if a chemist should set down 

 in a table all the acid, alkali, alcohol, etc. consumed in his laboratory 

 in the course of the year, with all the resulting substances of what- 

 ever kind, and undertake to derive from these data the great funda- 

 mental doctrines of his science. 



No record of the height of the mercury in the barometer can be 

 of value unless the instrument is both carefully purged of air and 

 filled with mercury whose specific gravity has been correctly ascer- 

 tained. It is somewhat remarkable, when we consider the many la- 

 borious observations that have been made in various parts of the 

 world, the plans that have been proposed for clearing the tube com- 

 pletely of air, and the contrivances added for reading off the altitude 

 to the thousandth part of an inch, that the constant and perhaps ne- 

 cessary uncertainty of the standard itself should have been so much 

 neglected. The mercury employed by Mr. Daniell in filling the 

 tube of the barometer recently made for the Royal Society, was 

 found by Mr. Faraday to have a specific gravity of 13.624 at 40° 

 of Fahrenheit's thermometer, the temperature of the water being the 

 same — or of 13.609 at the temperature of 60°. He states the 

 mean height of the barometer in London at 29.881 inches. But 

 Dr. Henry in the last edition of his Chemistry makes the specific 

 gravity of mercury 13.545 at 47° of Fahrenheit, or 13.538 at 60°. 

 If Daniell's mean altitude was obtained with the same metal, or with 

 a metal of the same specific gravity with that employed in filling the 

 Royal Society's barometer ; the mean height with Dr. Henry's nler- 

 cury would be 30.038. But Dr. Thomson, remarking that the spe- 

 cific gravity of mercury " varies considerably like that of all other 

 metals,''^ says that he once obtained it as low as 1 3.4228. This would 

 stand according to Daniell at 30.295. We have therefore the fol- 

 lowing numbers : 



Height of the barometer in London as deduced by Luke Howard 

 from the register kept at the rooms of the Royal Society, a mean 

 of twenty years, _ _ _ - 29.8655 



Height stated by Daniell — supposed specific gravity 

 of the mercury, 13.609, - - - 29.881 



Height with Dr. Henry's mercury — same pressure, 30.038 

 Height with Dr. Thompson's mercury, - - 30.295 



Uncertainty depending upon the variable specific 

 gravity of the mercury, _ - - - =4295 



