46S Barometric Observations. 



Mr. Cornfield's name, which by mistake you have made Cornfield 

 in last Number, p. 398. 



Upon examining the construction of the best portable baro- 

 meters, I have noticed one probable source of error in the posi- 

 tion of the attached Thermometer which is fixed near the basin 

 of mercury, and possibly may give nearly the heat of that portion 

 of the mercury. But when it is considered that all fluids com- 

 municate heat very slowly downwards, and that much of the 

 upper part of the tube is exposed to the atmosphere, and may 

 become several degrees warmer than the quantity in the basin, 

 or the wood in which the attached thermometer is fixed, it 

 may cause the average heat of the mercury in the tube to be se-^ 

 veral degrees above that in the basin : but in the necessary cal- 

 culations respecting the relative height of two places, it is the 

 heat of the mercury in the tube that forms one part of the data, 

 and not the temperature of that in the basin. 



On the 1 1th instant, I suspended two thermometers near the 

 centre of my tube, one on each side, and found a difference of 

 4 to 5° between the heat in that part, and the heat indicated by 

 the attached thermometer : it is well known that a difference of 

 4 or 5° will frequently affect the result of a calculation ten or 

 twelve feet, and may perhaps have caused some of those ano- 

 malies, which may in part be avoided by having the attached 

 thermometer so fixed as to give the mean heat of the mercury in 

 the tube, instead of that in the basin. 



I am not so sanguine as to imagine the improvement above 

 suggested will reconcile all the differences found; but it is a maxim 

 with me to correct an error when 1 find it, and view it as one 

 step permanently gained. 



The customary correction for the heat of the atmosphere, as 

 pointed out by the detached thermometer, I have long viewed 

 with doubts. 



I have calculated the relative heights of Crumpsall and Leigh- 

 ton, exclusive of fractions, to be nearly as follows: 

 Crumpsall above LfiglUon. 

 Feb. .. 174 feet. 

 March.. 161 

 April ..159 

 May . . 123 through the medium of Bushy. 



Col. Beaufoy has calculated the height of his instrument above 

 that of Mr. Gary's in February 487*4 feet. 

 March 481-0 

 April 487'6 

 May 480-5 



484*1 mean of 4. 



^ Mr. 



