CORRECTIONS HOW OBTAINED. 



these corrections have been applied, as they are doubtless 

 sufficiently accurate for all practical purposes. 



For the Dry-Bulb Thermometer the table of corrections 

 obtained from the Greenwich observations would not satisfy 

 the observations made at Liverpool, but as this correction 

 appeared to depend on the diurnal range of the thermometer, 

 a table of factors was formed by dividing the corrections given 

 in the table for each hour at Greenwich, by the diurnal range 

 of the thermometer at Greenwich. These factors when multi- 

 plied by the diurnal range of the thermometer at Liverpool, 

 give corrections which satisfy the Liverpool observations. 

 We can prove the truth of this in the following manner: — The 

 mean of the readings of the dry-bulb thermometer taken daily 

 at one o'clock should, when corrected for diurnal range as 

 above described, give the true mean temperature. Again, 

 the mean daily readings of the maximum and minimum ther- 

 mometers, after the application of a correction found from 

 the Greenwich observations and given in the above-named 

 tables, should also give the true mean temperature. ' 



Now, for each of the five years over which our observations 

 extend the mean temperature obtained by the last named 

 method has been in excess of the former, but the difference 

 has never exceeded 0*4°, and it has never been less than 0*1°. 

 The correction for diurnal range found in this way may 

 therefore be considered sensibly perfect, for the index cor- 

 rections which have been applied to the thermometers, not- 

 withstanding the care that has been taken to obtain them 

 correctly, may possibly be in error sufficiently to cause a 

 difference in the mean temperatures derived by the two 

 methods, as large as that which has been found to exist. 



The corrections for diurnal range which have been applied 

 to the Wet-Bulb Thermometer have been found in the foUowing 

 manner: — Li the tables before named (Transactions Koyal 

 Society, part 1, 1848), Table V. shews the mean depression 

 of the temperature of evaporation below that of air at Green- 

 X 169 



