1844.] Hygrometric Tables. 137 



In the above equations, however, the value of the co-efficient (m) 

 depends upon the assumed values of a and e which, Dr. Apjohn re- 

 marks, are in all probability not yet known with great precision, and 

 accordingly he proceeds to deduce values for the co-efficient (m) in 

 the general equation directly from experiment in three separate ways 

 as follows; viz. 



1. By observations in air, in reference to which t and t' had been 

 accurately noted, the temperature of which was afterwards raised and 

 the observations repeated; the value of/" is here constant for both 

 observations. 



2. By observations of t and t' in perfectly dry air where the value 

 of/" is of course == o. 



3. By observations in air saturated with moisture, where/" is ob- 

 tained from a simple observation of the temperature, and in which 

 after its temperature has been raised, the values t and t' were observed. 



From the above experiments, using Anderson's Table of the elastic 

 force of vapour, Edinburgh Encyclopedia, Art. ' f Hygrometer," three 

 separate values of (m) are obtained ; viz. 



1st Series, 11 observations m = '01151 



2nd Ditto, 19 ditto -01150 



3rd Ditto, 24 ditto -01140 



The Arithmetical mean of which is .. .. g=qg or -01147 



The most probable value of the final mean, as deduced by Professor 

 Lloyd, according to the calculus of probabilities, from the means of all 

 three values of m is *01145, but since in the second series the result is 

 affected by the full tabular error in the value of/ whereas in the 1st 

 and 3rd series as m is expressed in terms of the difference of two 

 values of/, the tabular error will not sensibly affect the result. Pro- 

 fessor Lloyd considers that the secoifd series should be omitted, and 

 combining the results of the 1st and 3rd series by the same method, 

 obtains for the value of m *01 145. 



As my object in the construction of the accompanying Tables has 

 been solely to enable observers to apply Dr. Apjohn's formula, without 



