272 Dr Andrew Fyfe on the Comparative Value of 



extremity as to quality, made it appear, that there was some 

 deviation from the rule I have laid down.* 



From the foregoing experiments, it is evident that the du- 

 rability of a gas can be easily ascertained, by using a burner 

 with an aperture of a certain diameter, having a scale for 

 measuring the height of flame, and a gauge for observing the 

 pressure, at which the gas is burning, at that height of flame ; 

 of course taking care to make the measurement with accuracy. 

 The consumpt of any gas by this burner, with the height of 

 flame fixed on, is first to be determined by metx'e, numerous 

 trials being made ; while, at the same time, at each trial, the 

 pressure is also noted. By this means, and taking the ave- 

 rage of alltheconsumpts and pressures, a standard is obtained. 

 Suppose that with a flame of 5 inches, and a pressure of 

 rVffths at the burner, the consumpt of a gas is, by metre, 

 found to be one foot in 65 minutes ; and that another gas 

 requires, with the same bui'ner and height of flame, a pres- 

 sure of iVoths ; the square roots of these pressures are 

 8560254 and 9695359 ; then, as the latter is to the former 

 so is G5, the time for the consumpt of the latter to 57"21, 

 which is the time for the consumpt of the former. Now, 

 8560254 X 65 = 5564165; accordingly, this number divided by 

 the square roots of the pressures necessary for the consumpt 

 of other gases, will give the times for the consumpts of these 

 gases, when consumed with the same burner, and with the 

 same height of flame. 



The jets now in general use, vary from the 28tli of an inch, 

 as recommended by Christison and Turner, to the 45th of an 

 inch. Most of the trials which 1 have now recorded, were 

 made with a jet of the 33d of an inch ; not because I prefer 

 it, but because I had begun my investigations with it, without 

 being aware of the practical applications to which the results 



* Since writing the above, I have fortunately had an opportunity of pro- 

 curing the coal alluded to, and of jjreparing gas from it. All the trials were 

 i-epeated with great care, and were found to give results corresponding with 

 those in the above table, so that there must have been an error in noting the 

 results of the first trials. In the subsequent trials, repeati^d again and again, 

 the consumpt by metre and calculated by pressure differed only in 40 seconds. 



