208 ME. W. F. G. SWANN ON THE SPECIFIC HEATS OF AIR AND 



standard alone until its temperature had risen about 5 C., this being about the 

 maximum variation in temperature experienced by the coil during the research, 

 and the coils were again compared. The results of several observations, full 

 details of which have been preserved in the archives, lead to the expression 

 E = 1-03899 + 0-000084 (-187) ohms for the resistance of the secondary standard in 

 terms of its temperature t. 



(10) The Pressure Regulator. The pressure regulator was of the form designed by 

 Prof. CALLENDAK. It consisted essentially of an inverted bell (a, fig. 2) weighted at 

 the bottom, and suspended in oil by a steel tape from a wooden beam b, which was 

 supported on friction wheels, and which formed one arm of a balance. From the other 

 arm a support (c) for weights was hung by another steel tape. Gas could be blown 

 into the bell either directly through pipes, or through the fixed inverted bell shown 

 at e in the figure. So long as the suspended bell was floating the pressure of the 

 gas inside it was constant, and independent of its degree of immersion, except to a 

 very slight extent, owing to the buoyancy effect on the thin copper of which the bell 

 was made. The buoyancy effect was easily overcome by attaching a clamp (/c) 

 asymmetrically to one end (li) of the beam, and adjusting its asymmetry so as to 

 automatically effect the compensation. 



It was most essential that the oil should contain no volatile constituent. To test 

 this point, the full current of gas was passed over a weighed quantity of oil for half 

 an hour, and the oil was reweiglied. In this way it was possible to form an estimate 

 of the amount of evaporation which would take place in the bell. An evaporation of 

 oil amounting to one part in 10 5 by weight of the total quantity of gas passing over 

 the oil could easily have been detected, but no alteration in weight was found. 



(11) The Oil Gauge. The gauge which served to measure the pressure difference 

 between the ends of the fine tubes contained Fleuss pump oil. It consisted of a 

 U-tube graduated in millimetres on both limbs, and was fixed to a board which was 

 clamped to a massive cast-iron support in order to keep it firmly fixed throughout the 

 work. The fractions of a millimetre were obtained by two microscopes, with scales in 

 the eye-pieces ; by removing the front lenses of the objectives it was easy to place the 

 microscopes so far away from the limbs of the gauge that the oil menisci and the 

 graduations were in focus at the same time. The gauge graduations were carefully 

 calibrated by means of a small cathetometer giving a fairly large magnification. Two 

 calibrations were performed in which no two corresponding readings differed by more 

 than 0'04 mm., and the agreement was generally to 0'02 mm. Full details of the 

 calibrations have been preserved. In order to measure the pressure difference between 

 the ends of the fine tubes, taps T 2 and T 3 (fig. 2) were closed, tap Tj being open. By 

 closing TI and T 3 and opening T 2 the excess pressure at the high-pressure ends of the 

 tubes over that of the- atmosphere was obtained. T 3 was to enable the pressure in the 

 bell to be measured if required. The mean of the pressures at the two ends of the 

 tubes was obtained by subtracting half the pressure difference between the ends from 



