416 ME. J. H. BKINKWOETH ON THE SPECIFIC HEAT OF STEAM AT 



obtained, thus correcting them to' the middle of the interval in order to correspond 

 to the time of average flow. The potential readings were not taken as frequently as 

 the thermometric readings, this being unnecessary, since the latter were irregular 

 whilst the former gradually changed in one definite direction ; they generally showed 

 a slight fall. The electrical energy supplied per second was determined from the 

 mean values of the potentiometer readings ; the necessary temperature corrections 

 for the resistance of the standardised coil and the E.M.F. of the cadmium cells having 

 been previously made. The mean thermometric readings give the temperatures 

 of the steam when hot and when cold, and from other measurements the rise of 

 temperature due to the heating effect of the current in the leads connected to the 

 heating coil can be obtained. All these temperatures are expressed in terms of 

 centimetres of the bridge-wire, the calibration corrections of the bridge itself being 

 included. 



The rise of temperature due to the passage of the current through the leads is 

 subtracted from the observed "hot temperature," the value thus obtained and the 

 mean " cold temperature " are then reduced to platinum temperatures, and these 

 again to temperatures measured on the air scale, the latter reduction being made by 

 means of the well-known parabolic formula tpt = 1'50 x I0~ 4 t (t 100). The flows 

 given are corrected to correspond to weighings in vacuo. 



The number of flows taken for a complete experiment was always three, each flow 

 being repeated twice. The concordance between the values of CE/Qc0, calculated 

 from two consecutive flows of the same magnitude, gives a criterion of the accuracy 

 of the experimental measurements. Whenever possible, three flows were taken on 

 the same day, as it was thought that the heat-loss might vary from day to day ; 

 also the order of the flows was usually medium, minimum, maximum, in order to 

 reduce the time taken between experiments made with different flows. For when 

 the flows are taken in this order, the initial heating up is fairly quick, and less time 

 is lost in refilling and reheating the boiler between successive flows. Hence there is 

 no long interval during which steam is not passing round the outer jacket, and the 

 calorimeter does not cool to any very great extent between consecutive experiments. 

 Even when taking the flows in this order it was impossible to complete the three 

 experiments in less than eight hours. 



Preliminary Experiments. 



A long series of preliminary experiments was made in order to find the best 

 method of mixing the steam, to investigate the relationships between the heat- 

 loss and the flow, and between the heat-loss and the distance separating the 

 thermometer from the heating coil. Several different arrangements of the 

 apparatus were made, and various methods of stirring employed. It will suffice 

 to summarise these experiments in the order in which they were carried out, by 



