- ATMOSPHERIC PRESSURE BETWEEN 104 C. AND 105" C. 417 



referring to the main details and arrangement of the calorimetric apparatus, and 

 to the results obtained. All the results quoted have been calculated in the same 

 way as in the final experiments. 



(a) A few experiments were made with the apparatus used by SWANN for air 

 and described by him in his paper. In this arrangement bare wire thermometers, 

 wound on mica frames, were used. No satisfactory results were obtained owing 

 to the condensation of steam upon the mica frames and the consequent impairment 

 of their insulating properties. 



(6) The apparatus was dismantled and set up again with a single thermometer 

 (that used in the final experiments) enclosed in a glass sheath passing centrally 

 along the flow-tube and carried by a rubber cork fixed in the outflow end of the 

 tube. The stirring arrangement consisted of the pinched spiral of compo-tubing, 

 to which reference has been made. The main flow-tube was fixed vertically, as 

 is indicated in fig. 7, with the non-vacuum jacketed sheath. 



The current and potential leads entered the flow-tube in the manner indicated 

 in this figure, but they were carried by a cork (see Sw ANN'S paper, fig. l) fitting 

 into the top of the outer brass jacket, and were not bent round and carried 

 down outside the flow-tube as in fig. 7. The disc, K, had not been soldered 

 across the jacket at this time, but its function was performed by a rubber bung 

 which fitted tightly to and was carried by the current and potential leads. The 

 initial temperature of the steam was about 103 C., and the three flows varied 

 from 0'57 to 0'14 gr. per second. The values of the constants deduced were 



S c at 103 C. = 2'021, h = G'0400, k = 0'00400. 



In these experiments the heating coil had a resistance of about an ohm, and 

 the standard resistance used as a comparison was 1'039 ohms at 19 0. The 

 temperature variation of the specific heat was about 0'0035 per degree, a value 

 three times as great as that found in the later experiments. Later, the separator 

 was re-designed, and in all the subsequent experiments it was of the form shown 

 diagrammatically in fig. 7. A new heating coil and a new standardised coil of 

 about 5 ohms resistance were made and used in all the experiments made 

 afterwards. In addition, the regulator was altered to give a " cold temperature " 

 of 104'5C. instead of 103 C. The effect of these changes was to reduce the 

 temperature coefficient of the specific heat to about O'OOIO joules per degree. 



(c) Many experiments were made in which the temperature of the steam was 

 measured by means of a bare wire thermometer consisting of about 3 cm. of 

 0'001-inch platinum wire. Spiral mixing was impossible, and in consequence the 

 gauze method of mixing was employed. Measurements of the heating effect of the 

 current in the thermometer were made. It was found that although this heating 

 effect was slightly different in the different flows, it was the same on both hot 

 and cold temperatures observed with any one flow, and no systematic error was 



3x2 



