412 MR. J. H. BRINKWORTH ON THE SPECIFIC HEAT OF STEAM AT 



the heating-coil and the thermometer. Until agreement in the calculated values of 

 the specific heat obtained under the different experimental conditions, and a definite 

 relationship between the values of the heat-loss and the distances separating the 

 heating-coil from the thermometer are obtained, it cannot be assumed that the mixing 

 is adequate. The methods of mixing employed may be conveniently referred to as 

 (a) gauze mixing ; (b) spiral mixing. The method (a) was employed by SWANN in 

 his experiments on the Specific Heat of Air and Carbon Dioxide, and has been used by 

 other observers in their determinations of Specific heats by the continuous-flow method. 

 Circular discs of copper gauze are cut with a diameter slightly greater than the 

 internal diameter of the flow-tube. These discs are then bent just to fit the flow- 

 tube, each disc having a small flange. The space between the heating-coil and the 

 end of the thermometer tube is then filled with these discs, and since any one disc 

 only touches the flanges of the discs on either side of it, longitudinal conduction from 

 disc to disc is necessarily very small. At the same time, in virtue of the high thermal 

 conductivity of copper, a uniform distribution of the heat throughout the gas, as it 

 passes from the heater to the thermometer, might be expected. This method of 

 mixing was employed in a large number of experiments extending over a period of 

 more than two years. In some experiments, discs were also fitted over that part of the 

 thermometer tube surrounding the ivory cylinder ; in other experiments the ther- 

 mometer previously described was replaced by a bare-wire thermometer consisting of 

 a single loop of 0'001-inch platinum wire. Although, in most cases, very good agree- 

 ment between the results of experiments under any given conditions was obtained, 

 no such agreement was found between these results and those obtained under different 

 conditions. A further consideration of the results obtained with this form of mixing 

 will be found later. 



(6) Spiral mixing. A very efficient method of mixing the steam consisted in 

 tightly wrapping around the thermometer tube a spiral of small bore compo-tubing, 

 . the walls of which were then pinched together to form a screw thread closely fitting 

 the flow-tube and with a pitch of about 1 cm. The only apparent disadvantage of 

 this was that conduction along the spiral might be considerable, so another form of 

 spiral was designed in which this error was minimised. 



The mixing arrangement as used in the final experiments is indicated in fig. 8. A 

 number of circular discs, each one having a diameter equal to that of the flow-tube, 

 were punched out of thin copper sheet. These were then clamped one above the 

 other and a hole of diameter equal to that of the thermometer sheath was drilled 

 eccentrically through all of them, fig. 8, D. Each disc was then cut along the 

 diameter common to the disc and the hole, and was slipped on to the thermometer 

 sheath. By slightly bending the discs in the manner shown, they, when soldered 

 together, form a kind of spiral which mixes the steam most efficiently ; moreover, 

 since the discs are only connected by small solder junctions, longitudinal conduction 

 is greatly reduced. 



