EXPERIMENTAL STUDY OF HEAT LEAICVGE. 793 



constant temperature, decreased slightly during the course of the 

 experimental work, doubtless because of an aging effect, recovering a 

 portion of the decrease during idle periods. An increase in the 

 energy input at constant flow resulted in a decrease of the measured 

 resistance in five cases and in an increase in three. The evidence is 

 thus against the existence of a heat-leak of the kind suspected; it is 

 probable that the variations of resistance actually noted were due to 

 other causes than variations of the temperature of the coil. 



g. The ivet-steam question. Radial flow plugs Ul and U2. 



It is improbable that perfectly dry superheated steam is ever secured 

 in practice, at least at temperatures within thirty or forty degrees 

 Centigrade of saturation. Particles of liquid are always brought 

 away, suspended in the vapor, during vaporization. These particles, 

 spherical in shape, are able to persist in the liquid form after the appli- 

 cation of superheat because of the excess pressure on the interior of 

 the drop produced by the surface tension. The effect of evaporation 

 from the surface of the drop is a reaction against a continuation of the 

 process, because of the increase in the curvature of the surface. 

 Further, if, in some manner, steam free from moisture were actually 

 obtained, it would be very difficult to demonstrate the fact. 



As far as calorimetric effects are concerned, the presence of a small 

 amount of moisture in superheated steam is not objectionable, pro- 

 vided the quantity present does not change as a result of the experi- 

 mental processes. For example, if the steam contains 0.1 per cent 

 of water, any temperature change undergone by the fluid will differ 

 by only about 0.1 per cent, from what it would be if no water were 

 present at all, provided the water-content is not changed; but if, for 

 each degree of temperature change, as little as 1 per cent, of the 0.1 

 per cent, water-content is vaporized, the effect on the temperature 

 change will amount to 1 per cent. (These statements suppose, as is 

 roughly the case, that the latent heat of vaporization and the specific 

 heats of water and steam are in the proportion 1000:2:1.) 



\Yhat an experimenter should aim at, therefore, is to reduce the 

 amount of moisture to as low a figure as possible and to endeavor to 

 have the moisture which is unavoidably present in a form such that 

 no change in its amount is likely to occur during passage through the 

 calorimetric apparatus. After taking all the precautions which seem 

 necessary to secure these conditions, they should be doubled. If no 

 nQticeable changes result, it may be regarded as reasonably certain 



