ATMOSPHERIC PEESSUEE BETWEEN 104 C. AND 115 C. 



411 



the values taken for the boiling-point and fundamental interval are those corre- 

 sponding to the given amount of exposure. A few determinations of these 

 constants obtained on different dates are included in order to show the constancy 

 with which they could be repeated. 



It will be noticed that there is an increase in the values of the freezing- and boiling- 

 points but that the fundamental interval remains remarkably constant. The increased 

 resistance is easily explainable. It is due to the slight squeezing of the wire against 

 the side of the glass sheath, whereby a gradual reduction in the cross-section of the 

 wire is produced after many movements of the thermometer in its sheath. During 

 the autumn of 1913 the thermometer was broken ; after mending, by fusing on a 

 short length of the same kind of platinum wire, the constants were : 



Fundamental 

 interval. 



538-29 

 538 22 

 538-21 



Exposure. 



cm. 



2 

 8 



8 



Methods used to Obtain a Uniform Temperature Distribution in the Steam Floiv. 



By far the most troublesome source of error to eradicate was that due to imperfect 

 mixing of the steam. When it is remembered that the velocity with which the 

 steam is flowing may be as great as 80 cm. per second, it is obvious, that in order to 

 ensure a uniform radial distribution of temperature throughout the steam, in the 

 short time of its passage from the heating-coil to the thermometer, the mixing 

 arrangement must be extremely good. The only method of testing the excellence of 

 the stirring arrangement is to conduct experiments with different distances between 



