284 ' SOSMAN AND HOSTETTER! A VACUUM FURNACE 



is evident from .the drawing. This cutoff, like the gages, is 

 operated by a plunger instead of a movable reservoir. The 

 glass traps on either side prevent mercury from being shot into 

 the furnace or the. pumps by any accidental rush of gas. 



With, the mercm-y cutoff closed and the joint of the crucible 

 suspension sealed with kollolith the apparatus is absolutely 

 free from leaks. It has stood 27 days without showing a pres- 

 sure greater than 0.0052 mm. of mercury. This pressure seems 

 to have been produced by the slow evolutiom of gas from the 

 large surface of glass exposed. This matter will be discussed 

 more in detail in a later paper on the iron oxides. 



The plunger-lift for the large McLeod gage, Mi, consists of a 

 steel plunger in a smooth wrought-iron pipe, as the necessary 

 dmiensions are too large to permit of the use of glass. The 

 dimensions need to be carefully proportioned, since the dis- 

 placement of mercury by the plunger must be equal to the volume 

 of the gage plus the volume of mercury in the annular space 

 between plunger and cylinder when the gage is at its highest 

 reading. The greater the clearance between plunger and cylin- 

 der, therefore, the greater must be the diameter of both. Ours 

 consists of a plunger 50.5 mm. in diameter and 100 cm. long 

 in a cylinder 53 mm. inside diameter and 102 cm. long. The 

 steel plunger is just floating in the mercury when the gage is 

 nearly filled, so that the mercury columns can be set accurately 

 by a slight pressure on the plunger. The use of a plunger-lift 

 of this kind does away with the raising and lowering of an awk- 

 ward mercury reservoir, with its accompanying rubber tubing, 

 leaks, and sulfur contamination of the mercury. 



Oxygen is admitted by way of one or both of the calibrated 

 bulbs at the right of the apparatus. We have made no provision 

 for exact measurement of the amount of oxygen to be admitted, 

 as it can be just as exactly measured by its pressure after ad- 

 mission to the apparatus. It is possible to get any desired 

 amount of oxygen by expanding slowly from either the small or 

 large bulb. These bulbs also serve, together with the space 

 between the mercury cutoff and the pump stopcock, for the 

 withdrawal of known amounts of oxygen from the apparatus. 



