66 THE ROYAL SOCIETY OF CANADA 



slowly into the manometer. From time to time the pressure was 

 measured by means of a McLeod gauge and kept below one-half a 

 micron by the pump. After three months no difference could be 

 detected between manometers in use and one recently made ; the sur- 

 face of the mercury remained perfectly clean. (Fig. 1.) 



Our constant temperature bath (Fig. 2) consisted of a three inch 

 cylindrical Dewar flask, silvered except for two strips on opposite 

 sides so that the contents might be easily seen. Ether and solid 

 carbon dioxide were used as the bath liquid, and with a May-Nelson 

 pump temperatures between — 80° and — 116°C could be obtained. 

 At these low temperatures the bath warmed one degree in 25 minutes, 

 but by regulating the pressure over the carbon dioxide the temperature 

 could be kept to • 1 °. The bath was stirred by forcing in dry air 

 through the tube C. Such an arrangement will be found satisfactory 

 only when a relatively small amount of solid is present. The mixture 

 must not be viscous. 



The hydrogen thermometer was of the well known Travers' type, 

 and was filled with hydrogen from palladium. Its ice-point was deter- 

 mined from time to time. By forcing the mercury up into the vacuum 

 tube gases liberated from the walls were pushed into a "trap" and there 

 retained. 



The hydrides were made in the ordinary ways and dried 1 by 

 phosphorus pentoxide. They were fractionated three times in vacuo, 

 the middle portions only being retained, and were finally distilled into 

 the vapour pressure bulb E. They were then frozen by liquid air and 

 the apparatus was exhausted to 0-01 mm. by the Toepler pump. 

 They were melted, re-frozen, the system re-exhausted and finally 

 sealed off from the pump. The gases came in contact only with the 

 glass of the tubes and of the spiral. 



The spiral and method of using it have been described fully by 

 Johnson. With our apparatus 0-8 mm. on the manometer corre- 

 sponded to one millimeter on the scale. As the scale could be read to 

 0-1 mm. pressures could be determined to 0-08 mm. On account 

 of the variations in the temperature of our bath, we have thought it 

 best to give our results only to the* nearest millimeter. 



1 The minute amounts of phosphorus compounds formed are removed during 

 the distillation. 



