HOGG. — FRK.'TION AND FORCE DUE TO TRANSPIRATION. 133 



day. There was no cracking of the glass, so that one was assured 

 that the joints had all been well annealed. 



Drying Process. 



This process consisted in first pumping out the air until a pressure 

 of a few millimeters of mercury was reached, then forcing in through 

 the tube G a large quantity of dry air. This was pumped out again, 

 and more dry air then forced in. This process was repeated many 

 times. After this preliminary drying, and when the pressure was low, 

 the front of the oven was closed and a current turned on sufficient to 

 raise the temperature to 200° C. This temperature was maintained 

 for several hours, and during this time the pump was kept going, and 

 for about one hour the connecting tubes were kept hot by a Bunsen 

 flame. 



Perhaps it may be well to state here, that in the effort to distinguish 

 between a real leak in the apparatus and what will produce a similar 

 eff'ect, viz., the slow "evaporation " of the gas from the inner surface of 

 the vessel, it was found expedient to remove the sulphur tube and silver 

 tube. Under these conditions the pumping can be done much more 

 quickly. When the heating and pumping process above described had 

 been resorted to, and after the oven had been allowed to cool down to 

 room temperature, it was found that a vacuum of about 0.0003 mm., as 

 indicated by the gauge, had been reached. When, however, the appara- 

 tus was simply allowed to stand the pressure greatly diminished, and a 

 condition was finally reached when a slight rise in the temperature of 

 the room would cause the pressure to increase greatly, much more than 

 could be accounted for by the temperature coefficient. Then again a 

 small decrease in temperature, say four or five degrees, would cause a 

 corresponding anomalous decrease in the pressure. In one case where 

 the temperature of the room reached 14° C. and remained there for 

 over a day, the pressure was less than one third of what it was when 

 the temperature had reached 18° C. for the first time after the oven 

 had been allowed to cool down. This is mentioned merely to show 

 how important a part the temperature of the glass plays in determining 

 what the vacuum in the apparatus at any time is.^o 



After the complete drying process the spectrum showed that the 

 moisture had been removed, and the vessel was then filled with 

 dry air. 



The suspensions were then raised to the proper height by turning 



" After it had been proved that there was no leak in the apparatus the 

 sulpliur and silver tubes were replaced. 



