102 PHENOMENA, ATOMS, AND MOLECULES 



vacuum give off gases almost indefinitely, and that it is impossible to free 

 metals from gas by heating. Still I thought that 7000 times its ov^^n volume 

 of gas was an entirely unreasonable amount to obtain from a filament. I 

 spent most of the summer in trying to find where this gas came from, and 

 never did investigate the different samples of wire to see how much gas 

 they contained. How much more logical it would have been if I had dropped 

 the work as soon as I found that I would not be able to get useful informa- 

 tion on the "offsetting" problem by the method that I had employed. 



What I really learned during that summer was that glass surfaces which 

 have not been heated a long time in vacuum slowly give off water vapor, 

 and this reacts with a tungsten filament to produce h}'drogen, and also that 

 the vapors of vaseline from a ground-glass joint in the vacuum system give 

 off hydrocarbon vapors, which produce hydrogen and carbon monoxide. 



That summer's work was so interesting that I dreaded to return to the 

 comparative monotony of teaching, and gladly accepted Doctor Whitney's 

 offer to continue at work in the laboratory. No definite program of work 

 was laid down. I was given first one assistant and then others to continue ex- 

 periments in the sources of gas within vacuum apparatus, and a study of 

 the effects produced by the introduction of various gases into tungsten 

 filament lamps. The truth is that I was merely curious about the mysterious 

 phenomena that occur in these lamps. Doctor Whitney had previously 

 found that gases have a habit of disappearing in lamps, and no one knew 

 where they went to, so I wanted to introduce each different kind of gas 

 that I could lay my hands on into a lamp with a tungsten filament and find 

 out definitely what happened to that gas. 



It was the universal opinion among the lamp engineers with whom I 

 came in contact that if only a much better vacuum could be produced in 

 a lainp a better lamp would result. Doctor Whitney, particularly, believed 

 that every effort should be made to improve the vacuum, for all laboratory 

 experience seemed to indicate that this was the hopeful line of attack on the 

 problem of a better lamp. However, I really didn't know how to produce 

 a better vacuum, and instead proposed to study the bad effects of gases by 

 putting gases in the lamp. I hoped that in this way I would become so 

 familiar with these effects of gas that I could extrapolate to zero gas pres- 

 sure, and thus predict, without really trying it, how good the lamp would 

 be if we could produce a perfect vacuum. 



This principle of research I have found extremely useful on many oc- 

 casions. When it is suspected that some useful result is to be obtained by 

 avoiding certain undesired factors, but it is found that these factors are 

 very difficult to avoid, then it is a good plan to increase deliberately each 

 of these factors in turn so as to exaggerate their bad effects, and thus 

 become so familiar with them that one can determine whether it is really 



