440 
PHYSICS: L. B. LOEB 
Proc. N. a. S. 
mobility computed therefrom are greater than 750 cm. /sec. The con- 
clusion to be drawn is that in pure nitrogen and hydrogen gas the elec- 
trons do not attach themselves to the molecules to form ions in any ap- 
preciable quantities. 
The results obtained in nitrogen are in agreement with those of Franck.^ 
The latter used a much more laborious method for the purification of 
his nitrogen than did the writer. He also used electrons generated by the 
action of alpha particles on the gas molecules. It is possible that the use 
of liquid air in the process of purification employed above simplified that 
process. Both Wellisch^ and Haines^ found that in hydrogen even at 
atmospheric pressure a considerable number of their carriers were elec- 
tronic in nature. Both of them, however, also found a large proportion 
of normal negative ions. That they did not obtain the complete absence 
of ions here obtained is not surprising. For if the purity of their gases 
had been a little less than that used by the writer, the fact that they 
generated their ions in an auxiliary chamber where they had consider- 
able opportunity in picking up molecules of the impurity before reaching 
the measuring field would have given them appreciable numbers of ionic 
carriers. In his monograph on positive rays J. J. Thomson^ states that 
he never observed a single instance of negatively charged nitrogen mole- 
cules in his positive ray spectrographs. He did in some cases find mole^ 
cules of oxygen carrying a negative charge. In a few vary rare instances 
he found molecules of hydrogen with a negative charge. It is, of course, 
doubtful whether the results of Thomson with positive rays are in any 
sense contradictory to the results obtained above for hydrogen. The 
carriers which Thomson observed were generated in electrical fields of 
much greater intensity and at much lower pressures than were used in 
this work. Furthermore Thomson was able to observe carriers whose 
existence must have been much more transitory than those that the mo- 
bility measurements could detect. 
According to Bohr's^ theory the hydrogen molecule permits the incor- 
poration of a third electron into the system to form a stable negative 
hydrogen ion. It is interesting to note that this does not occur to any 
noticeable extent under the conditions of the writer's experiments. 
In conclusion the writer wishes to thank Professor R. A. Millikan for 
his very kind advice and criticism. 
* Nationai. Research Fei^low 
t The values of k obtained (p. 438) are somewhat greater than those given for normal 
ions. These abnormal values of the mobilities are the result of the low pressures used. 
Their explanation on the basis of the Thomson theory will constitute the body of a 
later paper. 
§ It is impossible to locate the value of Fo for the curves in Nitrogen. They are too 
small. The curve in hydrogen obtained with 712 cycles appears to cut the axis near 
2 volts. The value of k for the hydrogen carrier deduced from this lies above 900. 
