Chemistry and Physics. 11 



pressure of about O-Ol" 1 " 1 . It has three pronounced peaks or 

 maxima which pertain to H,, H s and H 3 . The maximum for IT l is 

 nearest to the axis of ordinates while the peak for II s is reached 

 at a more intense magnetic held than is required for either H, or 

 II . The value of the charge at the maximum for II 2 is slightly 

 greater than for H,, whereas the greatest ordinate for H 3 is 

 decidedly higher than for Il 2 . It thus appears that the particles 

 of molecular weight 3 were present in relatively large numbers. 

 The second curve was obtained with a gas pressure of 000l7 m,n . 

 It too presents three peaks, but now the H 2 maximum is about 

 two and one-half times as high as either one of the remaining 

 peaks. The third curve indicates the relative proportions of H,, 

 H 2 and H s when a pressure less than 0005 mm was used. Under 

 these circumstances the H, and H 3 peaks have nearly disappeared 

 while the H a maximum has retained its former value. That this 

 change was caused by decreasing the pressure, and not by the 

 removal of some constituent of the gas by the charcoal was shown 

 by the fact that when hydrogen was admitted, while the charcoal 

 and liquid air were kept in action, H, and H 3 regained their 

 original relative intensities. 



The writer of the paper accounts for the phenomena in the fol- 

 lowing way. Since in the high vacuum the free-path of the 

 molecules is very great, the positive ions which are still formed 

 by the dense stream of electrons coming from the Wehnelt 

 cathode make very few collisions with the hydrogen molecules. 

 Hence these positively charged diatomic molecules are analyzed 

 in the condition in which they were at the instant of their forma- 

 tion. We must conclude, then, that electrons ionize only by 

 detaching a single electron from a molecule, and are not able to 

 dissociate a molecule into atoms. When the pressure is greater, 

 some of the positive ions collide with the neutral molecules of the 

 gas before the cathode and dissociate them. A positively charged 

 atom thus formed may attach itself to a neutral molecule and 

 give rise to II 3 . The author also concludes that H 3 cannot be 

 regarded as a stable gas since it is not present when there is no 

 dissociation of the hydrogen molecules. — Phil. Mag., xxxi, p. 

 438, May, 1916. h. s. u. 



7. The Structure of Broadened Spectrum Lines. — In the pre- 

 ceding number of this Journal the results obtained by Nicholson 

 and Merton were given, but details of the experimental method 

 were lacking in the original paper. Consequently it may not be 

 superfluous to give a brief description of the essential feature of 

 this method as explained in another article by T. R. Merton. 

 A neutral-tinted wedge (of the type used for recording the sensi- 

 bility curves of photographic plates) of density graded from 0'2 

 to 4*2 was mounted on the slit of a single-prism spectograph with 

 its refracting edge at right-angles to the axis of the slit. Before 

 the wedge was placed in position care was taken to have the slit 

 uniformly illuminated over its entire effective length. Under 

 these conditions each line of the spectrum appears brightest at 



