652 



SCIENCE 



[Vol. LV, No. 1433 



Physical Laboratory this problem is being at- 

 tacked from three different angles. This paper 

 presents some discoveries relating to the excita- 

 tion of radiation and ionization in hydrogen and 

 nitrogen. 



Hydrogen : Two methods of investigation have 

 been employed. In the first, an arc was produced 

 in hydrogen by the electronic discharge from an 

 incandescent tungsten wire to a surrounding co- 

 axial tungsten tube, which could be electrically 

 heated. The voltages at which discontinuities 

 appeared in the current between the electrodes and 

 especially tlie voltage at which the arc struck 

 indicated the critical potentials for the setting 

 in of radiation or ionization. With the outer 

 tube cold the hydrogen was in the ordinary 

 molecular state. With the outer tube at a 

 temperature near the melting point of tung- 

 sten, the hydrogen was completely dissociated 

 into atomic hydrogen. Thus the effects due to the 

 molecule and those due to the atom could be 

 definitely distinguished from each other. This 

 is the first experiment ever performed in an at- 

 mosphere of pure atomic hj'drogen. The follow- 

 ing results were obtained: (1) An are can not 

 be produced or maintained in molecular hydrogen 

 at voltages less than 16 volts, which is the 

 ionizing potential of the molecule. (2) In atomic 

 hydrogen the arc struck easily at 13.5 volts and, 

 with very large electronic currents, at 10.1 volts. 

 These are, respectively, the ionizing and radiating 

 potentials of the hydrogen atom as given by 

 Bohr's theory. (3) The hydrogen line spectrum 

 was observed whenever the are struck. It was not 

 observed below 16 volts in molecular hydrogen, 

 but was observed as down to 10 volts in atomic 

 hydrogen. (4) The hydrogen secondary spectrum 

 was not observed below 16 volts and the only 

 lines found in this spectrum were those of the 

 group which shows no Zeeman effect. (5) The 

 Balmer series lines were reversed in the hot tube, 

 provided the gas was ionized. The second method 

 was that of Franek and Hertz, modified to per- 

 mit a variation in the relative proportions of 

 atomic and molecular hydrogen by use of a grid 

 of hot tungsten wires, and to enable effects of 

 radiation to be distinguished from those of ioniza- 

 tion. These results corroborated those of the 

 above method and showed, further, that the hydro- 

 gen molecule can be ionized without dissociation 

 and that the lines of the Lyman series can proba- 

 bly be separately excited at successively higher 

 voltages. 



Nitrogen : In the hot tungsten tube, there was 

 no certain evidence of dissociation into atomic 



nitrogen by heat alone, but there was evidence 

 that nitrogen was more easily dissociated by 

 electron impacts in the hot than in the cold tube. 

 The atomic nitrogen was eliemically active, com- 

 bining with the tungsten of the tube furnace, and 

 it greatly increased the conductivity of the gas 

 between the electrodes. The presence of atomic 

 nitrogen was indicated by this increased conduc- 

 tivity of the gas or by the emission of lines of the 

 nitrogen line spectrum. The following conclusions 

 have been reached with regard to the nitrogen 

 spectrum: (1) The three groups of positive 

 bands are all due to the neutral nitrogen mole- 

 cules. (2) The negative bands are due to the 

 ionized nitrogen molecules. (3) The bands of 

 the third positive group are excited at about 

 7 volts, those of the second positive gi'oup are 

 excited below the ionizing potential and decrease 

 in intensity as the voltage is raised above the 

 ionizing potential, those of the first positive group 

 were not observed below the ionizing potential and 

 increased in intensity with increasing voltage, and 

 the negative bands were first observed at one or 

 two volts above the ionizing potential and in- 

 creased greatly in intensity mth increasing volt- 

 age. (4) Several new components of bauds in the 

 first group of negative bands were discovered, 

 and their wave lengths agreed accurately with 

 those predicted by Deslandre's formula. (5) 

 The line spectrum was not observed below 70 

 volts, which is also the voltage at which evidence 

 of atomic nitrogen is obtained. The minimum 

 arcing voltage, about 16.5 volts, is due to ioniza- 

 tion without dissociation of nitrogen molecules. 

 The relation of these results to observations made 

 in other connections is briefly considered. 



Becent developments in vacuum tubes and 

 their use: J. H. Morecroft. 



A primary standard of light: Herbert E. 

 IVES. The standard investigated is one developed 

 after the suggestion of Wardner and Burgess, 

 namely, the black body or complete radiator at 

 the melting point of platinum. In order to 

 realize this practically, hollow cylinders of plati- 

 num are raised to the melting point by the pas- 

 sage of a heavy electric current. The light 

 emitted from a small opening is observed by a 

 photometer upon whose field an image of the 

 cylinder is thrown by a lens. It is found that 

 with highly purified platinum the value obtained 

 for the brightness of the black body is 55% 

 candles per square centimeter. This standard 

 appears to be more reproducible than any now 

 available, and can be directly correlated with 

 other physical constants. 



