146 TRANSACTIONS OF SECTION A. 



atomic or molecular vibration, as distinguished from what in comparison might 

 be called molar vibration, could be excited. An illustration that suggested 

 itself was this : Take a vibrator composed of a series of small masses with 

 spring connections. If these masses are of atomic or molecular dimensions any 

 ordinary impulse or impact would leave them unaffected, while vibrations 

 of large groups of them, vibrations depending on the connections, would result. 

 But the impact on one of the masses of a hammer of sufficiently small dimensions 

 and mass would give vibrations depending on the structure of the mass struck, 

 and independent o't the connections, just as the bars of a xylophone ring, while 

 the suspended series of bars, if it swings at all, does so without emitting any 

 audible sound. This is, I be'lieve, in accordance with the theory now held as 

 to X-rays. We now have some information as to the mode of producing a local 

 excitement so intense as to cause not merely atomic disturbance, but actual 

 disruption of the atomic structure. Further developments of Sir Ernest Ruther- 

 ford's experiments and of his theory of their explanation will be eagerly awaited. 



The following Reports and Papers were then read : 



1. Report of Committee on Radiotelegraphic Investigations. 

 See Eeports, p. 40. 



2. The S'pednim of Nova Geminorum. 

 By P. J. M. Steatton, D.S.O., M.A.^ 



The following types of spectrum occur in the course of the star's history : 

 (1) An absorption spectrum of hydrogen and enhanced lines of calcium, iron, 

 and titanium, displaced towards the violet by amounts varying on different 

 dates between 0.0035A.and 0.0005X. (2) An absorption spectrum of hydrogen, 

 oxygen, nitrogen, carbon, and helium displaced towards the violet by amounts 

 varying on different dates from 0.0061 A to 0.0027 A. (3) A spectrum of bright 

 bands corresponding to both sets of lines represented in the absorption spectrum. 

 The bands were generally about 24 tm. wide, were slightly displaced to the 

 red, and appeared later than the corresponding absorption lines. For some 

 elements these bands were flanked by faint wings at each end, giving a wider 

 band of double the width. In the bright central band two maxima appeared 

 which for hydrogen varied in brightness with the strength of the two 

 absorptions. (4) A bright band spectrum, consisting of hydrogen and helium 

 lines and the nebulium lines known in the planetary nebulae. The structure 

 of the bright bands is maintained unaltered, the same maxima showing as 

 in stage (3). (5) A bright band spectrum, with the nebulium giving place to 

 the lines typical of the bright-line Wolf-Rayet stars. 



In addition, there are present in the early stages certain undisplaced narrow 

 dark lines, notably D,, D, of sodium and H and K of calcium. All structure 

 in the bright bands and displacements of dark lines of an element vary directly 

 as the wave-length of the presumed source; velocity is the only physical 

 cause which seems capable of producing the results. The velocities are so 

 large, reaching up to 2 x 10' cm./ sec, that electrical causes are suggested 

 for them. The sequence of spectral type from heavy elements to light ones 

 and the maintenance of a common structure in the bright bands for many 

 months after the initial changes seem typical of all Novas, but the structure 

 of these bands is different for each Nova. 



'■ To be published in Annals of Solar Physics Observatory, Cambridge, Vol. 4. 

 See also Monthly Notices. B. Astronomical Soc, Vols. 73, 79. 



