356 SECTIONAL TRANSACTIONS.— A. 



Tuesday, September 12. 



17. Mr. H. W. EicHMOND, F.R.S. — The Problem of Expressing any 



Rational Number as a Siim of Powers of sucli Numbers. 



18. Mr. M. A. GiBLETT. — Some Recent Developments in Synoptic 



Meteorology. (With reference to the Meteorological Exhibit 

 in the Guildhall, Hull, arranged by the Air Ministry during the 

 meeting.) 



19. Mr. J. Jackson. — Double Stars. 



20. Mr. E. A. Milne. — The Escape of Molecules from an Atmo- 



sphere, with special reference to the Boundary of a Star. 



21. Mr. J. E. P. Wagstaff. — The Determination of Dielectric Con- 



stants and Susceptibilities by Valve Methods. 



22. Mr. J. EwLES. — Kathode Luminescence and its Relation to States 



of Molecular Aggregation. 



The experiments were begun with the object of determining whether the 

 luminescence excited by cathode rays appeared only when the rays possessed 

 a minimum speed, i.e., whether a definite quantum of energy was required to 

 excite the kathode luminescence. 



The source of high potential was a Mercedes electrostatic machine, and the 

 voltage applied to the tube controlled by the pressure and measured by a Kelvin 

 and White electrostatic voltmeter. 



Many metallic oxides and other substances were investigated. It was found 

 that each substance requires a definite speed of cathode rays to excite 

 fluorescence. 



The speed is characteristic of the substance, but depends also on the manner 

 in which it is prepared. Evidence is put forward to show that the method of 

 preparation affects the characteristic speed by determining the state of molecular 

 aggregation. 



It is suggested that kathode luminescence is a manifestation of the energy 

 change involved when a substance changes from one state of aggregation to 

 another. 



To test this hypothesis substances were taken which are known to undergo 

 such a change at a certain temperature. 



It has been found in every case that above the temperature of transformation 

 the kathode luminescence disappears. 



A striking confirmation was provided by powdered quartz in another way — 

 a change of refractive index corresponding to the formation of a certain amount 

 of tridymite accompanying the fluorescence. 



Assuming that the least energy of the cathode rays which excites fluorescence 

 represents the amount of energy required to break up a molecular aggregate, 

 and assuming equipartition of energy among the constituent molecules of the 

 aggregate, it is possible from a knowledge of the molecular heat of transformation 

 to calculate the number of molecules per aggregate. 



It is shown that the theory of kathode luminescence here put forward offers 

 a reasonable explanation not only of the above results but of those of other 

 workers on this and other types of luminescence. 



Moreover, remembering that fluorescent spectra are always banded, it is 

 hoped, later, to link up this work with the molecular rotation theory of band 

 spectra of emission and absorption. 



23. Joint Discussion (Cosmical Physics Sub-section) with Sec- 



tion E on Monsoons. Opener: Dr. G. C. Simpson, C.B.E., 

 F.R.S. 



