568 REPORi — 1903. 



between the electrified mountain and the similarly electrified incandescent sand ; 

 and also to the attraction of the mountain for the oppositely electrified cloud 

 above. The principal cause of death is ascribed to electric shock, and the over- 

 throw of the cannon and image on the battery hill of St. Pierre is cited as an 

 instance of destruction that could only have been brought about by electric means. 



TUESBA r, SEPTEMBER 15. 



Depaetment of Mathematics and Physics. 

 The following Report and Papers were read : — • 

 1, Report of the Committee on Electrical Standards.— ^ee Reports, p. 33. 



2. Note on Carbon and Iron Are Spectra at High Gaseous Pressures. 

 By R. S. HuTTON and J. E. Petavel. 



The paper gives an account of a preliminary investigation of arc spectra at 

 pressures up to 100 atmospheres. 



Two points are worthy of note. Firstly, a sharp reversal of the cyanogen 

 band 3883 in the carbon arc spectrum. Secondly, the reversal of a large number 

 of lines in the ultra-violet part of the iron spectrum. 



Incidentally it was found that the discharge of a high-tension alternating 

 current of large intensity between iron electrodes fills the whole of the partially 

 evacuated enclosure with a brilliant yellow glow. A study of this glow has shown 

 a considerable simplification of the iron spectrum. 



3. Hoiv to Exhibit in Optical Instruments the Resolution of Light into its 

 Component Undulations of Flat Wavelets, and hovj to Employ this 

 Resolution as our Guide in Making and in Interpreting Exjyeriments. 

 By G. Johnstone Stoney, M.A., Hon.Sc.D., F.R.S. 



The present communication is the third of a series of papers on a new method 

 of dealing with optical problems, based on the fact that, however complex the 

 light traversing a given space may be, it always admits of being resolved into 

 components, each of which is an undulation of uniform flat wavelets sweeping 

 across that space. The chief peculiarity of this resolution is that the components 

 undergo no change as they advance. This gives the resolution into flat wavelets 

 an advantage over every other method of resolving light. 



The first paper, published in the ' British Association Report ' for 1901, p. 570, 

 gives a more direct and more easily understood proof of the fundamental theorem 

 than that which had some years previously been given by the author. The second 

 paper, of which an abstract is given in lastyear's * British Association Report,' and 

 which appears in full in the ' Philosophical Magazine ' for February 1903, explains 

 how to construct a reference hemisphere and indicator diagram, which makes it 

 easy to employ the new method of resolution in the chamber study of optical 

 pi-oblems ; and in the present paper an explanation is given of how to exhibit the 

 contents of this indicator diagram when making experiments with optical instru- 

 ments. Thus, in the case of the microscope, the light from the objects upon the 

 stage has to pass through a stratum of either air, water, or oil before entering the 

 objective ; and if wo conceive the light traversing this stratum to be resolved into 

 its component undulations of flat wavelets, then portions of these — that is to say, 

 beams which we may imagine to be cut out of them — enter the objective and are 

 brought to a focus by it, in an image which may be seen by removing the eyepiece 

 and looking down the tube of the instrument. 



