30 On the Spectra of Iron, $., radiating in a Magnetic Field. 



order in the spectrum. They may all exist in a single group of lines 

 of nearly the same wave-length, and the magnitude of effect, even in 

 the case of the lines which triple, cannot at present be stated to 

 follow any law. Such a law, if one exists, can be discovered only by 

 a complete survey of the spectra of various substances with a very 

 powerful magnetic field capable of wide limits of variation. Such a 

 field I hope to possess in an electromagnet which is being built for 

 the purposes of further inquiry in this direction. 



In the meantime it may be taken as thoroughly determined that it 

 is untrue to assume that there is any such law as that the effect is 

 uniform for the various spectral lines of any substance, or that the 

 effect varies as the wave-length or as the square of the wave-length. 

 No doubt it may be possible in the future to divide the spectral 

 lines into groups such that the members of each group are similarly 

 affected ; and it may be that a certain homology may be found to 

 exist between tLe groups of one substance and those of other sub- 

 stances, and hence the effect produced by the magnetic field may be 

 brought into line with those chemical and physical properties of the 

 elements which have already thrown them into periodic groups. 



I am at present very hopeful that some law will be obtained in 

 this department for the phenomena attending the action of a mag- 

 netic field on the radiation from a source placed in it, for the results 

 which I have already obtained show certain correspondences in the 

 effects produced on the spectra of allied elements as well as certain 

 recurrences of the same effect in the spectrum of the same substance. 



The amount of ground so far covered has not, however, been suffi- 

 cient, nor indeed has my magnetic field been sufficiently strong, to 

 enable me to lay hold of any general principle controlling the 

 observed effects ; nevertheless a useful purpose will be served in pub- 

 lishing this preliminary statement, as it will, at least to some extent, 

 clear the ground for those who are at present considering the pheno- 

 mena from a purely theoretical standpoint, and it may also to some 

 extent serve a perhaps equally useful purpose in diverting a useless 

 expenditure of energy on the part of those who may be engaged in 

 formulating the laws of the phenomena without sufficient knowledge 

 of the facts. 



In conclusion I may mention that the spectral lines of air and 

 hydrogen are rendered diffuse by the action of the magnetic field. 

 The broadening here is not of the character exhibited by spectral 

 lines of metals in which the lines remain sharp though easily broadened, 

 but it is more of the character of a broadening produced by increase 

 of pressure of the gas. 



I may also add that the magnetic effect does not appear to be in 

 any way parallel with the pressure effect, that is, the displacement of 

 the lines caused by increase of pressure in the gas surrounding the 



