Mr Purvis, The influence of a strong magnetic field, etc. 41 
The influence of a strong magnetic field on the spark spectra 
of titanium, chromium and manganese. By J. E. Purvis, M.A., 
St John’s College, Cambridge. 
[. Read 26 November 1906.] 
The author has given accounts in previous communications of 
investigations on the influence of a magnetic field on the spark 
spectra of various metals*. The distances between the consti- 
tuents of a considerable number of the divided spectral lines were 
measured, and the values of dX/X 2 were calculated therefrom. 
One of the results was to shew that some of the lines could be 
correlated and compared in possessing similar general appearances, 
similar corresponding constituents, and essentially identical values 
of dX/X- for these constituents. 
The investigations of Runge and Paschen'f' testing and con- 
firming Preston’s Law shewed that the lines of different elements 
corresponding to the laws of series are separated in the same 
manner in the magnetic field and, as a check on the present series 
of observations, the author has repeated their experiments with 
several lines of the second subordinate series of magnesium, zinc 
and cadmium. The field strengths utilised by them varied between 
28000 and 38000 C.G.S. units, and the spectra were produced 
simultaneously by passing the spark from one metal to the other. 
They claimed the advantage of thus avoiding the determination 
of the field strength, which cannot be made with the same relative 
accuracy as the measurements of the distances of the components. 
The only essential was to keep the field strength constant during 
an exposure, whilst the former might be different for different 
exposures. The following table, extracted from Runge and Pas- 
chen’s paper {Astrophysical Journal , Vol. xv. p. 335), gives the 
distances of the components of the lines on the scale of vibration 
numbers ; where, by the vibration number, is meant the number 
expressing the vibrations of light in passing over a distance of 
1 cm. The distances were computed from the centre of gravity 
of the components, and are designated by the -f- sign when the 
components lie on the side of larger vibration numbers. 
* Trans. Camb. Phil. Soc. Vol. xx. No. vm. p. 193; Proc. Camb. Phil. Soc. 
Vol. xm. Pt vi. p. 325. 
t Astrophys. Journ. Vol. xv. pp. 235, 333; Vol. xvi. p. 123. 
