124 
ASTRONOMY: HALE AND BABCOCK 
The Stark effect may be briefly described. Positively charged par- 
ticles (canal rays), emitted from the anode of a vacuum tube, pass 
through perforations in the kathode into an intense electric field. When 
viewed across the lines of electric force, the series lines of such an ele- 
ment as hydrogen are split into two sets of components, polarized in 
planes at right angles to one another. When seen along the lines of 
force, one set of components disappears, while the others are present 
but impolarized.2 Thus, while the phenomenon resembles the Zeeman 
effect, a fundamental distinction lies in the fact that the components 
observed along the lines of force of a spectrum line resolved by a mag- 
netic field are circularly polarized in opposite directions. 
Our studies of the magnetic phenomena of sun-spots and of the sun 
as a whole have been based mainly upon this circular polarization, thus 
eliminating any possibility of attributing the observed effects to electric 
rather than magnetic fields. Many other criteria, such as differences 
in the number of components and the variation of separation with 
wave-lengths are also available to remove possible doubts, which may 
enter when the observations are made at right angles to the lines of 
force. 
The Ha line of hydrogen, when observed by Stark with moderate 
dispersion in an electric field of 28,500 volts per centimeter, was re- 
solved into three components. The two outer components are polar- 
ized parallel to the field while the central line (which is double under 
higher dispersion) is polarized in the opposite plane. Hence the total 
width of the resolved line may be greatly varied by rotating the Nicol 
prism mounted above the spectroscope slit, since in one position the two 
outer components v/ill be transmitted, while if the Nicol is turned ninety 
degrees these will be cut off and the central line transmitted. 
In the sun the only known cases of line resolution (other than un- 
polarized reversals) are those found in the spectra of sun-spots. Im- 
mediately after the announcement of Stark's discovery, we examined 
our photographs of spot spectra to determine whether any anom- 
alous cases of widened or resolved lines might be attributable to an 
electric rather than a magnetic field. In general, however, it was 
found that the outer components of spot triplets were sharply and com- 
pletely cut off by the Nicol and quarter-wave plate under favorable 
conditions of observation, and are thus circularly polarized. Even in 
the case of spots near the middle of the sun, the central line of these 
triplets is usually present, apparently indicating that the lines of mag- 
netic force are not exactly radial. But this component is very nar- 
