SOLAR ECLIPSE EXPEDITIONS— MITCHELL I55 



candles as the result; in 1926, 0.14 foot-candles; and in 1929 and 1930, 

 0.15 and 0.38 foot-candles, respectively, the eclipses from 1926 to 

 1930 being measured by the same instruments. 



These measures seemed to show that although the 1926 corona 

 was 40 percent brighter than that of 1925, the total illumination of 

 corona and sky together was 40 percent fainter. 



Following the 1932 eclipse, when many measures were made in the 

 United States, one competent illumination expert voiced his opinion 

 of attempts to measure the total light of the corona by the use of 

 illumination meters in the following words: ^ 



There can be no objection to anyone measuring the normal illumination on a 

 completely exposed test plate at the time of totality if he is convinced of the 

 utility of so doing. Trouble, however, may ensue vi^hen the results of such 

 measurement are seriously put forward as indicative of the amount of light which 

 the corona is giving. 



Hence, it is evident that we must not now take too seriously the 

 attempts to correlate intensity of coronal radiation with sunspot 

 activity. It seems very liighly probable that the inner corona at 

 sunspot maximum must be brighter than at spot-minimum. More- 

 over, as the inner corona contributes the greatest part of the energy 

 of the total coronal radiation, we would logically expect that the total 

 energy at maximum of spots is greater than at minimum. Unfortunately, 

 eclipses come but seldom and there are few observations with which 

 to derive correlations. Many astronomers (and other scientists as 

 well) have come to hasty conclusions tlirough the discussion of 

 observations affected by large accidental and systematic errors. 

 When we consider the paltry amount of time devoted to the measure- 

 ment of the light of the corona, we must not be too much discouraged 

 at the meager and confhcting results obtained. 



In spite of all the inconsistencies of the results, observers through- 

 out the past 50 years have been in agreement that the total light of 

 the corona is roughly one-half that of the full moon. It is with a note 

 of surprise that we read ^ the statement by Menzel and Boyce that 

 the latest corona observed, that of 1936, shone with an illumination 

 of 50 to 100 times that of the full moon. We shall await the detailed 

 results with great interest. 



POLARIZATION OF THE LIGHT OF THE CORONA 



Polarization measures must be the result of a combination of the 

 partly polarized light from the corona and the unpolarized light in the 

 earth's atmosphere. The measures made of eclipses in the early years 

 of the present century seemed to show that the percentage of polar- 

 ization in photographic light was about three times greater than in 



' Journ. Optical Soc. Amer., vol. 23, p. 234, 1933. 

 ' Technology Rev., November 1936. 



