110 



Dr. W. Huggins. 



giving again and again the impression that my efforts were about to 

 be crowned with success." 



The spectroscopic method by which the prominences may be seen 

 without an eclipse, fails for the corona, because a small part ouly of 

 the coronal light is resolved by the prism into bright lines, and of these 

 lines no one is sufficiently bright and coextensive with the corona to 

 enable us to see the corona by its light. 



Let us look at some of the conditions of the problem. As the 

 obstacle to our seeing the corona consists of the bright screen of 

 illuminated air which comes in before it, it is of importance to 

 consider the relative degree of brightness of the air-glare, under 

 favourable conditions, to that of the corona behind it. 



During the eclipse of 1878, Professor J. W. Langley found the 

 apparent brightness of the coronal light at 1' from the limb of 

 the moon to be six times greater than that of the full moon, but at 

 3' distance, the light to have fallen off to one-tenth of the light 

 of the full moon.* Professor Harkness concludes for the same 

 eclipse : — (1.) The total light of the corona was 3*8 times that of the 

 full moon, or 0-0000069 of that of the sun. (2.) The coronal light, 

 varied inversely as the square of the distance from the sun's limb. 

 (3.) The brightest part of the corona was 15 times brighter than the 

 surface of the full moon. (4.) The corona of December 12, 1870, 

 seems to have been 7^ times brighter than that of July 29, 1878.f 

 In his report on the eclipse of 1883, M. Janssen says : — " Cette ex- 

 perience a montre qu'a Caroline l'illumination donnee a ete plus 

 grande que celle de la pleine lune."J 



The chief point of importance for this inquiry lies not so much in 

 the actual value of the coronal light as in the relation of that value to 

 the brightness of the illuminated air near the sun. Many observers 

 have borne testimony to the continued visibility of the corona for 

 some minutes (from three minutes to twelve minutes) after the end of 

 totality. 



The observations which give to us direct information on this point 

 are those which have been made of the planets Yenus and Mercury 

 when they come in between us and the sun. It is obvious that as the 

 planet approaches the sun it comes in before the corona and shuts off 

 the light which comes from it. Under these circumstances the 

 observer sees the sky in front of the planet to be darker than the 

 adjoining parts, that is to say, the withdrawal of the coronal light 

 from behind has made a sensible diminution of the brightness of the 

 sky. It follows that the part of the sky about the sun, behind which 

 the corona is situated, and to which its light is added, is brighter than 



* Washington Observations. Eeports of Solar Eclipses, 1878 and 1880, p. 214. 

 f Ditto, p. 392. 



+ " Annuaire pour l'An 1884" (B. des Longitudes), p. 875. 



