PORTION OF THE COSMOS. -THE SUN. 287 



designed not only for testing his own views, but also for 

 reducing the results of observation to exact numerical 

 proportions. 



The comparison of the Sun's light with the two most 

 intense artificial lights which have yet been produced,, gives 

 (according to the still very imperfect state of photometry) 

 the following numerical results. In the ingenious experi- 

 ments by Eizeau and Eoucault, Drummond's light (produced 

 by the flame of an oxy-hydrogen lamp directed upon lime) 

 is to the light from the Sun's disk as 1 to ]46. The in- 

 tensity of the light produced between two charcoal points 

 by a Bunsen's pile in Davy's experiment, with a battery of 

 46 small plates, was to the solar light as 1 : 4*2, and 

 with large plates as 1 : 2*5, or more than one-third of 

 the Sun's light ( 485 ). If we still hear with astonishment 

 that Drummond's dazzling light appears as a black spot 

 when projected on the Sun's disk, we may regard with the 

 higher admiration the genius of Galileo, in drawing, in 1612, 

 from a series of inferences respecting the smallness of the 

 distance from the Sun at which Venus would cease to be 

 visible to the naked eye, the conclusion, that the blackest 

 nucleus of a solar spot is brighter than the brightest part of 

 the full moon ( 486 ). 



Taking the intensity of the whole lighi, of the Sun as 

 equal to 1000, William Herschel estimated that of the pe- 

 numbras on the average as 469, and that of the black nuclei 

 themselves as 7. According to this assumption, which of 

 course can only be regarded as a very conjectural one, and 

 taking with Bouguer the light of the Sun to be 300000 

 times as strong as that of the full moon, a black nucleus 

 would still possess 2000 times more light than the full 



VOL. in. s 



