OF ARTS AND SCIENCES. 



429 



centre of this screen was cut a circular hole 2 cms. in diameter, and 

 the light passing through this aperture was received upon a Bunsen's 

 photometer disk, placed at a distance of 25 cms. behind it. The 

 porte-lumiere was then moved until the desired portion of the image 

 coincided with the centre of the aperture in the screen, and the image 

 kept at rest by a slight movement of the telescope whilst the pho- 

 tometer reading was taken. The light used for comparison was a 

 standard candle, which was placed in the photometer described above 

 (page 421). Much difficulty was experienced on account of the 

 difference in color of the light from the sun and candle, in obtaining a 

 satisfactory disappearance of the spot of the photometer disk. Various 

 attempts were made to remedy this trouble, by using colored paper, 

 disks of colored plaster, &c., none of which succeeded very well, and 

 the ordinary white disk was finally adopted. A large number of pre- 

 liminary series of observations were made, and rejected as not being 

 sufficiently accurate. 



The results of the last three days' observations are given in Table I. ; 

 the first column giving the percentage distance from the centre 



TABLE I. 



towards the edge, and the succeeding ones the intensity of the light 

 compared with that of the centre taken as unity. The second column 

 gives the mean of the first day's, the third of the second day's, and the 

 fourth of the third day's observations. Upon the second day, there 

 were several times as many observations taken as upon either of the 

 others, and its mean is correspondingly more reliable. A portion of 

 the observations were taken upon or near the polar, and others upon 

 or near tlie equatorial, diameter of the sun. Column 5 gives the 

 mean of the measurements taken upon the equatorial, and number 6 

 the mean upon the polar, diameter. As these are the results of but a 



