the peesedent's addeess. 317 



All the early spectroscopic observations were made by direct 

 vision, the spectroscope having been fitted with prisms of a high 

 dispersive power, and attached to a telescope in the place of the 

 ordinary eye-piece. When astronomical photography was first 

 introduced in these experiments, great difficulty was found, by 

 most observers, in obtaining a good image of the spectrum on 

 the photographic plate, owing to the care required in controlling 

 the clock, so that a sensibly uniform motion of the telescope may 

 be preserved while following the star.* Dr. Huggins, however, 

 in 1876, had so far successfully arranged the control of the 

 driving clock of his telescope, that he was able to obtain a satis- 

 factory photograph of the spectrum of the bright star Vega, and 

 also of a few of the other principal stars. The solar spectrum, 

 in each case, was taken on the same plate, so that a direct com- 

 parison could be made, in order to identify the separate stellar 

 lines with the corresponding solar lines. Two dark lines crossing 

 the spectrum of Vega were thus easily recognised to be coinci- 

 dent with the position of the two solar lines of hydrogen. Five 

 other lines were also detected in the spectrum. These early 

 photographs were examined with great interest at the meetings 

 of the Eoyal Society, but more recently, owing to the adoption 

 of improved methods of observation, the number of lines 

 recorded in most of the photographs are greatly increased. 

 Sometimes they exceed a hundred. 



The mapping of the lines in the photographic spectra of the 

 stars is a research of great delicacy, as it is necessary, for a proper 

 comparison of the lines, that their definition in so small a spec- 

 trum should be as perfect as possible. The photographic plates 

 in these early experiments were only 1^-inch long by ^-ineh 

 wide. This requisite precision in the adjustments arises partly 

 on account of the small quantity of light at the disposal of the 



* The labour involved in work of this delicate nature is very great, as it is 

 necessary to keep a constant watch on the star's image when projected upon the 

 slit of the spectroscope, in order to correct by hand any small inequalities of the 

 motion of the telescope, which might throw the star off the slit. When it is 

 considered that this slit may be only -g-^-g-th of an inch in width, and that the 

 star's image must remain steadily upon it for perhaps an hour or more, it is not 

 difficult to perceive how necessary it is to have the power of continuous super- 

 vision, and of instant control by hand, in order to give sufficient time to record 

 the character of the spectrum upon a gelatine plate of extreme sensitiveness. 



