1903 - 4 .] Mr Milne on a New Form of Spectrophotometer . 343 
p and q, fig. 4, which represent the intersections by the 
plane of the paper of the two spectra formed by the cylindrical 
lens C. 
In the case of a spectrophotometer where the light intensities 
are regulated and measured by means of a Yieroidt double slit on 
the collimator, the latter cannot be removed and placed in front of 
the absorption vessel without the loss of this means of controlling 
the light intensity. Of course the collimator might be left on the 
spectrophotometer and another collimator might be arranged in 
front of the absorption vessel, the two beams of light from the 
latter being directed upon the two Vieroidt slits respectively. 
With such an arrangement, however, the intensity would be 
reduced by the narrow openings of the Vieroidt slits, as well as 
by the small rectangular opening of the first collimator. The 
writer tried a modification of the above plan designed to obviate 
this loss, in which the second collimator being provided with 
Vieroidt slits, the latter were made to open at the maximum to a 
width equal to that of the two beams of light, while the lens of 
this collimator was discarded. Those changes are legitimate 
because the light rays have already been made parallel by the first 
collimator, and all that we wish to retain of the Vieroidt double 
slit collimator is its power to regulate the intensities of the two 
beams. The, difficulty with this plan is that the beam of light 
produced by the first collimator is apt not to have the same 
intensity at every point across a normal section, and if, for example, 
the jaws of one of the slits be closed together till only the half of 
that beam is permitted to pass through, we shall not in general 
have reduced the total light intensity of that beam by one half. 
The uniformity of the distribution of intensity in the cross 
section of the beam of light after leaving the first collimator 
depends to a large extent on what source of light is employed ; 
lime light, owing to the small area of its light source, being 
markedly inferior for such a purpose to a flat acetylene flame. 
Even with the latter, however, a doubt may exist as to the perfect 
equality of the intensity throughout the cross section of the beam, 
and so this modification of the Vieroidt double slit was abandoned 
and another device for intensity regulation was substituted which 
will be discussed later. 
