1878.] 



of determining Ratio of Dispersions. 



489 



zontal projection. I do not know that anything would be gained by 

 this ; I have found it sufficient to attend to the green. 



The light employed had best be that of the sky, reflected horizon- 

 tally by a looking-glass. It should be fairly bright, but not approach- 

 ing to dazzling. Thus the light reflected from the sky near the sun 

 would be "too bright. The room need not be darkened ; in fact, it is 

 better that the eye should be kept fresh for the appreciation of colour 

 by habitually looking about on ordinary objects. A simple collimating 

 lens such as I have described is sufficient, though doubtless an achro- 

 matic would theoretically be an improvement. It should be of longish 

 focus, at least in the case of a simple lens, lest any slight displacement 

 right or left of the middle of the incident beam should introduce a 

 minute dispersion due to the lens acting as a prism. If the middle of 

 the beam be not quite central, that does not signify, provided the 

 eccentricity be constant ; for then the minute dispersion is merely 

 added to that of the primary prism, which is arbitrary. To ensure 

 constancy of incidence on the lens, it is well to limit it by an aperture 

 with vertical sides, and to take care that the beam employed is wide 

 enough to fill the aperture. Similarly it is well to take care that the 

 beam falls centrically, or pretty fairly so, on the achromatic viewing 

 telescope. But what is of much more consequence is that the ray 

 passing through the optical centre of the object-glass should pass 

 centrically through the eye-piece, as otherwise the dispersion of the 

 eye-piece for eccentrical pencils would alter the secondary tint. It is 

 well, therefore, that the viewing telescope should be furnished with 

 cross wires. The telescope is then moved a little till the fiducial edge 

 is on the cross wires when the tint is observed. 



The determination, of the azimuth of achromatism is the capital 

 observation on which the accuracy of the result depends ; in com- 

 parison with it, the rest of the measurements required may be deemed 

 exact. Accordingly a number of observations of this azimuth should 

 be taken, and the mean adopted. The mean error will vary with cir- 

 cumstances, but it may be taken ordinarily as a few minutes. 



vol. xxvn. 2 K 



