and Vegetable Colouring -matters by the Spectrum Microscope. 1 47 



are at once apparent that otherwise might have been entirely over- 

 looked ; and on this account I shall adopt this system in those figured 

 in this paper. It is, in fact, merely representing the actual measure- 

 ments by drawings, without being at the trouble of distorting them, 

 so as to be like naturally distorted spectra. 



Since the number of divisions depends on the thickness of the in- 

 terference-plate, it became necessary to decide what number should 

 be adopted. At first I thought that ten would be most suitable ; but, 

 on trying, it appeared to me too few for practical work. Twenty 

 is too many, since it then becomes extremely difficult to count them. 

 Twelve is as many as can be easily counted ; it is a number easily 

 remembered, gives sufficient accuracy, and has a variety of other 

 advantages. With twelve divisions the sodium-line d comes very 

 accurately at 3 J ; and thus, by adjusting the plate so that a bright 

 sodium-line is hid in the centre of the band when the Nicol's prisms 

 are crossed, it is accurately at 3| when they are arranged parallel 

 so as to give a wider field. The general character of the scale will 

 be best understood from the following figure, in which I have num- 

 bered the bands, and given below the principal Fraunhofer lines. 



12 3 4 5 



10 11 12 



Blue end. 



The centre of the bands is black, and they are shaded off gradually 

 at each side, so that the shaded part is about equal to the inter- 

 mediate bright spaces. Taking, then, the centres of the black bands 

 as I, 2, 3, &c, the centres of the bright spaces are H, 2£, 3 J, &c, 

 the lower edges of each f, If, &c, and the upper 1|, 2|, &c. We 

 can easily divide these quarters into eighths by the eye ; and this 

 is as near as is required in the subject before us, and corresponds as 

 nearly as possible to y^- part of the whole spectrum visible under 

 ordinary circumstances by gaslight and daylight. Absorption-bands 

 at the red end are best seen by lamplight, and those at the blue end 

 by daylight. 



On this scale the position of some of the principal lines of the 

 solar spectrum is about as follows : — 



A.... | B....-1* C....2I D 



5H b ....6J* 



.7i. 



G 



io| 



At first I used plates of selenite, which are easily prepared, because 

 they can be split to nearly the requisite thickness with parallel faces ; 

 but I found that its depolarizing-power varies so much with the tem- 

 perature, that even the ordinary atmospheric changes alter the posi- 

 tion of the bands. Quartz cut parallel to the principal axis of the 

 crystal is so slightly affected in this manner as not to be open to this 



L2 



