OF THE DOUBLE SELENATES OF THE SERIES R.>M(SeO,).:,6HoO. 
2(19 
from the red end of the spectrum as far as wave-length 466 in tlie blue ; beyond this 
the symmetry axis h becomes the second median line. The first median line lies in 
the symmetry plane for all wave-lengths, in the acute angle of the morphological 
axes ac, and is inclined to the axis a 20° 30'. The second median line so long as it 
remains in the symmetry plane lies in the obtuse angle of the axes ac, and for 
sodium light (to which the measurements of extinction refer) is inclined 36° 47' to 
the vertical axis c. 
The double refraction is of positive sign. 
Refractive Indices .—These were determined with six excellent prisms, ground on 
six different crystals selected from three different crops. Each of the two prisms 
prepared to furnish a and /3, and whose refracting edge was in each case parallel to 
the second median line (for wave-lengths as far as 466) and vdiose bisecting plane 
was that containing this edge and the symmetry axis, appeared to exhibit only one 
image of the Websky slit of the refractometer instead of the usual two; and this 
image had the further pecidiarity of remaining permanent at all positions of the 
nicol, and for all wave-lengths, although its position naturally varied with the wave¬ 
length. (3n using tlie high-power eyepiece, the single image was clearly resolved 
into two images separated by l' for red light, approximating again to a single image 
as the greenish-lDlue was approached. One of the two images corresponded to a, and 
extinguished with the nicol at 90°; the other, corresponding to (I, extinguished 
when the nicol was rotated to 0°. For F-light the positions of the two images were 
identical, and for G-light the two images were found to have passed each other, the 
one which was formerly right being now left and vice versa, the amount of separation 
being about 1°. Hence for G-light the directions in the crystal parallel to which the 
a and (I vibrations occur are interchanged, compared witli tliose corresponding to 
wave-leno-ths on the red side of the crossino- wave-leno-th. 
O O O 
The actual results of the refractive index determinations are set forth in the 
accompanying table. It will he observed that the dilference between a and (3 works 
out to 0’0002 for Li-light. Considering the minuteness of this ([uantity, it is as 
well to hawe an independent check upon its accuracy, and this is alTorded by calcula¬ 
tion from the optic axial angle for this wave-length, assuming also the accuracy of 
one of the two indices, either a or and of course of y, with the aid of the 
formula 
The value of a tlius obtained is 1'51433, or OUOOlZ less than /3jj, a value for the 
difference which satisfactorily confirms that (0'()002) given in the table as the result 
of the determinations. 
The wave-length for whicli the indices a and ^ are truly identical is 466 in the blue, 
as will subse(|uently be shown in considering the optic axial angle. 
