Sornas—On the Law of Gladstone and Dale. 165 
Taking the following equivalents, Cu, = 11:72, Cl, = 11 
Ky = 8:2, H~= 1:3, O, = 2:8, we have >,:= 82°72; and for 
molecular weights, Cum, = 63°15, Cl, = 35°37, K,, = 39°03, On 
= 15°96, H,, = 1, we have 3, = 318°64 and 82:72 / 318°64 
= 0°2596 = «; also 
1:623 - 1 
1°6311 x 2+ 1°6070 / 3 = 1:6231 and RRO aa 0:2596 = k. 
Then for 
Cu,ClH, =“ = 14:253, and &,, = 46:9, and =" = 0°3082 = ky, 
(n,) 1:6811 -—1 / 0°3082 =d, = 2:081, and 46:9 / 2: a4 = 9, = 22°53. 
Also for 
Cu,OK, 34 = 12-958, Sp = 65°63, and = Eile ues 
(n.) 16070 / 0:1974 = d, = 3:025, and 65-68 / 3:025 = » = 21.42. 
Finally, 
v, x 4 = 90°12, and », x 2 = 42°84, and 90°12 + 42°84 = 1382°9. 
But 318°64 / 2°4 (the density of the salt) = 132°8. 
There is thus between the sum of the volumes of the com- 
ponents and the value of the volume of the whole salt a corre- 
spondence almost exact. 
The special interest of the salt lies, however, in the fact that 
it is dichroic, displaying a green tint when illuminated by the 
ordinary ray, and a sky-blue when seen by the extraordinary ray. 
If our hypothesis have in it any truth, then this difference in tint 
should be correlated with the difference in chemical constitution 
which we have imagined to exist in the direction of the vertical 
and horizontal crystallographic axes. Let us see. Along a 
vertical range lies Cu,OK; this gives for the copper-oxygen 
compound, Cu,0,, or CuO; ;* along a horizontal range, Cu,Cl,, or 
CuCl; ;! evidently the copper acting vertically is more of a cupric 
nature than that acting horizontally, but cupric salts are charac- 
teristically blue, and cuprous salts are as commonly green in 
colour. Thus the correspondence predicated actually exists, and 
in stumbling on this confirmation of our hypothesis we have 
discovered a theory of pleochroism. ‘The commonest pleochroic 
1 Of course these are impossible compounds, and are only to be regarded as 
shadowing forth the true relations, which are more fully discussed in the Memoir. 
