Opacity of Tourmaline Crystals. 12 



tion (15) becomes 



log _ fi 1 Y 1 

 logE H Y 2 ' 



(18) 



But fjb x and fi 2 are so nearly equal to 1 in all cases hitherto 

 observed, that the ratio of fi 1 to fi 2 is far less than that of V 2 

 to Y\i and therefore the greater opacity to the ray cannot be 

 accounted for on the supposition that it depends on differences 

 in the magnetic inductive capacity. 



19. There appears, then, to be no other explanation of the 

 observed opacity of coloured tourmaline crystals for rays 

 polarized in a principal plane of section than that afforded by 

 the hypothesis that the electric conductivity is greater in direc- 

 tions at right angles to the crystallographic axis than along 

 it. One fact of some importance bearing upon this point is 

 an observation which I have made, that the more opaque the 

 crystal in general the greater is the inequality of absorption 

 of the ordinary and extraordinary rays. The green tourma- 

 line " C " in my possession, which is about 2 millim. thick, 

 transmits both rays partially; but cuts off about three fourths 

 of their intensity from ordinary rays, and above half their 

 intensity from extraordinary rays. It is about 30 millim. in 

 length, and appears to be perfectly opaque in that direction. 

 On the other hand, a slice of a jet-black tourmaline, the pro- 

 perty of University College, Bristol, not more than 0*2 millim. 

 in thickness, is perfectly opaque to the ordinary ray ; but it 

 transmits extraordinary rays at about half of their intensity. 

 Not only must the conductivity as a whole be higher in the 

 latter case, but the ratio of the two conductivities must be 

 higher in the latter than in the former ; and the transverse 

 conductivity must be greater than the longitudinal. The 

 latter crystal was too small to permit of verification of these 

 inferences by direct experiment. 



III. Electric Conductivity of Tourmaline Crystals. 



20. The conducting-power of the tourmaline for electricity 

 is very feeble ; and the only determinations that have been 

 made of its relative conducting-powers in longitudinal and 

 transverse directions are determinations, not of the flux of elec- 

 tricity through a slab or film of the substance, but of the 

 superficial conductivity along a face of the crystal. This 

 superficial conductivity has been examined by two slightly 

 different processes, by de Senarmont* and by Wiedemann")". 



21. De Senarmont's process consisted in covering the face 



* Ann. de Chim. et de Phys. 3 e s<Srie, t. xxYiii. p. 257. 



t Ibid. t. xxix. p. 229 j and Pogg\ Ann. lxxvi. p. 404, and lxxvii. p. 534. 



