3S2 A. W. Ewell — Rotatory Polarization. 



described above. The amount of twist was measured by means 

 of a mirror attached to the lower wood block and a distant 

 telescope and horizontal scale. Successive twists in both direc- 

 tions were applied up to a twist of 3°'l produced by weights 

 of :2030 s applied by the cords to the opposite sides of the 

 lower clamp of 4:-58 cra diameter.* On applying this twist in 

 the opposite direction the glass rod broke. No evidence of 

 rotatory polarization was observed. Observations reported in 

 a previous paperf indicated that twist produces rotatory 

 polarization in glass, but glass is obviously a very unsuitable 

 material, since we have seen above that at least in jelly tubes 

 the rotation is closely proportional to the fourth power of the 

 twist and therefore appreciable results would hardly be expected 

 with glass, which only permits the smallest twists. 



The rigidity of these jellies was carefully studied. 

 Relation between Twisting Couple and Tioist. 



Mar. 1]. Independent jelly cylinder. Length = 5*6, diameter = 



2*0. Moment of couple = weight applied to each cord X 4*58, 



the diameter of the lower clamp upon which the cord was 



wound. The same couple was applied in both directions. 



Moment 22'8 45*8 96 169 238 306 376 466 561 



— Twist 9 40 73 102 129 147 179 207 



+ Twist 11 21 44 73 102 128 152 186 209 



These measurements are plotted in fig. 7, I. 



Independent cylinders could not endure such twists if the 

 cylinders were vertical supporting the lower clamp; a modified 

 form of apparatus was used in the above measurements, in 

 which the cylinder was horizontal and the movable clamp 

 rested upon friction wheels similar to those of an Atwood's 

 machine. 



July 21. Independent jelly cylinder ; (approximate) length = 8, 

 diam. = 2*0. 



Moment 22-6 48'7 75 101 



+ Twist 33 69 110 145 



— Twist ... 32 71 111 148(broke) 



These measurements are plotted in fig. 7, II. 



These results show that the rigidity remains very nearly 

 constant for twists of the general magnitude of those which 

 produced the rotations, and before the rotation is in practically 

 the same relation to the twisting moment as to the twist. 



It would have been interesting to have measured the rigidity 

 of jelly in rubber tubes but this did not prove feasible. The 

 soft rubber tube is so much more rigid than even the stiffest 



* These figures give as the rigidity of this glass 2*97 x 10 11 . 

 f This Journal (3), viii, 89, 1899. 



