414 DR. F. HORTON ON THE MODULUS OF TORSION A L RIGIDITY OF 



turned-up end serves as an index for counting the revolutions of the fibre. It is 

 viewed by means of a horizontal microscope not shown in the figure. The table D 

 has next to be adjusted in height and in horizoutality until the fibre rolls quite freely 

 on sliding the table F along its guides. When once these adjustments have been 

 made they serve for the whole fibre in fact, until it becomes necessary to renew the 

 glass capillary tubes on account of their becoming dirty. The small table D is covered 

 with a glass shade which extends over the tubes A and B and protects them from 

 dust (see fig. 3). 



The Zeiss scale is 1 centim. in length. It is divided into millimetres, the last of 

 which is subdivided into tenths. The scale was standardised by comparison with the 

 standard metre of the Laboratory, several centimetres of the latter being measured in 

 terms of the Zeiss scale at a temperature of 15 C. The agreement, after the 

 necessary temperature corrections had been made, was within the limits of experimental 

 error, and the scale was therefore taken to be correct at 15 C. The temperature of 

 the scale during the measurements of the radii of the fibres was noted, but the 

 temperature corrections to the length of the scale were always too small to be 

 observed. 



Before taking an observation it is first necessary to see that the scale is fixed on 

 the sliding table so as to be parallel to the direction of motion of the latter. In order 

 to enable this to be easily done, a strip of metal was screwed on to the table so as to 

 be accurately parallel to the slides, and the scale was held against this by means of 

 two brass springs. 



The whole of the apparatus, except the microscope M, is fixed on one heavy brass 

 base, and thus motion between the fixed table D and the moving one F, other than 

 that measured on the scale, is avoided. The apparatus can be levelled so that the 

 scale is horizontal. This is tested by the definition of the image in the microscope 

 remaining perfect as the scale is moved along underneath. The microscope M is 

 rigidly attached to a heavy cast-iron support, which rests on levelling screws, so that 

 the axis of the microscope can be adjusted into the vertical. The apparatus, when in 

 use, stands on a stone slab which cannot be moved by accidental blows. 



In taking an observation, the moving table is drawn back until the zero on the scale 

 has passed the cross-wire of the first or vertical microscope. The head H is then 

 turned until the zero is seen approaching the cross-wire, and then the fibre is viewed 

 through the second or horizontal microscope. The head H is further turned until the 

 index appears vertical, i.e., parallel to a vertical cross- wire in the second microscope. 

 This is a position which can be obtained to within a very small fraction of a revolution. 

 The image of the zero of the scale is then adjusted on to the cross-wire of the 

 microscope M by means of the rocking plate, and the position of the latter is noted 

 by reading the position of the arm P on the scale N. The head H is then slowly and 

 uniformly rotated and the revolutions of the fibre counted. This is continued until 

 such a number of exact revolutions have been made that the last '1 millim. of the 



