260 Mr. S. Bidwell. On Changes produced by [Mar. 25, 



actuates a short arm E attached to the hack of a small circular mirror 

 M; the mirror is capable of turning about its horizontal diameter 

 upon knife-edges, resting upon brass planes not shown in the figure. 

 By means of a lantern illuminated by a lime-light, the image of a 

 horizontal wire is, after reflection from a mirror, projected upon a 

 distant vertical scale ; a very slight deflection of the mirror causes a 

 considerable movement of the image. The dimensions are as follows : — 

 The distance SD = 10 mm., SE = 170 mm., ME = 7 mm. ; the dis- 

 tance from the mirror to the scale = 6400 mm., each scale division 

 = 0*64 mm., and the length of the experimental rod between the 

 clamps == 100 mm. The movement of the focussed image through one 

 scale division therefore indicates a difference of about one five- 

 millionth part* in the length of the rod. The mirror is very ac- 

 curately worked, and is silvered upon its outer surface ; the lens used 

 for the projection is a compound achromatic of high quality, and it is 

 easy to read with accuracy to a half or even a quarter of a scale 

 division. 



The ndagnetising coil is 11 '5 cm. long between the boxwood 

 ends; its external diameter is 5*2 cm. and internal diameter 1*9 cm. 

 A current of C amperes produces at its centre a field of about 92 C 

 units. I 



It will 'be seen from the above description that the wire under 

 examination sustains the whole weight of the magnetising coil as well 

 as that of the lever R. In order to ascertain the amount of the 

 tension thus produced, the brass rod P was suspended from a hook 

 beneath one pan of a large balance, and it was found that in order to 

 maintain the lever R in a horizontal position it was necessary to 

 place weights amounting to slightly more than 3 lbs. in the other 

 scale pan. In all experiments with this apparatus, therefore, the 

 iron wire is stretched by a minimum initial load of 3 lbs. For some 

 reasons this is a disadvantage, and the arrangement in question was 

 not adopted until many experiments had made it evident that by no 

 other method was it possible to avoid with certainty the source of 

 error introduced by the electromagnetic action commonly known as 

 solenoidal suction between the coil and the wire. When the coil is 

 fixed independently of the wire, the smallest trace of this action 

 produces upon the lever an effect which is enormously exaggerated in 

 the deflection of the image upon the scale. The wire may be placed 

 as accurately as it is possible to do so by measurement, with its 

 middle point in the centre of the coil ; but a change in the stretching 

 weight will at once displace it to a small but material extent ; and 

 even if the geometrical coincidence were perfect, a slight want of 

 uniformity in the physical qualities of the wire would still render 

 the objectionable action possible. Under ordinary circumstances the 

 * More exactly 0-00000020588. 



