2 1)6 . Prof. W. Stroud. ) [June 19, 



no observation can be made unless the needle is within at the most 

 2 mm. of the centre of the magnet. 



To read the deflections 0, 0, a lamp and scale, or telescope and scale, 

 may be used. A slight difficulty arises with a single telescope and a 

 single scale when setting up the instrument for the first time, owing 

 to the two mirrors not usually making the same or sufficiently nearly 

 the same angles with the vertical. Either two telescopes and one 

 scale, or two scales and one telescope, may be used ; but the best 

 plan is to use one telescope and one scale, and to bend the aluminium 

 wire supporting the lower mirror till the latter occupies a suitable 

 position with respect to the vertical. 



Corrections will have to be made for (1) the torsion of the silk 

 fibre susoending the needle, (2) the torsion of the silk fibres suspend- 

 ing the magnets, and (3) the couple which the little needle exerts on 

 the magnet. So far as the first two corrections are concerned, they 

 can be allowed for in the ordinary way. For the present instrument 

 the first correction affects H to the extent of one part in a thousand ; 

 the second is utterly negligible. It is a matter of interest to deter- 

 mine the magnitude of the couple exerted by the needle on the 

 magnet compared with that exerted by the earth. Now, the earth's 

 couple on the magnet = MH cos 6, and on the needle = mH sin 0, so 

 that the fractional error in equation (1), made by neglecting this 

 effect, would be m sin 0/M cos 0, or, practically, m0/M, which, for the 

 present instrument, would amount to about 3^, since by experiment 

 M = 91 c.g.s. units, and m = 1'5 c.g.s. units and = 10. This 

 correction is then, by no means, negligible, since it would affect H to 

 the extent of 1 in 700. The error arising from this source could, 

 however, be made very much smaller by diminishing m or, preferably, 

 by increasing M. 



The following results have been obtained for the value of H in the 

 physical laboratory of the Yorkshire College, Leeds, which was 

 designed by Professor Riicker, so as to be as free as possible from 

 iron which could not be removed if necessary. The instrument was 

 set up in the middle of the room, the nearest iron being some 4 metres 

 distant, and consisting of a grate, roughly in the same magnetic 

 meridian as the instrument. This will clearly give a higher value 

 for H than if the grate had been removed ; but the object of the 

 experiments was to determine H, not for Leeds, but for one place in 

 the physical laboratory. 



I. May 17th, 1890. Scale, a metre long, placed roughly 97 cm. 



from the centre of the instrument. 



cm. 



Reading for magnet before reversal . . 10'33 

 after 52'59 



Difference . 42'26 



