A NEW CURRENT WKIUHKK, ETC. 477 



angles, so that there was no mutual induction t>etween the members of one pair and 

 those of the other pair. 



After the complete ampere balance was set up at the National Physical Lalwratory, 

 further magnetic tests were made on the stand and surrounding parts by sending* 

 current of 1 ampere (approximately) round one of the suspended coils only, and 

 observing whether the rest-point of the balance was affected thereby. The same test 

 was made with the current reversed, and the whole repeated on the other suspended 

 coil. In neither case could any change in the rest-point be detected. Experiments 

 were also made by bringing masses of iron in proximity to the balance when the 

 suspended coils were carrying their normal current. The effects of these masses were 

 much smaller than expected ; in fact, the iron had to be placed very near a current- 

 carrying coil to produce any observable change on the rest-point. It may, therefore, 

 be concluded that there can be no appreciable error in the t>alance due to magnetism 

 or diamagnetism of the phosphor-bronze support. 



Magnetic tests on marble were made at the Central Technical College in 1897, 

 using the large marble cylinder employed in the Lorenz apparatus constructed by 

 Messrs. N ALDER BKOS. & Co. for the McGill University, Montreal.* Its permeability 

 differed from that of air by an amount too small to be detected, t This fact, together 

 with the high specific resistance of marble, decided the material to be used for the 

 cylinders of the proposed current weigher. 



All the marble used in the fixed and suspended cylinders of the ampere balance 

 was tested at the National Physical Laboratory, when received from the merchants, 

 by observing the swing (if any) of a galvanometer in the secondary circuit of a pair of 

 coils when the marble was quickly inserted as a core, the current in the primary 

 circuit being kept quite constant. The primary coil had 1000 turns of No. 32 S.W.G. 

 copper wire, and the secondary 10,000 turns of No. 42 S.W.G. With a current of 

 0'5 ampere in the primary the arrangement was extremely sensitive, as a change in 

 the primary current of 1 part in 10,000 produced a swing of 5 '4 millims. The scale 

 could be read to 0'2 millim., so that a change of flux of 1 part in 270,000 could be 

 detected. The tests showed that the permeability of the marble did not differ from 

 that of air by 1 part in 100,000, a result which is in agreement with the American 

 measurements mentioned above. 



By means of the same coils the susceptibility of solid ferrous sulphate was measured 

 as 73 x 10"*, crystallised salt being used, and the air space determined by the aid of 

 alcohol. KONIGSBERGER! gives 37 x 10~* as the susceptibility of powdered ferrous 

 sulphate. 



* 'B.A. Report,' 1897, p. 218. 



t More recently tests made in America by WILLS, GUTHK, and STKBBINS, show the magnetic 

 susceptibility of several kinds of marble to be extremely small, probably less than 1 x 10"'. Se ' Bulletin 

 of Bureau of Standards,' vol. 2, pp. 52, 89. 



I WlED. Ann.,' 66, 698, 1898. 



