294 Dr. Tyndall on the Laws of Magnetism. 



that on which the ball rests ; which circumstance, when uncom- 

 pensated by pressure, exhibits itself in the results. This being 

 the case, a small constant would have to be added to the distances 

 denoted by the leaves in those cases where the second mode of 

 experiment was adopted ; that is to say, in Tables VI. and VII. 

 The alteration thus effected will exhibit itself most sensibly in 

 the immediate neighbourhood of the poles, where alone the dis- 

 crepancies occur j indeed, its immediate effect will be to lessen 

 these discrepancies ; in Table VI. it will make the quotients at 

 the top of the columns less, and in Table VH. it Avill make the 

 products at the top of the column greater ; while at a distance 

 from the poles the change caused by the addition mil be scarcely 

 appreciable. There is another slight coi'rection to be made with 

 regard to the paper. When a good many leaves, say fifteen or 

 twenty, are upon the magnet, the space occupied by these is 

 somewhat more than fifteen or twenty times the space occupied 

 by one leaf. On trial mth the sphereometer, it was found that 

 fourteen leaves occupied nearly fifteen times the space of one 

 leaf. Taking this into account, and supposing the true pole in 

 the case of ball No. 2 to be about ^ono'^^'^ "^ ^'^ vach. beneath 

 the surface, and for the large ball a little deeper — for it must 

 be remembered, that the position of the pole depends as well 

 upon the surface of the ball as upon the magnet — the discrepan- 

 cies arising from the second mode of experiment mil entu-ely 

 disappear. 



40. Some of the laws expressed in the foregoing memoir will 

 naturally be limited liy the dimensions of the magnetic plane 

 from which the attraction proceeds, and also by the size of the 

 attracted ball. If the views of INI. Miiller, already adverted to, 

 be correct, a pohit of saturation for very small balls must soon 

 exhibit itself*, after which the law would be no longer appUcable. 

 As far as I am able to judge, the balls used in the present case, 

 although of very different volumes, follow the same law ; these, 

 however, and other matters of a similar bearing, I hope to make 

 the subject of a future commimication. 

 "* In stating the case, I have made use of those experiments 

 merely which appeared best calculated to illustrate the several 

 laws and the apparent deviations from them. To the latter, 

 indeed, I have given the most prominent place, as their expla- 

 nation appeared to me to be most important. The experiments 

 recorded constitute, however, but a small fraction of the number 

 actually made. My aim has been to embrace in one investigation 

 the whole of a subject whose separate details have occupied the 

 attention of many experimenters. One law alone of those ex- 



* Some interesting remarks on tliis subject occur in a paper by Professor 

 W. Thomson, Phil. Mag. vol. xxsvii. p. 241. 



