388 Professor J. A. Ewing [May 22, 



When a piece of iron or nickel or cobalt is magnetised by 

 induction, the magnetic state permeates the whole piece. It is not 

 a superficial change of state. Break the piece into as many fragments 

 as you please, and you will find that every one of these is a magnet. 

 In seeking an explanation of magnetic quality we must penetrate 

 the innermost framework of the substance — we must go to the 

 molecules. 



Now, in a molecular theory of magnetism there are two possible 

 beginnings. We might suppose, with Poisson, that each molecule 

 becomes magnetised when the field begins to act. Or we may adopt 

 the theory of Weber, which says that the molecules of iron are 

 always magnets, and that what the field does is to turn them so that 

 they face more or less one way. According to this view, a virgin 

 piece of iron shows no magnetic polarity, not because its molecules 

 are not magnets, but because they lie so thoroughly higgledy-piggledy 

 as regards direction that no greater number point one way than 

 another. But when the magnetic force of the field begins to act, the 

 molecules turn in response to it, and so a preponderating number 

 come to face in the direction in which the magnetic force is applied, 

 the result of which is that the piece as a whole shows magnetic 

 polarity. All the facts go to confirm Weber's view. One fact in 

 particular I may mention at once — it is almost conclusive in itself. 

 When the molecular magnets are all turned to face one way, the 

 piece has clearly received as much magnetisation as it is capable of. 

 Accordingly, if Weber's theory be true, we must expect to find that 

 in a very strong magnetic field a piece of iron, or other magnetisable 

 metal, becomes saturated, so that it cannot take up any more mag- 

 netism, however much the field be strengthened. This is just what 

 happens : experiments were published a few years ago which put the 

 fact of saturation beyond a doubt, and gave values of the limit to 

 which the intensity of magnetisation may be forced. 



When a piece of iron is put in a magnetic field, we do not find 

 that it becomes saturated unless the field is exceedingly strong. A 

 weak field induces but little magnetism ; and if the field be strength- 

 ened, more and more magnetism is acquired. This shows that the 

 molecules do not turn with perfect readiness in response to the deflect- 

 in" magnetic force of the field. Their turning is in some way resisted, 

 and this resistance is overcome as the field is strengthened, so that 

 the magnetism of the piece increases step by step. What is the 

 directing force which prevents the molecules from at once yielding to 

 the deflecting influence of the field, and to what is that force due? 

 And again, how comes it that after they have been deflected they 

 return partially, but by no means wholly, to their original places 

 when the field ceases to act ? 



I think these questions receive a complete and satisfactory answer 

 when we take account of the forces which the molecules necessarily 

 exert on one another in consequence of the fact that they are 

 magnets. We shall study the matter by examining the behaviour of 



