j 
NE ————S ——<—<—<<_- -——S — 
ON MOLECULAR PHENOMENA IN MAGNETISED IRON. 157 
at a lower temperature than it otherwise would, until at last a sort of 
explosive action occurs, and the change rapidly runs throughout the 
whole mass, analogous to what takes place in supersaturated solutions. 
This is what we may expect to occur in the magnetic metals which 
exhibit the phenomena of hysteresis when under stress.! The recent 
paper of Professor Ewing’s on the Molecular Theory of Induced Magne- 
tism,” to which we have already referred, throws remarkable light on the 
various phenomena we have been studying. By means of his beautiful 
experimental model, Professor Ewing has shown that the intermolecular 
magnetic forces alone are sufficient to account for the known facts of 
magnetisation, and that magnetic hysteresis is not due to anything in the 
nature of frictional resistance to the rotation of the molecular magnets, 
but simply to the molecular instability which results from these inter- 
molecular magnetic actions. And, further, that the same cause explains 
why there is ‘in magnetic metals hysteresis in physical quality generally 
with respect to stress, apart from the existence of magnetisation.’ We 
shall probably have occasion in our next report to deal more fully with 
Ewing’s explanation, and here can only congratulate the author on the 
value and beauty of his suggestive experiments. 
Connected with this part of the inquiry, we may refer to the inter- 
esting results Hopkinson has obtained with an alloy of iron and nickel. 
This alloy Hopkinson finds to have two stable conditions, one being 
magnetic and the other non-magnetic ; a high temperature destroys the 
magnetic state, which can only be resumed by lowering the temperature 
considerably below the freezing-point; the remarkable fact, now expli- 
cable by Ewing’s experiments, being that this iron-nickel alloy may be 
either magnetic or non-magnetic at the ordinary temperature, its previous 
history determining the state in which it remains.? 
Here we must leave the subject at present; we are well aware that 
many matters of interest have necessarily been omitted, and that we have 
inadequately dealt with those that have come under consideration. So 
many issues of importance, both to the chemist and metallurgist, as well 
as the physicist, have been opened up by this inquiry that we trust the 
Committee, which will be enlarged, may next year present a fuller report. 
APPENDIX. 
A proof of the foregoing report having been forwarded by us to Mons. 
Osmond he has sent us the accompanying notes, which we have thought 
desirable to add to the report :— 
, Page 145,—II ne parait pas possible que la récalescence fasse remonter 
la 
température au-dessus de a; , c’est-d-dire au-dessus du point 
_ Yéciproque pendant le chauffage; car, aussitét qu’on atteint ce point 
_ réciproque, il se produit une absorption de chaleur qui doit limiter la 
» In a paper by one of us (Newall), we pointed out some time ago that the pheno- 
_ mena observed in recalescence ‘ were really signs of something of the nature of what 
Professor Ewing calls hysteresis.’ 
2 Phil Maq., Sept. 1890. 
’ The electric resistance of this alloy at different temperatures is shown in the 
top curve of Fig. 2. When heated in a dry atmosphere of hydrogen the resistance 
regularly increases; when heated in an undried atmosphere a singular difference is 
observed during cooling as shown in the ‘ modified’ curve. 
