326 REPORTS ON THE STATE OF SCIENCE, ETC. 
beyond the limits of elasticity . . . and still remaining of the nature of one 
continuous crystal. What... may be the nature of the change of molecular 
arrangement induced by bending them I cannot say; but I suppose that, in 
their yielding, their crystalline structure is materially altered, and rendered 
discontinuous where, before, it was continuous.’ 
Where Thomson used the word ‘ discontinuous,’ current theory substitutes 
‘ vitreous’ or ‘amorphous,’ with many associated ideas and more or less 
satisfactory definition. The physical and chemical properties of vitreous metals 
are still in some measure uncertain, because the vitreous phase tends to 
recrystallise; but it is known that the physical characteristics include great 
hardness, the limited mobility of a highly viscous fluid, and, in the case of 
iron, low magnetic permeability. The change is not merely a matter of 
pulverising. Without entering on controversial matter that has collected round 
the definition of an ‘allotropic’ change, it may be accepted that the change 
that occurs when a ductile metal is strained is purely physical in the sense 
that it involves only a rearrangement of the molecules without change of 
internal molecular structure. 
If this theory be accepted as an explanation of permanent strain, it is 
evident that the change from the crystalline to the vitreous state must precede 
the gliding action; the essential condition for gliding movement cannot be 
merely a consequence. Without denying the probability that further quantities 
of metal may svffer the same change during the gliding movement, it is clear 
that the initial change of state must be a direct result of the preceding elastic 
stage of straining. It follows that, in the course of any alternative process 
by which the metal can be strained to its different elastic limits by the applica- 
tion of different stresses, the metal that suffers the change must absorb—as heat 
or as mechanical work—such quantities of work as enable it to change from the 
stable crystalline state to the vitreous phase, which, at ordinary temperatures, 
is known to be ‘metastable,’ i.e. unstable but restrained from change by 
‘internal viscosity.’ 
Thermodynamic Principles. 
When any physical or chemical change is produced mechanically, by forces 
exerted by external bodies acting on a mass of the substance in question 
(c.g. by a piston acting on vapour enclosed in a compressor cylinder), the work 
done by the forces during the change is ordinarily greater than the quantity 
that can be regained by allowing the change to occur in the reverse direction 
at the same temperature. The difference between the two quantities is con- 
verted to heat by the action of friction or other ‘ irreversible ’ effects. Under 
ideal frictionless conditions the two quantities of work may be equal, and the 
process of change is then said to be ‘ reversible.’ 
A well-known application of the second law of thermodynamics states that 
when a given change can be produced by different reversible processes, carried 
out at the same temperature, the quantity of work that must be done by the 
forces acting on the substance must be definitely constant, i.e. independent 
of how the forces act. If one and the same change of physical state can be 
caused by stresses of different kinds, pull or shear or combinations of these. 
the quantities of work done by these forces on unit mass of substance suffering 
the change must always be the same, provided that the action is reversible 
and is carried out at the same temperature. If the action be not perfectly 
reversible, somewhat more work may have to be done; but the quantity will 
approximate to the ideal constant value if the action is nearly reversible. 
It is clear that this thermodynamic law affords a theoretical relation between 
the elastic limits of a ductile metal. To develop such a relation, however, it 
is necessary to equate expressions for the work done by different forces in 
changing unit mass of metal from the crystalline to the vitreous state in 
reversible and isothermal processes. 
Prof. James Thomson’s Work on ‘ Regelation.’ 
In 1849,3 in applying thermodynamic principles to problems involving change 
from the crystalline to the fluid state, Professor James Thomson showed that 
the melting-point of ice should be lowered by the application of fluid pressure: 
* James Thomson. 7’rans. Roy. Soc., Hdinhurgh, 1849. 
