STRAIN ON THE PHYSICAL PROPERTIES OF MATTER, 
835 
tudes through which the logarithmic decrement is calculated diminish, and, moreover, 
the range through which the logarithmic decrement is a constant becomes wider as the 
amount of “ elastic fatigue ” is diminished. We may see here analogies between the 
increase of the proportionate diminution of amplitude caused by “ elastic fatigue ” and 
the increase of the ratio of the temporary molar deformation, whether torsional or 
longitudinal, to the stress producing it, as the latter increases vdrenever the wire acted 
upon has recently suffered permanent longitudinal extension or torsion.* 
Lastly, it is desirable to draw attention to the marvellously small deformations 
which can produce “elastic fatigue” in nickel as compared with those which are 
attended with corresponding effects in the case of the other metals. 
The Effect of Rest and of frequent Oscillations on the Internal Friction. 
It has been already observed that rest after strain of any kind invariably diminishes 
the internal friction, but rest may be considerably aided in this effect by agitating the 
molecules either by thermal or by mechanical agency. With the part played by 
thermal agents I shall deal in a later memoir, and now turn to agitations produced by 
oscillating the wire torsionally. 
Experiment XXIV. 
The wire used in Experiment XVI., and which had suffered the permanent exten¬ 
sion mentioned in this experiment, was allowed to rest for 24 hours with a load of 
12 kilos, on, and w-as then tested with the same load still on. 
Logarithmic 
decrement.t 
Range of arc. 
Number of oscillations 
before the 
logarithmic decrement 
was determined. 
•003050 
537-4-500-9 
4 
•003108 
411-5-346-6 
40 
•002617 
346-6-299-9 
64 
•002875 
299-9-255-8 
88 
•002905 
255'8-217’9 
112 
•002553 
217-9-189-2 
136 
Started again with an initial arc of 600 
scale-divisions, and left vibrating until the 
arc was reduced to 379'0 scale-divisions. 
•002033 
379-0-302-7 
•001911 
302-7-245-1 
* ‘ Phil. Trans.,’ 1883, Part I., pp. 7, 8. See also Wiedemann’s paper loc. cit. 
-j- In this experiment the effect of the resistance of the air was not eliminated, as it was impossible to 
have done so accurately, and the elimination would have been of little importance. 
5 O 2 
