818 
MR. H. TOMLINSON ON THE INFLUENCE OF STRESS AND 
Now, if we allow for the difference of temperature, the value of a given in (11) will be 
found to accord almost exactly with the values of a given in (9) and (10). 
The effect of change of temperature is very considerable, for a rise of 6 0, 85 C. 
caused an increase of 10‘05 per cent, in the internal friction, and proved to be 
nearly the same for the three vibration-periods: 1 - 913 second, 4’366 seconds, and 
7’989 seconds. 
With zinc the logarithmic decrement also increases with the vibration-period, but 
not to the same extent as with tin, the formula in this case becoming 
X T ='01019+'0022274 t-”0000941t 2 . ...... (12) 
With the metals possessing less internal friction, steel and annealed copper, the 
variation of the logarithmic decrement with the period of vibration is so slight as to 
be masked by errors due to small fluctuations of temperature* and other causes ; 
and, had not a great many other observations than those here recorded proved 
that the internal friction does slightly increase with the vibration-period, it would 
have almost appeared not to do so. If, however, we compare in Experiments III. 
and IV. the mean value of the first half of the logarithmic decrements with that of 
the second half, we may trace indications of the truth of the above statement. To 
sum up, then, these and many other experiments of a like kind have nearly satisfied 
me that, with all metals, the logarithmic decrement does increase with the period 
of vibration, but that, the less internal friction the metal possesses, the more nearly 
is the former independent of the latter. 
Effect of Fluctuations of Temperature on the Internal Friction. 
Perhaps one of the most remarkable features presented by a torsionally vibrating 
wire is the ease with which the molecules are disturbed by changes of temperature 
from those positions of stable equilibrium into which they will, after a period of time, 
whose length depends upon the nature of the metal, naturally creep. It is truly 
astonishing to observe the profound change in the internal molecular friction of such 
an elastic metal as unannealed piano-steelt which is wrought by a sudden rise or fall 
of one or two degrees of temperature. I have explained that, though the wires were 
shielded by a long box, this box was itself exposed, for the greater part of its length, 
to the external air. Now, the first wire examined was one of unannealed piano-steel, 
and some hundreds of trials were made with it during a period of three months, and 
* See what follows shortly. 
t I do not know that unannealed piano-steel is more remarkable than other metals in the respect 
mentioned.—H. T. 
