282 ON THE RIGIDITY 
Table IIT gives the results of the observations. There is a 
ereater uniformity in the way in which the values of the 
rigidity vary with change of tension than there was when the 
tension was increasing, which may be ascribed in part to the 
somewhat greater uniformity of the time intervals between 
changing the load and determining the rigidity, and in part to a 
greater permanence of internal structure produced by the 
previous prolonged extension. It will be noticed (i) that the 
values of the rigidity run through pretty much the same 
course as they did in Table II, when the tension was being 
increased, though the final values of Table III, in the case of 
the static rigidity for the greater angle of torsion and in the 
ease of the kinetic rigidity for the greater angles of oscillation, 
are less than the initial values of Table II, and (2) that the 
minimum point is given not only by the kinetic results for the 
amplitudes 180° and 90°, but also by the static resuits, which 
shows that the occurrence of the minimum point is not due to 
a defect peculiar to the kinetic method. 
To see what effect the time interval between the putting on 
of the load and the finding of the rigidity, had on the rigidity, a 
new cord was experimented with in the following manner:—A 
weight having been put on the cord, the rigidity was deter- 
mined both immediately afterwards and after the lapse of certain 
intervals of time. Then another weight was added and the pre- 
vious process repeated. Owing to lack of time, only the kinetie 
method was used. Table IV gives the results. 
