526 
MESSES. FLEEMING- JENKIN AND J. A. EWING ON EEICTION 
steel (II.) and wet agate (IX.), the coefficient of friction remained entirely unaffected 
by changes in the velocity except where the bearings were made of wood, and that even 
in these circumstances no change of the coefficient could be detected so long as the 
surfaces were dry ; but when oil or water was present the coefficient of friction increased 
in a very marked manner as the velocity diminished, this increase of the coefficient of 
friction taking place under a limit of velocity of about 0-01 foot per second. It is to he 
observed that the cases in which this increase of the coefficient at low speeds occur are 
precisely those in which Coulomb and Morin have found that there is a very marked 
difference between the static and kinetic values of the coefficient of friction. In the 
case of dry metal surfaces both these writers are agreed that no difference can be detected 
between the friction of rest and that of motion ; and we find that in such cases no change 
takes place in the value of (m as the velocity varies. Morin also found that the friction 
between unctuous metallic surfaces was the same or nearly the same for rest as for motion, 
a result which agrees with the absence of almost any change of the coefficient of friction 
in our investigation of Cases II. & V. The friction between wet metallic surfaces is 
not spoken of by Coulomb, and received comparatively little attention from Morin, who 
does not say whether he found any difference between its static and kinetic values. 
We have not seen any account of experiments on the friction between metals and agate. 
As to the remaining sets of conditions, Morin says that he found no sensible difference 
between the static and kinetic values of the friction of dry metals on wood (Memoire 
i. pp. 104, 106). In fact, out of all the cases which we have examined, the only ones 
in which there is known to be a marked difference between the friction of rest and that 
of motion are those in which steel slides on oiled or wetted surfaces of wood, and in 
these cases, and these only, we have detected a very considerable increase in the 
coefficient of friction at very low speeds. 
Although the varying want of balance of the disk, which has been already referred 
to, was never so great as to throw any doubt on the fact that the rate of acceleration 
did change in these particular cases (and we are unable to see any cause for the change 
of the rate of acceleration except a change in the coefficient of friction), still it was 
considerable enough to make the determination of the precise way in which the 
coefficient changed with changes in the velocity a matter of the greatest difficulty. We 
have been able to draw only very roughly curves connecting the coefficient ^ with the 
relative velocity of the sliding surfaces. These all agree in showing that or the rate 
of change of the coefficient of friction relatively to change of velocity, becomes greater 
as the velocity becomes less. This being so it is perfectly possible that at velocities 
below 00002 foot per second, which is the lowest limit to which our observation may 
be taken as extending, a considerable further increase of the coefficient may take place 
before motion entirely ceases. 
It is in fact highly probable that in those cases in which the static coefficient of 
friction is greater than the kinetic (that is, the coefficient which is observed when the 
