The same tractive force that has previously been applied to increase the tractive adhesion now exercises its influence upon the brake shoes and wheels, with the result of not only causing a very powerful pressure between the two surfaces due to the magnetic attraction, but offering an extremely large frictional resistance in virtue of the molecular interlocking action before referred to. As shown in the present instance, a portion of the current still flows through the traction circuit and prevents the skidding of the wheels. The method thus described is equally applicable to increase the coefficient of friction in apparatus for the transmission of power, its chief advantage for this purpose being the ease and facility with which the amount of friction between the wheels can be varied to suit different requirements, or increased and diminished (either automatically or manually) according to the nature of the work being done. With soft iron contact surfaces the variation in friction is very rapid and sensitive to slight changes in current strength, and this fact may prove of value in connection with its application to regulating and measuring apparatus. In all cases the point to be observed is to maintain a closed magnetic circuit of low resistance through the two or more surfaces the friction of which it is desired to increase, and the same rule holds good with respect to the electric system, except that in the latter case the best effects are obtained when the area of surface in contact is smallest. For large contact areas the magnetic system is found to be most economical, and this system might possibly be used to advantage to prevent slipping of short wire ropes and belts upon their driving pulleys, in cases where longer belts are inapplicable as in the driving of dynamos and other machinery. Experiments have also been, and are still being, made with the object of increasing friction by means of permanent magnetism, and also with a view to _diminishing_ the friction of revolving and other moving surfaces, the results of which will probably form the subject matter of a subsequent paper.