Rolling friction is of a different kind. The intermeshing inequalities or microscopic teeth are lifted out of contact with, one another just as the teeth of gear wheels are carried out of mesh. But there is another cause of friction due to the fact that no objects are so microscopic that they do not sink into each other to some extent. A wheel is always rolling out the surface it is turning on just as a rolling pin rolls out dough. If the surface is of elastic material such as a steel rail, it springs back into place immediately after the passage of the wheel, but the wheel must constantly travel in the trough of a wave which accompanies it along the rail. There is a similar wave in the wheel itself and this ironing-out action produces heat in the wheel and the rail. It is particularly noticeable in the flexible tires of automobiles which, after a run on even a smooth road, become too hot to be grasped with the bare hand merely because of the waves of compression and decompression to which they are subjected. Both rolling and sliding friction are increased by pressure because the depression is greater and because inequalities are brought into more intimate engagement with one another. The degree of friction also depends upon the nature of the substances in contact, but theoretically the area of contact does not make any difference. It is just as hard to push a block along a smooth surface on its edge as on its side. PUTTING ROLLERS BETWEEN LOAD AND ROAD As intimated in Chapter I the forerunner of the wheel was probably the roller. It was much easier to move a heavy object on rollers because rolling friction was substituted for sliding friction, but the rollers would not stay under the object; they traveled only half as fast as the load they carried. To make them keep up with the load they had to be mounted on axles which were fastened either directly to the load or to a cart body on which the load was supported. Thus the wheel came to be invented, but except for the fact that it stays by its load and does not roll out from under it a common wheel is not to be compared with a roller for efficiency. To be sure, it substitutes rolling friction for sliding friction where it contacts with the road, but the friction at the axle is sliding rather than rolling. However, drawing an object on wheels is a decided improvement over sliding it along the road, for two reasons: the sliding friction at the axle is reduced to a minimum by choosing materials that will slide upon each other with comparatively little resistance, by polishing them smooth and by lubricating them. But even if these precautions were not taken there would be a distinct advantage in the use of wheels because of the relatively shorter travel at the axle than at the rim of the wheel. If a wheel is thirty inches in diameter and turns on an axle one inch in diameter, it will travel thirty times as far at the rim as it does at the axle; hence the