THE STEAM-ENGINE. 



533 



of the cylinder, the foot F, being placed on the road, is pressed backward by 

 the force of the piston-rod at K ; but the friction of the ground preventing its 

 backward motion, the reaction causes the engine to advance : and in the same 

 manner this process is continued. 



Attached to the thigh at N, above the knee, by a joint, is a horizontal rod 

 N R, which works a rack R. This rack has beneath, it a cog-wheel. This 

 cog-wheel acts in another rack below it. By these means, when the knee K 

 H driven from the engine, the rack R is moved backward ; but the cog-wheel 

 acting on the other rack beneath it, will move the latter in the contrary direc- 

 tion. The rack R being then moved in the same direction with the knee K, it 

 follows that the other rack will always be moved in a contrary direction. The 

 lower rack is connected by another horizontal rod with the thigh of the leg 

 L F', immediately above the knee at N'. When the piston is forced inward, 

 the knee K' will thus be forced backward ; and when the piston is forced out- 

 ward, the knee K' will be drawn forward. It therefore follows, that the two 

 knees K and K' are pressed alternately backward and forward. The foot F' 

 when the knee K 7 is drawn forward, is lifted by the means already described 

 for the foot F. 



It will be apparent, from this description, that the piece of mechanism here 

 exhibited is a contrivance derived from the motion of the legs of an animal, 

 and resembling in all respects the fore legs of a horse. It is, however, to be 

 regarded rather as a specimen of great ingenuity than as a contrivance of 

 practical utility. 



It was about this period that the important fact was first ascertained that 

 the adhesion or friction of the wheels with the rails on which they moved was 

 amply sufficient to propel the engine, even when dragging after it a load of 

 great weight; and that in such case, the progressive motion would be effected 

 without any slipping of the wheels. The consequence of this fact rendered 

 totally useless all the contrivances for giving wheels a purchase on the road, 

 such as racks, chains, feet, &c. The experiment by which this was deter- 

 mined, appears to have been first tried on the Wy lam- railroad ; where it was 

 proved, that when the road was level, and the rails clean, the adhesion of the 

 wheels was sufficient, in all kinds of weather, to propel considerable loads. 

 By manual labor it was first ascertained how much weight the wheels of a 

 common carriage would overcome without slipping round on the rail, and hav- 

 ing found the proportion which that bore to the weight, they then ascertained 

 that the weight of the engine would produce sufficient adhesion to drag after 

 it on the railroad the requisite number of wagons. 



In 1814 an engine was constructed at Killingworth, by Mr. Stephenson, 

 having two cylinders with a cylindrical boiler, and working two pair of wheels, 

 by cranks placed at right angles ; so that when the one was in full operation, 

 the other was at its dead points. By these means the propelling power was 

 always in action. The cranks were maintained in this position by an endless 

 chain, which passed around two cogged wheels placed under the engine, and 

 which were fixed on the same axles on which the wheels were placed. The 

 wheels in this case were fixed on the axles, and turned with them. 



In an engine subsequently constructed by Mr. Stephenson, for the Killing- 

 worth railroad, the mode adopted of connecting the wheels by an endless chain 

 and cog-wheels was abandoned ; and the same effect was produced by con- 

 necting the two cranks by a straight rod. All such contrivances, however, 

 have this great defect, that, if the fore and hind wheels be not constructed 

 with dimensions accurately equal, there must necessarily be a slipping or 

 dragging on the road. The nature of the machinery requires that each wheel 

 should perform its revolution exactly in the same time ; and consequently, in 



