134 EXPERIMENTS ON FRICTION. CHAP. VIII. 



position, had at all times nearly the same amount of 

 weight to bear, and consequently the entire weight was 

 pretty equally distributed amongst the four wheels of 

 the locomotive. Thus the four floating pistons were 

 ingeniously made to serve the purpose of springs in 

 equalising the weight, and in softening the jerks of the 

 machine ; the weight of which, it must also be observed, 

 had been increased, on a road originally calculated to 

 bear a considerably lighter description of carriage. This 

 mode of supporting the engine remained in use until 

 the progress of spring-making had so far advanced that 

 steel springs could be manufactured of sufficient strength 

 to bear the weight of locomotive engines. 



The result of the actual working of the new locomo- 

 tive on the improved road amply justified the promises 

 held forth in the specification. The traffic was con- 

 ducted with greater regularity and economy, and the 

 superiority of the engine, as compared with horse trac- 

 tion, became still more marked. And it is a fact worthy 

 of notice, that the identical engines constructed by 

 Mr. Stephenson in 1816 are to this day to be seen in 

 regular useful work upon the Killingworth Railway, 

 conveying heavy coal-trains at the speed of between 

 five and six miles an hour, probably as economically 

 as any of the more perfect locomotives now in use. 



Mr. Stephenson's endeavours having been attended 

 with such marked success in the adaptation of locomotive 

 power to railways, his attention was called by many of 

 his friends, about the year 1818, to the application of 

 steam to travelling on common roads. It was from this 

 point, indeed, that the locomotive had first started, 

 Trevithick's first engine having been constructed with 

 this special object. Stephenson's friends having ob- 

 served how far behind he had left the original projector 

 of the locomotive in its application to railroads, perhaps 

 naturally inferred that he would be equally successful 

 in applying it to the purpose for which Trevithick and 



