CHAP. XVII. UNDULATING RAILWAYS. 307 



he held that an undulating railway was much better than 

 a level one for purposes of working. 1 For a time, this ) 

 theory found favour, and the " undulating system " was / 

 extensively adopted ; but Mr. Stephenson never ceased 

 to inveigh against it ; and experience has amply proved 

 that his judgment was correct. His practice, from the 

 beginning of his career until the end of it, was to secure 

 a road as nearly as possible on a level, following the 

 course of the valleys and the natural line of the country ; 

 preferring to go round a hill rather than to tunnel under 

 it or carry his railway over it, and often making a 

 considerable circuit to secure good, workable gradients. 

 He studied to lay out his lines so that long trains of 

 minerals and merchandise, as well as passengers, might 

 be hauled along them at the least possible expenditure 

 of locomotive power. He had long before ascertained, 

 by careful experiments at Killingworth, that the engine 

 expends half of its full power in overcoming a rising 

 gradient of 1 in 260, which is about 20 feet in the mile ; 

 and that when the gradient is so steep as 1 in 100, not 

 less than three-fourths of its propelling power is sacri- 

 ficed in ascending the acclivity. He never forgot the 

 valuable practical lesson taught him by those early 

 trials made and registered in the company of Nicholas 

 Wood, long before the advantages of railways had 

 been recognized. He saw clearly that the longer flat 

 line must eventually prove superior to the shorter line 

 of steep gradients as respected its paying qualities. He 

 urged that, after all, the power of the locomotive was 

 but limited ; and, although he and his son had done 

 more than any other men to increase its working 

 power, it provoked him to find that every improve- 

 ment made in it was neutralised by the steep gra- 

 dients which the new school of engineers were setting 



1 ' Treatise on Railway Improvements.' By Mr. Richard Badnell, C.E. 



