198 EIGHTH REPORT — 1838. 



duced on a straight and level railway by the combination of all 

 the above-mentioned causes. It is a matter of regret that the 

 obstacles to experiment which are produced by the great amount 

 of traffic on the principal railways are sucli that, notwithstanding 

 the lapse of time wliich has taken place since the commencement 

 of this inquiry, means have not been obtained for making such 

 an extensive and various course of experiments as would be sufli- 

 cient to solve this question. Besides the obstacles produced by 

 the traffic on the different lines of railway, difficulties also arose 

 in obtaining the means of experimenting, owing in some cases 

 to the inability of railway companies to spare the necessary en- 

 gines, carriages, and wagons. It is nevertheless due to these 

 companies to state their general willingness to facilitate the in- 

 vestigation, and this acknowledgement is especially due to the 

 Boards of Directors of the Grajid Junction and the Liverpool 

 and Manchester Railway Companies. 



Three methods for discovering the amount of resistance op- 

 posed by a train to the tractive power have been proposed : 



1". By a djmamometer, interposed between the tractive power 

 and the load, which should measure and record the force exerted 

 by the tractive power in drawing the load along a level and 

 straight line of railway. 



2°. By observing the motion of a load down an inclined plane 

 sufficiently steep to give it accelerated motion, and comparing 

 the rate of its acceleration with that which it ought to receive 

 from gravity, if it were subject to no resistance. 



3°. By putting a load in motion on a straight and level line 

 of railway, so as to impart to it a certain known velocity, and 

 then permitting it to run until it is brought to rest by the re- 

 sistance gradually destroying the velocity imparted to it. 



Each of these methods of experimenting was attended with 

 difficulties and objections. In practice, a line of rails is never 

 truly level. That which is commonly called level is a line 

 which, being examined from point to point at intervals — say of 

 quarter miles, is found neither to rise nor fall upon the whole. 

 But the surface of the rails along the intermediate parts is sub- 

 ject to considerable departures from an uniform level. How- 

 ever accurately they may have been laid when the line is first 

 constructed, the traffic upon them soon impairs their evenness, 

 and the inequalities of level become so considerable, that it fre- 

 quently happens that a wagon will not rest in certain positions 

 upon them, but will roll until its wheels get at the lowest point 

 of a part of the rail which has sunk. In the use of any form 

 of a dynamometer this circumstance produces extreme variations 

 in the index, so much so that in most of those which have been 



