321 



the square of the length, and it was predicted that in em- 

 ploying a traction tube three miles long, a velocity of twelve 

 miles per hour could not be attained at three miles from the 

 air pump, on account of the resistance of the friction of the 

 air in the pipe : but the parties who made such calculations 

 appear to have overlooked the fact, that as air in tube be- 

 comes more rare, the resistance is diminished, probably in 

 the ratio of the squares of the density, and also that air may 

 offer less resistance to the action of pumping it out of a tube 

 than to that of forcing it through; in like manner as sand may 

 be poured out of a tube, but is only with great difficulty 

 forced through. 



But practice is better than theory, and the triumphant 

 speeds obtained at Croydon prove that the friction of the air 

 in the tube does not prevent the attainment of a high speed, 

 but is merely a source of the loss of some power, inasmuch 

 as the barometric gauge at the end of the tube stands some- 

 what lower at the end of the tube than in the engine-house, 

 I believe about two inches, thus showing a loss of about 10 

 per cent. ; but this amounts to little more than a retardation 

 during the prior exhaustion, and also to some extent during 

 the running of the train, and is rapidly diminished as the 

 piston advances in the pipe, so that five per cent, may cover 

 the amount of the loss by friction of the air in the pipe. 



We have, therefore, the following losses per cent. 



Friction of engine and pump 10 



Adhesion of pump valves, and heating of air 10 



Leakage 15 



Friction of travelling piston 2 



Friction of air in traction tube 5 



Together 42 



But the three latter items are deductions from the residual 

 power after 20 per cent, has been deducted; their amount 



A A 2 



