Proceebings of the Polytechnic Association. 941 



the shaft. A A represent the brake wheel, made in two parts, and 

 bolted together as represented. B B represent the two cheeks refer- 

 red to. C represents a pin, inserted throngh both the boss of A, and 

 also through B, into a shallow hole bored for the purpose. In each 

 of the shafts, D, D, D, is an iron strap, to the inside of which is 

 secured a lagging of hard wood, wliich, fitting tightly to the wheel, 

 forms the brake. The lever, E, having a fulcrum at E', is held in a 

 horizontal, or nearly horizontal position, by the spring balance, F. 

 The corresponding lever, G, attached in a similar manner to the other 

 end of the strap, D, was variably weighted, according to the amount 

 of resistance required. The rubbing surfaces were frequenty "lubri- 

 cated " witli water, by leading a small stream through a hose to the 

 highest point in the wheel. 



f 



The strap, D, is therefore pulled vertically downward at each 

 extremity, and by subtracting from the effect of G the effect of E 

 (which latter, at any moment, was to have been ascertained by read- 

 ing the scale, F), the true resistance in pounds offered to the rotation 

 at the surface of A, was intended to have been easily ascertained. As 

 the experiments finally resulted, this portion was an entire failure ; 

 and the fiict goes to argue with some force against the practicability 

 of making scientific experiments under such circumstances. A pre- 

 liminary test of all portions of the apparatus was made in advance, on 

 each of tlie engines ; but in repeating the experiments formally, on 

 the days of trial, the scale, F, which had been thus tested, was 

 exchanged for another; and on account of this apparentl_y 'trifling 

 circumstance, all expectation of deducing the power from the friction 

 brake was, as it ultimately proved, completely frustrated. The new 

 scale employed, which was supposed to be better than its predecessor, 

 because more readable at a distance, employed an index, which, as it 

 subsequently appeared, became deranged, and shifted its place on its 

 axis at each of the many slight blows and collisions to which it was 

 subjected. Tliat portion of the notes taken during the trials, which 

 were intended to show the pull due to the^lever, E, were at different 



