Proceedings of the Polytechnic Association. 947 



eiglit varieties of conditions very commonly met with, and if the results 

 could he carefully tahnlated or otherwise " salted down " for the 

 information of the engineering world, it would do much to extinguish 

 one of what the same immortal author designated, by a less Latin hut 

 almost e(|uall_y mysterious term, the " idol of the tribe." In plain 

 language mankind are prone to find a greater degree of order, 

 regularity simplicity, and certainty, than is actually indicated when the 

 facts are closely observed. It is an " idol " of the whole human race to 

 " go off at half-cock," and to base on a single experiment, or on 

 no experiment at all, a hideous pile of falsehood, in lien of the 

 beautiful truth which would be developed b}^ a proper combination 

 and application of patience and industry. The absence of gain by 

 expansion in the engines in question was partly due to its being, 

 under the circumstances, carried in every instance too far. jSTeither 

 engine was capable of following the piston beyond half stroke. 

 When the load was light, they each " cut off " the steam at a very 

 small fraction of the stroke, and as the load was augmented, autom- 

 atically increased this fraction until it reached about one-half; but 

 if loaded l)eyond their ability, neither could follow further, but 

 must stop. Both engines were thus brought to rest by the brake at 

 several periods during the tests ; but it is hardly necessary to say, 

 that no diagrams, taken when the engine Avas unable to continue its 

 motion, were made the subject of calculation. The*^' ratio of expan- 

 sion " given in one column of the above table relates not to the 

 fraction of the stroke during which the steam valve w-as actually 

 open in these engines in tlie several instances, for that was not easy 

 to ascertain definitely, either hy observation at the time or by 

 examining the diagram produced; but relates solely to the expansion 

 which would, in a theoretically iwrfect engine., of the same dimen- 

 sions, have consumed the same quantity of steam, or, to be more 

 critically exact, which would have reduced the steam to the same 

 terminal pressure. As will be observed, it ranges in the table from 

 3.6G (implying a cutting off of the steam at a little later than one-fourth 

 of the stroke from the ct>m men cement) to as high as S.45 implying, 

 that the cut-off 0})erated at a little more than four inches from the 

 commencement. In jjoint of fjict, both engines closed tlieir ports 

 earlier with the lightest, and held them open considerably later with 

 the heavy loads than these figures indicate; but the clearance or 

 equivalent space at the end of the cylinder in the one class, and the 

 insufficiency of tlie supply of steam, while the valve was held open. 



