944 Transactions of the American Institute. 



steam is used in a well-constructed steam engine, so as to utilize any 

 considerable portion of its power, less heat will be found in the 

 escaping steam, however carefully the radiation from tlie pipes and 

 cylinders may be guarded against. It is reported to have been found, 

 in some experiments instituted by the French government, that only 

 from one-lialf to three-quarters of the water disappearing from an 

 ordinary boiler, existed in the form of vapor at the termination of 

 the stroke of an engine working very expansively. Whatever may 

 have been the loss of heat from these causes, it is evident that the 

 system adopted in the tests under consideration measured only the 

 steam existing at the end of the stroke, and are thus liable to consid- 

 erable error in attempting to estimate tlie water disajipearing from 

 the boilers. 



For the sake of comparison of the engines with each otlier, in 

 the plainest form, however, and also to approximate to the true con- 

 dition of affairs, I have assumed a well-arranged stationary boiler to 

 be capable of evaporating just .eight pounds of water for each pound 

 of fuel, or rather of evaporating so much more than eight pounds, 

 that one pound of fuel would be required for each eight pounds of 

 stecmi actually discliarged from the engine. Whether any boiler is 

 capable of performing so well as this, or whether, on the other hand, 

 a majority of good stationary boilers do not -actually do better, is a 

 question which each reader must decide for himself. As usually 

 tested, i. e., by the disappearance of water poured into the boiler, 

 boilers are reported capable of evaporating every quantity, from four 

 to twelve pounds of water for each pound of fuel, and that the abso- 

 lute theoretical effect of a pound of pure carbon would be to perfectly 

 evaporate some two and a half pounds more than this greatest quan- 

 tity ; the water being in each case assumed to be pumped into the 

 boiler at a temperature of 212 ° Fahrenlieit. The evaporation (from 

 a temperature of 100 ® Fahrenheit) in the boilers of some half a 

 score American naval and marine steamers, a few years ago, averaged 

 Y,235 pounds per pound of antliracite ctml. 



Owing to the great importance attached to the indicator diagrams 

 in the tests here adopted, it may be well J^o explain that large indi- 

 cators were attached, one at eacli end of the respective cylinders. 

 The area of each piston was one inch, and the scale of motion twenty 

 pounds per inch. To adapt these large instruments, which were 

 designed for slow-working engines, to record perfectly the effect of 

 the steam in those of quicker action, the cocks were bored out of a 



