41 G 



REPORT — 1894. 



The percentages of moisture have been calculated from the mean 

 saltness of the boiler water. It will be seen they vary considei'ably. 

 Further, the moistness from the salt test is only one-hundredth as great 

 as that shown by the wire-drawing calorimeter. 



The temperatures for the Barrus calorimeter are plotted in fig. 7, in 

 order to show how remarkably regular and consistent the observations 

 with a wire-drawing calorimeter are. 



SitnuJtaneous Wire-drmving and Salt Test. — In the following test the 

 Barrus wire-drawing calorimeter was used, the observations being taken 

 by Professor Unwin. Simultaneously Professor Kennedy took samples 

 of the condensed steam and boiler water, and determined the moisture in 

 the steam by the salt test. 



Test of Water Tube Boiler, rising Wire -drawing Calorimeter, 

 April 11, 1891. — The temperatures of the calorimeter thermometers 

 given are the means of two to five observations at intervals of about 

 one minute, taken after steam had been flowing through calorimeter for 

 about twenty minutes. They were taken as soon as the temperatures 

 had become steady. Barometer 29-936 in. = ll-70 lb. per sq. in. Tem- 

 perature of saturated steam corresponding to pressure in second chamber 

 assumed 214°. 



It is obvious that the salt test shows much less moisture than the 

 calorimeter, the mean moisture by the former being 1 per cent, and by 

 the latter only about one-tenth as much. Naturally the salt test which 

 shows only mechanical priming should give a lower result than the 

 calorimeter. But the salt tests are so irregular that it is impossible not 

 to doubt a little the accuracy of the method. 



