370 
PROFESSOR 0. REYNOLDS AND MR. W. H. MOORBY 
supply and discharge and the speed gauge in the order named, and after reading the 
three I entered them in the note-book. This generally took about a quarter of a 
minute. If then a reading of the atmospheric temperature was due, it was next 
taken and entered. After that the temperatures of the stuffing-box cap and of the 
bearing were noted in their turn, the whole series of observations being made in 1 or 
Ik minutes. 
The interval which then elapsed before the next series of observations became due 
was often fully occupied in making adjustments of the regulating valve controlling 
the main water supply to the brake ; of the cock regulating the supply to the 
stuffing-box ; and of the speed of the turbine driving tlie pump, small alterations at 
all these pcints being frequently necessary. 
At the head and foot of Cols. 3 and 8 will be seen observations in brackets. These 
observations were taken at the moments of starting and ending the trials, and were 
required in the calculation of a terminal correction to be referred to later. 
At the close of each trial a mean of the observations occurring in Cols. 2, 3, 4, 5 
and 7 was made, the two observations in brackets in Col., 3 beino- omitted in 
calculating these means. 
On pages 372 and 374 additive corrections to the weights and to the mean tempera¬ 
tures of supply and discharge are given. These will be referred to later. 
It will be noticed that in neither of the trials chosen was there any leakage of 
water from the stuffing-box. 
The observations are given again in the partially reduced form which has been 
adopted for the final tabulation of the results on p. 376. 
Cols. 1 to 8 should be self-explanatory. 
Col. 9 gives tlie first approximation to the heat generated, obtained by multi¬ 
plying the weight of water by its mean rise in temperature. 
Col. 11 gives the difference of the temperature of the stuffing-box (supposed to 
be a measure of that of the water leaking from it), and the temperature 
of supply. 
Col. 12 gives the loss of heat due to this leakage, and represents the product of 
Cols. fO and 11. 
Col. 13 gives the rise of temperature of the brake during the trial, and is 
assumed to be equal to the difference of the two temperatures given in 
brackets in the table of temperature observations (Col. 3). 
Col. 14 gives the terminal correction to the heat required on account of the 
increase of heat in the brake itself during the trial. 
Col. 15 gives the difference between the mean temperature of the stuffing-box 
and of the shaft bearing. As already explained the loss of heat by 
conduction has been assumed proportional to this difference, and a 
determination of its amount will be given later. At present it is 
