ON THE MECHANICAL EQUIVALENT OF HEAT. 
369 
If the capitals H and U refer to trials with a large turning moment on the brake, 
and the small letters h and u refer to trials with a small turning moment, then for 
our value of the mean specific heat of ■umter in mechanical units we have 
K = 
U — u 
TL-h * 
This quantity K is not strictly the same as the mechanical equivalent of heat, of 
which other determinations have been made, since we are here dealing with the mean 
specific heat of water betwmen freezing and boiling-points. 
For this reason it has been decided not to use the usual symbol J, at any rate at 
this stage of the research. 
19. As an illustration of the method of tabulating and reducing the observations, 
I append all that were taken in trials 69 and 72 made on the 7th and 8th July, 1896, 
respectively. 
It will be seen that all the observations of temperature, together with the readings 
of the speed indicator, which were made during the actual progress of each trial, are 
given on pages 373 and 375 respectively. 
With the exception of the two readings of the speed indicator taken at the 
moments of starting and finishing each trial, and shown in brackets at the top and 
bottom of column No. 8, I was personally responsible for all observations recorded. 
These two observations wmre made by the assistant in charge of the hand brake and 
artificial atmosphere. 
In the tables of temperature and speed observations 
Col. 1 gives the times at which observations became due, the whole period of 6,2 
minutes being divided into 31 two-minute inteivals. 
Col. 2 gives the temperatures of supply of the water to the brake. 
Col. 3 ,, ,, discharge of the water from the brake. 
Col. 4 „ ,, the air in the engine room. 
Col. 5 ,, ,, the stuffing-box cover. 
Col. 6 ,, ,, the lower brass of the main bearing. 
Col. 7 ,, fall of temperature between the stuffing-box and bearing, being 
the difference of Cols. 5 and 6. 
Col. 8 gives the readings of the speed indicator. 
Observations of the revolution counter and of the weight of the tank before and 
after each trial, are given on pages 372 and 374 respectively. 
As I had to take all the observations myself, it was, of course, impossible to make 
them simultaneously at the times indicated in Col. 1. They were, however, always 
taken in the same order, as follows. 
When the time for the next ensuing series of observations had arrived as given by 
a watch lying on the table at my side, I immediately read the temperatures of 
VOL. CXC.—A. 3 B 
