272 
PROFESSOR OSBORNE REYNOLDS AND MR. J. H. SMITH ON A 
diameter, which were fitted into cases specially cast on the under side of the bed¬ 
plate for them, their other extremities being sunk into the floor of the laboratory. 
The machine was mounted on the bed-plate in such a manner as to bring the vertical 
line of the oscillating piece into the centre line of the bed-plate, and about 3 cwts. of 
cast-iron was then bolted to the inside of the bed-plate, and its position adjusted so 
as to bring the surface horizontal. 
The machine was driven from the countershaft by means of a f-inch rope and large 
stepped grooved pulley, 30 inches diameter, a movable pulley being used to adjust 
the tension of the driving cord. 
At the left side of the bed-plate there was attached a speed indicator designed and 
constructed by Mr. T. Foster, Mechanical Expert in the Whitworth Engineering 
Laboratory. The form of the indicator is simple, the speed being indicated by the 
rise of water in a tall glass tube due to the “ centrifugal force ” produced by setting 
it in rotation by means of a spindle driven by the machine. The author is indebted 
to Mr. Foster, not only for this, but for many valuable suggestions and many 
excellent pieces of his workmanship during the construction of the apparatus. 
A central gunmetal spindle, driven from the crank shaft by a small gut band, has 
attached to it four radial vanes which revolve in a cylindrical brass box filled with 
water. A glass tube rising vertically is connected to the lower part of the case, and 
has a scale attached which is graduated by means of a revolution-counter to measure 
revolutions per minute. The case containing the water was arranged so that the 
amount of water used could be accurately regulated. Coloured water was first used, 
but it was found better to use pure water as the colouring matter was deposited on 
the tube, and after a few weeks made the taking of readings difficult. 
In addition to this, at the back end of the machine a cast-iron bracket is bolted to 
the standard. This has its upper surface planed, and on it a small table having a 
revolution-counter (T, fig. 2), attached to it. The table slides on the surface, being 
guided so as to allow the counter to jmss in and out of gear with the end of the 
crank shaft. A steady pin was used to hold the counter in its different positions. 
A lead buffer is used to receive the blow from the vertical oscillatory weight when 
the specimen breaks. Two cast-iron pieces F and G (fig. 5, p. 279), keep the buffer 
central and are so arranged that when F is lifted G can be removed. A conical piece 
of lead is inserted in the centre of the piece F, and is directly under the vertical 
spindle. The pieces of lead can easily be replaced, and it was found necessary to 
replace them after every three or four tests. 
Method of Lubrication. 
A great amount of difficult}^ was experienced in suppl} T ing the oil to the various 
bearings. A very thick oil was used for the crank pin, and an ordinary machine oil 
for the other parts. In the final arrangement of the apparatus, the oil was supplied 
