ON" THE MECHANICAL EQUIVALENT OF HEAT. 
377 
The Barometer .—(Part I., par. 47.) 
20. Before dealing with the thermometers and their corrections, it becomes neces¬ 
sary to describe a combined barometer and manometer which was constructed to 
measure the pressures of steam employed in the determination of the boiling-points 
on the thermometer used to measure the discharge temperature. 
The structural details of this instrument are given in Professor Reynolds’ paper. 
At present it is sufficient to say that it consisted of a cast-iron, bottle-shaped reser¬ 
voir, through the neck of which the glass tube holding the mercury column was 
carried in a stuffing-box, which made a perfectly air-tight joint between the glass 
and the reservoir. The pressure to be measured was introduced through a small 
iron pipe, which penetrated horizontally the cast-iron wall of the reservoir, and then 
turned vertically upwards till its open mouth stood above the level of the mercury 
inside. Two circular plate-glass Avindov/s in the reservoir walls provided a means of 
ascertaining the level of the mercury surface. In order to measure the height of the 
mercury column supported by any external pressure, a brass sleeve was made, Avhich 
fitted outside the glass tube and the upper part of the reservoir. This sleeve con¬ 
sisted of a piece of |-inch diameter brass pipe fixed into a conical brass casting, 
which carried a truly-turned bevelled edge at its lower extremity. This conical 
casting engaged by an internal screw of twenty threads to 1 inch Avith the neck of 
the cast-iron reservoir. The upper part of the sleeve carried an internal thread of 
the same pitch, and into this Avas screwed a second piece of pipe through AAdiich two 
long narroAv slits Avere cut at opposite extremities of a diameter. A third piece of 
brass pipe engaged Avith the upper end of the piece just mentioned, and AA^as pi’ovided 
at its lower end Avith a truly-turned bevelled edge. 
In use the bevelled edge on the conical brass casting was first adjusted to the 
surface of the mercury in the reservoir, and then the upper bevelled edge Avas 
adjusted to the surface at the top of the mercury column. Suitable horizontal and 
vertical scales Avere provided to enable me to measure the vertical distance between 
these two bevelled edges to xirn'o of an inch. 
It was necessary to standardise this scale (Part I., par. 44). There is a Whit¬ 
worth measuring machine in the laboratory, which is provided amongst others Avith 
standard end gauges of 9 inches and 3 inches long respectively. 
Two new steel standards were made by Mr. Foster as nearly as possible of the 
same length as the 9-inch Whitworth, and by means of the measuring machine I 
determined their exact lengths as follows, three comparisoiis being made of the two 
new gauges with the standard. The table shows the readings obtained. 
3 c 
VOL. CXC.—A. 
