512 TRANSACTIONS OF SECTION G. 
The main experiments were to determine the quantity of water pumped per 
minute, and the pressure at the pump delivery side and at the nozzle. The 
quantity of water pumped was measured either by passing it through a Venturi 
meter on its way to the nozzle or by discharging the water from the nozzle into 
a big swimming-bath which had been carefully calibrated beforehand. The 
Venturi meter was tested for the accuracy of its records both before and after 
the experiments were carried out. 
In order to determine frictional and other resistances to the flow of water 
through the canvas fire-hose, different lengths of hose were tested, all the other 
conditions being kept practically constant, and it was thus possible to eliminate 
the resistances at the point of entry into the hose, and in the meter when the 
water was delivered through a meter, and at the entry to the nozzle. Experi- 
ments were made with ordinary canvas hose and with rubber-lined canvas hose. 
The hose experimented with was 2% inches in internal diameter, and various 
diameters of nozzles were employed in the tests. The lengths of hose tested were 
100 feet, 500 feet, and 1,000 feet, and some experiments were made with double 
lines of hose, each delivering to a nozzle of the same diameter. 
The results obtained as to quantities of water pumped, pressures at pump and 
at nozzle, and speed of engine are given in the form of tables, and a final table 
gives the coefficients of frictional and other resistances to the flow of water 
through canvas fire-hose. 
5. Australian Timbers. By Professor W. 1. Warren. 
In Sub-Section G II. the following Report and Papers were read :— 
1. Report on Gaseous Haplosions.—See Reports, p. 177. 
) 
2. Temperature Cycles in Heat-Engines. 
By Professor E. G. Coxer and W. A. ScoBux. 
Experimental investigations of the cyclical variations of heat-engines and 
heat-pumps have received much attention, and numerous methods and instru- 
ments have been devised to give records of their cyclical changes, such as those 
of pressure and volume of the working fluid, changes of angular velocity of 
the crank-shaft, and the like. Temperature changes in the working fluid may 
usually be inferred very accurately from the pressures recorded on an indicator 
diagram, since there is usually a definite relation between pressure and tem- 
perature of a vapour, as, for example, in heat-engines using steam direct from 
a boiler without the intervention of a superheater. In other heat-engines, such 
as those using superheated fluids, and also those of the internal-combustion 
type, the temperature is more difficult to determine, and it becomes important 
to measure it directly. Platinum resistance thermometers and thermo-electric 
couples have been frequently employed for measuring cyclical changes of 
temperature in heat-engines, and a complete record from point to point of a 
cycle may be obtained if the engine is working with absolute uniformity. As 
it is usually impossible to prevent some amount of variation in the working 
of the engine while the measurements are in progress, the resulting curve is a 
composite one, since each measurement corresponds to a different cycle. 
The possibility of obtaining an instantaneous automatic record with an 
Einthoven type of galvanometer was considered in our early experiments on the 
cyclical variations of temperature of the working fluid of a gas-engine, and in 
the walls of the cylinder, but the difficulties then appeared to be so great that 
a potentiometer balance method was used instead. Recently, by the kindness 
of the Cambridge Scientific Instrument Company, we have been able to make 
experiments with their latest form of short-period Einthoven galvanometer, 
and this has enabled us to obtain instantaneous records of the temperature-cycles 
of the working fluid of steam- and gas-engines, and also the variations of 
temperature in the walls. Some of these photographic records are shown, and 
