438 TRANSACTIONS OF SECTION B. 
unnecessarily diminishing the radiant efficiency of the fire. In other words, 
whereas it is comparatively easy to achieve ‘ventilation’ at the expense of 
‘radiant efficiency,’ a really .scientifically constructed fire should ensure an 
equally good ‘ ventilation’ without sacrifice of radiant efficiency, which, although 
not so easy, is by no means an impossible matter. 
In this connection my own researches have convinced me of the importance 
of preserving a certain adequate vertical distance between the top of the 
radiants and the bottom of the canopy of a gas fire, so as to avoid drawing the 
induced ‘ventilating’ air over the upper portion of the radiants; experiment 
has proved that such a constructional feature, combined with an adequate flue 
vent, ensures a much higher radiant efficiency than another type (in point of 
fact older, but which it has been recently sought to revive), in which the 
canopy is brought down to overlap (or nearly so) the top of the radiants. As 
the relative merits of these two types of construction have recently been under 
discussion, it may be of interest if I append the results of an investigation of 
the matter in my own laboratory, which seem to prove conclusively the marked 
superiority of the first-named type ef construction. 
To ascertain the total heating efficiency, each type of construction was tested 
by determining the radiant efficiency, using the Leeds University method, and 
at the same time determining the amount of heat lost through the chimney 
flue. The total of these two, deducted from the heat developed by the com- 
bustion of the gas, gives the amount of convected heat. It is obviously impossible 
to estimate directly the convected heat, since the radiant energy also eventually 
makes itself sensible in this form. ; 
The fires were tested under a series of parallel conditions. In the first case 
the flue outlets were blocked up; this neutralised any possible cooling action of 
the flue draught on the radiants. The second test was one with the flue 
outlets open. In the third the two stoves were connected to a chimney. It 
ought to be remarked that changes in the meteorological conditions sometimes 
influence the amount of air withdrawn from the room by a given chimney to a 
surprising degree, 
1t will be seen from the figures that the lowering of the canopy, as had been 
anticipated, resulted in a lowered radiant efficiency and a lowered total heating 
efficiency, due to the cooling effect exercised on the radiants by the air drawn 
over them. 
Fic. 1. Canopy well above Radiants. Fig. 2. Canopy brought down to top 
of Radiants. 
