TRANSACTIONS OF SECTION B. 437 
Various other methods, both electrical and calorimetrical, have been sug- 
gested, but so far as J am aware no authoritative vindication of their reliability 
has as yet appeared. The most recent suggestion—viz. that of a calorimeter to 
absorb all the radiation from a fire—fails for one or other of two reasons. In 
the one case if it be larger than the fire then the water capacity becomes very 
great, and difficulties occur in ensuring perfect circulation; also heat is lost from 
the outer surface by convection and radiation. Alternatively, if the instrument 
be so small as to necessitate its being placed very close to the fire the free access 
of air to the fire is interfered with, thus upsetting the natural action of the air 
currents and setting up abnormal conditions. 
The testing of gas fires is a much more difficult matter than might be 
imagined. Considerable experience in heat measurement is called for, and until 
recently this has not been available in commercial testing laboratories, and even 
now the number of trained workers who have turned their attention to this 
branch of research in connection with gas fires is very small. In the absence of 
such special experience results obtained even by otherwise careful and competent 
workers are of doubtful value, owing to unsuspected errors in judgment, or to 
inherent defects in the methods employed. 
Ten years ago even the best deep fires did not afford more than 30 to 33 per 
cent. of the net heat of combustion of the gas in the form of radiant energy. 
In the effort to secure increased radiation (by which I mean a higher percentage 
of the heat developed by the combustion of a given amount of gas, delivered 
as radiant energy) it has been found, as was to be anticipated, that to advance 
from these low figures to 45 per cent. is much more easy than to make a 
further increase above 45 per cent. The adoption of the shallow-fire principle, 
and the dispensing with the front bars, to which I have already referred, was 
responsible for an increase from 30 per cent. to somewhere about 42 per cent. 
From this point, by attention to the perfecting of the design and proportion 
cof the burners and backbricks, and—most important of all—that of the 
radiants, we have been successful in further raising this figure to 48 to 50 per 
cent., and indeed in some instances to as high as 55 per cent. 
Irom the first gas fires had been made considerably smaller in width than 
the average coal fire, and when the new shallow-fronted radiation gas fire had 
been evolved there still remained a general tendency to keep these fires down to 
a similar narrow width, the impression probably being that inasmuch as gas 
fires give a more concentrated heat, a smaller firespace was adequate. I became 
convinced, however, that if gas fires were to take the place of coal fires for 
heating the largest domestic apartments equally well as the smallest, it would be 
necessary that gas fires should be made available having a firespace as wide as 
that of the coal fire. Although this development was simple in appearance, it 
involved constructional problems which were only solved after considerable 
experiment ; the result has been that gas fires of the new type are now made 
as wide as 21 inches, fires of that size being capable of heating rooms up to a 
cubical content of at least 4,000 cubic feet. 
The problem of total heating efficiency is, however, not the only one which 
makers of gas fires have to solve; the equally important question of ventilating 
effect must also be considered, for a properly constructed gas fire should effec- 
tively ventilate as well as heat an apartment. It is obvious that no fire could 
be considered as hygienically perfect which, when connected with a chimney 
flue in the ordinary way, and subjected to a moderate chimney draught, allows 
any products of combustion to escape into the room; bat provided that this 
elementary hygienic requirement is fulfilled, the question of ‘ hygienic efficiency ’ 
resolves itself into the amount of excess air, over and above that required for 
combustion, which can be drawn up the flue, per cubic foot of gas burned, when 
the fire is so connected. There is obviously no object served in testing or dis- 
cussing ‘hygienic efficiency’ or ‘ventilating effect’ except in relation to con- 
ditions of ordinary chimney draught, because no gas fire ought ever to be used 
except it be connected with a chimney or flue leading into the outside atmo- 
sphere. 
; It is not difficult to design and proportion the flue vent and the canopy of a 
gas fire so as to ensure the drawing up the flue of a large volume of air, thus 
producing good ventilation. The real difficulty is to avoid drawing this large 
excess of air over the upper portion of the radiants, thus cooling them and 
