the observed temperature of the exhaust gases was 1,229°. The fraction
then becomes (3443 - 1229)/3443 = 64 per cent. If we multiply this by
0.37, as we did in the case of the steam-engine, we get 23.7 per
cent., or approximately the same as that arrived at by direct
experience. Indeed, if the consumption is, as sometimes stated, less
than 18 feet, the two percentages would be exactly the same. I do not
put this forward as scientifically true; but the coincidence is at
least striking.

I have spoken of the illuminating power of the gas as of importance;
for the richer gases have also more calorific power, and an engine
would, of course, require a smaller quantity of them. The heat-giving
power does not, however, vary as the illuminating power, but at a much
slower rate; and, adopting the same contrivance as that on which the
absolute scale of temperature is formed, I would suggest a formula of
the following type: H = C (I + K), in which H represents the number of
heat-units given out by the combustion of 1 cubic foot of gas, I is
the illuminating power in candles, and C and K two constants to be
determined by experiment. If we take the value for motive power of the
different qualities of gas as given in Mr. Charles Hunt's interesting
paper in our Transactions for 1882, C might without any great error be
taken as 22 and K as 7.5. With Pintsch's oil gas, however, as compared
with coal gas, this formula does not hold; and C should be taken much
lower, and K much higher than the figures given above. That is to say,
the heating power increases in a slower progression. The data
available, however, are few; but I trust that Mr. Hartley will on
this, as he has done on so many other scientific subjects, come to our
aid.

I will now refer to the valuable experiments of Messrs. Brooks and