18 
greater number of oxygen atoms to combine at the surface of 
the flame, that is to say, gives greater energy of combustion, 
in proportion to the pressure; and hence will result a more 
rapid distillation, and a greater consumption of material. 
The fact that the size of the flame is reduced by compres- 
sion, does not alter this result at all, though it may modify it 
somewhat. ‘The mere fact of condensation of volume would 
tend to increase the temperature of the flame, apart from en- 
ergy of combustion. Moreover, the surface does not diminish 
under such condensation in proportion to the volume. If 
four flames have diameters as 1, 2, 8, 4, their surfaces will 
vary as the squares of those numbers, and their volumes as 
the cubes. The comparative volume of the last flame would 
be 4°,—64, and its surface 4,16; half that volume would 
be 82, but the surface of this latter flame would be much 
more than 8. Applying these principles to the experiments 
described, we may construct another table, as follows : | 
Surfaces; ob- : 
Absolute ihe lune er tained by taking mage ae Te 
Pressures. y i ‘ u the } power of “ 1 Gays mcs 
w. Lite! volumes. 
15 Ibs. i oe ike ts 1 
23 43 = 0°652 0-752 1153: 1 
25 #3 = 06 0-711 1185: 1 
46 15 = 0°326 0-474 1-453: 1 
Of course it is assumed in this table, that the temperature 
of the air does not vary, and that no other accidental circum- 
stances affect the result. 
Mr. Collingwood regretted that his observations had been 
necessarily so limited, on so interesting a topic, but desired 
to place them on record. 
Pror. SEELY remarked upon the paper as follows :— 
The effect of varying pressure on combustion, was first 
investigated by Sir Humphrey Davy; and his conclusion, 
that increased pressure favors combustion, has been accepted 
without dissent, until the present time. Frankland, Deville, 
and others have reinvestigated the subject, but it is doubtful 
whether anything has been developed by them, which was 
