458 
SIR ANDREW NOBLE: RESEARCHES ON EXPLOSIVES. 
the Norwegian ballistite, the volume of hydrogen falls from over 20 per cent, to about 
9 per cent., in the Italian the H rises from about 8 per cent, to about 10 per cent., 
falling slightly at higher densities. 
In both explosives the N is practically constant at about 12 per cent, and 16 per 
cent, respectively, but there is a very great difference as regards the H 2 0. In the 
Norwegian the H 2 0 is constant at 14 per cent., there being no greater difference 
than inight be expected from errors of observation, while in the Italian the H 2 0, 
which commences at density 0'05 with a volume of 29 per cent., falls at a density of 
0H5 to about 24'5 per cent. No other explosive approaches the Italian ballistite in 
respect to the large volume of aqueous vapour formed, especially at low densities. 
As regards the other explosives, the differences, although not so pronounced as 
those I have just quoted, are still remarkable. Thus, comparing Cordite Mark I and 
Norwegian 167 (see Plate 11), in the cordite the volume of C0 2 commences at 21 per 
cent., rising to 31|- jDer cent, at d = (P45, while CO, commencing at 26V per cent., 
falls to 16^ per cent., the two gases being of equal volume at a density of OH9. In 
the case of Norwegian 167, the C0 2 commences at 15^ per cent., rising to 31 per 
cent., while the CO, commencing at 35^ per cent., rapidly falls to 21 per cent., the 
two gases being equal in volume at a density of about 0 - 34. 
H 2 0 and N may, in both explosives, be regarded as nearly constant, H 2 0 being 
with the first at a volume of about 21 per cent., with the latter at a volume of about 
15^ per cent., N at volumes of approximately 15*3 per cent, and 13 per cent. 
CH 4 rises from a trace in I. and IV. explosives to 5'5 per cent, in the case of 
cordite and to 9 per cent, in the case of the Norwegian, while the H falls from nearly 
16 per cent, to 10‘5 per cent, in I. and from 19 per cent, to 9'5 per cent, in the case 
of IV. 
With respect, however, to Explosives II. and VI. (M.D. and nitrocellulose) there 
is, in the transformation, a remarkable similarity. In both explosives (see Plate 12) 
the C0 2 rises from about 15 per cent, to about 29^ per cent., while the CO falls from 
about 35 per cent, to between 21 and 22 per cent., equal volumes being at the 
densities of about 0‘32 and 0’36 respectively. 
In both the marsh gas rises from a trace to about 9 per cent., while the hydrogen 
falls from about 20 per cent, to about 11 per cent. In both the H 2 0 and N are 
nearly constant, the H 2 0 being about 17 per cent, in each, while the N is about 
12i per cent, with M.D., and about 11|- with nitrocellulose. 
In the tables are given the cubic centimetres per gramme of the permanent and 
total gases, and in Plate 13 are drawn curves representing for the six explosives the 
observations of these total volumes. It will be noted that in the case of the 
nitrocellulose and Norwegian ballistite 165 there is, with increasing density, a very 
considerable decrease in volume, but with the Italian ballistite throughout the range 
of the experiments there is hardly any change. The curves in this plate, it will be 
observed, are concave to the axis of abscissae. 
