SIR ANDREW NOBLE: RESEARCHES ON EXPLOSIVES. 
459 
In the tables are shown the units of heat both for the water fluid and the water 
gaseous, and in Plate 14 are drawn curves of the units of heat, water gaseous, for the 
whole of the six explosives. 
The Italian ballistite develops more heat than any of the other explosives, being a 
little higher than Mark I cordite. This is, of course, objectionable as regards erosion, 
but recently, when in Italy, I was informed that this heat had been greatly reduced 
by the incorporation of 12|- per cent, of carbon. 
I had already made some experiments with percentages of carbon, introduced as 
suggested by Lord Rayleigh. 
The curves in this instance are all convex to the axis of abscissae, and it is 
interesting to note that where the volume of gas per gramme is large the units of 
heat are low, and that where the volumes of gas are rapidly decreasing the curves 
representing the amount of heat developed show a rapid increase. 
Thus, with the new explosives, the point to which I so often drew attention in the 
case of the great variety of the old gunpowders with which I experimented is 
confirmed, namely that, with an explosive, if ever a large volume of gas is formed the 
heat developed will be low, if a small volume be generated the heat developed will 
be high. 
Coming now to the relation existing between the density of the charge and the 
pressure developed by the explosion, I have had some difficulty in regard to the 
abnormal pressures obtained with some explosives at high densities, but it must he 
remembered that the pressures to which I refer are very far above those with which 
artillerists are concerned, and do not materially affect the conditions of service of the 
propellants used in this or other countries. 
If reference be made to the tables, or preferably Plate 19, it will be observed that, 
up to densities of 0 - 45, the Explosives I., II., and VI. are perfectly regular; there are 
no abnormal pressures in any of the three, and the curves fairly represent the 
observations. In the case of cordite, which has a considerably higher potential energy 
than the other two, I have, without any abnormal pressure, gone to densities of 0'5 
and 0‘55, representing pressures of 53 and 60 tons per sq. inch (8078 and 9145 
atmospheres), but when M.D. and nitrocellulose were fired at the density of 0\5, the 
pressure of M.D. rose to 57 tons per sq. inch (8688 atmospheres) instead of the normal 
pressure, which should be about 49 tons per sq. inch (7469 atmospheres), and the 
pressure of the nitrocellulose rose to 56 tons per sq. inch (8535 atmospheres) 
instead of to what should have been the normal, about 47 tons per sq. inch 
(7164 atmospheres). 
Referring again to the same plate, on which are also drawn curves representing the 
relation of pressure to density of Explosives III, IV., and V., it will be seen that, 
while the Italian ballistite follows apj)roximately, up to a density of 0’45, the same 
law as rules in Explosives I., II., and VI., the two Norwegian ballistites show an 
abnormal, though small, increase in pressure when the density exceeds 0’3, 
3 N 2 
