52 
CAPTAIN NOBLE AND ME. E. A. ABEL ON EIEED GUNPOWDER. 
powder was nearly exactly equal to that of the same volume of water, — that is to say, 
the gravimetric density was about equal to unity. 
The curve drawn on Plate 13 exhibits the results of the first and most reliable series 
of Count Bumford’s observations. It shows the relation he believed to exist between 
the density of the gas and its pressure, and is expressed by the empirical formula 
^=l , 841# 1+ ' 0004,,! , 'p being the tension and x the density of the gas. 
The charges with which Bumford experimented were very small ; the largest, with 
one exception (by which his vessel was destroyed), was 18 grains (1T7 grm.). The total 
quantity of powder required to fill the vessel was about 28 grains (1 - 81 grm.). It will 
be observed that, if the curve (Plate 13) were supposed to be true up to the point when the 
chamber is completely filled, the pressure exhibited would be about 29,000 atmospheres. 
But, high as this result is, Bumford considered it much below the truth. In addition to 
the series the results of which are graphically represented, a second series was made, 
the results of which were very discordant. 
From Plate 13 it will be observed that, with a charge of 12 grains (0'78 grm.) 
(equivalent to a mean density in the products of combustion of 0’428), the tension of 
the gas was in the first experiment about 2700 atmospheres; but in this second series 
the tension with the same charge was repeatedly found higher than 9000 atmospheres. 
The discrepancies between the two series of experiments are not explained ; but, relying 
upon the second series, and on the experiment by which the cylinder was destroyed, 
Bumford calculated that the tension of exploded gunpowder, such as that employed by 
him, when filling completely the space in which it is confined, is 101,021 atmospheres 
(662 tons on the square inch)*. He accounts for this enormous pressure by ascribing 
it to the elasticity of the steam contained in the gunpowder, the tension of which he 
estimates as being doubled by every addition of temperature equal to 30° F. He further 
considers the combustion of powder in artillery and small arms to be comparatively slow, 
and that hence the initial tension he assumes is, in their case, not realized. 
In 1823 Gay-Lussac appears to have communicated to the “Comite des Poudres et 
Salpetres” a report of his experiments upon the decomposition of gunpowder f. Gay- 
Lussac’s products were obtained by allowing small quantities of gunpowder to fall into 
a tube arranged to receive the gases, and heated to redness. The collected permanent 
gases, when analyzed, gave in 100 volumes 52‘6 volumes of carbonic anhydride, 5 of 
carbonic oxide, and 42 - 4 of nitrogen. Gay-Lussac gave the volume of these gases, at 
a temperature of 0° C. and 760 millims. barometric pressure, as occupying 450 times 
the space filled by the powder, the gravimetric density of which was *9. Piobert, 
however, points out that Gay-Lussac’s results, thus stated, are not possible, and suggests 
that, by an error, the quantity of gas actually found has been doubled. 
Piobert’s suggestion is, from various corroborative circumstances, exceedingly pro- 
* Rumford, loc. cit. p. 280. 
t We have been unable to obtain the original of this report ; see, however, Piobert, loc. cit. p. 293. 
