January 25, 1ST3-] 
THE PHARMACEUTICAL; JOURNAL AND TRANSACTIONS. 
5S5 
cal examination has failed to afford any reason whatever 
for doubting the stability of gun-cotton under all con¬ 
ditions of storage which are likely to occur. The ex¬ 
perience thu3 gained applies even more favourably to 
gun-cotton reduced to pulp according to the system 
lately in use, whereby the uniform purification of the 
gun-cotton is more effectually secured. Compressed 
gun-cotton has not only been stored extensively in dif¬ 
ferent piwrts of Great Britain; it has also been exported 
in considerable quantities to Australia, India, the West 
Indies, South America, and other distant countries, and 
has been used under circumstances specially trying to 
any material of uncertain stability. 
The explosions which occurred at Stowmarket nine 
months ago had the natural effect of dispelling from the 
public mind the great confidence which was becoming 
very generally entertained in the stability of gun-cotton. 
Fortunately the facts which were elicited in the course 
of the inquiry constituted so complete a chain of evidence 
as to place the first cause of the explosion beyond any 
reasonable doubt, and to demonstrate that it was quite 
independent of any want of stability of the properly- 
manufactured material. A supply of gun-cotton delivered 
from the works at Stowmarket, forming part of a quantity 
of which there remained a store in the magazines that 
exploded, was found to contain a proportion of disks in 
a highly impure condition. The proportion of free 
(sulphuric) acid existing in some of these was so con¬ 
siderable that it could not possibly have been left in the 
gun-cotton after the first rough washing which it receives 
immediately on removal from the acid, and before conver¬ 
sion into pulp in the rag-engines, where it is beaten up 
for several hours with a very large volume of water. 
Supposing, therefore, that the gun-cotton pulp compos¬ 
ing these disks had been submitted to the compressing 
process without passing through the intermediate and 
principal purifying operation, it could not possibly have 
contained even a small proportion of the sulphuric acid 
discovered in the impure disks, and the same would have 
been the case even if the M/i-pulped gun-cotton, after the 
preliminary washing and ringing, could have been con¬ 
verted into compressed disks. It was indisputably esta¬ 
blished, therefore, that the sulphuric acid discovered in 
the impure gun-cotton, and which could not have been 
generated by any decomposition of the substance, must 
have found its way into the finished material in some 
manner totally unconnected with the process of manu¬ 
facture, and that no amount of carelessness in manufac¬ 
ture, even to the extent of partial omission of the purify¬ 
ing processes, could have led to the existence of the acid 
found in the impure gun-cotton. That this impurity was 
sufficient to establish rapid change was sufficiently 
proved by the condition of some of the disks; and that 
this chemical change, accelerated as it was by the great 
heat of the weather at the time, gave rise to a develop¬ 
ment and accumulation of heat inevitably culminating 
in the ignition of some portion of the stored gun-cotton, 
was readily demonstrated by simple experiments with 
some of the impure disks themselves. But although the 
ignition of the store of gun-cotton in the lightly-built 
magazines at Stowmarket was completely accounted for, 
the very violent character of the explosions, and especi¬ 
ally that of the second explosion of a small store, which 
was burning for a considei'able time before its contents 
detonated, were results quite unexpected to those well 
acquainted with the properties of gun-cotton in the com¬ 
pressed form. Many practical experiments had demon¬ 
strated that it might be submitted to extremely rough 
treatment without any risk of explosion, and single pack¬ 
ages of the closely-confined material had been repeatedly 
ignited, from within and without, no other result than 
an inflammation and a rapid burning of the gun-cotton 
having ever occurred. These demonstrations of the ap¬ 
parent immunity from explosive properties of compressed 
gun-cotton, unless very strongly confined or fired by 
detonation, appeared fully confirmed by the results of a 
somewhat extensive experiment made at Woolwich a 
year ago with gun-cotton packed in firmly-closed wooden 
boxes, of the kind which Government proposed to use 
for storing the material. Eight such packages, each 
containing 28 lb. of gun-cotton, were enclosed in a pile of 
similar boxes loaded to the same weight, and the con¬ 
tents of the centre box were ignited: no explosion re¬ 
sulted, and the contents of some of the boxes even 
escaped ignition A second experiment, in which the 
centre box was surrounded by inflammable matter, so 
that a fierce fire burned within the heap for many 
minutes before the gun-cotton ignited, was also un¬ 
attended by any approach to an explosion. The ap¬ 
parently conclusive nature of these experiments un¬ 
doubtedly encouraged a false confidence in the non¬ 
liability to explosion of stores of gun-cotton in the event 
of accidental ignition, and the Stowmarket catastrophe 
demonstrated the imperative necessity for a more exten¬ 
sive investigation of the subject. The results of some 
experiments recently instituted near Hastings by the 
Government Committee on gun-cotton have served to 
throw great light upon the manner in which the explo¬ 
sions at Stowmarket were brought about. In the first 
instance, twenty-four boxes (containing 6 cwt. of gun¬ 
cotton) of the kind used in the Woolwich experiment 
were stored upon tables in a small wooden shed of light 
structure, and a heap of shavings and light wood wa3 
kindled immediately beneath the boxes, two of which 
were left partly opened. After the fire had been burning 
for about seven minutes the gun-cotton inflamed and 
continued to burn with very rapidly-increasing violence 
for nine seconds, when a sharp explosion occurred; a 
very similar result was furnished by a second experi¬ 
ment, in which the same number of boxes of gun-cotton 
was stored in a small magazine of stout brickwork. By 
subsequent comparative experiments it was judged that 
a considerable proportion of the gun-cotton had been 
burned in both instances before the explosion occurred, 
but these were nevertheless of such violence as to pro¬ 
duce large craters in the shingle on the site of the 
buildings and to project the debris with much force to 
considerable distances. Two repetitions were afterwards 
made of the first experiment, in wooden sheds ot similar 
structure, and with corresponding quantities of gun¬ 
cotton similarly arranged in boxes of the same size, and 
fastened down just as securely as those in the former ex¬ 
periment, but the boxes were made of somewhat thinner 
wood and were constructed less strongly. In neither 
of these experiments did an explosion occur. In the one 
instance the fire was burning in the building for more 
than half an hour before the gun-cotton became ignited, 
and three minutes after the first great blaze had subsided 
there was a second blaze of gun-cotton. Although the 
latter must have been exposed to intense heat, no explo¬ 
sion was produced. In the second experiment the gun¬ 
cotton burned in three successive portions, the last having 
been exposed for many minutes to very fierce heat, yet 
burning non-explosively. The first two of these experi¬ 
ments demonstrated that if, in a store containing pack¬ 
ages of gun-cotton in somewhat considerable number, 
the material became accidentally inflamed, the intense 
heat developed by the burning gun-cotton in the first 
instance might raise some portion, still confined in boxes, 
to the inflaming point, and that then, the mass of the 
confined gun-cotton being in a heated condition, the igni¬ 
tion would proceed with such rapidity as to develope the 
pressure essential to explosion while the gun-cotton was 
still confined, the resulting explosion being instantane¬ 
ously transmitted to other boxes. When the magazines 
at Stowmarket exploded, a large volume of flame was 
observed to precede the explosion by a very distinct 
interval. The two other experiments described appear 
to demonstrate that with such quantities of gun-cotton as 
were stored in the experimental sheds, the fact of _ the 
material being confined in boxes of comparatively light 
structure constitutes a safeguard against explosion, the 
