the ignition point of his di-nitro-cellulose, and found it to be 194 deg. to 198 deg.  C., and he is therefore of opinion that the explosion was due to the treatment of the partially washed material with ammonia.  A certain quantity of ammonium nitrate was probably formed, and subsequently dried upon the nitro-cellulose, in a state of very fine subdivision.  The faintest trace of acid would then be sufficient to bring about the explosive ignition of the ammonium nitrate.

The drying of gun-cotton or collodion-cotton is also a somewhat dangerous operation.  A temperature of 40 deg.  C. (104 deg.  F.) should not be exceeded, and thermometers should be placed in the nitro-cotton, and the temperature frequently observed.  An electric alarm thermometer is also a useful adjunct to the cotton drying house.  Great care must also be taken that there are no exposed hot-water pipes or stoves in the drying house, as the fine gun-cotton dust produced by the turning or moving of the material upon the shelves would settle upon such pipes or stoves, and becoming hot, would be very sensitive to the least friction.  The floor also should be covered with linoleum or indiarubber.  When hot currents of air are made to pass over the surface of gun-cotton, the gun-cotton becomes electrified.  It is important, therefore, to provide some means to carry it away.  Mr W.F.  Reid, F.I.C., was the first to use metal frames, carriers, and sieves, upon which is secured the cloth holding the gun-cotton, and to earth them.

The compression of gun-cotton into blocks, discs, &c., is also attended with considerable risk.  Mr O. Guttmann, in an interesting paper upon “The Dangers in the Manufacture of Explosives” (Jour.  Soc.  Chem.  Ind., No. 3, vol. xi., 1892), says:  “The compression of gun-cotton into cartridges requires far more care than that of gunpowder, as this is done in a warm state, and gun-cotton even when cold, is more sensitive than gunpowder.  When coming out of the centrifugal machines, the gun-cotton should always pass first through a sieve, in order to detect nails or matches which may by chance have got into it.  What has been said as to gunpowder presses applies still more to those for gun-cotton, although the latter are always hydraulic presses.  Generally the pistons fit the mould perfectly, that is to say, they make aspiration like the piston of a pump.  But there is no metal as yet known which for any length of time will stand the constant friction of compression, and after some time the mould will be wider in that part where the greatest compression takes place.  The best metal for this purpose has proved to be a special steel made by Krupp, but this also is only relatively better; for pistons I prefer hard cast iron.  If the position of the moulds and pistons is not exactly the same in all cases, what the Germans call ‘Ecken’ (English ‘binding’) will take place, viz., the mould will stand obliquely to the piston, and a dangerous friction will result.”  “Of