Concerning the foregoing formulas, it should be noted that all discussion of the proportions of the different ingredients has been purposely avoided, for the reason that the practice of the manufacturers varies in this respect almost from day to day..

The percentages mentioned with the various formulas represent average practice rather than actual proportions used. The manufacturers state that the variations in the quality in the different lots of chemicals obtainable by them require constant changes in their composition. Their results are a matter of trial only, and they vary their compositions until proper functioning is obtained, often groping in the dark for some time before succeeding. They state that this applies particularly to certain chemicals of American manufacture, the grade of which is more uneven than that of the chemicals of foreign origin obtainable before the war.

Regarding the relative safety during storage of pyrotechnic pieces containing these formulas, it pan be said in general terms that those including fulminating caps or matches may possibly be regarded as a little more dangerous than the others. Instances are known where these matches have detonated merely from dropping on the floor.

All mixtures represented by formulas that do not contain chlorates may probably be regarded as relatively safe, since decompositions are less likely to take place. The presence of moisture to the extent of 0.5 per cent. or more, in mixtures containing chlorates or nitrates, is regarded by some as contributing elements of danger during storage, owing to the decompositions that may take place after a lapse of time.

In the foregoing formulas we have quoted the opinions of manufacturers who have handled these various compositions: some for a period of many years, in the manufacture of fireworks of various sorts; others for only a short time, in the manufacture of military pyrotechnics. Their suggestions are based wholly on experience and not, on the possible or probable chemical reactions which might take place under adverse conditions of storage. It is our opinion that we must not neglect the study of these compositions from the standpoint of the intimate contact of the various substances which is produced by the pressure used in the formation of the different pieces. The interpenetration of the substances may be increased, but certainly will not be decreased, by the lapse of time. Take the two well-known cases where the intimacy of contact between two substances results in a mutual exchange, such as carbon into iron and iron into carbon during the old and well-known cementation process for the manufacture of steel; and the well-known experiment of Rose, establishing the mutual interpenetration of lead into gold and gold into lead, when plates of lead and gold are placed in intimate contact. A very striking case is the interaction between dry sodium carbonate and barium sulphate when compressed; as high as 80 per cent. of the barium sulphate being converted into barium carbonate under these conditions.

Taking these examples into consideration, and the further fact that a large majority of these compositions exist in a state of compression the possibility of chemical changes must be considered. The growing intimacy of the mixtures may have a tendency toward spontaneous combustion on the one hand, or toward a deterioration of the composition on the other.