M. Berthelot (i. 100 ff.) gives the best existing text of the foregoing tract, founded on Paris MSS. 7156 and 7158 collated with the Munich MS. 267. He adds extracts from the Munich MS. 197, dated 1438. Herr von Romocki gives the text reproduced here and the Nürnberg MS. of a somewhat later date than the Paris MSS., say 1350.[137]
A glance at the text given here shows that, far from being a formal and connected treatise, it is a medley of recipes thrown together with very little method and without any literary skill. Of the thirty-five recipes (in Du Theil’s MS.) fourteen are war mixtures, six relate to the extinction of incendiaries or the prevention and cure of burns, eleven are for lamps, lights, &c., and four describe the preparation of certain chemicals—one of them, No. 14, giving a mode of refining saltpetre. The war mixtures consist of nine recipes for various fires, Nos. 1, 2, 3, 6, 7, 8, 9, 24, and 26; one for fire-arrow composition, No. 10; and four for rockets and Roman candles (including a “cracker”), Nos. 12, 13, 32, and 33. Nos. 9, 15, and 24 contain quicklime; 12, 13, 14, 32, and 33 contain saltpetre.
A closer examination leads to the conclusion that the tract is the work of neither one author nor one period. As we read of such ingredients as weasel’s gall (17) and paste of glow-worms (16); of the mercury to be found in black and green lizards’ tails (19); of the mixture which ignites incontinently at sunrise, wherewith crows are to be anointed and despatched against the enemy (3), we seem to hear the chant of the witches in “Macbeth”:—
“Eye of newt and toe of frog,
Wool of bat and tongue of dog,
Adder’s fork and blind-worm’s sting,
Lizard’s leg and owlet’s wing.”
These recipes embody the same traditions as the war recipes of the “Kestoi” of Sextus Julius Africanus, which belong to the seventh century. But on turning to Nos. 32 and 33, we find recipes as precise and formal as those of Hassan er-Rammah, who wrote in the last quarter of the thirteenth century. The description of the rocket and its composition (13) is as definite and intelligible as many a recipe of the seventeenth century: recipes 8 and 17, with their allusions to Hermes, the mythical Alexander the Great, Aristotle the wizard and Ptolemy the magician, belong to a far earlier period. In short, the extraordinary contrast in style and matter, phraseology and diction, between certain recipes and others, leads irresistibly to the conclusion that the oldest recipes are separated from the youngest by several centuries, and that the tract (as we possess it) was not the work of one man, but of several. There is a kernel of old recipes, to which others were added from time to time. This conclusion receives strong support from the fact that no two of the MSS. are of the same length. The Munich MS. contains twenty-two, Berthelot’s text thirty-five, and the Nürnberg MS. twenty-five recipes.
The best judges date the oldest existing MSS., Paris, 7156 and 7158, at about 1300 A.D., and Abd Allah tells us that saltpetre was known to the Spanish Arabs in the second quarter of the thirteenth century.[138] The saltpetre recipes, therefore, 12, 13, 14, 32, 33, lie between the years 1225 and 1300. We shall call them, for convenience of reference, the “Late Recipes.”
No one who carefully studies the remaining recipes can fail to observe that many of them are marked by archaism of style, form and matter, and that they hand down to us ancient alchemical traditions, or traces of them; while others display no such peculiarities. Let us then, again for mere convenience, divide them into two series—the “Early Recipes,” which possess these peculiarities, and the “Middle Recipes,” which do not. To what periods do these two series belong?