Complete Extant Works of Hero of Alexandria, Vol. 2

XXII

INTRODUCTION.

manuscripts, this is also confirmed as certain, which Baron C. de Vaux also acknowledges in a letter to me. In the note to p. 73 of the introduction to the separate edition, he did indeed try to defend the reading Praxidamas or to want to make Posidonius into a painter, but with little luck. For the Arabic relative clause after the name not only "seems" to reproduce the Greek apo tes stoas from the Stoa, but corresponds to it entirely, as the literal translation of it is: "who belongs to the associates of the Stoa." The other reading cited loc. cit., "qui était peintre" who was a painter, does not result at all so simply from the Arabic. If the word corresponding to "Stoa" becomes "painter" by the addition of other reading marks, which are often missing in the manuscripts, then the preceding words do not yet mean "who was," but the whole thing would mean: "who belongs to the associates of the painter." Then one would first have to know a Greek who was simply called the painter and was of such importance that one designated his students with the epithet "associates of the painter," perhaps hoi peri ton zographon those around the painter. De Vaux's latter conjecture is therefore in a very bad state, while the difficulties with the reading "Posidonius, a Stoic" vanish. Thinking of Archimedes instead of Posidonius or Praxidamas, as was recently suggested in the Berl. phil. Wochenschrift 1899 p. 1540 et seq., is forbidden by the Arabic characters of the name, from which even "Praxidamas" can only be read out with force. Let us therefore leave the Stoic Posidonius his place in the Mechanics of Hero.

III. THE AUTHENTICITY OF THE MECHANICS.

That the book before us, translated from Greek into Arabic by Kosta ben Luka and transmitted under Hero's name, is genuine is evident from the points mentioned below and in the appendix in the Greek text by the