This library is built in the open.
If you spot an error, have a suggestion, or just want to say hello — we’d love to hear from you.

XVII
I could not determine anything more precise about the Arabic works and their authors mentioned under 2. and 3.
The sources thus offer us a total of six titles of Hero's writings; of these, the oldest source accounts for five, the middle source for two, and the youngest for three, of which one—admittedly a doubtful one—is not named by the others. The fact that a book of war machines, mentioned last by Hadji Khalfa, is to be ascribed to a Hero is supported in particular by the circumstance that under the name of Hero a book belopoiika art of throwing missiles and another poliorcetika siegecraft are named; then also that the name Hârûn stands without any addition, which surely indicates that it is a foreign name, not the Arabic name Hârûn, which without any such distinguishing feature would give rise to confusion. Which of the two Heroes mentioned in Steinschneider (loc. cit. p. 347, No. 6) is meant here, I cannot decide, as the name and title alone offer no handle for this. The deviation of Hârûn from the otherwise common transcription (ʾÎron or ʾÎran, the second vowel is not written) would in itself be no obstacle to identifying the two names.
Whether all the mentioned writings were available in Arabic cannot be said based on the sparse information in the source works. Fragments of the book "Solution of Doubts in Euclid" may perhaps be provided by the sentences of Hero found in the Codex Leidensis 3991, published by Besthorn and Heiberg (Hauniae 1893, 1897; cf. ZDMG loc. cit. and Hero Vol. I, Suppl. p. 68, note 3). No manuscripts of the book itself are known, and the sources name no translators of it. Only one of the six mentioned writings is certainly translated, because it is still extant: namely, the one concerning the pulling or lifting of loads. Steinschneider (loc. cit. p. 347, No. 4) identified the "pneumatic machines" with the Mechanics, which the Arabic title already forbids, and which turns out to be entirely erroneous. For the work by Steinschneider...