540 APPENDIX used in various ways ; and the confusion of them under the common name of Greek or marine fire has led to some misapprehensions. The simplest fire weapon was probabl>- the "hand tube" {xf'poffi.^xoi'),^ a tube full of combustibles, which was flung by the hand like a "squib" and exploded on board the enemy's vessel. The marines who cast these weapons were the ' ' grenadiers " of the Middle Ages.^^ " Artificial fire " — probably in a liquid state — -was also kept in pots (xvTpai), which may have beon cast upon the hostile ships by engines. Such pots are re- presented in pictures of warships in an old Arabic Ms. preserved in the Biblio- tlieque Xationale, and reproduced by M. Schlumberger in his work on Nicephorus Phocas.'- But there was another, and more interesting, method of hurling " artificial fire ". This method anticipated the principle of later ilrearms : gun- powder was used to propel the missiles of destruction through a tube ((ri(pciju). This is the only reasonable inference from the two certain facts that gunpowder was one of the artificial explosives used by the Byzantines in their naval warfare (see above, p. 11, note 22), and that combustibles which exploded when they reached the enemy's ships were propelled through tubes, which were managed by a gunner (siphonator). Thus the Byzantines just fell short of revolutionising warfare, by failing to apply their propelling powder to leaden missiles. 6. THE PAULICIAN HERESY— (C. LIV.) In Gibbon's day the material for the origin, early history, and tenets of the Paulicians consisted of Bk. i. of the work of Photius on the Manichaean-s, and the History of the Manichaeans by Petros Sikeliotes. The work of Photius was edited by J. C. "Wolf in his Anecdota Graeca, i., ii. (1722) ; ^ but Gibbon did not consult it (above, chap. liv. note 1). There was further the account of the Bogo- mils in the Panoplia of Euthymius Zigabenus, a monk who lived under Alexin- Comnenus and is celebrated in the Alexiad of Anna. A Latin translation wa- published by P. F. Zinos in 155-5 ; the Greek text edited by a Greek monk (Metrophanes) in 1710. It may be read in Migne, P. G. vol. 130. The section on the Bogomils was edited separately by Gieseler in 1841-2. The documents which have come to light since are closely connected with th<^ accounts of Photius and Peter; they bring few new facts or fictions, but the bring material for criticizing the facts and fictions already known. (1) In 184: Gieseler published a tract- of a certain Abbot Peter, containing an account of the Paulicians similar to that of Photius and Peter Sikeliotes (with whom Gieseler identified the .author). (2) The publication of the chronicle of George Monachu^i by Muralt in 1859 showed that this chronicle had incorporated a similar account in his work. "We have then four documents, which presume one original account whereon all depend, directly or indirectly, if indeed one of them is not itself the original source. The problem of determining their relations to one another and the common original is complicated by (1) the nature of Photius, Bk. i., and (2) the variations in the !Mss. of George Jlonachus. The "First Book" of Photius falls into two parts : I. chaps. 1-15, which con- tains {a) a history of the Paulicians, chaps. 1-10; and (?>) an account of earlier Manichaean movements, chaps. 11-14 : II. chaps. 15-27, a history of the Paulicians. going over the same ground, but differently, and adding a brief notice of the revolt of Chrysocheir. Part I. (a) corresponds closeh- to the accounts of Abbot Peter, Peter Sik.,^ and George Mon. ; and its Photian authorship seems assured by the testimony of Euthymius Zigabenus. Part II. was a distinct composition originally, and was tacked on to the Photian work. Thus " Photius " resolves itself into two documents, one Photian, the other Pseudo-Photian. 10 Tactics, 19, § 57. 11 Some Arab grenades (first explained by de Saulcy) still exist. Cp. illustration in Schlumberger, Nicephore Phocas, p. 59. 1- P. 55, 57. 1 Reprinted in Mipne, P. G. vol. 102. - Title : Vl^Tpou eAaj^ttrrou fxoraxo^j *Hyou/xeVou nepi llav*.Kiaviov tC>v Ka "iiavixaiiav. 3 Peter Sifc. reverses the order of [n) and {h).
