We have seen in Julien’s translation what the encyclopædist actually says, but what meaning did he intend to convey by his words? Did he mean to say the shell exploded? The passage may be divided into two clauses: in the first he explains generally the action of the ho-pao, and in the second he gives a particular example of its use. In the first clause he says that “no sooner was a light applied to it than the fire burst forth on all sides” (_le feu éclatait de toute part_): in the second clause he says, “the pao burst forth” (_les pao éclataient_). But the effect produced by the shell shows that this latter phrase is simply an elliptical way of saying, “the fire of the mixture contained in the pao burst forth.” On this point Reinaud and Favé are clear: “_Les pao à feu éclataient_ s’applique aux éclats de la flamme qui sortait par les ouvertures,”[293]—holes in the shell which were probably numerous. Mayers agrees: the pao were lowered into the excavations, “when the fire burst out from them, utterly destroying every fragment of the hides,” &c.[294] The Chinese writer was describing an incendiary, not an explosive. Gunpowder would have left in the hiding-place of the Mongols a tangled mass of charred human remains and scorched cowhide: only an incendiary could have destroyed its contents so that “not a vestige remained.” Father Gaubil and M. Berthelot acquiesce in this conclusion:[295] Herr von Eomocki dissents from it.[296] There is nothing in the military history of China in the thirteenth and fourteenth centuries to lead us to suppose that the Chinese possessed an explosive during that period. In 1255 Prince Hulágu had 1000 Chinese arbalisters in his pay to work his incendiaries,[297] and it may be presumed that he would have learnt the secret of gunpowder from them if they had known it; but he possessed no explosive. Father Carpini, _cir._ 1250, states that when hard-pressed the Tartars had recourse to incendiaries, and Rashid ed-Din, in his history of Hulagu’s campaign of 1260, makes no allusion to explosives.[298] The Chinese had only reached the same stage as Marcus Græcus in 1257: in this year they had Roman candles.[299] During the siege of Siang-yang-fu, 1268-73, “Khubelai sent to his nephew Abaka, in Persia, for engineers skilled in making catapults, called mangonals[300] by Marco Polo. Two such engineers were sent.”[301] We have three different notices of this siege, Chinese, Persian, and Venetian, and “they all concur as to the employment of foreign engineers from the West,”[302] but none of them mentions the use of explosives by either side. “The Chinese at that period,” says Sir John Davis, “were as little acquainted with firearms as Europeans.”[303] When Chang-chi-ki’s fleet on the Kiang River was destroyed a few years afterwards by Atchu, it was by means of fire-arrows.[304] In a word, during the thirteenth century, the Chinese made a free use of various incendiaries already noticed in the chapters