shall have been any more successful than Goropius Becanus, Schrieckius, Herr Rudbeck and the Abbe de la Charmoye. 144. Let us return to Zoroaster, who led us to Oromasdes and Arimanius, the sources of good and evil, and let us assume that he looked upon them as two eternal principles opposed to each other, although there is reason to doubt this assumption. It is thought that Marcion, disciple of Cerdon, was of this opinion before Manes. M. Bayle acknowledges that these men used lamentable arguments; but he thinks that they did not sufficiently [214] recognize their advantages or know how to apply their principal instrument, which was the difficulty over the origin of evil. He believes that an able man on their side would have thoroughly embarrassed the orthodox, and it seems as though he himself, failing any other, wished to undertake a task so unnecessary in the opinion of many people. 'All the hypotheses' (he says, _Dictionary_, v., 'Marcion', p. 2039) 'that Christians have established parry but poorly the blows aimed at them: they all triumph when they act on the offensive; but they lose their whole advantage when they have to sustain the attack.' He confesses that the 'Dualists' (as with Mr. Hyde he calls them), that is, the champions of two principles, would soon have been routed by _a priori_ reasons, taken from the nature of God; but he thinks that they triumph in their turn when one comes to the _a posteriori_ reasons, which are taken from the existence of evil. 145. He treats of the matter with abundant detail in his _Dictionary_, article 'Manichaeans', p. 2025, which we must examine a little, in order to throw greater light upon this subject: 'The surest and clearest ideas of order teach us', he says, 'that a Being who exists through himself, who is necessary, who is eternal, must be single, infinite, all powerful, and endowed with all kinds of perfections.' This argument deserves to have been developed more completely. 'Now it is necessary to see', he goes on, 'if the phenomena of nature can be conveniently explained by the hypothesis of one single principle.' I have explained it sufficiently by showing that there are cases where some disorder in the part is necessary for producing the greatest order in the whole. But it appears that M. Bayle asks a little too much: he wishes for a detailed exposition of how evil is connected with the best possible scheme for the universe. That would be a complete explanation of the phenomena: but I do not undertake to give it; nor am I bound to do so, for there is no obligation to do that which is impossible for us in our existing state. It is sufficient for me to point out that there is nothing to prevent the connexion of a certain individual evil with what is the best on the whole. This incomplete explanation, leaving something to be discovered in the life to come, is sufficient for answering the objections, though not for a comprehension of the matter.