spite of the example set by George Eliot's hero, may refrain from marrying Dinah for fear of breaking his brother Seth's heart. Equally vain would be the search for any rule invariably applicable to political affairs. Even general propositions which sound like truisms are not universally true. It cannot even be said that misgovernment always produces discontent, or that the combination of superior strength and superior strategy is always successful in war; for examples might be cited of nations remaining patient under an iron despotism, and perhaps also of campaigns lost by armies with every advantage of skill, numbers, and discipline on their side. No better specimen can be given of what are popularly spoken of as historical laws than one propounded by Mr. C. Merivale, whose careful study of Roman annals has taught him to regard it as 'a condition of permanent dominion that conquerors should absorb the conquered gradually into their own body, by extending, as circumstances arise, a share in their own exclusive privileges to the masses from whom they have torn their original independence.' The principle thus laid down is of great value, but it must not be mistaken for an index pointing unerringly to a goal which will certainly be obtained by following its direction. At least the offer of Austrian citizenship had no perceptible effect in overcoming the exclusiveness of Hungarian nationality; nor in inducing Venetia to become a willing member of a Teutonic Federation, and to lend the same assistance to the House of Hapsburg, as Gaul and Spain did to the Caesars, in suppressing insurrection on the banks of the Danube. History supplies many principles similar to the one evolved by Mr. Merivale, all more or less useful for the guidance of the statesman. So far as they are just, they indicate the results which would spring from the adoption of certain lines of policy, unless something unforeseen should happen. It is true that something unforeseen is almost sure to happen and to divert or impede the course which events would otherwise take; but still, it is most important to be able to perceive clearly the influence exerted by certain causes, how much soever that influence may be disturbed by other causes; since, if it does nothing else, it will at least prevent the disturbing causes from producing what would otherwise have been their full effect. On principles which indicate only a few out of many causes in simultaneous operation, it is evident that nothing deserving to be called predictions can be founded; but from them, nevertheless--inasmuch as they teach that some causes act for good and others for evil, as far as their action extends--practical rules of government may be deduced. Such rules, however, which at best can only furnish a loose and shifting basis for doubtful conjectures, stand without the confines of positive knowledge; they occupy a middle-ground between science and nescience,