[pg 84] and it is probably not without some reason that Wolowski and other recent French economists attributed to this law of replacement an important effect in preventing disturbance in the relations of gold and silver.” ¹⁵⁷ _Theory of Political Economy_, 4th ed., 1911, pp. 134–36. ¹⁵⁸ It is this artificiality of the bimetallic system which unfortunately befogs the minds of some people and prejudices those of others. Some do not understand why the price determination of two commodities used as money should be so different from the price determination of any other two commodities as to be governed by a ratio fixed by law. Others are puzzled as to why, if gold and silver are a pair of substitutes, should they require a legal ratio while other pairs of substitutes circulate without a legal ratio, merely on the basis of the ratio of their utility. These difficulties are well explained away by Prof. Fisher thus: “… two forms of money differ from a random pair of commodities in being substitutes. Two substitutes proper are regarded by the consumer as a single commodity. Thus lumping together of the two commodities reduces the number of demand conditions, but does not introduce any indeterminateness into the problem because the missing conditions are at once supplied by a _fixed ratio of substitution_. Thus if ten pounds of cane sugar serve the same purpose as eleven pounds of beet-root sugar, their fixed ratio of substitution is ten to eleven. … In these cases the fixed ratio is based on the relative capacities of the two commodities to fill a common need, and is quite antecedent to their prices. … The substitution ratio is fixed by nature, and in turn fixes the price ratio. “In the single case of money, however, there is no fixed ratio of substitution. … We have here to deal not with relative sweetening power, nor relative nourishing power, nor with any other capacity to satisfy wants—no capacity inherent in the metals and independent of their prices. We have instead to deal only with relative _purchasing power_. We do not reckon a utility in the metal itself, but in the commodities it will buy. We assign their respective desirabilities or utilities to the sugars … before we know their prices, but we must inquire the relative circulating value of gold and silver before we can know at what ratio we ourselves prize them. To us the ratio of substitution is incidentally the price ratio. The case of the two forms of money is unique. They are substitutes, but have no natural ratio of substitution, dependent on consumers’ preferences.