When Ricardo suddenly gets off the straight path of his presentation and takes the very opposite view, he does so to turn his attention to the international circulation of precious metals and thus brings confusion into the problem by introducing considerations that are foreign to the subject. Let us follow his own course of reasoning, and, in order to remove everything that is artificial and incidental, let us assume that the gold and silver mines are located in the interior of the countries in which the precious metals circulate as money. The only inference which follows from Ricardo’s reasoning as so far developed, is that, the value of gold being given, the quantity of money in circulation will be determined by the prices of commodities. Thus, at a given moment, the quantity of gold in circulation in a country is simply determined by the exchange value of the commodities in circulation. Let us suppose now that the sum total of these exchange values has declined either because there are less commodities produced at the old exchange values, or because, in consequence of an increased productivity of labor, the same quantity of commodities has a smaller value. Or, we may assume on the contrary that the sum total of exchange values has increased, either because the quantity of commodities has increased while the cost of their production has remained the same, or because the value of the same or of a smaller quantity of commodities has risen in consequence of a diminished productivity of labor. What becomes in either case of the _given_ quantity of metal in circulation? If gold is money merely because it is current as a medium of circulation; if it is compelled to remain in circulation like government legal tender paper money (and that is what Ricardo has in mind), then the quantity of money in circulation will rise above the normal level, as determined by the exchange value of the metal, in the former case, and fall below that level in the latter. Although possessing a value of its own, gold will become in the former case a token of a metal of lower exchange value than its own, and in the latter, a token of a metal of higher value. In the former case it will remain as a token of value less than its own, in the latter greater than its own (again an abstract deduction from legal tender paper money). In the former case it is the same as though commodities were estimated in a metal of lower value than gold, in the latter, as though they were estimated in a metal of higher value. In the former case, prices of commodities would rise therefore, in the latter they would fall. In either case the movement of prices, their rise or fall, would appear as the effect of a relative expansion or contraction of the volume of gold in circulation above or below the level corresponding to its own value, i. e. above or below the normal quantity which is determined by the proportion between its own value and that of the commodities in circulation.