Under which term, observe, we include all documents of debt which, being honest, might be transferable, though they practically are not transferred; while we exclude all documents which are in reality worthless, though in fact transferred temporarily as bad money is. The document of honest debt, not transferred, is merely to paper currency as gold withdrawn from circulation is to that of bullion. Much confusion has crept into the reasoning on this subject from the idea that withdrawal from circulation is a definable state, whereas it is a gradated state, and indefinable. The sovereign in my pocket is withdrawn from circulation as long as I choose to keep it there. It is no otherwise withdrawn if I bury it, nor even if I choose to make it, and others, into a golden cup, and drink out of them; since a rise in the price of the wine, or of other things, may at any time cause me to melt the cup and throw it back into currency; and the bullion operates on the prices of the things in the market as directly, though not as forcibly, while it is in the form of a cup, as it does in the form of a sovereign. No calculation can be founded on my humour in any case. If I like to handle rouleaus, and therefore keep a quantity of gold, to play with, in the form of jointed basaltic columns, it is all one in its effect on the market as if I kept it in the form of twisted filigree, or steadily amicus lamnae, beat the narrow gold pieces into broad ones, and dined off them. The probability is greater that I break the rouleau than that I melt the plate; but the increased probability is not calculable. Thus, documents are only withdrawn from the currency when cancelled, and bullion when it is so effectually lost as that the probability of finding it is no greater than that of finding new gold in the mine.