Front Page Titles (by Subject) Systems of Fractional Money. - Money and the Mechanism of Exchange
The Online Library of Liberty
A project of Liberty Fund, Inc.
Systems of Fractional Money. - William Stanley Jevons, Money and the Mechanism of Exchange 
Money and the Mechanism of Exchange (New York: D. Appleton and Co. 1876).
About Liberty Fund:
The text is in the public domain.
Fair use statement:
Systems of Fractional Money.
A unit of value having been chosen, there are three competing methods according to which it might be subdivided, the binary, duodecimal, and decimal. The first system is carried out most perfectly in our avoirdupois weights, in which sixteen ounces make a pound; but it is also freely employed in our monetary system, the sovereign being divided into half-sovereigns, crowns, and half-crowns, the shilling into sixpences and threepenny pieces; and the penny into halfpence and farthings. At the same time, the duodecimal method is represented in our money by the division of the shilling into twelve pence, of which the third part is still in circulation as the groat, or fourpenny piece, now being withdrawn.
Each system of subdivision has its own advantages, and there must always exist a kind of natural competition between them. They have thus competed from the earliest times. In ancient Italy the duodecimal system predominated to the south of the Apennines, while the decimal division was in use to the northward. In Sicily the two methods were confused together. China has had a purely decimal system from an unknown epoch in antiquity. In England duodecimal and binary divisions have existed from very early times. It will be readily allowed that the binary system is most simple and natural, involving as it does the least possible factor above unity. The duodecimal system also has marked advantages, because it allows of division into several aliquot parts, involving the factor 2 twice over, and the next higher factor 3 once. Thus the shilling is divisible exactly into two sixpences, three fourpences, four threepences, and six twopences.
The decimal system is far less simple, and in some ways less convenient. Ten admits of only two factors superior to unity, namely, 2 and 5, and 5 is a more complex prime factor than appears in either of the previous methods. But the system has the supreme advantage of exactly falling in with our decimal system of numeration and calculation. Although probably not the best method which might have been selected, had selection been open to us, decimal numeration is firmly fixed among the institutions of the human race, as an hereditary habit, derived from the early practice of counting on the fingers. We have no choice but to accept the inevitable, and as all our arithmetical processes are conducted on the decimal method, there is an overwhelming advantage, as education and the use of writing advance, in making all our weights, measures, and coins conformable to the same system.
A perfectly and purely decimal system, indeed, would admit only the decimal multiples and submultiples, thus:—1000, 100, 10, 1, 0.1, 0.01, 0.001. But it is so troublesome to have to count out as many as ten coins, before coming to the next higher unit, that the rigour of the decimal divisions has always been relaxed. In the French system, the half and the double of each multiple are allowed to be represented by intermediate coins, the series being 1, 2, 5, 10, 20, 50, 100, 200, 500, etc. The American coinage is less simple and symmetrical, since it admits the half and quarter eagle, half and quarter dollar, the ten and five-cent pieces, and also a three-cent piece. I am inclined to prefer the French method, and to think that the American mint has issued too many dominations of coins.