"But what I'm really curious about is 0.999... Who first noticed that Stevin's system gave certain numbers two different decimal expansions? How was the anomaly dealt with?" Alexander Malcom's "A new system of arithmetick" (1730) states matter-of-factly: "If the repetend of any circulate is 9, the value or sum of that series is a unit of the place next to that repetend on the left hand, so, .9 [a dot over the 9] = 1". Richard Gadesby (A treatise of decimal arithmetic, 1757): "'Tis evident that .9 = 9/10 wants only 1/10 of unity; and .99 wants 1/100; .999 wants 1/1000; so that if the series were continued to infinity, the difference between that series of nines and an unit, would be equal to unity divided by infinity, that is, nothing at all." S.F. Lacroix (An elementary treatise on arithmetic, 1825 translation of 1818) concedes the result but feels it necessary to explain that unity is "but a limit".