<P> Another possible source of the suanpan is Chinese counting rods, which operated with a decimal system but lacked the concept of zero as a place holder . The zero was probably introduced to the Chinese in the Tang dynasty (618--907) when travel in the Indian Ocean and the Middle East would have provided direct contact with India, allowing them to acquire the concept of zero and the decimal point from Indian merchants and mathematicians . </P> <P> The normal method of calculation in ancient Rome, as in Greece, was by moving counters on a smooth table . Originally pebbles (calculi) were used . Later, and in medieval Europe, jetons were manufactured . Marked lines indicated units, fives, tens etc. as in the Roman numeral system . This system of' counter casting' continued into the late Roman empire and in medieval Europe, and persisted in limited use into the nineteenth century . Due to Pope Sylvester II's reintroduction of the abacus with modifications, it became widely used in Europe once again during the 11th century This abacus used beads on wires, unlike the traditional Roman counting boards, which meant the abacus could be used much faster . </P> <P> Writing in the 1st century BC, Horace refers to the wax abacus, a board covered with a thin layer of black wax on which columns and figures were inscribed using a stylus . </P> <P> One example of archaeological evidence of the Roman abacus, shown here in reconstruction, dates to the 1st century AD . It has eight long grooves containing up to five beads in each and eight shorter grooves having either one or no beads in each . The groove marked I indicates units, X tens, and so on up to millions . The beads in the shorter grooves denote fives--five units, five tens etc., essentially in a bi-quinary coded decimal system, related to the Roman numerals . The short grooves on the right may have been used for marking Roman "ounces" (i.e. fractions). </P>

1. who was the first to use an abacus