Besides the cheapness and celerity with which this machine will perform its work, the _absolute accuracy_ of the printed results deserves especial notice. By peculiar contrivances, any small error produced by accidental dust, or by any slight inaccuracy in one of the wheels, is corrected as soon as it is transmitted to the next, and this is done in such a manner as effectually to prevent any accumulation of small errors from producing an erroneous figure in the result. In order to convey some idea of this stupendous undertaking, we may mention the effects produced by a small trial engine constructed by the inventor, and by which he computed the following table from the formula _x^2_ + _x_ + 41. The figures, as they were calculated by the machine, were not exhibited to the eye as in sliding-rules and similar instruments, but were actually presented to the eye on two opposite sites of the machine, the number 383, for example, appearing in figures before the person employed in copying. _Table calculated by a small Trial Engine._ 41 131 383 797 1373 43 151 421 853 1447 47 173 461 911 1523 53 197 503 971 1601 61 223 547 1033 1681 71 251 593 1097 1763 83 281 641 1163 1847 97 313 691 1231 1933 113 347 743 1301 2021 While the machine was occupied in calculating this table, a friend of the inventor undertook to write down the numbers as they appeared. In consequence of the copyist writing quickly, he rather more than kept pace with the engine, but as soon as five figures appeared, the machine was at least equal in speed to the writer. At another trial _thirty-two_ numbers of the same table were calculated in the space of _two minutes and thirty seconds_; and as these contained _eighty-two_ figures, the engine produced thirty-three figures every minute, or more than one figure in every two seconds. On another occasion it produced _forty-four_ figures per minute. This rate of computation could be maintained for any length of time; and it is probable that few writers are able to copy with equal speed for many hours together.