avoided, from an apprehension that there might be in innovation, the seeds of confusion. In the ledger books of merchants and tradesman at the end of the fifteenth, and the beginning of the sixteenth centuries, (should there be any such remaining) items of receipts and payments in the vulgar figures may be discovered; but I somewhat question their being met with in any other MSS.
Printing must have accelerated the progress of Arabic numerals. By that excellent invention there would be an increase of scholars in arithmetic, and the knowledge of it attained with greater facility. Nor could the compositors of the press have had the same prejudices against these figures that the writers and transcribers of MSS. might entertain, because it might affect their livelihood. A cursory view of a proof meet, upon which were impressed the same sums of money in the common cyphers, and in Roman capitals, would immediately satisfy an impartial examiner which class was the more eligible.
In the middle of the sixteenth century Robert Record, fellow of All Souls college in Oxford, and doctor of physic, published his perfect Work and Practice of Arithmetic. It was dedicated to king Edward the VIth, and contained two dialogues between the master and scholar, the former of them teaching the art and use of arithmetic with the pen, the second the accounting by counters. Of this book the first edition in 8vo. was in 1540; the second in 8vo. with Augmentations, by John Dee, in 1590; and a third, in 12mo. enlarged by John Mellis, in 1658, of which I have procured a copy. And as the dialogues are printed in black letter, it may be presumed that all the numerals, whether letters or figures, correspond with those characters in the original edition.
Both Wallis and Ward mention their having seen in MSS. the Roman characters blended with Arabian figures; and it has been shewn that there is the like mixture in inscriptions on monuments and coins. Record has given all his examples in common figures,but
