Page:Archaeologia Volume 13.djvu/206

156" -Additional Remarks on the the letters, and on abbreviations. This ought however to be more clofely examined, and if poffible determined, in order to afcertairt the knowledge, though not the ufe, of the Arabic numerals at that, period. In the account prefixed to the calendar of the fourteenth century, referred to by Mr. North, there is added a paragraph ftill more explicit of the threefold divifion of numbers into digits, ar- ticles, and compound or mixed. Record's defcription cited in my letter to Mr. Gough has rendered into Engliih a fimilar paragraph ; and it feems to merit a refearch which may be the earlieft MS. in Latin in which this elucidation of the Arabic numerals occurs, and whether a tranilation of it is to be found in any book previous to- the introduction of printing. But fmce it is clear that Arabic numerals were not unknown to Matt. Paris, how are we to account for Chaucer in his Dreme ftyling them newe figures ? The probable folution of this difficulty is, that thefe figures ten had been then but lately ufed for the ad- dition and fub traction of pounds, millings, and pence ; and the fur- mife acquires weight from the concomitant words of the poet, that " Argus, the noble countour, counted with them in his countour." And yet proof is wanting of this ufe of the vulgar figures from the time of Chaucer to an advanced period of the fixteenth century. In books (writes Mr. North) they were doubtlefs firft ufed, and books have been examined to no purpofe. Imagining that though the fearch had been fruitlefs in the libraries of feveral colleges in. Oxford, fome examples might be traced in the books of accompts upon the ihelves and in the boxes of the Burfar's apartments, I hinted a wifh to Mr. Ellis to purfue, when quite convenient, this new line of inquiry. He willingly acceded to it, and this is the re- fult of his examination. " In the enumeration of thofe colleges whofe burfary accounts you wiihed me to fearch, Merton (the moft ancient Society) ap- pears