76 Journal of Philology. and the months and days in each year are counted forwards in the natural way in which the people who acted in them counted them : years are therefore the present retrogressive unit. Now the word century for 100 years is a very common word among us, and a most useful and vivid idea, wonderfully helping our historical conception : the word chiliad for 1000 years would, now that history rather tends to enlarge its temporal dimensions, and we are endeavouring, for the purpose of finding laws of it and making it a science, to conceive it in large portions, be equally real and useful. And if we made chiliads the retrogres- sive unit for pre-Christian dating, and counted the years forward in each chiliad, we should have all the advantages which arise from having only one arbitrary independent epoch, and none of the disadvantages which arise from continuous retrograde dating. All the positive ancient chronology which we have at present, independent of the Sacred Writings and of any Assy- rian and Egyptian discoveries, would come within the last chi- liad, and the dating of it would run straightforward. It is as if we assumed a possible succession of sub-epochs, each of them fixed by being 1000, 2000, &c. complete solar years distant in previous time from the fixed point or epochal night which marks our era, the yesterday of which was Dec. 31, B. C. 1 (retrograde) and the morrow of which was Jan. 1, A. D. 1. The signature of such a way of dating would not be difficult. 1 would propose to write the retrograding chiliads in Roman numerals, and the progressive years of the chiliad in common ones: and then to prefix A. C. where we wish fully to mark it: the signature then A. C. I. 500 (for instance) might be supposed to mark either ' Ante Christum I. 500/ or * Anno Chiliadis Primae 500/ But supposing any general use of dating such as this, the signature would be commonly no more necessary than it is for dates A. D. : we might talk of the battle of Salamis having been in I. 521, five hundred and twenty-one of the 1st Chiliad, which we should understand to mean the 1st ancient one (retrograde), as easily as of the battle of Pa via having been in 1525, fifteen hundred and twenty-five. The Roman numeral stands before the epoch as a subtracting figure as IX is one less than 10, and IIX 2 less than it : the other numerals mark an addition to the time signified by this. The advantages of such or some such way of dating are these.
