"^7"^ Miscellaneous Observations. that the division of the months by Nones, Ides, and Calends, arose from the attempt to preserve this ancient division in combination with lunar months and a lunar year. Each month was supposed to contain four periods of 8 days ; but as this would have made the months too long, the first of the four periods, from the Calends to the Nones, was arbi- trarily shortened; in March, May, July, and October, by one day ; in the rest of the months by three days. Never- theless, the last day of the first period retained its significant name, Nonce^ the ni7ith day; that is, according to the Latin idiom, by which both extremes of any period are counted in. This curtailment aiFords the reason of the solemnity, the account of which is preserved by Macrobius (Sat. I. 15); that the pontiff's, after observing the new Moon {J ana Novella)^ gave notice to the people on what day the Nones were to be reckoned. They knew the length of the other constant periods without notice. With respect to the Romulian year, which is said to have been divided into ten months, I would utterly reject, as Niebuhr seems to do (Vol. i. p. ^73), the account of Ma- crobius and Solimus, by which they make out the 304 days by assigning 31 days to the four months already named, and 30 days to each of the others. This allotment is inconsistent with a division into eight day periods. If the religious year were divided into months at all, they probably consisted in general of 32 days ; and then, there must either have been one of only 16 days, or two of 24 days each. Plutarch, Numa c. 18. says that some months consisted of fewer than 20 days, while some were extended to S5^ and others even to more. Before I quit the subject, I will observe that there was a certain symmetry in the mode in which the months of 29 days in the old lunar year were lengthened out in Julius Caesar'^s solar year ; which will be best understood by inspection. January .31 days. April 30 days. June 30 August 31 September 30 November 30 December 31 IL M.
