Ianuarium

Haec fuit a Romulo annua ordinata dimensio, qui, sicut supra iam diximus, annum decem mensium, dierum vero quattuor et trecentorum habendum esse constituit, mensesque ita disposuit ut quattuor ex his tricenos singulos, sex vero tricenos haberent dies. sed cum is numerus neque solis cursui neque lunae rationibus conveniret, non numquam usu veniebat ut frigus anni aestivis mensibus et contra calor hiemalibus proveniret, quod ubi contigisset, tantum dierum sine ullo mensis nomine patiebantur absumi quantum ad id anni tempus adduceret quo caeli habitus instanti mensi aptus inveniretur. sed secutus Numa, quantum saeculo rudi et adhuc impolito solo ingenio magistro comprehendere potuit, vel quia Graecorum observatione forsan instructus est, quinquaginta dies addidit, ut in trecentos quinquaginta quattuor dies quibus duodecim lunae cursus confici credidit annus extenderetur. atque his quinquaginta a se additis adiecit alios sex retractos illis sex mensibus qui triginta habebant dies, id est de singulis singulos, factosque quinquaginta et sex dies in duos novos menses pari ratione divisit. ac de duobus priorem Ianuarium nuncupavit primumque anni esse voluit, tamquam bicipitis dei mensem, respicientem ac prospicientem transacti anni finem futurique principia; secundum dicavit Februo deo, qui lustrationum potens creditur. lustrari autem eo mense civitatem necesse erat, quo statuit ut iusta dis Manibus solverentur.
(Macrobius, Sat. 1.12.38-1.13.3)

This was the measure of the year set in order by Romulus, who, as I said above, established that the year should be reckoned at ten months or 304 days, and arranged the months so that four of them had thirty-one days, while six had thirty. But since that number matched neither the revolution of the sun nor the phases of the moon, it sometimes happened that the cold season fell out in the summer months and, conversely, the hot season in the winter months: when that happened, they let pass as many days—assigned to no particular month—as were needed to bring them to the time of year when the condition of the heavens matched the month they were in. But Numa followed Romulus. With all the understanding he could muster, following his wits alone in an age still uncouth and unrefined—or perhaps because he learned from the Greeks—he added fifty days, extending the year to the 354 days he thought comprised twelve lunar cycles. To these additional fifty days he added another six taken from the months that had thirty days—that is, one day from each—and divided the resulting fifty-six days into two months of equal length. The first of these he named January and decided it should be the first month of the year, the month of the two-headed god, looking back to the end of the year past and ahead to the start of the year to come. The second he dedicated to the god Februus, who is believed to control rites of purification: the community had to be purified in that month, when he determined that the Good Gods be paid the offerings due to them. (tr. Robert A. Kaster)