Our calendar is a historical absurdity, without even the redeeming feature of accuracy at least on the cosmological timescales which I intended to challenge.

I wrote furious letters to The Times, proposing reforms which would enable us to function accurately and without ambiguity on timescales of genuine value to the modern scientist. To begin with, I said, let us discard all this nonsensical clutter of leap years. The year is close to three hundred and sixty-five days and a quarter; and that accidental quarter is the cause of all this ridiculous charade of leap year adjustments. I offered two alternative schemes, both guaranteed to remove this absurdity. We could take the day as our base unit, and devise regular months and years based on multiples of days: imagine a three-hundred-day Year made up of ten Months, each of thirty days. Of course the cycle of seasons would soon drift out of synchronization with the structure of the Year, but — in a civilization as advanced as ours — that would surely cause little trouble. The Royal Observatory at Greenwich, for example, could publish diaries each year to show the dates of the various solar positions the equinoxes and so forth just as, in 1891, all diaries showed the movable feasts of the Christian churches.

On the other hand, if the cycle of seasons is to be regarded as the fundamental unit, then we should devise a New Day as an exact fraction — say a hundredth — of the year. Naturally this would mean that the diurnal round, our periods of dark and light, of sleep and wakefulness, would fall at different times each New Day. But what of that? Already, I argued, many modern cities operated on a twenty-four-hour schedule. And as for the human side of it, simple diary keeping is not a difficult skill to acquire; with the help of proper records one would need plan one’s sleeping and wakefulness no more than a few Days in advance.