I begin with time; and the question shortly is, What was the measure of time before artificial measures were invented? and, What is the measure at present when these are not at hand? I speak not of months and days, which we compute by the moon and sun; but of hours, or in general of the time that runs betwixt any two occurrences when there is not access to the sun. The only natural measure we have, is the train of our thoughts; and we always judge the time to be long or short, in proportion to the number of perceptions that have passed through the mind during that interval. This is indeed a very imperfect measure; because in the different conditions of a quick or slow succession, the computation is different. But however imperfect, it is the only measure by which a person naturally calculates time; and this measure is applied on all occasions, without regard to any occasional variation in the rate of succession. This natural measure of time, imperfect as it is, would however be tolerable, did it labour under no other imperfection than the ordinary variations that happen in the motion of our perceptions. But in many particular circumstances, it is much more fallacious; and in order to explain these distinctly, I must analize the subject. Time is generally computed at two different periods; one while time is passing, another after it is past. I shall consider these separately, with the errors to which each of them is liable. It will be found that these errors often produce very different computations of the same period of time. The computation of time while it is passing, comes first in order. It is a common and trite observation, That to lovers absence appears immeasurably long, every minute an hour, and every hour a day. The same computation is made in every case where we long for a distant event; as where one is in expectation of good news, or where a profligate heir watches for the death of an old man who keeps him from a great estate. Opposite to these are instances not fewer in number. To a criminal the interval betwixt sentence and execution appears miserably short; and the same holds in every case where one dreads an approaching event. Of this even a schoolboy can bear witness: the hour allowed him for play, moves, in his apprehension, with a very swift pace: before he is thoroughly engaged, the hour is gone. A reckoning founded on the number of ideas, will never produce computations so regularly opposite to each other; for a slow succession of ideas is not connected with our wishes, nor a quick succession with our fears. What is it then, that, in the cases mentioned, moves nature to desert her common measure for one very different? I know not that this question ever has been resolved. The false reckonings I have suggested are so common and familiar, that no writer has thought of inquiring for their cause. And indeed, to enter upon this matter at short hand, without preparation, might occasion some difficulty.