The Italians, till recently, counted the hours in a single series, between two settings of the sun. The only gain in such a method would be to sailors, that they might know how many hours they had before night overtook them; the sun always setting at twenty-four o'clock; if the watch marked nineteen or twenty, it would mean they had five or four hours to see by—but such a gain would be very small against the necessity of setting their watches differently every morning, and the inconvenience of never having fixed hours for meals.

Among the Babylonians, Syrians, Persians, the modern Greeks, and inhabitants of the Balearic Isles, &c., the day commenced with the rising of the sun. Nevertheless, among all the astronomical phenomena that may be submitted to observation, none is so liable to uncertainty as the rising and setting of the heavenly bodies, owing among other things to the effects of refraction.

Among the ancient Arabians, followed in this by the author of the Almagesta, and by Ptolemy, the day commenced at noon. Modern astronomers adopt this usage. The moment of changing the date is then always marked by a phenomenon easy to observe.

Lastly, that we may see how every variety possible is sure to be chosen when anything is left to the free choice of men, we know that with the Egyptians, Hipparchus, the ancient Romans, and all the European nations at present, the day begins at midnight. Copernicus among the astronomers of our era followed this usage. We may remark that the commencement of the astronomical day commences twelve hours after the civil day.

Of the various periods composed of several days, the week of seven days is the most widely spread—and of considerable antiquity. Yet it is not the universal method of dividing months. Among the Egyptians the month was divided into periods of ten days each; and we find no sign of the seven days—the several days of the whole month having a god assigned to each. Among the Hindoos no trace has been found by Max Müller in their ancient Vedic literature of any such division, but the month is divided into two according to the moon; the clear half from the new to the full moon, the obscure half from the full to the new, and a similar division has been found among the Aztecs. The Chinese divide the month like the Egyptians. Among the Babylonians two methods of dividing the month existed, and both of them from the earliest times. The first method was to separate it into two halves of fifteen days each, and each of these periods into three shorter ones of five days, making six per month. The other method is the week of seven days. The days of the week with them, as they are with many nations now, were named after the sun and moon and the five planets, and the 7th, 14th, 19th, 21st, and 28th days of each month—days separated by seven days each omitting the 19th—were termed "days of rest," on which certain works were forbidden to be done. From this it is plain that we have here all the elements of our modern week. We find it, as is well known, in the earliest of Hebrew writings, but without the mark which gives reason for the number seven, that is the names of the seven heavenly bodies. It would seem most probable, then, that we must look to the Accadians as the originators of our modern week, from whom the Hebrews may have—and, if so, at a very early period—borrowed the idea.

It is known that the week was not employed in the ancient calendars of the Romans, into which it was afterwards introduced through the medium of the biblical traditions, and became a legal usage under the first Christian Emperors. From thence it has been propagated together with the Julian calendar amongst all the populations that have been subjected to the Roman power. We find the period of seven days employed in the astronomical treatises of Hindoo writers, but not before the fifth century.

Dion Cassius, in the third century, represents the week as universally spread in his times, and considers it a recent invention which he attributes to the Egyptians; meaning thereby, doubtless, the astrologers of the Alexandrian school, at that time very eager to spread the abstract speculations of Plato and Pythagoras.

If the names of the days of the week were derived from the planets, the sun and moon, as is easy to see, it is not so clear how they came to have their present order. The original order in which they were supposed to be placed in the various heavens that supported them according to their distance from the earth was thus:—Saturn, Jupiter, Mars, the Sun, Venus, Mercury, the Moon. One supposition is that each hour of the day was sacred to one of these, and that each day was named from the god that presided over the first hours. Now, as seven goes three times into twenty-four, and leaves three over, it is plain that if Saturn began the first hour of Saturday, the next day would begin with the planet three further on in the series, which would bring us to the Sun for Sunday, three more would bring us next day to the Moon for Monday, and so to Mars for Tuesday, to Mercury for Wednesday, to Jupiter for Thursday, to Venus for Friday, and so round again to Saturn for Saturday.

The same method is illustrated by putting the symbols in order round the circumference of a circle, and joining them by lines to the one most opposite, following always in the same order as in the following figure. We arrive in this way at the order of the days of the week.