Now suppose the sun to be in the same latitude as the Pleiades on some particular day; owing to the proximity of the group to the ecliptic, it will be then very near the sun, and it will set with it and be invisible during the night. If the sun were to the east of the Pleiades they would have already set, and the first view of the heavens at sunset would not contain this constellation; and so it would be so long as the sun was to the east, or for nearly half a year; though during some portion of this time it would rise later on in the night. During the other half year, while the sun was to the west, the Pleiades would be visible at sunset, and we immediately see how they are thus led to divide the whole year into two portions, one of which might be called the Pleiades below, and the other the Pleiades above. It is plain that the Pleiades first become visible at sunset, when they are then just rising, in which case they will culminate a little after midnight (not at midnight, on account of the twilight) and be visible all night. This will occur when the sun is about half a circle removed from them—that is, at this time, about the beginning of November; which would thus be the commencement of one half of the year, the other half commencing in May. The culmination of the Pleiades at midnight takes place a few days later, when they rise at the time that the sun is really on the horizon, in which case they are exactly opposite to it; and this will happen on the same day all over the earth. The opposite effect to this would be when the sun was close to the Pleiades—a few days before which the latter would be just setting after sunset, and a few days after would be just rising before sunrise.

We have thus the following observations, that might be made with respect to this, or any other well-marked constellation. First, the period during which it was visible at sunset; secondly, the date of its culmination at midnight; thirdly, its setting in the evening; and fourthly, its rising in the morning: the last two dates being nearly six months removed from the second. There are also the dates of its culmination at sunrise and sunset, which would divide these intervals into two equal halves. On account of the precession of the equinoxes, as explained in the last chapter, the time at which the sun has any particular position with respect to the stars, grows later year by year in relation to the equinoctial points. And as we regulate our year by the date of the sun's entrance on the northern hemisphere, the sidereal dates, as we may call them, keep advancing on the months. As, however, the change is slow, it has not prevented years being commenced and husbandry being regulated by the dates above mentioned. Any date that is regulated by the stars we might expect to be nearly the same all over the world, and the customs observed to be universal, though the date itself might alter, and in this way. So long as the date was directly obtained from the position of the star, all would agree; but as soon as a solar calendar was arranged, and it was found that at that time this position coincided with a certain day, say the Pleiades culminating at midnight on November 17, then some would keep on the date November 17 as the important day, even when the Pleiades no longer culminated at midnight then, and others would keep reckoning by the stars, and so have a different date.

With these explanations we shall be able to recognise how much the configurations of the Pleiades have had to do with the festivals and calendars of nations, and have even left their traces on customs and names in use among ourselves to the present day.

We have evidences from two very different quarters of the universality of the division of the year into two parts by means of the Pleiades. On the one hand we learn from Hesiod that the Greeks commenced their winter seasons in his days by the setting of the Pleiades in the morning, and the summer season by their rising at that time. And Mr. Ellis, in his Polynesian Researches, tells us that "the Society Islanders divided the year into two seasons of the Pleiades, or Matarii. The first they called Matarii i nia, or the Pleiades above. It commenced when, in the evening, these stars appeared at or near the horizon, and the half year during which, immediately after sunset, they were seen above the horizon was called Matarii i nia. The other season commenced when at sunset these stars are invisible, and continued until at that time they appeared again above the horizon. This season was called Matarii i raro, i.e. the Pleiades below." Besides these direct evidences we shall find that many semi-annual festivals connected with these stars indicate the commencement of the two seasons among other nations.

One of these festivals was of course always taken for the commencement of the year, and much was made of it as new-year's day. A new-year's festival connected with and determined by the Pleiades appears to be one of the most universal of all customs; and though some little difficulty arises, as we have already pointed out, in fixing the date with reference to solar calendars, and differences and coincidences in this respect among different nations may be to a certain extent accidental, yet the fact of the wide-spread observance of such a festival is certain and most interesting.

The actual observance at the present day of this festival is to be found among the Australian savages. At their midnight culmination in November, they still hold a new-year's corroboree, in honour of the Mormodellick, as they call the Pleiades, which they say are "very good to the black fellows." With them November is somewhat after the beginning of spring, but in former days it would mark the actual commencement, and the new year would be regulated by the seasons.

In the northern hemisphere this culmination of the Pleiades has the same relation to the autumnal equinox, which would never be taken as the commencement of the year; and we must therefore look to the southern hemisphere for the origin of the custom; especially as we find the very Pleiades themselves called Vergiliæ, or stars of spring. Of course we might suppose that the rising of the constellation in the morning had been observed in the northern hemisphere, which would certainly have taken place in the beginning of spring some 5,000 years ago; but this seems improbable, first, because it is unlikely that different phenomena of the Pleiades should have been most noticed, and secondly, because neither April nor May are among any nations connected with this constellation by name. Whereas in India the year commenced in the month they called Cartiguey, which means the Pleiades. Among the ancient Egyptians we find the same connection between Athar-aye, the name of the Pleiades, with the Chaldeans and Hebrews, and Athor in the Egyptian name of November. The Arabs also call the constellation Atauria. We shall have more to say on this etymology presently, but in the meantime we learn that it was the phenomenon connected with the Pleiades at or about November that was noticed by all ancient nations, from which we must conclude that the origin of the new-year's spring festival came from the southern hemisphere.

There is some corroboration of this in the ancient traditions as to the stars having changed their courses. In the southern hemisphere a man standing facing the position of the sun at noon would see the stars rise on his right hand and move towards his left. In the northern hemisphere, if he also looked in the direction of the sun at noon, he would see them rise on his left hand. Now one of a race migrating from one side to the other of the equator would take his position from the sun, and fancy he was facing the same way when he looked at it at noon, and so would think the motion of the stars to have altered, instead of his having turned round. Such a tradition, then, seems to have arisen from such a migration, the fact of which seems to be confirmed by the calling the Pleiades stars of spring, and commencing the year with their culmination at midnight. In order to trace this new-year's festival into other countries, and by this means to show its connection with the Pleiades, we must remark that every festival has its peculiar features and rites, and it is by these that we must recognise it, where the actual date of its occurrence has slightly changed; bearing, of course, in mind that the actual change of date must not be too great to be accounted for by the precession of the equinoxes, or about seventy-one years for each day of change, since the institution of the festival, and that the change is in the right direction.

Now we find that everywhere this festival of the Pleiades' culmination at midnight (or it may be of the slightly earlier one of their first appearance at the horizon at apparent sunset) was always connected with the memory of the dead. It was a "feast of ancestors."

Among the Australians themselves, the corroborees of the natives are connected with a worship of the dead. They paint a white stripe over their arms, legs, and ribs, and, dancing by the light of their fires by night, appear like so many skeletons rejoicing. What is also to be remarked, the festival lasts three days, and commences in the evening; the latter a natural result of the date depending on the appearance of the Pleiades on the horizon at that time.