No long experience was needed to discover the fact that the majority of solar eclipses were followed some fourteen days and a half after by an eclipse of the moon; but they were unable to take sufficient advantage of this experience to predict with certainty the instant of a future eclipse of the sun, although they had been so struck with the connection of the two phenomena as to believe that they were in a position to announce it approximately.* They were frequently deceived in their predictions, and more than one eclipse which they had promised did not take place at the time expected:** but their successful prognostications were sufficiently frequent to console them for their failures, and to maintain the respect of the people and the rulers for their knowledge. Their years were vague years of three hundred and sixty days. The twelve equal months of which they were composed bore names which were borrowed, on the one hand, from events in civil life, such as “Simanu,” from the making of brick, and “Addaru,” from the sowing of seed, and, on the other, from mythological occurrences whose origin is still obscure, such as “Nisanu,” from the altar of Ea, and “Elul,” from a message of Ishtar. The adjustment of this year to astronomical demands was roughly carried out by the addition of a month every six years, which was called a second Adar, Blul, or Nisan, according to the place in which it was intercalated. * Tannery is of opinion that the Chaldaeans must have predicted eclipses of the sun by means of the period of two hundred and twenty-three lunations, and shows by what a simple means they could have arrived at it. ** An astronomer mentions, in the time of Assurbanipal, that on the 28th, 29th, and 30th of the month he prepared for the observation of an eclipse; but the sun continued brilliant, and the eclipse did not take place. The neglect of the hours and minutes in their calculation of the length of the year became with them, as with the Egyptians, a source of serious embarrassment, and we are still ignorant as to the means employed to meet the difficulty. The months had relations to the signs of the zodiac, and the days composing them were made up of twelve double hours each. The Chaldaens had invented two instruments, both of them of a simple character, to measure time--the clepsydra and the solar clock, the latter of which in later times became the source of the Greek “polos.” The sun-dial served to determine a number of simple facts which were indispensable in astronomical calculations, such as the four cardinal points, the meridian of the place, the solstitial and equinoctial epochs, and the elevation of the pole at the position of observation. The construction of the sundial and clepsydra, if not of the polos also, is doubtless to be referred back to a very ancient date, but none of the texts already brought to light makes mention of the employment of these instruments.*