From this actual coincidence, at this remote period, M. Fourier, who has confirmed all these accounts by new calculations, concludes, that, since the length of the year of Sirius was so perfectly known to the Egyptians, they must have determined it by observations made during a long series of years, and conducted with great accuracy; observations which must be referred to at least 2500 years before the present time, and which could not have been made long before or long after this interval of time[187]. This result would assuredly be very striking, had it been directly, and by observations, made upon Sirius itself, that they had fixed the length of the year of Sirius. But experienced astronomers affirm it to be impossible that the heliacal rising of a star could afford a sufficient foundation for exact observations on such a subject, especially in a climate where _the circumference of the horizon is constantly so much loaded with vapours, that, in clear nights, stars of the second or third magnitude can never be seen within some degrees of the verge of the horizon, and that the sun itself is completely obscured at its rising and setting_.[188] They maintain, that, if the length of the year had not been otherwise ascertained, there would have been a mistake of one or two days.[189] They have no doubt, therefore, that this duration of 365 days and a quarter, is that of the tropical year inaccurately determined by the observation of the shadow, or by that of the point where the sun rose each day, and through ignorance identified with the heliacal year of Sirius; so that it would be mere chance which had fixed with so much accuracy the duration of this latter for the period of which we speak.[190] Perhaps it will also be judged, that men capable of making observations so exact, and which they had continued during so long a period, would not have attributed so much importance to Sirius, as to pay him religious homage; for they would have seen that the relations of the rising of this star with the tropical year, and with the inundation of the Nile, were merely temporary, and took place only in a determinate latitude. In fact, according to M. Ideler’s calculations, in the year 2782 before Christ, Sirius appeared in Upper Egypt, on the second day after the solstice; in 1322, on the third; and in the year 139 after Christ, on the twenty-sixth.[191] At the present day, its heliacal rising is more than a month after the solstice. The Egyptians would therefore set themselves by preference to finding the period, which would bring about the coincidence of the commencement of the sacred year, with that of the true tropical year, and then they would discover that their great period must have been 1508 sacred years, and not 1461.[192] Now, we assuredly do not find any traces of this period of 1508 years in antiquity.