I do not know of any book or published paper devoted to Cleostratus. In the indexes to most histories of astronomy you will seek his name in vain, and, where you do find it, you are referred to a few jejune paragraphs or more often to a single sentence. Boll in his Sphaera (1903) honours him with three pages (191–194), based on a passage in Pliny and a scholium on Euripides, but he misinterprets both passages and holds one of them to be based on a misunderstanding of some older writer. Nearly all the passages bearing on him are to be found with notes of varying value in Diels, Fragmente der Vorsokratiker, ii. (1912), pp. 197, 198, where they occupy rather more than a page. Some valuable comments and one reference which is not in Diels will be found in Breithaupt's treatise De Parmenisco Grammatico (1915). And yet, for all this neglect, there are attributed to Cleostratus two capital contributions to Greek astronomy, viz. the introduction of the signs of the zodiac and the authorship of the eight years' cycle of intercalations.
1 So Breithaupt, op. cit. p. 31.
2 Diels, , Fragmente der Vorsokratiker, ii. 198Google Scholar, argues that several octaeterides must have passed before the necessary corrections could be discovered, and, therefore proposes to place Cleostratus about 520 B.C. The date is reasonable enough, but the argument implies that Cleostratus's octaeteris was used in practice and Harpalus's corrections were based on experience. There is no ground for either assumption. As will be seen later, the octaeteris remained an astronomical conceit, and there is every reason to believe that the succession of octaeterides and other cycles produced by the astronomers of the fifth century B.C. did not owe their origin to defects in earlier systems proved by experience, but were exercises in the art of combining days, months, and years, of which the relative mean durations had been learned from Babylon.
3 Cleanthes is included among the ὰρχαȋοι in Geminus, xvi. 2. From the way in which the ἀρχαȋοι are habitually criticised by Hipparchus, it would appear that they were differentiated from the more modern astronomers by the inferiority of their mathematical methods.
4 I am assured by Dr. Langdon that the evidence in the case of Cancer is unsatisfactory.
5 Herodotus acknowledges the sun-dial, the gnomon, and the twelve hours of the day.
6 According to Drachmann this is the MS. reading, which he takes to represent ἀρχομένης. Weniger on the authority of Tycho Mommsen gives ὰρχόμϵνα.
7 This subject will be more fully treated in Mr. Forbes's book on the Attic Calendar and Chronology of Thucydides.
8 Martin, in Revue Archéologique, nouv. série, ix. (1864), pp. 170–199Google Scholar, makes much of these difficulties. Newcomb, , Researches on the Motion of the Moon, Part I. (1878), pp. 28–30Google Scholar, rejected the prediction for the reason given in the text, but in Part II. of the same work (1912), p. 231, he stated ‘There can be little doubt that Thales predicted this eclipse.’