We have already noted that historical names frequently occur in this series, and it is obvious that if they can be identified with the actual historical owners of such names much valuable information in regard to the chronology of Greek vases will be gained. The question has already been discussed in a previous chapter (Vol. I. p. 403), and the principles there laid down need not be repeated. It is sufficient to say that so far only two or three names have been identified with those of historical personages, though more results may yet be obtained. Of these one is Stesileos, occurring on two vases in Berlin, and identified with a _strategos_ who fell at Marathon in 490.[2215] On two lekythi (one late B.F., the other R.F.) the name of Glaukon son of Leagros[2216] appears, and these two names have also been identified with Athenian _strategi_, Leagros having fallen in battle against the Edones in 467, while Glaukon commanded at Kerkyra in 433–432 B.C. It may be roughly inferred that Leagros was a boy (παῖς) about 510 B.C., and his son Glaukon about 470 B.C., which gives an approximate date (within ten years or so) for these two groups of vases. It is, however, obvious that much at present only rests on hypothesis. It is curious to note that nearly all these names have an aristocratic sound: thus we have Alcibiades, Alkmaeon, Hipparchos, and Megakles, besides those already quoted. Miltiades καλός occurs on a R.F. plate at Oxford,[2217] but there seems hardly sufficient evidence for referring it to the youth of the conqueror of Marathon (cf. Vol. I. p. 403). The table at the end of this chapter may be found useful as giving a _conspectus_ of the principal names and their relation to the artists. * * * * * It is now necessary to discuss some of the principal peculiarities of the Attic vase-inscriptions, in regard to palaeography, orthography, and grammar.[2218] The variety in the forms and uses of the letters is somewhat surprising at first sight, but it must be remembered that non-Attic influences were always strong, as has indeed already been pointed out.