A study of any collection of Greek vases will make it apparent that there is a great variety in the forms of the different periods. This is especially marked in the earliest ages of Greece, in which the variety is almost endless, and the adoption and development of certain recognised forms practically unknown. It must therefore be evident that the statements of ancient writers must always be used with caution, and that a shape described by an early writer must not be taken as representing the same in a later period, even if the same word be used, or _vice versa_. For instance, the δέπας ἀμφικύπελλον of Homer, which finds a curious parallel in the gold cup with the doves discovered by Schliemann at Mycenae, is, whatever view we may take of the Homeric civilisation, only an example of a passing fashion. Or again, many of the drinking-cups described by Athenaeus in his eleventh book are doubtless only instances of new experiments in pottery or metal-work characteristic of the Hellenistic age, with its tendency to strive after novelties. Many of his names are little more than nicknames for familiar shapes, which enjoyed a temporary popularity. Some information may be derived from the vases themselves by means of inscriptions, specimens of which are given in Chapter XVII. Thus on the François vase the three-handled pitcher used by Polyxena is inscribed [ΥΔΡΙΑ], or “water-pot,” and enables us to apply the name hydria with certainty to a three-handled vase, of which many black- and red-figured specimens exist.[495] Then we have the _lekythos_ of Tataie, and the _kylikes_ of Philto and Kephisophon,[496] which testify by inscriptions to the name by which they were known. The names incised in _graffito_ on the feet of vases[497] are a more doubtful source of evidence, inasmuch as they may refer either to mixed batches of vases or to the names of measures of capacity. Examples of cursory mention of names in the ancient writers, such as Aristophanes, are innumerable, but seldom explicit, and the scholia on these writers are hardly more useful, inasmuch as the grammarians probably knew little more about obsolete shapes than we do ourselves, and their commentaries have little critical weight. The _loci classici_ on the subject are the book of Athenaeus already referred to,[498] in which he gives a list of over one hundred names, with more or less full explanation and commentary, most of the forms being apparently varieties of drinking-cups, and the _Onomasticon_ of Pollux.[499] Notices of vases are also to be found in the lexicographers, such as Hesychius and Suidas, and the _Etymologicum Magnum_.