The whole subject has been treated exhaustively—especially from a philological point of view—in a valuable treatise by P. Kretschmer,[2061] to which we shall have occasion to make constant reference in the following pages. He classifies them under two main headings: (_a_) inscriptions incised with a sharp tool in the hard clay; (_b_) inscriptions painted with the brush after the final baking. They are also found in very rare instances impressed in the soft clay and varnished over.[2062] In later times inscriptions in relief are actually found, sometimes painted with thick white pigment, sometimes gilded.[2063] On the so-called Megarian bowls and on the Arretine and other wares of the Roman period they are stamped from the moulds. Lastly, there are the stamps imprinted on the handles of wine-amphorae, which have been discussed in Chapter IV. The =incised inscriptions= are of three kinds: (1) those executed by the maker of the vase; (2) those scratched under the foot; (3) those incised by the owner. As these represent a much smaller class than the painted ones, they shall be dealt with first. (1) Inscriptions incised by the maker before the final baking. These are found on the handles and feet, round the edge of a design, or interspersed therewith like the painted inscriptions. Generally they represent the signature of the potter, as in the case of the early Boeotian vase signed by Gamedes,[2064] the vases of the fifth-century artist Hieron,[2065] and those of Assteas, Python, and Lasimos in Southern Italy.[2066] On the vases of the latter class explanatory inscriptions seldom occur, but when they do (as on the vases of Assteas) they are always incised. Of their palaeographical peculiarities we will speak later. On a vase in the South Kensington Museum[2067] the words Βραχᾶς καλός are incised and painted red, and on the pottery found on the site of the Kabeirion at Thebes the same process is often adopted, except that the paint used is white.[2068] (2) Of inscriptions scratched under the foot a considerable number remain, especially on B.F. vases. They are often difficult to decipher, being in the form of monograms, and frequently appear to be meaningless. In many cases they may have been private marks of the potter or his workmen; others, again, are evidently private memoranda made by the workman, relating to the number of forms of vases in his batch, or by the merchant respecting the price to be paid. Commonly they take the form of names of vases,[2069] such as [ΗVΔΡΙ] for ὑδρία (_hydria_), [ΛΗΚ] or [ΛΗΚV] for λήκυθος (_lekythos_), [ΣΚV] for σκύφος (_skyphos_),[2070] and so on. Many of the inscriptions give the words in full, with numbers and prices, and we may obtain from them some curious information. Among the more elaborate examples given by Schöne in his valuable monograph is one from a krater in the Louvre[2071]: [ΚΡΑΤΕΡΕΣ : ΠI] κρατῆρες ἑξ [ΤΙΜΕ : ͰͰͰͰ ΟΞΙΔΕΣ : [Π]ΙΙΙ] τιμὴ τέσσαρες ὀξίδες ὀκτώ