Sun-dried bricks were known as πλίνθοι ὠμαί (lateres crudi); baked bricks as πλίνθοι ὠπταί (lateres cocti or coctiles). The Romans also used the word testa for baked brick, corresponding to the Greek κέραμος. Vitruvius^[324] distinguishes three varieties of unburnt bricks, as used by the Greeks. One, known as “Lydian,” was also used by the Romans, who named the bricks from their length sesquipedales; their size was 1½ by 1 ft. The other two, exclusively Greek, were known as πεντάδωρον and τετράδωρον, the word δῶρον signifying a “palm” or three inches; in other words, they were respectively fifteen inches and one foot square. The former was used for public buildings, the latter for private houses, and they were arranged in the walls in courses of alternate whole and half bricks, as is frequently done at the present day. Vitruvius also speaks of bricks made at Pitane in Mysia, and in Spain, which were so light that they would float in water.^[325] He advises that bricks should not be made of sandy or pebbly clay, which makes them heavy and prevents the straw from cohering, so that they fall to pieces after wet. Many other directions are given by him,^[326] but are too lengthy to quote here. Bricks were made in a mould called πλαίσιον, a rectangular framework of boards^[327]; and the sun-dried bricks were, as we learn from the passage quoted above, made by collecting the clay with shovels (ἄμαι) into troughs (λεκάναι) and working it with the feet.^[328] It is probable that we have some allusion to the use of moulds in certain passages from the Latin writers.^[329] The final proceeding was the drying in the sun.

An important branch of the subject is the use of terracotta for roof-tiles and other architectural decorations of temples and other buildings. On this point our knowledge has during the last five-and-twenty years been marvellously increased, the extent of its use in architecture having been hitherto but little suspected.^[330] The generic term for a roof-tile is in Greek κέραμος; they are generally divided into flat tiles (στεγαστῆρες or σωλῆες, tegulae) and covering-tiles (καλυπῆρες, imbrices). Besides the ordinary roof-tiles there must also be taken into consideration four varieties of ornamental tiles which found their place on a classical building. They are: (1) the covering-slabs arranged in a row along the γεῖσον, or raking cornice of the pediment; (2) the κυμάτιον or cornice above the γεῖσον; (3) the cornice along the sides of the building, with spouts in the form of lions' heads, to carry off rain-water; (4) the row of antefixal ornaments or ἀκρωτήρια surmounting the side-tiles.^[331]

The flat roof-tiles or σωλῆες, as in the Heraion of Olympia and other early buildings, are square and slightly concave, so that the raised edges placed side by side may catch under the semi-cylindrical καλυπῆρες, and so be held in their place. The latter are of plain semi-cylindrical form, except the row at the lower edge of the roof, which have attached to them the vertical semi-elliptical slabs known as “antefixae,” of which more later.

The κυμάτια were painted with elaborate patterns of lotos-and-honeysuckle, or maeanders, in red, blue, brown, and yellow, the principle being preserved (as always in Greek architectural decoration) of employing curvilinear patterns only on curved surfaces, rectilinear only on flat surfaces.^[332] At the back was the gutter for collecting rain-water, which ran off through the holes pierced at intervals in the cornice, passing through the mouths of lions’ heads, moulded in very salient relief. These correspond to the gurgoyles of Gothic architecture. Many specimens have been found at Olympia, Elateia, and elsewhere in Greece; one of the finest, from a temple of Apollo at Metapontum, is now in the Bibliothèque Nationale at Paris. It is very finely modelled, and the whole, with the background, richly coloured in red, yellow, and black.^[333] Spouts were sometimes modelled in other forms, such as a Satyric mask, or the fore-part of a lion; of the latter there are some examples in the British Museum.^[334] In the accounts for the erection of the arsenal at the Peiraeus there is an interesting entry relating to these lions’ head spouts, in which they are described as κεραμίδες ἡγέμονες λεοντοκεφάλαι, “principal tiles with lions' heads.”^[335]

The invention of antefixae is attributed by Pliny^[336] to Butades of Sikyon, who is also credited with the invention of modelling in clay, in a well-known story; “he was,” says Pliny, “the first to place masks on the extremities of the roof-tiles, which were at first called bas-reliefs (protypa), but afterwards alto-reliefs (ectypa).”^[337] It is possible that the ἀγάλματα ὀπτῆς γῆς seen by Pausanias in the Stoa Basileios at Athens^[338] were ἀκρωτήρια or antefixal ornaments at the angles of the cornice, but they are more likely to have been modelled free and in the round than in relief on a background.^[339] Such sculptured groups were not uncommon in Greek architecture; thus the cornice of the pediment of the temple of Zeus at Olympia was adorned with a series of figures of Victory. The groups above mentioned represented Theseus slaying Skiron and Eos carrying off Kephalos; and it is interesting to note that a terracotta group with the latter subject found at Cervetri^[340] also undoubtedly came from the cornice of a building.

The manner in which the antefixae were treated by the Greeks and Etruscans for purposes of decoration is well illustrated in the British Museum collection. In Cases 64–71 of the Terracotta Room may be seen a series from Capua of archaic style, the front part being semi-elliptical in form, having within an ornamental border a female bust, Gorgon’s head, or other design in relief, all being richly coloured (Plate II.). The back projects in a semi-cylindrical termination, forming the covering-tile, with an arched support to the upright piece. Similar antefixae were found by Lord Savile at Civita Lavinia (see below), and some have elaborate subjects, such as Artemis with two lions, or a Satyr and Maenad with a panther (Plate II.).^[341] Many have also been found at Cervetri, from which site came some interesting friezes of terracotta now in the British Museum (B 626) and at Berlin. These works of art, with which we must rank for their style the reliefs on the archaic terracotta sarcophagus in the British Museum (see Chapter XVIII.), show throughout a strong influence of Ionic art; though all of local manufacture, their style is purely Greek, as is the case with many of the contemporary works in bronze found in Italy.^[342]

Antefixes from Hellenic sites are not so common, nor do they present the same variety of subject or richness of colour. In many cases, as in the fourth-century British Museum specimens from Asia Minor,^[343] the decoration is confined to scrolls and floral patterns in low relief, the palmette being regarded as the most appropriate decorative motive for this form of tile. An example of this type in the British Museum (C 902 = Fig. 10.), found on the field of Marathon, is inscribed with the name Athenaios. Many later antefixes with remains of colouring have been found at Tarentum, the subjects being chiefly heads of women or mythological personages.

Roof-tiles proper have been discovered in large numbers both in Greece and Italy. Olympia has proved the richest site in this respect, and there are many specimens in the Museums of Athens and Palermo.^[344] Many of them have coloured decoration, and these terracotta remains are almost the only evidence we now have of the extensive system of colouring applied by the Greeks to their temples.^[345]

At Olympia all the buildings have terracotta roofs except the temple of Zeus and two others, the dates varying from the seventh century B.C. down to Roman times. We know from Pausanias^[346] that the temple of Zeus was roofed with marble tiles in imitation of terracotta, an invention traditionally attributed to Byzes of Naxos. The covering-tiles of the Heraion roof (see Fig. 9.) end in semicircular discs painted with ornamental patterns; the flat roof-tiles are of the concave type described above. The normal sixth-century type of roof is seen in the Treasury of the Megarians, which has smooth flat tiles and covering-tiles ending in antefixes with palmette-and-lotos ornament, and a kymation cornice with lion’s head spouts.

A greater variety of tiles is to be seen in the Treasury of Gela. Here for the first time we note the introduction of a new system, which consists in nailing slabs of terracotta over the surface of the stonework, or, to use the convenient German term, “Bekleidungstechnik.”^[347] It is obvious at the first glance that the origin of this practice dates from the time when buildings were largely or wholly of wood, which required protection from the weather. When the wood was replaced by stone, the fashion held its ground for a time; but with the more extensive use of marble, which could not well be covered in this manner, it disappeared altogether in Greece.

But the Treasury of Gela is by a Sicilian architect, and it seems highly probable that the method of decoration employed was not one usually practised in Greece, but was introduced from the Western Mediterranean. Though rare in Greece, it is exceedingly common in Sicily and Southern Italy. The middle temple (known as C) on the acropolis of Selinus, and buildings at Gela and Syracuse, may be cited as examples. The principle is also well illustrated in the terracotta remains of the temple at Civita Lavinia, excavated by Lord Savile in 1890–94, which are now in the British Museum. They have, as far as possible, been incorporated in a conjectural restoration in the Etruscan Saloon (Plate III.).^[348] It will be noted that most of the slabs are pierced with holes, by means of which they were attached to the walls or surface of the entablature; they are mostly decorated with lotos-and-honeysuckle and other patterns, in relief and coloured, the same being repeated in colour only on the back of the overhanging edges of the cornice. These remains belong to two periods, the end of the sixth century and the fourth century B.C.; they may be easily distinguished by the differences in the treatment of the ornamental patterns, while there is a marked absence of colouring in the later remains. Similar architectural remains in terracotta have been found in Etruria, and are described in Chapter XVIII. It should be noted that the Civita Lavinia slabs are flat, whereas those used at Olympia, and many others in Southern Italy and Sicily, are three-sided.

Specimens of ordinary Greek tiles have been found in many parts of the ancient world, besides those for special architectural purposes already discussed. Avolio^[349] mentions many examples from Acrae and elsewhere in Sicily, stamped with emblems or names of officials and of makers. At Olbia, in Southern Russia, tiles were found stamped with names of Greek aediles (ἀστυνόμοι), like the amphora-handles described below (p. 158),^[350] and in Corfu tiles and bricks with names of magistrates (πρυτάνεις), indicating in each case the existence of public regulations concerning the potteries.^[351] At Kertch (Panticapaeum) Dr. Macpherson discovered large numbers of tiles with labels on which was stamped the word ΒΑΣΙΛΙΚΗ, “Royal,” together with other inscriptions.^[352] These tiles showed the manner of their attachment one upon the other, and their dimensions answered to the Lydian variety mentioned above. Other tiles discovered by Mr. Burgon at Athens, by Sir Charles Newton in Kalymnos, and by Mr. Colnaghi at Kandyla (Alyzia) in Acarnania, bore labels with inscriptions and designs in relief.^[353] On one of the latter series in the British Museum is the inscription ALYZEIÔN, “of the people of Alyzia” (Fig. 11); on another was inscribed in the manner of the Athenian vases (see Chapters X. and XVII.) ΙΠΠΕΟΣ ΚΑΛΟΣ ΑΡΙ[Σ]ΤΟΜΕΔΕΙ ΔΟΚΕΙ, “Hippeus seems handsome to Aristomedes.”^[354]

Inscribed tiles from Greece proper are somewhat rare, and the best-known examples, to the number of sixteen, have been collected by M. Paris^[355]; they are usually inscribed with the word ΔΗΜΟΣΙΑ or ΔΗΜΟΣΙΟΣ, as a sort of Government stamp. Others have magistrates’ names, as ΦΡΟΔFΙΣ, Ἀ]φροδ(ε)ισίου, on a tile at Corinth, or the maker’s name, FΑΣΤΟΥΚΡΙΤ, Fαστουκρίτ[ου, on one from Thisbe in Boeotia.^[356] Those found by M. Paris at Elateia have either the word ΔΗΜΟΣΙΟΣ or Ε[Π]Ι with the name of the magistrate; though all are fragmentary, it is possible to restore the full formula as πλίνθος δημοσία ἐπὶ Ἀπελλέα, “government bricks, in the year of Apelleas’ office.”^[357] A remarkable tile or stele, found near Capua and now in the British Museum, has an inscription in Oscan, and two stamps of a boar and a head of Athena, resembling types on Italian coins of the early part of the third century.^[358]

We may recall the fact that it was with a tile that Pyrrhus met his death when besieging Argos. Nor is this the only occasion on which these humble objects have played a part in history. In the well-known Athenian institution of Ostracism the act of voting was performed by writing on fragments of tiles or potsherds the names of those whom it was desired to banish. Recent excavations have yielded more than one actual specimen of these ὄστρακα or sherds,—one bearing the name of Megakles (Fig. 12.); another, part of a painted vase from the pre-Persian débris on the Athenian Acropolis, the name of Xanthippos, the father of Perikles (Fig. 13); and a third, that of Themistokles.^[359]

It is also probable that in Greece, as among the Romans, the hollow floors of the hypocausts, as well as the flue-tiles of the hot baths, were made of terracotta. The same material was also used for the pipes, by means of which water was conveyed from aqueducts or drained from the soil. A drain-pipe from Ephesos in the Museum at Sèvres is noted by Brongniart and Riocreux,^[360] and others have been found at Athens^[361] and in the Troad.^[362]

Tiles were also employed for constructing graves, as has already been noted in Chapter II. (see p. 34). In some tombs the floor was paved with flat tiles, and the roof was constructed of arched tiles forming a vault. The flat and square tiles were not used for tombs until a comparatively late period. Some graves had a second layer of tiles to protect the body from the superincumbent earth.^[363] We shall have occasion to make further allusion to the use of painted terracotta slabs in Etruscan tombs (Chapter XVIII.).

The sarcophagi which played so important a part in the tomb were also frequently made of terracotta, this material being most commonly employed in Etruria. We have already mentioned (p. 62) the series of archaic painted sarcophagi, which have all come from Clazomenae, near Smyrna, and furnish us with much valuable information on the art of painting in Ionia in the sixth century B.C. They will receive some attention from this point of view in Chapter VIII. The British Museum contains two very remarkable examples of Etruscan terracotta sarcophagi, which are described in Chapter XVIII., as well as a series of smaller examples, which are mere cinerary urns. Among other examples of terracotta as used in tombs may be mentioned here a series of small reliefs found in tombs at Capua and elsewhere in Southern Italy. They consist of masks of Satyrs, river-gods, and Gorgons, and are often highly coloured in red and blue. They are of late archaic work, about 480 B.C., but the exact way in which they were used to decorate the tombs is uncertain. The British Museum collection contains many specimens of these objects.^[364]

There is a curious class of objects which hardly come under the heading of any other category, but may be conveniently discussed here. Complete specimens are very rare, but there is one in the Museum at Geneva which has been identified as a brazier (πύραυνος or ἐσχάρα), and more recently as a baking-oven (κλίβανος).^[365] The form is that of a large basin on a high stand, hollow underneath, with three square solid handles projecting upwards from the rim. These handles, of which over a thousand examples are to be found in various collections, are usually the only part remaining, sometimes with part of the rim attached. They are decorated with heads and other devices, usually in relief on square panels, and the majority of these heads are of a Satyric or grotesque character, wearing conical caps or adorned with ivy-wreaths. They probably represent demons of some kind, and are placed there with superstitious intent, to avert evil influences from whatever was baked or cooked in the vessel. Similar masks are usually seen attached to representations of forges and ovens on the painted vases,^[366] and remind us of the pseudo-Homeric invocation of evil deities against the potters of Samos (see also p. 213 below). Professor Furtwaengler has identified the heads as those of the Kyklopes, the attendant workmen of Hephaistos.^[367]

These objects are found all over the Mediterranean, especially at Halikarnassos, Naukratis, and Delos, and the last-named place has been regarded as the centre of their manufacture. They are all of the same brick-like, coarse, red clay. Some bear the name of their maker, Hekataios or Nikolaos. Besides the heads already mentioned, heads of goats or oxen, or of Sirius, thunderbolts and rosettes are used by way of devices. They have been collected together, and illustrations of all the different types given by Conze in the Jahrbuch for 1890, p. 118 ff.: two specimens are given on Plate IV. They belong to the Hellenistic Age.

Other objects that exemplify the use of clay or terracotta in Greek daily life are: moulds for vases and terracotta figures, lamps, weights, and stamps for various purposes. Many flat discs of terracotta have been found at Tarsus, Gela in Sicily, Tarentum, and other places, pierced with two holes and about three inches in diameter.^[368] They are stamped with various devices and inscriptions, but their use is unknown. Other discs of convex form found at Halikarnassos and stamped with heads in relief are supposed to have been weights ([λεῖαι) to hold down the threads of the loom (ἀγνύθες),^[369] such as are used by the Greeks at the present day; others again may be the weights used for keeping the ends of the folds of a himation in position. Small pierced cones of terracotta often found in the fields of Greece have been supposed to have been suspended round the necks of cattle, but are probably weights of some kind.^[370] Lastly, terracotta egg-shaped objects have been found in Sicily inscribed with various names, and are supposed to have been voting-tickets used for the ballots of the tribes.^[371]

Many examples have been found of terracotta impressions from coins, which may have been the trial-pieces of die-sinkers or forgers, since persons of that class, as among the Romans, seem to have employed this material for their nefarious practices. They are more fully discussed in Chapter XIX. The British Museum contains a large collection of these found in the Fayûm in Egypt, all of Roman date; also a copy of a coin of Larissa from Acarnania. Terracotta medallions with impressions of gems or seals are not uncommon, especially in Asia Minor and at Naukratis, and among the latter are many lumps of clay actually used as seals, with the pattern of the substance in which they were impressed adhering to the back of them, while on the front is a design from a signet-ring.^[372]

The subject of Lamps is one that is more conveniently and appropriately treated in the Roman section of this work (see Chapter XX.), almost all existing examples in terracotta being of that period; it may not, however, be out of place to include here a few general remarks on the subject, pointing out the distinctive features of those of purely Greek origin.

The invention of lamps was ascribed by Clement of Alexandria to the Egyptians; and they were certainly in common use among the Greeks. Herodotos^[373] describes those which he saw in Egypt as simple saucers filled with oil in which the wick floated, and this statement is partly supported by the form of the lamps found in the earlier tombs of Cyprus and on sites under Phoenician influence.^[374] He also uses the phrase περὶ λύχνων ἁφάς, “about the time of lighting lamps,” to denote the evening.^[375] The Greek comic writers allude to the use of lamps of terracotta or metal,^[376] and they played a part in religious ceremonies.

The regular Greek name for a lamp was λύχνος (not λαμπάς, which means a torch), and a lampstand was called λυχνοῦχος; the spout or nozzle in which the wick was placed was known as μύξος or μυκτήρ, the wick itself as ἐλλύχνιον.^[377] A lamp with more than one nozzle was known as δίμυξος or τρίμυξος.^[378] The simple form was that derived from the Phoenician lamp, an open saucer with a bent-up lip in which the wick was placed; but commonly the Greek lamp had a circular or oval body (the receiver) with flat covered top, in the centre of which was the filling-hole. To this was sometimes attached a handle permitting the insertion of a finger, and the nozzle was usually very small and quite plain. An epithet applied by Aristophanes^[379] to a lamp is τροχήλατος, “made on the wheel”; but evidence points to their being always made in moulds.

The majority of the lamps which have been found on Greek sites are of Roman date, and they frequently bear Latin inscriptions; those of the Hellenic period are seldom ornamented, and are usually covered with a thin black glaze. Others are modelled in the form of human figures, animals, heads, or sandalled feet; the British Museum possesses a good example of grey ware from Knidos in the form of a figure of Artemis (Cat. C 421), with the oil-receptacle on the top of her head; another from Naukratis represents Eros (see for these Plate IV.). One from Athens was inscribed ΜΗ ΑΠΤΟΥ, “Do not touch,”^[380] an inscription of similar import to those on the Roman lamps from the Esquiline described in Chapter XX.

Little has at present been done in the way of a scientific investigation of Roman lamps, but the results of a rough classification according to shapes show that certain forms are more specially associated with Greek sites, and moreover frequently bear names of makers in Greek letters. This is particularly the case with one form, which appears to be confined to Athens, Corfu, the coast of Asia Minor, and Cyprus. These lamps, of a pale yellow clay, have a circular body with flat top, round the edge of which runs a border of impressed egg-pattern, interrupted on either side by a small plain raised panel.^[381] The handle is small and pierced with a hole, the nozzle also small, with straight sides. These lamps bear the makers’ names (in the genitive), Primus (ΠΡΕΙΜΟΥ), Abaskantos (ΑΒΑCΚΑΝΤΟΥ), etc., the former being especially common; all are in Greek letters. Some again only have a single letter or monogram engraved underneath. They are often very carefully executed, with sharply cut details, and the subjects are usually mythological (see Plate IV. fig. 1); they appear to be of very late date, not earlier than the third century after Christ.

Another form which appears to be specially characteristic of Greek sites is that with a plain or heart-shaped nozzle, sometimes with a groove incised at the base, but without a handle. They are usually quite small, with circular bodies. Large numbers of these were found by Mr. Newton at Knidos in 1859,^[382] and by Mr. Barker at Tarsos in 1845.^[383] The subjects are mostly poor and devoid of interest, including animals, rosettes, and various floral patterns. Many of these lamps bear the signature ROMAINE(N)SIS, the form of the word indicating that they were made by a Roman residing abroad (i.e. at Knidos), not in Rome.^[384] A third form, approximating to the Christian type, has a small solid handle and plain nozzle, and is confined to sites on or near the coast of Asia Minor. These, with the remaining types of lamps, will be more fully dealt with in the Roman section of this work. It may, however, be worth while mentioning here that Mr. Newton found at Knidos several lamps of a coarse black ware, covered with thin glaze, which are mostly of large size. They are circular, and convex above, and are supplied with two or more long nozzles with blunt terminations radiating round them (see Plate IV. fig. 6). Between the nozzles are roughly stamped devices of Satyrs’ heads, flowers, etc., in relief. These may fairly be regarded as a Greek type.

The subject of Greek sculpture in terracotta is so wide as to demand a volume to itself; but a discussion of the uses to which clay was put by the Greeks would not be complete without some mention of their achievements in this direction. We propose therefore briefly to review the main features of Greek terracotta statuettes and reliefs, by way of illustrating the purely artistic use which they made of this material.

The subject may be divided under four heads: (1) Large statues; (2) Statuettes or figurines; (3) Reliefs; (4) Moulds. Large or life-size statues belong more particularly to the earlier phases of Greek art, but appear again in its later developments, under Italian influences. Statues of terracotta were also a common feature of Italian art, being, in fact, the usual material employed by Etruscan statuaries, as well as for the decoration of temples (see Chapter XVIII.). Greek terracotta statues are practically non-existent; and although there are some female figures nearly life-size and a male torso of almost colossal proportions in the British Museum, also a Hermes in the Vatican, these were found at Rome, belong to the Roman period, and, though Greek in style, are really following an Etruscan fashion.

It is characteristic of the Hellenic race that from its earliest beginnings it did not employ clay for utilitarian purposes exclusively, but, influenced partly by the natural imitative instincts of man, partly by the anthropomorphic tendencies of the Greek religion, soon learned the value of this easily worked material for producing images of deities, animals, and other objects. Although an equally high antiquity may be claimed for images of wood, and the word ξόανον used for a primitive cult-statue argues for the frequent use of this material, yet the history of the word πλάσσειν tells equally in the other direction. Originally used of moulding wet clay, it came by degrees to denote modelling in general, and finally its derivative πλαστική became the authorised classical word for sculpture.

Lactantius^[385] speaks of Prometheus as the inventor of fictile images for religious purposes, and of figures in bronze and marble as a later development; the Latin poets^[386] bear similar witness to the primitive use of clay for sculptured images, and Pliny marvels at its long-continued employment in Italy.^[387] Among early Greek legends the most noteworthy is that of Butades, the potter of Sikyon, to whom the invention of modelling clay in relief was ascribed by Pliny^[388] and Athenagoras. The story as told by the former was that, in order to preserve the likeness of his daughter’s lover, he moulded in terracotta the shadow of his profile which the girl drew on the wall. This account, however, is not very intelligible, and the clue is perhaps to be found in the words of Athenagoras,^[389] who says that he hollowed out the lines of the face in the wall, filled in the grooves with clay, and so obtained his relief as from a mould. This primitive work of art was said to have been exhibited in the Nymphaeum at Corinth.

But this same invention was also claimed by the Samian sculptors, Theodoros and Rhoikos, who flourished about the end of the seventh century. They were pre-eminently artists in bronze, and were associated with the introduction of hollow-casting in that material into Greece; it may therefore be supposed that they actually were among the first to use clay models for statues, this being an essential preliminary to the hollow-casting process. This would not be incompatible with the invention of moulding reliefs by Butades, admitting the truth of his story. The latter was also credited with the invention of antefixal ornaments (see above, p. 98) and the introduction of a mixture of red ochre or ruddle with clay in order to give it a warmer tone.

The clay models used by sculptors as the basis of their work, which were known as προπλάσματα, were probably made on the same lines as the large works of art in clay. We read that Lysistratos of Sikyon, the brother of Lysippos, was the first to make casts of statues by means of terracotta moulds,^[390] implying that it was about this time that the practice arose of multiplying the principal statues in the same manner as is now done by means of plaster casts. Some of the latter artists combined the plastic art with that of painting, and Zeuxis is said to have previously modelled in terracotta the subjects which he afterwards painted. Pasiteles, an artist who lived at Rome in the first century B.C., always first modelled his statues in terracotta, and spoke of the plastic art as the mother of statuary.^[391] But it must not be supposed that as a general rule the Greek sculptors worked their marble statues from models; rather, the contrary was the case, and Pasiteles seems to have been peculiar in this respect.

The statue of Zeus, which has already been mentioned as made by Theokosmos for Megara (p. 92), appears to have been made from a clay model. It was intended to be of gold and ivory, but the breaking out of the Peloponnesian War prevented the artist from carrying out his intention, and only the head was completed, the other portions being of gypsum and terracotta. At a later period gypsum was sometimes used for sculpture, as in the case of an Apollo mentioned by Prudentius,^[392] and some fragmentary remains from Cyprus in the British Museum.

The clay models were sometimes made entirely by hand, but more usually on a wooden core known as κάναβος,^[393] which we may conjecture to have been formed of two rods in the form of a cross, from the use of the Latin word crux in this connection.^[394] It was certainly a framework, not a solid core, and must be carefully distinguished from κίνναβος, a lay-figure. Aristotle, in an interesting passage, uses the word in speaking of skeletons drawn on a wall.^[395] The modelling of details was done partly with tools, partly with the finger. The use of the finger-nail for this purpose became proverbial, as in the saying attributed to Polykleitos: “When the clay has reached the finger-nail stage, then the real difficulty begins.”^[396]

The chief attention of inferior artists was directed to the production of small terracotta figures, which the Greeks used as ornaments or household gods, buried in their tombs, or dedicated in their temples. They follow the same lines of development as the larger sculptures, beginning with the columnar (ξόανα) and board-like (σανίδες) types found in the primitive tombs of the Mycenaean and early Hellenic civilisation. Originally they seem to have been manufactured purely for religious purposes, but in course of time, with the gradual rationalising of religious beliefs and consequent secularisation of art-types, they lost this significance, and, while the types were preserved, they were converted into genre figures from daily life.

These statuettes have been found on nearly all the famous sites of antiquity from Babylonia to Carthage and Kertch; the most fruitful have been Tanagra in Boeotia, Rhodes, the Cyrenaica, Capua and Canosa in Italy, and various sites in Sicily. In Cyprus, Sardinia, and to a great extent also in Rhodes, Phoenician influences seem to have been dominant, and the earlier types bear a markedly Oriental character. For beauty and charm the palm has by general consent been given to the Tanagra statuettes of the fourth and third centuries, which were known in antiquity as κόραι or “maidens,” from the prevalence of the seated or standing types of girls in various attitudes.

The makers of these charming figures, known as κοροπάσται or κοροπλάθοι, were, like the vase-painters, quite in a subordinate position in the artistic world, and are spoken of with some contempt by Isokrates, as if it would be absurd to compare them with a Pheidias or a Zeuxis.^[397] A fable of Aesop’s^[398] represents Hermes being offered a statue of Zeus for a drachma and one of himself for a mere song; the low price seems to suggest that they were of terracotta, but the vendor is called an ἀγαλματοποιός, not a κοροπλάθος. Demosthenes^[399] condemns the Athenians for voting for figure-head generals like makers of toys for the market; and in further illustration of the uses to which they were put, we may cite the definition of Suidas, of “those who fashion little images out of clay of all kinds of creatures, with which to trick children”; and the remark of Dio Chrysostom, who speaks of those who buy the “maiden” figures for their children. A pretty epigram in the Anthology^[400] tells how Timareta, when about to marry, dedicated to Artemis the playthings of her childhood, including her terracotta dolls (κόρας). Lastly, Plato speaks of κόραι and images hung up in shrines.^[401]

The processes employed in the manufacture of terracotta statuettes were five in number: (1) the preparation of the clay; (2) moulding; (3) retouching; (4) baking; and (5) colouring and gilding. It does not follow that all five were employed in the production of any one object; on the other hand, all processes necessary to the completion of any one object fall under one or other of these heads.

There were many varieties of clay in use among the Greeks, some being considered more suitable for one purpose, some for another. These clays vary in their characteristics in different parts of the Greek world, and this may often be an important criterion for distinguishing fabrics and detecting instances of importation. The clay of Cyprus differs much from that of Rhodes, and that of Naukratis again from either, being of a dark, coarse, and brick-like consistency. M. Pottier noted nine varieties of clay in use at Myrina in Asia Minor, and M. Martha distinguishes five in the terracottas of Athens. But these differences may be explained by variations in the length or temperature of the firing rather than in the clay.

Generally speaking, the clay of the terracottas is softer and more porous than that of the vases. It is easily scratched or marked, and does not ring a clear sound when struck; nor does it when submitted to a high temperature become so hard as the pottery.^[402] Its colour ranges from deep red to a pale buff colour, and its texture and density vary greatly in different localities. It was prepared by being washed free of all granular substances, and then kneaded with the aid of water. So, as we read in Hesiod’s account of the creation of Pandora,^[403] the god directed the mixing of clay and water, in order to form his new creation.

The modelling of the figures was done by hand in the case of the earlier fabrics, and of small objects such as the toys and dolls; the clay was worked up into a solid mass with the fingers, and the marks of these, left while it was wet, may still be often seen. Subsequently the use of moulds became universal, the final touches being given to the figure either with the finger or with a graving-tool, traces of which are often visible on the faces and hair of the Tanagra figures. These were invariably moulded, and the finer ones show traces of having been most carefully touched up.

There is a pretty epigram in the Anthology,^[404] which seems to imply that the wheel was sometimes brought into use for modelling figures, perhaps for the first rough outlining. A statuette of Hermes is supposed to say: