A Dictionary of Greek and Roman Antiquities (1890)
William Smith, LLD, William Wayte, G. E. Marindin, Ed.

---

TELA

TELA (ἱστός), a loom. The elementary principle of weaving being merely the crossing of threads over and under, it is probable that it first took the form of simple plaiting (Lucret. 5.1349; cf. the term ἔμπλεξις τοῦ στήμονος, Plat. Polit. p. 282 E); but we have no record of a time when the real loom in some form or other was unknown to the Greeks and Romans. Its construction in many points is clear, but there are also several questions which cannot be answered with certainty, and about which we must be content with conjectures. Even now the dispute whether writers of the Augustan age are speaking of the upright or the horizontal loom cannot be said to be ended.

From plaiting comes naturally the idea of stretching fixed threads and working a cross thread alternately over and under: for everything [p. 2.765]woven consists of two parts, the fixed thread or warp (stamen, στήμων), and the woof or weft (subtemen, later trana,¹ κροκή). Instead of κροκὴ we sometimes find ἐφυφὴ used (Plat. Legg. v. p. 734 E), and in this passage, as well as in Plat. Polit. p. 283 E, we find noticed one of the most important differences between the warp and the weft; viz. that the threads of the former are strong and firm in consequence of being more twisted in spinning, while those of the latter are comparatively soft and yielding. This is in fact the difference which in modern silk manufacture distinguishes organzine from tram, and in cotton manufacture twist from weft. Another name for the weft or tram was ῥοδάνη (Batr. 181; Eustath. ad Il. 23.762; Od. 5.121).

It may facilitate reference to arrange the parts of the loom under different heads, noting the terms discussed in each:--I. Words connected with the arrangement of the stamen (στήμων, warp); viz. the framework, jugum, insubuli, scapi, κελέοντες; ordiri, διάζομαι, καῖρος: pondera ἀγνῦθες, λεῖαι: II. Those connected with the licia or μίτοι ( “shedding” by leashes or heddles); viz. arundo, liciatorium, perhaps insilia (κανών, perhaps ἀντίον = “heddle-leaf” ): III. With the radius (κερκίς, shuttle); viz. πήνη, panus (bobbin or spool), subtemen, trama (κροκή, weft, woof or tram): IV. With the spatha, and the later pecten (σπάθη and κτείς = reed, lay, batten): V. The question of upright and horizontal looms: VI. Style and pattern.

I. The threads of the warp were called stamina, στήμονες, because they were, at any rate originally, fixed at certain intervals in a row, upright, i.e. perpendicularly from the top to the bottom of the loom (Varr. L. L. 5.113). For the same reason the very first operation in weaving was to set up the loom, ἱστὸν στήσασθαι (Hom. Od. 2.94; Hes. Op. 779); and the web or cloth, before it was cut down or “descended” from the loom (κατέβα ἀφ̓ ἱστῶ, Theoc. 15.35), was called “vestis pendens” or “pendula tela” (Ov. Met. 4.395; Ep. 1.10), because it hung down from the transverse beam, which was probably the jugum, our “yarnbeam.” This transverse beam with the two upright side-posts (ἱστοπόδες or κελέοντες, Theoc. 17.34) formed the whole framework (ἱστὸς or tela) of the primitive loom. Blümner indeed denies that jugum had this meaning, on the ground that Ovid (Ov. Met. 6.55) uses it in speaking of what he believes to be the horizontal loom. But (1) from analogy of the other senses of the word [see JUGUM], the only natural view is, that in the loom it means the bar connecting the uprights (Ahrens well compares the jugum under which the vanquished passed): (2) this explains the jugum of the lyre, which we may suppose to have been named from a resemblance to the loom, the strings stretched from the jugum being compared to the threads of the warp: (3) even if Blümner's theory were right as to the horizontal loom--(for arguments against it, see under V.)--the jugum might still be the cross-bar joining the side-pieces, on which the warp-threads (here called collectively “tela” ) are bound: and (4) the name of tela jugalis (Cat. Agr. 10, 14) is then easily explained as being the primitive loom, in which the warp was fastened directly to the jugum, with no second cross-bar or “yarn-beam” underneath, as in the woodcut below. The doubtful words to be noticed in the structure of the loom are insubuli and scapi. The former is explained by Blümner as = κανόνες: the words of Isidore, however (Or. 19.29), “insubuli quia infra supra sunt,” seem rather to indicate that the yarnbeam and cloth-beam were together known by this name. In the earliest and simplest frame, as will be shown, where the web was not longer than the loom itself, the top-bar acted as the yarn-beam, and there was no cloth-beam at all, but (as in the Chiusi vase, and in the Icelandic loom represented in fig. 1) it was a useful addition to have a second upper bar as yarn-beam, which might take the form of a roller with a reserve of warp, and again, instead of the καῖρος and weights at the bottom of the loom, to have a beam on which the cloth could be rolled as it was made. Scapi, according to Blümner, also = κανόνες, as in the gloss “scapi, κανονες γερδιακοί:” Rich and Monro (ad Lucret. 5.1361) translate it “yarn-beam:” the use of scapus in Plin. Nat. 13.77 for the roll of a book would rather suggest the cloth-beam as its part in the loom; but on the whole, if we reason from the ordinary sense of scapus, we may best suppose the scapi of the loom to be the side-posts = κελέοντες, for which a Latin name is wanted: the epithet “sonans” may refer to the rattling of the loom generally.

The fastening of the warp to the top-bar or jugum was called specially διάζεσθαι, στημονίζεσθαι, and in Latin ordiri, exordiri (Plaut. Pseud. 1.4, 6; Bacch. 2.3, 116; cf. Cic. de Or. 2.3. 3, 145): the handing of the threads for this process, when two persons were setting up the loom, is προφορεῖσθαι, which involved some running backwards and forwards, which is the meaning of the word in Aristoph. Birds 4 (Schol. ad loc.; Hesych. sub voce the rendering in L. and S. is at variance with these authorities). The process is well illustrated by Nonnus (Dion. 6.150):-- καὶ ποσὶ φοιταλέοισι παλίνδρομος ἀκρὸν ἀπ̓ ἀκροῦ
πρωτοπαγῆ ποίησε διάσματα, φάρεος ἀρχήν,
ἱστῷ δ᾽ ἀμφὶς ἕλισσεν.

This moving backwards and forwards (ἱστὸν ἐποίχεσθαι, Od. 10.222) belonged to the old fashion of standing to weave, before the fashion of sitting and beginning the web at the bottom of the loom was introduced from Egypt. Nonnus describes it in Epic manner, though the loom at which the worker sat no doubt prevailed in his [p. 2.766]time. In setting the warp for lighter fabrics the threads were stretched fewer and further apart, and the web was then ἀραιόστημος or μανόστημος, as opposed to the thicker and coarser (στημόνιον, πολύστημος or πυκνόστημος. But the warmth required in winter was secured by driving the softer weft threads closer, στημόνι δ᾽ ἐν παύρῳ πολλὴν κρόκα μηρύσασθαι (Hes. Op. 538).

It may be supposed that the threads of the warp would easily fall out of place and become entangled unless they were secured at both ends: this in more modern looms is effected by the “yarn-roll” or “yarn-beam” at one end, and the “cloth-roll” at the other. In the older Greek and Roman looms the warp was fastened to the jugum at the top, and the lower end of each warp-thread was passed through a loop (καῖρος), and also had a weight attached to it to make it hang straight. This lower row of loops (καῖροι, καίρωμα) must (as Blümner rightly shows, Techn. 1.126) be distinguished from the μίτοι or licia with which they are sometimes confused. This is clear from the explanation in Etym. Mag. and Eustath. ad Od. 7.107, παρὰ τὸν μίτον [i. e. parallel to, but below, the μίτοι] ὑπὲρ τοῦ μὴ συγχεῖσθαι τοὺς στήμονας. Hence it is probable that the Homeric adjective καιροσέων is a mere synonym for ὑφασμένων. The weights attached to the end of each warp-thread were called ἀγνῦθες or λεῖαι (Poll. 7.36), in Latin merely pondera (Sen. Ep. 90, 20): they were either simple stones with a hole bored through them, or made of pottery: a great number of these have been found: Blümner refers to Ritschl, Uebcr antike Gewichtsteine, Bonn, 1866; see also the account of those found at Hissarlik (Schliemann, Troja, p. 163); it is possible that many of the terra-cotta “whorls” which he thinks intended for spindles may have been weights for weaving (ib. p. 41). In the Scandinavian ode translated by Gray as The Fatal Sisters, the weights are warriors' skulls. Perhaps the expression there, “the weights that play below,” may explain the λίθον ὀρχηστῆρα of Nonn. Dion. 24.254.

Whilst the improvements in machinery have to a great extent superseded the use of the upright loom in all other parts of Europe, it remains almost in its primitive state in Iceland. The following woodcut is reduced from an engraving of the Icelandic loom in Olaf Olafsen's Economic Tour in that island, published in Danish at Copenhagen, A.D. 1780, which will probably illustrate the earlier Greek and Roman loom better than any of the few representations on ancient vases which have been discovered. (For the best of these, Penelope's loom, on a vase from Clusium, see Baumeister, Denkm. fig. 2332.) We observe underneath the jugum a roller which is turned by a handle, and on which the web is wound as the work advances. The threads of the warp are divided into thirty or forty parcels, to each of which a stone is suspended for the purpose of keeping the warp in a perpendicular position and allowing the necessary play to the strokes of the spatha, which is drawn at the side of the loom: they correspond to the καῖρος or καίρωμα described above. These knotted bundles of threads to which the stones were attached often remained after the web was finished, in the form of a fringe. [FIMBRIA.] In the centre of the web we see the attachment of the threads of the warp by means of leashes

Fig. 1. Icelandic loom.

to three rods (κανόνες, liciatoria). This important and intricate part of the loom needs some explanation.

II. In the most primitive method of weaving it is probable that the passage for the weft was opened merely by a transverse rod (arundo, κανὼν) passed through the warp, separating the threads so that they were alternately on either side of the arundo. This seems to be shown in Circe's loom (fig. 4), for we can hardly think Ahrens (Philog. xxxv. p. 391) right in taking it to be Circe's magic wand. Such a method of course only admits of plain weaving without a pattern, and moreover the shifting of the arundo would be slow and tedious: in order that the weft might be taken backwards and forwards across the warp passing over or under as might be required, it would be necessary laboriously to raise or depress each thread separately, as in plaiting, unless the improved plan, which was already in use in the Homeric age, of “decussating,” or as it is now called “shedding,” the warp by leashes (μίτοι, licia) had been invented. By a leash, or as weavers term it “a heddle,” we are to understand a thread having at one end a loop, through which a thread of the warp was passed, the other end being fastened to a straight rod (κανών, arundo; later liciatorium). Thus, supposing that only plain weaving without a pattern is required, so that the weft is merely to pass over and under alternately, and we number the warp-threads 1, 2, 3, 4, &c., all the leashes holding the threads of uneven numbers 1, 3, &c. are tied to one rod or liciatorium, while all those holding the threads 2, 4, &c. are tied to another, and by simply moving one rod forward and the other back a free passage is opened for the weft to shoot through. But here, though there might be a coloured stripe by changing at regular intervals the colour of the thread in the weft, or other variations by colouring different threads of the [p. 2.767]warp (see below), there could be no elaborate colour pattern and no pattern at all of the texture. This was produced, just as it is now, by a contrivance for passing over at requisite places a number of warp-threads together, so that the weft might pass under one and over two or under one and over three, and so on. Since it is obvious that it must not be the same single threads that are raised (or, in the upright loom, brought forward) and the same two or more that are depressed (otherwise there would be no weaving at all), it is necessary that there should be an additional set of leashes or “heddles” for every increase of variation, so as to vary the threads which are raised or depressed. When there was one additional set, the weaving was called bilix, δίμιτος, of which the Icelandic loom in the woodcut above gives an example: with two additional sets it was trilix, and then could pass under one and over three: for great complexity of pattern a great many sets of leashes were used (see further below). The details of this part of the subject can be studied in modern weaving. The principle of varying the pattern was really exactly the same as in the loom of to-day: the only difference lies in the mechanical contrivances which make the work vastly more rapid. In the earliest times not only the shuttle, but the liciatoria, κανόνες, or leash-rods ( “heddle-leaves” ), were worked by the hand. This is signified by Homer: ἐπὶ δ᾽ ὤρνυτο δῖος Ὀδύσσευς
ἄγχι μάλ̓ ὡς ὅτε τίς τε γυναικὸς ἐϋζώνοιο
τανίον ἐξέλκουσα παρὲκ μίτον, ἀγχόθί δ᾽ ἴσχει
στήθεος.--(Il. 23.760, imitated by Nonn. Dion. 6.152, 631.)

Here, as rightly explained by Blümner and Marquardt, Odysseus is near to Ajax, as the κανὼν or leash-rod is to the breast of the weaver, when she brings it forward with one hand (ἀγχόθι ἴσχει στήθεος), in order with the other hand to draw the weft through the “shed,” or, as it is expressed, “behind the leashes,” i. e. behind the warp held in the leashes (παρὲκ μίτον). Heyne and others, who make the κανὼν the shuttle, not only give the word a wrong meaning, but miss the point, since at the upright loom part of the warp must always be between the shuttle and the weaver, and it is only the κανὼν which would be moved nearer the body. It is possible, perhaps probable, that the Greeks and Romans of later times used treadles for moving the liciatoria, but we have no direct evidence of it, nor any word to express it: as, however, they worked many machines with the feet, it is unlikely that they omitted to do so in weaving. Some indeed explain INSILIA in Lucret. 5.1352 as treadles, but we believe it to be more correctly understood as=liciatorium (see Munro ad loc.). The Icelandic loom, like the Homeric, has no treadle, and hence we see two rods at the side which can give some help to the single weaver by fixing the leash-rod, as required, while he works the shuttle; but this is clearly a slower process than using the feet to release the hands. The ἀντίον (Aristoph. Thes. 822) Blümner explains as a special name of one of the κανόνες (cf. Poll. 7.36).

III. We have described the comparatively coarse, strong and much-twisted threads designed for the warp arranged in parallel lines, and the leashes ready to “shed” them: we have now to speak of the shuttle which conveyed the weft or woof across. This implement was called κερκὶς in Greek and radius in Latin (Hor. Od. 5.62; Plat. Polit. p. 281 E, Cratyl. p. 388 C; Ov. Met. 4.275, compared with Horn. Il. 22.448): it is imagined of gold in Homer (l.c.), but was usually of wood (Plat. Cratyl. p. 389; Ov. Met. 6.132): the end pointed (Soph. Ant. 976; Ov. Met. 6.56): the humming sound of its passage is expressed by Aristoph. Frogs 1315. The κερκὶς or radius was strictly, like our shuttle, the receptacle for the “bobbin” (πήνη πηνίον, panus, panuvellium) on which the weft was wound (Hom. Il. 23.762; Eur. Hec. 470; Anth. Pal. 6.288; Varr. L. L. 5.114; Isid. Or. 19.29). The annexed woodcut shows the form in which

Fig. 2. The shuttle.

it is still used in some retired parts of our island for common domestic purposes, and which may be regarded as a form of great antiquity. An oblong cavity is seen in its upper surface, which holds the bobbin. A small stick, like a wire, extends through the length of this cavity, and enters its two extremities so as to turn freely. The small stick passes through a hollow cane, which our manufacturers call a quill, and which is surrounded by the woof. This is drawn through a round hole in the front of the shuttle, and, whenever the shuttle is thrown, the bobbin revolves and delivers the woof through this hole. The ancient “shuttles” in the Mayence Museum (Blümner, p. 146) are probably, though not certainly, rightly so named. They are pen-shaped and would have to be turned round for the return passage, only one end being pointed. The process of winding the yarn so as to make it into a bobbin or pen was called πηνίζεσθαι (Theoc. 18.32), or ἀναπηνίζεσθαι (Aristot. H. A. 5.19). The reverse process by which it was delivered through the hole in front of the shuttle (see the last woodcut) was called ἐκπηνίζεσθαι. Hence the phrase ἐκπηνιεῖται ταῦτα means “he shall disgorge these things” (Aristoph. Frogs 586; Schol. in loc.).

IV. Supposing the warp to have been thus adjusted, and the pen or the shuttle to have been carried through it, it was then decussated or “shedded” by drawing forwards the proper rod, so as to carry one set of the threads of the warp across the rest, after which the weft was shot back again (the shuttle being thrown by the hand, as was the case even down to 1738), and by the continual repetition of this process the warp and woof were interlaced; and in “fancy” weaving, with several sets of leashes, the pattern was produced. It was necessary further to close up the weft threads. It has been said above that, after the weft had been conveyed by the shuttle through the warp, it was driven originally in Rome and Greece upwards, as is represented in the first woodcut, but afterwards, according to the prevalent [p. 2.768]fashion in Egypt, downwards, as in the second (Isid. Orig. 19.22; Hdt. 2.35). Two different instruments were used in this part of the process. The simplest and most ancient was in the form of a large wooden sword (spatha, σπάθη, dim. σπάθιον, Plato, Lysis, p. 208; Aesch. Choeph. 226). From the verb σπαθάω, to beat with the spatha, cloth rendered close and compact by this process was called σπαθητός (Athen. 12. 525 d): when the weft is not driven close, as in light, transparent fabrics, it is called λεπτοσπάθητος (Soph. Fr. 400): the close texture πολυσπαθής (Anth. Pal. 6.39). This instrument is still used in Iceland exactly as it was in ancient times, and a figure of it, copied from Olafsen, is given in the first woodcut.

The spatha was, however, superseded by the comb (pecten, κτείς), the teeth of which were inserted between the threads of the warp, and thus made by a forcible impulse to drive the threads of the woof close together. (Ovid, Ov. Fast. 3.820, Met. 6.58; Juv. 9.26; Verg. A. 7.14; Nonn. Dion. 24.253; Poll. 7.35.) As to its form, we are told only that it was an implement with teeth. Blümner doubts if the example from an Egyptian tomb figured by Rich is a pecten, but we think that the correctness of Rich in this point is established by the very similar pecten of which we give an illustration. It is the comb now used in parts of Asia Minor in the loom shown below (fig. 6), which we believe to resemble closely the later form of the Greek and Roman loom. As a late introduction, instead of the σπάθη, it is mentioned only in late Greek writers: it originated in Egypt, whence it is called Niliacus in Mart. 14.150 (cf. Verg. Cir. 179).² Among us the office of the comb is executed with greater ease and effect by the reed, lay, or batten.

Fig. 3. Weaving comb used in Asia Minor. (Benndorff.)

The lyre [LYRA], the favourite musical instrument of the Greeks, was only known to the Romans as a foreign invention. Hence they appear to have described its parts by a comparison with the loom, with which they were familiar. The terms jugum and stamina (Ov. Met. 11.169) were transferred by an obvious resemblance from the latter to the former object; and, although they adopted into their own language the Greek word plectrum (Ovid, Ov. Met. 11.167-170), they used the Latin PECTEN to denote the same thing, not because the instrument used in striking the lyre was at all like a comb in shape and appearance, but because it was held in the right hand and inserted between the strings of the lyre as the comb was between the stamina of the loom (Verg. A. 6.647; Pers. 6.2).

V. The two kinds of upright looms and the supposed horizontal loom.--At some time or other a more convenient form of loom was introduced into Europe, in which the web was worked in a flat horizontal frame instead of hanging vertically in front of the weaver. The parts of this loom are the same in nature and object as those described above, except that, as the warp frame lies flat, the leashes or heddles must be worked vertically up and down instead of backwards and forwards; and if the Romans used such a loom, the licia and liciatorium depended from a cross-beam raised above the flat tela. But when this change came, and even whether it belongs to anything earlier than mediaeval times, is a matter of doubt. The view of Blümner and Marquardt is that in the Augustan age the horizontal loom had already superseded in ordinary use the upright loom. We are led to conclude, though with diffidence in opposing such authorities, that the evidence is not only too slight to warrant such an assertion, but that it points the other way: we go far beyond Rich in this view, and hold, with Ahrens, that the horizontal loom does not belong to ancient Greece and Rome at all, and was probably introduced into Europe by the Arabs.

First, as to the passages which speak definitely of a change in looms: Artemidorus speaks of two kinds of looms, the ἱστὸς ὄρθιος, at which the weaver is said περιπατεῖν, and the ἕτερος ἱστός at which she sits (Oneirocr. 3.36). Similarly, standing to weave is called “oldfashioned” by Festus, pp. 277, 8; 288, 33; Serv. ad Aen. 7.14; Hesych. sub voce ἐποιχόμενοι; Isid. Orig. 19.22. But sitting does not imply the horizontal loom: it merely distinguishes the “Egyptian” fashion of beginning the cloth at the bottom of the loom, as will be seen below. Nor can any argument be drawn from the implements used. Blümner (as was said before) believes jugum to be only the beam for hanging licia above the horizontal loom, and therefore assumes that all looms, in which a jugum is mentioned, are horizontal: but we cannot accept his view about the meaning of jugum as proved or even likely: if it were, Ovid's “tela jugo vincta est” (see above) must refer to binding the licia on to this beam: but to use tela for licia would be strange, when we compare “addere licia telae.” Then again it is said that the mention of pecten implies the horizontal loom because Hesychius says, s. v. σπαθητόν, τὸ ὀρθὸν ὕφος σπάθῃ κεκρουμένον οὐ κτενί: but this refers to the tunica recta, or regilla (see below), woven from the top downwards (the weft being driven upwards) in the fashion of the time when the spatha only was used; weaving from. the bottom upwards, front which it is distinguished, belonged equally to the upright loom, and in the Epithalamium Laurentii et Mariae (Poet. Lat. Min. 3.295, Bahrens) the tela is suspensa, but still the pecten is used.

We may remark also that not only does no Greek or Latin writer mention any difference of weaving beyond this upwards and downwards weaving and the positions of sitting and standing, [p. 2.769]but of the few representations of weaving which are given in ancient art, none show anything but the upright loom; and lastly, the following passage from Theophylact, archbishop of Bulgaria A.D. 1070, shows that he knew nothing but the two kinds of upright looms: ἄλλοι δέ φασιν ὅτι ἐν Παλαιστίνῃ ὑφαίνουσι τοὺς ἱστοὺς οὐχ ὡς παῤ ἡμῖν, ὄντων ἄνω μὲν τῶν μίτων καὶ τοῦ στήμονος, κάτω δὲ ὑφαινομένου τοῦ πανίου καὶ οὕτως ἀναβαίνοντος, ἀλλὰ τοὐναντίον κάτω μέν εἰσιν οἱ μίτοι ἄνω δὲ ὑφαίνεται τὸ ὕφασμα (ad Joann. xviii. p. 825).

The changes in ancient Greek and Roman looms from the earliest to the latest period of literature we believe to have been as follows. The earliest loom (the Homeric loom and the

Fig. 4.

early Roman loom, the tela jugalis of Cato) resembled the Icelandic loom (fig. 1) except that it was a simpler framework without the yarnbeam, and so far like the representation of Circe's loom taken from the Vatican Aeneid (fig. 4). But this representation is an anachronism in making the web begin at the bottom. The author of this ancient picture (whom it is of course absurd to make an authority as to the Homeric loom) has in this point adhered to the fashion of his own day: but in the simpler frame he has probably come near to the primitive pattern. Homer's loom, however, had leashes, which are not given here, besides the simple κανών. The essential distinction between the early Greek and Roman looms and the later was that pointed out by Herodotus (2.35), that the web began at the top, and therefore the weaver always thrust the weft upwards (ἄνω τὴν κροκὴν ὠθοῦσι) in striking it close with the σπάθη. The tunica recta or regilla, enjoined with the conservatism of religion for the marriage garment, was woven at this ancient loom ( “sursum versum,” “in altitudinem,” Isid. Orig. 19.22; Fest. p. 277, 8: the words

Fig. 5. Loom from an Egyptian painting.

=the ἄνω ὠθοῦσι of Herodotus): it was woven also in one piece of the size of the loom frame, as there was no rolling up of the cloth-beam or unrolling of the yarn-beam. At such a loom also, as was said above, the weaver stood, and possibly, in the lack of well-arranged leashes and heddles, had to walk round the loom for the adjustment of the warp threads.

At a later time, probably quite at the end of the Republic, the Egyptian fashion (Herod. l.c.), of beginning the web at the bottom and so weaving in a sitting posture, was introduced. The cut (fig. 5) of an Egyptian weaver from a wall-painting (Wilkinson, Ancient Egyptians, vol. 2.170) illustrates this kind of loom. We think that Wilkinson is right in considering the painting, which Rich (s. v. subtemen) gives as an instance of Egyptian horizontal weaving, to be not weaving at all, but the plaiting of mats. An even better illustration of the Egyptian loom as adopted by the later Romans and Greeks is afforded by a sketch of the modern Lycian weaving (fig. 6), which we have taken from Benndorff (Reise). The weaver is using the comb described on page 768 a;

Fig. 6. Weaver in modern Lycia. (Benndorff.)

We have little doubt that this faithfully reproduces in its form the Roman loom which is characterised as the later kind, though the arrangements of leashes, &c., may often have been more elaborate: this pattern may well have been introduced into Asia Minor at som date later than the time of Herodotus, and have lingered there since. With this Egyptian form came in the other improvements described in I. and IV., the substitution of the pecten for the spatha, and the discontinuance of the weights (ἀμνῦθες, λεῖαι, pondera). It is needless to say that the changes were not made all at once all over the Roman Empire: the older form no doubt lingered in many places, particularly in the more remote countries. Hence the stone weights found in Germany and elsewhere may well belong to a date when at Rome itself the later form of loom prevailed and weights were no longer of any use. Rejecting, as we feel compelled to do, the idea of a horizontal loom, [p. 2.770]we believe that no further change in the loom took place except the development of dexterity in its manipulation.

VI. After enumerating those parts of the loom which were necessary to produce even the plainest piece of cloth, it remains to describe the methods of producing its varieties, and more especially of adding to its value by making it either warmer and softer, or more rich and ornamental. If the object was to produce a checked pattern (scutulis dividere, Plin. Nat. 8.196; Juv. 2.97), or to weave what we should call a Scotch plaid (and it is worthy of notice that Pliny attributes this pattern to a Celtic people), the threads of the warp were arranged alternately black and white, or of different colours in a certain series according to the pattern which was to be exhibited. On the other hand, a striped pattern (ῥαβδωτός, Diod. 5.30; virgata sagula, Verg. A. 8.660) was produced by using a warp of one colour only, but changing at regular intervals the colour of the weft. Of this kind of cloth the Roman trabea (Verg. A. 7.188) was an example. [TOGA] Checked and striped goods were no doubt, in the first instance, produced by combining the natural varieties of wool, white, black, brown, &c. [PALLIUM]. The weft also was the medium through which almost every other diversity of appearance and quality was effected. The warp as mentioned above was generally more twisted, and consequently stronger and firmer than the weft: and with a view to the same object different kinds of wool were spun for the warp and for the weft. The consequence was, that after the piece was woven, the fuller drew out its nap by carding, so as to make it like a soft blanket (Plato, Polit. p. 302) [FULLO]; and, as stated above, when the intention was to guard against the cold, the warp was diminished and the weft or nap (κρόξ, κρόκυς) made more abundant in proportion (Hesiod, Op. 537; Proclus ad loc.). In this manner they made the soft χλαῖνα or LAENA [PALLIUM]. On the other hand a weft of finely twisted thread (ἤτριον） produced a thin kind of cloth, which resembled our buntine ( “lacernae nimia subteminum tenuitate perflabiles,” Amm. Marcell. 14.6). Where any kind of cloth was enriched by the admixture of different materials, the richer and more beautiful substance always formed part of the weft. Thus the vestis subserica, or tramoserica, had the weft of silk [SERICUM]. In other cases it was of gold (Verg. A. 3.483; Servius in loc.)--the invention of Attalus, according to Plin. Nat. 8.196, and thence called vestes Attalicae, but it was probably older in the East and got its name because Attalus prized it; of wool dyed with Tyrian purple (Ovid, Ov. Met. 6.578; Tyrio subtegmine, Tib. 4.1, 122; picto subtegmine, V. Fl. 6.228); or of beavers'-wool (vestis fibrina, Isid. Orig. 19.22). Hence the epithets φοινικόνροκος, “having a purple weft” (Pind. O. 6.39), ἀνθοκρόκος, “producing a flowery weft” (Eur. Hec. 470), χρυσεοπηνήτος, “made from bobbins or pens of gold thread” (Eur. Orest. 841), εὔπηνος, “made with good bobbins” (Eurip. Iph. in Taur. 1465), κερκίδι ποικιλλοῦσα, “variegating with the shuttle” (Eurip. Iph. in Taur. 223), &c.

But besides the variety of materials constituting the weft, an endless diversity was effected by the manner of inserting them into the warp. The terms bilix and δίμιτος, the origin of which has been explained, probably denoted what we call dimity or twilled cloth, and the Germans Zwillich, where by missing over a certain number of warp-threads a ridged pattern is produced. The poets apply trilix, which in German has become Drillich, to a kind of armour, perhaps chain-mail, no doubt resembling the pattern of cloth which was denoted by the same term (Verg. A. 3.467, 5.259, 7.639, 12.375; Val. Flaccus, 3.199) [LORICA p. 81]. All kinds of damask were produced by a very complicated apparatus of the same kind (plurimis liciis), and were therefore called Polymita (Plin. Nat. 8.196; Mart. 14.150), for which multicia (Juv. 2.66) is probably, as Blümner thinks, an equivalent (cf. Gloss. Philox. s. v.).

The sprigs or other ornaments produced in the texture at regular intervals were called flowers (ἄνθη, Philostr. Imag. 2.28; θρόνα, Hom. Il. 22.440) or feathers (plumae). Another term, adopted with reference to the same machinery, was ἐξάμιτον, denoting velvet. In the Middle Ages it became ξάμιτον, and thus produced the German sammet, our samite.

As far as we can form a judgment from the language and descriptions of ancient authors, the productions of the loom appear to have fallen in ancient times very little, if at all, below the beauty and variety of the damasks, shawls, and tapestry of the present age. In addition to the notices of particular works of this class, contained in the passages and articles which have been already referred to, the following authors may be consulted for accounts of some of the finest specimens of weaving: Euripid. Ion, 190-202, 1141-1165; Aristot. Mir. Auscult. 96, = p. 838; Athen. 12.541; Verg. A. 5.250-257, Cir. 21-35; Ovid, Ov. Met. 6.61-128; Stat. Theb. 6.64, 540-547; Auson. Epiq. 26; Lamprid. Heliog. 28; Claudian, in Stilich. 2.330-365.

Although weaving was amongst the Greeks and Romans a distinct trade carried on by a separate class of persons (ὑφάνται, textores and textrices, linteones; cf. even in the Homeric age the γυνὴ χερνῆτις, Il. 12.433), who more particularly supplied the inhabitants of the towns with the productions of their skill (Cato, Cat. Agr. 135; Plat. Phaed. p. 87 B, Rep. ii. p. 370 D; Paus. 7.21), yet every considerable domestic establishment, especially in the country, contained a loom (Cato, Cat. Agr. 10, 14), together with the whole apparatus necessary for the working of wool (lanificium, ταλασία, ταλασιουργία). (Hesiod, Op. 779; Verg. G. 1.285, 294.) [CALATHUS] If in the more luxurious age the most ornamental work was purchased, the slave household (familia rustica) at least was thus clothed, and the commoner stragula were made (Dig. 33, 7, 12, 5; Paul. Sent. 3, 6, 37). In Greece as at Rome in earlier times the matron and her daughters, assisted by female slaves, wove garments for husband, sons and brothers (Plat. Legg. vii. p. 805 E; Aesch. Cho. 231; Eur. Ion 1417): so of the Roman matron weaving in the atrium, Liv. 1.57; Ascon. in Milon. p. 43, and even in later times, Arnob. 2.67; C. I. L. 6.1527, 11602. [p. 2.771]

When the farm or the palace was sufficiently large to admit of it, a portion of it called the ἱστῶν (histones, Varro, R. R. 1.2), textrina or textrinum, was devoted to this purpose (Cic. Ver. 4.26, 58, 59; Isid. Or. 14.8; cf. Hor. Od. 2.18, 6). The work was there principally carried on by female slaves (quasillariae, αἳ ἔριθοι, Theoc. 15.80; Hom. Od. 7.235, 21.350; C. I. L. 6.6639-6646) under the superintendence of the mistress of the house, who herself also together with her daughters took part in the labour, both by instructing beginners and by finishing the more tasteful and ornamental parts (Vitr. 6.7, p. 164; Symmachus, Epist. 6.40). But although weaving was employed in providing the ordinary articles of clothing among the Greeks and Romans from the earliest times, yet as an inventive and decorative art, subservient to luxury and refinement, it was almost entirely Oriental. Persia, Babylonia, Egypt, Phoenicia, Phrygia, and Lydia, are all celebrated for the wonderful skill and magnificence displayed in the manufacture of scarfs, shawls, carpets, and tapestry. [CHLAMYS; PALLIUM; TAPES.]

For the weaving of sacred robes in Greek temples, see ARRHEPHORIA, HERAEA, PANATHENAEA; and cf. Paus. 3.16, 2. (On the construction of the loom, see also Blümner, Technologie, i. pp. 120-157; Marquardt, Privatleben, 519-527; Ahrens, in Philolog. 35.385 ff.)

[J.Y] [G.E.M]