BELTS (Mid. Pets, kamar, NPers. kamar-band)
Our knowledge of the early history of garments and accessories, including the belt, is rather restricted. Before the production of metal belts in Urartu we have no extant examples; our only sources are representations in art from two regions in Iran: Susiana, Elam, and adjacent districts (Fārs, Luristan), on one hand, and northeastern Iran (cf. E. Porada in EIr. II, pp. 549-60), on the other. A large proportion of these representations is found on seals and sealings that do not show much detail (P. Amiet, Glyptique susienne, MDAFI 43, 1972; H. Pittman, Art of the Bronze Age, New York, 1984, pp. 52-65).
From this material it seems that the belt underwent a development in Iran similar to that in Mesopotamia (P. Calmeyer, in Reallexikon der Assyriologie III, 1957-71, pp. 689-93 s.v. Gürtel). During the third millennium b.c. garments were frequently made with waistbands of the same material that functioned as belts: They included the well-known short kilt apparently of mesh (“Netzrock”; probably from southeastern Iran: Porada, p. 555, pl. X, 7), the zigzag-patterned long skirt (from Susa: P. Amiet, Elam, Auvers-sur-Oise, 1966, p. 183 fig. 134 A-B), and the full-skirted sheepskin garment (ibid., pp. 190f. fig. 141: the ends of the belt are fastened in a hump at the back).
From the early second millennium there is no evidence for belts. An important change must have occurred in this period, for after about 1500 B.C. (ibid., pp. 248, 308f., figs. 213, 232: from Susa) all Middle Elamite monuments with sufficiently detailed rendering show belts of different textures from those of the garments (ibid., pp. 418-21 figs. 318; Porada, p. 550 fig. 26k: in all these instances the upper and lower garments, as well as the belts, are of different materials). As in Mesopotamia, the otherwise naked bull-man always wears a broad belt (Amiet, pp. 309 fig. 232C, 376 fig. 248, 396f. fig 299; Porada, p. 550 fig. 26k). Such belts often follow the body’s contours closely, suggesting flexible materials: leather or textiles.
Numerous human figures on the so-called “Luristan bronzes” and other objects from the art market have been collected by B. Goldman (“Origin of the Persian Robe,” Iranica Antiqua 4, 1964, pp. 133-52). Most of them must represent inhabitants—and sometimes deities—of southwestern Iran, but the garments are the same as in neighboring Babylonia (Goldman, figs. 1, 6, 14, 23-26), Assyria (ibid., figs. 7-11, 27-28), and Syria (ibid., fig. 2). The varied types of belts were thus probably also derived from those regions.
Neo-Elamite belts are narrower and sometimes decorated with a zigzag or rosettes (Amiet, pp. 540 fig. 413, 566 fig. 431; Porada, pl. XII, 18-19). They also sometimes hold daggers (Amiet, p. 531 fig. 407: the form of the dagger’s hilt is similar to that of later Persian ones). There are also Babylonian parallels for all these features (F. Wetzel et al., Das Babylon der Spätzeit, Berlin, 1957, pls. 43-44: genii and a ziggurat on the belts of two deities), but it is also possible that this new fashion of decorating belts was connected with a sharp change that had taken place in the northwest in the early first millennium b.c.
There, in and around the ancient kingdom of Urartu, bronze was used to cover (and protect?) the inner belt, which was apparently of leather or woven textile; the metal was hammered into a continuous strip of sheet bronze between ten and twenty cm wide and provided with numerous small holes along the edges, so that it could be sewn onto the more flexible inner material. Very often these strips were decorated with ornaments or figures, either repoussé or chased. Two leather belts plated with silver from the Early Dynastic cemetery at Ur (L. Woolley, The Royal Cemetery, Ur Excavations 2, 1934, pp. 51f. fig. 4, pl. 13b, 156f. fig. 35) may have been forerunners of this use of metal sheathing. Similar belts are shown on a silver and bronze fragment and statuettes with belts from Hattuša (R. M. Boehmer, Die Kleinfunde von Boğazköy, Berlin, 1971, nn. 546ff.) and Susa (MDAFP XXV, Paris, 1934, pp. 208f. pl. X 4, 5). The numerous first-millennium examples come from the region between Transcaucasia (J. de Morgan, Mission au Caucase I, Paris, 1889, p. 116 figs. 17-19, 23, 27; idem, La préhistoire orientale III, Paris, 1927, pp. 276f.; R. Virchow, Über die culturgeschichtliche Stellung des Kaukasus, Berlin, 1895, pls. II-III; A. M. Tallgren, Eurasia septentrionalis antiqua 5, Helsinki, 1930, pp. 165f.; B. A. Kuftin, Archaeological Excavations in Trialeti [in Russian] I, 1941, pl. XXV), where several styles of decoration occur, and Assyria (A. H. Layard, Nineveh and Babylon, London, 1853, p. 180), where they also seem to be depicted on the reliefs. Urartu (modern Armenia, eastern Turkey, and Iranian Azerbaijan) must have been the production center of these metal belts. Several styles of decoration can be distinguished: a “court style” (M. N. van Loon, Urartian Art, Leiden, 1966, pp. 121ff. pls. XXXf.), actually the style of the royal fortresses (H.-V. Herrmann, “Frühgriechischer Pferdeschmuck vom Luristantypus,” Jahrbuch des Deutschen Archäologischen Instituts 81, 1966, pp. 107f. nn. 81-83; G. Azarpay, Urartian Art and Artifacts, 1968, pp. 47ff.), one example of which has been excavated at Qūščī near Lake Urmia (R. W. Hamilton, “The Decorated Bronze Strip from Gushchi,” Anatolian Studies 15, 1965, pp. 41ff.: explained as a rim, because of a faulty reconstruction; see P. Calmeyer, Reallexikon der Assyriologie III, s.v. Gūščī); several local styles; and a distinct group of narrow strips with ritual scenes (for a corpus of Urartian belts, see H.-J. Kellner, in Prähistorische Bronzefunde, Mainz, 1988; cf. E. O. Negahban, A Preliminary Report on Marlik Excavation, Tehran, 1964, fig. 55). This last group is related to Scythian art. A prominent example must have belonged to the hoard alleged to have come from Zīvīa (Ziwiyeh) in Iranian Kurdistan; the gold strip of this belt was cut into pieces by plunderers and reconstructed by P. Amandry (“A propos du trésor de Ziwiyé,” Iranica Antiqua 6, 1966, pp. 113ff.). The decoration of these belts consists of prancing animals and fabulous creatures contained within compartments created by a network of vegetal motifs. Amandry has shown that a “sacred tree” originally formed the right end of the strip and, when worn, would have functioned as the central motif in front.
P. R. S. Moorey, who has treated the general development of metal belts (“Some Ancient Metal Belts. Their Antecedents and Relatives,” Iran 5, 1967, pp. 83ff.), has collected the scanty evidence for the western Iranian local styles of decorated belts. One group, with coarse repoussé decoration (pp. 87f.), has Caucasian connections, although the pieces are said to come from Luristan and Ḵorvīn. Another group is related to repoussé reliefs on circular pins from Sorḵ Dom in southern Luristan (pp. 89ff., esp. n. 78); it seems possible that it belongs to the end of the second millennium. A third group is characterized by animals with bodies patterned in geometric designs (p. 90 with nn. 85-86). There is also a strip that is certainly genuine but with a relief of doubtful authenticity (p. 86 fig. 1, pl. lb), as well as another piece, of outstanding workmanship (R. Dussaud, “Ceinture en bronze du Louristan avec scènes de chasse,” Syria 15, 1934, p. 187 fig. 1, pl. XXV), that is related to a 10th/9th-century group of metal vessels from the region between Zālūāb and Sorḵ Dom. None of these objects comes from a controlled excavation.
Under the domination of the early Achaemenids another change in costume seems to have taken place. There are no more metal strips or plaques, either properly excavated or from the art market, except for those found together with objects of undoubtedly Achaemenid inspiration at Pazyryk, far from Iran (G. Azarpay, “Some Classical and Near Eastern Motifs in the Art of Pasyryk,” Artibus Asiae 22, 1959, pp. 313ff.). A belt made of silver-gilt plaques, from the de Walden collection (Moorey, 1967, pp. 91ff. pls. I-II), has been shown by Moorey to be a modern forgery (“"Some Ancient Metal Belts"—A Retraction and a Cautionary Note,” Iran 7, 1969, p. 155). The change is, however, clearly reflected on the monuments, especially the facades of royal tombs (see P. Calmeyer in EIr. II, p. 576 fig. 43) and the gold plaques from the Oxus Treasure (ibid., p. 579 fig. 45): All Iranian peoples represented, with the exception of the Persians, wear basically the same costume of trousers and overcoat, though there are variations in the styles of these garments and especially in those of the headdresses. But all these “many sword-bearing peoples of the East,” as Aeschylus describes them, have their overcoats girdled with a narrow belt from which the typical short sword, the akinakes, is suspended; the belt is fastened by a button at the side (ibid., p. 576 fig. 43; fourth figure from the right on the bottom line). Often one end of the belt is seen hanging down from the waist in front (p. 575 fig. 41). It is apparently this Median, or war, dress of Darius III that is described by Curtius Rufus (3.3.7), who says the belt is “knotted as a woman’s” and “golden,” probably referring to the color, not the material. In contrast to these western, northern, and northeastern Iranian peoples the Persians of Fārs (p. 575 fig. 41: left side) and the Susians, probably Kissians, from Susiana (p. 574 fig. 38) wear not trousers, but wide, flowing robes, perhaps made from single pieces of cloth, girdled with belts of soft material, a style that they seem to have inherited from the Elamites. This Elamite-Persian outfit is worn by the king on formal occasions, as on the Persepolis reliefs and in the gate hall at Susa (p. 575 fig. 40). There and elsewhere the short, broad dagger that belongs to this costume is held by the belt in the same simple way that was depicted in Babylonian, Assyrian, and Sumerian reliefs from the third millennium onward.
Bibliography: Given in the text.
Investigation of representations of belts in Iran between the fall of the Achaemenid dynasty in the 4th century b.c. and the coming of Islam in a.d. the 7th century reveals that they were almost exclusively male accessories. Depictions of females wearing belts are rare, and most are heavily influenced by Hellenistic and Roman styles, with simple, draped garments or robes girt below the breast (Ghirshman, 1956, pl. VII; 1962, pp. 92-93 figs. 103-04, p. 147 fig. 186, p. 193 fig. 235; Peck, pls. IXb-XI).
In the Parthian period two basic types are found: a tied sash and an elaborate wide belt. At Tang-e Sarvak (Tang-i Sarwak) in Elymais a late a.d. 2nd-century rock relief portrays a Parthian ruler and his retinue clad in tunics tied with tasseled sashes, which recalls the Achaemenid fashion for tied belts of cloth or leather (Ghirshman, 1962, p. 55 fig. 68; 1964, p. 84 fig. 109, pp. 157-58 fig. 209, p. 188 fig. 235; Survey of Persian Art IV, pls. 108 A, B; also see i above). At Dura Europos and Hatra, as well as at Palmyra just west of Parthia, 1st- and 2nd-century images of deities are shown in Roman armor bound with the cingulum, or belt of honor, worn by the Roman emperor and his troops (Ghirshman, 1962, p. 71 fig. 84; Colledge, fig. 10a; Downey, p. 203 pl. L/1; Pauly-Wissowa, III/2, p. 2561; Brilliant, figs. 2.63, 4.50, 4.89). The most common belt represented at Palmyra, however, is a sash derived from the cingulum, worn round the hips and tied with a square “herakles” knot, the long ends tucked up over the belt (PLATE II): In 2nd- and 3rd-century funerary sculpture it is worn by figures clad in Parthian dress (Ghirshman, 1962, pp. 78-79 figs. 90-91; Ingholt, 1954, pl. 11; Schlumberger, p. 90). Almost identical sashes, of similar derivation, appear on the stelae of Antiochus I of Commagene (69-34 b.c.) at Nimrud Dağ (Ghirshman, 1962, p. 67 fig. 80; Rosenfield, fig. 154; Houston, figs. 168, 169).
As for the wide belts, bronze buckles with movable tongues securing the girdles were excavated at Palmyra (MacKay, pl. LXIII/3); to the east, a gold-inlaid buckle from the so-called “treasure of Nihavand” (1st-3rd centuries) attests to a taste for elaborate wide belts in Iran (Ghirshman, 1962, p. 100 fig. 112). From approximately the same period in Iran are bronze belt plaques in openwork with representations of human and animal figures and bronze buckles in the shape of confronted animal heads formerly in the Foroughi collection (Ghirshman, 1979, p. 171, pls. I-IV). Geo Widengren distinguishes between the Parthian kamar, a sword or tunic belt, and the hemyān, or sacred girdle worn by priests (Widengren, p. 254). Funerary monuments at Palmyra, on which only members of the priestly caste wear wide, rolled belts, seem to support this differentiation (Ingholt, 1928, pl. IV). The fashion for decorated belts, probably of jointed metal on a leather backing, is known from the bronze statue of a Parthian ruler found at Shami; a similar belt, tied with ribbons, appears at Bīsotūn on a rock relief depicting a Parthian prince (Widengren, p. 252; Fukai, 1960, p. 172; Ghirshman, 1962, p. 53 fig. 66, p. 88 fig. 99). Also dated to a.d. the 2nd century are the elaborate belts girding the hips of ruler statues found at Hatra. The belts seem to be of metal, in openwork (Ghirshman, 1962, p. 94, fig. 105) or ornamented with reliefs of hippocamps and portrait busts of Greco-Roman design (PLATE III; Ghirshman, 1962, p. 89 fig. 100; Rosenfield, fig. 136). The distinctive semicircular clasps, tied with strips of material or leather thongs with dangling ends, suggest that these belts are very closely related to those found on 2nd-century Kushan monuments in India and Afghanistan (Ghirshman, 1962, p. 89 fig. 100; Rosenfield, fig. 136; Kushan belts: Rosenfield, figs. 2, 3, 8, 12, 20, 62a, 67). A precursor of these sumptuous girdles is a gold belt adorned with plaques depicting goddesses riding lions, found at the site of Tillya Tepe in northwestern Afghanistan (1st century b.c. to a.d. 1st century; Sarianidi, pp. 38-40, 37 ill.). Rosenfield called attention to classical imagery in the decorations of a Kushan ruler’s belt, suggesting that it symbolizes the well-being of the kingdom (Rosenfield, p. 183 figs. 3, 3a); perhaps the classical motifs on the Hatra belts have comparable symbolic meanings. This similarity between belts depicted at Hatra and in the distant Kushan empire may be explained either by contact through trade or by a common derivation from a now lost Parthian site (Rosenfield, pp. 170-71).
The most characteristic Sasanian girdle consists of ribbons tied in a bow in front, the ends often pleated horizontally. It is worn by queens and goddesses, as well as by kings and gods, and appears fully developed, without apparent prototypes, as early as the reign of Ardašīr I (a.d. 224-41) on the rock relief at Naqš-e Rostam (Naqsh-i Rustam; Schmidt, pl. 90). The fashion was followed by Šāpūr I (241-72) at Bīšāpūr and Naqš-e Rajab (Naqsh-i Rajab; Ghirshman, 1962, p. 165 fig. 209; Schmidt, pl. 90), by Hormozd I (272-73) at Tang-e Qandīl (Hinz, 1973, pls. 46-48), and by Bahrām II (27693) at Sar Mašhad, Barm-e Delak, and Naqš-e Rostam (Hinz, 1969, pls. 135, 137, 117). Such a ribbon belt also encircles the waist of the goddess Anāhīd in the investiture relief of Narseh (293-302) at Naqš-e Rostam (PLATE IV: Schmidt, pl. 90). The latest representation of the type is on the image of Ohrmazd in the large niche at Ṭāq-e Bostān, which should probably be dated to the reign of Ḵosrow II (591-628). Prudence Oliver Harper has noted that this late appearance of the belt demonstrates the conservative nature of sacred representations (p. 65 n. 126).
The popularity of the ribbon belt was rivaled by that of the girdle tied with a ribbon in such a way that two loops were pulled up through circular clasps (described by Grenet as two cabochon jewels regulating the ribbon length, p. 197 pl. 1) and the ends were left dangling (PLATE IV). This belt can be seen on the reliefs of Hormozd I at Tang-e Qandīl, the investiture relief of Narseh, and the rock carvings of Ardašīr II (379-83) and Šāpūr II (309-79) and III (383-88) at Ṭāq-e Bostān (Hinz, 1973, pls. 47, 48; Ghirshman, 1962, p. 176 fig. 218, p. 190 fig. 223; Fukai, 1972, pls. LXVI, LXVIII). It was on silver “hunting plates,” however, that this ribboned belt with double clasp was most fully delineated, as Harper has shown (pp. 54-87). It appears first on vessels of the late 3rd and 4th centuries (Harper, pls. 10, 12, 15, 16, 23, 28) and continues on examples from the mid-5th to 7th centuries (Harper, pls. 25, 26, 31).
The taste for jeweled belts, already apparent at Parthian Hatra, persisted at the early Sasanian royal court, as is clear from the reliefs of Ardašīr I and Šāpūr I at Dārāb and Naqš-e Rajab (Hinz, 1969, pls. 57, 58; Schmidt, pls. 97A, 101B). On “hunting vessels” of the 5th century and later jeweled belts with square buckles are illustrated (Harper, p. 87 pls. 18, 20, 21), but in the reliefs of the boar hunt at Ṭāq-e Bostān the buckles are hidden under vertical panels of patterned or embroidered cloth (PLATE V; Peck, pl. XV). This style is repeated only once, on a vessel of about the same date in the Hermitage Museum (Harper, pl. 19). Ḵosrow II wears a most elaborate belt in the depiction of his investiture at Ṭāq-e Bostān. Studded with rows of pearls and huge, square gems, it is the perfect visual reflection of the kamar, which Widengren describes as a jewel-encrusted warrior’s belt worn in the Sasanian period. The high priest Kartīr received a kamar together with a kōlāf “hat” from Hormozd I, son of Šāpūr I, as a symbol of his high rank (inscription on the Kaʿba-ye Zardošt at Naqš-e Rostam, line 4, Back, p. 394; cf. Widengren, p. 260).
A new style of belt with lyriform buckle and movable tongue appears first on a “hunting vessel” dated to the reign of Pērōz (457-84) or Kavād I (488-96 and 498-531) in The Metropolitan Museum of Art (Harper, p. 87 pl. 17) and later, at the end of the Sasanian dynasty, on plates in the Bibliothèque Nationale and the Hermitage Museum (Harper, pls. 22, 27). Harper has pointed out that, though the buckle with movable tongue does not appear on Sasanian reliefs before those at Ṭāq-e Bostān, Marshak has traced it to 5th-century Kucha in Central Asia (Harper, p. 65 n. 126). Ghirshman published a gilded iron buckle with movable tongue that he had excavated in level IV at Susa; he dated it to the mid-4th century (1979, pp. 183, 184 fig. 2). This type of buckle is depicted in the reliefs of boar and stag hunts at Ṭāq-e Bostān on belts worn by the king and his male entourage. As Widengren was the first to note, this kind of jeweled leather belt with pendant thongs tipped and adorned with metal plaques and gems is of Central Asian origin (Widengren, pp. 270-72; PLATE VI; Peck, pls. III, IV, VI, IXa, XIII, XVI-XIX). Although often portrayed on Central Asian wall paintings of the 6th-8th centuries as merely an ornamental fashion (von Le Coq, pls. 14, 16, Bäzäklyk; Grünwedel, fig. 56, Ming-Oï, fig. 654, Bäzäklyk, fig, 666, Idyqutšahri), this belt was essentially a nomadic accessory, weapons being suspended front the thongs during battle and the hunt (Bussagli, p. 59, Dandan Öiliq; Belenitsky, pls. 9, 11, 12, Pyandzhikent). Ghirshman observed that the earliest known examples came from 5th- and 6th-century burials of the nomadic Avars in Mongolia and southern Siberia (1953, p. 69; 1963, pp. 305-06, fig. 13); among these tribes two belts were worn, the upper one as a symbol of station, the lower one for attachment of weapons (1963, p. 305). In the sculpture of the royal mounted knight carved in high relief below the investiture scene at Ṭāq-e Bostān, there is a lower belt with long lappets (PLATE VII; Ghirshman, 1962, p. 192 fig. 234). On the other hand, the thonged girdles worn by Ḵosrow II in the hunting scenes seem to denote his high station; they are more elaborate and highly decorated than those of the courtiers. In addition, only the royal belt is used to suspend objects: a piece of cloth attached to a ring and a flint or whetstone (Gropp, p. 276; Fukai, pl. XLVII). Two other belts are depicted in the large ayvān at Ṭāq-e Bostān: a simple ribbon sash, most typical of the Sasanian period, and the nomadic girdle of eastern derivation (Peck, pp. 118, 120, pls. IXa, CIII, XIX).
M. Back, Die sassanidischen Staatsinschriften, Acta Iranica 18, Tehran and Liège, 1978.
A.M. Belenitsky, Monumental’noe iskusstvo Pyandzhikenta, Moscow, 1973.
M. Bieber, Entwicklungsgeschichte der griechischen Tracht, Berlin, 1934.
R. Brilliant, Gesture and Rank in Roman Art. The Use of Gestures to Denote Status in Roman Sculpture and Coinage, Memoirs of the Connecticut Academy of Arts and Sciences 14, New Haven, 1963.
M. Bussagli, Painting of Central Asia, Geneva, 1964.
M. A. R. Colledge, Parthian Art, London, 1977.
S. B. Downey, The Excavations at Dura-Europas. The Stone and Plaster Sculpture, Monumenta Archaeologica V, 1977.
S. Fukai, “The Artifacts of Hatran and Parthian Art,” East and West 11, 1960, pp. 135-81.
Idem and K. Horiuchi, Taq-i-Bustan I-II, Tokyo, 1969, 1972.
R. Ghirshman, “Notes iraniennes V. Scènes de banquet sur l’argenterie sassanide,” Artibus Asiae 16, 1953, pp. 51-71.
Idem, Fouilles de Châpour. Bîchâpour II. Les mosaïques sassanides, Paris, 1956.
Idem, Persian Art. The Parthian and Sassanian Dynasties, New York, 1962.
Idem, “Notes iraniennes XIII. Trois épées sassanides,” Artibus Asiae 26, 1963, pp. 293-311.
Idem, The Arts of Ancient Iran. From Its Origins to the Time of Alexander the Great, New York, 1964.
Idem, “La ceinture en Iran,” Iranica Antiqua 14, 1979, pp. 167-95.
F. Grenet, “Un plat sassanide d’Ardašir II (379-383) au bazar de Kabul,” Studia Iranica 12, 1983, pp. 195-205.
G. Gropp, “Der Gürtel mit Riemenzungen auf den sasanidischen Reliefs in der grossen Grotte des Taq-e Bostan,” AMI, N.S. 3, 1970, pp. 273-88.
A. Grünwedel, Altbuddhistische Kultstätten in Chinesisch-Turkistan, Berlin, 1912.
P. O. Harper, Silver Vessels of the Sassanian Period I: Royal Imagery, New York, 1981.
W. Hinz, Altiranische Funde und Forschungen, Berlin, 1969.
Idem, “Das Sasanidische Felsrelief von Tang-e Qandil,” AMI, N.S. 6, 1973, pp. 201-12.
D. Homès-Fredericq, Hatra et ses sculptures parthes. Etude stylistique et iconographique, Istanbul, 1963.
M. G. Houston, Ancient Egyptian, Mesopotamian and Persian Costume and Decoration, London, 1920.
H. Ingholt, Studier over palmyrensk skulptur, Copenhagen, 1928.
Idem, “Parthian Sculptures from Hatra,” Memoirs of the Connecticut Academy of Arts and Sciences 12, 1954, pp. 5-46.
Idem, Gandharan Art in Pakistan, New York, 1957.
A. von Le Coq, Die buddhistische Spätantike in Mittelasien. Die Wandmalereien, Berlin, 1924.
D. MacKay, “The Jewellery of Palmyra and Its Significance,” Iraq 11, 1949, pp. 160-87.
E. H. Peck, “The Representations of Costumes in the Reliefs of Taq-i-Bustan,” Artibus Asiae 31, 1969, pp. 101-23.
J. M. Rosenfield, The Dynastic Arts of the Kushans, Los Angeles, 1967.
V. Sarianidi, “The Treasure of the Golden Mound,” Archaeology 33/3, 1980, pp. 31-41.
D. Schlumberger, L’Orient hellénisé, Paris, 1969.
F. F. Schmidt, Persepolis III: The Royal Tombs and Other Monuments, Oriental Institute Publications 70, Chicago, 1970.
G. Widengren, “Some Remarks on Riding Costume and Articles of Dress among Iranian People in Antiquity,” Studia Ethnographica Upsaliensia 11, 1956, pp. 228-76.
Originally Published: December 15, 1989
Last Updated: December 15, 1989
This article is available in print.
Vol. IV, Fasc. 2, pp. 130-136
Multiple Authors, “BELTS,” Encyclopaedia Iranica, IV/2, pp. 130-136, available online at http://www.iranicaonline.org/articles/belts (accessed on 30 December 2012).