Craft, Handloom, Art, Micro-history, Art History, Monographs, Research

Cotton: A Probe into Archaeological Evidences

Gupta, Dr. Charu Smita, Pathak, Anamika


The cotton fibres (Pl. 1) belong to the genus Gossypium of family Malvaceae. About 39 species are known worldwide, native to the tropics and warm temperate region (http./ Some scholars have discussed cotton varieties in detail (Watt 1966: 575-589). Among all, only four popular varieties have been used for domestication. These are Gossypium arboreum, Gossypium herbaceum, Gossypium hirsutum and Gossypium barbadense. First two varieties belong to old World and are usually known as desi cotton, while other two are of new World. (Cotton committee report, 1963:1)


Some scholars are of the opinion that Gossypium herbaceum is a wild ancestor to old World cottons, which is endemic to South Africa (Phillips 1976: 196), whether this is a place of origin for all old World cottons is contested (Santhanam and Hutchinson 1974: 90). Few other scholars think that Gossypium arboreum is a domestic form of cotton, which is widely cultivated in Indian sub-continent region from very early period (Possehl 1999: 249). Physical characteristic of the Gossypium herbaceumcotton is generally much longer in staple. On the other hand Gossypium arboreum variety of cotton is predominantly coarse short stapled; though a few among them are medium to long stapled and fairly fine. Some scientists feel there are weedy, ‘feral’ varieties of Gosssypium arboreum available in the fields of India and Pakistan. These have a range of variations, but it cannot be shown that, are truly wild progenitors of desi cotton of the region today. At present wild species of cotton in Pakistan called Gossypium stocksii yields short, useful fibres. G. Willcox is of the opinion that this could have been the cotton used during the Indus period (Willcox 199293). Traditional cotton growing areas are located in Saurashtra especially the region of Sindh and Gujarat (Hutchinson 1976:136).


Antiquity of cotton can be traced from early Chalcolithic period (ca.5,500-4,300 B.C.) from the entire Indian sub-continent region. There are some direct and indirect sources, which establish the fact that cotton is the biggest contribution of India to the world.
Discovery of cotton seeds from excavated sites of early Chalcolithic period, early to mature Harappan culture (ca. 2600-2500 B.C.) and late Harappan culture (ca. 2000-1200 B.C.) is the most noteworthy find. Traces of cotton fibres and textile fragments provide evidences of production of cotton textiles from early Harappan period onwards. Apart from these direct evidences, there are some indirect references of textile such as impressions on potsherd, which have been found from sites of Harappan, late Harappan, Chalcolithic, Iron Age and Sunga- Kushan periods.
Earliest Indian historical cotton textiles have been found from Fustat, Quseir-al-Qadim and Qasr Ibrim and Gebel Adda in Nubia, Egypt (Guy 1998: 39). These thousands of fragmentary pieces are dated from ninth century A.D. to seventeenth century A.D., which are housed in different collections of the World. Research Laboratory for Archaeology and the History of Art, Oxford had done the analysis of these objects with radiocarbon dating (Guy 1998: 186).
Find of actual cotton fragment from ninth century A.D., examples of cotton seed and weave impression on pottery from Chalcolithic period onwards indicate towards few important points like production of cotton, loom weaving, production for domestic market etc. References of cotton fibres and cotton garments in Greek indicate that cotton was probably exported to Greece from India during that period. The microscopic and scientific examination of few textile fragments and its impression on potsherd reveals that the people of the Harrapan and the Late Harappan periods had knowledge of ‘plain weave technique’, which is considered the foremost weaving technique. Later on several complex and advance weaving technique were developed. Cotton seems to have been the most preferred material used during that period and still remains so.


Charred cotton seeds have been found from three sites belonging to the early Chalcolithic and mature Harappan periods. The earliest evidence of several hundred charred cotton seeds, which were associated with a fireplace in one of the compartmented buildings, was found from Mehargarh, an early Chalcolithic site in Pakistan (Possehl 1999: 251; Jarrige, Meadow and Quirron 1995: 248). The next important finding of cotton seeds is from an inter-phase period between early and mature Harappan (ca. 2600-2500 B.C) site of Kunal (Hissar, Harayana). This site provides two types of cotton seeds; two carbonized cotton seeds (Gossypium arboreum L/ herbaceum L) (Saraswat 2003: 112) and two seeds of semal (Bombax Ceiba L) (Saraswat 2003: 119). Some cotton seeds were also found from Hulas (Uttar Pradesh), a site dated in the mature phase of Harappan culture (ca. 2,000-1,200 B.C) (IAR 1986-87: 132).

Several cotton threads, fibres and textile fragments have been found at Mohenjodaro and Harappa (now in Pakistan) from Harappan period (c.2700-2000 B.C), at Sringaverapur (District Allahabad, Uttar Pradesh) from levels corresponding to Iron Age (c.1050-1000 B.C), and at Narhan (District-Gorakhpur, Uttar Pradesh) from Sunga-Kushan period (ca.200 B.C- 400 A.D). Most of these evidences are in the form of thread found around the handle of copper mirror, razor, knife, rod and blade. These articles were found wrapped with the cotton fabric probably with the idea of preserving their edges from accidental damage. Similar kind of practice is prevalent even today, when not only are weapons covered with cloth sheath but also silver ornaments and metal zari (where metal thread is twisted and winded on silken thread) brocade sarees are wrapped into white muslin cloth to prevent tarnishing.
Among variety of cotton fibres evidences, only four such examples have been scientifically examined so for. Two of them are from Mohenjodaro and one each from Sringaverapur and Narhan. Mohenjodaro specimens have been found adhering to various copper objects and the later two evidences are found from pottery pieces. Mohenjodaro fibres were preserved by the metallic salts, which were created by contact of metal with alkalis in the damp soil of the Mohenjodaro site. Amarnath Gulati of Cotton Technological Laboratory, Bombay examined these samples and their summary reports have been published by Mackey (Mackay 1998: 591-594), the extracts of which are noted below:

  1. Copper broken knife (DK 83761) reveals four bits of string adhering to one side and on other side seventeen grooves in the patina possibly present the position of string that has disappeared. Microscopic examination of the fibres revealed, ‘the convoluted structure charactenrstic of cotton. The ribbon-width commonly varied between 15 and 20. The string appeared to be four fold; the count value of each strand would correspondingly be 8s”.
  2. The blade of copper razor (DK 11985) appears to have been wrapped up in a coarse fabric. ‘The fabric could be distinguished in a few places only and appeared as an impression of fabric elsewhere or was hidden under a coating of clay. The weave of this fabric was plain, i.e. with an almost equal number of picks and ends per unit length. They were also counted with the help of a micro-linen tester. The mean of these observations gives 44 picks and 43 ends per inch. The diameter of yarn varies between 1/50″ and 1/80″ of an inch. It has already been shown that with ordinary twists and counts lying between 20’s and 40’s, the diameter of yarn (d) and its counts (c) are connected by the formula d=l/c Applying this formula, we find that the counts of the yarns used in the manufacture of the fabric under examination lie between 13’s and 17’s. Taking 15’s as the average value. The weight of the fabric would be 4ozs. per square yard. Few small pieces of yarn were scraped off and microscopic examination was also done The majority appeared as transparent crystalline needles, but a single fibre was ultimately found which was definitely convoluted. None of the other fibres examined exhibited the characteristics of fibres of flax. hemp etc, It is, therefore, to be concluded that the material used in the manufacture of this fabric was cotton”
  • Cotton fibres from a small pit of pottery piece have been found from Sringaverapura (c.l050-1000 B.C) (Saraswat 81: 79-81). Under a steriobinocular it was revealed that, ‘these are light to deep brown coloured fibres and coarse in nature. They were attached to the small pieces of carbonised seed coat. The long and short fibres on seed coat surface and the flat, long fibres with convolutions are the diagnostic features for the identification of true cottons of commerce i.e. the cultivated specimen of Gossypium suitable for spinning and twisting into yam (Carpenter and Leney 1952; Purseglone 1974). The ancient cotton fibres from Sringaverapura reveal all the features of cultivated cotton, quite suitable for the textiles and it falls under the category of old World cottons and has been placed under Gossppium herbaceum-arboreum ”.
  1. Last specimen in this group is a potsherd with an impression of textile and a few pieces of fibre, found at Narhan, a site dated in the Sunga-Kushan period (ca. 300 B.C-400 A.D) (Saraswat etal. 1994. 319). On microscopic analysis of mud from selected areas of impression of textile on potsherd, a few pieces of fibre were found. These fibres were exceedingly tender, flat, having a wide lumen and convolutions throughout their length. In the light of these diagnostic features, the fibre has been identified as one of cultivated species of cotton (Gossypium), suitable for spinning and turning into yam. The cotton from Narhan may safely be refered to the category of Gossypium arboreum or Gossypium herbaceum.
  2. Two silver vases, one somewhat larger than the other and closed with its lid of the same material have been found from Mohenjodaro. Both the vases were wrapped in cotton cloth.

These direct references of cotton fibres and its variety reflects two important facts; occurrence of cultivated cotton variety from the Harappan period onwards and secondly, after scientific examination, it becomes clear that spin thread was used, which proves the development of technology. Finding of spinning tools (will be discussed further) from a number of Harappan sites also support the view that from the Harappan period onwards spinning was practiced. However spinning tools were found from various sites of the Neolithic period of northern Indian region.


Apart from these fibre and fragment examples, several woven fabric impressions on faience potsherd and pottery have been found from most of the Harappan, late Harappan, Chalcolithic, Iron Age, Sunga, Kushan and later historical sites of India. Few important evidences are:

  1. From Lothal, Gujarat (ca. 2,200-1,700 B.C), around 65 terracotta sealings were recovered from warehouse, which have the impressions of Indus seals on obverse, and some impressions of packing materials such as bamboo matting, reed, woven cloth and cord on the reverse (Ghosh1989: 259).
  2. Impressions of cloth fragments on potsherds have been found from the inter phase of between early and mature Harappan (ca. 2600-2500 B.C) site of Kunal, Hissar, Haryana (Saraswat 2002-2003:105)
  • Recent excavations at Harappa, have revealed numerous examples of textile impression on the interior of faience vessels and inadvertently pressed onto pottery containers. The uniform thickness of threads in a single piece of fabric and the tight weave reflected by these impressions indicate the use of spinning wheels. Hand spun thread usually is quite irregular; spinning wheels allow more uniform tension and twisting, which results in standardized threads (Kenoyer 1998:159).
  1. From Harappa, there are several traces of the woven materials and threads- presumably of cotton with which cores of sand were secured for moulding on the faience vases (Vats 1997: 466).
  2. A temple ornament bears an impression of cloth on its underside, while imprints of woven materials are found on two bricks (Vats 1997: 466).
  3. A pottery fragment of Harappa and Bara Culture having textile impression has been found from Alamgirpur, Uttar Pradesh. The clear depiction of warp and weft on the fragment indicate fine example of ‘Plain weave’ (Tandon 1967-68:54-60 Ghosh 1989: 12).
  • A copper vase has been found from Rakhigarhi (Hissar, Haryana) site belonging to the mature Harappan period. This vase, sealed by a miniature pot buried under a mud floor, contained a hoard of jewellery and a bunch of rings clustered together, bearing cloth impression. This vase is displayed in the Harappan gallery of the National Museum, New Delhi.
  • Although no direct evidence for the cultivation of cotton is available from Megalithic culture sites (c.700 to 300 B.C), however, its use for cloth may be inferred from the traces of cloth found on some bronze objects at Adichanulllar and in the Nilgiri hills (Thapar 1985: 111).
  1. One specimen of potsherd of burnished black and red ware from T. Narasipur showed a cloth impression on the interor of the base. This specimen belongs to an early stage of the South Indian Iron Age dated c 750 B.C. Another such impression on a sherd from Piklihal showed a cloth impression on the interior of the base of the bowl. This specimen is slightly late; it belongs to the final centuries of the Iron Age, or opening of the subsequent early historic period, probably of third century B.C. Miss G M. Crowfoot has examined both these specimens. She reports that the, ‘T. Narasipur specimen has an area of c 2.3 x 3.5 cm. The impression is on the inner surface of the sherd and is clearest near the wall of the pot, where presumably pressure would have been greatest. It is a textile impression. The fabric must have been a plain (tabby) weave, count 20/10 threads per cm the latter are clearly 5 on 5mm. The impression is not deep enough to show spinning direction. The Piklihal specimen has an area of c. 5.0 x 3.0 cm .It is a similar impression of a cloth with plain (tabby) weave, with as far as can be seen in the better preserved parts, a similar cotton or flax fabric in the Karnataka region during the Iron Age (Allchin 1997:428-29)
  2. Cotton impression was also found on pottery piece at Ropar (c.500 B.C) in Punjab (Ghosh 1989: 377).
  3. A potsherd with textile impression has been found from excavation at Narhan. This piece showed a good example of ‘plain weave technique’, which belongs to the Kushan period (c. 300-400 A.D) (Saraswat etal.1994: 319, Pl.1)

On the basis of above evidences and references of textiles from Pre-Harappan period to Kushan period the continuity of cotton weaving in Indian sub-continent is conclusively proved.

Process/ Technique
These indirect woven impressions also reveal one of the most important facts about the textile technology that technique of ‘plain weaving’ was consistently prevalent from Harppan period onwards. Practically for any kind of weaving, three basic things are required; yarn of two kinds (warp and weft), loom and the technique of weaving. So far as the first is concerned, cotton yarns have been found from some of the excavated sites. The loom was made of wood from ancient times till date. Since wood is perishable, it is impossible to get traces of wooden loom from any excavated site. However, after analyzing the excavated fabric, it has been already established that these fabrics are made with plain weaving technique for which looms are required. Moreover, excavation finding of loom weight, bobbin and point also support the view of use of loom during the Harappan period. Third is the weaving technique, which was practiced by the Harappan people. From cultivation of cotton, cleaning, combing, ginning, carding, twisting, silvering, preparation of warp and weft yarn to weaving and making of the final product is a lengthy process. Some of the processes are done just by hand, while some are done with implements. Entire process can broadly be divided into following categories;

  1. Pre-spinning: Fibre to yarn includes cleaning, combing, ginning, carding and silvering.
  2. Spinning implements: spindle and spinning wheel.
  3. Pre-loom preparation includes yarn reeling, warp preparation, making of weft
  4. Loom weaving: Yarn to fabric is done on different type of looms and with various weaving techniques.
  5. Ornamentation of fabric through three types;Pre-loom (Colour dyeing of yarn)Loom (Needle and Points)Post-loom (Dyeing, painting, printing, embroidery, and applique).

Pre-spinning: Fibre to yarn
The main process of obtaining cotton starts right from the cultivation of cotton balls, which are hand picked. The cotton is generally spread out in the sun for three to four hours. Once it is dry, some of the impurities are removed. The cotton fleece is combed with the ‘fish jaw’ manually; so that fibres on the seeds are made parallel as well as weaker fibres are removed. The ‘fish jaw’ is tied to a stick so that it can conveniently comb the cotton, which is done at least four to five times to remove some impurity. Generally it has been observed that yarns, which are processed in this way, remain stronger. Tradition of combing the cotton fibres with jaw of the Valugu fish (Pl. 2) is still practiced in Ponduru village of Andhra Pradesh, South India (Menon 2003: 4).





Once preliminary cleaning of cotton (Pl.3) fibres is done, comes the stage of ginning, in which ‘cotton is separated am cotton seeds’ with the help of ginning implement. Like most of the Asian countries, in India also, process of ginning is done with two type of implements known as; ‘board and roller’ (Prattipita Dukka & Gunupu in Telegu; Salai patri in Hindi) and ‘cotton-gin’ (Chakri in Hindi) (Alam 1990: 94). Both these basic implements have ancient origin and their tradition is still continuing. The ‘board and roller’ ginning implement was present in nearly all the ancient civilizations. The earliest example of this implement is evident from Ajanta frescoes titled, as Mahajanka Jataka (Mid sixth century A.D), which portrayed ‘three women at work’ (Yazdani 1930-55: 17). It depicts one of the ladies busy in ginning the cotton with roller and board type of implement. Similar method is well represented in fifteenth century manuscript (Steingass 1981: 403) These early examples are clear depiction of ginning implement and its process of using it. However, what material had been used for making this basic implement is not clear from these early examples. In later period such implements were made of wood, iron or other material. One such example is presently housed in National Gandhi Museum of New Delhi. This twentieth century implement ‘board’ is made of wood and its ‘roller’ stick is of aluminium. This type of traditional ginning implement is still practiced by Pattushali, Sali and Devangi communities of Ponduru town, Srikakulam, Andhra Pradesh, South India. (Pl 4)


While examining the archaeological material one interesting reference has been noticed from Bara (District Rupnagar, formerly in district Ambala, Ropar) of Harappan culture as described by Y.D. Sharma. It is an unidentified terracotta object, 10 cm square from the base and 17.5 cm high and has tapering outline, with two bifurcated curved upper ends; it has three-finger depression on the front and back. Such objects are reported from Sanghol and Chandigarh (Sharma 1989: 52) also. These could be used as ‘board and roller’ implement of ginning implement.

Method of using the implement
The ‘board and roller’ or ‘salai patri’ implement consists of a slanting wooden plank having horizontal grooves in it and a wooden or metal stick. The combed cotton with seeds is spread on the wooden plank, pressed and rolled with the metal rod towards forward motion with the help of slight pressure of palms. In this process the seeds are separated from the fibres. This sounds like a simple process but it calls for a skilled hand (Rao 1982: 12).
Cotton-gin’ or ‘Charkhi’ is another variety of ginning implement, which was also prevalent probably from mid-sixth century A.D. The earliest evidence of charkhiimplement is depicted in one of the fresco paintings of Ajanta, titled, ‘Dancing Girls’, which depicts three women at work (Yazdani 1930-55: 17). One of these women is shown working with charkhi, with its characteristic rectangular frame. She is feeding cotton between two rollers and probably moving the rollers with the other hand. The charkhi in the painting clearly shows both the upper rollers although these are cylindrical in shape. Here the charkhi is without crank-handle, which appears to be a later attachment to the instrument. It is not certain when the charkhi obtained its crank-handle. Bahar-1-Ajam (1740) manuscript talks about the weaving and its technique, but it does not speak about how it was used. There is a mid-18 century miniature painting of Kangra, Himachal Pradesh housed in the collection of National Museum, New Delhi, which shows the use of implement by the people. This painting titled, ‘migration of villagers from Gokula to Vrindavan’ belongs to Bhagavata Purana series (Randhawa 1973: 50). It shows one lady carrying charkhi, the ginning implement, and another one carrying the charkha. Both the implements appear to be made of wood. Charkhi in this painting has two rollers and crank-handle, which is fixed in ‘T’ shaped frame. Cotton ginning and spinning appears to be cottage craft till eighteenth century, mainly done by women only, as in this painting also both these implements are carried by ladies only. Three similar kinds of charkhries are housed in Himalyan Art Museum, Mandi, Himachal Pradesh, Dogra Musuem, Jammu and National Gandhi Museum, New Delhi, which belongs to last century. All three charkhis are made of wood and iron and have rollers, crank-handle, which are fixed in a frame. In Himalayan Art museum charkhi has twisted grooved angular rod in between the rollers, so that seeds may pass smoothly.
Although no such ginning implement has been found from any Harappan site, but if excavated materials were examined in the light of shape of charkhi and its parts, it may be possible that some missing link can be traced from unidentified objects reported by archaeologist from all the excavated sites. While going through the reports of Lothal, Gujarat, one such object attracted attention. One bronze drill of the auger type with twisted grooves has been found from period B of Lothal of mature Harappan period (Ghosh 1989: 258). This could be a part of ginning implement in which iron twisted grooves rod is used for separating the seed from the seeded cotton, It is quite similar to twisted angular rod of charkhi of Himalayan museum. Moreover it was found from the area which provided accommodation for craftsmen-coppersmith, goldsmith and other crafts.


Charkhi implement is generally made of wood or bamboo. It consists of two vertical wooden posts, fixed to a flat piece of wood and the rollers are geared to move in opposite direction. Later on, sometimes iron auger type rod is also used in between the rollers to expedite the ginning process. The operator sets the rollers in motion by means of a crank attached to the end of one of the rollers, moving the crank with one hand and feeding cotton in the machine with the other. Once the seeds are removed from the cotton, it is cleaned further and sun dried.
Carding is the next stage of processing the cotton fleece, in which fibres need to be separated by two methods; ‘scotching’ with wooden stick and with carding implement known as ‘bowstring device’ (Alam 1990:101). Scotching with stick is one of the most primitive methods of removing dirt and dust from cotton. However, this is a simple method and has quick results, but fibres break easily and become weak and not remain good for spinning. The bowstring device is simpler and more popular and prevalent till last century. Basically it is a kind of small bow consisting of a bamboo stick to which a cane string is attached and can be used by the mallet. The bowstring device is quite ancient. The two Sanskrit dictionaries of 11 and 12 centuries give textual references of bowstring device (Chandra 1973: 126). One Persian manuscript named Miftah-ul-Fuzala of fifteenth century depicts the illustration of bowstring device. Two such implements are also housed in the National Gandhi Museum, New Delhi. These implements are named as ‘Majhlapinjan’ and ‘Yudha pinjan’. Both of these were made of wood, bamboo and jute string and were in use till the start of this century.

Spinning implements

Spinning is the next most important process prior to weaving. This stage requires the maximum amount of skill as fine weaving depends on good quality of spun yarn. Spinning is done with most simple implement, which has not changed much as known so far. ‘Spindle whorl is basic instrument used for spinning, which is in two parts; spindle and whorl (PI.5). Spindle is a kind of stick fixed in the centre of whorl, which is circular in shape. Basically spindle’s weight and size of whorl facilitates the spinning process and acts both as weight and flywheel helping in the motion of the spindle. Other end of the spike has a notch for the purpose of holding the thread. Thus the ‘drop spindle’ is spun with the right hand and the cotton piece is held in the left hand. The strand of cotton, being spun, passes from the left hand to the spindle. The yarn thus produced is wound round the wooden or spike from time to time and thus works as bobbin.

The earliest evidence of terracotta spindle whorl has been found from Late Mesolithic phase. It was found lying with burial of Phase II of Bagor, Bhilwara, Rajasthan (Misra 2002: 120). However, some archaeologists are of the opinion that though it is a late Mesolithic site, some of the material found from here belongs to early Chalcolithic phase, with whom these
habitants might have had contacts.
The stone and pottery spindle whorls of Late Neolithic period have been found from Gufkral, Kashmir, (Sharma, 1982:23). Several terracotta spindle whorls have been found from Harappan period sites namely Mohenjodaro (Mackey 1973:470), Lothal (Ghosh 259), Amra (IAR; 1955-56), Bagasra (Sonawane 2003:21-50), Dhatwa (Mehta and Chaudhary 1975), Kanewal (Mehta, Momin and Shah 1980), Kanmer (L4R 1985-86), Lakhabawal (IAR 1955-56), Nageshwar (Bhan and Kenoyer 1984:115-120), Nesdi/Valabi (Mehta 1984: 243-251), Padri (Shinde 1992: 79-86), Pithad/Jaidak (IAR 1992-92), Vagad (Sonawane and Mehta 1985: 38-44), Valabhipur (Anderson and Afonso 1990) and Rakhigarhi (Nath 1999: 41) sites.
Spindle whorls of different size, design and various materials such as shell, faience and pottery were also found from Mohenjodaro. The pottery whorls are of three types:
(i) With a single hole in the middle. (ii) With two holes in the middle. (iii) With three holes in the middle.
The shell spindle whorls are somewhat rare. They range in size from 1.5 inches in diameter and the single hole in the centre averages 0.8 inch in diameter, which suggests that a metal and not a wooden spindle whorl are all slightly protruding out owing to the curvature of the shell. Spindle whorls of faience have been found from D.K. and SD area of Mohenjodaro. Such spindle whorls are rare and smaller in size. Spindle whorls having a very small size hole suggest that they were fixed to a metal rod. Metal rods were also used in Sumer, as reported on the ‘A’ cemetery at Kish (Mackey 1973: 469).
The medium size spindle whorls having holes have been found from Gufkral (Sharma 1982: 24). The Khairadih (Ballia, Uttar Pradesh) site have rich repertoire of iron objects of the Sunga-Kushan level (2-1“ century B.C to 1-2 century A.D) and from this site also spindle whorl has been found (Tripathi 2001: 143). During exploration of Mahanad, District Hoogly, West Bengal, some very important terracotta weaving implements have been found. These belong to 5-6′ century A.D. (Majumdar 1934-35: 43). One spinning implement made of wood and iron is housed in National Gandhi Museum, New Delhi.
Once the cotton is processed from all these stages it becomes ready for weaving, which is done on loom.

  1. Pre-Loom preparation includes reeling, warping, reed preparation and heddle.
  2. Loom weaving: Yarn to fabric with different types of looms, various weaving techniques and implements

The archaeologists and the textile technologists have analyzed the pre-Harappan excavated cotton fabric and fragments, and opine that the Harrappan people were very familiar with the ‘plain’ or the ‘tabby’ weaving technique. In the plain weaving technique warp and weft threads are almost equal in thickness and closeness and pass over and under each other alternately. Basically the warps are set evenly on the frame and the shuttle carrying crosswise weft goes over one warp and then under. This process is continued until a pattern is woven (Barve 1976: 137). Such a fabric would appear to allow only slight ornamentation, but actually many patterns are possible even with this simple weave, just by reducing or increasing the thickness of the yarn.

The headband in the stone image of bearded priest from Mohenjodaro is a very good example of use of narrow strips of fabric or ribbon, which was prevalent during the Harappan period apart form headgear or cap. Such strips of fabric can be made either through weaving on ‘back strap loom’ or by ‘knotting’ and ‘knitting’ with hands.
Cotton fragment woven with tabby weave have been reported from Chalcolithic phase of Harappan period, which signifies the use of loom and the looms are made of wood. Harappan’s familiarity with timber yielding plants and other types of plants such as ‘rosewood’, ‘deodar’, ‘bamboo’, ‘neem’, ‘pipal’ etc. cannot be ignored (Bhardwaj 1990: 5; Saraswat 2002-2003: 114). Even today most of the looms were made of these woods, as discussed by the weavers of Varanasi and Ahmadabad region.
Terracotta and stone loom weight were found from Mohenjodaro (1973: 461; Mackey 1998.601) and Harappa (Vats 1997:388).
Three bobbins made of steatite (two) and terracotta (one) were found from Harappa (Vats 1997: LXXII, 15, Mackey 1998: 420) (PL6).


Terracotta balls, variously described as marbles, pellets, sling balls have been found from most of the sites of Harappan culture as well as sites of early historical periods. They are solid, handmade, well rounded and baked mostly in oxidizing condition with red colour. Such terracotta balls have been found from Vagad (Sonawane and Mehta 1985:38-44), Zekhda (IAR 1977-78), Lakhabawal (IAR 1955-56), Bokhira (Gaur 2006:33-39), Bagasra/ Gola Dhora (Sonawane 2003: 21-50) Amra (IAR 1955-56), Taxila (Ghosh 1948: 78), Rajghat (Narain 1972: 65) and Narhan, from circa. 800 B. C to 600 B.C to circa 300 A.D to 600 A.D (Singh 1994: 209), Vaisali (Sinha and Roy 1969: 202, pl.XCII) and from Andhra levels at Brahmagiri (Wheeler, No. 4: 270, pl.CXXII). Identical pieces have been reported from Nipania (district Bhairhwa) and Kadzahawa (district Taulihawal) in the Nepalese Terai. Archaeologists feel that such terracotta balls have been described as net sinkers or loom weights (Mitra 1972, Pl. CXXXII 5 & CXXXVI.4). In later period sand bags were also used as loom weight.

Ornamentation of fabric

Pre-loom ornamentation of fabric deals with the dyeing of yarns. Dyeing of yam is done first and with later dyed yarns are woven in geometric, floral and figurative patterns. This technique is known as
patola in Gujarat and internationally ikat. The term ikat is a Malay word, which means tie and dye.
Loom ornamentation is done by needle, which doesn’t have a hole. Basically while weaving, warp has to go through needle and weft has to interlace in it. The first material used for needles was bone as evident from Gufkral site of Kashmir from Neolithic period. Later on needles made of other materials were also found from early Chalocolithic, Harappan, and Iron Age sites. (Pl 7)


Post-loom decoration is fabric dyeing, painting, embroidery and applique work for which various types of implements are required. For dyeing fabric bronze vessel is an essential tool in which dyes are prepared. Needle with a hole is essential for thread embroidery and applique work, however, in later period wooden frame also became necessary for zari -embroidery.
Ornamentation techniques/tools

Starching the yarn as well as fabric is an important stage, which is considered to be apart of ornamentation as it provides extra luster and improves the texture of ‘fabric.

Dye and Dyeing

Evidences of madder dyed cotton fabric, dyeing bath and ochre from Harappan period and seed of Indigo from early Harppan phase at Kunal, Haryana, provides important aspect of colour knowledge and aesthetic sense of Harppan people. However, the concept of colour and its use with proper contrast is evident from the Mesolithic period onwards. Some of the examples are as follows:
From Mohenjodaro the madder dye cotton fabric has been found. The cotton resembles the coarser varieties of present-day Indian cottons, and was produced from a plant closely related to Gossypium arboreum or one of its varieties (Maity and Maity 1996:288).


Twelve carbonized seeds of Indigo (Indigo fera L) have been found from Early Harappan site of Kunal, Hissar, Haryana. Ten seeds are from early Harappan (ca. 2850-2600 B.C) phase, while two are from the inter phase between the early and mature Harappan phase (ca. 2600-2500 B.C.) (Saraswat 2002-2003:115).
Dyeing bath from Harappa depicts the so-called circular working platform area and workmen’s quarters are seen in the mid-ground. New excavations of a circular working platform were begun in 1998 and continue in 1999. Initial results suggest that the platforms were not used for processing of grain, but were more likely associated with processing something using water. One possibility is the preparation of indigo dye or dyeing (www/
Two cakes of red ochre were found from the Great Granary Area of the Harappa site (Vats 1997:258).
Finding of dyeing courtyard from Early Harappan phase of Rakhigarhi, Harayana, depicts the huge burnt brickbat floor sloping southwest. There were four circular pits cut in the floor, two of bigger size were aligned north-south, while the other two of smaller diameter were aligned east-west. There were few postholes around these pits. The sharp gradient in the floor towards the public drain suggests its possible use as textile dyeing courtyard. The circular pits in the floor were specially cut to the size of open mouth receptacles containing dyeing solution. An equally wide veranda was noticed on the northern side of the courtyard, which possibly served the purpose of stacking textile (Nath 1999:47).
All these evidences of red ochre colours, dyeing bath and red madder dyed cotton cloth fragment provide the testimony that dyeing was prevalent from Harappan period onwards.

Block printing and Stamping

The earliest Indian block printed cotton textiles have been found from 11 -12 century A.D from Foustat, housed in museums of France and Germany. These examples establish the fact that tradition of block printing existed in the eleventh century (PI.9). Here it would not be out of context to draw attention towards the findings of the excavations, which the archaeologists had identified as ‘Potter Stamp’, ‘Moulds’ or ‘Dyes’ found in most of the archaeological sites as early as early as first few centuries of the Christian era. Although these terracotta stamps were used for stamping the pottery, one cannot deny the familiarity of ‘concept of stamping’ among the Harappans.


Such terracotta stamps were reported from several historical sites. At Kumrahar such antiquity has been described as a ‘mould or ‘die’ for stamping, which is dated to Period III (circa.100 A.D – 300 A.D) (Altekar and Mishra 1959: 129). From Kaus ambi such antiquities have been found from S.P. VII level datable to circa 200 A.D. The designs comprise radiating lines with horizontal/transverse incision within concentric rings and have beautifully made handles (Singh 1994:191). Such stamps were also found from Rajghat and Hastinapur.


Apart from dyeing work, there are references of stitching, embroidery, knitting, knotting and weaving activity also, for which the basic implement required is the needle. This needle could be with or without hole. Needle having hole is used for stitching and embroidery and without hole is used for knitting and weaving. Several stone, bone, iron and gold needles have been found from Neolithic, Early Chalcolithic, Harappan and Chalcolithic culture and Iron Age sites.
Needles without hole are of two size; large and small.
The earliest bone needles are found from Neolithic period of Gufkral site, Kashmir (Sharma 1982: 21). These needles are of different varieties; with and without eyes. Finds from late Neolithic phase include small fine sized bone needles, indicating refineness in making of this tool. Tradition of the use of bone needle continued in Neolithic Culture of Mid-eastern region. The copper and bone needles, very roughly made and not completely round, with polished tip have been found from Mohenjodaro (now in Pakistan). Copper needles are of 1.95 inches and 1.81 inches long and diameters of 0.5 inches (Mackay 1973; 470).



Few copper or bronze needles have been found from Harappa. One is made of rectangular bar, but its pointed end retains a vestige of the needles. Eye is round in the cross section. The eye being at the point, the stitching was presumably done with two threads. Another is made of a round bar, and shows more of the eye, which is near the point. Metal needles were found Rom Rakhigarhi (Haryana) (Nath 1999:48), faience needles from Dholavira (Gujarat) and gold needles from Mohenjodaro and also Dholavira.


Late Harappan and Painted Gray Ware period of overlap phase revealed bone needles from Bhagwanpura. From Painted Gray Ware Culture (from the beginning to 7-6 century B.C) iron needles have been found from Atranjikera (Uttar Pradesh) and Takalghat-Khapa and Naikund (Tripathi 2001:175).


Awl is another kind of implement used for embroidery, apart from needle. Awl is an important tool for embroidery even today, especially in Kashmir and Gujarat area, where embroidery is done with awl tool and this work is known as mochiembroidery. The earliest example of awl, which is made of copper, has been found from advance Neolithic period from phase II of Bagor (Rajasthan) and from Adamgarh (Madhya Pradesh).
Few copper or bronze awls have also been found from Harappan sites, some have slight inward cut near the point, perhaps for putting the thread. Sometime it appears that it has handle also (Vats 1997: 390).
From Mohenjodaro four awls were reported; two each awl is of copper / bronze and bone. One copper awl has a blue point at one end and a graduate one at the other. Bone awls have a long round in section with one side is slightly thick and good polished point at one end. The roughness of the other end of the bone awl suggests its insertion in a wooden handle (Mackey 1973: 470). Copper or bronze short, broad and straight edged knife, curved knife, razors were also found from Harappa (Vats .:1997 388). One of the possible function of such curved knifes are at the time of carpet weaving. So it may be inferred that the carpet weaving was in practice during the Harappan period. Bone awl has been found from the Megalithic period of Gufkral, Kashmir. (Sharma 2000:110)

Find of cotton seed, cotton fragment, dyed fabric fragment supported by spinning, weaving and ornamenting tools like spindle and spindle whorl, bobbin, terracotta balls, sling balls, weight, needle, awl, terracotta and copper vessel etc from archaeological excavations provides a clear picture of cotton production and existence of cotton weaving tradition during the Harappan civilization. The bearded priest wearing trefoil patterned shawl, mother goddess decorated with ornamental head dress, male wearing conical caps and a seal depicting seven females wearing tunic are few significant examples found from different sites of Harappan period. All these examples give glimpses of different type of textiles manufactured and used by the people of the Harappan era. Their knowledge from production to weaver and their taste for adorning themselves in a befitting manner clearly show that people of the Harrappan era were very skilled and had artistic taste.

In the end we would like to say that since this paper is the first attempt for interpretation of excavated materials particularly from the published reports of the important sites, it may bring forth many stimulating and welcome discussions. We have to collectively build up on our traditional knowledge and establish that the cotton technology not only emerged in India but also continues to remain the prime technology even today.


Alam, I. 1990. Textile tools as depicted in Ajanta and Mughal painting. In History of science and technology in India, Vol. III, ed. G. Kuppuram and K. Kumudamani. Delhi: Sandeep Prakashan.
Allchin, F.R. 1997. Textile Impressions from the South Indian Iron Age. In A Sourcebook of Indian Archeology, Vol. II, ed. D.K. Chakrabarti. Delhi: Munshiram Manoharlal.
Altekar and Mishra. 1959. Report on Kumarahar Excavations 1951-55. Patna: Asian Educational Services.

Anderson, B. and J.C. Afonso.1990. Heras Indological Studies. New Delhi.
Barve, V.R. 1976. Complete Textile Encyclopaedia. Bombay: D.B. Taraporevala & Sons.
Bhardwaj, H.C. 1990. An assessment of science and technology during Harppan phase. In History of Science and Technology in India, Vol. V, ed. G. Kuppuram and K. Kumudamani. Delhi.
Bhan, K.K. and J.M. Kenoyer. 1984. Nageshwar: A mature Harappan shell working site on the Gulf of Kutch. Journal of Oriental Institute 34: 1-2Carpenter, C.H and L. Leney. 1952. 382 Photomicrographs of 91 paper making fibers, Vol XXXV, No. 1 Technical Publication No. 74 of the State University of New York. College of Forestry, Syracuse.
Chandra, M. 1973. Kashmir Shawl, Costume, Textiles, Cosmetics and Coiffure m Ancient and Mediaeval India, ed. S.P. Gupta, V.P. Dwivedi and S.P. Asthana. Delhi. Oriental Publishers.
Committee on Natural Resources Planning Commission. 1963. Central Cotton Committee report. Bombay.
Dutt, M.N., trans. 1998. Manu’s Manusmriti. DeKi: Chaukhamba Sanskrit Pratishthan.
Gaur, A.S., Sundaresh and P.P. Joglekar. 2006 Excavations at Bokhira (Porbandar) on the Saurastra Coast. Man and Environment VIII (1)
Ghosh, A. 1989 An Encyclopedia of Indian Archaeology, Vol. II. Delhi. Munsihiram Manoharlal.
Guy, J 1998.Woven Cargoes. London
Hutchinson, S.J. 1976. India: Local and introduced crops Philosophical Transactions of the Royal Society of London. Bulletin of Biological Sciences.
Jarrige, C., J.F. Jarrige, R.H. Meadow and G. Quirron. 1995. Mehragarh: Field reports 1974-1985, from Neolithic times to the Indus Civilization. Karachi.
Kenoyer, J.M. 1998. Ancient cities of the Indus Valley Civilization. Karachi: Oxford University Press.
Krishna, Kalayan and K. Talwar. 2007. In Adoration of Krishna. Surat: Indica Books & Abhidha Prakashan.
Mackey, E. 1973. Household objects, tools and implements. In Mohenjodaro and the Indus civilization, Vols I, II and III, ed. J. Marshall. Delhi: Indological Books House.
– 1998. Further Excavations at Mohenjodaro. New Delhi: Munshiram Manoharlal.
Maity S. and Sujata Maity. 1996. Textile. In History of Science and Technology in Ancient India, ed. D.P. Chattopadhyay. Calcutta: Firma Klm Pvt. Ltd.
Majumdar N G., Exploration in Bengal. Annual Report of ASI: 1934-35
Mathpal, Y. 1984. Prehistoric Rock Art of Bhimbetka, Central India. Delhi.
Mehta, R.N.and S.N Chaudhary. 1975. Excavation at Dhatwa. Vadodara: M.S. University, Baroda.
Mehta, R.N., K.N Momin, and D.R. Shah. 1980 Excavation at Kanewal, Vadodara. M.S. University Archaeological Series, No.17.
Meta, R.N, 1984. Valabhi: A Station of Harappan Cattle Breeders. In Frontiers of the Indus Civilization, ed. B.B Lal and S.P. Gupta. Delhi
Menon, M. 2003. Caught in a time warp. The Hindu Magazine, June 15
Misra, V.N 2002. The Mesolithic Age in India. In Prehistory Archaeology of South Asia. Ed S. Settar and R. Konsettar. Delhi: Manohar Publishers
Mitra, D. 1972. Excavations at Tilaura-kot and Kodan. Katmandu
Narain, L. A. 1972. A Study in the Techniques of the Neolithic Bone tools making at Chirand and their probable use. Journal of The Bihar Research Society LVIII.
1979. The Neolithic Culture of Eastern India. In Essays in Indian Proto History, ed. D.P; Agrawal and Dilip K Chakrabarti. Delhi; B.R. Publishing Corporation.
Nath, A. 1998. Excavation at Rakhigarhi. Puratattva 28.
1999. Further excavation at Rakhigarhi. Puratattva 29.
Philips, L.L. 1976. Cotton. Evolution of crop plants. Longman.
G.L. 1999. Indus Age – The Beginnings. Delhi: Oxford & IBH Publishing.
Purseglove, J. W. 1974. Tropical Crops – Dicotyledons (Vol.I & II combined). London.
Randhawa, M.S. 1973. Kangra Paintings of the Bhagvata Purana. New Delhi: National Museum.
Rao, K. 1982. Handspun and Hand woven Textiles of Ponduru. New Delhi.
Santhanam, V. and J.B. Hutchinson. 1974. Cotton in World Crops: Diversity and Change in the Indian Subcontinent. Evolutionary Studies. Cambridge University Press.
Saraswat, K. S. 1981. Plant economy in ancient Srinagaverapura. Phase-I (c. 1050-1000 B.C.). Puratattva 12.
Saraswat, K.S., N.K. Sharma, and D.C. Saini. 1994. Plant Economy at Ancient Narahan ca. 1,300 B.C – 300/400 A.D. In: Excavations at Narahan- 1984-89, ed. P. Singh. Varanasi: Banaras Hindu University.
Saraswat, K.S. and Pokharia. 2003. Palaeoethnobotanical Investigations at Early Harappan KunaL Praghara13.
Sharma, A.K. 1982. Excavations at Gufkral. Puratattva 11.
– 2000. Early Man in Jammu, Kashmir & Ladakh. Delhi: Agam Kala Prakashan.
Sharma Y.D. 1989. A Note on Bara. In An Encyclopedia of Indian Archeology,Vols. I and II. Delhi: Munshiram Manoharlal.
Shinde, V. 1992. Excavations at Padri 1990-91: A Preliminary Report. Man and Environment XVII (1).
Singh, K. 1994. Textiles of Ancient India. Varanasi: University Publication.
Sinha, B. P. and S.R. Roy. 1969. Vaisali Excavations (1958-1962). Patna.
Sonawane, V.H. and R.N. Mehta. 1985. Vagad – A Rural Harappan Settlement. Man and Environment IX.
Sonawane, V.H. 2003. Excavations at Bagasra 1996-2003: A Preliminary Report. Man and Environment XXVIII (2).
Steingass, F. 1981. Persian-English Dictionary (2nd ed.). New Delhi.
Tandon, O.P. 1968. Alamgirpur and the Iron Age in India. Puratattva 54-60.
Thapar, B.K. 1985. Recent archeological discoveries in India. Japan: UNESCO.
Tripathi, V. 2001. The Age of Iron in South Asia – Legacy & Tradition. Delhi: Aryan Books International.
Vats, M.S. 1997. Excavations at Harappa, reprint. Delhi: Munshiram Manoharlal.
Watts, G. 1966. The Commercial Products of India. New Delhi: Today & Tomorrow’s printers & publishers.
Wheeler, R.E.M. 1948. Brahmagiri and Chandravalli 1947: Megalithic and other Cultures in the Chitadurig District, Mysore State. Ancient India 4:270.
Willcox, G.H. 1992. Some differences between crops of Near Eastern origin and those from the tropics. South Asian Archaeology. ed. Catherine Jarrage.
Yazdani G., Ajanta. 1930-55. The colour and monochrome reproductions of the Ajanta Frescoes based on photography.London.  (http/www/

First published in Puratattva, No 38, pp136-149

Pictures by Charu Smita Gupta & Anamika Pathak

Your views


Access 70,000+ practitioners in 2100+ crafts across India.


10,000+ listings on arts, crafts, design, heritage, culture etc.


Rich and often unfamiliar vocabulary of crafts and textiles.

SHOP at India inCH

Needs to be written.