[Pramāṇa](https://indike.org/pramana/) › [Studies](https://indike.org/pramana/ins) › [Science, technology and education](https://indike.org/pramana/ins.science) › [Minerals and Metals in Ancient India](https://indike.org/pramana/ins.science.biswas-1996) › v1.c19.s5 # Minerals and Metals in Ancient India v1.c19.s5: PGW Culture Minerals and Metals in Ancient India v1.c19.s5 - **Language:** English ## Contents - [v1.c19.s5.t1 Terracotta](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t1) — 2 parts - [v1.c19.s5.t2 Stone Objects](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t2) — 1 part - [v1.c19.s5.t3 Iron Objects](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3) — 15 parts - [v1.c19.s5.t4 Copper Objects](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4) — 8 parts - [v1.c19.s5.t5 Glass Pieces](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5) — 6 parts ### PGW Culture ### [v1.c19.s5.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.1) Of the protohistoric deposits at Atrañjikhera the richest is that of the Painted Grey Ware Culture. The total area under occupation during this period was more than that during the preceding two periods. ### [v1.c19.s5.2](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.2) In comparison to the previous two periods, a larger number of things have been unearthed from this stratum. ### Terracotta ### [v1.c19.s5.t1.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t1.1) Terracotta objects were apparently not popular during this period. Not a single terracotta human figure was found. Even animal figures are rare during this period at Atrañjikhera. They could possibly be connected with some ritual, and perhaps were not merely children toys. As for toys, terracotta balls were probably intended for children, while the game-pieces of various shapes used for a primitive class. Though wheels have been found, toy-carts are missing. This suggests that the wheels are probably spindle-whorls. Similarly other terracotta objects were made to meet the needs of life. These objects include stoppers of containers, reels, discs, net-sinkers, crucibles and clay-stands. ### [v1.c19.s5.t1.2](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t1.2) In all, twenty-six terracotta beads have been recovered from the different phases of Period III. Sixteeen of them are broken or damaged. Generally, the beads are well-baked and eighteen of them are slipped. On the basis of their shapes, the beads may be divided broadly into eleven types. Apart from this, twelve beads of semi-precious stones have been recovered from the different levels of Period III. Out of these, six beads are broken or damaged. On the whole, the beads show a high order of workmanship and skill. Agate is represented by eight beads of different shapes including a small spherical etched bead. While carnelian is the material for another two beads, quartz is represented by a standard barrel triangular bead and marble by a short circular bead. Besides these finished beads, waste cores and raw material have also been found. ### Stone Objects ### [v1.c19.s5.t2.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t2.1) In all, eleven stone objects and twenty-two ordinary stone pieces have been unearthed from the different levels of Period III. The objects include six pestles, two white stones, one small ball and two indeterminate pieces, one of which is of soapstone. All the objects, except the last, are of sandstone. The pestles of this period can be grouped into several types. ### Iron Objects ### [v1.c19.s5.t3.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.1) The most significant discovery of the period is that of iron tools and implements. These have been found in practically all the layers suggesting that the use of iron was common throughout the PGW period. However, a majority of them came from the Upper phase. In all, one hundred and thirty-five iron objects were recovered including seven pieces of iron slags and fourteen indeterminate objects. The identifiable finds comprised of fourteen kinds of objects viz, arrowhead, spearhead, shaft, pair of tongs, clamp, chisel, bar (rod), borer, needle, hook, nail, axe, knife, and bangle. One may broadly divide them into three groups: (i) Weapons, (ii) Craft tools and implements, (iii) Household objects and ornaments. ### [v1.c19.s5.t3.2](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.2) The presence of iron objects in such profusion (vide Table 11.1) and the discovery of furnace, slags (Table 11.2) and certain specific tools used by blacksmiths suggest that not only were iron goods manufactured at the site, but that the smelting of iron ore was also carried out here. It has been suggested that the main source of iron for PGW people of Atrañjikhera was most probably the region extending from south of Agra, to Gwalior containing rocks in which iron content is 'high'. The extensive deposits of banded haematite-quartzite found in these rocks were considered important and have been worked out on large scale by the indigenous smelters. Chemical analysis of iron shales collected from Agra-Gwalior range shows that the material contains 21.7 per cent iron, 41.9 per cent SiO₂, 14 per cent Al₂O₃, 6.3 per cent CaO, 2.3 per cent MgO. 1.4 per cent Na₂O and 6.8 per cent K₂O. O.P. Agrawal claims that this shell might have been used for iron-making at Atrañjikhera in view of the fact that SiO₂, Al₂O₃, CaO and MgO are present in the above material as well as in the finished objects and slags. (Gaur, 1983, 487-500) ### [v1.c19.s5.t3.3](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.3) However, the proportions of SiO₂, Al₂O₃, CaO and MgO are lower in four of the five slags (exception being slag No. 3 vide Table 11.2) compared to the raw material, and iron percentage much higher. Therefore, the raw material used must have been lot purer than the iron shale (as analysed above) to produce slag containing as high as 63-66 per cent Fe₂O₃. Alternatively it may be suggested that only No. 3 is acceptable as a representative slag. ### [v1.c19.s5.t3.4](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.4) Table 11.1 ### [v1.c19.s5.t3.5](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.5) \*\*Analysis of Iron Objects (Chemical and Spectrographic)\*\* \*\*from Atrañjikhera\*\* ### [v1.c19.s5.t3.6](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.6) | | Samples | | | | | | 1 | 2 | 3 | 4 | | Fe₂O₃ | 89.36 | — | — | — | | SiO₂ | 0.53 | 1.80 | 1.51 | 1.32 | | Al₂O₃ | 1.33 | .82 | .96 | 1.25 | | CaO | 0.60 | 1.31 | .82 | .73 | | MgO | 0.12 | 0.19 | .71 | .62 | | Cu | — | — | — | — | | Ni | — | — | — | — | | TiO₂ | Tr. | Tr. | — | Tr. | | P₂O₅ | 0.33 | 0.21 | 0.18 | 0.21 | | S | 0.08 | 0.10 | 0.09 | 0.12 | | C | 0.33 | 0.18 | 0.26 | 0.31 | | MnO₂ | Tr. | Tr. | Tr. | Tr. | | Co | — | — | — | — | | Zn | — | — | — | — | | As | — | — | — | — | ### [v1.c19.s5.t3.7](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.7) Percentage Composition of Slags from Atrañjikhera ### [v1.c19.s5.t3.8](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.8) | | Samples | | | | | | | Slag No. 1 | Slag No. 2 | Slag No. 3 | Slag No. 4 | Slag No.5 | | Fe₂O₃ | 63.76 | 65.36 | 30.4 | 66.6 | 56.0 | | SiO₂ | 29.39 | 25.34 | 57.3 | 25.6 | 30.5 | | CaO | 0.86 | 2.41 | 8.4 | 1.3 | 2.2 | | MgO | 0.64 | 0.97 | 0.6 | 0.2 | — | | MnO₂ | 1.42 | Trace | Trace | — | — | | Al₂O₃ | 3.14 | 4.76 | 3.1 | 5.0 | 4.4 | | P₂O₅ | 0.77 | 0.69 | — | — | — | | Total | 99.98 | 99.53 | 99.8 | 98.7 | 93.1 | ### [v1.c19.s5.t3.9](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.9) It seems likely that the ore was put at first along with fuel in an open circular or oval-shaped pit-like furnace, and then reduced at high temperature raised with the help of a leather bellow. The metal thus formed might have been then forged repeatedly and hammered out to be in proper shape for preparing the desired objects. ### [v1.c19.s5.t3.10](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.10) This iron industry must have maintained a number of black-smiths. Another important class of skilled workers would have been that of carpenters, since several items like axe, nails, clamps, chisels, and borers apparently belong to their tool-kit. The presence of arrowheads and spearheads were probably used in both war and hunting. Though the presence of cultivated wheat, rice, barley and pulses has been shown during this period, it is surprising that not a single iron implement irrevocably associated with agriculture has been found so far. The association of iron with the painted gray ware has how been attested from other sites. Atrañjikhera has yielded what is so far the earliest reported find of iron, placed stratigraphically in twelfth century BC. (Gaur 1983, 219-20) ### [v1.c19.s5.t3.11](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.11) There is no doubt that iron implements were being made in India since very early times, but in comparison to copper and its alloy-bronze, the use of iron started rather late. This may be so on account of the problems presented in the working of iron. It melts at a high temperature of 1537°C. The Indus Valley Culture sites have not yielded iron objects. ### [v1.c19.s5.t3.12](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.12) Some of the iron objects as well as slags, excavated at Atrañjikhera were examined to gain some knowledge about the metallurgy of iron in that remote period. The aims of the metallurgical analysis were to know the nature of iron used; whether it is meteoric, wrought iron, cast iron, carbon steel, alloy steel; details of metallurgical operations, nature of the ore and if possible, its location, the use of flux, metallurgical refinement, whether any slag was present in the metal or it was completely removed, working temperatures, etc. ### [v1.c19.s5.t3.13](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.13) Samples were taken from four iron objects: No. 1 from the upper phases and nos. 2-4 from the middle phase of the PGW stratum. All samples had some metallic core to allow taking good representative samples. The chemical analysis of these iron objects show that silica, lime, alumina, magnesia and manganese are always present in the metal, although as impurities. Silica is in the range of .53 to 1.8 per cent, lime ranges from 0.6 to 1.31 per cent, and magnesia is .12 to .7 per cent. The phosphorous oxides are ranging from .18 to .33 per cent. Sulphur is also present. Titanium is present in traces. The percentage of carbon is low. Nickel, cobalt, zinc, arsenic are all absent. (\*vide\* Table 11.1) ### [v1.c19.s5.t3.14](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.14) It is clear from the above account that the iron used for these objects was not of meteoric origin. Meteoric iron is always associated with nickel and very often cobalt. Even otherwise, it is known that the working of meteoric iron is rather difficult. Furthermore, meteoric iron is found rarely, and would never be available in sufficient quantities for general use. ### [v1.c19.s5.t3.15](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t3.15) Metallurgical examination of the four samples gave some very interesting results. Sample No. 1 was clearly a sample of wrought iron showing no pearlite formation but containing multiple slag inclusions throughout the body. The other samples showed apart from slag inclusions, lamellar and spherodized pearlite in zones near the surface and ferrite in other areas. From this examination it appears that these objects were first made of wrought iron and later on were carburized by some technique whereby the surface or the zone nearest to the surface turned into carbide. This happens when the iron object is kept on charcoal bed for a long time at high temperature. For sample 1, either it did not pass through this process or else the surface has by now completely corroded. (Gaur 1983: 487-89; 495-96) ### Copper Objects ### [v1.c19.s5.t4.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.1) In all, twenty-two copper objects have been recovered from the different levels of Period III. It appears that the introduction of iron, which was the most suitable metal for arms and implements, drastically reduced the use of copper. Its use was now more specialised, it serving as the material chiefly for ornaments, toilet-goods, decorative objects, kitchen-utensils, and the like. More than half the items recovered were found in badly damaged conditions. However, these offered good evidence of the copper-smith’s craft. The fragment of a copper dish is particularly worth mentioning. The presence of fish-hooks show that the fish, bivalve, etc. formed part of the diet. Other copper items include rings, antimony rods, bangles, clamps, antimony rod-cum-nail parers, a pin and a celt. (Gaur, 1983: 230-32) ### [v1.c19.s5.t4.2](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.2) Analysis of four copper objects are presented in Table 11.3. ### [v1.c19.s5.t4.3](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.3) The analysis of Sample No. 2 shows that the copper was almost without any deliberate alloying. The presence of iron, tin, lead and aluminium is evidently on account of impurities coming from the ore. No arsenic or antimony could be detected. On the other hand, Sample No. 1 showed the inclusions of tin to the extent of 11.68 per cent and lead and zinc 9.00 per cent and 6.28 per cent respectively. ### [v1.c19.s5.t4.4](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.4) Table 11.3 ### [v1.c19.s5.t4.5](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.5) Chemical Analysis of Copper Objects from Atrañjikhera ### [v1.c19.s5.t4.6](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.6) | | 1 | 2 | 3 | 4 | | Cu | 70.99 | 97.33 | 52.42 | 71.6 | | Sn | 11.68 | 0.49 | 20.72 | 12.9 | | Pb | 9.00 | 0.57 | 9.77 | Tr. | | Zn | 6.28 | — | 16.20 | Tr. | | Fe₂O₃ | 1.93 | 1.34 | 1.23 | 9.7 | | Al₂O₃ | 0.18 | 0.15 | 0.01 | — | | MnO₂ | — | — | — | — | | Ni | — | — | — | — | | Co | — | — | — | — | | As | — | — | — | — | | Sb | — | — | — | — | | CaO | — | — | — | — | | MgO | Tr. | — | — | — | | S | 0.12 | — | 0.08 | 0.3 | | Total | 100.18 | 99.88 | 100.43 | 94.5 | ### [v1.c19.s5.t4.7](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.7) The result of the analysis of Sample No. 3 was very curious inasmuch as it contained a very high amount of tin (20.72 %) and also zinc which is 16.20 per cent. There is no arsenic or antimony detected in this sample. These figures indicate heavy alloying with tin and use of zinc ore. ### [v1.c19.s5.t4.8](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t4.8) It is found that sulphur is present in Sample Nos. 1, 3 and 4 although not in appreciable quantities. The probability is that a sulphide ore was used in smelting copper. Since iron is present in all the four samples, (in three of them ranging between 1.23 per cent and 1.93 per cent and in the fourth to the extent of 9.7 per cent), the sources could be the iron-bearing chalcopyrites. Normally iron is removed at the time of smelting. However, if the process of melting is not thorough, the product would contain a higher percentage of iron as in the case of Sample No. 4. (Gaur 1983, 483-90) The copper samples might have come from Ahar. ### Glass Pieces ### [v1.c19.s5.t5.1](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.1) A small piece of glass was examined and analysed. It was dark green in colour. It is obvious that it formed part of a bottle or some other receptacle. The results of chemical analysis give 56.96 per cent SiO₂, 4.46 per cent Fe₂O₃, 3.62 per cent Al₂O₃, 1.09 per cent Cr₂O₃, 1.06 per cent PbO, 0.89 per cent SnO₂, 7.00 per cent CaO, 3.75 per cent MgO, 17.55 per cent Na₂O and 4.09 per cent K₂O. ### [v1.c19.s5.t5.2](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.2) The amount of alkalies in the glass piece is rather high. It is a soda-lime soft glass. The colour is due to the presence of iron and chromium. (Gaur: 1983, 490) This is one of the earliest evidences of glass being made in ancient India. ### [v1.c19.s5.t5.3](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.3) Hegde (1987) has suggested that glass in India was an innovation of the PGW period, and \*sajjī maṭṭi\*, also called \*reh\*, \*khari\*, was used for making the early variety of soft glass. \*Sajjī maṭṭi\* is an alkaline, natural efflorescence, profusely found in the Ganga Valley, chiefly in mid-eastern and eastern Uttar Pradesh and Bihar. During the summer months, it comes out as a superficial incrustation on the alluvial soil as a 2-3 cm thick layer of glistening, white, crystalline salt, shining brightly in sunlight. It assays 20-40 per cent soda, variable proportions of Na₂CO₃ and Na₂SO₄, 5-10 per cent lime and 50-70 per cent silica; its use with some raw material containing Fe and Cr in making the dark green soft glass, as mentioned in the previous paragraph, can be easily surmised. \*Sajjī maṭṭi\* was also used in providing the liquid glaze on the later day NBP wares. ### [v1.c19.s5.t5.4](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.4) Hegde (1987) examined thin sections of PGW and found ‘insignificantly small number’ of larger and heavier particles of feldspar, quartz or limestone, thus proving the thoroughness of labour that the ancient potters had put in to prepare well-elutriated fine clay before making PGW. Red ochre containing Fe₂O₃ must have been used as the painting medium to decorate PGW before baking in a reducing atmosphere (in a \*gajapuṭa\* type of sealed furnace in which carbonaceous matter burned in an insufficient supply of oxygen producing carbon monoxide) since the black paint on the PGW has been proven to be Fe₃O₄ or magnetite. It is noteworthy that in this PGW era, one further step of reduction of iron oxide led to the first ever production of iron metal in India. ### [v1.c19.s5.t5.5](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.5) Fe₂O₃ → Fe₃O₄ → Fe ### [v1.c19.s5.t5.6](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5.t5.6) Although not much gold and silver have been found from any PGW site, which shows that the Harappan type of affluence re-emerged only much later in the NBPW era, Atrañjikhera excavation has yielded an interesting soapstone fragment of a mould which could have been used to prepare gold jewellery. ## Linked texts (1) - [Atranjikhera](https://indike.org/pramana/inp.site.atranjikhera) --- Source: [https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5](https://indike.org/pramana/ins.science.biswas-1996.v1.c19.s5) · Pramāṇa