Antiquity of the Narmada Homo erectus, the early man of India

Arun Sonakia* and S. Biswas

Palaeontology Division, Geological Survey of India,
Nagpur 440 006, India

The fossilized skull of Narmada Homo erectus was found embedded in a conglomerate bed in the Narmada valley of central India. This bed occurs at the basal part of a formation sandwiched between two other formations of 0.73 Ma and 74000 yrs BP. The conglomerate bed has also preserved fossils of Hippopotamus namadicus, Equus namadicus, Stegodon namadicus, Sus namadicus, etc. having Middle Pleistocene affinity. All these stratigraphic and palaeontological evidences point to a Middle Pleistocene age of the Narmada Homo erectus.

A hundred years of investigation on human ancestry in India was rewarded in the year 1982 with the discovery by the first author of a skull cap of Homo erectus from the Quaternary alluvial deposits of the Narmada valley

*For correspondence.

in central India1. In its first report, a Middle Pleistocene age, was assigned to it based on a preliminary study of the associated mammalian fossils. But being the only authentic record of Homo erectus from the Indian subcontinent, its precise taxonomic position and age is still under discussion amongst palaeoanthropologists and palaeontologists the world over.

Subsequent multidisciplinary studies on lithostratigraphy, tephrastratigraphy, magnetostratigraphy and bio-stratigraphy of these deposits have thrown fresh light on the problem of age of the Narmada Homo erectus. Discoveries of a layer of volcanic ash and palaeomagnetic reversal in these deposits and for the first time for peninsular India are breakthroughs in fixing the chronologic position of the skull. Fresh data on the age of the Narmada Homo erectus are also available by additional finds of fossil mammals and a detailed taxonomic study along with a thorough review of some families of mammals.

The Homo erectus skull was discovered embedded (Figure 1) in a conglomerate bed within the Quaternary alluvial deposits of the Narmada Valley. Several lithostratigraphic classifications have been suggested for these deposits2 (Table 1). The conglomerate bed occurs at the basal part of a formation designated Surajkund

Table 1.  Litho, chrono and biostratigraphy of the Narmada Valley alluvial deposits

t392.gif (13444 bytes)

392.jpg (25043 bytes)

Figure 1. a, Narmada Homo erectus skull in situ in conglomerate bed. b, Narmada Homo erectus skull, after recovery from in situ position –
right lateral view.

Formation. This formation has a 19 m thickness at its type section and a 50 m compiled thickness. This Suraj-
kund Formation is sandwiched between two formations, the older one showing Matuyama reversed polarity chron3 (0.73 Ma) and the younger one with tephra layer equivalent to Toba ash4,5 of 74,000 yrs BP. Some upper layers of these Narmada Valley alluvial deposits have also been proved to be older than 25,000 yrs BP by 14C dating6,7.

Mammalian fossils of the families Bovidae, Elephantidae, Hippopotamidae, Equidae, Stegodontidae and Cervidae were found associated with the Hominid skull. The Surajkund Formation, which has yielded the hominid skull, has also preserved fossils of Suidae and Canidae which play a significant role in biozonation.

Two extinct species of Hippopotamidae, namely, Hippopotamus namadicus and Hippopotamus palaeindicus occur in successive biozones in the Narmada Valley Quaternary alluvium8. H. namadicus is associated with the Homo erectus skull. This H. namadicus stands in between the other two Indian species H. sivalensis and H. palaeindicus in skull characters, nature of incisors, etc.9,10. While H. sivalensis is reported from the Lower Pleistocene Siwalik deposits, H. palaeindicus is reported from the adjacent Upper Pleistocene to Lower Holocene Son Valley deposits also11. Thus H. namadicus occupies a position between Lower Pleistocene and Upper Pleistocene.

Species of the Equidae family found associated with the Hominid skull is Equus namadicus. It is either closely related12 to Lower Pleistocene E. sivalensis or inseparable13–15 from that. E. namadicus also shows resemblance to Late Villafranchian E. stenonis of Europe and Central China16. This E. namadicus was subsequently replaced by E. hemionus khur during the Upper Pleistocene.

A preliminary study indicated the presence of Stegodon ganesa associated with the Hominid skull1. Further study, however, shows that these specimens do not preserve all diagnostic characters of S. ganesa and should better be assigned to Stegodon sp. A new species Stegodon namadicus with all diagnostic characters is also reported from the Narmada valley17. This species shows close affinity to Middle Pleistocene S. orientalis of South East Asia.

Fossils of Suidae, on the other hand, were not recorded during excavation of the hominid skull. However, a fair number of specimens of fossil Suidae, Sus namadicus have been reported from the Quaternary alluvial deposits of the Narmada valley elsewhere. Later a fragmentary RM3 (A993, GSI, Nagpur) of this species has been collected from the same locality and same litho-unit which has yielded the Hominid skull. Sus namadicus shows affinity to Lower Pleistocene Siwalik Sus falconeri; the species is also reported from the Upper Pleistocene Kurnool Cave deposits18. S. namadicus possesses similar characters to Middle Pleistocene Chou-Ken-Tien form Sus lydekkeri19.

A single specimen (GSI Type No. 20425) of fossilized mandible of Cuon aplinus has been collected20 from the litho-unit which has preserved the Hominid skull. The specimen showing M3 absent and M2 much reduced is somewhat intermediate between Middle Pleistocene European C. a. priscus and C. a. fossilis21. Though slightly larger, the Narmada form is also comparable to Middle Pleistocene Chinese C. a. antiquus22,23.

Most common fossils in the Narmada valley are of Bovidae. Major members of this family, namely, Bos namadicus and Bubalus palaeindicus are reported to range from Middle Pleistocene to Lower Holocene. However, during the last two decades, a large number of collection of fossil Bovidae has been made which deserves a thorough review to make this prolific family more useful for precise biozonation. The other dominant member, Elephas namadicus is also considered to be of Middle Pleistocene to Lower Holocene age. But this form possibly includes two24 or three25 species or subspecies. On the other hand, it is again suggested that European E. antiquus and some Japanese forms belong to E. namadicus26. These controversies are to be resolved to understand the role of this form in biozonation. Another family, Cervidae is not very abundant in Narmada valley. An up-to-date review work or even a proper taxonomic description is lacking for this family and thus its due significance in biozonation is yet to be recognized.

Studies on lithostratigraphy, magnetostratigraphy and tephrastratigraphy reveal that Narmada Homo erectus is from a bed just above a formation of 0.73 Ma and at least 19 m below a layer of 74,000 BP.

Fossils belonging to Hippopotamidae, Equidae, Stegodontidae, Suidae and Canidae are mainly relied upon for biozonation of the Narmada deposits. Fossils of these families collected along with the Homo erectus skull or from the same geological horizons containing the skull point to a Middle Pleistocene (in all probability its lower horizon) age of the Narmada Homo erectus.

The discovery of Indian Homo erectus bridges the gap between African H. erectus in the west and Chinese and Javan H. erectus in the east and south east respectively. There is a general consensus of opinion that Afro-Asian H. erectus ranges in age from Lower Pleistocene to Middle Pleistocene. Indian H. erectus falls within this range.

  1. Sonakia, A., Rec. GSI, 1984, 113, 159–172.
  2. Tiwari, M. P. and Bhai, H. Y., Spl. Publ. GSI, 1997, 46, 41–54.
  3. Rao, K. V., Chakrabarti, S., Rao, K. J., Ramani, M. S. V., Marathe, S. D. and Borkar, B. T., Spl. Publ. GSI, 1997, 46, 65–78.
  4. Basu, P. K., Biswas, S. and Acharyya, S. K., Indian Miner., 1987, 41, 66–72.
  5. Acharyya, S. K. and Basu, P. K., Quat. Res., 1993, 40, 10–19.
  6. Agrawal, D. P., Kotlia, B. S. and Kusumgar, S., Abstract, International Symposium on Palaeoclimatic and Palaeoenvironmental Changes in Asia during the last 4 Ma, P.R.L. Ahmedabad, 1986, p. 36.
  7. Mishra, S. and Rajaguru, S. N., Man Environ., 1993, xviii, 7–12.
  8. Biswas, S. and Dassarma, D. C., Rec. GSI, 1984, 114, 19–24.
  9. Lydekker, R., Palaeontol Indica, 1882, 10, 35–104.
  10. Hooijer, D. A., Zool. Verh. Liden, 1950, 8, 34–55.
  11. Dassarma, D. C. and Biswas, S., J. Earth Sci., 1977, 4 122–136.
  12. Biswas, S., Rec. GSI, 1988, 118, 53–62.
  13. Matthew, W. D., Bull. Am. Mus. Nat. Hist., 1929, 56, 443–447.
  14. Colbert, E. H., Trans Am. Phil. Soc., 1935, NS 36, 1–401.
  15. Hooizer, D. A., Arch. Neerl. de Zool., 1949, 8, 243–263.
  16. Azzaroli, A., Pal. Ital., 1982, 72, 74–97.
  17. Biswas, S. and Dassarma, D. C., Proceedings of Field Conference on N/Q Boundary India 1979, 1981, pp. 15–20.
  18. Pilgrim, G. E., Palaeontol Indica, 1926, NS 8, 64–65.
  19. Zdansky, D., Palaeontol Sin., 1928, 4, 91–100.
  20. Biswas, S. and Basu, P. K., Spl. Publ. GSI, 1997, 46, 117–121.
  21. Kurten, B., Pleistocene Mammals of Europe, Weidenfeld and Nicolson, London 1968, pp. 108–118.
  22. Matthew, W. D. and Granger, W., Bull. Am. Mus. Nat. Hist., 1923, 48, 563–598.
  23. Colbert, E. H. and Hooijer, D. A., Bull. Am. Mus. Nat. Hist., 1953, 102, 41–43.
  24. Tripathi, C. and Basu, P. K., J. Palaeontol. Soc. India, 1983, 28, 63–66.
  25. Pilgrim, G. E., Rec. GSI, 1905, 32, 199–218.
  26. Maglio, V. J., Trans Am.. Phil. Soc., 1973, NS 63, 31–50.


Received 16 March 1998; revised accepted 29 June 1998