Melacheri, a tiny village situated in Gingee taluk of Villupuram District has a rock–cut cave temple dedicated to Siva. This place is also known as Siva Gingee. This cave is excavated from the natural rock lying on the outskirts, north of this village. Locally, it is known as the Mattisvarar temple. It has a square garbhagriha with an oblong mukhamandapa in front of it. This mandapa is supported by four square pillars having neither ornamentation nor an octagonal shape in the middle, which we distinctly see in the Mahendra type cave temples. An inscription is found engraved on the western face of the Southern pillar1. The epigraph is written in Sanskrit and in the Grantha character. The said inscription, written in four lines, is stated to have read as follows2.

  1. Karitam-idan-nripatina
  2. Chandradityena Sarva(nathe)na (1*)
  3. Sri Sikhari-Pallavesvaram-iti
  4. Saivam-dha(ma) Simhapure (11*)

The meaning of this inscription is given as follows: “This home of Siva named Sri-Sikhari-Pallavesvaram, was caused to be made at Simhapura (Singavaram) by King Chandraditya who was a Sarvanatha “.

Dr. T V. Mahalingam, in his ‘Inscriptions of the Pallavas’3, has published this inscription with the same version and meaning. But, Mayilai Seeni Venkatasamy in his Tamil book entitled ‘Mahendravarman’4 has given a different version of this same epigraph. His reading is as given below:

  1. Karitam Itam nrupatina
  2. Chandradityena Sarvabhaumena
  3. Sri Sikhari Pallavesvaramiti
  4. Saivantama Sittir Ashtukra

But he also has given the same meaning as stated by K.R. Srinivasan and T.V. Mahalingam. His reading differs in two lines. In the second line, he says ‘sarvabhaumena’ as against sarvanathena, and in the fourth line instead of ‘Simhapuri’ he read it as ‘Sittir Ashtukra’. However, he has not explained his different reading and has also not changed the meaning of the inscription.

With the help of the estampage5 prepared by Viraraghavan of Villupuram, it can be read as follows:

  1. Karitamita Nrupatena
  2. Chandradityenu Sarvva (bhaume) na
  3. Sri Sikhvari or (Sri Sisvari)
  4. Saivandhama Simhapure

Accordingly, the meaning of the inscription could be changed as follows:

This home of Siva named Pallavesvara at Singapura (Singavanam) was caused to be made by one Sikhisvara, who was a Chandraditya Sarvabhauman. The letters after Chandraditya are mutilated but sarvva and na are readable. Hence, the reading of the second line by Mayilai Seeni Venkatasamy may be taken as granted.

If the revised reading and meaning are acceptable, the excavator of this temple emerges. It appears that Sikhisvari was the excavator, and she could be the wife of a vassal of the Pallava ruler.

While narrating the details of the architectural features of this temple the author of this cave temple, K.R. Srinivasan has given the following remarks:

“The facade of the excavation consists of two massive pillars and two pilasters-one at either end. All of them are not differentiated into sadurams and kattus, but are uniformly four-sided throughout, a feature not common in the Mahendra type of temples. The massive corbel (potika) is of the bevelled type and has an angular profile without the taranga ornamentation and patta. The beam on top of the corbel is about the same width as the corbel. Cut on the central portion of the hind wall of this mandapa is a plain oblong opening into the cubical cell, in excavating which a large portion of the central mass of rock has been retained and moulded into a cylindrical ‘linga’ with an octagonal ‘avudaiyar’ below. This is the only example of a rock-cut cave temple in Tondaimandalam where the ‘linga’ and ‘avudaiyar’ are cut out of the same rock of the cave temple – a feature which is not found in the Mahendravarman caves or even in the caves of the era of Mamalla alias Rajasimha. Carved on the ceiling over the Linga is a lotus6”.

Regarding the authorship of this cave temple, his statements are more interesting. “The absence of any clear birudas of Mahendravarman, the extreme simplicity of the cave temple and the presence of the rock-cut linga with avudaiyar would all tend to show that this cave temple is later than the time of Mahendra, perhaps the time of Mamalla, if not even later, like the Singavarman cave temple. But though this cave temple is aberrant and has nothing characteristically Pallava in it and was not excavated by a Pallava king, it has to be considered here only because of its location among other Pallava cave temples in the Pallava country7”. Since this cave temple is situated in the Pallava region he was tempted to identify it with a Pallava King, which he otherwise could not have done.

Assuming that the word Chandraditya denoted the name of a king, he could not identify the king. But he says that there was a Chandraditya, son of Pulekesin II, known from two records of his queen Vijayamahadeva, who was a contemporary of Mammalla, and speculates about her relationship with the Pallava monarch if, indeed, she existed. Finally, he himself opines that it has to be explained. Hence, it may not be possible to presume that he was Chandraditya, the son of Pulakesin II. In that case, who is Chandraditya?

The word Chandraditya may mean a dynasty to which Kalabhra kings are said to have belonged. The inscriptions of Kokkandan Ravi8 and Kokkandan Viranarayanan,9 the Kalabhra kings of Ponnivadi and Vellalur respectively, distinctly name them Kali Nrupa Kalvan and mention that they are the thilakams (forehead mark i.e. the supreme kings) of Chandraditya kula and Sarvabhauma, one who controls the whole earth. Kali Nrupa Kalvan is the Sanskrit version of Kalabhranenum Kalaiarasan10 written in Tamil in the Velvikkudi Copper-Plates of the Pandyan Parantaka Nedumjadaiyan. Hence, it leads us to assign this temple to a Kalabhara vassal.

Simhavishnu, the establisher of the Pallava rule in Tamilnadu is stated, in the Kasakkudi plates, to have vanquished the Kalabhras, Malavas, Cholas, Pandyas and Simhalas in the course of his extension of the Pallava territory as far as the Kaveri River. Therefore, the presence of Kalabhras during the period of Simhavishnu is clearly attested. Considering the architectural feature of Melaccheri cave temple it should be assigned to a period earlier, than Mahendravarman, i.e. either to Simhavishnu or still earlier, as the pillars and the corbels are square-shaped and the ‘linga’ and ‘avudaiyar’ have also been scooped out of the natural rock. These architectural features can be found only in the earlier cave temples. Hence, in all probability, this cave could have been excavated by Sikhesvari the wife of a Kalabhra vassal for the benefit of his master, a Pallava monarch.


  1. K. R. Srinivasan, Cave Temples of the Pallavas, p. 115 (Reprint 1993).
  2. S.I.I. Vol. XII No. 115 ARE. 234 of 1916, Part II, para 6, Dubreuil, Pallava Antiquities, 1, p. 66.
  3. Dr. T.V. Mahalingam, Inscriptions of the Pallavar, No. 249, p. 623 (pub. 1988).
  4. Mayilai Seeni Venkatasamy, Mahendravarman (Tamil) p. 74, (pub 1959) by Sava Siddhanta Publications.
  5. V. Viraragavan
  6. K. R. Srinivasan, Op.Cit., p. 116.
  7. Ibid., p. 118.
  8. Natana Kasinathan, Kalappirar (Tamil), p. 30. (pub. 1981) by the Tamilnadu State Department of Archaeology.
  9. Natana Kasinathan, Kalleluthukkalai (Tamil), pp 183-187. (pub. 1989) by Manivasakar Patippakam. 55, Lingi Street, Chennai – 1.
  10. Ten Pandya Copper Plates, p. 22 line 40 (pub. 1967) by the Tamil Varalatru Kazhagam, 40, Peters Road, Chennai – 14.

Natana Kasinathan
Former Director of Archaeology
Government of Tamilnadu

Source: Journal of Indian History and Culture, March 2004.

You can leave a response, or trackback from your own site.

Leave a Reply