Before the Nagarjuna Sagar Reservoir Rose, the Archaeological Survey Dug 130 Sites at Nagarjunakonda, Recovered the Ikshvaku Capital Vijayapuri and Rebuilt Its Monuments on Higher Ground

The Archaeological Survey of India (ASI) dug 130 sites across 24 square kilometres of the Nagarjunakonda valley in Andhra Pradesh between 1954 and 1960, found the remains of more than 70 structures, and rebuilt some of the best of them on higher ground before the Nagarjuna Sagar reservoir covered the valley. The survey’s museum portal describes the rebuilding, done with original material and a similar setting and alignment, as the first of its kind in India, and compares it with the rescue of the temples at Abu Simbel in Egypt. What you can see today is an island museum, built like a Buddhist monastery, and a handful of rebuilt monuments, reached by motor launch.

The valley was Vijayapuri, the capital of the Ikshvaku dynasty. The portal dates most of the finds to the 3rd and 4th centuries CE and says the dynasty ruled the lower Krishna basin for about a hundred years.

A valley first noticed by a schoolteacher

The ASI portal says the ruins were first brought to light by S. Venkataramayya, a teacher, and made known to scholars by A. R. Sarasvati in March 1926. A. H. Longhurst and M. H. Quraishi then dug there for the survey, and T. N. Ramachandran followed in 1938. The portal gives the first team’s years as 1924 to 1931; ThePrint and Live History India both date Longhurst’s excavation to 1927 to 1931. Longhurst’s report was published in 1938, and Zin’s paper cites it.

The portal says the hill was named after the Buddhist teacher Acharya Nagarjuna and that his efforts made the valley a centre of Buddhist study. The portal also gives Sriparvata as an ancient name of the Buddhist centre here.

Six years against the water

The ASI’s Indian Archaeology review for 1954 to 1955 reports that intensive excavation began in October 1954 and had opened seven important sites in that first season, among them monasteries, stupas, an assembly hall, a closed bath and two temples. ETV Bharat dates the foundation stone of the Nagarjuna Sagar dam to 10 December 1955, so the dig had started about fourteen months earlier.

The project, under Dr. R. Subrahmanyam, grew each year. By 1955 to 1956 it had uncovered at least fifteen monasteries, eight stupas, a medieval temple and a bathing ghat on the Krishna, with a stepped brick embankment leading toward what looked like the palace area. The 1957 to 1958 review puts the excavated area at about 1,200 acres, and the 1958 to 1959 review at about 1,800 acres (roughly 730 hectares).

What the Ikshvaku city looked like

By 1957 to 1958 the survey had uncovered the citadel, with a fortification wall, a ditch, gates and barracks, which it said showed the town planning of an Ikshvaku capital in the third century. Beside it lay a brick-built oblong arena 309 feet long, 259 feet wide and 15 feet deep (about 94 by 79 by 4.6 metres), ringed by flights of steps with a pavilion on the west side. The museum portal calls it the amphitheatre or stadium.

The 1956 to 1957 review describes a pattern in the layout. The northern reaches along the river, including the fortified zone, held mainly Brahmanical shrines and ritual structures, while the interior held countless Buddhist monastic establishments, in the review’s words. One inscription from that season records a temple of Mahadeva, called Pushpabhadrasvamin, built in the sixteenth year of the king Ehavala Chantamula, with a village given for its upkeep. The portal adds that the Ikshvaku kings worshipped Kartikeya and Shiva and performed the ashvamedha and vajapeya sacrifices, while the women of the royal household leaned toward Buddhism.

The inscriptions also filled in the dynasty. The 1955 to 1956 review reported a new Ikshvaku name, Rudrapurushadatta, and the next year’s review added a fifth, Virapurushadatta II, a son of Ehavala Chantamula. The review for 1961 to 1962 reports the only Satavahana record from the site, a pillar inscription of Gautamiputra Vijaya Satakarni dated to his sixth regnal year. The portal says the city’s name came from him. The Satavahanas, who held the region before the Ikshvakus, left a much earlier record at Naneghat in Maharashtra.

Nearly forty Buddhist sites and the schools that lived in them

Monika Zin, a scholar at the Saxon Academy of Sciences and Humanities in Leipzig, counts nearly 40 Buddhist sites excavated at Nagarjunakonda, built within about a hundred years. In a 2022 paper she notes that only six of them carry inscriptions naming a school, which gives four schools in all: the Aparamahavinaseliyas, Theravadins, Mahisasakas and Bahusrutiyas. She finds that the monasteries with narrative relief carvings share one layout, with a stupa outside a U-shaped block of cells and two apsidal shrines facing each other, and that one of them names the Aparamahavinaseliya school, which she suggests was the one that favoured such carvings.

The survey’s reports record the variety. The 1954 to 1955 review describes a stupa, typical of the Andhra country, with a diameter of 48 feet laid out as a wheel with a hub and ten spokes. The 1955 to 1956 review describes a slab carved with a pair of Buddha’s footprints, inscribed in the middle third century, naming monks of the Theravada-Vibhajyavada school of Sri Lanka. The Great Stupa of Amaravati, on the same river, belongs to the same artistic tradition, and the museum portal calls Nagarjunakonda the final phase of it.

Gold from Rome in a Krishna valley

The 1956 to 1957 review lists two Roman gold coins from the dig, one of Tiberius (16 to 37 CE) and one of Faustina the Elder, issued after 141 CE. The next year’s review notes sprinklers and a few handles from Roman amphorae, and clay bullae (lumps pressed with a seal) that show a trident and a human head, possibly copied from a Roman coin. The portal attributes the city’s prosperity to uninterrupted sea trade with the Roman and Byzantine empires along the east coast. The Kerala side of that trade is described in the piece on Muziris and the Pattanam digs.

What was saved, and where it is now

The portal says some of the outstanding structures were rebuilt on the island and on the eastern bank of the lake near Anupu. ThePrint and Live History India each say nine monuments were reconstructed; the ASI portal does not give a number. Zin writes that some sites were taken apart and rebuilt on higher ground, that half-size models were made of several others, and that most survive only as small models in the museum garden.

The museum was built in 1966, according to the Museums of India listing. It is laid out like a Buddhist vihara among the remains of a medieval fort, with five galleries. Gallery 1 holds carved slabs from the platforms of the stupas. Gallery 3 holds a model of the submerged valley, 6.01 metres square, which marks about 120 excavated sites. Gallery 5 holds the inscriptions, including the Satavahana pillar record. The portal’s digital catalogue counts 8,450 objects, of which 3,636 are coins. The listing adds that some excavated pieces are now in Delhi, Chennai, Kolkata, Paris and New York.

Reaching the island

Practical details below were checked on 29 October 2026 on the two official museum pages, which do not agree on every point.

The ASI portal says the island is approached by motor launch from Vijayapuri South, and that Macherla, 24 kilometres west of Vijayapuri South, is the nearest railway station. It gives the museum’s opening hours as 10:00 am to 5:00 pm. The second Museums of India listing gives 9:00 am to 4:00 pm and names the open days as Sunday to Thursday and Saturday, with no Friday. Travel magazines give different launch timings and fares, and none of them shows a date, so this article gives none. Check the launch schedule with Andhra Pradesh Tourism and the museum hours with the ASI museum, on the contact number shown on the portal, before travelling.

What the survey could not save

Zin writes that the excavations did not meet today’s standards and were done under heavy time pressure, and that it is futile to expect any new material from the valley. ThePrint reports that in 2015, when the reservoir fell below its minimum level, a megalithic burial ground of more than 200 graves came back above the water. It cites Telangana’s Department of Archaeology and Museums for a date of about 1000 BCE to the 2nd century CE for the stone circles.