we may presume that the hall which originally stood on the square pavement in the centre of the monastery was entirely of wood (Plate V (b)). The walls are of brick and plaster and traces of a few plain mouldings discovered along the plinth of the cells show that the ornamentation was in plaster. Since no tiles were found in the débris, the wooden roof over the cells was probably covered with thatch. In one of the cells belonging to this monastery a large number of small lead coins of the usual Andhra type of about the second century A.D. were found, including a lump of lead ore and an earthenware die for the manufacture of coins of this size and pattern, indicating that the monks made their own coins. A large grindstone for crushing grain, a small broken limestone image of the Buddha, a few pieces of ornamental -pottery, a broken pottery roof finial of the usual kind, and the curious little doll-like terra-cotts female figurine shown in Plate VII (a) were also recovered from this site. This figurine is only 3 inches in height and was found 10 feet below the foundations of Monastery 1 and therefore, may belong to an earlier period. Another interesting terra-cotta figure discovered during the excavation of Stapa 9, which I will describe later, is shown in Plate VII (6). The head only was recovered and from the position in which it was found at the bottom of the stupa it appears to have got there by accident’ when the interior of the stupa was filled in with earth, and therefore may have no real connection with the stupa,
Monastery 2 is a small but well-arranged vihara (Plate V (c)). The court- yard contains two apsidal temples facing each other, the usual central hall, and a row of cells and rooms all round the enclosure. An entrance on the east side leads into a second open court-yard containing a long building abutting onto the eastern wall. This structure has a stone seat all round the room and a big stoné table outside the doorway so we may presume it was the refectory. On the opposite side of the enclosure is a long stone bench built against the west wall. On the south side are two cells or store-rooms, a kitchen and a small lavatory im the northern corner of the enclosure, and as usual, a stupa stood close to the monastery. (Stupa 4.) All the buildings are roofless, The two apsidal temples had barrel-vaulted roofs of brick construction but the rest of the buildings had wooden roofs covered. with thatch. The pillars of the central hall were of stone and its floor was of the same material. When first discovered the entire building was underground and covered by a mound of brick débris. The difficulty of excavating this monument was greatly enhanced owing to treasure-seekers having broken up the stone pavement of the central hall and scattered the broken slabs all over the place. All this stone had to be dug out, collected and stacked before it was possible to trace out the plan of the monastery (Plate V (d) ).
Ruined pillared halls of all sizes abound at Nagarjunakonda, many of them with their pillars still standing erect. A few inscriptions found on some of the pillars show that many of these buildings belonged to the monasteries and were erected by pious donors as resthouses for pilgrims and visitors. A good example of this type of building is illustrated in Plate VIII (a). This hall formed part of Monastery 3 which is located at the foot of Nagarjuna’s Hill, An inscription
