There are the ruins of an ancient Buddhist monastery or some establishment of the kind in the jungle near the spill of the madukanda tank. Madukanda, or Mandukoddai, as the Tamils call it, is a sinhalese village about three miles south east of Vavuniya, off the horawapotana and trincomalee road. The following description of the ruins is based on the Ratemahatmaya's official report dated October 16, 1890.
An embankment of considerable size encloses a square of about 200 yards' length of sides. The inner slope of the embankment is faced with rough slabs of stone. The square is divided into two by a cross embankment, part of which is not now discernible.
In the western half is another square enclosure with the remains of a wall of brick and rough stone. Close up to the western wall of this smaller enclosure there appears to have been a pond, the bed of which is now filled with broken bricks and other debris. On the other side of the pond are to be seen the remains of what was probably at one time the “Viharage”, all that is left of it now being a single upright pillar with a carved top, and another broken one just opposite it, with five or six others lying nearby. Many broken bricks and fragments of stonework lie scattered about the place.
If the other half of the large square was originally a “vila” (tank) for growing lotuses in, it does not appear how, with an embankment on all four sides, it could have obtained a constant supply of water. But, on the other hand, it is provided with a sluice or culvert just in the centre of the eastern embankment. This sluice is constructed of rough slabs of stone, which form the sides and covering of two parallel channels, each a few feet in width and one or two deep. Besides this sluice there are no signs of there having been any other channel through the embankment. It may have served merely as an outlet for rain water collecting within the enclosure.
If this half of the enclosure was not a tank, it may have been the part allotted to the priests' residence or “Arama”, the western half being that set apart for religious purposes only. This may account for the fact that there is more debris of masonry in the western half than in the other.
The eastern half is covered with thick jungle, but there do not appear to be any ruins of buildings in it. On clearing away the jungle in the western half, and excavating the neighbourhood of the pillars already described, a good many fragments of tiles and earthen pots were discovered, and among them, completely buried, what appears to be a sort of earthenware oven, divided into two compartments, was found. The greater portion of it was intact, one end only having been found broken. The face is ornamented with a pattern of lines.
The back wall of one compartment is broken and detached. The length of the front, which projects an inch or two beyond the side wall at each end, is 16 inches; the height 6 inches; and the depth of the chambers about 6 inches.
I think the site of these ruins was once surveyed. If so, a correct plan might perhaps be obtained from the Surveyor General's Office.