Buddhist monks is evident from his statement that "three li before you reach the top of Mount Gṛidhrakûṭa there is a cavern in the rocks facing the south in which Buddha sat in meditation; thirty paces to the north-west there is another where Ânanda was sitting in meditation when the Dêva, Mâra-Pisuna, having assumed the form of a vulture took his place in front of the cavern and frightened the disciple; going on still to the west they found the cavern called Sritapara, the place where after the nirvâṇa of Buddha 500 arhats collected the sûtras." ³ The Buddhist priests of later years than the time of the great founder appear to have followed the same practice, and the hands of the devotees developed the rude natural caves into habitable dwellings befitting their residents. Whether they were primarily designed as the provision for the annual "retreat" initiated by Buddha when it was ordained that the monks were to keep vassa and refrain from peregrination during the rains, or were intended to give a cool resort during the hot season, cannot now be easily determined. Besides being watertight, convenient for human habitation and far above any possible accident from the rains and floods of the monsoon, to this day they are agreeably cool even in the hottest weather. The doubt raised in the first part of the passage quoted here, whether the caverns were designed for the annual "retreat" or were intended to give a cool resort, can be cleared from the reply which Mâhinda gave to Tissa when the latter requested the saint to halt in the beautiful garden adjoining his capital on a certain night. The statement⁴ of the thêra shows that the Buddhist monks were prohibited by the rules of their order to stay even in the immediate proximity of cities or villages, and it also accounts in a way for the necessity for the caverns.

In the general forms of these, viz., one boulder overhanging another, a flat one on which it rests at one extremity, in the cutting of the projecting rock to a certain depth in order to prevent the rain water from gliding into the cavern, in the existence on the bottom boulder (1) of smoothly chiselled beds with a slightly raised portion for the head, just sufficient for a man to lie down, (2) of the groove immediately in the outer fringe of the cave quite below the cutting on the upper rock for carrying away the drippling of the rain water to a distance, (3) of big holes cut on the open yard intended perhaps for fixing poles or railings, and (4) of a number of smaller holes for other works of protection-in all these details the caverns of the Pâṇḍya country resemble those in Ceylon, which are assuredly Buddhistic in their character. As Ariṭṭa and his followers,

------------------------------------------------------------------------------------------------------------------

³Ibid.

⁴ Mahâvaṁśa, Wijesinha's translation, p. 54.