250 



JOURNAL, R.A.S. (CEYLON). [VOL. XIV. 



Completely, I say, in contradistinction to the unavoidable 

 4t half -measures " of 1895 ; when, with fewer hands and less 

 light, mere trenches were dug along walls and steps, and the 

 nett result was but partially satisfactory. Plotted on paper 

 the general plan of the ground dug last year can be fairly 

 well understood ; but go to the spot itself, and all seems 

 confusion — indeed, must be confused, so long as the inter- 

 vening earth and debris are not fully removed. 



In this — our second — season, working with a larger force, 

 and upon the useful experience acquired in 1895, the 

 mistake was not repeated. From the outset the coolies were 

 put to the slow but sure task — galling from its wearisome 

 monotony and inherent delay — of digging out the whole 

 mass of hard-caked brick and earth hiding walls and stair- 

 cases, down to the original floor level. 



The area excavated this year covers (omitting the pokuna) 

 acres and upwards. As the depth of earth-cutting varied 

 from 5 ft. to as much as 20 ft. in places ; and as, again, every 

 basket of " spoil " had to be carried to the east, or south, edge 

 of the rock and thrown over, the completion of the heavy 

 task before the south-west gale fell on us taxed our powers 

 to the utmost. 



But the ultimate benefit of this wholesale "sweep " of the 

 debris, laborious and slow though it be, cannot be exaggerated. 

 Nothing is missed ; walls and foundations can be perfectly 

 differentiated; forms of moulding examined properly; above 

 all, thus only can there be got a comprehensive and intelli- 

 gible view of the trend of walls and cross-walls, and the inter- 

 communication of a perfect labyrinth of stairs and passages. 



The " Plan of the excavations on the Summit of Sigiri- 

 gala, 1895-96,"* and the composite photographs,! best give 

 an idea of the lie of the several terraces and the direction of 

 the stairways descending from the level of the pokunahund. 



Not to burden a provisional Report, such as this, with a 

 load of dry details and measurements, it will suffice to draw 



* Reproduced. 



f C 593-595, 597-602 : not reproduced. 



