30 



JOURNAL, K.A.S. (CEYLON). 



[Vol. VIII. 



have been observed, the ashes and charcoal being in nearly 

 all cases undisturbed. The charred remains of bones which 

 have been roasted were contained in a few ; all the larger 

 ones have been broken, apparently in order to extract the 

 marrow. 



Though not belonging to these houses, a good many 

 pieces of tile have been found in this cutting, of a shape 

 long since obsolete, yet a very effective one. Four parallel 

 grooves, which seem to have been made with the fingers, 

 run along one s ide of the upper surface, the outer one 

 being deeper and wider than the others. At the opposite 

 side, on the under surface, a similar deep groove ran 

 close to the edge of the tile, so that, each tile overlapped 

 and fitted into the groove of the adjoining one on its left 

 side. 



This arrangement must certainly have prevented any 

 leakage, while at the same time it was so simple as almost 

 to be worth adoption at the present day, were it not for 

 the thickness of tile which it requires. In the upper part 

 of each tile a hole was made to receive a wooden peg for 

 holding the tile in position. The tiles were of large size, 

 being probably 12 inches long, inches wide, and from 

 three-quarters of an inch to an inch thick. They all 

 appear to be well burnt. As none of these houses of the 

 workmen, or other dwellings of the poorer classes which 

 have been cut through in another channel, were covered 

 with tiles, although there was a manufactory on the 

 spot, it must be presumed that tiles were employed for 

 roofing only wiharas and the dwellings of the wealthier 

 classes. 



No bricks were used in building these houses of the work- 

 people ; but in a series of better-class structures cut through 

 in a high-level channel from the western sluice, all the floors 

 were laid with them. The remains of a dagaba have been 

 found near this site ; so that these may have been wiharas, 

 or other baildings connected with them. As they are part 

 of the subject of this report, I have measured the bricks at 

 the different dagabas and those found at various ruins 

 in the city of Magama. The following table gives their 

 mean dimensions ; in each case, except where otherwise 

 specified, this is the mean of from 15 to 20 bricks. The 



