18 JOURNAL, R.A.S. (CEYLON). [VOL. XI. 



the same day for want of a lucky hour, a small dam (liyadda) 

 in the field which is not damp is selected, the cocks (keta- 

 kandu) are removed there, and heaped into ricks or stacks 

 ( wata-malu). 



At the lucky hour for threshing (which always occurs in 

 the evening) the paddy is taken from the wata-malu and tied 

 into small bundles, and carried on the head to the kamata. 

 Sometimes the paddy is put on the kamata in semi-circular 

 shape, or, as the natives say, " like a rainbow or half-moon " 

 ( rena-kanda-ekatukaranawd ) ; and sometimes heaped in the 

 centre (koleta-ekatukaranaivd) after a diagram has been 

 drawn with ashes on the threshing-floor.* 



The paddy is then tossed and mixed together with hands 

 (kola-wadanawa). But if it is a large stack two " teams " of 

 buffaloes ( dekerella ) are yoked together to thresh the paddy : 

 otherwise one team suffices. The boys who are employed to 

 drive the buffaloes over the paddy (kola-madawanawd, 

 goyan-madawanaivd ) are first fed. Should the animals void 

 their dung whilst being driven the boys are taught to take a 

 little straw and hold it on both the upturned palms of the 

 hands to the sterns of the buffaloes (to prevent the dung falling 

 on the straw and fouling the paddy), and to throw it over the 

 other side of the wdriya-kanda. They are very particular in 

 this respect until the whole threshing is over. 



Two or three pe after first driving the buffaloes over the 

 paddy, what has worked to the edge of the threshing-floor is 

 tossed with the deti-goyiyd, or flail, on to the heap in the centre 

 ( wata-kadanawa J, and the buffaloes again driven over it. 



After threshing for some time several men (according to 

 the number of men employed), picking up a little straw on 

 their flails, and placing them on their shoulders, walk round 

 repeating the kalawiti-kima or threshing-floor refrain ("Deti 

 deti kumana deti" &c), whilst the buffaloes are moving. 



* Plate I. — This diagram differs from that given by Mr. levers chiefly in 

 having only three concentric circles instead of seven. Three seem so 

 universally the number drawn that it is possible Mr. levers may have 

 been misled in making his diagram contain more circles. 



