l i:> 



The unavoidable mistakes caused by the rough measurements 

 in the mills and those caused by portions of the prepared 

 product sticking to the sides of the tubes through which 

 it passed, were as much as possible avoided by treating 

 large quantities — 160 and 180 K° — at a time. In spite of 

 this, it was still necessary to test the results of this weighing 

 by means of the laboratory-balance, in order to determine 

 the average weight of the different grains before and after 

 treatment. 



The following table I gives, in per centum, calculated 

 from the original weight of gaba, what was the product 

 obtained under the measures taken, after each manner of 



Refuse obtained 

 ) 6% Chaff (dedekpesak). 

 (gaba). )17% dedek sedeng. 



Unpolished rice . . 

 (pitjah koelit). 



White rice 



(bras sosok). 



The results were for both kinds of rice practically the 

 same, in consequence of identical adjustment of the different 

 parts of the machine, except that during the first treatment, 

 the larger Boeloe grain produced slightly more refuse than 

 the Idjoean rice. This first refuse was roughly divided by 

 means of a winnowing-fan into light chaff (dedek pesak) 

 and into the heavier, coarsely-powdered bran (dedek sedeng). 

 By a slightly different adjustment of the machine during a 

 former treatment, practised on white rice destined for the 

 army, this riee did not give 2°/ but 4% refuse, consisting 

 of the finest kind of bran (dedek bekatoel). 



Although the difference in adjustment of the respective 

 parts of the machine naturally influences the given percentage, 

 this may nevertheless be considered as pretty well the 



LXXIV 10k * 



