— 184 — 



a watch glass, previously weighed, or in a small evaporating 

 basin the weight of which is known. To avoid the danger 

 of the spirit creeping up over the edge during this evaporation, 

 care must be taken that the level of the liquid remains 

 witbin the circle enclosed by the ring of the water bath. 

 The extract in finally dried during a quarter of an hour at 

 a temperature of 100 — 105°C. and then weighed after cooling 

 in the exsiccator, after which a simple calculation shows just 

 the amount of dry-residue which would have remained from 

 25 ccni., and afterwards the percent cipher of that 

 dry-residue. 



All these manipulations do not need to last altogether 

 longer than 1 or 2 hours, while a P 2 5 estimation in rice 

 takes at least three, generally 4 days; sometimes it even 

 lasts much longer. 



This manner of testing does not only take up very little 

 time, but it is also very simple, — it requires so little work, 

 that even a layman would without difficulty be able to carry 

 it out — and the expenses are scarcely worth reckoning. 

 But besides this, it has appeared that this test makes it 

 possible to judge of the protection which a certain kind 

 of rice affords against beri-beri, better than can be done 

 by a P 2 5 estimation, — a circumstance which from the nature 

 of the matter gives the precedence. 



The results obtained by this new test, even during its 

 first applications, raised high expectations. Later investiga- 

 tions have certainly shown that some of these advantages 

 were illusory, and have also brought to light a few drawbacks 

 in connection with the spirit-proof, but, notwithstanding all 

 this, the new test has not been so much disparaged that it 

 would have to be considered, generally speaking, as inferior 

 to the P 2 5 test. 



So the first comparisons between the P 2 5 amount and 

 the dry-residue cipher of some kinds of rice seemed to show 

 the superiority of the spirit-proof, in so far that hereby 

 greater differences between unpolished rice on the one side 



