( 912 ) 
anology I conclude that Kadjang hidjoe *) will have a similar preventing 
influence on the formation of starch. *) 
Moreover it is by no means indifferent whether these substances 
are mixed with the rice before it has been cooked, or after the 
cooking has taken place. The starch that has been formed already *) 
cannot be precipitated, either the starch that is in process of forma- 
tion is precipitated, or its development is prevented. 
It is very remarkable that the natives, when left to themselves, 
never cook rice but steam it, whilst the steaming is variegated by 
washing, the consequence of which must be that the starch dis- 
appears, is washed away. Rice for prisoners and soldiers on the 
contrary is cooked, and though one tries afterwards by evaporation 
to give to the grains of rice the dry appearance that steamed rice 
has, yet the starch remains in the food. The first group is 
consequently guaranteed against Beri-beri, the other exposed to it. 
The danger augments considerably, if one eats by preference cooked 
or steamed rice, after it has entirely cooled down. On such rice the 
acetifying bacteria of the air have fallen down, not only does it 
smell sour, but it obtains that agreeable flavour reminding of fruit, a 
consequence of fermentation that is highly praised by gastronomers *). 
Food containing amylum in which few substances are found that 
form starch guarantees against Beri-beri, and there exists moreover a 
sort of starch from amylum that ferments, but through which only 
very little acid is formed. So our bacillus very easily produces 
water and gaz from starch of potatoflour, but by this process only 
a very small quantity of acid is formed. Consequently it is not strange 
that with potato-flour EiJkMaN cannot excite Beri-beri ®). 
A controversy arose between ElskKMAN on the one hand and Grins 
with Horst on the other, whether Beri-beri can occur when feeding 
with sterilized meat. J have not repeated these experiments but I 
draw the attention to the fact that our bacillus grows vigorously on 
1) ROELFSEMA, GRIJNS, HursnHorr-Por. 
2) The influence of Kadjang hidjoe and rice-bran may however be quite different, 
see beiow. 
3) For this reason the extract of dedek, or the salts found in dedek added to 
rice that has already been cooked, will not prevert Beri-beri (GRIJNS contra E1JKMAN). 
') Tamil rice on the contrary which protects against Beri-beri stinks (FRASER, 
STANTON). It is submitted to a treatment by which the starch-substances are 
lixiviated, by this process the fermentation that gives an agreeable flavour is rendered 
impossible. 
>) Sterilized potato-flour ferments somewhat quicker. 
6) Nor is it strange that neither this flour nor gaba can absolutely protect 
against Beri-beri (GRIJNS). 
