62 Water-Soluble B 
It is, therefore, possible to exhaust beans successively with ether, 
benzene, and acetone, and then to extract the physiologically 
active substance with alcohol. If the alcohol is then evaporated 
upon some substance, as dextrin, to distribute the alcohol-soluble 
material over a large surface, the desired substance can then be 
extracted with benzene (17). It is obvious that by this procedure 
and with the introduction of still other solvents the substances 
which accompany the physiologically active compound can in 
great measure be eliminated. By the introduction of precipita- 
tion methods at that poimt where interfering substances have 
been eliminated as far as possible, the chances of success in the 
preparation of a pure product or the preparation of derivatives, 
and with a minimum of Joss in manipulation, are much greater, 
we believe, than in the methods of procedure hitherto employed 
by others. 
In the experimental part of this paper it is shown that the 
water-soluble B is not extracted directly from beans, wheat germ, 
or pig kidney by ether, benzene, or acetone, but is readily ex- 
tracted in great part by alcohol. After being removed by alcohol 
it is shown to be soluble in benzene, but very slightly soluble in 
acetone. The probability that there should be two or more 
physiologically indispensable substances in what we term water- 
soluble B, both or all of which should show the same solubility 
relations with three solvents, is relatively small and lends sup- 
port to our view that the substance which protects animals 
against polyneuritis is the only essential complex in the extracts 
described. In other words, the data support the view that there 
are no specific substances present in these extracts which pro- 
tect against such diseases as scurvy, rickets, pellagra, sprue, etc., 
and tend to confirm our contention that the latter are not due to 
specific starvation as is the case with beri-beri and xerophthal- 
mia (16). 
In interpreting the charts it should be remembered that at the 
point marking the beginning of the second period the animals 
were at or near the point of showing loss of muscular control, 
and that response with even a slow rate of growth on the ad- 
dition of a preparation is conclusive evidence of the presence of 
the physiologically active substance, water-soluble B.! 
~ 1 Credit is due to Mr. H. Steenbock for the preparation of the extracts 
employed in this work. j 
