60 Water-Soluble B 
studies have shown that there is degeneration of the motor cells 
of the cord in animals restricted to such a diet. We see, there- 
fore, reason to believe that our animals in the experiments de- 
scribed are actually in the polyneuritic state. When kept upon a 
diet of polished rice, young rats do not live long enough to become 
paralyzed owing to the deficiency of poiished rice in respect to 
protein, inorganic elements, and both the A and B (6). Rats 
suffering from polyneuritis are promptly relieved by the adminis- 
tration of the water-soluble unknown. When young rats are fed 
the diet of purified foodstuffs, plus butter fat, as employed in 
this work, polyneuritis occurs ordinarily between the 30th to 
the 40th day. 
Animals given a diet which is properly constituted iba that 
it is lacking in the fat-soluble A, may increase in weight during 
the first 2 weeks, then they gradually lose weight, become emaci- 
ated, and suffer from edema of the eyelids (15). We have re- 
cently (16) described this condition as xerophthalmia and believe 
it is due to this specific dietary deficiency. There is no loss of 
motor function in these animals. Their eyes are very much irri- 
tated and they scratch them and thereby cause mechanical in- 
jury. Recovery is prompt when the missing dietary factor is 
supplied, otherwise permanent blindness ensues and death super- 
venes, usually within 3 months. 
Since we have no chemical methods for the detection of either of 
these two unidentified dietary essentials we must employ biologi- 
cal tests as a guide to their isolation. The period required in 
order to make the test complete must extend over several weeks 
because of the ability of the more vigorous animals to grow to a 
slight extent during the first 2 or 3 weeks, even when the sub- 
stance to be tested for is absent from the diet. Preparations 
made with a view to separating the factor B in a pure form fre- 
‘quently led, if fed from the beginning of the experiment, to long 
maintenance and confusing data. This is due, on the one hand, 
to the difficulty, if not impossibility, of making complete ex- 
tractions, so that the residues still contain traces of the factor B, 
and, on the other hand, to the fact that in all probability solvents 
in which this substance is regarded as insoluble remove small 
amounts when the extractions are greatly prolonged. Such resid- 
ual amounts may well be wholly inadequate to cause recovery of 
