We are now ready to appreciate the misleading nature of the test which many investigators have employed in their efforts to determine the relative values of different foodstuffs, or of preparations made with the object of isolating and identifying the anti-neuritic substance. For this purpose the procedure of Funk has been almost universally followed. This consists of feeding pigeons on a dietary of polished rice until acute symptoms of impaired functional activity of the nerve cells appear. The substance to be tested is then administered and the recovery of the bird's powers of locomotion interpreted to indicate that the anti-neuritic substance has been made available - in other words that the test is positive. Failure of the bird to respond to this treatment is accepted as a negative test for this vitamin.

It was pointed out in 1918 by McCollum and Simmonds (28) that this test is inadequate and misleading, but no attention has as yet been paid by other investigators to this criticism of the classic "vitamin test." At that time they accepted the prevailing view that nerve degeneration was the most pronounced lesion in avian polyneuritis, and emphasized the probability that substances of a stimulating nature might whip up motor nerve cells which had lost their functional capacity, and might temporarily restore the birds to an apparently normal condition. They pointed out the high degree of improbability that so many totally unrelated chemical substances as pyrimidins, purins, and pyridins could fulfill the same physiological role. They were led to suspect that the "curative" substance was not, in some cases at least, identical with the indispensable nutritive principle, essential for growth and normal functioning, but rather bodies which possessed the pharmacological property of goading to renewed activity, nerve cells which had been injured and had ceased to functionate. The observations of Uhlmann (29) and of Dutcher, made later in the same year, that pilocarpin and other physiological stimulants are actually able to induce "cures" of pigeons in polyneuritis, supports this view. The findings of McCarrison, that the nervous symptoms in avian beri-beri are of secondary importance, further emphasizes the rationality of the view that the pigeon test, as ordinarily carried out, is utterly without value for the specific purpose for which it has been so widely employed.