METHODS OF ASSESSING B VITAMIN REQUIREMENTS 249 



cal. Whether or not the finding that thiomalic and thiolactic acids afford 

 no protection against the latter but seem to promote the fatty infiltration 

 actually supports this belief remains to be seen. 



It is seldom as simple as it might appear to those unfamiliar with the 

 field to exclude one factor from the diet without simultaneously omitting 

 another. 19 In recent studies using dogs, a diet apparently complete in all 

 necessary factors but biotin produced paralysis and death. Biotin, more- 

 over, appeared to prevent these effects. 20, 21 Later work 22, 23 showed, 

 however, that the symptoms were primarily due to a potassium deficiency. 

 While biotin produced some temporary responses for several hours, the 

 protective effect of a single adequate dose of potassium lasted for six to 

 ten weeks. Similar cases are well known throughout the history of the 

 assessment of B vitamin requirements by depletion methods. 



It is only in recent years that it has been possible to apply this type 

 of study directly to humans. This is due to a considerable extent to the 

 difficulty in controlling a multitude of variables which must be controlled 

 in order to obtain significant data. Realizing this, the National Research 

 Council's Committee on Diagnosis and Pathology of Nutritional Defi- 

 ciencies has outlined certain of these factors which they feel to be of the 

 greatest importance. 24 While primarily designed for assessing conditions 

 of deficiency, these factors apply in every sense to the assessment of 

 requirements in animals as well as humans. Briefly they are: (1) the 

 adequacy of the criterion for determining the nutritional status; (2) the 

 nutritional status previous to the experimental period; (3) the diet prior 

 to and during the experimental period; (4) the conditions which influence 

 the relationship between the supply and requirements; (5) the indices set 

 by the observer to measure the criteria; (6) the initial status of the sub- 

 jects with regard to growth, physical performance, resistance to disease, 

 etc.; (7) the method of selection of subjects; (8) the number of subjects; 

 (9) the nature and potency of supplements; and (10) the length of the 

 experiment. 



In addition to these factors, adequate controls are essential. This is 

 particularly true in human experimentation when psychic factors may be 

 magnified. 25 The supplement must be matched in such control groups with 

 placebos, indistinguishable from the supplement in taste and appearance. 

 Not only the subjects, but equally important, the observers, must be 

 ignorant of which individual receives a placebo and which a supplement. 

 Even in experimentation designed to determine the essentiality of a nutri- 

 tional source of some B vitamin, it is necessary to observe the factors 

 outlined above. In a recent study of the effects of a vitamin B 6 deficient 

 diet in humans, 26 almost no definite conclusions could be reached because 



