256 THE BIOCHEMISTRY OF B VITAMINS 



report that pyramin excretion increases exponentially with increases in 

 thiamine intake, approaching linearity at normal levels, as contrasted 

 with linear increases in thiamine excretion throughout the range studied. 

 Moreover, pyramin excretion remains measurable at low intakes of thi- 

 amine, when thiamine excretion falls to essentially zero. For these reasons, 

 pyramin would seem to be a better end product to study than thiamine. 



Comparing the excretion technique with others, Berryman et al. 39 find 

 that urinary excretion of B vitamins falls off rapidly following the change 

 of men to a deficiency diet, but that fecal levels remain unchanged. 

 Decline in physical and mental states are more gradual, and these states 

 improve more slowly when the vitamin supply is increased than do the 

 urinary excretion levels. The great individual variability in response was 

 again noted by these workers. Still other studies 40> 41 indicate quite 

 clearly that a decreased dietary intake is rather rapidly followed by a 

 decrease in urinary output, followed only gradually by decreases in tissue 

 content, and only much later by the manifestations of a clinical deficiency. 



The most advantageous use of excretion studies in the assessment of 

 vitamin requirements comes about in those cases in which, unlike thi- 

 amine, other methods of approach to the requirement are less feasible. 

 The riboflavin requirement is an example of this situation. Hagedorn, 42 

 in a study of prison inmates, found great variability in riboflavin excre- 

 tion (0.05-2.4 mg/24 hrs) and in the retention of test doses without 

 apparent cause, and experienced further difficulty in obtaining useful 

 results from fasting subjects due to the large errors in measuring such 

 low riboflavin concentrations as were obtained. Despite the apparent 

 advantages of dietary control, prison inmates have seldom been a source 

 of entirely satisfactory data, and other workers have obtained better 

 results with more satisfactory subject material. Oldham et at. concluded 

 from a study of the riboflavin excretion of institutionalized children 43 

 that 1.15-1.6 mg/day of riboflavin met their nutritional requirements. 

 Other workers, by studying excretion at different levels of intake and the 

 per cent of a 3-mg test dose of riboflavin excreted in 24 hours at each 

 level of intake, decided that 1.3-1.5 mg/day is adequate for an intake of 

 2100-2300 Calories. 44 Oldham et al., in their earlier studies, estimated 

 that the riboflavin nutritional requirement was about 1 mg/day or 0.50 

 mg per 1000 Calories. A summary of their results is given in Table 6. 



All these results on riboflavin agree well with the earlier observations 

 of R. D. Williams et al., 46 but the requirements arrived at are less than 

 those proposed by Sebrell et al. i7 It is noteworthy that nearly all workers 

 are in agreement that fecal riboflavin does not vary with the intake. 



Briggs 48 studied two subjects who had been pellagrins, and who were 

 placed on a "corn-poor" diet containing about 2.4 mg of nicotinic acid 



