Harris and Leong (1), in 1936, suggested that the state of vitamin B1 deficiency in man could be detected by measurement of the amount of the vitamin excreted in the urine in the twenty-fourhour period following a single oral test dose of 900 micrograms of thiamine chloride, and since then many other procedures have been suggested for the detection of vitamin B1 deficiency by means of a test dose (2 to 8). Oral and parenteral routes have been used for administration of test doses varying from 1 to 10 mgm. of thiamine chloride and various criteria have been set up for estimation of the state of the subject's nutrition with respect to vitamin B1.The value of a laboratory test for vitamin B1 deficiency lies largely in its application to the detection of those cases in which a chronic mild deficiency may have existed for some time but in which cleancut symptoms or signs of deficiency are not evident. The urinary excretion of thiamine under various conditions has been used as a suggestive index of the state of thiamine nutrition but, unfortunately, most of the control tests reported in the literature have been made on persons partaking of their customary diets of unknown or unspecified vitamin B1 content, or on hospital patients. We believe that it is necessary to determine the response to the test of individuals maintained on various levels of thiamine intake for varying periods of time in order to assess the validity of the test. Only in this way can the response be correlated satisfactorily with the nutritional history and clinical state.We are prompted to report our experience with test-dose procedures because we have had the opportunity to study the responses of fifteen female subjects maintained on rigidly controlled diets of known thiamine content for intervals varying from thirteen weeks to five months. Factors of the vitamin B complex other than thiamine were supplied in purified form or in the form of autoclaved yeast; therefore, the dietary restriction was limited as nearly as possible to thiamine alone. Since the duration and degree of deficiency were known accurately, we believe that the value of test-dose procedures or of the ordinary twentyfour-hour excretion of thiamine as a laboratory aid can be accurately assessed.
EXPERIMENTALThe present study is part of a larger one on induced thiamine deficiency in human subjects. The selection of subjects and the diets utilized in this study have been described in previous reports (9,10,11 Table I shows the values of thiamine excretion in two groups of four subjects who were maintained on a diet which contained not more than 150 micrograms of thiamine per day. The first group (Subjects 1 to 4) received the test dose after 155 days and again after 169 days at this intake level. Symptoms of severe deficiency of 247 HAROLD L. MASON AND RAY D. WILLIAMS thiamine were evident on both occasions. The results of the test were essentially the same in both instances. The excretions of thiamine in the twenty-four hours preceding the test were 11 to 26 micrograms.1 In the tw...