l TH - URINARY EXCRETION OF RIBOFLAVIN OF COLLEGE WOMEN ON A BELT-“CHOSEN DIET AND THE RESPONSE TO A SATURATION DOSE OF RIBOFLAVIN AND THIAMlNE Thesis for the Degree of M. 3. MICHIGAN STATE COLLEGE Ruih Louise lngalla I945 THESIS° This is to certify that the thesis entitled Urinary Excretion of Riboflavin or Coilege “omen on a coir—Chosen List eno the Heaponse to a saturation Lose of Fibofiavin ano Thiamine. presented by Ruth Louise Ingalls has been accepted towards fulfihnent of the requirements for M. 8. degree iWutx'ition ' ‘dinLaA4fllALjL{TIEBLQ;H?71\~‘ Majoj professor Datebeptember l, 1345. URINARY EXCRETION 0E RIBOELAVIN OE COLLEGE WOMEN ON A.SELF-GHOSEN DIET AND THE RESPONSE TO.A SATURATION DOSE OF RIBOFLAVDI AND THIALIIUE by RUTH LOUISE IFGALLS ‘4 '3‘"— A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfilment of the requirements for the degree of MASTER OF SCIENCE Department of Foods and Nutrition 1945 THESIS The writer wishes to eXpress her appreciation to Dr. Margaret A, Ohlson for her interest, supervision and helpful suggestions, to hiss uilma D. Brewer for her valu- able assistance, encouragement and advice throughout the study, to Dr. W. D, Baten for help with the statistical analysis of the data and to all the students for their willing cooperation in the study. 1 EFL”: ‘22“ ‘3 4.. ‘ 3J‘.J O. ' TABLE 9;; commms Page Introduction ........................... 1 Literature ............................. 5 Experimental Procedure ................. 15 Results ................................ 21 Discussion ............................. 33 Summary ................................ 42 Literature Cited .QOOOOOOOOOOOOOOOOOOOO. 44 LIST 9;; TABLES Table Page 1;. Range of Urinary Excretion of Riboflavin Reported in the Literature as hormal ......................... 8 2. Stability of Urinary Riboflavin .............. 17 5. Urinary Excretions of Riboflavin by College Women 22 4. Urinary Excretions of Riboflavin ............. 23 5. Summary of Average Riboflavin moretions OOOOOOOOOOOOOOOOOOOOOO0000000000coo 24 5. Range of Riboflavin Excretion Found by Grouping the Data ................... 25 7. Average Daily Dietary Intakes of Protein, Thiamine and Riboflavin ............. 29 8. Urinary Excretions of Riboflavin Calculated to 70 Kilograms Weight ............ 34 LIST 93 FIGURES Figure Page I. II. III. Daily Dietary Intake and 24-Hour Urinary Excretion of Riboflavin ................ 30 Percentage of Saturation Dose Excreted in 24 Hours and Daily Dietary Intake of Riboflavin ................... 31 Urinary Excretion of Riboflavin during 24-Hour Periods before Test Dose and Percentage Excre- tion of Test Dose in 24-Hours .................. 37 Urinary Excretion of Riboflavin during 24-Hour Periods before Test Dose Recalculated for 70 Kilograms Weight and.Percentage Excretion of Test Dose in 24 Hours .0.000000000000000000000000000000000000000 59 URINARY EXCRETION OF RIBOFLAVIN OE COLLEGE WOMEN ON A SELF-CHOSEN DIET AND THE RESPONSE TO A.SATURATION DOSE OF RIEOELAVDT AND TEIIAI-JIIIE INTRODUCTION Riboflavin has been known as an essential dietary factor for a comparatively short time. Though early seen as a fluorescent substance in whey (Blyth, 1879), recog- nition of this vitamin as part of the B complex came within the last decade (Gydrgy, '35). Now it is conceded that riboflavin functions in the oxidation and reduction system as a catalyst in the energy transfer of phosphorylated hexoses (Quastrel and Webley, '41). The importance of the vitamin to man became evident when symptoms were observed which were due directly to a dietary deficiency of riboflavin (Sebrell and Butler, '39). This has led to investigations of man's requirement of riboflavin and to experimental studies attempting to meas- ure deficiency and saturation as evidence of riboflavin status (Axelrod, Spies and Elvehjem, '43). These studies were given new impetus when the National Research Council in 1941 attempted to establish standards of vitamin re- quirement. Urinary excretion of riboflavin has been used as an index to riboflavin nutrition. The riboflavin excreted in a 24-hour urine sample has been used generally as the cri- terion of riboflavin status (Ehmerie, '36; Eerrebee,h'4l; 2 Williams, and others, '43), but the response to a satura- tion dose of riboflavin has been found.more satisfactory in some cases (Gybrgy, '42). The micrograms of riboflavin ex- creted per milliliter of urine in the one-hour sample after a night of fasting has been found by one investigator (Feder, Lewis and Alden, '44) to give good agreement with the micrograms excreted per.milliliter during 24 hours. The purpose of this study was to determine the range of daily riboflavin excretion of college girls on a self- chosen diet, the correlation of the 24-hour sample with a one-hour fasting sample, and the response of each girl to a saturation dose of riboflavin. Since thiamine and ribo- flavin are thought to be closely related in animal meta- bolism, the effect of a saturation dose given simultaneous- ,ly as it affects the riboflavin excretion was studied. An- other investigator from.this laboratory will report on thiamine excretion and the response to a saturation dose of thiamine. L Imam-1mm Riboflavin Excretion Studies The excretion of riboflavin does not depend directly on intake, but is complicated by many factors. It has been demonstrated that urinary excretion parallels intake (Se- brell, and others, '41; Williams and co-workers, '43; Naj- jar, Holt and others, '44) and that fecal excretion remains nearly constant at varying intakes (Najjar and others,'44). The total urinary and fecal excretions, however, plus the small amount excreted in perspiration (Tennent and Sil- ber, '43) nermally represents only part of the total in- take. Sebrell and others (1941) in their studies with humans noticed that with increasing riboflavin intakes there was an increase in the amount of riboflavin not ac- ‘counted for when the amount excreted was totaled. This was more pronounced at the higher intakes. That is, the ribo- flavin which could not be measured in the excretions was much more at an intake of 0.11 milligram.per kilogram.than at 0.085 milligram.per kilogram. Each additional dietary increment was accompanied by a greater proportion of unmeas- ured riboflavin. Even at 0.025 milligram per kilogram.in- take there was some unaccounted loss. Sure and ford (1943) fed rats 1000 micrograms ribo- flavin and thiamine for 30 days. At the end of that time, 4 the bodies of the animals were assayed for these vitamins. The entire body of an animal contained only 783 micrograms riboflavin. On the same intake the daily excretion of the animal showed a utilization of 76 to 88 per cent of the amount ingested. In vitro studies made at the same time demonstrated a destruction of riboflavin in tissues (liver, heart, lung, stomach, large and small intestine). When 1000 micrograms riboflavin were incubated with these tissues at 37° centigrade for 24 to 48 hours, three to 28 per cent destruction occurred. These workers concluded that there is destruction of riboflavin by the tissues, perhaps after the tissue has made use of the vitamin. Another factor to be considered when measuring ribo- flavin excretion is the possibility of bacterial synthesis. Najjar and others (1944) showed that bacterial synthesis does occur in the large intestine and at times may be enough to lower dietary requirement. When urinary excretion is used as an index of ribo- flavin status, the loss of part of the intake through tis- sue destruction and the augmentation of the vitamin supply through bacterial synthesis should be remembered. Other factors, shch as poor absorption, the effect of impaired kidney function, and the influence of the other vitamins or of any of the nutrients also may have a bearing on the problem (Seyle, '43; Unna and others, '44; Mannering, Orsini and Elvehjem, '44). Thus, the desirability of using urinary excretion 5 alone to measure the adequacy of riboflavin in the diet of the individual has been questioned. Axelrod, Spies and Elvehjem (1941) found that the blood of twenty normal in- dividuals contained an average of 0.42 micrograms ribo- flavin per milliliter of blood while the blood of deficient patients contained 0.40 micrograms per milliliter. Strong, Feeney and hoore (1941) found human blood to contain 0.5 micrograms per milliliter. Rajjar and others (1944) found that fecal excretion of riboflavin remained at 200 to 600 micrograms per day even when riboflavin was injected. These findings lead to the conclusion that riboflavin sta- tus cannot be measured by analysis of fecal excretion of riboflavin or of riboflavin concentration in the blood. The early investigations of riboflavin status were made using individuals with what appeared to be marked de- ficiency. Oden, Oden and Sebrell (1939) noted cases de- scribed as ariboflavinosis and recorded the symptoms. Axelrod, Spies and Elvehjem (1941) also described ribo- flavin deficiency in hospital patients. Both investigators cured the syndrome with a therapeutic dose of three to five milligrams of riboflavin. Sydenstricker, hruse and co- workers (1940) cured certain ocular changes with four to 10 milligrams of riboflavin. Sanstead (1942) failed to effect a cure with the same dosage. Recently, Machella and McDonald (1943) attempted to remove symptoms ascribed to ariboflavinosis by riboflavin therapy and failed. The work with cases of extreme deficiency symptoms probably is not 6 comparable to what has been described as borderline defi- ciency. Since urinary excretion follows intake closely, the individual's daily urinary excretion may be used as a meas- I ure of dietary intake (Williams and co-workers, '43). fhe amount excreted varies from day to day with variations of intake but, in general, remains within a certain range for the individual. This range of excretion should give an in- dication of the adequacy of riboflavin in that individual's diet. It also has been found (Feder, Lewis and Alden, '44; Oldham and others, '44) that the riboflavin in a one-hour fasting sample and the four-hour and the 24-hour samples .after a saturation dose may be used as a measure of nutri- tional status in regard to riboflavin. Excretion of less than 0.3 microgram of riboflavin in the one-hour sample or of less than 35 per cent of the saturation dose was inter- preted to indicate deficiency. As early as 1936 an attempt was made to find the range of urinary excretion of riboflavin in apparently normal in- dividuals when Emmerie reported the average daily excre- tion of males to be 819 to 1250 micrograms per 24 hours. Helmer (1937), using the rat-assay method, estimated ex- cretions of 120 to 175 Sherman-Bourquin units or 360 to 875 micrograms per day. The methods used were questionable but the findings are interesting when compared with more recent data. Ferrebee (1940) found the range of excretion to be 700 to 1700 micrograms. Axelrod, Spies and Elvehjem (1941) 7 found 477 to 835 micrograms or an average of 625 micrograms excreted by two patients on a regular hospital diet. At the same time Sebrell and others (1941) found the range of excretion in eleven young men to be 234 to 1740 micrograms daily. Strong, Feeney, Moore and Parsons (1941) found 500 to 800 micrograms excreted by wamen. Klopp, Abels and Rhoads (1943) found 210 to 1200 micrograms per 24-hour urine sample. Williams and others (1943) recorded the range of urinary excretion compared directly with intake. When riboflavin intake was one milligram per day or less, 150 to 350 micrograms were excreted. At an intake of 1.6 milligrams per day excretion was raised to 400 to 700 micrograms. When 3.6 milligrams was the total intake (diet plus supplementary administration of the vitamin), the excretion was 1600 to 1800 micrograms. Feder, Lewis and Alden (1944) stated that consideration of these data sets the range of urinary excretion at 500 to 1000 micro- grams with an average of 800 micrograms when the riboflavin intake meets the National Research Council standard of two to three milligrams per day. These data are summarized in table 1. Some investigators have used the one-hour fasting sample rather than the 24-hour sample as a measure of ribo- flavin status. Feder, Lewis and Alden (1944) found that the micrograms per milliliter in the fasting hair sample agreed well with the per milliliter excretion in the cor- responding 24-hour sample. Feder also stated that when emueaago spa: sense saaea Isomsaeupsa empoew InH on 0.0 nadehspdm amputee“ eased“ Ionwn aenmaaafie use no emoe soapeuspew mama Isa emHHonpqoo hp oopmasmmn soapenoan qumfieeo Hmpwmmos so moaeopw mvsefiuea Hepwmmos no meaesvm mpnmfipem Hmpfim Imon no soapwbnowpo nee munch no hespw emvoen Ina aehmwaafie one we mmoe soapmnspwm epwpm mnapmem ea swamp pom we: soapmnoxo upon one pnmaaoo mmlmm mm mmIom Anon H mason em mason a ZOHB>.HIO no emanm>w oom e.¢Haue.a oooauoom we. .noeoe "' "' Ed. hands Imam: one monounem In- oomauoma me. .mesaaafie III cowauoaa we. .mmesosm .mH094 «macaw III oomuoom He..ueomnem .maoos «hoaeeh ammonpm III oesHIemm Ha. .Hamsnom III owenepe mum He. «amn£m>am mnmIeee ens moaam .eoeflmxe In- ooeHIooa He. .omnmsnme were II: He. .esnuez III mamuomm em. .nmaamm omuom ommanoam em. .oawmsam .m: .m& madame“ nsomIono esomlem hopemapmobnn zOHBmmoxm szdqhomHm ho mcz¢m Hashes we easaenepaa esp ad vouuoaeu nabaamonan mo soapenoxo madman: mo omswm H mamas D 9 riboflavin excretion falls below 0.3 micrograms per milli- liter in the fasting hour, the individual may be considered deficient. The values found in that study for the fasting hour sample range from 0.00 to 1.77 micrograms per milli- liter. The range in total micrograms per hour was 1.4 to 114.4 micrograms for four subjects. Najjar and.Holt (1941) found six to 49 micrograms excreted during the fasting hour. The excretion of a saturation dose of riboflavin dem- onstrates the ability of the body to excrete any excess of the vitamin. It is thought that the amount of the test dose excreted depends on the individual's need for the vitamin. The amount of the test dose has varied as has the method of administration. Klein and Kohn (1941) found 25 per cent of two milligrams of riboflavin excreted. Axel- rod, Spies and E1vehjem.(l94l) injected 300 to 400 micro- grams riboflavin per kilogram.body weight intravenously and found no correlation between the per cent excreted by nor- mal and by deficient subjects. Najjar and Holt (1941) in- jected one milligram.intravenously. Individuals having an adequate dietary intake retained 32 to 72 per cent of the test dose at the end of the four-hour period, while defi- cient individuals retained 81 to 93 per cent. Ferrebee and weisman (1943) injected one milligram.ribof1avin and one milligram thiamine. They reported 30 to 35 per cent of the test dose of riboflavin excreted in one hour and values for the two, three and 24-hour response in agreement with the 10 data of Najjar and.Holt. Feder and co-workers injected 0.016 milligram per kilogram.by the intramuscular route and found that an excretion of less than 35 per cent of the test dose indicated deficiency; Oldham and co-workers (1944) gave a test dose of 75 micrograms per kilogram oral- ly to two normal children and found 23 to 32 per cent re- spectively excreted in 24 hours. In summary of these studies, the 24-hour excretion of riboflavin by normal subjects has been found to average 500 to 1000 micrograms. The amount found in the one-hour fas- ting sample varies widely, but less than 0.3 micrograms per milliliter seems to indicate low storage of the vitamin,and less than 25 to 35 per cent excretion of an injected satura- tion dose has been interpreted to denote the same thing. However, the application of the last figure to this study is uncertain,since the only response to an oral test dose reported in the above literature was in the study with children. Very little is known of the relationships existing between the vitamins of the B complex. There has been some study of the interrelationship between thiamine and ribo- flavin. This work has been chiefly with animals. Sure (1944, 1945) noted that riboflavin excretion in rats is in- creased two to three times in chronic thiamine deficiency, but found no difference in riboflavin content of tissues or in absorption. Singher, and others (1944) found a 50 per cent increase in livers of thiamine-deficient animals. ll Sure (1945) maintained this to be transient riboflavin found during digestion and absorption. Ferrebee and weiss-' man (1943) could find no difference in the riboflavin excre- tion of human subjects with thiamine deficiency. Following an injection of one milligram of each vitamin the ribo- flavin response was similar to the response reported by Najjar and Holt (1941), who injected only riboflavin. Klopp, Abels and Rhoads (1943) found an increased excretion of riboflavin in human subjects after an injection of thia- mine. Continued daily injections, however, did not produce a riboflavin deficiency. At the present time the analysis of riboflavin excre- tion in the urine appears the most satisfactory method of judging riboflavin status. The range of riboflavin excre- tion in 24 hours, the amount of riboflavin excreted in a one-hour fasting sample, the response to a saturation dose of the vitamin, the effect of a saturation dose of thiamine on this response, and the existing relationships between these factors have not been established conclusively. This study has attempted to correlate these scattered data by the investigation of all these techniques within one experi- ment. Evaluation of Chemical Method The choice of a method for any routine analysis de- pends on the accuracy and convenience of administration of 12 that method. In vitamin assay there are three types of method used, biological, microbiological and chemical. The last two of these have been used successfully in analyses of riboflavin in the urine. The microbiological method of Snell and Strong (1939) has been used by many workers (Axelrod, Spies and Elvehjem, '41; Strong, Feeney, Moore and Parsons, '41; 01dham.and co-workers, '44). The results of the fluorometric method of riboflavin determination com- pared well with the microbiological (Najjar and.Holt, '41). The fluorometric method depends on the fact that pure riboflavin in aqueous solution shows a yellow-green fluo- rescence which can be measured in a fluorOphotometer. It has been used for the estimation of riboflavin in food (Hand, '39; Hodson and Norris, '39; Connor and Straub, '41) and in urine(Najjar and Holt, '41; Ferrebee, '40; Sure and Ford, '42; Keys and co-workers, '44). The accuracy of this method depends on the removal of certain interfering fac- tors, so that the fluorescence measured will be due to ri- boflavin alone. Some of these interfering factors were found to be the pigments in the urine, turbidity of the so- lution, the formation of gaseous emulsions during the ex- traction process and the scattering of light rays through the solution (Najjar and Holt, '41). The pigments found in normal urine are urochrome, urobilin and uroerythrin. The last of these is non- fluorescent. The turbidity of urine is due to a precipita- tion of calcium.phosphate in urine voided directly after a 13 meal, also to an alkalinity resulting from the decomposi- tion of urea to form ammonia. Other sedimentary substances also are present in varying amounts (Hawk and Bergeim, 1942). Some of these are easily filtered out of the urine, but some remain in the sample. Various methods have been used to remove these substances. Sure and Ford (1942), us- ing a modification of the method given by Hodson and Nor- ris (1939), oxidized the interfering substances in the fil- tered urine sample with potassium.permanganate and sub- tracted a blank reading from the sample. Several workers have used the adsorption technique to remove the pigment and to concentrate the riboflavin in urines of low value. Emmett and.McKim.(l9l7) found that fuller's earth adsorbed the vitamin freeman of yeast. Narayanan and Drummond (1930) and Salmon and co-wonkers (1928) also showed that fuller's earth adsorbed certain of .the B vitamins. Supplee, Bender and Jensen (1939) and Con- ner and Straub (1941) demonstrated the use of adsorption in the estimation of riboflavin in food. Ferrebee (1940), Najjar and Holt (1941) , and Keys (1944) have used it in the riboflavin analysis of urine. Since all interfering substances are not removed by adsorption and by oxidation with potassium.permanganate, a blank reading has been found advisable. The blank has been obtained by various methods. Sure and Ford (1942) reduced the riboflavin in the sample with sodium hydrosulfite after the sample reading had been taken. This reduces the ribo- l4 flavin to a non-fluorescent leuco form which may be re- oxidized by atmospheric oxygen (Hodson and Morris, '42), thus introducing an error in the blank reading (Uajjar and Holt, '41). Ferrebee (1940) used a constant blank obtained by passing pyridine-acetic acid solution thraigh the ad- sorption columns and then treating this solution as a sample. Najjar and Holt (1941) prepared blanks by exposure of the sample or of a duplicate sample to light, using ei- ther a mercury vapor lamp or direct sunlight for one to two hours. {eys (1944) irradiated a duplicate of each sample for one to two hours under a mercury vapor lamp at an acid pfl. Deflerre and Brown (1944) found complete destruction of riboflavin at all pH values upon exposure of the solution to strong daylight. The adsorption technique of ferrebee with the modifi- cations suggested by Keys appear to eliminate most of the interfering substances in urine and to give accurate re- sults as judged by the recovery of added riboflavin. MEMBER-EAL 131:0 0131mm Subjects The urinary excretion of riboflavin of 20 junior and senior college women was studied. The subjects ate their customary dict obtained at the college dormitories or at the Home management houses. All the subjects were in ap- parent good physical health at the time of the study. The importance of accuracy in making collections was impressed on the subjects; each kept a record of time of collection, noting any error made in collection of a.sampla and re- ported regularly to the laboratory. Each subject kept a record of her diet during the five-day collection period.' This explained any marked deviation in riboflavin excretion and at the same time gave a pattern of the subject's dieta- ry habits. In addition, a history of past illnesses, food likes and dislikes and general information concerning the health and habits of the subject were obtained. The length of the experimental period was five days for each subject. This was divided into two periods, a pre-saturation period and a period of saturation response. Twenty-four hour urine samples were collected for three days. On the fourth day each subject was instructed to drink a glass (200 cc.) water on arising. At the end of an hour a fasting urine sample was obtained. At this time the subject was given three milligrams riboflavin and three 16 milligrams thiamine in aqueous solution, orally, as a satu- ration test dose. Urine was collected for the four hours after the saturation dose and kept separate from the col- lection for the remaining 19 hours. The urine for the fol- lowing 24 hours also was colledted in order to observe any delayed response to the saturation dose. Collections were made directly into amber jars with screw tops containing five milliliters glacial acetic acid as a preservative. The subjects were instructed to keep the containers in the dark and away from radiators. Each sample was returned to the laboratory and refrigerated as soon as possible after each period. Chemical Method Each 24-hour collection was measured, made up to a volume of 2000 milliliters (2500 if the original volume was over 2000 milliliters), mixed well and a filtered aliquot of 150 milliliters saved. The three pre-saturation ali- quots were combined and analysis of this composite made. Trial analysis showed that urinary riboflavin was not de- stroyed by this procedure (Table 2). Two types of eXperi- ments were tried, In the first,2Mrhour urine samples for three successive days were analyzed separately for riboflavin and on the third day a composite of the three days' samples was analyzed. In the second type of experiment a single day's urine sample was analyzed at the end of the:2éhour pe- riod and again three days after the collection. There was no 17 TABLE 2 Stability of urinary riboflavin Experiment 1. Analyses of three successive 24-hour urine samples and the three-day com- posite of these samples Sample Riboflavin excretion fig./24 hrs. First day 1610.4 Second day 1434.0 Third day 1250.0 ' Average 1431.5 Three-day composite 1397.0 Per cent difference -2.4 Experiment 2. Analyses of same urine sample at end of 24-hour period and after three days Subject I II III kg. riboflavin at end of 24-hour col- - lection 153.5 832.6 366.6 pg. riboflavin in same sample after 3 days 161.0 812.9 350.0 Per cent difference +5.2 -2.4 -4.5 1 _ll....!0n.L.k J.‘ 18 loss of riboflavin after storage of the urine sample. The collections for the other periods were measured and made to volume as follows: One-hour fasting sample ------ to 100 milliliter Four-hour saturation period -- to 1000 milliliter Nineteen-hour period --------- to 2000 milliliter Twenty—four hour period after saturation --- volume recorded and sample used without dilution to any given volume. This dilution made the urine concentration of all subjects comparable and facilitated judgment of the size of the sample taken for analysis. The size of the aliquot taken for analysis depends on the amount of riboflavin expected in the urine sample. Three to 15 milliliters filtered urine were pipetted into a 50-milli1iter beaker and made to a volume of 20 milliliters with distilled water. The sample was adjusted to pH6 with sodium hydroxide, using Nitrazine paper as indicator. Ad- sorption columns were prepared with activated florisil and the urine sample passed through. The beakers were rinsed twice with warm distilled water and these rinsings passed through the columns also. The riboflavin was eluted with 50 to 35 milliliters of an aqueous solution of 20 per cent pyridine and two per cent glacial acetic acid, the eluate collected into 50 milliliter volumetric flasks and made to volume with the pyridine-acetic acid solution. A.l5- milliliter aliquot of this eluate was pipetted into a 25- milliliter volumetric flask, one milliliter of four per 19 cent potassium permanganate solution added to oxidize inter; fering substances and the flask shaken for three minutes. The solution was decolorized with a three per cent hydrogen peroxide solution, usually one milliliter being needed. The sample was made to volume with the pyridine-acetic acid so- lution and was ready to be read in the photo-electric fluo- rescence meter. Correction for interfering fluorescent substances was made by using a separate blank for each sample and reading this blank against its duplicate sample. .A duplicate ali- quot of each sample was pipetted into a 250-milliliter beak- er, the volume made to 20 milliliters and the beaker placed under a mercury vapor lamp for two hours to destroy the riboflavin (Keys, '44; DaMerre and Brown, '44). The fluo- rescence which remained was due to the interfering sub- stances. After the two hours any loss in volume due to evaporation was replaced with distilled water. The blank 'was then made to pH6, passed through the adsorption columns and treated in the same manner as the sample. A reagent blank was prepared by passing 30 to 35 milli- liters of the pyridine-acetic acid solution through an ad- sorption column, making to a volume of 50 milliliters with the same solution and treating like the samples. All samples and blanks were read in a photoelectric colorimeter equipped with a mercury vapor'lamp (Lumetron). The instrument was standardized at 100 with a solution of pure riboflavin containing 0.1 gamma per milliliter. The 20 zero setting of the instrument was made against the reagent blank to suppress any incidental fluorescence from the re- agents. Figures in the slide dial were in logarithmic re- lationship so that readings were directly in terms of per- centage concentration of the standard solutions. The micro- grams riboflavin for the total volume were calculated using this formula: Final reading x 5ojvolume of eluate) x 25(final volumg) Aliquot of urine sample x 15Tmilliliters of eluate) x total volume = total micrograms riboflavin 21 ‘J‘T‘JT" r1 luau UL: .LS Pre-Saturation Period The range of daily urinary excretion of riboflavin of the twenty subjects during the pre-saturation period of three days was 58.2 to 1030.0 micrograms (Table 3). The average excretion was 446.1 t 238.7 micrograms (Table 5). Since this range was so wide with a few subjects at the two extremes in excretion, an attempt was made to find a.more representative range of excretion for this group of college girls. When the data were grouped and the figures for the highest and the lowest excretions eliminated, the range of riboflavin excreted by 14 subjects was 250 to 750 micro- grams per day (Table 6). The range in.micrograms of ribo- flavin per milliliter of urine excreted during this period was 0.078 to 0.757 micrograms (Table 4) with an average ex- cretion of 0.444 t 0.213 micrograms per milliliter (Table 5). The range of urinary riboflavin excreted during the fasting hour was 0.4 to 37.8 micrograms (Table 3) with an average of 12.7 t 10.8 micrograms. The micrograms of ribo- flavin per milliliter of urine excreted during the fasting hour ranged from 0.023 to 1.686 micrograms (Table 4) with an average excretion of 0.488 1 0.263 micrograms. The cor- relation between the average total excretion for this fasting hour and for the previous 24-hour periods was not o.oewa H.5ma a.mm o.aaa m.os a.mwm o.mam a.mae m.oma ¢.nnm o.mam w.omm a.amm o.Hms a.mmm a.msm a.mam a.mam o.aaa o.HHo a.mmm m.amm .ma O.Hm a.mH «.mm H.NH m.Hm N.mH H.HN H.>¢ h.¢H m.¢m H.mm o.N¢ m.m¢ a.mn a.mw a.mm down m.>m p.0H n.0m m.ww m.ommfl w.omma o.mmm o.omm a.smm n.5maa m.Hms a.mmm a.sosa «.mmm a.smaa a.mama s.omsa ¢.sama m.meH a.ssma n.u¢mm a.mmma a.mmea a.mmm a.mwba N.momm .m1 .neoa can aoam asap“ on» no nowasabov chevnepm* 0.5H a.md «.mH w.m m.¢H m.m m.HH a.mm m.¢ N.OH a.mm b.mm m.ow N.HN a.mH $.om o.mH a.md w.» m.Hw w.Hn 5.HnH« m.mmb a.mmw a.mwm o.¢mw 0.5dm >.Nmm a.mmm o.¢wOH a.mmw m.mm¢ m.NmHH m.meH >.wwmd m.NHm w.¢m> a.sbh m.bbm a.mmw N.mmm «.mNNH 0.5H6H .ml m.OH% a.mH 00000 O l"‘| O C C I O r401H mmwwmmmmommmmm¢ O .comHmr-INo-lmd'oow HNH o F- N w.hw camadm on H.0H n.HH w.mu .mk *a.mmam H.m¢¢ N.wm a.mm b.mmH m.mom o.mmm m.0mm N.mmm m.mH¢ m.bmd o.wmw a.mmw «.mbv a.mmw m.me m.mwm n.06m a.mww 0.00m w.nmm 0.0n0H .m« omwum>< OH mm mm M! m: Hm mm am am 33 a: an Uh an an mm mm mm mm mm sowpmnspwm mason em ma soapsnspsm mason d ma aoapdusesm mnavmwm haamn nevus woponoxm nouns ucaonoxo nouns nsonlmso omshoba poonnsm mason mwldm pace 9mm mason wmlw pnmowumm mason via ZOHammon zOHBHMUNm ZOHBuamonau mo maoavohoxo hasnwha n quda NN 23 TABLE 4 Urinary excretions of riboflavin* POST-SATURATION EXCRETION FEE-SATURATION EXCRETION 1-4 hours 4-24 hours 24-48 hours Subject Daily One-hour after after after average fasting saturation saturation saturation )‘s/ml. fiso/ml. rs/ml. Hes-hal- Ha/ml. EH. 0.756 1.686 2.516 0.849 0.482 AB 0.737 0.753 2.005 0.847 0.632 FR 0.591 0.055 0.972 0.980 0.568 GH 0.670 no sample 1.026 0.902 0.389 ER 0.356 0.266 1.015 0.556 0.309 LM 0.560 0.821 1.353 4.332 0.768 JC 0.757 0.746 4.749 0.528 0.396 JG 0.600 0.722 1.383 0.843 1.019 JF 0.310 0.927 1.318 0.434 0.514 14D 0.423 0.509 1.066 0.906 0.625 132 0.453 0.242 2.801 0.667 1.080 BJ 0.502 1.410 0.906 0.595 0.409 RN 0.436 0.067 0.497 0.482 0.351 KB 0.330 0.314 2.331 0.957 0.431 ET 0.390 0.213 0.547 0.454 0.367 IEP 0.503 0.098 1.764 1.156 0.618 135 0.194 0.707 1.609 0.833 0.601 FF 0.142 0.023 0.689 0.194 0.078 LH 0.083 0.158 0.735 0.303 0.188 I0 0.078 0.040 2.137 0.221 0.068 Average 0.444 0.488 1.571 0.852 0.495 *Micrograms of riboflavin excreted per milliliter of urine. Summary of Period 24-hour excretion before saturation One-hour fasting excretion Four-hour excretion after saturation 24-hour excretion after saturation 48-hour excretion after saturation 24 TABLE 5 average riboflavin excretions Micrograms excreted per milliliter during period Micrograms excreted during period Average S.D. Average S.D. yg./24 hrS. fig./ml. 446.1 2238.7 0.444 10.213 12.7 110.8 0.488 10.263 789.8 1131.7 1.571 $1.003 1380.6 1560.6 0.852 10.86 487.1 1240.0 0.495 £0.477 [ill-III! 11.1. 25 TABLE 6 Range of riboflavin excretion found by grouping the data Range in 24-hour period before saturation Group Range Frequency Average within group "go o I 750-999 2 9 1.6 II 500-749 6 578.0 III 250-499 8 398.0 IV 0-249 4 137.7 Average 449.0 1215.0 fig. Range in 24-hour period of fifth day Group Range Frequency. Average within group Pg‘ ”go I 750-999 3 853.7 II 500-749 8 595.7 III 250-499 7 332.1 IV 0-249 2 64.2 Average 524.0 1166.5‘pg. 26 significant (r = 0.34) for the number of samples studied. The lowest correlation at which significance can be assumed for this size of sample (19) is 0.37. The correlation of 0.34 found in the comparison of the one-hour fasting sample and the preliminary 24-hour urine sample is very near to a significant figure. When the average micrograms of ribo- flavin per milliliter excreted during the fasting hour and the pre-saturation 24-hour periods were compared, there was good correlation (r = 0.53). There was further confirma- tion of this relationship when no significant difference (t : 0.99, D.F. = 18) was found between the two averages, the average micrograms of riboflavin excreted per milli- liter of urine during the one-hour and during the 24-hour periods. There was no correlation between the volume of urine and the micrograms of riboflavin excreted during the one-hour fasting period. Saturatiop Period During the four hours after the three milligram test dose of riboflavin was given, the range of excretion was 263.9 to 1717.0 micrograms of riboflavin with an average of 789.8 1 131.7 micrograms. This represented 3.6 to 40.6 per cent, or an average of 17.9 per cent, of the three milligram test dose excreted in the four hours (Table 3). At the end of the first 24 hours after the test dose the total riboflavin excretion ranged from.524.6 to 2647.3 micrograms, an average of 1380.6 t 580.6 micrograms. The fraction of the test dose that had been excreted by the end 27 of 24 hours was 10.7 to 69.5 per cent or an average of 31 per cent (Table 3). The second 24-hour excretion immedi— ately after the test dose gave a high correlation (r = 0.68) with the 24-hour average of the presaturation period suggesting that the test dose was excreted within the first 24 hours. Thus, during the fifth day, which was the second 24 hours after the administration of the test dose, the uri- nary excretion of riboflavin appeared to have returned to the range found during the 24-hour period before the test dose was given. The range of riboflavin excretion of the fifth day of the experimental period was 58.4 to 966.6 micrograms (Table 3) with an average of 487.1 i 240.0 mi- crograms (Table 5). This average was not significantly different (t = 0.55, D.F. I 19) from.the average excretion of the pre-saturation period. Fifteen of the subjects ex- ' creted between 250 to 750 micrograms, or an average of 524.0 1 166.5 micrograms of riboflavin during this period (Table 6). The micrograms of riboflavin per'milliliter of urine excreted during this day ranged from.0.068 to 1.080 micrograms with an average of 0.495 t 0.477 micrograms. This average was not significantly different (t = 0.55, D.F. = 19) from the average micrograms per milliliter of the pre-saturation period, though there was some individual variation. 28 Dietary,Study: The dietary intakes of protein, thiamine and ribofla- vin were calculated from food consumption records (Short method, Reynolds,'44). AAn average of the dietary intake of each subject for one day was obtained from the intake of the five-day experimental period. This average was used as an indication of the subject's dietary habits. The average intake of protein was 58.2 grams and for thiamine was 1.056 milligrams. Riboflavin intake varied from 1.03 to 2.2 mil- ligrams, or an average of 1.55 i 0.339 milligrams per day (Table 7). The protein and thiamine in the individual diet appeared to have no relationship to the amount of ribofla- vin. The correlation between the number of glasses of milk in the diet and the riboflavin.excreted in the preliminary period was high (r : 0.668). This seemed to indicate that milk was an important constituent of the diet in determin- ing the amount of riboflavin excretion. There was good correlation (r = 0.525) between dietary intake of ribofla- vin and the micrograms of riboflavin excreted in the 24- hour period before the test dose (Figure 1). There was possibly some relationship between dietary intake of ribo- flavin and the percentage of the test dose excreted during the first 24 hours after the awninistration of the test dose (Figure 2). Relationship between Riboflavin and Thiamine No effect on riboflavin excretion due to the adminis- tration of three milligrams of thiamine could be observed. 29 TABLE 7 Average daily dietary intakes of protein, thiamine and riboflavin Subject Protein Thiamine Riboflavin gms. mgs. mgs. Ifii 62.6 1.65 2.113 AB 51.5 1.011 1.269 FR 57.0 1.0 1.865 GH 71.8 1.221 1.888 ER 67.0 1.086 1.715 LM 71.8 1.294 2.203“ JC 46.0 0.916 1.458 JG 57.1 0.861 1.707 JF 71.1 1.479 1.846 MD 55.5 1.128 1.267 in? 72.6 0.946 1.419 BJ 58.7 0.926 1.456 IR! 61.5 0.847 1.614 KB 70.2 1.170 1.623 BT 42.7 0.932 1.477 MP 44.9 0.783 1.401 MK 51.9 0.868 1.451 FF 55.3 0.910 1.030 LH 48.9 1.041 1.045 ID A 46.4 1.052 1.123 Averages 58.2 1.056 1.584 *Liver eaten on day of saturation; value for pre-saturation days . 1.71 Dietary Intake of Riboflavin 3O DAILY DIETARY IETAKE AND 24—HOUR URIK RY EXCRETION OE RIBOELAVIH IxZg . 2.1 1.7 : 1P 1 0 l 1.5 0 1.3 = / 1.1 = . 0.6 I 0.5j ‘yg. 100 200 300 400 500 600 700 800 900 1000 Riboflavin Excreted in 24-Hour Urine Samples of Pre-Saturation Period Figure I Dietary Intake of Riboflavin 31 PERCENTAGE or sartaxrrcn DOSE xxcnarnb In 24 nouns RED DAILY DIETARY IETAKE OE RIBOELAVIE a; O 0 105 ' 1.3 ‘ 1.1 4 - éer Cent 10' 20 50 4o 50 60 70 Percentage of Saturation Dose Excreted in 24 Hours \ Figure II 32 Also, the dietary intake of thiamine seemed to have no rela- "I tionship to the dietary intake of riboflavin. The thiamine excretions will be reported in detail by another investiga- tor from this laboratory. 33 DISCUSSION The range of urinary excretion of riboflavin found in this study was lower than that previously reported (Table 1). This may be due to the fact that the findings in the lit- erature have been reported on studies made almost entirely with men (Sebrell, '41; Williams, '43). The lower values previously reported were from the studies with hospital pa- tients (Axelrod, Spies, and Elvehjem, '41). In an attempt to compare the figures obtained in this study with the data reported in the literature, the riboflavin excretion of each subject was recalculated on the basis of 70 kilograms of weight (Table 8). The adjusted range of excretion found was 65 to 1273.7 micrograms or an average of 539.3 micro- grams. These values compared with the lower range of pre- sumably normal data reported in the literature. The fact that the 24-hour urine samples were held for three days before analyses did not explain the lowered ex- cretion values. Hagedorn and co-workers (1945) stored urine samples for three weeks with no significant variation in riboflavin content. Trial analyses in this laboratory indicated that there was little variation between the aver- age riboflavin content for three 24-hour urine samples ana- lyzed immediately at the end of each 24-hour period and the riboflavin content of a composite of three days analyzed at the end of a three-day period. TABLE 8 Urinary excretions of riboflavin calculated to 70 kilograms weight Subject Height Weight Riboflavin Per 70 kilo- Inches Kilograms excretion grams weight pg./24 hrs. fig./24 hrs. HR 66 55.9 1030.0 1273.7 AB 64 63.6 833.3 917.2 FR 61 63.6 666.6 733.7 GH 66 52.3 649.9 869.9 DR 60 54.5 576.3 740.2 Iii 61 51.8 563.8 761.9 JC 65 61.8 516.3 584.8 JG 67 65.9 499.9 530.2 JF 66 67.3 472.2 491.1 IMD 64 50.0 468.3 655.6 1m” 65 59.1 466.6 552.1 BJ 64 55.0 437.9 557.3 RH 66 75.0 415.5 387.8 KB 63 54.6 338.2 433.6 RT 64 62.7 320.8 358.1 RP 66 58.1 266.6 321.5 FF 64 54.6 159.7 204.7 LH 67 64.6 95.5 103.5 I0 62 62.7 58.2 65.0 Average 446.1 539.3 35 Feder and co-workers (1944) reported that the micro- grams of riboflavin excreted per milliliter in the fasting hour sample gave good agreement with the micrograms of riboflavin excreted per milliliter in the 24—hour sample. Hagedorn, and others (1945) found fair agreement in the two measurements but stated that the low concentration of the one-hour sample led to errors in analyses. In this study there was no significant difference between the average mi- crograms of riboflavin excreted per milliliter of urine in the two periods. The individual variation in riboflavin excretion per milliliter during the two periods appeared greater than statistical analysis would indicate (Table 4). This one-hour fasting excretion seemed, therefore, to be an adequate measurement of riboflavin excretion when a group was studied but inadequate for individual comparisons. The three-milligram.test dose of riboflavin has been referred to in this paper as a "saturation" dose. Keys, and others (1944) found that the 24-hour urinary excretion of riboflavin averaged 12 per cent of the dietary intake when the subject was on a low riboflavin diet. The per- centage recovery of a one-milligram test dose was similar. For 26 days the subjects in Key's study were given 11.2 milligrams of riboflavin per man per day. The response to the test dose was 19.7 per cent six days after the excess, intake was stopped. When Hagedorn, and others (1945) used a two-milligram test dose given orally, there was 64 per cent excretion of the test dose in 24 hours compared to a 56 20 per cent excretion when one milligram of riboflavin was used as the test dose. Others (Ferrebee, '41; Williams and others, '45) have injected one milligram of riboflavin as the test dose. In this study three milligrams of ribofla- vin were given orally as a test dose. In 24 hours after administration, an average of 51 per cent of the test dose had been excreted, a figure which would not suggest marked depletion of body Stores of the vitamin. The excretion of riboflavin in the four hours after the test dose was lower than that reported in the litera- ture (Ferrebee and Wakxman, '45; Najjar and Holt, '41; Feder and others, '44). These investigators injected the test dose. A.de1ay in excretion of the oral test dose was to be expected,since the time needed for absorption could not be determined. In this study 24 hours after adminis- tration of the test dose an average of 51 per cent was ex- creted which is comparable to the excretions reported in the literature (Table 1). However, only seven of the sub- jects responded to the test dose by an excretion of 25 per cent or more and also excreted 500 micrograms of riboflavin or over per 24 hours during the preliminary period. These limits, 500 micrograms riboflavin excreted per 24 hours and 25 per cent excretion of a test dose in 24 hours, are the lower limits of riboflavin excretion proposed in the lit- erature as indicating good riboflavin nutrition. Figure 5 illustrates this. This would mean, if former investiga- tions are used as the criteria, only seven of the twenty sub- jects studied were in good riboflavin nutrition. Six of the 57 UBITARY ELCRETIOl-I OF RBOL‘LAVIN DURIKG 24—ZOUR PERIODS LJEORE TEST DOSE AED Pfifi‘;nTAfi£ EhURfiTIOK 0E TEST DOSE IN 24 HOURS 70 . 60 5O . 4o ' in 24 Hours 20 , - . P Q 0 Percentage of Test Dose Excreted r8. 100 200 300 400 500 600 700 800 900 1000 Riboflavin Excreted during 24—Hour Period before Test Dose -—— Cited in literature as lower limit of per- centage response to test dose. --- Cited in literature as lower limit of micro- grams riboflavin excreted in 24 hours. FIGURE III Viv. .r :a' .3 . {it 38 subjects excreted less than the minimum 500 micrograms in the 24 hours but showed a 25 per cent or more response to the test dose. When the 24-hour urinary excretion of riboflavin was recalculated on the basis of 70 kilograms weight, all but four subjects (Figure 4) fell within the standards proposed by the literature. Three of these four subjects excreted less than 500 micrograms per 24 hours even with the new calculations, but the response of these three subjects to the test dose was above 25 per cent. This may have been due to a lower kidney threshold for riboflavin in these subjects than in the other subjects, if riboflavin can be considered a "threshold" substance. The response of the one subject who excreted 666.6 micrograms of riboflavin in 24 hours during the preliminary period and only 10.7 per cent of the test dose could not be explained. The recalculation of the riboflavin excretion values on the basis of size was done arbitrarily. The effect of size on the urinary excre- tion of riboflavin is not known, but the results as ex- pressed in Figure 4 seem to indicate that there may be some relationship, particularly since no complaints attributable ' to riboflavin deficiency were made by the subjects. There appeared to be some relationship between diet- ary intake of riboflavin and the percentage of the three- milligram.test dose excreted in 24 hours after administra- tion (Figure 2). This relationship might have been more pronounced if there had been more predise dietary data. Percentage of Test Dose Excreted Hours in 24 PERCENTAGE ERURQTIUN OF 59 URINARY EXCRETIOE ‘F RIBOELAVIE DUKIZI} BIL-{1-01111 PERIODS BEFORE TES‘I‘ DOSE RbOALCULATEU FOR 70 LILOGRALS WEIGHT AND TEST DOSE IE 24 HOURS 70 6O 50 4O 30 20 H-CC-lp—I_-—‘——-——‘—--—‘—+h——— fig. 150 300 450 600 750 900 1050 1200 1350 1500 Riboflavin Excreted in 24-Hour Period before Test Dose uecalculated for 70 Kilograms Weight Cited in literature as lower limit of per- centage response to test dose.. . n . Cited in literature as lower limit or micro- grams riboflavin excreted in 24 hours. 40 The diets for the five-day experimental period were calcu- lated. The calculations were made on the basis of records kept by the subjects and all measurements of food intake were estimated. This method gave an indication of the sub- ject's dietary habits, but did not furnish absolute values of intake. The urinary excretion of riboflavin on the fifth day of the experimental period showed that there was very little delay in the excretion of the test dose (Table 5). At the end of the first 24 hours after administration of the test dose, the riboflavin excretion returned to the av- erage excretion found during the 24-hour periods before the test dose. This suggested that the excretion on the fifth day might be eliminated from a similar experimental study without sacrificing useful information. The fadt that there is no significant difference between the average mi- crograms of riboflavin excreted per milliliter of urine during these two periods emphasizes this relationship. In this study it was found that the daily variation in urinary excretion of riboflavin between 20 college women was over 900 micrograms. There was also individual varia- ‘fien in the amount of riboflavin excreted in one period in relation to the amount excreted in another period. One subject excreted a little over 500 micrograms riboflavin daily and excreted 47.1 per cent of the test dose in 24 hours. Another subject excreted over 600 micrograms of riboflavin daily, but excreted only 27.9 per cent of the 41 test dose in 24 hours. Still, the riboflavin excretion of the group of subjects as a whole fell into a definite pat- tern. This is shown in figure 5, but is more apparent after the figures were recalculated to'a uniform weight basis as is shown in Figure 4. The subjects who excreted the most riboflavin daily were, in general, the same ones who ex- creted the highest percentage of the three milligrams of riboflavin in the first 24 hours after the test dose was given. The six subjects with low daily excretion were the same subjects excreting below 25 per cent of the test dose in 24 hours. Figure 4 shows only four subjects of the 20 who did not follow this pattern. There were five subjects (Figure 4) who might be said to have poor riboflavin nutrition according to previously reported data (Table 1). .All of the subjects were selected for this study because of their apparent good health. No indication of riboflavin deficiency could be noted. Each of the subjects had a record of previous good health. Is the degree of saturation a measure of riboflavin deficiency when no deficiency can be detected in subjects who, by lab- oratory measurements, are desaturated? Others (Hagedorn and co-workers, '45) have questioned the validity of a sat- uration test when the effects of tissue desaturation cannot be measured. There is a definite need for further investi- gation of methods of measuring deficiency and the effects of riboflavin deficiency on the human subject before any conclusions concerning the riboflavin status of the indi- vidual may be made. 42 SUMMARY The usual range of urinary excretion of riboflavin in twenty college women on a self-chosen diet and the response (to a test dose of three milligrams of riboflavin and three milligrams of thiamine were studied. The range of urinary riboflavin excreted in 24 hours was found to be 58.2 to 1050.0 micrograms or an average of 446.1 micrograms. Four- teen of these twenty subjects excreted between 250 and 750 micrograms for the 24 hours. The fasting hour excretion ranged from 0.4 to 57.8 micrograms, an average of 12.7 mi- crograms. Four hours after a test dose of three milligrams of riboflavin was given, 17.9 per cent of the amount had been excreted. The average excretion of the test dose of riboflavin in 24 hours after administration was 51 per cent which compared with the 25 to 55 per cent output reported in the literature. The range of urinary excretion of ribo- flavin during the second 24 hours after administration of the test dose returned to the range found during the 24- hour period previous to the test dose. There was no significant difference between the av- erage micrograms of riboflavin excreted per milliliter of urine during the 24-hour periods before the test dose, the one-hour fasting period and the second 24-hour period after the test dose. The dietary intake of riboflavin gave a high correla- 45 tion with the daily 24-hour excretion and was reflected in the response to the saturation dose. No effect on the urinary excretion of riboflavin due to the administration of a three-milligram test dose of thiamine could be observed. A wide variation was found in the range of riboflavin excretion of the group and in individual excretions after the test dose. However, a certain pattern of excretion was shown, the subjects with the higher daily excretions having the greater percentage excretion of the test dose. No sign of riboflavin deficiency could be detected in the subjects who had maintained a lOW'riboflavin excretion in all of the periods. The reliability of using the degree of saturation as an index to riboflavin status of human subjects should not be accepted without further study. 44 LITERATURE CITED Axelrod, A. E., T. D. Spies, and C. A. Elvehjem. 1941 Riboflavin content of blood and muscle in normal and malnourished humans. Proc. Soc. Exp. Biol. and 13136.0, VOlo 4:6, pp. 146-1490 Axelrod, A. E., T. D. Spies, and C..A. Elvehjem, 1941 A.study of urinary riboflavin excretion in.man. J. Clin. Invest., vol. 20, pp. 229-252. Baten, W. D. 1958 Elementary;Mathematical Statistics. John Wiley and Sons, Inc” New York. Blyth, A. W. 1879 The composition of cow‘s milk in health and disease. J. Chem. Soc., vol. 55, P. 5500 Conner, R. T., and G. J. Straub 1941 Combined deter- mination of riboflavin and thiamine in food ducts. Ind. and Eng. Chem., Anal. Ed., vol. 15, pp. 585-588. De Merre, Leon J., and William S. Brown 1944 Effect of various lighting conditions on riboflavin solu- tions. Archives of Biochem., vol. 5, pp. 181- 190. Emmerie,.A. 1956 Determination and excretion of flavins in normal human urine. Nature, vol. 158, p. 164. Emmett, A. D., and L. H. McKim 1917 The value of the yeast vitamine fraction as a.supplement to a rice diet. J. Biol. Chem., vol. 52, pp. 409-419. Feder, V} H., G. T. Lewis, 81d H. S. Alden. 1944 Studies'w on the urinary excretion of riboflavin. J. Nutrition, vol. 27, pp. 547-555. Ferrebee, J. W. 1940 The urinary excretion of riboflavin: fluorometric methods for its estimation. J. Clin. Invest., vol. 19, pp. 251-256. Ferrebee, Joseph W., and.Norman Weissman 1945 Riboflavin and thiamine interrelationships in rats and in man. J. Nutrition, vol. 26, pp. 459-469. Gyfirgy, Paul 1955 The differentiation of lactoflavin and the rat anti-pellagra factor. Biochem. J., vol. 29. PP. 741-759. Gyflrgy, Paul 1942 The water soluble vitamins. Ann. Rev. Biochem., vol. 11, p. 509. 45 Hagedorn, D. R., E. D. Kyhos, O. A. Germek, and E. L. Sev- ringhaus 1945 Observations on riboflavin ex- cretion by the adult male. J. Nutrition, vol. 29, PP. 179-189. Hand, David B. 1959 Determination of riboflavin in milk. Ind. and Eng. Chem., Anal. Ed., vol. 11, p. 506. Hawk, Phillip E., and Olaf Bergeim 1957 Practical Physiologica1_0hemistpy, 11th ed. Tie Blakis- ton Company, Philadelphia. Helmer, 0. M. 1957 The determination of vitamins B and G in human urine by rat growth.method. J. Nutri- tion, vol. 15, pp. 279-286. Hodson,.A. Z., and L. C. Norris 1959 A fluorometric method for determining the riboflavin content of foodstuffs. J. Biol. Chem., vol. 151, pp. 615- 650. Keys, A., A. F. Henschel, 0. Mickelsen, J. M. Brozek, and J} H. Crawford 1944 Physiological and bio- chemical functions in normal young men on a diet restricted in riboflavin. J. Nutrition, vol. 27, pp. 165-178. Keys, Ancel 1944 Personal communication. Klein, J. R., and.H. I. Kohn 1940 The synthesis of flavin-adenine dinucleotide from riboflavin by human blood cells in vitro a1d in vivo. J. Biol. Chem., vol. 156, pp. 177-189. Klopp, C. T., J. C. Abels, and C. P. Rhoads 1945 The re- . lationship between riboflavin intake and thiamine excretion in man. Am. J. Med. Sci., vol. 205, pp. 852-857. Machella, T. E., and P. R42McDonald 1945 Failure of ribo- flavin therapy in patients with the accepted pic- ture of riboflavin deficiency. Am. J. Med. Sci., vol. 205, pp. 214-225. Mannering, G. J., D. Orsini, and C. A" Elvehjem. 1944 Effect of the composition of the diet on the ribo- flavin requirement of the rat. J. Nutrition, vol. 28, Pp. 141-156. Najjar, V. A., and E. L. Holt 1941 A.ribof1avin excretion test as a measure 0f riboflavin deficiency in man. Johns Hopkins Hosp. Bull., vol. 69, p. 476. 46 Najjar, Victor.A. 1941 The fluorometric determination of riboflavin in urine and other biological fluids. J. BiOl. Chem., V01. 141, pp. 355-564.“. Najjar, Victor A., G. A. Johns, Geo. C. Medairy, G. Fleischmann, and.L. Emmett Holt 1944 The biosynthesis of riboflavin in man. J. Am. Med. Assn., vol. 126, pp. 557-558. Narayanan, B. T., and J. C. Drummond 1950 The concentra- tion of vitamin B2. Biochem. J., vol. 24, pp. 19-26. Oden, J. W., L..H. Oden, and W} H. Sebrell 1959 Report of three cases of ariboflavinosis. U. S. Pub. Health Rpts., vol. 54, pp. 790-792. Oldham, H., F. Johnston, S. Kleiger, and Hernaanedderich- Arismedi 1944 A study of the riboflavin and thiamine requirements of children.of pre-school age. J. Nutrition, vol. 27, pp. 455-446. Parsons, Helen T. 1944 Further studies on human require- ments for riboflavin. Federation Proceedings, V01. 3, pp. 162‘1710 Quastel, J. E., and D. M. webley 1941 Vitamin Bl and bacterial oxidation. Biochem. J., vol. 55, pp. 192-206. Reynolds, May, Chairman, Diet Therapy Section, American Dietetic Association 1942 Food composition table for short method of dietary analysis. Revised, 1944, by Eva G. Donelson and JaneId. Leichsering (mimeographed). Salmon, W. D., N. B. Guerrant, and I. M. Hays 1928 The effect of hydrogen ion concentration upon adsorption of the active factors of vitamin B complex by fuller's earth. J. Biol. Chem., vol. 80, pp. 91-101. Sandstead, H. R. 1942 Superficial vascularization of the cornea. The results of riboflavin therapy. U. S. Public Health Rpts., vol. 57, pp. 1821- 1825. Sebrell, W}‘H., R. E. Butler, J. G. wooley, and.Harris Isbell 1941 Human riboflavin requirements es- timated by urinary excretion of subjects on con— trolled intake. U. 8. Public Health Rpts., vol. 56, pp. 510~519. 47 Selye, Hans 1945 The role played by the gastrointestinal .tract in the absorption and excretion of ribo- flavin. J. Nutrition, vol. 25, pp. 157-142. Singher, H. 0., C. J. Kensler, H. Levy, 3. Poore, 0. P. Rhoads, and.Hlaus Unna 1944 Interrelationship between thiamine and riboflavin in the liver. J. Biol. Chem., vol. 154, pp. 69-77. Snell, E. E., and F. H; Strong 1959 A.microbiologica1 assay for riboflavin. Ind. and Eng. Chem., Anal. Ed. , V01 0 ll , pp 0 346-349 0 Strong, F. M., R. E. Feeney, B. meore, and H. T. Parsons 1941 The riboflavin content of blood and urine. (I. B101. Chemo, VOlo 137’ pp. 363-572. Supplee, G. C., R. C. Bender, and O. G. Jensen 1959 Determining riboflavin. Ind. and Eng. Chem., Anal. Ed., vol. 11, pp. 495-498. Sure, Barnett 1944 Influence of sub-optimal doses of B1 on urinary excretion of riboflavin. J. Nutrition, vol. 27, pp. 447-452. Sure, Barnett 1945 Vitamin interrelationships. IV Fur- ther studies on the influence of chronic thia- mine deficiency on riboflavin metabolism. J. Biol. Chem,, vol. 157, pp. 545-549. Sure, B., and Z. W} Ford 1942 Vitamin interrelationships. II Thiamine and riboflavin interrelationships in metabolism. J. Biol. Chem,, vol. 146, pp. 241- 250. Sure, Barnett, and Zenas W. Ford, Jr. 1945 Influence of increasing doses of thiamine and riboflavin on efficiency of their utilization. J. Nutrition, vol. 26. PP. 659-671. Sydenstricker, V. P., W; H. Sebrell, H. M. Cleckley, and H. D. Kruse 1940 The ocular manifestations of ariboflavinosis. J. Am. Red. Assn., vol. 114, pp 0 24:57-24:45 o Tennent, DavidIi., and Robert H. Silber 1945 The excre- tion of ascorbic acid, thiamine, riboflavin and pantothenic acid in sweat. J. Biol. Chem., vol. 148, pp. 559-564. Unna, K., H. O. Singher, C. J. Kensler3 H. C. Taylor, and A“ Rhoads 1944 Effect of dietary protein on liver riboflavin levels and inactivation of estradiol by liver. Proc. Soc. Exptl. Biol. Med., vol. 55, pp. 254-256. . 48 E'filliams, R. D., H. L. Llason, P. L. Cusick, and R. LI. Wilder 1945 Observations on induced riboflavin deficiency and the riboflavin requirement of man. J. Nutrition, vol. 25, pp. 561-577. .1 .lldn .nlu Ail II] . AIIIJ' rv ubfL iyi'r -Hgn-fi! s . . _ . ' ‘I -u‘.£.—--—_.______ "'TITI'ITII'flI’DLflflIMfi!flifljlflfflfljilflfljflflfflflmfl“