A STUDY OF THE STORAGE OF WTAMN A {N THE LNERS OF NETS FED VAREOUS COMMERCEAL £35232" AND CANNED DOG F’SC‘DS mask for fhe Dagree a5 M. S. s‘xfilCE‘flGAM STA? CGLLEGE Nit-Ea Yum Shamg L636 1949 This is to cortih; that the thesis entitled 'A Study of the Storage of Vitamin A in the Livers of Rats Fed Various Commercial Dry and Canned Dog Foods' presented ht] Alicia (Yun Sen) Lee has been acreptcd tuwzlrds fulfillment nl thv requirements far M.S. Chemistry degree in ,’ , / U ‘1] / [1/ j ‘MXF hliljul‘ [DI‘UI‘ SMjr Date May 11; 19L}? .m‘.--_—- a 5-”-— A STUDY OF THE STORAGE OF VITAMIN A IN THE LIVERS OF RATS FED VARIOUS COHHERCIAL RY AND CANNED DOG FOODS BY Alicia Yun Sheng Lee A Thesis Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in fulfillment of the requirement for the degree of Master of Science Department of Chemistry 19m CHEMISTRY DEPT. '3 r... 7-" (I w x, AC EII‘?O’.'~."I.BDGEIJZE1‘ITS The author wishes to express her sincere appreciation to Doctor C. A. Hoppert for his guidance, encouragement and kindly criticism throughout these studies and in the preparation of this thesis. IntrOduC-tionOOOOO...0.00.00...0..OOOOOOOOOOOOOOOO Review of Literature............................. I. II. Storage of vitamin A A. Methods of Assay for Vitamin A A. Biological assay............... B. Physieo-chemical method........ Distribution of vitamin A in the body of the ratOOOOOOO0....OOOOIOOOOOOOOOOOOOOOOOO Relation of liver storage of vitamin A to the intal:e.0.00.00.00000000000000000000000 Relation of plasma level and liver storage of vitamin A...................... Relation of liver storage of vitamin A to age and 81‘0Vu't1’l...........o....o. C. Factors affecting the accuracy of A determinations.......... 1. Liczl’ltoooooooooooooooooo Exposure to EuirOOOOOOIO Saponifiable material.. Deeompos storage..... Peroxide.... Carotenes... Cholesterol. ition of chloro 0.0.1 0 .III ..III .OO.V ....V ..VII .VIII OOOOX ..XII .XIII .XIII v- T ..JL 0 o 03W ...XV , o o cit . 3le I I Amneriuental IIethod.... I-i 1-! L |~ J I. Storage of vitamin \ in the liver 0; various dog foods A. Equipment......... C. sailpling.......... D. Saponifi cation.... E. JwitraCtionoooooooo F. Solvent removal... D. RGaSCntsoo........ D G. Reading of the unknown........... .........XVII rats H. Preparation of the calibration curve.. A. Comparison of tile an t imony G.D.H. reagents for determining A content of rat liver 3. Cholesterol content of 1 J. C. Comparison of tie ’- in the UWO methods. 0’} [—10 O s and Di cus n... (1' C3 esul smm‘:lary000000000000...... 'fi r- g ibli lor.rap11;J-QOOOOOOOOOOOO trichlo O W 1‘108 fed ...KVIII wr ‘7 o o o o oA.-.4\. ...KKI 1rvr .. ...xumI . .1. o oXQiI . . .KKIII ...XXIII ...KKIII and the vitamin SoooooooOOQOOQOOOOOOOOXZW the li \fOl s.. . .. . . ...IIKVI rforonce of cholesterol 0 . .:{:LJ.II ...XKVII o o o .33th . .XICXVII A STUDY OF THE STORAGE OF VITAKIN A IN THE LIVERS OF RATS FED VARIOUS COMMERCIAL DRY AND CANNED DOG FOODS INTRODUCTION "Fat soluble A", as a growth promoting factor was first named and studied by McCollum, Davis and Osborne and Mendel in 1913-1915. As a result of successive publications of the physiological effect of vitamin A on growth, health, reproduction, length of life, and the well being of the animals, vitamin A is generally recognized as an essential constituent of the diet. In nutritional studies, the direct estimation of vitamin A from the food material is complicated by the following reasons: the presence of pro-vitamin A, carotenoids, with different vitamin A potencies; the destruction of vitamin A during preparation and storage; the variable efficiencies of absorption and conversion of pro-vitamin A into vitamin A due to individual differences; and the presence of substances which would either increase or decrease the stability of vitamin A. The literature indicates that vitamin A can be stored in the body, particularly in the liver and used during a period of deficiency in the diet (1,2,3). The amount of storage in the liver is dependent on the II vitamin A and pro-vitamin A intake. The most common pro-vitamins A in foods are Alpha, Beta and Gamma carotenes and cryptoxanthin. They are converted into vitamin A in the intestine (h,5) or in the liver. Then they are stored in the liver mostly in the ester form, a small amount in free form and only relatively small amounts as the unchanged carotenoids (6). Among the methods used in the determination of vitamin A in the diet or in boilogical materials, the physico-chemical method is more sensitive and time saving. Under well controlled conditions, it can be independent from personal errors. Liver contains vitamin A and the amount stored reflects the intake of the available vitamin A. Thereof using the chemical method in determining the vitamin A content in the liver of the animal on the food under investigation is an indirect way of measuring the vitamin A in the food. Because of the importance of the manufacture of dog foods and the current interest in the relative merits of dry and canned dog feeds a study was undertaken to determine the storage of vitamin A in rats fed leading brands of both types of foods. Incidental to this part of the work the Carr-Price reagent for vitamin A was compared with a recently introduced reagent, glycerol di-chloro-hydrin and the influence of cholesterol on the color development with each of the reagents studied. III REVIEW or L TEBATURE I. Storage of vitamin A A. Distribution of vitamin A in the body of the rat In 191k Osborne and Mendels suggested that the liver is especially concerned with the storage of vitamin A. That stimulated the interest of several investigators working on the storage of vitamin A in the liver and other organs. Sherman and Boynton 1925 (1) fed the tissues of adult animals as the sole source of this vitamin to young rats which had ceased to grow on a diet otherwise adequate. They estimated the value of vitamin A by measuring the growth of the animals and showed kidney tissue to be at least hO times as rich as muscle; lung more than hO times, and the liver from 200 to MOO times as rich in vitamin A as muscle. By weight, blood contained more vitamin A than muscle and much less than lung and kidney. Only a minute quantity of this vitamin was present in brain, bones and other tissues. T. Koore, 1931 (2) found that liver contains a large amount of vitamin A and that it plays an tnportant role in regulating the concentration of this vitamin through- out the remainder of the body. The vitamin A concentra- tion in the liver may vary over a wide range without producing any obvious effect on the concentration in IV the remainder of the body. By using the antimony trichloride reaction he established that 95$ of the total vitamin A of the rat was stored in the liver. The remainder was located in lung and kidney tissues, whereas brain, blood, muscle and the organs of digestion gave I. negative results. In the light OI other experimental data the negative results are perhaps subject to criticism. Baumann, Riising and Steenbock, l93h (7) showed that vitamin A was never found in the non-hepatic tissues of an animal unless the liver also contained vitamin A. Lewis, et a1. 1942 (8) found that retinal concen- trations of vitamin A were usually low at zero unit intake and reached an optimal value at an intake of 2 units daily in the case of the rat. The retinal concen- tration of vitamin A remained relatively high despite the low vitamin A plasma concentration and the absence of vitamin A in the liver. Recently Radice and Herraiz (9) used fluorescent light to follow the fate and storage of vitamin A in the body. They revealed that the main storage depots were the liver, adrenals and adipose tissue, especially the abdominal fat. Each of the above experiments made some contribution to our knowledge of the distribution of vitamin A in the V body. They all confirmed that liver contains the major part of the vitamin A. B. Relation of liver storage of vitamin A to the intake. The difference in the amount of vitamin A found in the liver and the lung tissue was directly attributable to the different amounts of this vitamin in the food (1). No vitamin A was detected in the livers of rats receiv- ing 10 units or less daily. Then the intake was 25 units or more, vitamin A was found in the liver, the amount stored being directly related to the intake (10). Sherman et a1. 19LS (11) demonstrated that at an intake of 3 I.U. per gram diet there was very little storage of vitamin A and no increase in the sto age with it increasing age up to 300 days. This intake 0 vitamin A is obviously not optimal and in the light of the storage results appears to be near the minimal limit of adequacy. When the amount was increased to 6 and 12 I.U. Per gram.of food appreciable storage of vitamin A in the liver was noted. Similar results were obtained by Steenbock et a1. 193A (7). C. Relationship of plasma level and liver storage of vitamin A. Lewis, Bodansky, Falk and McGuire 1 A1 (10) studied the vitamin A in the blood in relation to liver storage. VI They found that groups of rats receiving 0, l, 2, 10, 2S and 50 units of vitamin A daily, for six weeks, had average blood levels of O, 7, 1h, 35, 69, and 100 units respectively, per 100 ml. of blood plasma. haximal blood concentration of vitamin A was obtained with an intake of 50 units daily. No vitamin A was detected in the livers of rats receiving 10 units or less daily. Ihen the intake was less than 50 units daily the vitamin concentration of the blood was directly related to the vitamin A intake. Increasing the feeding above 50 units daily had no appreciable effect on the vitamin A level of the blood. The amount stored, however, varied considerable and had no consistent rel.tion to the absolute blood level. These findings therefore indicate that the blood level may be of value in ascertaining whether there is any storage of vitamin A in the liver. McCoord and Luee-Clausen 193k (12) had also showed that the concentration of vitamin A in the blood is no indication of the amount that may be stored in the liver. here was no definite relation between serum levels and liver storage except that vitamin A deficient animals had low serum values. In the absence of liver stores for several weeks vitamin A was still present in the blood and eye in measurable amounts in stock rats (1h). VII The knowledge of vitamin A circulating in the blood normallv as alcohol led Glover, Goodwin and Morton 19k (15) to go further into the stud y of the relation of the liver level of unesterified vitamin to the plasma level. Their data showed that the plasma vitamin A levels of rats are proportional to the concen- trations of vitamin A alcohol in the livers, but are not proportional to the total liver stores of vitamin A which consist mainly of esters. Plasma vitamin A levels are maintained near normal (3L-i: LO I.U./100 ml.) even when liver store approach exhaustion. D. Relation of liver Stora.ge of vitamin A to age and growth. In 193A (7) Baumann, Riising and Steenbock observed that in new born young rats, the stores of vita min A were found to be very low, averaging 7 blue units per liver. No vitamin A could be detected in the livers at age of one or two weeks. After three weeks the storage was rapid and regular. Using the length of the surVival period as a criterion Sher :1an and Storms 1925 (16) found that the maximum.body store of vitamin A was at the age of six months or at the beginning of full adult life in rats. Accordin“ly in the case of animals fed the same diet age was ibund to have an important influence upon the VIII length of the survival period when the animals were given food devoid of vitamin A. Davis and Hoore 1937 (17) showed that both growth and survival usually paralleled the storage of vitamin A, but some instances of anomalous behavior were observed in rats with high reserves. Thus several deaths occurred while substantial liver reserves were still present. II. Methods of Assay for Vitamin A A. Biological assay. The quantative determination of vitamin A has been made on the following bases: (1) Growth; (2) survival period; (3) curative treatment; (A) liver storage. The gr wth method was almost simultaneously develop- ed by Steenbock, Mendel, Drummond and their associates (18). Sherman and Munsell 1925(19) refined the method and introduced the unit for the numerical expression of the result. In this method rats of a certain ge of known nutritional history are placed upon a diet adequate in all respects but free of vitamin A. After growth has ceased, the negative control rats are continued on the basal vitamin A-free diet until death, whereas the others are fed graded portions of the food to be tested, as the sole source of vitamin A during a period of 8 IX weeks. The minimum daily allowance of the test material which will induce an average gain in weight of 3 grams per week during the test period is considered as having a unit of vitamin A. The method of measuring vitamin A based on the length of survival period involves giving depleted animals graded amounts of vitamin A. The length of the survival period is found proportional to the dose of vitamin A. A more widely adopted procedure is the curative method, in which the rate of disappearance of symptoms of vitamin A deficiency constitutes a measure of the vitamin A activity of the test material. The criteria usually used are the disappearance of the abnormal estrous and keratinization of vaginal epithelial cells and disappearance of xerophthalmia. The United State Pharmacopoeia recommends a curative procedure which is more accurate but involves more refined conditions; however, the greater cost and labor prevent its extensive use. The liver storage method was originally proposed by Guffenhein and Koch l9kh (20). They determined the available amount of vitamin A in material by feeding them to rats and determining chemically the amount of V [X vitamin A stored in the liver. They found that it has an accuracy at least the same order as the three weeks curative growth test and has the advantage of being shorter and having a specific criterion of response (21). Its limitations lie in the fact that the materials to be tested must contain an amount of vitamin A which will cause measurable storage in the liver. B. Physico-chemical method. Takahashi et al. 1925 discovered that vitamin A concentrates show selective absorption in the ultraviolet spectral region with a maximum at 328 mm. It was later utilized as a measure of the concentration of vitamin A, since the concentration of vitamin A is proportional to the amount of absorption. This absorption spectrum is very specific for pure vitamin A. This method is not so commonly used for the several reasons. (1) The presence of impurities may shift the point of maximum absorption or increase the absorption. Consequently steps would have to be taken to eliminate this source of error. 4 (2) The factor of converting E}6m_(328 mp) into biological units of vitamin A has not been settled. (3) Expensive equipment is required. The chemical method involving the Carr-Price reaction was originated in 1926 (22). The method (23) is based on the fact that a chloroform solution of vitamin A reacts Q? with antimony trichloride producing a blue color which has selective absorption of light at a maximum of (620 mp). The vitamin A concentration is proportional to the value ,4 of E%§m (620 mp) within a certain range. A calibration curve is recommended for the direct computation of vitamin A value. The rapid fading of the blue color limited the accurate measurement in the tintometer in which the comparison is made with standard colored glass. This limitation was overcome by Dann and Evelyn 1938 (2h) by uSing a photoelectric colorimeter instead of a tint- ometer. In lQhS Sobel and Werbin (2h) proposed a new colorimetric reaction of vitamin A with glycerol 1-3 dichlorohydrin. The next year they made an improvement 26) by using activative glycerol dichlorohydrin, here- after designated as G.D.H., because commercial sources of the compound did not always give a blue color with vitamin A as at first claimed. The method has been applied to the estimation of serum.vitamin A, vitamin A ester in fortified poultry mashes and fish liver oils (27-29). This new reagent appears to be suitable for quantitative purposes. Beers law holds over a reasonable range. The authors offer the following advantages: 1. The agreement between the G.D.H. and antimony trichloride method is close (25-29). XII 2. The violet color produced is stable from 2-10 minutes after the addition of the reagent. This suggested that the possibility of employing the visual method. 3. The interference of vitamin D and related sterols with the G.D.H. reaction is negligible (30). N..- p. The reagent is not affected by traces of moisture. \j‘L . No film of antimony oxychloride is left on the cuvettes. The reagent is non-corrosive. -\1 O‘ O . This method gives the same results when applied to .D whole or saponiiied extracts. The disadvantage of the new reaction is that the ng 10m3\615 mm) of the color extinction coefficient ( produced is about one~fourth that of antimony trichloride 1% 6 blue color (Llcm,)‘ 15 mm), The interference by carotene could be allowed for in the same way as with the antimony trichloride reaction. C. Factors affecting the accuracy of vitamin A determinations. 1. Light --- Vitamin A can be easily destroyed by light. Fuchs and Soos 19h3 (31) found that after 48 hours in the dark, neither the alcohol nor the ester showed a deterioration of more than 3%, regardless of the solvent. After M8 hours exposure to light the loss of the ester ranged from 22 to 66% depending on the solvent used. Gallup and Hoefer 19h6 (32) determined that during a 2-hour XIII .0 n .0 - , o o [-1’ period, losses OI Vitamin A were approximately 38; greater when the saponified sample was extracted with ether and the extract washed and dried near a north window than when the same operations were performed in subdued light. 2. Exposure to air --- Oxidation usually occurs when vitamin A comes in contact with air. Oser, Melnick and Pader 19h3 (33) observed that when solutions of vitamin A acetate and vitamin A alcohol were vigorously aerated at hSOC in the absence of light, there were changes in the curve in the region below 328 mp. In the antimony trichloride reaction the rate of color development decreased, 30 seconds being required to reach.maximal color intensity in the sample aerated for lhO hours, as compared with h seconds in the case of-the non-aerated solution. They suggested that there may be more than one form of oxidized vitamin A, those first produced reacting to a less extent than vitamin A with antimony trichloride reagent, whereas other compounds formed in the later stages of oxidation react to an even greater extent than (0 o o o i s w , Vlttflln A. Bolomey found in 19h? (3m) tnat tne rate of oxidation of vitamin A in contact with air increases with rise of temperature. The time necessary for destruction of half the vitamin at 1000C, as measured by the absorption at 328 mp, varied from.§2 to 175 minutes in different XIV specimens of oil. However, the destruction of the vitamin A could be prevented by the presence of 0.5% of mixed toc0pherols. 3. Saponifiable material --- The inhibition of color development with antimony trichloride in7 low potency fats and oils was overcome by saponification. Thus it was possible to make a satisfactory determination of vitamin A in fortified margarine by using the unsaponifiable fraction for the test (35). When antimony trichloride is applied to fish liver oils directly, especially to low potency oils, inhibitors of the color development are often encountered (33). Upon saponification the inhibitors are removed and the vitamin develops its full color intensity with the reagent. In addition, nonspecific materials may be present in fish oils which react with antimony trichloride to form a blue color absorbing light at 620 mu. It has been shown that the lower values of the unsaponifiable xtracts were due to the existence of vitamin A degrad- ation products which are removed because of their appreciable solubility in water. It was suggested by Dann and Evelyn (2h) that if the vitamin A content of a material is small, then the fat must be saponified and the unsaponifiable residue used for the test. This XV eliminates interference of unsaturated fatty acid groups with the Carr-Price reaction. J h. Decomposition of chloroform due to storage --- I On storage chloroform will decompose and form phosgen gas which will destroy vitamin A. Tastaldi 193h (36) recommended that chloroform should be redistilled at leaSt every 15 days. 5. Peroxide --- The best grade of ethyl ether usually contains sufficient peroxide to destroy vitamin A. Gallup and Hoeffer (32) showed that when an ether solution containing 2.5 microgram of vitamin A was evaporated in the usual manner in colorimeter tubes in the presence of O, 2, h, 6, and 8 ml. of anhydrous ether which gave a positive peroxide test, losses of vitamin A were 0, 6, 18, 27 and 295 respectively. 6. Carotenes --- Both carotenes and vitamin A yield blue color with antimony trichloride. Vitamin A and carotene, in solution together, react in additive manner (37). However, the maximum color produced by 10 U.S.P. units of vitamin A an? measured at 620 mp is of approx- imately the same order as that produced by 60 micrograms of Beta carotene. Thus on the basis of equal weights, the ratio of the color intensity of the reaction product of preformed vitamin A to that of carotene is approx- imately 20-1. Similar results were obtained by Johnson XVI et al. (38) who found that the blue color from 100 micrograms Oi pure carotene equalled that from 5.6 micrograms Oi vitamin A. o a'e the color reactions. [—10 It is possible to di ferent The blue color formed in tre ( C's «Jing vitamin A with antimony trichloride reaches its maximum in 2-5 seconds, whereas carotene attains its maximal intensity only after stand- ing somewhat longer. Johnson and Baumann found it was 5 to 10 seconds with pure carotene and 15 to 120 seconds with carotene from crude natural products. Bernard et al. reported that Beta carotene produced a maximum blue color after an interval of 2 hours. It can also be differenti- ated by using selective filters between the color produced by vitamin A and by carotenes by measuring the absorption, the maximum being 620 mp for vitamin A and 590 mp for carotene. Gray et al. (6) studied the recovery of vitamin A from liver by the administration of 20,000 units of various forms of vitamin A and Beta carotene during 48 hours. In the case of Beta carotene he recovered 9. i as vitamin A esters and only traces of unchanged carotene from the liver. Consequently the interference of carotene in the vitamin A assay on liver is not as serious as in certain food products. XVII 7. Cholesterol --- Cholesterol exists in the unsaponifiable part of the extract with vitamin A. According to Corbet, Geisinger and Holmes (39) in the estimation of Vitamin A with antimony trichloride oxycholesterol produces a blue color with antimony trichloride, but pure cholesterol produces no color.- Even in the hign concentration tested, it had no effect. However, Kann and Yang suggested (hO) that the maximum absorption of the product of the reaction between antimony trichloride and cholesterol occurred at 620 mp with sufficient strength to interfere with the estimation of vitamin A. EXPER IIIIIN’I‘ AL MET HOD I. Storage of Vitamin A in the liver of Rats fed various dog foods. Weaned albino rats were fed the stock ration, containing 35% yellow corn meal; 25% ground wheat; 20% whole milk powder; 10% linseed oil meal; 12$ alfalfa; 2% brewers yeast; and 1% table salt, until the weight reached was hS-GO grams. Twenty-four were assembled into six groups of h each. Whenever possible two of the same sex were placed in the same cage. Each group was fed with one kind of dog's food and water ad libitum daily. Twice the amount was fed on Saturdays in order to avoid feeding on Sundays. In the first series three XVIII dry dog foods, Gaines Krunchon, Gro-Pup (Ribbon), and Gro-Pup (Pellet) and three canned dog foods, Ken-L- lation, Rival and Pard were used. The animals were weighed once a week. At the end of 7 to 8 weeks they were sacrificed with ether and the livers removed. The livers were washed, dried with blotting paper and weighed. They were kept in the refrigerator until the time of the assay. The same procedure was repeated with another group of rats on the same diets. Duplicate feeding of twenty four more was also done with another series of dog foods namely Gaines Meal, Gro-pup (Ribbon), Gro-Pup (Cooked Heal) and Gro-Pup (Uncooked Neal) and with the canned foods, Red Heart and Ideal. Six samples of livers from dogs on the diets, Gro- Pup (Ribbon), Pard, Red Heart, Ideal, and Ken-L-Ration were sent over twice from.the Kellog kennels. The livers were weighed and kept in the refrigerator until used for analysis. The vitamin A assay for the livers was carried out by the modified Carr-Price method (2h). Four to six samples were analyzed each day. A. Equipment A cenco electrophotometer with red filter (6&5 mp) wr- 7r 4\IA was employed for the determinations. Amber flasks and separatory funnels were used to avoid the effect of light. ( The lubrica Wn used f r the stepcock was prepared by mixing 9 grams of soluble starch with 22 grams of glycerol, heating to lhOOC and decanting after one-half hour. This grease is well suited for use w’th ether.) The flaslzs, funnels, test tubes, pipets and cuvettes used during the drying, diluting, and reading of the transmission were dried in the oven at 1000C. It was necessary to have dry glassware because antimony trichloride in the presence of traces of moisture will r” produce cloudiness which interferes with the determinations. A rapid delivery pipet was prepared for the addition of the antimony trichlor i3 e reagent. It produced a thorough mixing of tie reagent and the solutions to be tested within 1 to 2 seconds. 3. Reagents Ether. The diethyl ether used was tested for peroxide by shaking 20 ml. of ether with 5 ml. of a w..!.. - ‘ _ 1 __ [J (I I, mixture of e;ual volumes of fresnly prepared )0» x and l} alcoholic phenolphthalein. The presence of rm peroxide was indicated by a red color produced. The peroxide was removed by wa hing the other with aqueous XX NaHSO3 solution. Anhydrous sodium.sulfate. The anhydrous sodium sulfate used should not retain vitamin A. It was checked by shak’ng the anhydrous Na2SOh with about 50 ml. of ether containing cod liver oil with about 100 U.S.P. units of vitamin A. The ether was decanted as completely as possible and the residue washed several times with 10-15 ml. portions of ether. 2-3 ml. of saturated antimony trichloride in chloroform was added to the sodium.sulfate. If no color developed the sodium sulfate was considered satisfactory for use. The salt was kept in the oven at 1000C for a few hours, to assure efficient drying. Chloroform. On prolonged standing chloroform slowly decomposes with the production of phosgene. The latter rapidly destroys vitamin A. To avoid this trouble it is desirable to use the reagent grade and redistill at weekly intervals. The redistilled chloroform was allowed to stand over C.P. granulated CaCl2 to remove moisture. Ho rubber stoppers were used because materials may be extracted which react with antimony trichloride to give a blue green color (30). The intensity of the color unlike that of the vitamin A reaction, increases during the first few minutes. XXI C. Sampling For the first series of dog livers, duplicate samples of about 10 grams were removed from the same portion of each liver and weighed. In the second series, the whole liver macerated in a daring blender and 10 ' ..n .1: 0 .0 gram samples OI the homogenpus material used Tor each analysis. In the case of rats the entire liver was used for each determination. D. Saponification Each sample was placed in a 250 ml. amber flask. A volume of 503 KOH solution equal to the number of grams of the wet sample was mixed with six times of its volume of ethyl alcohol. The alcoholic alkali was added to the flask and heated on a steam bath for two hours. The completeness of the saponification was indicated by the production of a clear solution when a few ml. of water was added and the mixture shaken. E. Extraction The saponified mixture was cooled to room.temperature. 50 to 100 ml. of distilled water was added to the flask and cooled again after which the solution was transferred to a 500 ml. amber separatory funnel. The saponification flask was rinsed with a volume of ether equal to twice the volume of alcoholic alkali used in the saponification, but not less than 50 ml.and the mixture added to the XXII above separatory funnel. The separatory funnel was shaken cautiously, especially at the beginning and the stopcock opened carefully at intervals to release pressure. The funnel was then placed on a rack until the layers had clearly separated. Too vigorous shaking .usually caused an emulsion which was hard to separate. The addition of a few ml. of alcohol and water was found to be helpful in breaking the emulsion. Then the aqueous layer was drawn directly to a second separatory funnel ,ft in the first. The saponifica- and the ether extract 10 nsed with 35 ml. of ether and the M tion flask was again r rinsing added to the second separatory funnel. The mixture was carefully shaken and the two phases allowed to separate. The aqueous layer was drawn off and added to the main part of the mixture. The ether extract was transferred to the first separatory funnel. The extraction of the lower portion was repeated 3 times as stated above. The final aqueous layer was removed and disccrded. 50 ml. of water was poured through the combined ether extracts in the first sepa atory funnel without shaking. The aqueous layer was then drawn off and discarded. 50 ml. of 0.5 N KOH was added to the ether solution and the separatory funnel shaken gently to 1 remove the acid soaps which are soluble in ether. After I complete separation of the two layers, tee aqueous layer was drawn off as completely as possible and discarded. XXIII To remove the alkali, the ether solution was repeatedly shaken gently with 50 ml. of water. Five such washings vere required to remove the last traces of alkali as 0 determined by testing With phenolphthalein. F. Solvent removal The ether extract was filtered into a dry 100 ml. amber flask through several grams of anhydrous NaZSOk distributed on a filter paper. The funnel containing the NaQSOLL was rinsed twice with 15 ml. ether and the nsings added to the 500 ml. flask. To prevent bumping H. r a few glass beads were placed in the flask and the ether was evaporated over a steam bath in a hood. The flask was removed from the steam bath before the ether was completely evaporated in order to avoid possible oxidation of vi amin A. The small amount of ether which remained in the flask was removed by suction whereupon the chloroform was immediately added to the residue to make up a volume of 10 ml. in a volumetric flask. Two dilutions with a concentration between 7 and 18 U.S.P. units vitamin A per ml. were made for the color reactions. G. Reading of the unknown The cuvettes were first checked in the electro- photometer. Those with the same transmission of light were selected for subsequent use. The ralvanometer was set to zero, before turning on the light. 2 ml. of J i O n O 1 V O reptilled chloroxorm was intrOducee into a cuvette, XXIV placed in the instrument and a ml. of antimony trichloride reagent was added from a rapid delivery pipet. This served as a blank and the galvanometer was set at 100; transmission. To the second cuvette placed in the , instrument, 1 ml. of chloroform, 1 ml. of the unknown and h ml. of the antimony trichloride reagent were added from the rapid delivery pipet. The galvanometer was read at the temporary pulse point about 2-k seconds after the addition of the reagent. The per cent transmission of the two dilutions of each sample was obtained and recorded. H. Preparation of the calibration curve A U.S.P. vitamin A reference oil in a gelatin capsule containing a solution of crystalline vitamin A acetate in cottonseed oil was used as the standard. The vitamin potency was 2500 units per capsule. The whole capsule was put in a 250 ml. amber flask and soaked in a few ml. of warm water until the gelatin was softened. A volume of 5 ml. 503 KOH mixed with 30 ml. of alcohol was added to the flask and the contents saponified over the steam.bath for 30 minutes. The extractions with ether, washing and drying over anhydrous reason and evaporating were done as with the samples. A series of dilutions in duplicate with chloroform were made to obtain concen- trations of 25, 20, IS, 10 and S U.S.P. units per m1. «71 1h The per cent transmission of 1 m1. aliquots of these solutions was determined. The values were plotted against the corresponding unitages of vitamin A on a semilogarithmic paper. This curve was used as a standard to calculate the concentration of the unknown. A new calibration curve was made whenever a new batch of antimony trichloride reagent was used. II A. Comparison of the antimony trichloride and G.D.H. reagents for determining the vitamin A content of rat livers. Four rats, 2 males and 2 females, were fed the new Gro-Pup (Meal) for 8 weeks, then sacrificed. The livers were used for a compar’son of antimony trichloride and G.D.H. methods. The method of saponification, extraction, solvent removal and the procedure for the antimony trichloride reaction were the same as stated in Part I. The G.D.H. reaction was carried out as follows: To 1 ml. of chloroform containing 8-15 I.U. of vitamin A in a 10 m1. glass steppered graduate was added h ml. of activated glycerol dichlorohydrin. Then the mixture was shaken vigorously and its transmission read in a Cenco electrophotometer with a green filter (530 mp) XXVI 2 minutes from the time of mixing. A blank consisting of 1 m1. of chloroform.and a mi. of glycerol dichloro- hydrin was used for the 100$ transmission setting of the instrument. The reaction was also carried out with standard solutions containing 5 to 25 I.U. of vitamin A and a curve of the per cent transmission versus concentration was constructed for reference. B. Cholesterol content of the rat's liver. The remaining extracts from above were used for cholesterol determinations. To 5 ml. of the sample, 2 m1. of redistilled acetic anhydride was added and .well shaken. Then 0.1 ml. of concentrated sulfuric acid was added. Exactly 15 minutes after the addition of sulfuric acid, the transmission was taken by using the Cenco electrophotometer with the red filter (6&5 mp). A blank was prepared with 5 m1. of chloroform, 2 ml. of acetic anhydride and 0.1 ml. of concentrated sulfuric acid and set the galvanometer at a transmission of 100 per cent. Standard solutions containing 0.1 to 0.8 mg. of cholesterol per ml. in chloroform were treated in tie same way and a reference curve was constructed with per cent of transmission against the concentration of cholesterol of a semilogarithmic paper. XXVII C. Comparison of the interference of cholesterol in the two methods. 1. The U.S.P. vitamin A reference standard in a gelatin capsule was saponified and extracted as in Part I. Dilutions were prepared containing 5, 10, 15, 20, and 25 I.U. of vitamin A per ml. in chloroform. To 1 m1. of each dilution of vitamin A 1 m1. chloroform contained 2 mg. cholesterol was added and the vitamin A determined with antimony trichloride as before. In a similar way measurements were made using the G.D.H. reagent. 2. In compari.g the effect of different amounts of cholesterol on the same concentration of vitamin A, 0.1, 0.2, 0.k, 1.0 and 2 mg. of cholesterol in chloroform were added to samples of 10 I.U. of vitemin A in the same manner as before (0.1.). Transmission values were determined by uSing the antimony trichloride and the G.D.H. reagent. RESULTS AND DISCUSSION From the analyses it was found that the livers of rats fed Gaines Heal contained the highest concentration of vitamin A and Pard contained the lowest value. The decreasing order of vitamin A content with 11 kinds of dog food was Gaines Meal, Red Heart, Gaines Krunchon, Gro-Pup (Cooked), Gro-Pup (Ribbon), Rival, Ken-L-Ration, XXVIII Gro-Pup (Pellet), Gro-Pup (Uncooked), Ideal and Pard with the corresponding vitamin A values of 782, 719, 5k2, 306, 2A1, 232, 148, 129, 81, 71, 37 U.S.P. units of vitamin A per gram of liver respectively. The data are given in table I. In comparing the vitamin A potency of 5 kinds of canned food and Gro-Pup (Ribbon), the foods used in the feeding experiments with dogs shown on table II, the order of increasing value was Pard, Ideal, ien—L-Ration, Rival, Gro-Pup (Ribbon) and Red Heart. Essentially similar results were obtained with rats and dogs of series 2 except that Rival and Gro-Pup (Ribbon) were in the reverse order. However their values were quite close to each other in the livers of the rats. The order of vitamin A storage in the livers of the dogs in series 1 was quite different except that Red Heart . was still the highest. The disagreement in the results of the two series of dogs may be due in part to the uneven distribution of vitamin A in the livers. In es 1, random sampling of the livers was done, whereas I—Jo 81" m n the case of series 2 homogeneous mixtures of the Po whole livers were prepared in a Waring blender. Gallup and Hoefer (32) in comparing the values from different sections of the same liver with those of the well mixed sample, found an uneven distribution of vitamin A in different parts of the liver. However, Rouir (kl) noted ICCIK Table I Vitamin A content of the livers of male and female rats on different dog foods Exptt7Ration Ho.of Expt.period Size of liver Average vitamin A Ho. Rats week Average storage h Fihale Female ggm. _per gm. liver 1 HaleL Female Male 1 Female 1 Gaines 2 2 7 8 12.8 9.0 514 A76 Krunchon 2 " 2 2 8 8 10.8 9.2 684 k96 3 Gro-Pup (Ribbon) 1 2 7 8 11.9 7.1 15 151 t- " 2 2 8 8 10.2 7.5 190 191 S " 3 l 7 7 12.0 8.3 109 99 6 " 1 3 7 7 9.8 8.0 355 509 7 Gro-Pup Pellet) 2 2 7 8 12.0 8.1 166 132 8 " 2 2 8 8 11.3 7.9 1&3 76 9 Ken-L Ration 2 2 7 8 10.6 8.0 86 1zk 10 " 2 2 8 8 11.3 7.9 20k 2 9 11 Rival 2 2 7 8 9.1 6.6 17k 251 12 " 2 2 8 8 9.k 7.7 222 281 1 Pard 2 2 7 8 12.3 8'0 26 57 1+ ” 2 2 8 8 9.9 .3 22 T2 15 Gaines Meal k 7 11.8 838 16 " a 7 9.1 726 17 Gro-Pup H (Cooked) 2 2 7 7 12.5 2.6 369 A39 18 " 2 2 5 5 9.2 .6 214 209 19 Gro-Pup (Uncooked) 2 2 7 7 12.0 8.2 107 89 20 " 2 2 5 5 9.2 6.0 65 63 21 Red Heart 2 2 7 7 11.9 7.8 558 52a 22 " 2 2 7 7 10.1 7.1 cam 952 2 Ideal 2 2 7 7 12. 8. 68 6 2i " 2 2 7 7_ 9.8 7.8 56 9 Table II A comparison of the liver storage of vitamin A in dogs and rats No. Ration fit. of liver Vitamin A Sterage gm. 1 U.S.P. units Dogs Dogs Rats; gper liver ggper gmiliver SeriesSeries {Dogs *Dogs Rats Dogs *Dog§'—R§t§’ f 5 1 II T; ,f I g» 11 I II . 1 Pard 153 155i1o 2769 910 365 18 5.8 37 2 éIdeal 101 1953 9.k:10739 1279 677 105 6.5 71 3 Ken-L- i :Ration 10k 230 9.k;12k09 2892 1686 179 12.5 1&8 k :Rival 66 181, 8.1 1804 5125 1892 p27 28 232 5 :Gro-Pup . I I 1(R1bbon) 130 265; 9.k 1099 A616 2097 8.k 17.3 2&1 7 ‘ p .‘ 6 see Hean 103 112' 9.1351711 9313916581 111 539 112 KKKI that the values of vitamin A in the liver of different dogs varied from 18-78 micrograms per gram flesh tissue but the variations in seaples from the same animal fell within the range of experimental error. In the case of rats fed Gro-Pup (Cooked Ileal), the average vitamin A content of the livers was 3 or k times higher than those fed Gro—Pup (Uncooked heal). This may be due to the difference in the content of vitamin A at the time of feeding, the uncooked meal having been prepared at an earlier date permitting greater loss of the vitamin by oxidation. It is quite apparent that large variations occurred in the storage of vitamin A in rats fed various commercial dry and canned dog foods. In view of the fact that all of ‘he brands used in these studies contained enough to permit some storage of vitamin A in the livers it may be concluded that all of the foods contained an adequate supply of this v'tamin for growth and maintenance. It may well be however that in cases of greater physiological demand such as in reproduction and lactation certain of the brands of dog food would have failed to meet the requirements for vitamin A. In comparing the two reagents for the production of color, he values obtained with the G.D.H. reagent were 13.3-1M.7 per cent lower than those with antimony ICC/{I I trichloride as shown in table III. Although this difference is appreciable, certain advantages in the use of the G.D.H. may justify its selection for comparative Vitamin A studies. Knen 2 mg. of cholesterol was added to standards of various vitamin A concentrations, the G.D.H. method gave values from 12 to 23 per cent lower than without cholesterol. The data are given in table IV. The effect of cholesterol was greater at the low levels. At 5 to lifferences were 20 F" 10 I.U. of vitamin A per ml., the to 23 per cent and at 15 to 25 I.U. per ml., 12 to 12.5 per cent. In the case of the antimony trichloride reaction only the lowest level of vitamin A, that is, S I.U. per ml. showed an effect. The value found was hfl higher than with vitamin A alone. Enen amounts of cholesterol from 0.1 mg. to 2 mg. were added to a definite amount of vitamin A, 10 I.U. per ml., there was negligible effect with the antimony trichloride reagent. However, the values obtained with the G.D.H. reagent were lower. For 0.1, 0.2, O.h, 1.0, and 2.0 mg. of cholesterol, the percentage decreases . ,- - ,4 , . 7 were 3%, 3}, 126, 11; and 22fi respectively. The results are shown in table V. The cholesterol content of the livers of h rats on new Gro-Pup Meal were 12.9, lh.7, 11.3, and 12.3 XXKIII Table III A comparison of vitamin A storage in the rat's liver on new Gro-Pup Heal determined by antimony trichloride and activated dichlorohydrin reagents (Experimental period -- 8 weeks) Rat LiveerbCl3 reagent G.D.H.reagent Ho. Sex weight: USP units . USP units 5 Per cent ;, gm. _perggm liver '_per_gm liver 3 lower 1 5 i 15.351 592 ' 508 ? 1t.2 2 L 6 1A.18; 6A1 550 , 1A.2 l E 3 Q 9 oil-6 Ell-9 : 14-58 9 l’Jr- 7 I 3 _; _ h jg_( 8.655 560 9 A85 ' 13.3 Table IV Interference of cholesterol on vitamin A determination using U. S. P. vitamin A reference standard extract Expt. Vitamin A Vitamin A content per ml.sample No. USP units after 2 mg. cholesterol added .per m1. sample SbCl3 reagent GDH reagent GDH method ' per cent ; lower 1 5 5.2 h.o -2o 2 10 10.0 7.7 ~23 A 20 20.0 § 17.5 -12.5 25.0 22.0 I -12 U1 N \J'L .meav Table V A comparison of the effect of graded cholesterol on the determination of 10 U.S.P. units of vitamin A on reference standard extract with SbCl3 and G. D. H. reagent Expt.Cholesterol Vitamin A content U.S.P. units per ml. No. per ml.sample SbCl3 reagenti . G.D.H. reagent. ; mg. :Vitamin content fl‘dev1ation l 0.1 10.1 . 9.7 -3 2 . 0.2 10.0 9.7 '3 3 i 0.4 10.0 I 8.8 -12 A ! 1.0 10.1 g 8.9 -11 _5 2.0 ( 10.0 _j 7.8 i -22 Table VI Cholesterol content in the rat's liver Rat fit. of liver Cholesterol Cholesterol Cholesterol per No. gm. 3 per liver iper gm liverf ml. sample r mg. mg. mg. 1 15.35 g 198 12.9 0.39 2 1A.18 f 209 1A.17 0.A1 3 i 9.48 § 107 L 11.3 g 0.21 A l 8.85 107 i 12.3 i 0.21 Opwrvfi 4. I. "L 11 mg. per gram of liver (table VI). The dilution commonly ct ion of vitamin A is l to 500 ml. :3 CD CD OJ 1.. O m C F ,4 (J (D 0) Cl" E. 1a According to this dilution, there were 0.39, 0.ul, 0.21 and 0.21 mg. of cholesterol per ml. of sample used. These amounts of cholesterol had no effect in the estimation of vitamin A with antimony trichloride. The values were from 3 to 12% lower in the case of the J. G.D.H. reagent. It may therefore be that the low values obtained with the G.D.H. reagent are due partly to cholesterol interference altho there may be other factors which have not been found. According to Sobel, hayer and Kramer (30) cholesterol gave no color with the G. D. H. reagent. They did not however, determine whether the addition of choles ero l to vitamin A might affect the development of color. This may, therefore, account for the differences observed. SUIEIARY 1. The storage of vitamin A in the livers or rats fed commercial dog foods was determined by use of the Carr-Price reaction. The results of decreasing order of vitamin A content per gram liver with the various dog foods was Gaines Heal, Red Heart, Gaines Krunchon, Gro-Pup (Cooked Leal), Gro-Pup (Riboon), Rival, Iien-L- q. Hat ion, Gro-Pup (Pellet), Gro-Pup (Uncooked Heal), 3. XXXVI Ideal and Pard. '2‘] ive minds of canned foods, Pard, Ideal, Ken-L-Ration, J Lival, and Red Heart and the dry food Gro-Pup (Ribbon) )— were used in the feeding experiment with dogs and showed essentially similar results as with rats except that Rival and Gro-Pup (Ribbon) were in the reverse order. In comparing the Carr-Price with the G.D.H. (glycerol dichlorohydrin) reagent the latter was found to give values 13.3 to 1A.? per cent lower. The influence of cholesterol on the vitamin A determination was studied with both methods. The G.D.H. method gave values from 12 to 23 per cent lower in the presence of certain amounts of cholesterol. The effect of cholesterol was greater at he low level of vitamin A. Tith antimony trichloride, there was no effect on the vitamin A determination when the sample contained 0.1 - 2.0 mg. cholesterol per ml. 3. 1- The cholesterol content of the livers of rats was 12.9, 1h.7, 11.3, and 12.3 mg. per gram of liver. 0n the basis of the dilutions made this would pr vide 0.39, 0.hl, 0.21, and 0.21 mg. of cholesterol per ml. of chloroform solution used for analysis. This amount vould have no effect on the vitamin A determination with antimony trichloride reagent but would largely account for the lower values obtained with the G.D.H. reagent. F0 3. ‘P‘Pflr’f AHUXIIII BIBLIOGAAPLI Sherman and Boynton Distribution of vitamin A in the body. I / J. Amer. Chem. Soc. a7:l& LO (192F) Hoore, T. The distribution of vitamin L and carotene in the body of the rat. Biochem. J. 25:275 (1931) Davies, uJI., and hoore, T The distribution of vitamin A in the organs of the normal and hypervitaminotic ra Biochem. J. 28. 218 (193“) Hattson, F.H., Hehl, J.Y. and Deuel, H.J., Jr. The site of conversion of carotene to vitamin A in the rat. Arch. Biochem. 15:65-73 (19A?) Uiese, 0.3., Hehl, J.W., and Deuel, H.J., Jr. The in vitro conversion of carotene to vitamin A in tae intestine of the rat. Arch. Biocheu. 15:75-79 (19k7) Gray, E.L., Imic .1an, K.C. D. and Brown E.F. The state of V1 't mmin i in the liver of the rat after feeding various forms of the vitamin. J. Hutri. 19:39 (1M9 0) Baumann, C.A., Riisinm h.I.. and Steenbock, H. The absorption and storage of vitamin “ in the rat. J. Biol. Chem. 107:705 (193 1) Lewis, J.M., Bodansky, 0., Falk, K.G. and Hcguire,G Vitamin A requirement in the rat. The relation of vitamin A intake to growth and to concentration of vitamin A in the blood plasma, liver and retina. J. Hutri. 23: 351 (l9h2) Radice, J.C. and Herriaz, M.L. Absorption and storage of vitamin A in the rat; macroscopic and microsc0pic study by fluorescent light. Hutri. Abst. and Rev. 17:828 (19k8) 10. ll. 13. lb. 16. 17. ~r*'"x"v 4L. L1\ II Lewis, J.H, Bodansky, 0., Falk, K.F. and chuire,G. Vitamin A storage in liver. Proc. Soc. :xptl. 3101. Med., A6:2L8 (19A1) Caldwell, A.B., Macleod G. and Sherman, H.C. Bodily storage of vitamin A in relation to diet and age studied by means of antimony trichloride reaction using a photoelectric colorimeter. J. Hutri. 30:3L9 (19M HcCoord, A.B., and Luce- -Clausen, E.M. Variation of vitmnin A content 01 rats with age. J. Nutri. 7:537 (193L Sobel, 2., Sherman, K., Lichtblau, J., and Kramer,B. Comparison of vitamin A liver st01 age following administration of vit.min A in oily and aeueous media. ‘- 0 r-’ v—I J. flutrl. 39:22; (19A8) Brenner, S., Brookes, U.G. H., and Roberts, L. J. The relation of liver stores to the occurrence of early signs of vitamin A deficiency in the white rat. J. Nutri. .A-59 (19‘s) Glover, J., Goodwin, T.U. and Korton R.A. Relationship between blood vitamin A levels and liver stores in rats. ‘ Biochem. J. Alz97-lOO (1947) Sherman, H.C. and Storms L.B. Variation of the vitem lin A content of rats with age. J. Am. Chem. Soc. 17: 16 F3 (192F) Davis, A.W., and Moore, T. The influence of the vitamin A reserve on the length of the depletion period in the young rat. Biochem. J. 31:172 (1937) Symposia XII I-II, 19A? CU Pa O [.4 O C ”3 H. O r.) H ’|_1 herman, H.C. and Hunsell, H.1. The quantitative determination of vitamin A. J. Amer. Chem. Soc. 1039 (192F) Guggenhein, K. and Koch, 3. Liver storage test for the assessment of itamin A. Biochem. J. 38:256-60 (lgAAj 22. 28. 29. 30. ‘1.-f*:I:C AAAAA Fey, J.R. and Iorgareid , Biological assay of vit1"1 in Analytical chemistry 20 Wz30 Carr, F.H. and Frice, 3.1. Colour re1ctione attributed to vitamin A. Biochem. J. 20:19? (1926) Morris Dean Dnbree thsicochemic9l aSS9y of vitzmin A. ‘ Ind. Eng. Chem. anal. Ed. 13:1RU (1911) Dann and Evelyn The determination of vitamin A with the photoelectric colorimetcr. Biochem. J. 32:1008 (1938) Sobel, A.E. and Ucrbin, H. Spectrophotometric study of a new colorimetric reaction of vitamin A. *1 {’3 ‘r" J. Biol. bhem. 159:001 (1919) Sobel, A.E. and Ierbin, H. Activated glycerol dichlorohvdrin: A new colorimetric reagent for vitamin A. Ind. Eng. Chem. Anal. Ed. 18: S7O-V73 (1913.'3) Sobel, A.E. and Snow, S.D. The estim1tion of serum vitamin A with activated glvcerol dichlorol fdrin. J. Biol. Chem. 171: 617-632 (1917) Wall, M.E. and Killey, E.G. Determination of vitamin A ester in fortified poultry mashes with activated glycerol dichloro- hvdrin. Anal. Chem. 20: 757 (1918) Sobel, A.E. and Herbin, H. Determination of vitamin A in fish liver oils with activated glycerol dichlorohydrin comparison with Spectrophotometric and antimony trichloride methods. Anal. Chem. 19:107-112 (19‘?) Sobel, A.E., Mayer, A.M. and Kramer, B. New colorimetric reaction of vitamins D2 and D and their provitamins. Ind. Eng. Chem. Anal. Ed. 17:160 (1915) 31. 33. 37. XL Fuchs, L. and 8005, E. l. The rate of change of free and esterified vitamin A in various solvents. A contribution to the photometric and spectrographic estimation of vitamin A. 1. 2. Changes and anomalies in ultraviolet absor- ption of vitamin A concentrates and in relation to the Carr-Price reaction. 2. Nutri. Abst. and flev. 18:59 (1918) Gallup, U.D. and Hoefer, J.A. Determination of vitamin A in liver. Ind. Eng. Chem., Anal. Ed. 18:288-90 (1916) Bernard, L. Oser, Daniel Melnick and Morton Pader Chemical and physical determination of vitamin A in fish liver oils. Ind. Eng. Chem., Anal. Ed. 15:71? (1913) Bolomey, R.A. Oxidative decomposition of vitamin A. J. Biol. Chem. 189:323—329, 331-335 (1917) J.R. Edisburg Spectrophotometric Assay of vitamin A with special reference to margarine. Analyst 65:181 (1910) Tastaldi, H. Photometric estimation of vitamin A. l. The Carr-Price reaction. 2. Effect of various factors on Carr-Price reaction. 3. Natural interference with the Carr Price reaction. Nutri. Abst. and Rev. 18:59 (1918) Bernard, L, Oser, Daniel Melnick and Morton Pader. Estimation of vitamin A in food products. Ind. Eng. Chem., Anal Ed. 15:721 (1913) Johnson, R.M. and Baumann, C.A. ‘tudies on the reaction of certain carotenoids with antimony trichloride. J. Biol. Chem. 189:83-90 (1917) 39. XL I Corbet, R.E., Glisinger, H.H., and Holmer, H.N. Substances which interfere w’th antimony trichloride test for vitamin A. J. Biol. Chem. 100:857 (1933) K1nn, U. and Yang, E.F. Cholesterol in the Carr-Price reaction for the determination of vitamin A. Nutrition Abstracts and Reviews 18:60 (1918) Rouir, E.V. Distribution of vi tamin A in the liver of the dog. Nutri. Abst. and 18V. 18:86 (1918) I”EB 5 "57 , T612.015 ' L477 Lee 217602 )li'lllHllll'llllHIIIHIIIUIIIIHIIHHIIHIIIII)\llliimll 31293 02446 7973 . ._. - -.un‘_glg_‘l4_.a.1. . .~.