’Iy‘ \J“ NI k M MIN H x I 137 210 .THS. A STUQY OF THE PREPARATION OF THE UNSAPONEFIABLE MATTER OF FORTEHED MILK SUITABLE §IOR THE SPECTROGRM‘HIC DETERMWMTEON 0F ‘f'E‘E‘AMIN D T113333 fen" $129 Eegwa 65 M. .5. Qiifiifiiflzii‘é STfiiTE CGLL§GE Sing Few thmg ‘E‘Q‘é'fi? _‘ I. _ .-.-.'—.-— 1’7- __ I This is to certifg that the thesis entitled ’A Study of the Preparation of the Unsaponifiable Matter from Fortified Milk Suitable for the Spectrographic Determination of Vitamin D.‘ presented bl] Sing Pao Chiang has been accepted towards fulfillment of the requirements for 4.5;— degree in Chemi at m 0 Q Mfr/”WY Majur prulgésfiir Date May 270 191‘9 0-169 4.3; "-44 A- 4A A STUDY OF THE PREPARATION OF THE'UNSAPONIFIABLE MATTER OF FORTIFIED MILK SUITABLE FOR THE SPECTROGRAPHIC DEIERMINATI ON OF VITAMIN D By Sing Pee Chieng A THBIS Submitted to the School of Greduete Studiee of Michigan State College of Agriculture end Applied Science in partial fulfillmmt of the requiremente for the degree of ‘ MASTER OF scxmcn Department. of Chemie try 1949 ZELEHBTRY DEPr. u——- / v-‘a kn, I //_3 ~,~'r ACKNOWLEDGMENT The writer wiehee to expreee her eincereet epprecietion end gretitude to Dr. C. A. Hoppert, whose euggeetione end guidance made possible the completion or this work. titttttt *ttttt *it# it e: 216940 TABLE OF CONTENTS HISTORICAL AND INTRODUCTIONOOOOO0....OOOCOOOOOOOOOOOOOOOOOCO 1 EXPERIMENTAL PROCEDURE I. Preparation of the Unseponii’iable Matter.....................'................ 7 II. Removal of Interfering Substances and Spectrographic Study of the Prepared Unsaponifiable Matter...................... 8 IIIe Biological TCBt at Various Stageaeeeeeeeeee 8 IV. Determination of Fat, Unseponifiable Matter and Cholesterol of Some Vitamin D Milkleeeeeeeeeeeeeeeeeeeeeeeeeeeeee'eeeeeeee 10 RMTSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 12 DISCUSSIONeeeeeeeeeeeeeeeeeeeeeeeeoeeeeeoeeeeeeeeeeeeeoeeeee 15 MMYOOOOOOOOOOO0.00.0.0...OOOOOOOOOOOCO...0.000.000.0000. 17 BIBumR-WCCOOOOOOOOOOOOOO0..0......OOOOOOOOOOOOOOOOOOOOOO 18 e e O 2 I I 0 e I I O ' O O O I O . e I . e O Q . O a O a O I O O eovweOIeOOlOe I e. I 00 .00v HISTORICAL AND INTROIITCTION Among common natural food-stuffs Vitamin D occurs in some quantity only in eggs and milk. The amount they contain, however, is insuffi- cient to meet the nutritional requirements of infants and children. Consequently cod liver oil was used until relatively recent years to assure norml developmait-in infants and children. Because of the ease of providing vitamin D to infants directly in milk and because it has been demonstrated that milk is a better menstruun than oily carriers, such work has bem directed toward investigating the special role which the milk constituents might play in this connection. Supplee (l), and his associates reported that the vitamin D carried by the lactalbumin of milk showed greater effectiveness than that fed alone. They postulated that the vitamin D and the lactalbumin had formed a sym- plex, a system of a prosthetic group and a o'olloidal carrier, vitamin D being the prosthetic group and the lactalbumin the carrier. Thus the process of enriching fluid milk with vitamin D was started and so-called ”Vitamin D Milk" is now widely sold throughout this country, the potency Inving been standardised at 400 U.S.P. units per quart. There are several methods of enriching milk with Vitamin D. Most practical is that of the direct addition of a vitamin D concentrate. At first fish liver concentrates were used exclusively although in re- cent years activated ergosterol has come into extensive use. A solution of vitamin D in prepylene glycol or some other suitable solvent, is sim- ply added to yield milk of a standard potmcy. -1- I. A second method involves feeding irradiated material to the milk cow. Because of the high content of ergosterol in yeast, irradiated yeast has been fed to achieve a milk of standard vitamin D content. This method has not gained wide acceptance because of the inconvenience of adjusting the allowance of yeast to the level of milk production d’ each cow. In a third method milk may be activated directly through the ex- posure of a thin .film for a brief period to the active rays of ultra violet light from artificial sources. This type of milk was originally standardised at 135 units per quart, but for the last decade has attain- ed the level cf other milks. mmely 400 units. This method, although still used by a few ’of the larger dairies, has failed to gain popularity because of the high initial cost of the equipment and the difficulty in maintaining productim of milk of uniform potency. It was found (2) that in general all these types of milk are effec- tive to the utmt of their unitage, for both the prevention and cure of rickets. However, the results on chicks (3) dcnonstrated that "irra- diated milk" was approximately ten times more effective than the same lumber of rat units of “yeast milk“. It is now well known that vitamin D2 is much less effective in poultry than 03, although there is very little difference in the case of animals, including humus. Milk has long been subjected to rigorous control by state and local health authorities. It was only natural then that when the fertifica- tion of milk was permitted and introduced, steps would have to be taken to periodically check the vitamin D content of such milk. The exercise of this control work varies greatly. In the absence of a state law the commnities fix the number of assays which are required yearly. In the State of Michigan this varies from none to twelve. In states which do have a law for this subject two or three assays per year are required. The assaying of low potency mterials like vitamin D milk is not an easy task especially if one wishes to do it routinely and economic- ally. The biological method using either rats or chicks as test ani- mls is still the official method. It is both expensive and time con- suming imsnnch as each determination takes twenty-one days for the preparation of the rachitic rats and another ten days for feeding the test material. Consequently, a search for more practical and non-bio- logical method was soon mde. Within the last ten or fifteen years, a mmber of physico-chemical methods have been developed, each claiming good agreement with the bio- logical assay. Brochnmnn and Chen (19, 20, 21) discovered that vitamin D on addi- tion of a saturated solution of antimony trichlcride in chloroform developed a pink color with a characteristic absorption maximum at 500 nu. By using the appropriate correction factors it was found possible to eetiamte vitamin D in products of high potency. Milas, Heggie. md Reynolds (4) modified the above method by a preliminary treatment with maleic anhydride, thus destroying vitamin A, 7 - dehydro cholesterol and interfering carotenoids. Nield, Russell, and Ziminerli (5) used a new reagent consisting of a solution of antimony trichloride and acetyl chloride in chloroform. . . ’ v . ' Q ~ . ' O . i . .y . . I ' l . . . _ . _ r ' t . ' I ' e . o . _. ‘ V I . e ,_ ‘. ‘1 ‘ o ‘ Q . . ‘ a a . r l . . . , . . r r r . ' e , 4 A yellowish pink color was produced with a maximum.absorption at 500 mu. They claimed greater sensitivity for their method. Later the same authors (6) improved their reagent by adding a reducing agent, zinc or tin or antimony to reduce the concentration of pentavalent antimony, thus making it satisfactory for the determination of chol- esterol and other sterols as well as vitamin D2 and D3. Ewing, Kingsley, Brown and Emett, (7) removed carotenoids. vitamin.A and pigments by chromatographic adsorption. They used a two-step chromatographic treatment in which the 3 you was determined first for the combined vitamin D and sterols and then for the separated sterols. The value for the vitamin.was obtained by difference. Re- cently Ewing and his associates (8) inmroved the above method by passing the unsaponifiable fraction through a column of superfiltrol, a mined solvent composed of 50 parts of hexane, 10 parts of ethyl ether, and 1 part of absolute alcohol being used. The interfering substances were adsorbed, and vitamin D in.the percolate measured by its extinction at 265 mu. . DeWitt and Sullivan (9) modified the antimony trichloride method by using ethylene 'chloride as a solvent. This reath was found to be more stable and it gave a salmon pink solution exhibiting an absorp- tion.mlximum.at 500 mu. ‘ Sobel, beer, and Kramer (10) developed a neW'colorimetric method by using glycerol dichlorohydrin in the presence of acetyl chloride or other acid halides. Different color changes were obtained with calciferol, ergosterol, and 7-dehydrocholesterol. They claimed the reagent to be -4- a. more stable and the absorption curves more specific. Villar Palasi (ll) devised another color reaction for vitamin D by using saccharose and absolute alcohol. When concentrated sulfuric acid was added, the color first became red and then blue. Upon shak- ing, the blue color increased in intensity. The final sky-blue color was stable and its intmsity was found proportional to the amount of vitamin D. Most of the above methods. have been claimed to give satisfactory results for high potency materials. However, little success has bem had with those of low potency, largely because of the difficulty in the extraction and concentration of the vitamin D and subsequently the removal of interfering substances. Ken and Booth (12), (13), in studying the vitamin D activity of butter fat derived from ordinary milk, found that about 75% of the potency of autumn and winter butters was either destroyed or directed toward‘another fraction in the preparation of the non-saponifiable resi- due. In a study of the seasonal variation of the vitamin D content of cow's milk, Bechtel (14) had to resort to the concentration of vitamin D by warm alcohol extraction of the milk fat. This mde it possible to carry out the biological assay with a fair measure of success, and estab- lished that milk produced under common conditions of farm nanagement varied in vitamin D potency from 4 to 45 units per quart. The potency paralleled the exposure of cows to sunshine. It is quite likely that the butter fat used in the studies of Ken and Booth (12), (13) was de- rived from milk of fairly low potency. -5- Although failure attended the attempts to quantitatively recover vitamin D from.ordinary mdlk fat by the saponification procedure, it was nevertheless considered possible that in the case of fortified milk enough vitamin D was present to permit the method to be used. Accordingly it was the purpose of this investigation to apply a suit- able method of extraction of the fat from vitamin D milks of various types, to saponify, and to test biologically for the vitamin D at various stages in the procedure. .A quantitative study was also made of the amount of fat, total unsaponifiable matter, and cholesterol in the samples of milk which were obtained from various cities in Michigan. An attempt was also made to prepare the unsaponifiable matter from.a three quart sample of vitamdn.D:milk containing approximately 1200 U.S.P. units of vitamin D. It was hoped that this would supply a large enough amount of‘vitamin D for the spectrographic analysis of this vitamin. HPERITJENTAL PROCEDURE I. Preparation of the unsaponifiable matter A. Extraction of the milk fat 100 m1. of the milk was delivered into a 2-liter separatory funnel, 15 m1. of concentrated NH4OH added and the mixture well shaken to keep the protein in solution. 100 ml. 95% ethyl alcohol was added and the mixture again well shaken. Then 200 ml. of perox- ide-free ether was added, and the mixture carefully shaken. 200 ml. of Skelly-solve B was now added and the mixture again well shaken. After the two layers had clearly separated, the aqueous layer was re- moved and the extraction repeated twice. The combined extracts were transferred to a beaker and evaporated to dryness on a steam bath. This residie was used both for the determination of total fat and for the subsequent preparation of the unsaponifiable matter. B. Hydrolysis The extracted milk fat was transferred to an Erlenmeyer flask and 30 m1. ethyl alcohol 95% and 5 ml. 50% aqueous KCH were added. The mdxture was boiled for one hour under a reflux condenser. C. Extraction of the unsaponifiable matter After the hydrolysed mixture had cooled down, it was transferred to a separatory funnel. The Erlenmeyer flask used in the hydrolysis was mashed first with warm water, then with cold water, and finally ‘with ethyl ether. The washings were combined‘with the main hydrolysed mixture. The cooled solution was then extracted with seven 50 m1. por- tions of ethyl ether. Ench time after the addition of the other the rI separatory funnel ms slightly tilted to mix the two layers but avoid ,. the formation of an emulsion. In case an emlsion did form, one or two m1. of ethyl alcohol was added, and the anulsion broke within 2 or 3 minutes. The combined ether extracts were washed with distilled water until free from alkali as shown by testing with phenolphthalein. The washed ether extract was dried by pouring through a filter containing anhy- drous sodium sulfate. The solvent was then removed by evaporation and the residue weighed. II. Removal of Interfering Substances and Spectrographic Study of the Prepared Unsaponified Matter The unsaponifiable matter from three quarts of M.S.C. vitamin D milk prepared as above was dissolved in 10 ml. of a mixture of 50 parts of hexane, 10 parts of ethyl ether, and 1 part of purified ethyl alcohol. The solution was passed through a 9 on. column of superfiltrol prewashed with 15 m1. of the same solvent mixture. Vitamin D was eluted with the 50-10-1 solvent, and the aluant put under reduced pressure at 60°C. to remove the solvent. The residue was then taken into alcohol and its ab- sorption measured at 265 um with a Beskman quarts spectrophotometer. III. Biological test at various stages In orda' to test the procedure at each stage a check was made using the biological line test as a criterion. In each case these assays were ads on the original milk, on the extracted fat and on the unsaponifiable nutter. In the case of the chromtographic separation of interfering -8- substances biological assays were made on the solution before and after passing through the adsorption column. A. Preparation of test animals Albino rats weighing 45-55 g. (about 21 days of age) were made rachitic by feeding them a basal ration of the following: Cornmeal 72 Wheat gluten 20 Yeast 4 Ca 003 3 NaCL l The animls were put in separate cages and fed the rachitogenic diet for 21 days. Water and the basal ration were given ad libitum during this period. The animals usually gained from 20 to 30 gm. Those that gained less than 15 gms. were considered unsuitable and were discarded. B. Test period In each series the vitamin D milk was compared with equivalent amounts of the extracted fat and of the unsaponifiable utter. 10 m1. of milk containing 4 U.S.P. units of vitamin D was mixed with 40 g. of the basal diet, dried, and ground. All rate used throughout these tests were fed on this basis during the supplementary period. The extracted fat samples and the unsaponifiable matter were dis- solved in ethyl ether and a portion equivalent to 4 units was evaporated on 40 g. of the basal ration containing 3% milk powder. The milk was included in the assay of the extracts in order to sinulate the supple- ments of the rats receiving the milk. -9- The animals usually consumed the supplemented ration in 7 or 8 days and than were fed the basal ration for the remainder of the ten- day period. At least 7 animls were used for the assay of each sample and for each group of assays 4 negative controls were included. c. Line test On the tenth day of the supplemaitary feeding the animals were etherised and the wrist bones removed and cleaned of adhering tissue. The bones were kept in 95% alcohol for at least two days and then split longitudinally with a clean, sharp blade to expose a plane surface through the junction of the epiphysis and diaphysis. The sections were then immersed in 2% AgN03 solution for two minutes. Then they were put in distilled water, and exposed to light until the calcified areas were clearly distinguishable. The average response of the various groups was then evaluated. The extent of calcification reflects the amount of vita- min D contained in the samples. IV. Determination of fat, unsaponifiable matter, and cholesterol of some vitamin D milks. A. Fat Pat was extracted as previously described, followed by drying at 90°C. in an oven until constant weight was obtained. The drying was done at half hour intervals to minimise thermal decomposition. B. Unsaponifiable nutter The hydrolysis and extraction were as previously described. The unsaponifiable matter was also dried at 90°C. for half hour intervals until constant weight was obtained. C. Cholesterol Cholesterol was determined oolorimetrically. The unsaponifiable nutter was dissolved in 25 ml. CHCl3 (C.P. grade) and 6 m1. of this solution as taken for amlysis. Further dilution was rude if necessary. 2 ml. of redistilled acetic anhydried was added and the tube well shaken. In a similar way 0.1 m1. of concentrated £12304 was added. Exactly 15 minutes later the great color developed was measured in a photelometer using a red filter. From the pccent of transmission the amount of cholesterol cmld be read from a standard curve obtained simi- larly by using known concentratiom of cholesterol. The addition of the concentrated H2804 was usually made to a series of solution at 2 minute intervals so that successive readings on the photelometer could be completed at uniform times. -11- RESULTS A. The similarity in response of the various groups as judged by the line tests indicates that the extraction, as well as the saponifica- tion and extraction of the unsaponifiable nutter were quantitative with- in the limits of accuracy of the method employed. B. Complete recovery of vitamin D was also observed in the solution passed through the adsorption column and used for the spectrographic determimtion of vitamin D. Three fractions were collected and examined. None showed a distinc- tive absorption maximum at 265 m. This failure nay have been due to a combination of factors of which the presence of impurities in the solvents and the presence of excessive amounts of interfering substances in the un- saponifiable nmtter from the large sample of milk were perhaps the most serious. Time did not permit a solution of this part of the problem. C. The results of the determination of fat, unsaponifiable ntter, and cholesta'ol are shown in Table I. The fat content of the vitamins D milks examined varied from 3.21 g/100 ml. for the Central Creamery, Detroit, to 4.70 g/100 ml. for the Michigan State College Creamery. Most of the sampl. had a fat content betwem 3.50 g. and 3.80 g. The range for the unsaponifiable matter was from 40.4 mg./100 ml. for the Lansing Dairy Company, Lansing, to 75.0 mg./100 ml. for the Rebel Creamery, Detroit. When calculated on the basis of fat the varia- tion was from 1.06% for the lensing Creamery, to 1.97% for the Heather- wood Company, Lansing, as shown in Table I. The sajority of the samples examined had unsaponifiable matter values between 53.0 and 60.0 mg./ 100 ml. The cholesterol contents ranged from the 8.50 mg. per 100 ml. for the United Dairies to 16.0 mg. per 100 ml. for the 11.3.0. Creamery. When calculated on the basis of fat and of the unsaponifiable matter, the values were 0.222-0.436% and 14.4 - 35.7% respectively. Table I. Fat, Unsaponifiable matter, and Cholesterol Contents of Vitamin D Milks Fat Unsainatter Cholesterol_ “mp“ mg7 o/o mg/ ore ale 5 100 100 in 100 in unsap. No. Place Dairy m1 . m1. fat m1. fat matter 1 Detroit Rebel 3.99 75.0 1.88 12.8 0.321 17.1 2 Detroit Central 3.21 55.1 1.71 11.0 0.343 20.0 3 Detroit United 3.60 59.0 1.64 8.5 0.236 14.4 4 Detroit Detroit 3.56 68.6 1.93 13.3 0.374 19.4 5 DCbrOit R“ Obud 3.55 5708 1.63 12 e0 0.338 20.8 6 mrquOttO Northom 3e98 56.7 1.43 10.4 0.261 18e4 7 ‘Marquette Bancroft 4.28 62.8 1.47 9.5 0.222 15.1 8 Battle Creek Ashley 3.54 53.0 1.50 12.3 0.348 23.2 9 Grand Rapids Borden's 3.73 47.7 1.28 11.7 0.314 24.5 10 Lansing Lansing 3.81 40.4 1.06 12.6 0.331 31.2 12 Lansing Quality 3.57 55.8 1.56 15.2 0.426 27.2 13 JCCklon melid. 3e51 42e3 1021 15e1 0.430 55e 7 14 Jackson Loud and 3.51 53.8 1.53 15.3 0.436 28.4 Jackson 15 Jackson Servall 3.75 57.3 1.53 14.5 0.387 25.3 Jersey 16 Midland Midland 3.68 44.6 1.21 15.4 0.419 34.5 17 Pontiac Arctic 3.62 60.3 1.70 12.5 0.345 20.7 18 East Lansing M.S.C. 4.70 52.5 1.12 16.0 0.341 29.8 -14 - DISCUSSION The procedure used for preparing the unsaponifiable matter from vitamin D milk appears to allow complete recovery of the vitamin D as shown by the line tests at end of each stage. Ken and Booth (12),(13), did not get satisfactory results from their extraction probably due to the fact that they worked with materials of much lower potency. It my also be that the ethyl ether used in their extractions was not peroxide- free and might have contributed to loss of vitamin D. In the present study care was taken to use peroxide-free ethyl ether. Fritz, Halpin, Hooper, and Kramke (15) reported that the various forms of vitamin D, including crystalline activated ergosterol and 7 - dehydro-cholesterol as well as natural sources, are all susceptible to destruction. Avocado oil, which is being used increasingly in human diets, has been studied for its vitamin D content recently. Weatherby (16) concluded that this oil contains appreciable quantities of vitamin D. However, Lassen Bacon, and Sutherland (17) obtained negative results from the extraction of the unsaponifiable matter of the avocado. In view of these contradictory reports it would be of interest to apply the procedure used in this study to the avocado. The result of testing the unsaponifiable latter with the Becksmn quarts spectrophotometer after chromatographic separation of interfering substances proved to be disappointing. a number of reasons may be offered for the failure to find the characteristic absorption at 265 m. The ap- plication of this method to vitamin D concentrates involves such smaller quantities of the sample and consequently less possible interfering sub- stances. In the case of concentrates, considerably less than one gram -15- of the sample is needed. The use of 3 quarts of milk involved the processing of approximately 100 grams of milk fat. Obviously the quan- tity and nature of the interfering substances might be such as to make the chromtographic separation incomplete. Moreover, the interfering substances introduced by the use of the large volumes of solvents needed to process a three-quart sample of milk would be considerable. It is also quite likely that the amount of vitamin D supplied, namely about 1200 U.8.P. units, was insufficient for the spectrographic determination in its [resent state of develoment. It my be added that unless mach snaller amounts of vitamin D could be determined spectrophotometrically, the cost of the purified solvents alone would make this method prohibi- tive. The analysis of vitamin D milks for fat content shows that the varia- tion was from 3.21 to 4.69 g./100 ml. The average was about 3.60 g./100 ml. The extraction of fat by this method of solvents gave reliable re- sults for the duplicates checked quite well with each other. In the case of the‘unsaponifiable matter greater differences between duplicate determinations were observed. No satisfactory explanation is forthcoming for these results. There was little correlation between the fat content and the unsaponifiable matter. The cholesterol values ranged from 8.5 - 16 mg. per 100 m1. These agree very well with those found by Nataf, Mickelsen, and Keys (18) who reported values of 9 mg. to 17 mg. per 100 m1. Some correlation.was ob- served with the fat content of the milk as was reported by the above worker s . -16.. .o ‘. 1. 2. 3. 4. SUMT-KARY The unsaponifiable matter of fortified milk was prepared by extract- ing the milk fat with ethyl alcohol, ethyl ether, and Skelly solve B, followed by hydrolyzing with alcoholic potash, and final extraction with ethyl ether. The line tests performed with rats fed equivalent amounts of vitamin D in the form of fortified milk, extracted fat, and unsaponifiable matter indicated complete recovery of vitamin D. The unsaponifiable matter from a three-quart sample of vitamin D was also prepared. The spectrographic examination made after the attemp- ted removal of interfering substances by chromtographic absorption did not show an absorption maximun at 265 m. The failure was prob- ably due to excessive amounts of interfering substances introduced by the use of large volumes of solvents and the large quantity of sample. Eighteen samples of vitamin D milk from different dairies in various cities of Michigan were analysed for the fat content, the unsaponifi- able nmtter, and cholesterol. Results obtained were from 3.21 g. to 4.70 g./100 ml. for the fat content, 40.4 mg. to 75.0 mg./100 ml. for the unsaponifiable matter, and 8.50 mg. to 16.0 mg./100 ml. for cholester 01 . '17- y e e e O y o . ‘ . . I . p . . C .. .Y , u . s . n . . . U . l I II I O s s o ( v r . l ‘ I c _ . I o s I t . .s st. ' 0 v e \ A I . e _ a V . . e. r . . 1 s a s . . o p I . . \ . , 7 ‘ . . , I - . O . . . . I‘ n 1 . . \ n . o c . . . . l I O O' 1. 2. 3. 4. 5. 6. 7. B. 9. 10. 11. 12. 13. BIBLIOGRAPHY Supplee, G. C., Arisbacher, 8., Bender, R. 0., and Flanigan, E. B., The Influence of Milk Constituents on the Effectiveness of Vitamin J. Am. Med. Assoc. 106, 1664 (1936) Bethke, R. 11., Kraus, W. 3., Record, P. R., and Wilder, 0. H. 35., The Comparative Anti-rachitic BfficiencLof Vitamin D in Irradiated Milk, Metabolized (yeasflmk, and Cod Liver Oil. J. Nutrition 11, 21'(1936y‘ Milas, N. A., Heggie, R., and Raynolds, J. A., A Spectroscopic Method for the Quantitative Estimation of Vitamin D. Ind. . Chem. Anal. E. IS, 227 (1941) Nield, C. B., Russell, W. 0., and Zimmerli, A., The Spectrophoto- metric Determination of Vitamins Di and 133. J. Biol. Chem. 136, 73 (1940) Zimmerli, A., Nield,..C. R., and Russell, W. 0., A Mo_d__itfied Antimony Trichloride Reagent for the Determination of CertaTn sterols and VI ml]. g and Q1. Je BfOle Cfieme 148, 245 (1943)— Ewing, D. T., Kingsley, G. V., Brown, R. A., and Emett, A. D.. Physical-Chemical Method for the Determination of Vitamin D in Fish H791. 0m. Inde E3. Chane 5.11an He Is. 361 (1913; Ewing, De To. Pm11, Me Je. Brawn, Re As. and Barnett, As De MC,- mining Vitamin D2 by Two Physical Chemical Methods. Anal. Chem. 20, 3I77(194ey DeWitt, J. B., and Sullivan, M. K., S ectrophotometric Procedure for Quantitative Estinmtion of Vitamins . Ind. Eng. Chem. Anal. Ed. 18, UV (1946) Sobel, A. B., Mayer, A. M., and Kramer, 8.. New Colorimetric Re- action of Vitamins D1 and D1 and Their Provitamins, Ind. 7mg. Chem. Anal. EE;‘I7,*160 (1945) " Villar Palasi, V., A New Color Reaction for the D Vitamins, Nature, 160, 99 (1947) - Ken, 8. K., and Booth, R. G., The Vitamin D Activity of Butter, Bio. Chem. J. 28, 111 (1934) Ron, 8. K., and Booth, R. 0., Vitamin D Activity of Butter, 1. A Chemical Factor of Autumn and WTnter BuRer from Irradfafid Ergos- terol and the Vitamin D of Cod Liver 011, 810‘. Chen. J. 27, 1302 (1933) 14. 15. 16. 17. 18. 19. 20. 21. Bechtel, H. E., and Hoppert, C. A., A Study of TheSeasonal Variation of Vitamin D In Normal Cow's Milk, J. Nutrition 11, 337"(I§§€) Fritz, J. C., Halpin, J. L., Hooper, J. R., and Kramke, E. R., Oxidative Destructive of Vitamdn_D, Ind. Eng. Chem. 34, 979, Weatherby, L. ‘Vitamins C, D, and E, in Avocados, California Avocado.Assoc. Y QFBOOk 100 (1930) Lassen, 8., Bacon, R., and Sutherland, J., Vitamin D and Pro- vitamin D Content of the.Avocado, Food Research 10, I (1945) Nataf, B., Mickelsen, 0., and Keys, A., The Cholesterol Content of Cow's Milk, J. Nutrition 36, 495 (1948) Brockmann, R., Z. Physiol. Chem., 241, 104 (1936) Brockmann, H., and Busse, A., Z. Physiol. Chem., 256, 252 (1938) Brockmann, R., and Chen, Y., Z. Physiol. Chem., 241, 129 (1936) -19.. A 1 ‘ I Ira. . 4.... r V .1 WI'EL...IJe.b‘.M§A. ....Jd-Ir.‘ylltl.un. .AejfnLar . ...,O.¢..r CHEMISTRY LIBRARY .T637.1 216940 [0532 Chiang.