A ^JANTITATIVE SOTUr OF THE HEAT LABILE SULFIDES OF MILK A quantitative STu EY of h e a t l a b i l e s u l f i d e s o f m i l k by ROBERT C. TOWHLEY 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 DOCTOR OF PHILOSOPHY Department of Dairy Husbandry 1942 ProQ uest Num ber: 10008443 All rights reserved INFO RM ATIO N TO ALL USERS The quality o f this reproduction is dependent upon the quality of the copy subm itted. In the unlikely event that the author did not send a com plete m anuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest. ProQ uest 10008443 Published by ProQuest LLC (2016). C opyright o f the Dissertation is held by the Author. All rights reserved. This w ork is protected against unauthorized copying under Title 17, United States Code M icroform Edition © ProQ uest LLC. ProQ uest LLC. 789 East Eisenhow er Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 ACKN OWLE DGMENT S The writer wishes to express his sincere appreciation to Doctor Earl Weaver, Head of the Dairy Department, for making this study possible, and to Doctor I. A. Gould for his valued guidance in conducting this study and in the preparation of this manuscript. jL ij O 9 TABLE CF CONTENTS page 1 INTROrtfCTIQN.............................................. 2 HISTORICAL........................ Heat Formation of Hydrogen Sulfide andSulfhydryl Compounds.... 2 Early Observations...................................... 2 Recent Observations....................... 3 Observed Sources of Hydrogen Sulfides and Sulfhydryl Compounds and Factors Influencing Their Liberation by Heat.............. 3 Hydrogen-ion Concnetration............... 3 Fat Percentage......................................... 4 Proteins Associated with Fat Globules................... 5 Casein..................... 6 Whey Pro teins. ........................... 7 Influence of Copper and Iron............................ 9 Nitroprusside Test...................................... 9 Persistency of Sulfhydryl Compounds....................... 10 Miscellaneous Factors..................................... 10 Practical Applications of the Formation of Sulfur Compounds by Heat Treatment............................ Cooked Flavor............................................. 11 Activated Flavor.......................................... 12 Oxidized Flavor Prevent ion...................... Sulfur Distribution of Milk ...................... 15 ....................................... 15 Protein Sulfur............................................ 15 Total Sulfur Non-Protein Sulfur 18 - b - page Additional Heat Changes in Which Milk Proteins are Involved*.... Discoloration of Milk........................... 20 Heat Coagulation of Soluble Proteins............... 23 Influence of Sugars Upon the Protein Coagulation... 24 Influence of Formaldehyde Upon Protein Coagulation. 20 .... 25 Influence of Metals on the Heat Coagulation of Proteins.... 26 Influence of Alcohol on the Heat Coagulation of Proteins... 28 SCOPE OP INVESTIGATION.................................... 30 EXFERIMENTAL PROCEDURE............................ ............. 31 General Methods............... 31 qualitative Methods ............ 32 quantitative Methods. .................... 32 RESULTS...... 40 General Origin and quantitative Studies of Heat Labile Sulfides 40 Critical Temperature Studies............................. 40 Relationship of Heat Labile Sulfur to Total Sulfur and Total Nitrogen........... 45 Washing of Cream......................... 46 Separation Temperature................................... 47 Homogen izat ion 48 ........................................ Breed and Period of Lactation....... 49 Persistency of Volatile Sulfur Liberation................ 51 Influence of Variable Heat Treatment Upon the Labile Compounds.. 53 Function of Time and Temperature......................... 53 Influence of Previous Heat Treatment..................... 55 Prolonged and High Temperature Treatment................. 56 - c page Influence of Added Compounds Upon the Heat Labile Sulfur Compounds 62 Sugars............. 62 Alcohol................. 64 Salt.................................................... 66 Qystine and Cysteine................. 67 Reducing Substances................................ 69 Hydrogen Peroxide.................... 71 pH and Iron Salts....................................... 73 Additional Studies on Hie Influence of Metals............. 76 Miscellaneous Factors.................................... 81 DISCUSSION.............................................. 82 SUMMARY AND CONCLUSIONS........................... 88 LITERATURE CITED............................................... 91 INTRODUCTION The recent trends toward the use of higher temperatures in the processing of dairy products has caused considerable attention to be directed toward the volatile sulfides and sulfhydryl group appearing at such temperatures. These compounds have been shown to be especially significant in heat treated products since they appear rather abruptly at approximately the same temperature found to produce a cooked flavor, prevent an oxidized flavor, destroy certain enzymes and cause a redis­ tribution of nitrogen. In addition, these Sulfide and sulfhydryl com­ pounds are apparently associated with the purple ring formation in evaporated, and perhaps has some significant role in the browning of milk subjected to sterilization temperatures. The source of the volatile sulfides evolved from milk during heating has been shown by quantitative tests to be the serum proteins and fat globule membrane. However, the conclusions drawn in this regard are not entirely in harmony and a number of the factors influencing the heat liberation of sulfides have not been fully investigated. In addi­ tion, the amount of sulfides contributed by each of these constituents and the total amount of sulfides evolved by heat treated milk has not been measured quantitatively. This study was conducted to determine quantitatively the volatile sulfides of milk and the influence of certain factors on these sulfides. Also, efforts were made to ascertain the relationship between the heat liberated sulfides and the total sulfur content of milk. HISTORICAL Heat Formation of Hydrogen Sulfide and Sulfhydryl Compounds Barly observationsS %drogen sulfide was early identified by Schreiner (136)* Neimann (ill), hubner (l3l), Oppenheimer (113), Rettger (124), and Utz (l5l), as being among the volatile constituents liberated when milk is subected to high temperatures. Similar observa­ tions have been made more recently by Konig and Schreiber (88). In general, these experiments were concerned with the development of specific tests to differentiate between heated and unheated milk or to demonstrate general decomposition of milk by high teaperature treatment. Lead acetate paper or cotton soaked in lead acetate solution were used by the majority of the early workers. Oppenheimer ^113) in 1901, used lead acetate paper for sulfide detection and obtained from 3 to 6 mg. of sulfur by boiling 300 cc. of milk for 30 minutes. He de­ tected no sulfur compounds when the milk was held 30 minutes at 75° C. In a similar manner Utz (151) observed a small amount of hydrogen sulfide liberation from milk heated to near boiling, and points out that another investigator (Schreiner) observed hydrogen sulfide liberation from milk at tenperatures between 81 and 82° C. Kettger (124), by suspending cotton soaked in lead acetate solution above heating milk, was unable to detect hydrogen sulfide liberation until the temperature exceeded 85° G. This worker observed that pasteurized milk and condensed milk liberated only small amounts of hydrogen sulfide, and that upon prolonged heating of milk a point was reached at which hydrogen sulfide was no longer evolved. In 1927, Konig and Schreiber (88) boiled milk 4 hours in an atmosphere of nitrogen and collected the volatile sulfur in a train con- •• 3 •• taining a silver solution. By this procedure they obtained 3*8 mg* of sulfhr per liter of milk. Recent observations: A more recent study on the problem of sulfide liberation in milk is presented by Gould and Sommer (45). These workers used basic lead acetate paper and rath®* large volumes of milk to show that hydrogen sulfide liberation is a function of time as well as tenperatur e. Sulfides were detected when milk was heated from 76 to 78° C. momentarily, from 74 to 76° C. for 3 minutes, or from 70 to 72° C. for 30 minutes. Positive nitroprusside tests were obtained on milk heated to similar temperatures by Josephson and Doan (75) and Gould (48), indicating the formation of sulfhydryl groups. Plake, Jackson and Weckel (36) conducted limited quantitative studies on the heat liberation of volatile sulfur. Ry vacuum distillation at 77 to 79 C. they were able to recover 0.0036 mg. of sulfur from 350 cc. of milk. In other quantitative studies, Dieraair, Strohecker and Keller (27) added 6 cc. of 1 per cent sodium hydroxide to 10 cc. of milk and dis­ tilled over the volatile sulfur. cc. samples: They report the following values for 100 raw milk, 0.13 to 0.15 mg.; pasteurized milk, 0.11 to 0.14 n^. ; and cream, 0.07 to 0.11 mg. Observed Sources of Hydrogen Sulfide and Sulfhydryl Compounds and Factors Influencing Their Libera" tion by Heat Hydrogen-ion- Concentration: Hydrogen-ion concentration is generally known to influence the stability of proteins and observations on milk have shown this factor to influence hydrogen sulfide liberation by heat. Osborne (114), Zahnd and Clarke (166), Blumenthal and Clarke (15), - 4 Abel and Gelling (l)* Brand and Sandberg (18), Jones and Gersdorff (74), and Vickery and White (156) have shown cystine to be unstable to alkali. On the other hand, methionine has been r sported to be rather stable to alkali by Mueller (109), Painter and Pranke (118), and Masters (9v). Shinohara and Kilpatrick (140) found that casein would liberate hydrogen sulfide after heating at 80° C. for 6 hours in an acid medium. Bettger (124) concluded from his early observations that the liberation of hydrogen sulfide from milk was enhanced by a slight alka­ linity and retarded by slight acidity. According to Jackson, Howat and Hoar (69) sodium bicarbonate, disodium phosphate, or sodium citrate added to canned cream at the rate of 5 grams per gallon enhanced sulfide libera­ tion during sterilization, and decreased the temperature and period of heat exposure necessary to give a positive nitroprusside test. The work of Gould and Sommer (45) indicates that the pH of the sample may influence the lability of sulfur in milk. They observed that milk at a pH of 7.6 formed a cooked flavor at a lower temperature than normal milk, whereas a pH of 5.8 to 6.0 retarded cooked flavor formation. In other work Gould (48) states that whey has a normal critical temperature even when heated at pH 4.6. Pat percentage! One may logically conclude that increases in fat percentage of a sample of milk would decrease the amount of available sulfur since there would be a corresponding decrease in the protein content. In quantitative studies, Bettger (124) and Diemair, Strohecker and Keller (27) report that skimmilk liberates more hydrogen sulfide and cream less hydrogen sulfide than normal milk. In contrast, Gould and Sommer (45) and Gould (48) demonstrate that increases in butter fat decrease the temperature - 5 at which hydrogen sulfide appears. Skimmilk required a somewhat higher temperature than milk before evolving sulfides, and cream, buttermilk whey liberated sulfides at a lower temperature than milk. The following critical temperatures for a 3 minute heating period were reported by these workers* 35 per cent cream, 66 to 68° C.; 30 per cent cream, 70 to 72° 0.; whole milk, 74 to 76° C.; and skimmilk, 76 to 78° C. Joseph- son and Doan (75) also report that cream has a lower critical temperature than milk. Proteins associated with fat globules! The lower critical temperature for sulfide liberation from cream observed by Gould and Sommer (45) and Josephson and Doan (75) indicates that the material contributing greatly to the heat labile sulfides of milk products is associated with the fat globules. Gould and Sommer (45) report critical temperatures for 3 minute holding periods of 35 per cent cream and the buttermilk from this cream as 66 to 68° U., respectively, approximately 10° C. lower than that of normal milk, furthermore, these workers showed that normal 20 per cent cream had a critical temperature of 72° C. for a 3 minute holding period, whereas 20 per cent cream prepared from skimmilk and butter oil had a critical temperature of 79° C., a critical temperature identical to that of skimmilk. In addition, Gould (48) showed that the sulfide contributing material of cream is difficult to remove by washing. Three-times washed cream had a critical temperature practically identical to normal cream. In contrast, Josephson and Doan (75) were able to remove the sulfhydryl producing materials by washing cream several times and believe that the absorbed layer around the fat globules is a relatively unimportant source of heat-producing sulfhydryl groups. However, th^r s tate that this evidence is not entirely convincing since cream has a lower critical temperature than milk, and has a more sulfide odor and flavor than milk or skimmilk. Casein; Reasonably convincing evidence has been presented to show that casein as such plays no significant part in labile sulfur formation in heated milk. However, observations on the volatile sulfur of casein have been reported by Osborne (114), Zahnd and Clarke (166), and Bluraenthol and Clarke (16). These workers heated casein with basic lead acetate and report sulfur values ranging from 0.07 to 0.10 per cent. Rettger (124) suspended casein in a phosphate buffer (the mixture was still acid to litmus), and upon boiling 100 cc. of a 3 per cent suspension for 40 minutes obtained enough hydrogen sulfide to color lead acetate paper. Ansbacher, Flanigan and Supplee (4) eluted casein with dilute sodium chloride and obtained a protein fraction that appeared to display sulfur migration when exposed to ultra-violet light, heat or agitation. Weckel and Jackson (160) produced an activated flavor in casein by irradiation, and Flake, Jackson, Weckel (36) isolated the activated flavor material from irradiated locust-bean-precipitated casein by steam distilla­ tion and adsorption on activated charcoal. Early studies by Wright (163) and Utz (l5l) have demonstrated the heat stability of casein toward sulfideliberation. Utz (l5l) was un­ able to detect hydrogen sulfide liberation from casein heated to 100° C. in an amphoteric phosphate buffer. Wrigit (163) used racemisation curves to demonstrate that heat caused no alteration in the constitution of the casein molecule after heating at 60° C. for 1 hour, boiling under reflux for 30 minutes, or autoclaving at 120° C. for 30 minutes. Nickels, Bailey, Holm, Greenbank and Deysher (110) found that the particle size distribu­ tion of calcium caseinate in skimmilk was not affected by momentarily heating at 95° C. Studies by Gould and Sommer (45) and Gould (48) have shown that skimmilk and skimmilk whey have identical critical tempera­ tures, and that buttermilk whey possesses the same critical temperature as buttermilk. Therefore, they concluded that casein of milk contributes few if any of the sulfides liberated at the critical temperature. Similar conclusions were drawn by Josephson and Doan (75). Whey proteins: Lactalbumin, the principle whey protein has been shown to contain considerable quantities of sulfur, of which approx­ imately 60 per cent is due to cystine (Schmidt (134), Baernsteine (9) and Kassel and Brand (77) )• Blumethal and Clark (16) and Zahnd and Clarke (167) have recently reported that allof the sulfur of cystine is heat labile. Mlrsky and .Anson (106, 107, 108) and Hopkins (64) have demon­ strated that -SK and S-S groups appear upon denaturation and coagulation of albumin* Kassel and Branc (77) found that 5 per cent of the sulfur of lactalbumin to be due to cysteine. Globulin, the other major protein of whey, has been reported by Osborne and Wakeman (116) to contain 0.86 per cent sulfur. Palmer (119) reports a crystalline globulin, which he obtained from albumin, and states that this material appears to contain approximately 4 per cent cystine. Other sulfur containing materials have been reported as being present in whey and perhaps they too contain heat labile sulfur. Ans­ bacher, Flanigan, and Supplee (4) isolated the foam producing material from whey and found it to display sulfhr migration when subjected to ultra­ violet light, heat, or agitation. This material con tained no cystine. - 8 Observations on milk have demonstrated rather conclusively that whey proteins are responsible for a great portion of the heat labile sulfur obtained from milk. Gould and Sommer (45) and Gould (48) removed the casein from skimmilk and found that the serum thus obtained liberated hydrogen sulfide when heated to 76 to 78° C. for 3 minutes. The latter also found that the temperature at which sulfhydryls are formed is un­ changed when the pH is decreased to 4.6. He states that Halbumin may be expected to contribute sulfdr compounds on being subjected to heat since it undergoes denaturation and coagulation at the temperatures which pro­ duce the -SH groups and hydrogen sulfide, since it has a relatively high sulfur content, and, thirdly since this sulfur is largely supplied by the heat labile amino acid cystine*1. Josephson and Doan (75) attempted to determine the source of the sulfhydryl compounds formed in heat treated milk and concluded that lactalbumin is their principle origin. Hettger (125) early obtained hydrogen sulfide from lactalbumin even when it was boiled in an acid medium. Weckel and Jackson (160) were able to obtain an activated flavor in casein-albumin-globulin-free wheyseparated from irradiated milk. They believe this to be due to absorption. However, Doan and Meyers (28) believe this whey portion of milk to be the source of the “burnt” flavor produced Since vitamin in milk esposed to sunlight. contains the thiazole ring as a part of the molecule, there is a possibility that some of the heat labile sulfur of milk is derived from this source. Gortner (44) states that strong alkali and heat causes theliberation of hydrogen sulfide from the thiazole half of thiamin hydrochloride. When milk is subjected to temperatures ranging 9 from pasteurization to sterilization there may be as much as 10 to 50 per cent of this vitamin destroyed (Daniels, Gibt>ings and Jorden (24), Holli­ day (6), Butcher, Guerrant and McKelvey (29), Krauss, Erb and Washburn (89), and Houston and Thompson (65), Influence of copper and iron: Small amounts of copper and iron have been shown to retard and in some cases entirely prevent the lability of sulfur. Gould and Sommer (45) found that in normal milk, 76 to 78° C. were sufficient to liberate sulfides, but in the presence of 1.4 ppm. of added copper temperatures of 94° C. or higher are required. Similar results are shown by Gould (48), who found further that the addi­ tion of ferrous iron had only a slight retarding influence on sulfide liberation. He also states that copper added to milk after heating is more effective for inhibiting sulfide liberation than when added previous to heating. Nitroprusside test? The nitroprusside test was employed by Josephson and Doan (75) who found that milk containing 0.5 ppm. of added copper did not give positive tests until heated to 82° C., and that this milk remained positive to nitroprusside for only 4 hours, whereas un­ treated milk gave positive tests at 76 to 78° C.and remained positive for 48 hours. for as The test remained positive longas 144 hours in milk momentarily heated to temperatures above 80° C. The nitroprusside test was also used by Gould (48) who re­ ports a positive nitroprusside test on milk containing 1 ppm. of copper when heated to 84° C.; this milk, however, did not evolve detectable hydrogen sulfide until heated at least to 90° C. - 10 Persistency of sulfhydryl compounds? Gould (48) demonstrated that a positive nitroprusside test persists and hydrogen sulfide is evolved throughout a 6-hour aspiration period from milk previously heated momentarily at 90° C.» whereas the addition of 1 ppm. of copper immedia­ tely after heating and cooling prevented the liberation of hydrogen sul/ fide and decreased the intensity of the nitroprusside test. Dahle, Lawhorn, and Barnhart (22) report that the addition of 1 ppm. of copper to cream heated to 88° C. momentarily or to 76° C. for 10 minutes resulted in a negative nitroprusside test. Miscellaneous factors? Additional factors which may influence liberation of sulfides from milk products have been considered. Flake, Jackson and Weckel (36) found that prolonged irradiation of milk caused labile sulfhr formation at a temperature approximately 10° C, below normal. They recovered 0.0131 mg. of sulfur from 350 cc, of milk exposed to ultra violet rays for 90 minutes, whereas only 0.0036 mg. were recovered from normal milk. Jackson, Howat and Hoar (68) demonstrated that homogenization had no influence on sulfide liberation in sterilized canned cream. Certain reagents have been shown to bring about sulfhydryl formation in milk without the application of heat. Jackson (?0) and Gould and Sommer (45) obtained positive nitroprusside tests on milk by adding a small amount of sodium cyanide, and Greenstein (54) detected sulfhydryl formation in milk subjected to 8 to 16 molar guanidine hydrochloride. Flake, Jackson and Weckel (36) correlated sulfhydryl compounds with an activated flavor in irradiated milk. In a previous paper (39) they had demonstrated that the flavor could be removed from milk by adding a small amount of hydrogen or calcium peroxide. - 11 Practical Applications of the Formation of Sulfur Compounds by Heat Treatment Cooked flavori Considerable attention is now being given to the problem of cooked flavor in milk and milk products, and recent studies have indicated a relationship between sulfur compounds formed during heating and processing and the production of this flavor (Gould and Sommer (45), Josephson and Doan (75), Gould (48), and Ifehle, Lawhorn and Barnhart (22) ). Gould and Sommer (45) present the first positive evidence indicating that sulfur compounds are involved in the formation of a cooked flavor. When milk was subjected to high ternperatures the cooked flavor was found to appear rather abruptly with the liberation of hydrogen sul­ fide at an unusually constant temperature range of from 76 to 78° C. The critical temperature was largely unaffected by such factors as aeration during heating, dilution with as much as 20 per cent of distilled water, separation and remixing of the cream and milk, mastitis infection, seasonal variation, or breed differences. The development of a cooked flavor was enhanced by heating at a pH of 7.6 and retarded by heating at a pH of 5.8 to 6.0. The critical temperature was decreased with a longer period of exposure to heat or by marked increases in fat content. The addition of 1.4 ppm. of copper prior to hea.ting raised the temperature necessary to produce a cooked flavor to approximately 84° C. When the copper was added following heating, the cooked flavor rapidly disappeared. In other work by Gould (48) he showed that ferrous iron added to milk at the rate of 2*8 ppm. raised the only slightly. critical temperature - 12 The findings of Gould and Sommer (45) were largely sub­ stantiated in the studies of Josephson and Doan (75). In addition, the latter workers were able to demonstrate that the appearance of the cooked flavor was simultaneous with the occurrence of sulfhydryl groups as detected by a modified nitroprusside test. This was also shown by Gould (48), who states that this test is more closely correlated with the cooked flavor than the appearance of hydrogen sulfide especially in the presence of added copper. Several workers have reported on theuse of electrical methods of pasteurizing milk as a possible means of using slightly high­ er tenperature than normally employed when other methods of heating are utilized (Prescott (122), Supplee and Jensen (145), and Millenky and Brueckner (103) ). In general, it may be concluded from their reports that this method of subjecting milk t o heat has no marked influence on the temperature at which cooked flavors and heat-labile sulfur compounds appear. Activated flavor? Radiation of milk for prolonged periods may cause milk to acquire a flavor defect identified as activated, sun­ shine or burnt. This flavor is now usually differentiated from an oxi­ dized flavor, but at first was thought to be closely allied. The work of Flake, Jackson and Weckel (36) has shown that ultra-violet rays will liberate sulfhydryl compounds from milk, and have associated these com­ pounds with the activated flavor. Weckel and Jackson (160) produced an activated flavor in casein and albumin by irradiation. Mancovitz (93) studied ultra-violet rays as a means of pasteurizing milk and states that an exposure of 45 seconds produced - 13 disagreeable sulfur flavors and odors. In addition, Weckel and Jackson (160) , Flake, Weckel and Jackson (37), and Flake, Jackson and Weckel (35, 38) demonstrated that a distinguishing characteristic of the activated flavor is that it becomes more noticeable when irradiated milk is momen­ tarily exposed to temperatures above approximately 70° C. These workers found that homogenization rendered milk more susceptible to activated flavor development, and Weckel and Jackson (160) report that milk and buttermilk are more susceptible than cream or skimmilk. The addition of 2 to 3 ppm. of copper causes the activated flavor to disappear. Doan and Meyers (28) present a rather thorough study showing that sunlight produced the activated flavor in milk products, but made no attempt to correlate their findiigs with the formation of sulfhydryl com­ pounds. However, Young (l64) has demonstrated that sunlight has a de­ structive effect on albumin, and Ansbacher, Flanigan and Supplee (4) con­ cluded that ultra-violet rays alter the sulfur compounds of the foam pro­ ducing material of milk. Oxidized flavor prevention; Oxidized flavor often occurs in processed dairy pro diets especially in thepresence of certain heavy metals. The beneficial effect of high temperature pasteurization has been recogniz­ ed as a preventive measure. A temperature of 85° C# for 5 minutes was employed by Kende (79) to retard and in some cases prevent oxidized flavor development, even in the presence of as much as 2 mg. of CuSO^ per liter of milk. Kende was of the opinion that an enzyme which he called '’oleinase” was catalyzed by metals to develop the oxidized flavor, and that high temperatures inactivated this enzyme. Normal pasteurization temperatures were reported by Dahle and Palmer (23) to enhance the degree of oxidized flavor which may develop. These workers favor the enzyme theory of oxida- 14 tion since temperatures slightly above pasteurization temperature did not prevent oxidized flavor development, Pereas, temperatures above 73.° C. for 30 minutes were effective. Ritter (127) emphasizes the beneficial effect of high temperature pasteurization in preventing copper-induced oxidized flavor. Work dealing with sulfhydryl formation and the oxidized flavor has largely refuted the enzyme theory. Gould and Sommer (45), Josephson and Doan (75), Gould (48), and Swanson and Sommer (l4?), have shown that high temperature treatment of milk lowers the oxidation-reduction potential, and prevents copper induced oxidized flavor development. This lowering of Eh was correlated with reduced sulfur compounds formed in the milk upon heating. Copper added to milk after heat treatment tends to cause a more frequent and more intense oxidized flavor than when copper contamination occurs before heat treatment, and an oxidized flavor does not develop in milk even in the presence of small amounts of copper until after the sulfhydryl groups have disappeared (48). Gould and Sommer (4?) have also shown that milk containing 1.4 ppm. of copper had a higher critical temperature for sulfide liberation than for cooked flavor; that hydrogen sulfide may combine with the added copper. this indicates The importance of sulfhydryl compounds in preventing oxidized flavors in the storage of sweet cream has been demonstrated by Dahle, Lawhorn and Barnhart (22). The failure of milk to dissolve copper at high temperatures is also involved in the heat prevention of the oxidized flavor. Milk has been shown to practically lose the ability to dissolve copper when heated to high temperatures. (Rice and Miscall (126), Miscall, Cavanaugh and Carodemos (108), and Gebhardt and Sommer (43) )♦ The latter investigators point out that the addition of copper to milk increases the oxidation- - 15 reduction potential but that there is a decided drop in the potential of milk heated above 70° C. * temperatures at which the solubility of copper is greatly decreased. Sulfur Distribution of Milk Total sulfur; Although different methods have been em­ ployed to estimate the sulfur content of milk, the values reported by various investigators agree closely. several workers are as follows: Sutthoff (149), 0.037; The percentages reported by Sherman (l39)t 0.034; Tillsman and Revol and Paccard (125), 0.027 to 0.044; and Sullraann (145), 0.030; Kemmerer and Boutwell (78), 0.028; Steffen Masters and McOance (96), 0.029; Beach, Bernstein and Hoffman (ll), 0.033; and Hutchinson (66), 0.031. Masters (97) has presented an excellent review of the limitations of the methods most applicable to total sulfur deter­ minations of milk products. The sulfur distribution in milk has been studied by Tillsman and Sutthoff (l49) and Beach, Bernstein and Hoffman (lO). Tillsman and Sutthoff present the following percentages showing the sulfur distribution of milk containing 92.1 mg. of S0g per liter; organic sulfur 4.9; protein sulfur 84.9; other sulfuric acid already formed, 10.4; and ash 1.23* Beach, Bernstein and Hoffman obtained 7.3 mg* of cystine and 86.6 mg. of methionine from the casein present in 100 cc. of milk and obtained 15.5 mg. of cystine and 17.1 mg. of methionine from the whey proteins in 100 cc. of milk. The data of the latter workers show that approximately 80 per cent of the protein sulfur of milk is due tomethionine. Protein sulfur? The three recognized proteins of milk; i.e., <- 16 casein, albumin and globulin, are considered to be good sources of sulfur and have been studied considerably in this regard. A summary of the total sulfur and the portioh of the total sulfur contained in the amino acids of the principle milk proteins is reported in Table 1. The total sulfur values reported for casein are consistently near 0.80 per cent. However, the total sulfur values for lactalbumin vary from 1.22 per cent to 1.92 per cent. The cystine and methionine values also show considerable variation. Table 1. Total sulfur and amino acid sulfur of milk proteins. Portion of Milk Protein To tal Sulfur Cystine Per Cent Per cent 0.80-0.90 Casein ii 9.8 0.80 « 0.63 ii 0.71-0.78 n 0.70-0.80 12.0 ii 8.7 0.796 Albumin H II 1.45 1.22 1.92 1.42 Globulin 0.86 ii Total Sulfur Authority Methionine Per Cent 84.2-90.2 Baernstein (8) 83.4 Baernstein (10) Blumenthal and Clarke (16) Zahn and Clarke (166) 88.0 Kassell and Brand (78) 83.7 Beach,Bernstein and Hoffman (ll) 62.1 34.4 53.0 42.0 Baernstein (lO) Blumenthal and Clarke (16) Osborne and Wakeman (117) Kassell and Brand* (78) Osborne and Wakeman (l!7) * These authors report that 5 per cent of lactalbumin sulfur is due to cystine. Although Kassell and Brand (77) have been able to account for all of the sulfur in casein and albumin as cystine, cysteine and methionine, other investigators have been unable to find this to be true, especially in the case of casein. There are relatively few sulfur-containing amino acids, although the list is steadily being increased. Schmidt (134) states that cystine, - 17 cysteine, methionine, and djenkolic acid are the four sulfo-amino acids present in proteins even though indirect evidence points to the possibility of others. Van Veen and Ifyman (153) isolated djenkolic acid from the djenkol nut and Horn, Jones and Bingel (64) found a new sulfUr-containing amino acid (lanthionine) in sodium carbonate treated wool. Blumenthal and Clarke (l5) recently reported that proteins contain at least two sulfur bearing constituents besides cystine and methionine. According to Gulland and Morris (54) acid hydrolyzed casein yields a fraction from which sulfur in elemental form is readily split off. A summary of the cystine and methionine values of lactalbumin and casein is presented in Table 2. Table 2. Amino acid distribution in milk proteins. • « Milk Protein Casein 11 ii ii n ii Qystine : Methionine Per Cent : Per Cent : 3.31-3.10 0.30 : 3.4 0.21-0.47: 0.34 : 0.20-0. 301 0.28 : 3.10 Authority Baernstein (9) Schmidt (135) Vickery and White (l5v) Jones and Gersdorff (75) Polin and Marenzi (40) Beach,Bernstein and Hoffman (ll) * • • • Lactalbumin H II 4.30 2.65 : 2.60 s : 2.45-2. 32 Schmidt (135) Vickery and White (l5?) Baernstein (9) • • Considerable variation is evident in the case of the values reported for cystine. Vickery and White (156) believe the variation in cystine values reported for casein is due to methods of purification, and showed that the cystine content of casein decreased each time it was washed with dilute alkali. - 18 Early experiments by Osborne and Wakeman (116) on milk proteins have shown the presence of an alcohol-soluble protein in acidprecipitated casein. In another study (117) these authors found that this alcohol-soluble protein contained 0.95 per cent sulfur, distinctly more than casein. Several studies have been conducted dealing with the protein compounds in whey freed of casein, albumin and globulin. Osborne and Wakeman (l!7) indicated that a protein fraction remained after removal of casein, albumin and globulin from milk. Sure and 0*Kelly (146) re­ moved these three proteins from milk and report that 66 to 76 per cent of the sulfur remaining was of an organic source. In a recent study by Ansbacher, Flanigan and Supplee (4), a foam producing fraction was isolated from casein-albumin-globulin-free milk. This protein fraction contained 0*60 per cent sulfur, but no cystine. However, Josephson and Doan (75) were unable to obtain a positive nitroprusside test on such whey after heating at 180° C. for 10 minutes. Several workers have isolated the adsorbed material surrounding the fat globule and have indicated that this material is not identical with other milk proteins. The following percentages of sulfur have been re­ ported for the adsorbed materiall Hattori (56), 2.58; Hart (150), 0.74; 0.96; Palmer and Wiese (122), 0,96; and Palmer and Samuelson (120), 0.73. Titus, Sommer and Wiese and palmer (162), Davies (26) calls attention to a mucoprotein, which was isolated from butter. He states that the adsorbed protein on fat globules has been found, to be high in arginine and low in histidine and cystine. Hon-protein sulfur* Surprisingly little convincii^ information is available regarding the sulfur content of the non-protein fraction of - 19 milk* Early work on this problem by Tillsman and Sutthoff (149) has shown that cows milk contains 15.1 per cent of its sulfur in the form of non—protein organic sulfur. Sure and 0*Kelly (146) studied the sul­ fur distribution of the filtrate obtained from milk after precipitating the casein with acid and precipitating albumin and globulin by heating the acid whey. ©ley report that the dry residue of protein-free milk contains 0.11 to 0.14 per cent sulfur, and that 65.9 to 76.4 per cent of this sulfur is in the organic form. Viale (154) found that fresh milk reduced methylene blue and gave positive nitroprusside test, concluding from this test that cysteine was present. Martini (94) claimed to have isolated 8 mg. per cent of glutathione from a liter of milk. be a source of labile sulfur. If present, glutathione may Kinsey (8l) has demonstrated the destruc­ tive effect of x-rays on aqueous solutions of glutathione, and Hopkins (62) removed sulfur from glutathione solutions by gentle aeration of pH 716 with a small amount of iron, or by boiling in water. H0wever, Jack­ son (70) , and Gould and Sommer (45) could not obtain positive nitroprusside tests on milk unless a small amount of sodium cyanide was added. Jackson (71, 72) concluded that no cysteine or glutathione are present in free solution in milk, basing this conclusion on the f act that the protein-free filtrate of milk failed to give a positive nitroprusside test when treated with sodium cyanide, whereas a postive test was obtained on the filtrates from milk to which 1 mg. of either reduced glutathione or cystine hydro­ chloride has been added before removal of themilk proteins. Studies con­ ducted by Gould (46, 47, 49, 50) also indicate that milk does not contain glutathione. - 20 Vitamin B^ has been shown by Daniel and Munsell (25) to be present in milk to the extent of 20 Sherman units or 30 micrograms per 100 grams of milk} this vitamin contains about 12 per cent sulfur according to Winters teiner, Williams and Ruehle (162). Additional Heat Changes in Which Milk proteins are Involved The temperature producing hydrogen sulfide, sulfhydryl groups, cooked flavor, and decreased oxidation-reduction potential appears to coincide in a general way with those temperatures causing an increased tendency toward browning, coagulation of the soluble proteins and redis­ tribution of nitrogen. Pis coloration of milk t When milk is subjected to sufficiently high temperatures for a sufficient period of time a distinct browning occurs. numerous studies have been conducted in order to determine the nature of this discoloration, but as yet this phenomenon has not been fully explained. Webb and Holm (158), Webb (15?) and Ramsey, Tracy and Ruehe (123) have pointed out that this discoloration is a function of time as well as tenperature. Milk forewarraed at a high temperature for a relatively short time shows greater discoloration than when heated longer at a lower temperature. The latter authors demonstrate that sweetened condensed milk which had been forewarmed in the presence of corn sugar at 88° C. for 15 minutes was only slightly tan in color, whereas the same milk developed a dark brown color when held at 120° C. for 5 minutes. A protein-sugar complex has been thought by many to be the product responsible for the browning of milk. In this connection, rather comprehensive studies of the reaction between amino acids and reducing sugars have been conducted by Ambler (3), von Euler and Josephson (31), and Ramsey, Tracey and Ruehe (l23). These authors have demonstrated that sugars possessing either aldehyde or ketone groups will condense with the amino group of various amino acids to produce a complex, hi^ily colored substance. The latter authors state that this reaction is more pronounced with lysine and glycine and that sugars having a free aldehyde group, i.e., dextrose, glucose, and levulose, apparently react with free amino groups of the casein. With casein, this reaction may take place at temperatures as low as 40° C., but is enhanced by higher temperatures with milk; the reaction is accompanied by a decrease in pH. this complex unless it becomes inverted. Sucrose was found not to form Kometiani (87) states that caramelization of the lactose is the cause for the browning of milk by heat, and that the reaction is dependent upon the character of the dissolved salts. However, Leeds (9) attributes the formation of brownish-yellow color occurring during thesterilization of milk only partially to the formation of decompo­ sition products of milk sugar since mixtures of casein and albumin removed from discolored milk had a strong yellow color. in the whey thus obtained. No off color was apparent Ramsey, Tracy and Ruehe (123) and Webb (157) state that the complex formed during sterilization is removed with the proteins, and that the color-complex is so firmly attached to the milk proteins that it is not washed away with water, dilute acid, alcohol or ether. Ramsey, Tracy and Ruehe (123) found that browning could be prevented with a small amount of formaldehyde, citing this as evidence refuting the caramelizatiom theory. They believe that the formaldehyde blocks the amino groups and amino acids of proteins, thus preventing them from condensing with the lactose. As further proof that the off­ color of sterilized milk is not due to caramelization of lactose, they ob­ - 22 - served that solutions of either lactose, casein or albumin failed to change color when held at 250° F. for 30 minutes. Webb (l57) used formaldehyde in his experiment and states that "the action of formalde­ hyde appears to be more or less independent of the amino acid content of the lactose solution11. He found that small amounts of formaldehyde markedly increased discoloration, whereas larger amounts inhibited color formation. However, these investigators (123, 157) agree that the browning of sterilized milk is dependent upon free amino groups contained or formed by protein decomposition. Kass and Palmer (76) completed a recent study on the browning of autoclaved milk and concluded that the development and properties of the brown substance in autoclaved milk are not in harmony with the assump­ tion that it is a case in-lactose condensate. They conclude further that "the exact parallelism between the reactions of caramel, prepared either through the agency of heat alone or heat in the presence of buffer, and those of the brown substance present in heated milk point to the cohclusion that the origin and behavior of this coloration may be satisfactorily accounted for on the basis of the caramelization of lactose by the casein and the adsorption of the lac to-caramel by the colloidal caseinate". They present data demonstrating the essential role of casein in the discolora­ tion reaction. Von Euler and Josephson (3L) demonstrated that amino acidlactose condensation is enhanced with an increasing alkalinity which favors the formation of a free aldehyde group in the sugar molecule. Ramsey, Tracy and Ruehe (l23) found that the addition of minute amounts of carbonates or phosphates enhanced color formation in sweetened condensed milk, due to the increase in alkalinity. Webb and Holm (l58) found that small amounts of 23 phosphates, citrates or lime-water failed to produce any change when added to evaporated milk previous to sterilization; bicarbonate caused a darkening of color. however, sodium Further studies by Webb (l57) emphasizes that the discoloration during sterilization is of a catalytic nature and sulfite. is enhanced by phosphates and metals, but prevented by sodium Kass and Palmer (76) believe that the dissolved salts of milk have only a negligable influence on heat discoloration. Heat coagulation of soluble proteins; The soluble protein fraction of milk is often designated as albumin even though it includes albumin and globulin. Rupp (132), Grimmer, Kurtenacker and Berg (53), Matsua (98), Howland (128, 129, 130), Kieferle and Gloetzl (80), and Irvine and Sproule (67) subjected milk to various temperatures and found that albumin and globulin precipitation does not occur abruptly at any given time and temperature, but begins to precipitate at temperatures o o above 60 C. and is not completely precipitated until held at 100 C. for 10 to 30 minutes. Rupp (132) reported that no albumin was coagulated in milk during a 30 minute holding period at 62.8° C. However, 30.73 per cent of the albumin present in milk was precipitated when held for 30 minutes at 71.1° C* Similar observations are reported by Rowland (l28) and Irvine and Sproule (67). In other studies (129), Rowland found that maximum precipitation of albumin and globulin occurred when milk was heated to 80° C.for 60 minutes; 90° C. for 30 minutes; 95° C. for 10 to 15 minutes and 100° C* for 5 to 10 minutes. He found that approximately 75.2 per cent of the total soluble nitrogen was albumin and globulin nitrogen. Studies on the influence of heat on the nitrogen distribution 24 — of milk are reported "by Grimmer, Kurtenacker and Berg (53) and Kieferle and Gloetzl (80). A portion of the results reported by the latter workers on the nitrogen distribution of milk heated to various tempera­ tures and held for 30 minutes is presented in Table 3. Table 3. Influence of heat upon nitrogen distribution in milk. Total N Casein Albumin Proteose Peptone Total Resi­ dual N Raw mg. ft 540.4 348.3 75.7 44.6 45.2 123.6 63° C. mg. ft 537.8 335.6 71.7 46.0 59.4 138.0 85° C. mg. io 537.3 347.8 53.3 45.7 60.0 148.6 100° c. mg. ft 540.5 383.0 13.9 42.4 66.0 115° C. mg. ft 537.2 390.6 8.0 36.4 68.4 152.0 160.0 These data indicate that a small portion of albumin is precipated when milk is held for 30 minutes at 63° C.; approximately onethird of the albumin is precipitated when the milk was held at 85° C. However, the albumin was not completely precipitated even though the milk o was held at 115 C. for 30 minutes. Influence of sugars upon protein coagulation: Comparatively little information is available regarding the influence of sugars upon the heat stability of milk proteins. However, early observations by Leighton and Mudge (92), in a study of the influence of forewarming of milk on the heat stability of the evaporated product during sterilization, have shown that the heat stability of normal milk is increased to a maximum by heating to 95° C. for 10 minutes. However, milk treated with approximately 20 per cent sucrose did not reach maximum heat stability until held for 30 minutes at 95° C. These authors believe that the protec­ tive action of sucrose solutions are due to the affinity of sucrose for - 25 calcium and phosphates, which mgy shift the salt balance to increase the heat stability of the milk. The possibility exists that sucrose may exert some in­ fluence upon the amount of the soluble proteins coagulated during the heating of milk. Beilinsson (12), Biddles (30), and Pay (32) have demonstrated that sugars tend to inhibit the heat coagulation of egg albumin at temperatures of approximately 70° C* Biddles determined the coagulation of egg albumin at pH 4.6 when held at 70° C. for 10 minutes, and found that glucose and fructose exerted increased protective action against coagulation with increased concentration. Complete pro­ tective action was afforded when the saturation point was reached. mannose and mannitol exerted some protective action, but Sucrose, were less effec­ tive than glucose and fructose. Influence of formaldehyde upon protein coagulations Evidence is available indicating that formaldehyde influences the voagulation properties of serum proteins. directly applied to milk; This information, however, has not been althou^i, as indicated in theprevious section, the action of formaldehyde has been studied in connection with browning. Barly studies of Blum (14) showed that a few drops of formal­ dehyde prevented the heat coagulation of albumin. Pischer (33, 34) found that 0.0045 to 0.022 per cent of formaldehyde would delay or inhibit the o denaturation of serum globulin held at 70 C* This treatment was more effective when the formaldehyde was added soon after denaturation started and had no effect if added 5 or 6 minutes later. Hcwever, Freeman (40) demonstrates that serum proteins are 25 to 70 per cent denatured by treat­ ment with 0.5 formaldehyde and holding at 37° C. for 10 days. Hensley (58, 59), and Marton and Reiner (95) have demon­ strated that the addition of formaldehyde caused a gelation of the protein. Hensley believes this effect to be due to condensation of protein molecules. Clark and Shenk (21) believe that formaldehyde reacts at the amide nitrogen groups of protein, tying adjacent chains together. Schiff (133) and Sorenson (143) believe that proteins are denatured by formaldehyde reacting with the free amino group, thereby converting proteins to acids. Sorenson states that formaldehyde reacts with amino acids to form either a methylene derivative or an aldehyde of ammonia. Influence of metals on the heat coagulation of proteins! Numerous papers have pointed out that certain metals combine with proteins and amino acids in rather definite proportions to form stable complexes. However, the ratio at which different metals combine with protein has been shown to vary. Osborne and Leavenworth (ll5) demonstrated that copper com­ bined with the peptide nitrogen of edestin in the ratio of 1 to 4. Bom (16) added copper sulfate and sodium hydroxide to edestin, gelatin and serum protein to show that these substances formed compounds containing from 4.2 to 6.3 per cent copper. Schom (l35) and Bechhold (ll) found that zinc, chromium and aluminum combined with albumin at the rate of 1 gram equivalent of metal to 5100-5200 grams of protein. Silver also com­ bines with albumin at this ratio according to Schorn (135) and Heyman and Oppenheimer (60). The latter authors found iron to combine with albumin at approximately twice the above ratio. Northrop and Kunitz (112) found that copper, magnesium, aluminum and silver formed stable compounds with - 27 gelatin. The copper combined at the ratio of 0.9 millimoles per gram of gelatin. In general, an alkaline medium has been used to bring about the formation of the metal-protein complex. Vickery and Gordon (155) found that mercuric chloride formed complexes most favorably with amino acids at pH 9.3 to 9.8. The complex contains mercuxy and nitrogen In the ratio of 3 to 2, and also contains chlorine and the base employed for precipitation. Jesserer and Lieben (72, 73) have shown that a number of metals react with casein in the presence of alkali. Kober and Sugiura (86) prepared copper salts of amino acids, peptides and peptones by adding copper hydroxide and holding at low temperatures. Borsook and Thimann (17) studied the influence of hydrogen-ion concentra­ tion on the equilibrium relations existing between cupric Ions and glycine and alanine at room temperature, and concluded that at least four types of complexes may exist. The copper tended to combine with the carboxyl groups in acid solution and with nitrogen groups in neutral and basic solution. Metals are attached to thenitrogen of the peptide linkage according to Osborne and Leavenworth (115), Northrup and Kunitz (112) and Jesserer and Lieben (73). However, Smythe and Schmidt (141) noted a correlation between the iron bound casein and the number of free carboxyl and phosphorus groups in the casein molecule. Michaelis (102) has shown on the basis of oxidation-reduction reactions that cobalt and iron form complexes with cysteine within the pH range of 7 to 8.5. The latter catalyzes the oxidation of cysteine. The amino group plays no part in the reaction , which appers to be due to the simultaneous presence of an -SH and -C00H group in the molecule. Galuialo 28 and Dobrotworskaja (42) state that proteins increase the oxidative and catalytic action of metals by protecting them in solution. The influence of metals per se on the heat coagulation of proteins has not been fully established. However, Benedicenti and Rebello-Alves (13) state that metal protein complexes axe no longer coagulalable by heat. Numerous workers, including Sommer and hart (142), Webb and Holm (159), Miller and Sommer (104) have shown the salts of certain metals to influence the heat stability of milk proteins. However, this influence concerns the salt balance of themilk rather than the action of specif ic metals upon the proteins in question. Influences of alcohol on the heat coagulation of proteins: Al­ though alcohol is used in the precipitation of certain proteins, only limited attention has been given to its influence upon the heat coagulation of these proteins. Henkel (57) was one of the first to show that an equal volume of 70 per cent alcohol does not always completely coagulate fresh mixedherd milk. proteins. He believes the coagulation action to be due to dehydration of However, Vasil*ev (152) measured the coagulating action of various alcohols on albumin and casein and found that coagulation begins at such low concentration of alcohol that the action is not one of dehydra­ tion, but is a surface energy change resulting from the adsorption of alcohol on the protein micelle. Klobusitzlcy (85) and Teorell (148) have shown that even small amounts of alcohol lower the temperature of proteincoagulation. (85) found that pseudo globulin manifests Klobusitsky its greatest turbidity below the boiling temperature with 18 per cent alcohol. Teorell (148) shows that - 29 protein coagulum formed, in such concentrations of alcohol tend to dissolve when heated to boiling and reprecipitate on cooling. Harris and Matill (55) found that extraction of kidney and liver globulins with boiling 95 per cent alcohol produced no decrease of nitrogen in these ptoreins, but both of the proteins show a loss of cystine and hydrogen sulfide. Hopkins (65) states the precipitation of protein from solution with alcohol causes the appearance of — SH groups. - 30 - SCOPS OP INVESTIGATION This study was conducted to obtain information concerning the volatile sulfides of milk products and the influence of certain factors thereon. (a) Data were secured with three main objectives in mind: To determine in a general way the origin of the heat labile sulfur compounds of milk by means of a quantitative method, and to determine the relationship of volatile sulfide liberation to total sulfur, total nitrogen and albumin nitrogen. (b) To determine the influence of various heat treatments upon the heat labile sulfur compounds, cooked flavor and color changes of milk. (c) To determine the influence upon the heat volatile sulfides of the addition of (l) sugars, (2) ethyl alcohol, (3) sodium chloride, (4) cystine or cysteine, (5) reducing substances, (6) hydrogen peroxide, (7) acids or alkali and (8) metals. - 31 - EXPERIMENTAL PROCEDURE General methods: Milk used in this study was obtained either by random selection from the daily deliveries received by the College Creamery* or, in case of controlled e2q>eriments, directly from the college herd. Such milk was used immediately either directly in experimental studies, or for preparing other milk products. secured by mechanical separation of the ing the fresh cream; Cream and skimmilk were fresh milk; buttermilk by churn­ and wh^y and buttermilk whey by rennin coagulation of the casein in the skimmilk or buttermilk. Samples used in homogeniza­ tion studies were momentarily heated to 135° C. to inactivate lipase (Gould (51) ), All samples were secured and handled in well-tinned or glass containers to avoid copper and iron contamination. Heating of the samples to tenperatures below boiling was accomplished in a water bath. Unless otherwise stated the samples were placed in a round bottom flask and agitated with a glass-rod stirrer, or by aspiration with nitrogen or air. Approximately 15 minutes were re­ quired to bring the samples to the desired temperature. was controlled within an accuracy of — 0.5° C. The temperature Cooling was accomplished by replacing the water in the water bath wi th cold water, thus bringing the sample to room temperature within 5 minutes. When samples for flavor or nitroprusside determinations were required* the heating apparatus included a glass-tube siphon which per­ mitted the removal of samples at desired temperatures. Flavor observations were made by two reliable judges, after first adjusting the samples to approximately room temperature. High temperature heat treatment of the milk, ie. , temperatures above "boiling was obtained by means of a steam autoclave* The milk was usually sealed and heated in No* 2 C&nco, plain sanitary cans (American Can Company). Approximately 15 minutes were required to reach the desired pressure. Following heating, the samples were cooled by immediately trans­ ferring the cans to cold running water. Qualitative methods; The qualitative studies on sulfide libera­ tion were conducted according to the procedure adopted by Gould and Sommer (45). The volatile sulfides were recovered on basic acetate paper by bubbling air through 4.5 liter samples while heating to the desired tenperature in a 5-liter round bottom flask* The discoloration of the acetate paper indicated the liberation of sulfides. The basic acetate paper was prepared by the method described by Gould and Sommer (45). Sulfhydryl determinations were made as suggested by Gould (48). A 5 cc. sample was saturated with ammonium sulfate, 10 drops of 5 per cent sodium nitroprusside and 5 drops of concentrated ammonium hydroxide were added. The tube was inverted several times after adding each reagent. The intensity of the pink color appearing indicated the extent of sulfhydryl formation. Quantitative methods? After conducting a number of preliminary trials involving the use of colorimetric and titration methods for the quantitative determination of volatile sulfides the Sheppard and Hudson (38) modification of the Almy (2) method with some additional variations was adopted. Kirklenko (82) has shown this method to give an accuracy of 0.001 - 33 per cent when used to determine the sulfides of geletin. The method involves the collection of the liberated sulfides in alkaline zinc acetate and then the production of methylene blue by the addition of p-amino dimethyl aniline and ferric chloride. The intensity of the blue color varies directly with the amount of sulfides. Alkaline zinc acetate was prepared by adding 25 cc. of a 20 per cent stock solution of zinc acetate and 40 cc. of 10 per cent sodium hydroxide to a liter flask and making up to volume with distilled water. The indicator solutions consisted of a 0.1 per cent p-amino dimethyl aniline (Eastman Kodak, No. 1333), in Isl hydrochloric acid and 0.02 molar ferric chloride (hexahydrate), in 4 per cent hydrochloric acid. The apparatus used for the quantitative determination of volatile sulfides is diagrammatically shown in Fig* 1# The quantitative procedure involved the heating of a 2-liter sample in a 3-liter flask suspended in a water bath. Moderate agitation was pro diced and the evolved gases were carried off by passing nitrogen gas through the samples during the heating period and for 30 minutes thereafter. Before passing through the sample, the nitrogen gas was freed from oxidants by bubbling through alkaline pyrogallol and distilled water. The rate of nitrogen flow through the sample was regulated so as to keep 5 bubbles apparent on the surface of the distilled water* Hie alkaline pyrogallol was prepared as recommended by the chemist of the U. S* Steel Corporation (101), whereby 300 gm. of pyrogallic acid are dissolved in a liter of distilled water and 2.5 volumes of 50 per cent sodium hydroxide added. Rubber stoppers used in the apparatus were freed of sulfur by boiling in strong sodium hydroxide and rinsing in an abundance of distilled - 34 water. Sulfur-free rubber tubing was used for connecting glass tubing. Blank determinations with, distilled water showed the apparatus to be free of sulfur. In order to recover the volatile sulfides, the nitrogen gas was conveyed to the two receivers by a 3-foot condenser and glass tubing. An empty flask was connected in the line between the condenser and the receivers to serve as a trap. 20 mm. in diameter. The receivers were 50 cm. in height and Each receiver was half filled with 4 mm. glass beads and contained 75 cc. of the dilute alkaline zinc acetate. Preliminary trials revealed that practically all of the sulfides were removed in the first receiver. At the end of the aspiration period, the solution and glass beads were removed from the receivers and the receivers were rinsed with distilled water and then with dilute hydrochloric acid, the rinsings were collected in separate containers. The acid rinse was withheld from the alkaline receiving solution to avoid the possibility of an error due to the liberation of hydrogen sulfide before the indicators were added. The rinsings were then added to the alkaline zinc mixture, along with 25 cc. of the diamine reagent and 5 cc. of ferric chloride and the flask was immediately stoppered, A few minutes were allowed for color development, after which the blue solution was placed into a 300 cc. volumetric flask. The beads and flask were rinsed several times with distilled water, the rinsings were added to the volumetric flask and the flask made up to volume with distilled water. After standing over night the intensity of the blue color was determined with an Evelyn photoelectric colorimeter, using a 620 35 J »»«< ■'«■»*i w t » »i m l-.yrT7,!,., t>i,Yi d s z o ESET'V. l^CL/ )’ns ======?/ w m j s ^ ^ 0 0 o -c —cr 0 0 @ @ © @ @ d J x § o ) q @ ( 5 s ) “ . s'u . ' u ■ i ' I ' 1, : i I M M ; l ‘ t ' 1 1 ! •' t '• Fig. 1 / D ia g ra m m a tic sHetch of the apparatus used for quantitative determination of volatile sulfides. th •1'* '■1'i'I1i11111 - filter. 36 The photometer readings were referred to a previously prepared standard curve for the sulfur values* A filter showing maximum light transmission at a wave length of 630 millimicrons was selected after finding that the representative color standard exhibited maximum absorption at a wave length between 600 and 700 millimicrons when observed with a Cenco-Sheard Spectrophotelometer. The typical transmission curve is shown in Fig* 2. The known color standards were obtained by a variation of the Lachele (90) procedure. Accurate amounts of standard sample 14 c Bureau of Standards basic open-hearth steel were placed in the sulfide apparatus and 50 cc. of 1:1 hydrochloric acid were added, the acid alone gave a perfect blank. The sample was gently heated until dissolved and then boiled for 5 minutes. Nitrogen gas was passed through the system during the heating period and for 10 minutes after heating. The hydrogen sulfide thus produced was collected in the receivers and the blue color was developed as described above. The photometer readings produced by recovering the hydrogen sulfide from different weights of steel of known sulfur content are shown in Table 4. After obtaining the photometer readings of the color standards produced from steel samples containing amounts of sulfur ranging from 0.03 to 1.50 mgs., the standard curve was prepared by plotting the logarithm of the photometer reading of the known standards against their sulfur con­ tent. The standard surve is presented in Fig. 3. Total sulfur was determined on a 25 gram sample according to the official magnesium nitrate method (6). This method was found to be accurate when checked against cystine samples of sufficient size to give barium sulfate precipitates equal to those generally obtained for normal milk. Total and albumin nitrogan was determined on a 5 gm. sample by the - 37 - lo o TRANSMISSION (%) do 350 SJO WAVE LENG TH Fig® 2. 590 6 70 (M IL L IM IC R O N S ) Typical transmission curve of a representative color standard of known sulfur content„ 7S0 official method (6), Table 4. Photometer readings for standard methylene blue solutions prepared from steel samples of known sulfUr content*. Grams of steel used Milligrams of sulfur in Photometer readings using 1 cc of the standard solution and a (0.03 fo sulfur) the steel sample 620 filter 0.1 0.2 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.030 0.060 0.150 0.300 0.450 0.600 0.750 0.900 1.050 1.200 1.350 1.500 * Each sample was determined in triplicate. 97.0 95.5 90.5 81.0 72.0 62.0 52.0 47.5 35.0 27.0 20.0 12.7 O u n j l n s o o ^ jo SWWFoTtTTW - 40 - RESULTS General Origin and Qiantitative Studies of Heat Labile Sulfides Critical temperature studies! To study quantitatively the critical temperatures of milk, skimmilk and rennet whey,samples were heated momentarily to temperatures ranging from 72° C. to 90° C. The heat labile sulfides evolved during heating and for 30 minutes after cooling are shown in Table 5. Table 5. Volatile sulfhr recovered from milk, skimmilk and whey momentarily heated at different temperatures. Temperature °C 72 74 76 78 80 82 84 86 88 90 Milk** (mg/l) 0.000 0.000 0.014 0.025 0.053 0.076 0.100 0.130 0.202 0.247 Skimmilk* (mg/l) 0.000 0.000 0.000 0.000 0.012 0.026 0.045 0.054 0.065 0.158 Whey* (mg/l) 0.000 0.000 0.000 0.000 0.020 0.040 0.065 0.124 0.158 0.205 ** Average of 6 determinations. * Average of 3 determinations. These data show that a small but measurable amount of sulfides are liberated when milk is heated momentarily to 76 to 78° C. and that the amount of sulfides liberated increase with increases in heat exposure throughout the entire heating range. Ho sulfides were detected in milk o heated to 72 and 74° C., but 0,014 mg. per liter are evolved at 76 C. and - 41 0.247 mg. per liter are liberated at a temperature of 90° C. Skimmilk and whey have critical temperatures from 2 to 4° higher than whole milk9 and these two pro diets do not evolve as much sulfide as whole milk. However, these products show the same relationship as whole milk in regard to in­ creased sulfide liberation with increased heat exposure* The skimmilk evolved 0.012 mg. of sulfur as volatile sulfides when heated to 80° C. and 0.158 rag. at 90° C. Whey evolved slightly more sulfides than skimmilk. but less than whole milk, the whey liberating 0+020 mg. of sulfur at 80° C. and 0.205 mg. at 90° C. The data in Table 5 also show that casein per se is not a con­ tributor of heat labile sulfur. In fact, casein appears to slightly inhibit sulfide liberation as indicated by the difference in values obtained for skimmilk and whey. However, serum proteins, or those proteins in whey, apparently are a definite source of heat labile sulfur. The fact that milk has a lower critical temperature and liberates sulfides in greater quanti­ ties than skimmilk and whey indicates a correlation with the butterfat con­ tent. To determine the influence of butterfat and the membrane ad­ sorbed on the fat globules upon the critical temperature and quantity of sulfides evolved, milk, cream (both 20 and 30 per cent), buttermilk and buttermilk whey (rennet) were secured from the same source. Heating trials were th^i conducted and the amount of volatile sulfide evolved at different temperatures from each of these products are presented in Table 6. These results show that cream, buttermilk and buttermilk whey have a lower critical temperature than milk. Furthermore, by comparing the data in Tables 5 and 6, it is evident that cream, buttermilk and buttermilk - 42 whey also have lower critical tempera tares than skimmilk or whey. These results demonstrate that 30 per cent cream has a slightly lower critical temperature than 20 per cent cream; 30 per cent cream liberated the first detectable sulfides at a temperature of 66° C. f whereas 20 per cent cream showed the first measurable liberation of volatile sulfides when momentarily heated to 70° C» In like manner, the corresponding buttermilk and butter­ milk whey from 30 per cent cream was found to have lower critical tempera­ tures than the buttermilk and buttermilk whey from 20 per cent cream. The buttermilk has a slightly lower critical temperature than its corresponding cream, and the buttermilk whey has a critical temperature approximately 2° C. below the buttermilk from vfcich it came. Table 6. Volatile sulfur recovered from milk and its corresponding cream, buttermilk, and buttermilk whey momentarily heated at different temperature s*. Cream Buttermilk whey Temperature Milk 20$ fat Buttermilk Cmg/i; (mg/i; (rag/l) (mg/l> °C 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 0.000 0.000 0.000 0.000 0.000 0.012 0.031 0.052 0.065 0.090 0.124 0.196 0.243 0.000 0.000 0.000 0.000 0.006 0.012 0.028 0.033 0.095 0.151 0.196 0.237 0.260 0.275 0.296 * Average of two trials. 0.000 0.000 0.000 0.012 0.028 0.040 0.060 0.082 0.098 0.158 0.192 0.237 0.282 0.305 0.330 : 0.000 0.000 0.006 0.022 0.036 0.057 0.082 0.106 0.127 0.167 0.200 0.252 0.290 0.337 0.375 Cream Buttermilk 30$ fat Buttermilk whey (mg/l) Tmg/l) (mg/lT 0.000 0.000 0.000 0.006 0.010 0.019 0.045 0.067 0.151 0.252 0.345 0.400 0.430 0.462 0.480 0.000 0.000 0.006 0.012 0.036 0.067 0.114 0.135 0.176 0.275 0.350 0.405 0.435 0.487 0.512 0.000 0.012 0.026 0.049 0.071 0.106 0.151 0.192 0.245 0.305 0.360 0.412 0.445 0.525 0.575 - 43 Another point of interest revealed ty the data in Table 6 is that cream, buttermilk, and buttermilk whey liberate considerably more volatile sulfides than either milk, skimmilk, or whey. For example, when milk, cream, buttermilk, and buttermilk whey are momentarily heated at 90° C. the following amounts of sulfur per liter were obtained by collecting the volatile sulfides from these products: 20 per cent cream, 0.2#6 mg. ; milk, 0.243 mg.; the buttermilk from 20 per cent cream, 0.330 mg.; and the tuttermilk whey from 20 per cent cream, 0.375 mg. The values obtained for 30 per cent cream and its corresponding buttermilk and buttermilk whey are 0*480, 0.5/jL and 0*575, respectively. The above values further substantiate the fact that casein apparently does not contribute to the amount of sulfides evolved. HJiie is best observed in Fig. 4 by comparing the similarity in results obtained for buttermilk and buttermilk whey. In fact, buttermilk whey evolves sulfides at a lower temperature and in greater quantities than the butter­ milk. In additi cn this portrayal clearly demonstrates the significance the material adsorbed on butter fat as a major source of heat labile sulfides. The proteins associated with the fat appears to be the greatest contributor of heat labile sulfides of any of these fractions studied. In Table 6, the results show that as the fat content increases the amount of sulfur recovered as volatile sulfides is also increased. These results show further that buttermilk and buttermilk whey, which contain the fat associated proteins in greater amounts per unit volume than the cream, liberate heat labile sulfur in much greater quantities than an equal volume of cream* The fact that 30 per cent cream and its corresponding buttermilk and buttermilk uriiey evolve sulfides at lower temperatures and in quantities 44 9o A T /// 1 / / ' o - ■ / / If / / / TEMPERATURE CO o 80 — / / o • A • A A J 7S -/ A / J/ A 7o U A !// L/ 1- M ILK 2-CREAM 3-BUTTERMILK H-BUTTERMILK WWfK 6f 4o 0 .0 0 I o ./a 0.24- I 0 .3 4 I 0 .4 6 O.bO SULFUR PER LITER (MG.) I*ig0 4. Th© critical temperatures of milk and its corresponding cream (30 $) , buttermilk and buttermilk whey, and the relationship of the temperature of heating to the quantity of sulfur evolved as sulfides by these products. approximately twice as great as the milk from which it came is significant proof of the importance of the fat membrane, or at least the fat associated proteins, as a major source of sulfides. Relationship of heat labile sulfur to total sulfur and total nitrogen; In order to ascertain more about the origin of heat labile sul­ fides, and, also, to study the nature of the constituents evolving sulfides when milk products are exposed to heat, various milk products were secured as in the previous section and the volatile &ilfides, total sulfur and total nitrogen were determined. These results are presented in Table 7, which presents an average of four typical determinations. Table 7. Relationship of the volatile sulfides obtained by heating milk momentarily at 90° C. to the total sulfur and total nitrogen*. Ratio Sulfur recovered Total Nitrogen Volatile sulfur Total sulfur sulfur as volatile to to sulfide total sulfur total nitrogen (mg/l) Cgms/i; tmg/l) 291 1 J1074 5. 62 1519.31 0*227 Milk 5.89 1519.37 151737 304 0.175 Skimm ilk 15 984 1.76 189 0.192 is 9.31 Whey 1:15.87 15494 235 3.73 0.476 Cream** 15579 1:18.68 5.66 303 0.523 Buttermilk 2.12 15 422 220 is 9.64 0.521 Buttermilk whey Product * Average of 4 determinations. ** Between 30 and 35 per cent fat. The volatile sulfide values were obtained by aspirating the sample while heating to 90° C., and for 30 minutes thereafter. 5he total sulfur was determined on a 25 gram sample and the total nitrogen was de­ termined on a 5 gram sample of the original pro duct. 2!he results on the volatile sulfide determinations compare the "buttermilk whey, this fraction appears to be the factor responsible for the fact that the ratio of volatile sulfur to total sulfur of butter­ milk whey is only twice that of whey. However, the ratio of total sulfur to total nitrogen is approximately the same for these two products. These findings indicate that the membrane constituents contain a fraction that is not especially hi^i in sulfur, but is relatively unstable toward heat. Washing of cream: Trials were also conducted to determine the influence of washing upon the fat globule membrane. In this study, cream of approximately 30 per cent fat was washed three successive times at 37.5° C. with equal portions of distilled water of the same temperature, after each washing and separation the cream was adjusted to the original volume and original fat test with the liquid obtained by the corresponding separation. The results for volatile sulfur at 90° C., total sulfur, total nitrogen and albumin nitrogen are presented in Table 8. Table 8* Influence of washing of cream on the volatile sulfides, total sulfur, total nitrogen, and albumin nitrogen of the cream and its buttermilk*. Product Cream** Buttermilk Washed cream Washed cream buttermilk Albumin Ratio Sulfur recovered Total Total Volatile as volatile sul­ sulfur ni trogen ni trogen Total sulfur sulfur fides to to total albumin nitrogen nitrogen (mg/1) (gms/l) (gms/l) (mg/1) 0.602 1:21.74 212 4.610 1:1337 0.450 0.650 5.680 1:19.59 1 :1329 0.489 290 0.640 95 0.072 1: 6.74 1:196 0.368 0.470 104 * Average of 3 trials. ** Fat content of 25 to 30 per cent. 0.805 0.083 l: 7.74 1:17? - 47 These results show that the constituent of the fat membrane contributing heat labile sulftir is rather firmly attached, Ihe washed cream has only a slightly lower volatile sulfUr value than the original cream, although the values for total sulfur, total nitrogen and albumin nitrogen of the washed cream are considerably lower than those of the original cream. The washed cream buttermilk is also shown to be high in heat labile sulfur and low in total sulfur, total nitrogen and albumin nitrogen. Also, the washed cream and washed cream buttermilk are shown to agree closely in the ratio of total sulfur to total nitrogen and in the ratio of volatile sulfur to albumin nitrogen. Separation temperature; In a further study of the heat labile properties of the material adsorbed on fat globules, samples of milk were separated at 15,5° C., 35° C. and 57.2° C. In addition, a sample of cream obtained by separating milk at 15,5° C. was reseparated at 57,2° C. to a hi^ier fat content, and then restandardized with the original skimm ilk, which was secured by separating milk at 15,5° C. Such a study appeared desirable since Sharp (138) has pointed out that the globule membrane apparently contains an adhesive material that remains with the cream when milk is separated at low temperatures, this material passing into the skimmilk when milk is separated at high temperatures. The volatile sulfur liberated at 90° C. and the total sulfur were deter­ mined on the cream and skimmi lk secured by separating at each of the above temperatures. These results are presented in Table 9. The values obtained for the volatile sulfur and total sulfur of the cream or skimmilk obtained by separating milk at a low temperature 48 vary only slightly from those obtained for cream and skimmilk secured by high temperature separation* These data indicate that the adhesive material remaining in cream at low terqperature separation and passing into the skimmilk at high temperature separation is either not a major source of heat labile sulfides and is not especially high in sulfur content, or the methods used were not sufficiently sensitive to detect a difference in values* Table 9* Influence of separation temperature upon die volatile sulfides and total sulfur of the resulting cream and skimmilk*. Skimmilk Temperature of Cream (25 per cent fat) separation Volatile sulfur Total sulfur Volatile sulfur Total sulfur Jmg/l) (mg/l) UC Cmg/l) 245.5 324.1 0.156 0.428 15*5 0.156 318.6 35*0 0.397 263.2 0.175 320.0 247.0 0.408 57.2** 0.172 318.1 254.0 0.388 57.2 * Average of 2 determinations. ** This was a sample of 25 per cent cream obtained by separating milk at 15*5° C. This 25 per cent was th&i reseparated to a higher fat con­ tent after heating to 57*2° C.» the cream sample was then restandardized to 25 per cent butterfat with skimmilk obtained by low temperature separation* Homogenization: To determine the influence of homogenization on the critical temperature, milk was heated to approximately 57° C. to largely inactivate the lipase (51), and a portion was homogenized at 2500 pounds pressure* The homogenized and unhomogenized milk was then studied from the standpoint of sulfide liberation, nitroprusside reaction and cooked flavor after momentarily heatii^ to temperatures ranging from 70 to 90° C. Results are presented in Table 10. - 49 Table 10. Influence of homogenization upon the sulfide liberation, nitroprusside test and cooked flavor of milk momentarily heated to different temperatures*. Temperature 70 75 80 85 90 Homogenized* Unhomogenized Volatile Nitroprusside Cooked Volatile Nitroprusside Cooked sulfur flavor flavor sulfUr test test tmg/l) (mg/l) 0 0.000 0 0 0.000 0 ? 0.010 0 0 0.015 0 1 1 0.052 1 0.043 1 0.129 2 0.129 2 2 2 4 4 4 0.237 0.235 4 * Average of two trials. The intensity of the nitroprusside test and cooked flavor is indicated as follows: negative, 0; questionable, ?; slight, 1; definite, 2; pronounced, 3; and strong, 4. These results show that homogenization of milk has no significant influence on the critical temperature nor upon the amount of heat labile sulfur evolved at the various temperatures. Furthermore, the sulfhydryl reaction as detected by nitroprusside and the cooked flavor of the homogenized and unhomogenized milk are also shown to be practically identical. Breed and period of lactation: conducted on mixed herd milk. The previous studies were It seemed advisable to demonstrate the influsice of the breed of cows and theperiod of lactation upon the critical temperature, amount of volatile sulfides evolved by heating, and the total sulfur. For this study four cows were selected, two Holsteins and two Guernseys. Numbers 280 and 48 were Just entering their lactation periods and numbers 271 and 51 had been producing for approximately 5 months. This scheme of selecting the animals was followed in order to study the milk from one animal of each breed throughout the entire lactation period, * 50 and also to study the milk from one animal of each “breed as she was com­ pleting her lactation period and again at the beginning of her lactation period. Unfortunately, number 51 had to be slaughtered during the second month of her new lactation period and only one sample was obtained after she freshened. The milk from these animals was collected over a 24-hour period once each month and the volatile sulfides and also the total sulfur values were determined. In each case the determination was conducted in duplicate and the average results are presented in Table 11. Table 11. July August Sep tember October Nov saber December January February Influence of breed and period of lactation upon the volatile sulfides and upon the total sulfur*. Guernsey Holstein 271 51 : 48 280 Volatile Volatile?Total iVolatile Total Total Volatile Total sulfur sulfur sulfur sulfur sulfur sulfur :sulfur :sulfur mg/l i mg/l : mg/l mg/1 mg/l mg/l mg/T mg/1 277 0.222 : 270 : 0.223 0.201 273 208 0.208 275 277 0.287 ! 273 ! 0.228 0.238 0.218 210 **** ** ** 285 0.286 ! 273 : 0.253 0.245 210 **** **** 279 0.282 : 281 : 0.245 242 0.252 **** . **** ; 0.234 249 306 247 0.192 0.256 269 242 0.275 : 278 : 0.267 0.215 255 0.250 ___ * _,_• **** **** 241 244 0.238 0.245 __ * *«**# 0.238 243 252 0.252 » • • V *These data are an average of duplicate determinations . **** jjpy period. The volatile sulfur and total sulfur values are shown to increase as the lactation period progresses although some variations occur. The re­ sults show that the two samples of Holstein milk agree rather closely for both the total sulfur and volatile sulfur. Likewise, the two samples of Guernsey milk correspond fairly well for both the volatile and total sulfur. However, the Guernsey milk which was slightly higher in fat and thus also in 51 - solids-not-fat, is slightly higher in volatile and total sulfur. Ihrther- more, in each of the samples of milk there appears to he a rather constant ratio of volatile sulflir to total sulfhr. Persistency of volatile sulfur liberation: To determine the tenacity with which the volatile sulfides liberated on heating are retained by the milk, milk was momentarily heated to 90° C. and the volatile sulfides evolved during heating and for each 30 minute period throughout a 4-hour aspiration period were determined. In addition, the sample was checked for intensity of the color produced by the nitroprusside and for cooked flavor at the end of each 30 minutes of aspiration. The average results for two trials are shown in Table 12. Table 12. Sulfide liberation, nitroprusside test, and cooked flavor of milk momentarily heated at 90° C. and then aspirated con­ tinually for four hours*. Period of aspiration Minutes During heating: 30 60 90 120 150 180 210 240 Volatile sulfur mg/l 0.333 0.190_ 0.041 0.036 0.034 0.031 0.025 C.025 0.019 Nitroprusside Test 4 4 4 4 4 3 3 2 Cooked flavor 4 4 4 4 4 3 3 2 * Average of two determinations. The results of this determination show that only a relatively small portion (0.033 mg.) of the total amount of volatile sulfur is evolved during the 15 minutes necessary to bring the sample to 90° C., whereas the major portion is evolved during the first 30 minutes of aspiration. H0w- - 52 ever, the sample liberated sulfides in decreasing amounts during each sub­ sequent aspiration period throughout the entire four hours* with a small quantity still being liberated at the end of four hours of aspiration. According to these results the method arbitrarily adopted for estimating volatile sulfur actually secured only half of the sulfides made available by the heat treatment, The intensity of the nitroprusside test and cooked flavor were identical after each aspiration period and showed no decrease in intensity until after three hours of aspiration. However, there was a definite nitroprusside test and cooked flavor in milk practically free from volatile sulfides. The length of time necessary for milk heated to 90° C. to subside in sulfide liberation was also studied. In this experiment, milk was heated to 90° C. within 15 minutes and then immediately cooled. During heating and cooling the sample was well agitated by means of a glass stirring rod. This milk was then stored at 5° C. A two liter sample of this milk was removed, warmed to room temperature, and examined while fresh for sulfide liberation during 30 minutes of aspiration, nitroprusside reaction and cooked flavor. These determinations were also made on the stored milk each day for a period of seven days. These results are shown in Table 13. These results show that milk stored at 5° 0. exhibits a de­ crease in sulfide liberation. The fresh milk liberated 0.186 mg. of sulfides during 30 minutes of aspiration, and after one day of storage another sample of this same milk liberated 0.117 mg. of volatile sulfur as sulfides. Even after seven days of storage a sanple of this milk liberated 0.049 mg. of volatile sulfur. - 53 - Table 13* Persistency of sulfide liberation, nitroprusside test, and cooked flavor of stored milk previously heated momentarily at 90 C.* Age of sample Fresh 1 day 2 days 3 days 4 days 5 days 6 days 7 days Volatile sulfur Hitropruss ide mg/l 0.186 0.117 0.094 0*086 0.079 0.068 0.056 0.049 Cooked flavor 4 4 3 3 3 2 2 2 4 4 3 3 2 2 2 1 * Average of two trials. The intensity of the nitroprusside reaction and cooked flavor was also found to decrease as the storage period progressed. change was noted after the second day of storage* The first The nitroprusside re­ action was still definite after seven days of storage* wh^eas the cooked flavor had practically disappeared. Influence of Variable Heat Treatments Upon the Labile Sulfur Compounds Function of time and temperaturei Previous investigations have shown that the liberation of sulfides by milk products is a function of time as well as temperature* In these trials the amount of heat labile sulfur, cooked flavor and sulfhydryl formation were compared on milk heated to temperatures rangiig from 72 to 95° C. for 0 minutes, 30 minutes and 60 minutes. The sulfides were collected during the heating period and for 30 minutes after reaching the desired temperature, the sample having been cooled to room temperature immediately after the desired heat treatment. were conducted and the average results are presented in Table 14. The trials - 54 - Table 14. IVinction of time and temperature upon the heat labile sulfides, cooked flavor and nitroprusside test*. Heating Period : Momentary 60 minutes 30 minutes Temper­ Volatile C.F.* N.F.iVolatile C.F.* N.P. Volatile C.F.* N.P. test test:sulfur test sulfur sulfur ature mg/l mg/l mg/l 0.019 0.012 1 3 3 1 70 0 0.000 0 4 4 0.106 0.032 1 1 0 0 75 0.000 4 4 0.134 2 0.110 3 80 1 1 0.031 4 4 4 0.158 85 0.106 0.123 4 2 2 4 4 4 0.176 0.167 4 4 4 0.137 90 4 4 4 0.206 0.208 4 4 4 95 0.157 * Intensity of cooked flavor (C*F.) or color (H.p. — nitroprusside test); 0 - negative; 1 - slight; 2 - definite; 3 - pronounced; and 4 - strong. These results are on an average of two trials. These trials demonstrate again that milk does not liberate heat labile sulfides when subjected to momentary heat treatment until a temperature above 75° C. is reached, and that sulfhydryl formation and a cooked flavor occur simultaneously with sulfide liberation. However, when milk is heated for a period greater than 30 minutes sulfide liberation, sulfhydryl formation and the development of a cooked flavor take place at temperatures as low as 70° C. These results also show that the amount of volatile sulfur liberated, and the intensity of thenitroprosside reaction and cooked flavor development are increased either by a longer period of heat exposure or exposure to a hi^ier temperature. momentarily heated to 95° C. as heat labile sulfides. When this milk was 0.157 mg. of sulfiir per liter were recovered However, when this milk was subjected to 95° C. for 30 or 60 minutes approximately identical volatile sulfur values (0.207 mg. per liter), were obtained. - 55 - Influence of previous heat treatment; To determine the influence of previous heat treatment upon the formation of heat labile sulfUr compounds, milk was subjected to a mild heat treatment previous to determining the volatile sulfides evolved when momentarily heated to 90° C. Samples of milk were subjected to temperatures from 50 to 60° C. for 30 minutes and to 62 to 79.5° C. momentarily and then cooled before determining the volatile sulfides in the usual manner. These samples of milk were also observed for sulfhydryl formation and cooked flavor development t. Table 15. Results are presented in Table 15. Influence of previous heat treatment upon the heat labile sulfides of milk momentarily heated to 90° CL*. Previous heat treatment Holding time Temperature Control 50 55 60 62.8 68.5 73.6 79.5 — .— .--- 30 minutes 30 11 30 ” Momentary t! (1 tl After heating to 90° C.** Volatile sulfur mg/l 0.232 0.215 0.215 0.225 0.215 0.226 0.218 0.202 * Average of two trials. ** All of these samples had a distinct cooked flavor and gave a strong nitroprusside test. The heat treatments previous to the volatile sulfide de­ termination ranged from just under the critical temperature to just above the critical temperature. The data in TaDle 15 indicate that when milk is exposed to temperatures slightly below the critical temperature the subsequent volatile sulfide determination showed only a slight varia- 56 - tion from the normal value* However, when milk is heated to a tempera­ ture slightly above the critical temperature the subsequent sulfides evolved show a slight decrease. Prolonged and high temperature; Prolonged exposure of milk to 95° C. was studied to determine the influence upon the quantity of volatile sulfides liberated during 30 minutes of aspiration after cooling and upon the nitroprusside test, cooked flavor and extent of discoloration. In these trials a 2-liter sample of milk was placed in a 3-liter flask and agitated with a motor-driven glass stirring rod while exposing to 95° C. for the indicated period* Results presented in Table 16 represent the volatile sulfides liberated during 30 minutes of aspiration subsequent to heating and cooling. Table 16. Ihe influence of heating milk for prolonged periods at 95° C. upon volatile sulfides liberated during a subsequent 30 minutes aspiration period and upon the nitroprusside test, cooked flavor and discoloration.1*' Period of exposure to 95° C. Volatile sulfur 0 minutes ti 5 it 15 ti 30 ii 60 i i 90 H 120 II 150 II 180 n 210 * Average of two trials. mg/l 0.158 0.165 0.192 0.203 0*208 0.091 0.060 0.053 0.012 0.000 Nitroprusside test 4 4 4 4 3 2 1 1 1 ? Flavor cooked ti ti ii si.caramel ii ii it it it Browning 0 0 0 0 0 1 1 2 3 4 - 5? The data in Table 16 demonstrates that an increasing amounts of sulfides are made available by exposing milk to 95° C. for periods of time up to 30 minutes of exposure* However, when the samples were ex­ posed to 95° C. for a period of 90 minutes or longer the amount of sulfide liberated by aspiration are progressively decreased. 95 When milk is held at C. for 210 minutes no volatile sulfides were secured during a subse­ quent 30 minute aspriation period* In a similar manner, the sulfhydryl o groups are shown to decrease with increased period of exposure to 95 C* o Milk exposed to 95 C. for periods of from 0 to 30 minutes showed a strong nitroprusside reaction. However, the intensity of the nitroprusside test decreased with increasing periods of exposure. After 210 minutes of ex­ posure the milk itself had become discolored so that it was questionable whether or not the color produced with nitroprusside could be accurately determined. A strorg cooked flavor was observed in the samples held at 95° C* for period ranging from 0 to 30 minutes. After holding the milk at this temperature for 60 minutes or longer a slight caramel flavor similar to that of evaporated milk was observed. Milk held at 95° C. for 90 minutes or longer took on a brownish discoloration, which increased in intensity with increased periods of exposure. Since the volatile sulfides and, also, the sulfhydryl group were found to decrease simultaneously with the appearance of browning there seems to be a possibility that the occurrence of browning of milk may be associe/bed with the disappearance of the heat labile sulfur compounds. - 58 Additional studies were conducted to determine the influence of prolonged heat treatment of milk placed in varied types of containers. In these trials a portion of the milk was sealed in 18 ounce cans and quart fruit jars and other portions were heated in open 18 ounce cans and in an open flask in which tiie milk was agitated "by a motor driven glass stirring rod. These samples were held at 90° C. for periods from 0 to 240 minutes, then after cooling to room temperature the volatile sulfides were collected during 30 minutes of aspiration. The data are presented in Table 17. Table 17. Period of exposure to 90° C. Influence of the type of heating vessel upon the sulfides obtained from milk held at 90° C. for different periods of time.* Sealed cans Open cans Sealed jars** Agitated in open flask mg/l 0.107 0.088 0.040 0.006 0.000 0.000 0.000 0.000 0.000 0 minutes 30 fl 60 » 90 » 120 « 150 « 180 " 210 » 240 » mg/l 0.132 0.166 0.178 0.121 0.095 0.064 0.033 0.006 0.000 mg/l 0.125 0.175 0.183 0.129 0.097 0.061 0.031 0.009 0.000 mg/l 0.120 0.170 0.178 0.110 0.093 0.060 0.027 0.006 0.000 « * Average of two trials **A1 though these standard fruit jars were sealed in the usual manner, the rubber gasket did not maintain the pressure that developed during heat treatment. These results show that milk sealed in cans and exposed to a temperature of 90° C. has a relatively small amount of sulfides liberated during a subsequent 30 minute aspiration period in comparison to that heated in other types of containers. The samples heated in sealed cans liberated - 60 - 0.20 (M6J hSEALED CANS Z-OPEN CANS SULFUX PEKLim 0.12 0.08 0.04 0.00 o SO too ISO Zoo M I N U T E S OF EXPOSURE TO ?0°C. Fig. 5. Besults secured by exposing different samples of milk to 90° C„ for different periods of time. 250 - 59 - only 0.107 mg* of sulfur per liter upon momentary exposure to 90° C., and showed a progressive decrease in amount of sulfides liberated after periods of exposure* long Only a trace of sulfur was obtained in the form of sulfides after 90 minutes of exposure and longer periods of exposure caused a complete disappearance of the volatile sulfides. Samples heated either in open cans* sealed fruit jars or an open flask with agitation gave practically identical values upon momentary heating, and liberated increasing amounts of volatile sulfur for periods of exposure to 60 minutes. H0wever, longer periods of exposure caused a progressive decrease in sulfides, until no sulfides were liberated after an exposure of 240 minutes. The lack of similarity of values between samples heated in sealed cans and sealed fruit jars indicates that pressure developed in the can during heating plays some significant role in causing the disappearance of the volatile sulftir compounds. These results are graphically protrayed in Fig. 5. Whey was also found to respond to prolonged heat exposure in a manner similar to milk. that casein is Results are presented in Table 18. This indicates not responsible for the disappearance of the volatile sul­ fides during prolonged heat treatments. In other studies, the influence of prolonged heat treatment was ascertained from the standpoint of comparing the amounts of volatile sulftir liberated from milk heated in sealed cans and in sealed fruit jars to temperatures racing from 90° C. for 0 minutes to 126.6° G. for 30 minutes* In addition these samples were compared for intensity of the nitroprusside test, cooked flavor and discoloration. in Table 19, Results are presented - 61 Table 18, The influence of prolonged heat treatment upon the heat labile sulfhr compounds of whey samples held for different periods of time at 95° C. Period of ex^ posure to 96 C. 0 minutes 5 minutes ft 15 rt 30 it 60 ft 90 ti 120 it 150 n 180 it 210 Trial 1 Sulfhr as Nitro­ volatile prusside sulfides* test mg/l 0.126 4 0.135 4 0.172 4 0.185 4 0.186 4 0.050 2 0.045 2 0.037 1 ? 0.012 0.000 0 Trial Sul fir as volatile sulfides* mg/l 0.122 0.132 0.172 0.189 0.180 0.049 0.045 0.036 0.012 0.000 2 Nitro­ prusside test 4 4 4 4 4 2 2 1 ? 0 * These values were obtained by aspirating the sample for 30 minutes immediately after heating and cooling. Table 19, Influence of high heat treatment upon theheat labile sulfides of milk. • • Observations after heating Heating sHolding period Volatile sulfhr :Nitro- :Cooked:Browning temperatureiat heating tprus side Iflavor ! :temperature mg/l ! t : ■" "°C. Sealed cans 4 0 0.110 3 0 minutes 90 H 4 0 0.096 3 20 90 It 2* 4 0.045 1 20 108.3 H 3** 2 0.010 1* 20 126.6 II 1* 0.000 3 30 126.6 90 90 108.3 126.6 126.6 * ** 0 minutes tt 20 it 20 it 20 tt 30 0.138 0.170 0.052 0.012 0.000 Glass jars 4 4 3* 2* 1* 4 4 4 3** 3** Results not reliable due to off color of milk. Not a typical cooked flavor, but a slight caramel flavor. 0 0 ? 2 3 — 62 — These results are very similar to those previously reported regarding the influence of prolonged heat treatment upon the volatile sulfides# Milk heated in sealed cans show maximum sulfide liberations when heated momentarily to 90° C. * whereas milk heated at 90° C. in sealed glass jars showed maximum sulfide liberation when heated for 20 minutes. However, when milk was heated in either of these containers to temperatures above boiling the amount of sulfides liberated during a subsequent 30 minute aspiration was decreased. Either an increase in temperature or increase in the period of exposure to heat resulted in decreases in sulfide liberation. The intensity of the nitroprusside test and cooked flavor are shown to decrease slightly with increased browning. The browning prevented accurate observation of the color formation in the nitroprusside test and the flavor developed in the samples heated above boiling was not a typical cooked flavor but more of an "evaporated milk" flavor. These results demonstrate once more that the volatile sul­ fides and sulfhydryl groups progressively decrease as the intensity of the discoloration progresses. Influence of Added Compounds Upon the Heat Labile Sulfur Compounds Sugarsi Quantitative determinations of the volatile sulfides from milk samples containiig 5 per cent of added sugar were conducted by momentarily heatiig the milk to 90° C. presented in Table 20. Results of three trials are - Table 20. 63 - Influence of various sugars upon the volatile sulfides liberated from milk momentarily heated to 90° C* Control mg/SI 0.215 0.222 0.236 Trial 1 Trial 2 Trial 3 5 $ sucrose mg/SI 0.192 0.191 0.200 5 $ dextrose mg/SI 0.142 0.151 0.176 5 $ lactose mg/SI 0.122 0.136 0.151 These results show tiiat 5 per cent sucrose has only a slight retarding influence upon the volatile sulfides evolved from milk momen­ tarily heated to 90° C.» but that dextrose and lactose show definite inhibitive properties. 2’ urther studies on the influence of sucrose upon sulfide liberation are presented in Table 21. added to milk at the rate In these experiments, sucrose was of 20 per cent by weight and the vo la tile sul­ fides were determined at temperatures ranging from 74 to 90° C. Table 21. Influence of sucrose upon the sulfide liberation of milk momentarily heated at different tempeatures.* Sulfur recovered as volatile sulfides Control ! 20 per cent sucrose added mg/l : mg/l 0.000 i 0.000 0.006 0.012 i 0.019 0.053 : 0.034 0.082 : 0.090 0.222 : Temp era tur e 74 76 80 82 90 » * Average of three trials. The control sample is shown to evolve 0.222 rag. of sulfur as volatile sulfides when momentarily heated to 90° C., whereas the sample containing 20 per cent of added sucrose evolved only 0.090 mg. of sulfur at this temperature. These results are portrayed graphically in Tig. 6, Fairly smooth curves are obtained when the amount of sulfides evolved by each series of samples are plotted against the heating temperature. Through­ out the temperature range studied, the sucrose is shown to exhibit a definite decreasing action upon the volatile sulfides. Alcohol; Ethyl alcohol (95 $) was added to milk at the rates of 12.5, 25, and 50 per cent by volume, and the volatile sulfides were determined in the usual manner. Influence of alcohol upon the volatile liberation of milk momentarily heated at different temperatures*. Per cent ethyl alcohol 0 12.5 25 50 Sulfur recovered as volatile sulfides 90° C. mg/l mg/l 0.075 0.210 0.230 0.290 0.300 0.310** ——— —* — 0.260** 00 CO c o • Table 22. The data are presented in Table 22. * Average of three trials ** Considerable alcohol was driven off and some difficulty was encountered in reaching this temperature; therefore, these determinati ons are not comparable. These results show alcohol to increase the amount of volatile sulfides liberated. The control samples liberated 0.075 mg. of sulfides at 82° C. and 0.2L0 mg. at 90° C. Whereas, when 12.5 per cent alcohol was added 0.230 mg. of volatile sulfur was obtained at 82° C. and 0.290 mg. at 90° C. Some difficulty was encountered in heating samples containing 25 per cent or more of alcohol. In the case of these samples some alcohol was driven over at 82 and 90° C., but was generally withheld from the re- - 65 - 1. CONTROL SULFUR PIR LITER (MG.) 0.20 2. 2 0 % SUCROSE ADDED 0.15 0.10 0.05 0.00 78 82 TEMPERATURE Fig. 6. ?0 ( ° C .) Influence of 20 per cent sucrose upon sulfide liberation by milk when heated to different temperatures. - 66 - ceivers by the foam trap. However, when 50 per cent of alcohol v/as added to the milk considerable difficulty was encountered in heating the sample to 82 C* within the usual 15 minute heating period and good checks were never obtained due to considerable amounts of the alcohol evaporating over and condensing in the foam trap. For this reason, attests to conduct the volatile sulfide determination at 90° C. on samples containing 50 per cent of added alcohol were unsuccessful. The resul ts in Table 22 indicate that even small amounts of alcohol decrease the stability of the heat labile sulfur compounds. In other trials the filtrate obtained by adding 95 per cent alcohol to milk in equal quantities and removing the precipitate, was also found to evolve 0.028 mg. volatile sulfur per liter when momentarily heated to 82° C. This indicates that milk contains heat labile sulfur compounds that are alcohol soluble. Salt: Sodium chloride was added to milk at rates of 5 to 25 per cent and the sulfides liberated upon momentarily heating the milk to 90° C. were determined. Table 23. Results are shown in Table 23* Influence of sodium chloride upon the sulfide liberation of milk momentarily heated to 90° C.* Sodium chloride concentration (per cent) « * 0 : 5 • • mg/l 0.230 I : ft : ft • • • 10 : , • 15 # • Sulfur liberated volatile sulfides : mg/l * mg/l mg/l 0.088 0.053 : 0.026 : : * Average of three trials. : 25 • * : : mg/l 0.010 - 67 - These results show that salt exhibits ability to markedly decrease the heat sulf ide liberation. In milk heated to 90° C., 5 per cent salt reduced the amount of volatile sulfur to about one-third the normal value. When 25 per cent salt was added only a trace of sulfides were evolved. These results are also shown graphically in Fig. 7. Qystlne and cysteines Trials were conducted to determine the influence of cystine (Pfanstiehl) and cy stein e-hydrochloride upon the volatile sulfides liberated from milk momentarily heated to temperatures ranging from 76 to 90° C. Table 24. Results are shown in Table 24. Influence of cystine and cysteine upon the liberation of volatile sulfides from milk momentarily heated to different temperatures * Temperature Control mg/l 0.012 76 80 0.053 0.082 82 0.222 90 Sulfur liberated as volatile sulfides Qysteine-ii per liter Cystine per liter 0.25 gram 0.5 gram 0.5 gram 0.25 gram mg/l mg/l mg/l mg/l 0.025 0.020 0.000 0.000 0.051 0.082 0.006 0.006 0.106 0.158 0.008 0.009 0.232 0.370 0.026 0.023 These results show that either 0.25 or 0.5 grams of cystine per liter of milk practically but not entirely inhibit sulfide liberation. However, in contrast, the addition of cysteine-bydrochloride to milk definitely increased the amount of sulfides evolved. It may be observed that there is practically no difference in the amount of sulfides secured when either 0.25 or 0.5 grams of cystine per liter were added to milk, but only about 10 per cent of the normal sulfide liberation was secured. How­ ever 0.25 grams of cysteine-hydrochloride produced about 35 per cent increase ~ 68 - SULFUR PER LITER (MB.) 0,20 O .IS OJO 0.05 0.00 O 5 IO IS 2o SALT C O N C E N TR A TIO N (%) Fig* 7* Influence of salt concentration upon the volatile sulfide liberation of milk momentarily heated to 90° C. 25 - 69 - over the normal value for volatile sulfur and 0.5 gram show an increase of approximately 70 per cent in the amount of volatile sulfides evolved. The results in Table 24 indicating that as the enhancing ability of cysteine on sulfide liberation is due to the action of cysteine on the volatile sulfur compounds in the milk and not to the liberation of sulfides by cysteine itself. This is indicated by the data in Table 25 in which blank determinations of the volatile sulfidesliberated by cystine and cysteine in distilled water are presented. #hen either 0.25 or 0.5 grams of cystine were added per liter of distilled water, no volatile sulfides were evolved even byboiling for 15 minutes. Likewise, when 0.25 gram of cysteine-hydrocholoride was used no sulfides were evolved. How­ ever, when 0.5 gram of cysteine was added to a liter of distilled water, momentarily heating to 90° C. or boiled for 15 minutes liberated only a trace of sulfides. Blank trials were also conducted using sodium cyanide and sodium sulfite, and these results are also presented in Table 25* The sodium sulfite solution did. not evolve sulfides, but the sodium cyanide solution did produce a slightly positive result indicating that the method used in this study for estimating sulfides is influenced by other reducing compounds. Reducing substances! Observations presented in the above section indicate that reducing substances enhance sulfide liberation from heat treated milk. Also, Jackson (70) and Gould and Sommer (45) have demonstrated that reducing substances form - SK groups in milk without the application of heat. Therefore, trials were conduted to determine the volatile sulfides of milk when momentarily heated to various temperatures. The results are presented in Table 26. - 7 0 - Table 25. Stability of reagents toward sulfide liberation or formation of reducing compounds when heated in distilled water. Tenperatur e ° C. 80 * 90 * 100* * Hedncing Compuunds as hydrogen sulfide 0.5 gm. ; 1.0 gm 0.5 gm. 1.0 gm. 0.5 gm. 0.5 gm. cystine : cystine cysteine cysteine sodium sodium • sulfite cyanide mg/l ; mg/l mg/l mg/l mg/l mg/l 0 : 0 0 0 0 0 0 : 0 0 0.009 0.006 0 — — — “—“ 0 : 0 0 0,013 • » * Heated momentarily ** Heated for 15 minutes Table 26. Influence of certain reducing compounds upon the sulfide liberation of milk momentarily heated to 90° C. Volatile sulfur evolved as sulfides 0.5 ppm. of 0.25 gm. sodium 0.25 gm. sodium sulfite per liter Temperature Control sulfur added cyanide per liter as sodium sulfide mg/l mg/l mg/l mg/l 0.385 0.000 72 ------0.555 0.076 82 “ • — • — — 0.395 0.345 0.218 90 These results show that either sodium sulfide, sodium cyanide, or sodium sulfite enhance the amount of volatile sulfur compounds evolved from milk. When 0.5 ppm. of sulfur as sodium sulfide was added to milk which was subsequently heated to 72 C.» the amount of volatile sulfides evolved were found to be 0.385 mg. per liter; a value less than the amount added to the milk (0.5 ppm. equals 0.5 mg. per liter). However, no volatile sulfides were secured from the control sample heated to 72° C. the sodium sulfide sample was heated to 82 0.555 mg. per liter were secured. Also, when C.» a volatile sulfur value of This value is only slightly less than a - 71 total of the 0.5 ppm. (or 0.5 mg. per liter) added to the milk plus the 0.076 mg. per liter evolved by the control samples at this temperature. Trials involving the use of sodium cyanide and sodium sulfite at the rate of 0.25 gram per liter were found to evolve approximately 70 to 90 per cent more sulfides than the control when heated to 90° C. Data were not secured showing the influence of these compounds when milk was heated at temperatures lower than 90° C. Hydrogen peroxide? To determine the influence of certain oxidizing agents upon the heat labile sulfUr compounds of milk, 30 per cent hydrogen peroxide was added to milk and the volatile sulfides were determined quantitatively in the usual manner. Trials were conducted on samples to which the peroxide was added previous to heating and on samples to which the peroxide was added to the sample immediately upon reaching 90° C. These results are presented in Table 27. and represent the volatile sulfur secured by aspiration during heating and for 30 minutes thereafter, at which time the sanqple was at room temperature. Table 27. Influence of adding peroxide before or after heating milk momentarily to 90° C. upon the sulfide liberation*. Amount of hydrogen peroxide per liter of milk Control 0.25 c.c. 0.5 d.c, Peroxide added Before heating to 90° C. After reaching 90^ C. Volatile hitroCooked Volatile UitroCooked sulfur prusside flavor sulfur prusside flavor test test mg/l mg/l 4 4 : 0.236 0.236 4 4 0.089 0.047 3 1 3 3** 0.032 0.064 0 1 3 3** * Average of two trials. ** These samples developed a smoky flavor. - 72 The results presented in Table 2/ demonstrate that peroxide added either before heating the sample or after the sample reaches 90° C. will reduce the quantity of volatile sulfides evolved and also will de­ crease the intensity of the color produced with the nitroprusside reagent. When the peroxide was added after the sample reached 90° C. less sulfides were evolved and the ni troprus si de test was less pronounced than when the peroxide was added previously to heating. These results show that 0.064 mg. of volatile sulfUr per liter was liberated when 0.5 cc. of peroxide o per liter was added previous to heating at 90 C.» ana that only 0.032 mg. were evolved when 0.5 cc. of the peroxide was added after heating the sample to 90° C. The cooked flavor was reduced in intensity only sli^itly by adding peroxide either before or after heating. Hhen the peroxide was added to the sample after heating the flavor of the milk was not a typical cooked flavor, but can best be described as a smoky flavor. lUrther trials were conducted in a qualitative manner in order to determine the amount of peroxide necessary to prevent volatile sulfide liberation and destroy sulfhydryl groups. In this study the peroxide was o o eitii er added to milk before heating to 90 C. or after heating to 90 C. The volatile sulfides were collected on basic lead acetate paper during a subsequent 30 minute aspiration period. These results are presented in Table 28. The results of these quantitative trials again demonstrate that peroxide tends to prevent volatile sulfide liberation from milk. More peroxide was required to prevent sulfide liberation if added previous to heating than when added subsequent to heating. For example, the addition - 73 Table 28. Volatile sulfides, nitroprusside test, and cooked flavor of milk treated with different amounts of hydrogen peroxide eith^* before or after heating to 90 C. * Hydrogen peroxide per liter Before heating After heating Volatile NitroCooked Volatile Nitro­ Cooked pruss ide flavor** sulfides prusside flavor ** sulfides test test ml 0.0 0.25 0.5 1.0 1.5 2.0 2.5 3.0 4* 4 3 3 2 2 ? 0 4 3 3 2 1 ? ? 4 4 3 3 3 2 2 2 4 1 0 0 0 - 4 4 ? ? 0 - 4 4 2 3 3 - - - * * Average of three determinations. All numbers indicate intensity. **Samples to which hydrogen peroxide was added, either before or after heating, possessed an off flavor best characterized as smoky. of 2*5 cc. of hydrogen peroxide prior to heating prevented the appearance of sulfides and sulfhydryl groups, whereas, the addition of 0.5 of peroxide per liter of milk previously heated to 90° C. prevented the liberation of volatile suliides durirg the subsequent aspiration period. As in previous trials, the addition of peroxide either before or after heating reduced the intensity of the flavor produced by the heat treatment. Ph and Iron Salts: To show the influence of pH upon volatile sulfide liberation, rennet whey was adjusted to various pH values with concentrated ammonium hydroxide or hydrochloric acid. The volatile sulfides liberated by momentarily heating to 90° C. were then determined quantitatively in the usual manner. Also, in this study a series of trials were conducted in which 0.5 ppm. of ferric chloride and 3 ppm. of ferrous chloride were added to #iey sables and the influence of pH was studied in the presence of these compounds. These results are presented in Table 29. - 74 - Table 29. Influence of pH and ferric or ferrous chloride upon the vola­ tile sulfides liberated by whey momentarily heated to 90 C. Control* pH mg/l 0.026 0.063 0.097 0.137 0.190 0.202 0.260 0.330 0.350 0.260 0.010 2 3 4 5 6 normal 7 8 9 10 10.5 Volatile sulfUr recovered as sulfides 0.5 ppm of ferric 3.0 ppm of ferrous chloride chloride mg/l mg/l ----0.006 0.005 0.019 0.024 0.009 0.125 0.030 ----0.151 0.167 0.172 ----0.230 0.267 0. 275 ---- - ----- 0.208 0.177 * Average of three trials. The data presented in Table 29 show that as the pH of the sample is lowered the quantity of sulfides evolved at 90° C. is decreased. In contrast, an increase in pH above normal to about pH 9 was accompanied by an increase in the quantity of volatile sulfides evolved. However, at pH value above pH 9 (pH 10 and 10.5) a decrease in the amount of sulfides evolved was observed with only a trace of sulfides being evolved at pH 10.5. The results dealing with the influence of pH alone are graphically presented in Fig, 8. Normal whey was found to liberate 0.202 mg. of sulfur as vola­ tile sulfides. At pH 4 this whey liberated only 0.097 mg. per liter and at pH 9 the amount of sulfur liberated as volatile sulfides increased to 0.350 mg. per liter. The results presented in Table 29 demonstrate that the addition of 0.5 ppm of iron as ferric chloride or 3 ppm of iron as ferrous chloride will reduce the quantity of sulfides liberated by whey; however, the de - crease in sulfide liberation produced by these two compounds is only slight. - 75 - SULFUR PER LITER (MG.) 0.30 0.20 0.15 0.00 pH Xn.flu.ence of "oH u."pon "tin© libeX'cition of volatile s\ix,fldes by whey momentarily heated to 90 C. - 76 - The form in which the iron is added is apparently of some si^iificance since 0.5 ppm. of iron as ferric chloride was as effective in redning sulfide liberation from whey heated to 90° C. as 3 ppm. of iron when added as ferrous chloride. The pH of samples is shown to have only a limited influence upon the ability of these compounds to reduce sulfide liberations. At pH values above 8 the ferrous chloride is shown to in­ crease in effectiveness. Additional studies on the influence of metals? The in­ fluence of other metals on sulfide liberation from milk was also studied. The results secured when 0.5 ppm. of copper as copper sulfate is added to milk momentarily heated to temperatures ranging from 74 to 90° C. are presented in Table 30. Table 30. Influence of added copper upon the volatile sulfides, ni tro­ prus side test, and cooked flavor of milk when heated momentarily at 90° C.* » 9 Temperature 74 76 78 80 82 84 86 88 90 Control Co oked Volatile flavor sulf ides mg/l 0 0.000 0.012 ? 1 0.016 2 0.045 2 0.075 0.095 3 3 0.127 4 0. 185 4 0.235 * Average of three determinations. 0.5 ppm* of copper added Volatile Cooked sulf ides flavor Blg/1 0.000 0 0.004 0 0.006 0 0.008 0 ? 0.010 0.012 1 0.016 2 0.045 3 0.075 3 77 These results show that 0.5 ppm. of copper practically pre­ vents sulfide liberation from milk until the sample is heated to 86° C# or higher. A cooked flavor was not detected in this milk until heated to a temperature of 84° C. * whereas, the control sample developed a cooked flavor when momentarily heated to 76° C. The results presented in Table 30 are also presented graphically in Fig. 9. Additional trials were conducted to compare quantitatively the influence of various metals upon the sulfide liberation, sulfhydryl formation and cooked flavor of milk. 31. These results are presented in Table Copper, mercury and silver are equally effective in retarding sulfide liberation by milk momentarily heated to 90° C. The amount of sulfides liberated by milk is reduced with the addition of increased amounts of these metals* When 1 ppm. was added only a small amount of sulfides were liberated from the milk, but a definite ni troprus side test was apparent and the cooked flavor was pronounced. Also, in Table 31 additional data are presented showing the influence of ferric and ferrous iron. Although both of these iron compounds decrease sulfide liberation, the ferric iron was more effective. pounds were found to These com­ have only a slight effect upon the nitroprusside test and cooked flavor. The influence of the negative ion was also studied as a posslbl factor in the sulfide liberation of milk* qualitative trials were con­ ducted In which a number of copper salts were added in increasing amounts to milk previous to heating to 90° C* The amount of sulfides liberated were collected on lead acetate paper. The results as shown in Table 32 indicate that the negative ion exhibits no influence upon the amount of SULFUR PER LITER (MG.) 0.20 /. C O N T R O L 2. 0 . 5 P /> M . O f Cu 045 040 o- 0.00 3 62 6b T E M P E R A T U R E (°C) Fig. 9. Influence of 0* 5 ppm, of copper upon the quantity of sulfides liberated by milk when momentarily heated to different temperatures,' - Table 31. ppm. of metal. Control Copper sulfate » ii ii Silver nitrate n u ti ii Mercuric acetate ii tt it it Ferric chloride ti - Influence of different metals upon the volatile sulfides, nitroprusside test, and cooked flavor of milk momentarily heated to 90° C. * Metal salt added before heating ii 79 if n Ferrous chloride n ii h ii ii II If 0.00 0.50 0.75 1.00 0.50 0.75 1.00 0.50 0.75 1.00 0.50 0.75 1.00 0.50 1.00 3.00 5.00 Sulfur recovered as volatile sulfides mg 0.236 0.032 0.028 0.019 0.032 0.027 0.019 0.036 0.028 0.019 0.135 0.090 0.045 0.225 0.174 0.155 0.053 Nitroprusside test Cooked flavor 4 3 2 2 3 2 2 3 2 2 4 4 3 4 4 4 3 4 4 3 3 4 3 3 4 3 3 4 4 3 4 4 4 3 * Average of two trials. sulfides liberated. These metals salts were all effective in reducing the intensity of the cooked flavor and of the ni troprusside reaction. When 1.25 ppm. of copper was added to milk previous to heating the intensity of the ni troprus side test and cooked flavor were only slight in the fresh sariples and were absent after holding the samples for 48 hours* However, an oxidized flavor developed after 48 hours of storage in the samples con­ taining 1.25 ppm. or more of copper. In the literature review, certain metals have been reported to react differently when added to the proteins. To ascertain the in­ fluence of certain other metals upon the sulfides liberated by heat treated 80 - Table 32. Influence of various copper salts upon the sulfides liberated by milk while fresh and upon the sulfhydryl reaction and cooked flavor of milk after different periods of storage at 5° C. • • Amount of • After storage metal used Sulfides JNitroprusside :Cooked flavor (ppm.) liberatedsreaction : :4 hours 48 hours 14 hours 48 hours Compound used • • Control Copper nitrate Copper acetate Copp er chi or ide Copper sulfate 0,00 0.25 0.75 1.25 2.50 4 3 2 ? 0 : : : : ; 4 3 2 1 0 3 ? 0 0 0 : : : : i • • 4 3 3 1 0 3 1 0 ox** ox** • ♦ * Average of two trials. The relative intensity is reported by number, ** Indicates oxidized flavor. milk, trials were conducted in a qualitative manner and are presented in fable 33, Table 33, Influence of different metals upon the sulfides liberated by milk while fresh and upon the sulfhydryl reaction and cooked flavor of milk after different periods of storage at 5° C, i l After storage Amount of Cooked flavor metal used Sulfides :Nitroprusside liDeratedireaction (ppm.) 4 hours 48 hours 4 hours 48 hours « « Compounds used 4 « Control Nickel acetate Stannous chloride Alum inum chior ide Manganese chloride 0.00 5.00 5.00 5.00 5.00 4 4 4 4 4 4 4 4 4 4 3 2 3 4 4 4 4 4 3 2 3 Table 33 shows that nickel, tin, albumin, and manganese chloride are wi hiout material effect upon sulfide liberation, sulfhydryl formation and cooked flavor. 81 - Miscellaneous factor si Since sunlight has "been shown to affect the stability of proteins, a study was conducted to determine its influence upon the heat labile sulfides of milk. The milk was exposed to direct sunlight for three hours by placing a 2-liter sample into a 5-liter round bottom flask and agitating with a motor driven gLass stirring rod. In addition trials were conducted on milk to deteraine the influence of formaldehyde, mercurous chloride or iodine upon the volatile sulfides of milk. Table 34. These results are presented in Table 34. The volatile sulfides liberated by milk exposed to sunlight or treated with formaldehyde, mercurous chloride, or iodine. Sulfur evolved as sulfides at 90° C. Trial 2 Trial 1 Trial 3 mg/l mg/l mg/l 0.227 0.220 0.230 Treatment of sample Direct sunlight for 3 hours Formalin (0.5 cc/l) 0.060 0.055 0.050 Mercurous chloride (0*5 gram/l) 0.012 0.015 0.006 Iodine (0.5 ppm.) 0.082 0*086 ----- These results show that three hours of exposure to direct sunlight does not influence the amount of sulfides liberated by milk momentarily heated to 90° C. However, the addition of small amounts of formalin, mercurous chloride or iodine markedly reduce sulfide liberation. - 82 DISCUSSION Quantitative studies of the critical temperatures of milk products have pointed toward two general sources of heat labile sulfides. First, the serum proteins are designated as a source of heat labile sul­ fides by the fact that rennet whey evolves sulfides at the same tempera­ ture and in slightly greater quantities than skimmilk. Similarly, butter­ milk wh^r evolves sulfides at a slightly lower temperature and in greater quantities than the buttermilk from which it came. Bn.ese findings demon­ strate fairly conclusively that the serum proteins are a definite source of heat labile sulfides and that casein per se is not a contributor of heat labile sulfide evolved by milk at temperatures ranging from 76 to 90° C. The fact that cream, buttermilk and buttermilk whey all evolve sulfides at lower temperatures and in greater quantities than either milk, skimmilk or whey demonstrates that the material associated with the butterfat Is the second source of heat labile sulfides and the major source of sulfides in products containing rather high concentrations of the material adsorbed on butterfat. In addition, these critical temperature studies have shown that the membrane material is less stable to heat than the other proteins contributing heat labile sulfides. The lack of heat stability characteristic of the fat membrane is further demonstrated by comparison of the ratio of volatile sulfur to total sulfur and the ratio of total sulfur to total nitrogen of the various milk products. Skimmilk and IxLttermilk are higher in total sulfur and - 83 ~ nitrogen than milk, whey* cream or buttermilk whey. However, skimmilk liberates fewer volatile sulf ides than any of the above mentioned products and buttermilk is excelled in sulfide liberation only by buttermilk wh^r. This is significant proof that the membrane material is less heat stable than other milk proteins and that casein not only does not contribute sulfides itself, but actually slightly inhibits sulfide liberation. This conclusion is readily apparent when observing that whey and buttermilk whey liberate sulfides in slightly greater quantities than the skimmilk and buttermilk from which th^r were prepared and that the ratio of volatile sulfur to total sulfur of but term i3.k is about twice that of whey. Washed cream studies indicate that the material adsorbed on butterfat is significantly different in composition to the casein present in milk. The ratio of total sulfur to total nitrcgen for washed cream and washed cream buttermilk is about twice that of the cream and buttermilk. However, other data show that the ratio of total sulfur to total nitrogen of whey and buttermilk whey are approximately the same and are about twice that of the skimmilk and buttermilk from which th^r came* Therefore, it appears that the material associated with the fat phase of dairy products is not especially different in sulfur content from the serum proteins. The trials presented, in this paper, however, are not of sufficient number to permit the drawing definite conclusions regarding the composition of the membrane material. Sharp (137) describes an adhesive material associated with the fat phase of milk as being removed from the fat by high temperature - 84 separation. However, neither three successive washings with distilled water nor high tenp eratur e separation produced a significant change in quantity of volatile sulfides secured from cream. This indicates that the precursor of the volatile sulfides is rather firmly attached to the fat globules. Homogenization studies showed the fat membrane and also the other proteins of milk to be unaffected as far as heat labile sulfides were concerned* One might expect a change in critical temperature or quantity of volatile sulfides as Bull (19) has produced denaturation of proteins by shaking. However, Menefee, Overman and Tracy (100) found that homogenization had no significant effect upon the nitrogen distribu­ tion of milk. The fact that milk evolves sulfides after four hours of con­ tinued aspiration after heating indicates that volatile sulfides are made available by heat, but that th^r are not volatilized immediately. This phenomenon is perhaps due to protein rearrangement, as it is unlikely that the volatile sulfides are held in solution during this long period. Also, the fact that aspiration causes a disappearance of volatile sulfides at a much faster rate than the disappearance of sulphydryl groups is a further indication that volatile sulfides are freed by milk proteins for a rather long period after heat exposure. However, storage trials indicate prolonged holding of milk at 5° C. will either allow the sulfides to escape or allow changes to occur in milk whereby the sulfides are recom­ bined with the milk. The fact that protein rearrangement may occur in milk whereby the vola,ti le sulfides are no longer available is enphasized by studies - 85 - showing that the maximum amount of sulfides liberated by heat treatment fippear after 30 minutes of exposure to 90° C« After one hour of heating there was no change in the quantity of sulfides available and continued heating was found to cause a gradual disappearance of the volatile sul­ fides of milk. IJo sulfides were evolved by milk heated for 3^- hours* This fact can best be explained by suggesting that active groups appear in some milk constituents which combine with the sulfides. The appearance of active groups is indicated by the appearance of a slight discoloration of the milk. Discoloration and the disappearance of volatile sulfides was found to occur in a shorter time when milk was exposed to pressure crea,ted by heating in a closed container or by heating in either a closed or open container in an autoclave. well as the temperature In either case it appears that the pressure as evolved played a significant part in the more rapid disappearance of sulfides. Since the volatile sulfides and sulfhydryl compounds were found to disappear from whey in about the same time as from milk when these products were heated for prolonged periods indicate that casein is not necessary for the disappearance or recombination of sulfides. Studies bn the influence of various reagents upon the volatile sulfide liberation by milk demonstrate in a general way that reducing sub­ stances enhance sulfide liberation, \daereas oxidizing agents prevent sul­ fide liberation. Undoubtedly this action is due to the greater tendency toward the formation of sulfhydryl groups. Clark (20) was one of the first to demonstrate conclusively that sulfite acted upon cystine to form cysteine. In general, sugars even in relatively low concentration depress sulfide liberated from heated milk. Sugars having potential aldehyde or ketone groups, such as dextrose and lactose, have a more pronounced depress— - 86 - ing action upon sulfide liberation than sucrose. However, 5 or 20 per cent of sucrose did decrease sulfide liberation, but did not Influence the critical temperature. The exact mechanism whereby sucrose depressed sulfide liberation is not fully understood. However, sugars are known to exert a definite action toward inhibiting protein coagulation and the deveased coagulation of proteins in the presence of sugar appears to be the explanation for the decreased sulfide liberation. Sodium chloride and formaldehyde also inhibit sulfide liberation by milk. This, also, may be explained by the depressing action of the compounds upon protein coagulation as has been demonstrated by Klobusitzky (83, 94). In contrast, alcohol is known to enhance heat co­ agulation and in this study even small concentrations of alcohol enhanced sulfide liberation. Results of this study indicate that cysteine and certain re­ ducing agents favor sulfide liberation by their action on the milk proteins. Since sodium cyanide and sodium sulfite are known to cause the occurrence of sulfhydryl group in proteins even without the application of heat and, since the compounds are shown to enhance sulfide liberation, it appears that the formation of sulfhydryl groups is the first step in sulfide liberation. Therefore, any method of freeing sulfhydryl groups will enhance sulfide liberation. Cystine was found to inhibit sulfide liberation* Since Mirsky and Anson (l06) have shown that cystine acts as a milk oxidizing agent toward sulfhydryl groups, one may logically conclude that the blocking of free -SH groups will prevent the formation of sulfides. This conclusion would also explain the inhibitive action of hydrogen peroxide upon the - 87 liberatim of volatile sulfides. Since peroxide was more effective when added after heating than when added before heating indicates that the peroxide must be available to react with the -SH groups or the sulfides themselves in order to be effective. Copper, mercury and silver appear to combine with the sul­ fides liberated by heated milk. These metals retarded sulfide liberation when added at the rate of 0.5 ppm. but did not prevent sulfide liberation until used in concentration of 1.25 ppm. That the metal ion is the active agent is emphasized by the fact that the negative ion was shown to exert no significant influence upon volatile sulfide liberation. Other metals, less active oxidizers than copper, mercury and silver, were shown to be practically without effect upon sulfide liberation. - 88 - SUMMARY AND CONCLUS IONS The heat liberation of volatile sulfides by different milk products was studied quantitatively, When milk products are brought to temperatures of 64 to 80° C. in 15 minutes and then immediately cooled volatile sulfides are liberated abruptly and the amount liberated in­ creases with higher temperatures. By aspirating the sample with nitrogen during heating and for 30 minutes thereafter the amount of sulfide evolved by different milk products was found to vary considerably. Upon momentary heating volatile sulfides are evolved from milk products as follows: Product Critical Temperature Sulfur evolved as sulfides when heated to 90° c.____________ Milk Skimmilk Whey Cream (20 per cent) Cream (30 per cent) (a) Buttermilk (b) Buttermilk whey 76 to 78° C80° C. 80° C. 70° C. 68° C. 66° C. 64° C. 0.240 0.160 0.205 0.386 0.480 0.512 0.575 mg. per liter •' « « « « » Casein per se plays no significant role in the liberation of volatile sulfides by heated milk. The serum proteins and the material ad­ sorbed on the fat globules are the major sources of hea.t labile sulfur from mi 31c. The amount of volatile sulfides liberated by milk when heated to 90° C. shows no relationship to the total sulfur or total nitrogen con­ tent of the different milk products. - 89 - The material, associated with the fat membrane* which is a mao or source of volatile sulfiir, is not removed by washing three successive times with warm, distilled water. This material is not especially high in total sulfUr but differs considerably in heat stability from the other constituents contributing volatile sulfides. The amount of sulfides evolved by milk are decreased by con­ tinued aspiration or during storage at 5° C. ; however, a small amount of sulfides are evolved by milk after 4 hours of continued aspiration or after 8 days of storage at 5° C. The temperature of separation, homogenization at 2,500 pounds pressure, previous heating to temperatures below the critical temperature, or exposure to direct sunlight have no influence upon the temperature at which sulfides are liberated by milk products. The milk from Guernsey cows is slightly higher in total and volatile sulfur than the milk from Holstein cows. The amount of sulfides evolved by milk and the total sulfur content was found to increase as the lactation period progressed. When milk is heated at 90° C. the amount of sulfides evolved increases with an increase in the time of exposure up to 30 minutes. How­ ever, after 1 hour of exposure the amount of volatile sulfides decreases until no sulfides are evolved by milk held 3§ hours, at which time the milk was also negative to the nitroprusside test, had a caramel flavor and a brownish color. Milk held in sealed cans at 90 C. for 1^> hours did not evolve sulfides. Milk antoclaved in sealed cans or glass jars for 30 minutes at 126.6° C. did not evolve sulfides or contain sulfhydryl groups. - 90 Five per cent of sucrose, dextrose or lactose added to milk reduced the quantity of sulfides liberated. Hie addition of 20 per cent sucrose had no influence upon the temperature at which sulfides are liberated, but decreased the quantity by 50 per cent. Increased concentrations of sodium chloride progressively decreases the quantity of heat labile sulfides liberated by milk. a concentration of 25 per cent was used practically no sulfides When were evolved. The addition of 10 to 20 per cent of alcohol increases volatile sulfide liberation. The addition of a small amount of cystine to milk inhibits sulfide liberation, whereas, the addition of a small amount of cysteine enhances sulfide liberation. Sulfide liberation is increased by adding small amounts of sodium sulfide, sodium sulfite, or sodium cyanide. Sulfide liberation is increased by raising the pH above 6.5 and decreased by lowering the pH below 6.5. Sulfide liberation by heated milk may be prevented by adding a small amount of hydrogen peroxide, formaldehyde, mercurous chloride or as much as 1.25 ppm. of either copper, mercury, or silver as a salt. compounds are more effective when added after heating. These Iron has only a limited ability to decrease sulfide liberation and tin, aluminum, maganese, and nickel are without effect. 91 - LITERATURE CITED 1- Abel* J. J. and Geiling, E. M* K. 1925. Researches on Insulin. I. Is Insulin an Unstable Sulflir Compound? Jour* Riarmacol. and Exp. Ther. 25: 423-48. 2* Almy, L. H. 1925. Method for the Estimation of Hydrogen Sulfide in Proteinaceous Pood Products. Amer. Chem. Soc. Jour. 47: 1381-90. 3. Ambler, J. A1929. The Reaction Between Amino Acids andGlucose. and Eng in. Chem. 21: 4/-50, 4. Ahsbacher, S., Flanigan, G.E. and G. C. Supplee 1934. Certain Foam Producing Substances of Milk. Sci. 17: 723-31. Indus, Jour. Dairy 5* Ahson, M. L. 1939—40. The Reactions of Iodine and Iodoacetamide with Native Egg Albumin. Jour. Gen. Physiol. 23 : 321—331. 6. Association of Official Agricultural Chemist, Washington, D. C. 1940. Official and Tentative Methods of Analysis. Fifth Ed. p. 132, 270, 271. 7. Baernstein, H. D. 1932. The Sulphur Distribution in proteins. 97: 669-74. 8. Jour. Biol. Chem. _ 1936. A New Method for the Determination of Methionine in Proteins. Jour. Biol* Chem. 115: 25—32* 1936. The Sulphur Distribution in proteins. II. The Combined Method for the Determination of Cystine, Methionine and Silphates in Hydriodic Acid Digests. Jour. Biol. Chem. 115: 33-6. 10. Beach, E. P., Bernstein, S. S., Hoffman, 0. D. , Teague, D. M. and I. G. Macy. 1941. Distribution of Nitrogen and Protein Amino Acids in Human and Cow*s Milk. Jour. Biol. Chem. 139: 57-63. 11. Bechhold, H. 1928. Albumin-Metall sal ze. Biochem. Zescbr. 199: 451-8. - 92 - 12, BellInsson, A. 1929. ©lermostabilisation der Eiweisslosunge mit Rohrzucker und Glycerin* Biochem. Ztschr. 213: 399-405. 13. Benedicenti, A. and Rebello-Alves, B. 1922. Electric Cataphoresis of Metallic Protein Compounds Ob­ tained by Treatment with Powdered Metals. Arch. Intern. Pharra&co-dynamie, 26: 297-316. (Chem. Abs. 16: 2698). 14* Blum, P. 1896-97. Ueber eine Neue Klasse von Uerbindungen der EiweisskBrper. Ztschr. f* Physiol. Chem. 22: 127-31. 15. Bluraenthal, Doris and H, T. Clarke 1935. Unrecognized P o m s of sulfur in Proteins. 110: 343-49. 16. 17. Born, S. 1912. Jour. Biol. Protein—Copper Compounds. Biochem. Bull. 2, 166 (Columbia Univ.). (chem. Abs. 9: 3071). Borsook, H. and Tbimann, K. V. 1932* The Cupric Complexes of Glycine and Alanine. Chem. 98 : 671-705. Jour. Biol. 18. Brand, E. and Sandberg, M* 1926. The Lability of the Sulfur in Qystine Derivativies and its Possible Bearing on the Constituion of Insulin. Jour. Biol. Chem. 70: 381-95. 19. Bull, Henry B. 1937. The Denaturation and Hydration of Proteins. II* Surface Denaturation of !Egg Albumin. J 0ur. Biol. Chem. 118: 163-75. 20. Clark, H, T. 1932. The Action of Sulfite upon Cystine. 235-48. Jour. Biol. Chem. 17: 21. Clark, G. L. and Shenk, J. H. 193V. X-Rsy Diffraction Studies of Globular Proteins. III. The Action of Formaldehyde on Proteins. Radiology 28: 357-61. 22. Dhhle, C. D., Lawhorn, R. K. and Barnhart, J. L. 1940. The Effect of Certain Factors on the Keeping Qpality of Frozen Cream. Reprint from Vol. II of 40th Ann. Con. Internat. Assoc, of Ice Cream Mfrs., Atlantic City, New Jersey. Pages 7-23. 23. Dahle, C. D. and Palmer, L. S# 1937. Oxidized Flavor in Milk from the Individual Cow. Agri. Sxpt. Sta. Bull. 347: 30 pp. Penn. - 93 - 24. Daniels, A. D* • Giddings, M. L. and Jordan, D. 1929. Effect of Heat on the Autineuritic Vitamin of Milk. Jour. Nutr. 1: 455-66. 25. Daniels, E. P. and i&insell, H. E. 1937. Vitamin Content of Foods. Pub. 275* 175 pp. U. S. Dept. Agr. Misc. 26. Davies. W. L. 1939. The Chemistry of Milk:. Second Ed. 534 pp. illus. R. Van N0strand Co., New lork. 27* Diemair, W., Strohecker, R. and Keller, H. 1939. Beitrog. zur Kenntnis Fluchtiger Schwefelverbindungen. Ztschr. f. Analyt. Chem. 116S 385-403. 28. Doan, P. J. and Meyers, C. H. 1936. Effect of Sunlight on Some Milk and Cream Products. Milk Dealer 26: 76-87. 29. Dutcher, R. A., Guerrant, N. B. 1934. Vitamin Studies. XX. Pasteurization on the of Cowfs Milk. Jour. 30. Buddies, W. J. 1932. The Effect of Sugars and Mannitol upon Coagulation of IggAlbumin. Masters Thesis, Michigan State College. 32 pp. illus. 31. Euler, H. Von and Josephson, K. 1926, Uber Reaktionen Zwischen Zucherarten and Aminen I. Eine Reaktion Zwischen Glucose and Glykokoll. Ztschr. F. Physiol. Chem. 153: 1—9. 32. Fay, A. C. 1934. The Effect of Hypertonie Sugar Solutions on the Thermal Resistance of Bacteria. J0ur. Agr. Res. 48 : 453-68. 33. Fischer, Albert. 1937. The Action of Formaldehyde on Heat-denatured Proteins. Enzymologia, 1: 353—8. (Chem. Abs. 33.• 4682). and McKelvey, J. G. The Effect of Various Methods of Vitamin B and Vitamin G Content Dairy Sci. 17: 455-66. 34._________________ 1936. The Effect of Formaldehyde on Blood Coagulation. Enzymologia, 1: 85-91. (Chan. Abs. 30 : 7701). 35. Flake, J. C,» Jackson, H* C* and Weckel, K. G. 1940. Studies on the Source-Qrigin of Activated Flavor in Milk. Jour. Dairy Sci. 23: 1079-86. - 94 - 36. 37. 38. Flake, J, C., Jackson, H. C. and K. G. Weckel. 194D. Isolation of Substance Responsible for the Activated Flavor in Milk. Jour. Dairy Sci. 23: 1087-95. ________ •• Weckel, K. G* and H. C* Jackson. 1939. Studies on the Activated Flavor of Milk. Sci. 22* 153-61. ___________., Jackson, H. C. and E. G. Weckel. 1938. Mg. On the Activated Flavor of Milk. 21: 145-6. Jour. Dairy Jour. Dairy Sci. 39. Folin, 0. and Marenzi, A# D. 1929. An Improved Colorimetric Method for the Determination of Cystine in proteins* Jour. Biol. Chem. 83: 103-8. 40. Freeman, M. 1930. Action of Dilute Formaldehyde Solution on protein Derivatives. Australian J 0ur. Exptl. Biol. Med. Sci. 7: 117-24. 41. Garrett, 0. F. 194L. Some Factors Affecting the Stability of Certain Milk properties V. Interrelation of Certain Metals and Metallic Ions and the Development of Oxidized Flavor in Milk. Jour. Dairy Sci. 24: 103-9. 42. Galwialo, M. T. and Dobrotworskaja, R. 1929. Der Einfluss des Eiweisses auf die Ctxydatxon und Katalytisehen Eigenschaften Anorganischer Fermente. Biochem. Ztschr. 207: 146—50. 43* Gebhardt, H. T. and Sommer, H. H. 1931. Ohe Solubility of Metals in Milk I. TheSolubility of Copper Under Various Conditions. Jour. Dairy Sci. 14: 416-46. 44. Gortner, R* A. 1938. Outlines of Biochemistry. Second Ed. 1017 pp., Illus. John Wiley and Sons, Inc., New York. 45. Gould, I. A. and Sommer, H. H. 1939. Effect of Heat on Milk with Especial Reference tothe Cooked Flavor. Mich. Agr. Expt. Sta. Tech. Bui. 164: 48 pp. Illus. 46, 1940. ifhe Relationship Between the Cooked Flavor and Peroxidase Reaction in Milk, Skimmilk and Cream. Jour. Dairy Sci. 23: 37-46. - 47. Gould, 95 - I. a * 1940* Protective Influence of a Glutathione on CopperInduced Oxidation of Ascorbic Acid in Milk* J0ur. Dairy Sci. 23i 991-95. 48*____ __________ 1940. Cooked Flavor in Milk, A Study of Its Cause and Prevention. Lab* Sec. 33rd Ann* Conv. Internat* Milk Dealers, Atlantic City, N. J. pp* 553-64* 4 9 . ____________ 1940. 50* 51. A Study of the Reducing System of Milk* Sci. 23: 977-84. Jour* Dairy ___________ 1940. 3he Disappearance of Added Glutathione in Milk. Jour* Dairy Sci. 23 ClO): 985-89. ___ 1940* Uipolysis in Raw Milk, Influence of Homogenization Temperature. Indus, and Engin. Chem. 32: 876. 52. Greenstein, J* P* 1940. Sulfhydryl Groups of Serum Albumin, Serum, and Milk. Jour* Biol* Chem. 136: 795-96. 53. Grimner, W. * Hurtenacker, C. and R. Berg. 1923. Zur Kenntnis der Serumeiweisskftrper der Milch. Biochem. Zt schr* 137: 465—81. 54. Gull&nd, J. M. and Morris, C. J. O.R. 1934. B-^ydroxyglutamic Acid. Jour. Chem. Soc. Proc. (London) 1644-49. 55. Harris, R. L. and Mat ill, H. A. 1940* Effect of Hot Alcohol on Purified Animal proteins. Jour* Biol. Chem. 132: 477-85. 56. Hattori, K. 1925. Ueber die Hullensubstanz der Milchfettkugelchen. pharm. Soc. Japan Ho. 516: 123—70. (Chem* Abs. 19: 2380). 57. Henkel, T. 1906. The Coagulation of Milk with Alcohol* 3: 387-405 (Chem. Abs. 1: 2915). 58. 59. Milchw* Zentol. Hensl^» R« R* 1923. Changes in the Proteins and in the Gelatification of Formalized Blood Serum. Jour. Biol. Chem. 57: 139-5. ___________ 1924* Obser vat ions on the Mechanism of the Reaction Between Formaldehyde and Serum Proteins. Jour. Agr. Res* 29: 471-82 - 96 - 60* Hermann, E. and Oppenheiraer, S’, 1928. Gleichgewichte Zwischen Albumin and Metallsalzen (Silbernitrat, Eisen (3) Chlorid, Goldchlorid) Biochem. Ztschr. 199: 468-97, 61. Holliday, N* 1932. Effect of Heat at Varying Concentrations of Hydrogen Ion on Vitamin B (B^) in protein Milk, Jour, Biol. Chem, 98: 707-17. 62. Hopkins, F. G. 1929. On Gluta,thione: 84: 269-320. A Heinvestigation. Jour, Biol, Chem. 63. _____________ 1930. Denaturation of Protein by Urea and R elated Substances. Nature, 126: 328-30. 64. Horn, H. J., Jones, D. B. and S. J. Ringel. 1941. Isolation of a New SuliUr-containing AminoAcid (Lanthionine) from Sodium Carbonat e-treated Wool. Biol. Chem. 138: 141-49. Jour* 65. Houston, S, K. and Thompson, S* Y. 1940. The Effect of Commercial Pasteurization and Sterilization on the Vitamin and Riboflavin Content of Milk as Measured by Chemical Methods. Jour. Ifeiry hes. 11: 67—70. 66. Hutchinson, R. C. 1941. The Sulfur Content of Cow*s Milk. Med. Jour. Austral. 28! 229-31. 67. Irwine, 0. R. and Sproule, H. 1940. The Denaturation of the Soluble Proteins of Whey by Heat. Can. Dairy and Ice Cream Jour. 19: 62, 64. 68. Jackson, C. J., H0wat, G. R. and T. P. Hoar. 1936. Decoloration and Corrosion in Canned Cream. Res. 7: 284-90. 69. __________ _________________________ _____ 1937. Discoloration and Corrosion in Canned Cream. Jour. Dairy Res. 8: 324-30. 70. Jackson, C. J. 1936. Notes on the Sulphydryl Compounds of Milk. Res. 7: 29-30. 71. Jour. Dairy II. Jour. Dairy _____________ 1936. Factors in the Reduction of Methylene Blue in Milk. Jour. Dairy Res. 7: 31-40. - 97 72. Jesserer H. and Lieben, F. 1938. Studien zur Biuretreaktion. VI. Untersuchungen uber Eiweiss-Alkali-schwermetallverbindugen. Biochem. Ztschr. 297S 369-78. 73. __ _____________ ________ 1937. S-£Uden zur Biuretreaktion. IV. Eiweissverbindungen der Kupf ers, Nickels und Kobaits. Biochem* Ztschr. 292: 403-18. 74. Jones, D. B., and Gersdorff, C. E. F. 1934. The Effect of Dilute Alkali on tile Qystine Content of pasein. Jour. Biol. Chem. 104: 99-108. 75. Josephson, D. V. and Doan, F. J. 1939. Observations on Cooked Flavor in Milk: Its Source and Significance. Milk Dealer 29: 35-41; Jour. Series 928, Penn. Agr. E^pt. sta. 76. Kass, J. P. and Palmer, L. S. 1940. Browning of Autoclaved Milk. 32: 1360-6. Indus, and Engin. Chem. 77. Xassell, B. and Brand, E. 1938. The Distribution of the Sulfur in Casein, Lactalbumin, Edestin and Papain. Jour. Biol. Chem. 125: 435-43. 78. Keramerer, K. S. and Boutwell, P. W. 1932. Sulflir Content of Foods. Indus, and Engin. Chem. Anal. Ed. 4: 423-25. 79. Kende, S. 1932. Untersuclmngen uber 'olig-talgige' schmirgellige Veranderungen der Milk. Milchw. For ch. 13: 111—143, 80. Kieferle, W. and Gloetzl, J, 1930. "Veran derungen der Milch" in Handbuch der Milchwertschaft. edited by Winkler, Pub* by Springer, Vienna. Vol. 1, Part 1, p. 231* 81. Kinsey, V. E. 1935. The Effect of X-Rays on Glutathione. 110: 551-8. 82. Kirilenko, N. V. 1932* Determination of Labile Sulfur in Gelatin. JQur. Applied Chem. (U.S.S.K.) 5* 1119—26 (Chem. Abs. 25:1272.). 83. Klobusitzky, D. Von 1929. Einfluss der Wasserstoffionenkonzentration auf die Salzflockung der Serumeiseisskorper II. Biochem. Ztschr. 209: 304-11. Jour. Biol. Chem. 98 - 84* Klobusitzky, D. von 1930. EinlUss der Wasserstoffionenkonzentration auf die Salzflockung der Serumeiweisskorper. III. Biochem. Ztschr. 2235 120-9. 85* ___ _____ _______ 1934. Physicochemical Studies on Albumin-Alcohol Mixtures. I* Effect of Ethyl Alcohol on the Heat Coagulation of Serum Proteins. Biochem. Ztschr. 271: 385-94 (Chem. Abs. 29: 2186). 86. Kober, P. A» and Surgiura, K. 1912-13. The Copper Complexes Amino-acids, Peptids and Peptone. Jour. Biol. Chem. 135 1-13. 87. Kometiani, P. A* 1931. Voranderunger einiger Milchbestandteile durch Erhitzen. Milchw. Forsch. 12: 433^454. 88. Konig, J. and Schreiber, W. 1927. Die Fl&chtigen St^fl© der Nahrungsmittel. Untersuch. der Lebensmtl. 53: 1-44. Ztschr. f. 89. Erauss, W. E. # Erb, J. H* and R. 0. Washburn. 1933. Studies on the Nutritive Value of Milk. II. The Effect of Pasteurization on Some of the Nutritive Properties of Milk. Ohio Agri. Expt. Sta. Bui. 518: 33 pp. illus. 90. Lachele, C. E* 1934. Determination of Minute Quantities of Sulfide SulfUr. Indus, and Engin. Chem. Anal. Ed. 6: 200-1. 91. Leeds, A. R. 1891. The Chemical and Physical Changes Attendant upon Sterili­ zation of Milk. Amer. Chem. Soc. Jour. 13 5 34-43. 92. Leighton, A. and Mudge, C. S. 1923. On t h e E n d o t h e r m i c Reaction w h i c h A c c o m p a n i e s t h e Appearance of a Visible Curd in Milk Coagulated by Heat. Jour. Biol. Chem. 56: 53-73. 93. 94. Mancovitz, S* 1938. Ultra-Violet Ray Pasteurization of Milk. 27: 33, 58. Milk Dealer Martini, V* 1931. Photochemical Reduction of Methylene Blue by Milk and Gluthathione Content of Milk, Bull. Soc. Ital. Biol. Sper* 6: 773-5 (Chem* Abs. 26: 1631). - 99 95. 96. Mar ton A* and Reiner, L. 1923. Formaldehyde-Albumin. Kollo id Ztschr. 32 : 273-9. Masters, M. and McCance, R. A* 1939. Sulfur Content of Foods, Biochem. JQur. 33: 1304-12. 97. Masters, M. 1939. The Determination of Sulfhr in Biological Material. Biochem. Jour. 335 1313-24. 98. Matsua, Isamu. 1929. Biochemical Study' of Milk. IV. The Effect of Heating upon the Constituents of Milk.Osaka J0ur. Med. 28: 555-62. (Chem. Abs. 24: 1160). 99. Mattick, E. C. V. and Hallett, H* S. 1929. The Effect of Heat on Milk. 195 452-62. JQur. Agr. Sci. (England) 100. Menefee, S. G. , Overman, 0. R. and P. H. Tracy. 194L. The Effect of Processing on the Nitrogen Distribution in Milk. Jour. Dairy sci. 24: 953-68. 101. Methods of the Chemist of the U. S. Stell Corp. for the Sampling and Analysis of Gases, 1927. Pub. by Carnegie Steel Company, Pittsburg, Penn. Third Ed. p. 36-37. 102. Michaelis, L. 1929. Oxidation-Reduction Systems of Biological Significance. VI. The Mechanism of the Catalytic Effect of Iron on the Oxidation of Qysteine. Jour. Biol* Chem. 84: 777—87. 103. Millenky, A. and Brueckner, H. j. 194L. A Comparative Stucfer of High-Temperature, Short-Time and Holder Pasteurization. Cornell Univ. Agr. Expt. Sta. (Ithaca) Bull. 754: 26 pp. illus. 104. Miller, P. G. and Sommer, H. H. 1940. The Coagulation Temperature of Milk as Affected by pH, Salts, Evaporation and Previous Heat Treatment. Jour. Dairy Sci. 23: 405-421. 105. Mir sky, A. E. and Anson, M. I.. 1934-5. Sulfhdryl and Disulfide Groups of Proteins. I. of Estimation. Jour* Gen. Physiol. 18 : 307—23* 106. Methods __ 1935-6. Sulfhydryl and Disulfide Groups of Proteins. II. The Relation Between Number of SH and S-S Groups and Quantity of Insoluble Protein in Denaturation and in Reversal of Denaturation. Jour. Gen. Physiol. 19 5 42/—38. - 100 107. Mirsky, A. E. and Anson, M. L. 1935-6. The Reducing Groups of Proteins. 19: 451-59. Jour. Gen. Ihysiol. 108. Miscall, J., Cavanaugh, G. w. and P. P. Carodemos. 1929. Copper in Dairy Products and its Solution Under Various Conditions. II. Jour. Dairy Sci. 12: 379-84. 109. Mueller, J. H. 1923. A New Sul fur-Containing Amino Acid Isolated from the Hydrolytic Products of Protein. J0ur. Biol. Chem. 565 155-69. 110* Nichols, J. B., Bailey, E. D., Holm, **. E., Greenbank, G. R. and E. F. Deysher. 1931. The Effect of Preheating on the Dispersity of Calcium Caseinate in Skimmilk. J0ur. Phys. Chem* 35: 1303-7. 111. Niemann, cited by Rettger, L. F. 1901. The Liberation of Volatile Sulfides from Milk on Heating. Amer. Jour. Physiol, 6 5 450-7. 112. Northrop, J. H. and Kunitz, M. 1927-8. Combination of Salts and Proteins. III. The Combination of CuClg, MgCl2» CaClg* A1C1«, LaClg* KC1, AgNo3, and Na£So4 with Gelatin, Jour. Gen. Physiol. 11: 481—93* 113. Oppenheiraer, K* 1901. Uber die Zersetzung des Eiweiss beim Kochen. Wchenschr. 7: 105-6. 114. Osborne, T. B. 1900. Sulphur in protein Bodies. Sta. Ann. Rept* 443-71. Deut. Med. Conn. (New Haven) Agr. Expt. and Leavenworth, 115. 1916* Protein Copper Compounds. 116. 1918. and Wakeman, A. J. The Proteins of Cowfs Milk. C* A* Jonr. Biol* Chem, 28 5 109-23. J0ur. Biol. Chem. 33: 7-17. 117. _______ ______ __ 1918. Some New Constituents of Milk. III. A New Protein, Soluble in Alcohol. Jour. Biol. Chem. 33: 243-51. 118. Painter, E. P* and Franke, F. W. 1936. A Comparison between the Bene diet-Denis and parr Bomb Methods for the Determination of Total Silfhr in plants and Proteins. J q u t . Biol. Chem, 1145 235— 39. - 101 119. Palmer, A. H. 1934. The Preparation of a Crystalline Giobuline from the Albumin Fraction of Cow's Milk. Jour. -Biol, Chem. 104: 369—72* 120. Palmer, L. S. and Samuel son, E. 1924. The Nature of the Substance Adsorbed on the Surface of the Fat Globules in Cow's Milk, Soc. Exp. Biol, and Med. ^roc* 21: 53/1—39 (Cited by Palmer and Wiese). 121. 1935. 122. 123. audWiese,J. F. Substances Adsorbed on the Fat Globules in Cream and Their Relation to Churning. II. The Isolation and Identification of Adsorbed Substances. Jour. Daiiy Sci. 16! 41-57. Prescott, S. C. 1927. The Treatment of Milk by an Electrical Method. Jour. Pub. Health 3! 221-23. fiams^, R. J., Tracy, P. H. and H. A. Ruehe. 1933. Corn Sugar in Sweetened Condensed Skimmilk. Daily Sci. 16: 17-32. Amer. Jour. 124. Rettger, L. F. 1902. The L^beratioh of Volatile Sulphides from Milk on Heating. Amer. Jour. Physiol. 6: 450-7. 125* Revol, L. and paceard, R. 1937. Total Sulfur in Human and Cow's Milk. Soc. Biol. 126: 25—6, Compt. rend. 126. Rice, F. E. end Miscall, J. 1923. Copper in Dairy Products and its Solution under Various Conditions. I. Jour. Dairy Sci. 6* 261—77* 127. Ritter, W. 1939. The Influence of Pasteurization-Temperature on Certain Properties of Milk. Jour. Dairy Sci. 22: 117-8. 128. Rowland, S. J. 1933. Heat Denaturation of Albumin and Globulin in Milk. Jour. Daily Res. 58 46-53. 129._______________ 1937. The Heat Denaturation of Albumin and Globulin in Milk, II. Denaturation and Degradation of Protein at Temperatures of 75 to 120° C. Jour. Dairy Res. 8: 1—5. 130. 1937. The Soluble Protein Fraction of Milk. 8: 6-14. Jour. Dairy Res. - 102 131* Rubner, M. 1895* Notiz uber die Unterscheidnng Gekochter und Ugekochter Milch. % g . Rundschau 5: 1021-2. 132. Rupp, P. 1913, Chemical Changes Produced in Cow*s Milk by Pasteuriza­ tion. U„ S. Dept. Agr. Bur* Animal Ind. Bui. 166: 10 pp. illus. 133. Schiff, H. 1901. Trennung von Amin-und Saurefunc tion in Losungen von Eiweisskorpern. Ann. Chem. 319: 287—92. 134. Schmidt, C. A* 1938. The Chemistry of the Amino Acids and Proteins, pp. 1004, illus. C. C. Thomas, Springfield, 111. and Baltimore, Md. 135. Schorn, H. 1929. Untersuchung von Me tal1sal z—Album ini8sung en Mitt els der Auswaschmethode. Biochem. Ztschr. 199: 459—67. 136. Schreiner cited Dy Utz, K. S. 1887. Ensteht beim Kochen von Milch. 10 c.c.t. (152). Schwefelwasserstoff 137. Sharp, P. P. 1938. The Physical State of Milk Pat in Elation to Behavior of Milk Products* Proc. 30th Ann. Meeting Amer. Butter Inst. Chicago, 111., pages 6-18. 138. Sheppard, S. E* and Hudson, J. H. 1930. Determination of Labile Sulphur in Gelatin and Proteins. Indus, and Engin. Chem. Anal. Ed. 2: 73-5. 139* Sherman, H, C. 1941. Chemistry of Pood and Nutrition. MacMillan Co. , New York. 140. Sixth Ed. 611 pp. Shinohara, K. and Kilpatrick, M. 1934. The Stability of Cystine in Acid Solution. Chem. 105: 245—51* Jour. Biol. 141. Smythe, C. V. and Schmidt, C. L. A. 1930. Studies on the Mode of Combination of Iron with Certain Proteins, Amino Acids and Related Compunnds. Jour. Biol. Chem, 88* 241—69. 142. Sommer, H. H. and Hart, E. B. 1922. The Heat Coagulation of Milk. J0ur. Dairy Sci. 5: 525-543. 103 143. Sorensen, S, p, s. 1907. Ihzymstudien. Biochem. Z. , 7: 45-101. 144. Steffen, F* and Sullmann, H. 1931. ©le Differences in the Sulfur and Phosphorus Contents of Human and Cow*s Milk. Schweiz. Med. Wchnschr., 61; 1114-16. (Chem. Abs. 26: 3559). 145. Supplee, 0. C. and Jensen, C. G* 1940. The Instantaneous Heat '-treatment of Milk. Reprint 14th .Ann* Rept. Hew lork State Assn. Ihiry and Milk Inspectors, pages 123-142. 146. Sure, B. and 0*Kelly* R* E. 1923. The Distribution of Sulfur in Protein-free Milk. Jour. Metabolic Bes. 3! 365—71. 14?. Swanson, A. M. and Sommer, H. H. 1940. Oxidized Flavor in Milk. II. The Relation of OxidationRe&iction Potentials to its Development. Jour. Dairy Sci. 23! 597-614. 148. Teorell, T. 1930. Einwirkung von Alkaholen auf die Hitzegerinnung Acetatgepuffeter EiweisslSsungen. Biochem. Ztschr. 229; 1-15. 149. Tillsman, J. and Sutthoff, W. 1910. praformerte Schwefelsaure in der Milch. Ztschr. f. Untersuch. der Nahr. u. Genussmtl. 20: 49-63. 150* Titus, R* W., Sommer, H. H. and E. B. Hart. 1928. The Nature of the Protein Surrounding the Fat Globule in Milk. Jour. Biol. Chem. 76 I 237-50. 151. Utz, K. S. 1903. Ensteht beim Kochen von Milch Schwefelwasserstoff? Zeitung, (Leipzig) 235 354-5. Milch- 152. Vasilfev, S* 1938. The Coagulating Action of Alcohols on Protein Solutions and the Traube Rule. Acta. Physicochim. U.R. S.S. 9! 942-62. (Chem. Abs. 335 7828). 153. Veen, A* G. van and Hyman, A* J* 1935. Ajenkolic Acid, A New Sulfur-Containing Amino Acid. Rec. Trav. Chim. 54: 493-50. (Chem. Abs. 29: 5816). 154. Viale, G. 1926* Uber die Rednktionsfahigkeit der Milch. III. Vorhandesein der Schwefelgruppe der Milch und die Reduktion der Methylenbluas. Rer. uber die Gesam. Physiol. Ud. Exptl. Pharmakol. 38: 25.(Chem. Abs. 21* 1152). - 104 155. Vickery, H. B. and Gordon, W. G. 1933* Complex Compounds Formed "by Certain Amino Acids in the Presence of Mercuric Chloride and Alkali. J 0ur. Biol. Chem. 103: 543-47. 156. Vickey, H. B. and White, A. 1933. 2he Use of Cysteine Cuprous Mercaptide in the Determination of Qystine. J0ur* Biol. Chem. 99s 701-15. 157. Webb, B. H. 1935. Color Development in Lactose Solutions Hiring Heating with Special Reference to the Color of Evaporated Milk. Jour. Dairy Sci. 18S 81-96. 158.___________ and Holm, G. E. 1930. She Color of Evaporated Milks. Jour. Daily Sci. 13:25-39. 159.__________ _________________ 1932. The Heat Coagulation of Milk. II. The Influence of Various Added Salts upon the Heat Stabilities of Milks of Different Concentrations. Jour. I^iry Sci. 15 : 345-366. 160. Weckel, K. G> a n d J a c k s o n , H. C. 1936. Observations on the Source of Flavor in Milk Exposed for Prolonged Periods of Radiation* Food -“-es. is 419-26. 161* Weise, H* F* and Palmer, L # S. 1934. Substances Adsorbed on the Fat Globules in Cream and Their Relation to Churning. III. Analysis of the Adsorbed Proteins. Jour. 3^,iry Sci. 17: 29-32. 162. Wintersteiner, 0., Williams, B. R* and A» E. Ruehle. 1935. Studies of Crystalline Vitamin B^. II. Elementary Composition and Ultraviolet Adsorption. Jour. Amer. Chem. Soc. 5? I 517-20. 163. Wright, N. C. 1924. The Action of Rennet and of Heat on Milk. 18« 245-51. 164. 165. 166 . Young, E. G. 1922. The Coagulation of Proteins by Sunlight. (London) Proc., Ser. B 93: 235—48. Biochem. JQur. Roy. Soc. Zahnd, H. and Clarke, H. T* 1930. TheSstimation of Sulphur in Organic Compounds. Chem. Soc. Jour. 52: 3275—9. 1933. __________ Labile Sulfur in Proteins. Ambr. Jour. Biol. Chem. 102: 171-86.