THE ASSOCIATION COT AN UDDER BORNE MICROCOCCUS WITH THE OXIDIZED FLAVOR CF MILK By Albert Vernon tawuiMMnb Moore A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Dairy 1948 P roQ uest N um ber: 10008237 All rights reserved IN F O R M A T IO N TO A LL U SER S The quality o f this reproductio n is de p e n d e n t upon the quality o f the copy subm itted. In the unlikely e ven t th a t the au th o r did not send a com plete m anuscript and there are m issing pages, these w ill be noted. Also, if m aterial had to be rem oved, a note w ill indicate the deletion. uest P roQ uest 10008237 Published by P roQ uest LLC (2016). C opyright o f the D issertation is held by the A uthor. A ll rights reserved. T his w ork is protected a g a inst unauthorized copying under T itle 17, United S tates C ode M icroform Edition © ProQ uest LLC. P roQ uest LLC. 789 East E isenhow er P arkw ay P.O. Box 1346 A nn A rbor, Ml 4 8 1 0 6 - 1346 C O N T E N T S Introduction — — - — - — — — — — — — — — — — Historical Review — — - ]_ 2 Metals and Light as Catalysts Feeds, Vitamins and Oxidized Flavor Heat and Deaeration General Udder Flora and Milk Flavor Flavors Caused "by Specific Organisms Lipolytic and Oxidizing Organisms Oxygen Metabolism Bacterial Growth and Oxidation-Peduction Leucocytes Lecithin Orgin Of the Present Study- - - - - - - - - - - - - - - - - - - - - 17 Spontaneously Oxidized Flavor in Aseptically Drawn Fresh Milk Occurrence of Pin Point Colonies in Agar Plates, From Samples of Aseptically Drawn Milk Showing an Oxidized Flavor Summary of Preliminary Observations The Problem Scope of the Investigation Procedure - - - - - Results ----- 20 — - -- -- — - — — - — 21 - - - -- - - 26 Occurrence of Oxidized Flavor in Milk from Individual Cows Collection of Samples for Bacteriological Analysis Standard Bacteria Counts In Milk from Individual Ccrws The Effect of Lecithin Agar on Total Counts and on Pin Point Colony Development 2 1 7 8 7 1 Results, continued Inoculation of Milk and Milk Fractions with, an Organism Similar to Gaffkya Tardissima Heat Resistance of Udder Borne Organisms Use of the Oxidase Indicator P-Phenylenediamine Oxalate Dehydrogenation by Udder Borne Organisms Discussion - - - - --- Summary - - - - - — _ _ _ _ Conclusions - -- - - - ----- £8 - - ------ ------- ----- -- — - — — — ----- Bibliography Appendix — - — — 59 63 6U - — - — — — - — - — - — — — — — — — — — — — — i -x ACKMOWLED GEMEHT S The writer is indebted to Dr* G. M* Trout of the Michigan State College Dairy Department for his counsel in planning the research work done in this thesis, and for his suggestions on the preparation of the manuscript* Appreciation is accorded Dr. W* L. Mallmann of the Michigan State College Department of Bacteriology and Public Health for valuable suggestions on the selection of bacteriological tech­ niques used in this study* Credit is due Dr* C. M* Lyman of the Texas Agricultural Experiment Station for advice concerning the preparation of lecithin-enriched agars* - 1 - INTRODUCTION Studies on the occurrence and prevention or control of oxidized flavor in milk have "been reported for twenty years. The same flavor, though called "by several other names, has "been the "basis of research in the milk "by-products fields for a much longer time. This is especially true of the products "butter and powdered milk. The exact origin of this flavor has never "been determined, though it is now well known that a number of control measures will . prevent it. Oxidized flavor continues to be a major problem in the dairy industry. This is shown by the growing number of markets that re­ port having it and the more intense effort expended in its control. One intriguing observation has been made by many investigators. That is that the occurrence and intensity of the flavor increases, as quality measured in other ways, improves. Consumer education in nutrition and sanitation have led the entire dairy industry toward supplying a better product; therefore more emphasis is constantly being placed on the production, processing and distri­ bution of a more nourishing and a public health-wise safe product. The flavor is definitely associated with the rich portion of the milk and is further characteristic of milk that has been handled so as to make it bacteriologically good. It is unfortunate that a food industry constantly progressing toward the goal of perfection in product quality, should be simul­ taneously plagued by a serious fault. final basis of judgment. Flavor is the consumer's - 2 - HISTORICAL REVIEW Metals and Light as Catalysts Feeds, Vitamins and Oxidized Flavor Heat and Deaeration General TJdder Flora and Milk Flavor Flavors Caused by Specific Organisms Lipolytic and Oxidizing Organisms Oxygen Metaholism Bacterial Growth and Oxidation-Reduction Leucocytes Lecithin METALS AND LIGHT AS CATALYSTS Metals, primarily copper and iron, have "been given much attention as aids to oxidized flavor development. Guthrie and Brueckner (31) contributed the significant fact that the mechanism of oxidation catalysis by oleinase is entirely different from that of copper, since the former does not require high oxidationreduction potentials and the latter does. Of the many references to the catalytic effects of metals the more pertinent works of Kende (37) (38) (40) appear to be a logical summary. He showed that copper, either that contained in the udder or outside, was responsible for fat oxidation according to the relation of copper to the oleinase and/or the reducing substances present. Reductase present In the udder or produced by bacteria were factors recipro­ cating with oleinase and the metal. The progress ox* durability of the "oily" or "emery" flavor, its Intensity, the time of develop­ - 3 - ment and the eventual disappearance Here dependent upon and -were measured by this reciprocating action. Feeds may supply substances sufficient to protect the milk from becoming oxidized, or living cultures of reducing bacteria or their metabolic products may be used. Frazier (24) held that neither enzymes nor bacteria were necessary for milk fat oxidation. He kept raw and pasteurized milk samples at about freezing for eight hours in diffused light. Oxidation occurred more rapidly in pasteurized than in raw milk, irrespective of the feeds, cottonseed and linseed cake. He attri­ buted the oxidation to light, though there was not a clear definition of the type of flavor experienced. FEEDS, VITAMINS AND OXIDIZED FLAVOR Anderson, Hardenberg and "Wilson (l) used vitamin A to con­ trol oxidized and rancid flavors. Their claim that 8 pounds of carrots in the daily ration was superior to 500,000 U.S.P. units of vitamin A has been observed also under commercial conditions but not reported experimentally. While Whitnah, Martin and Beck (70) concluded that there was no certain relation between feed and oxidized flavor, they did. observe that all samples that developed the flavor were below the breed average in color of fat intensity (vitamin A) . They also ranked the breeds in order of vitamin C content of the milk— Jersey, Guernsey, Ay shire and - 4 - Holstein and found that the incidence of oxidized flavor from these breeds was in reverse order. Other evidence that vitamin A does inhibit or prevent oxidized flavor is presented by Brown, Thurston and Dustman (7) Kende (38) Garrett, Bender and Tucker (27). Prewitt and Parfitt (53) found both soybean, oil and un­ processed beans to be protective against the development of oxidized flavor, even when copper had been added to the milk. None of the samples to which copper had not been added showed any oxidized flavor after pasteurization and holding at 4.4° C. for 72 hours. Supplee and Beilis (60) found no difference be­ tween the copper content of cow’s milk when pasture and stall feeding were compared. They suggested (this was in 1922) that copper "may prove to be significant in connection with the high susceptibility of the antiscorbutic vitamin to oxidation." Since they found cow’s milk to contain from 0.2 to 0.8 milligrams of copper per liter, average 0.52 milligrams, there is a basis for the statement made in 1922 by Kende (40) regarding "inner and outer" metal contamination. HEAT AND DEAERATION Dahle (15) collected milk samples directly from cows into amber glass bottles and studied the influence of pasteurizing on oxidized flavor development. When heated at 71.1° C. for 30 minutes, cooled and stored at 4.4° C. for 3 days the flavor was decreased or - prevented. 5 - There -was no evidence of* oxidized flavor in the skim- milk so treated and the flavor was more pronounced in whole milk than in cream. Foremilk had less noticeable flavor than middle- drawn or last-drawn milk. Cows that produced milk with the flavor in winter did not produce it in summer. There was no statement made in this report indicating that the raw samples were tasted. McFarland and Burgwald (44) prevented oxidized flavor develop­ ment in storage cream by heating at 77.7° C. for 5 minutes or by homogenizing. The cooked flavor resulting from the high heat treatment was not noticeable after one month in storage. Kende (40) and G-ould and Sommer (30) explained that oxidized flavor is prevented by heating, owing to the liberation at temperatures in the range of 76° to 78° C. of reducing substances. Gould and Sommer (30) specifically named hydrogen sulphide as the protec­ tive reducing substance and showed also that the oxidation reduc­ tion potential is lowered at these high temperatures. Oleinase, named by Kende (40) as the agent responsible for the oxidized flavor, was apparently prevented. Leeder and Herrled (43) evacuated and stored.under a vacuum of 23 to 25 inches, milk susceptible to becoming oxidized. Bacteria decreased the oxygen content of milks held at atmospheric pressure and prevented or inhibited flavor development; under vacuum the milk did not be­ come oxidized and bacteria were not considered important as oxygen consumers in this milk. This work showed no relation between bacteria count and intensity of oxidized flavor or oxygen content. Reed (54) reports that evacuation of milk, when preceded by fortification with ascorbic acid, pasteurizing and homogenizing, prevents or retards oxidized flavor. Low storage temperatures delay the rate of change from ascorbic to dehydroascorbic acid; oxidized flavor is delayed simultaneously. GENERAL 'UDDER FLORA AND MILK FLAVOR Little has been done to show that organisms isolated from the udder have any relation on milk flavors. The bacteria counts from healthy udders determined by established methods are usually low; fLirther, it is usually believed that these organisms have no appre­ ciable effect on the properties of fresh milk, and that they do not thrive outside the animal body. However, Evans (20) in 1917 re­ ported a study made on 192 samples from 161 cows in 5 herds in which the organism Bacillus abortus variety lipolyticus was fre­ quently found. It decomposed fat and "imparted undesirable flavors and odors to cream kept under conditions to which cream is fre­ quently subjected." Since the Bacillus abortus strain was isolated from all cows at all 5 dairies it was assumed that it could, be isolated from all mixed milk even though the total count were low. In 1917 oxidized flavor in milk was not reported on as such. The "undesirable” flavor reported by Evans was prob­ ably rancid. In further work on lipolytic rod shaped bacteria, Evans (21) found as high as 112,000 per milliliter of the Bacillus -7 ~ abortus strain in the milk of an apparently normal cow. Again the "disagreeable" flavor was noted but it was observed that the organism was destroyed in 25 minutes at 51.7° C. or in 25 seconds at 62.6° C. Tracy, Ramsey and Reube (65) concluded that tallowy flavors could not always be produced to the same degree by add­ ing equal amounts of copper salt to aseptically drawn samples nor to mixed herd samples. They felt that a lack of bacterial metabolism accounted for oxidized flavor in low count raw or pasteurized milk and that leucocytes as well as bacteria func­ tioned as reducing bodies. Udder tissue did not function as a re due ing sub stanc e . FLAVORS CAUSED BY SPECIFIC ORGANISMS When Ruehle (56) inoculated Into intermittently heated sterile milk, single species of organisms, and made parallel plantings of each specie in association with Streptococcus lactis, a wide variety of flavors developed at room temperature in 48 hours. While the organisms were originally isolated from butter some of them could have been of udder origin. This Is not known, partially because the description of the organisms was limited to morphological characteristics. Oxidized flavors were not mentioned (the work was reported in 1920) but "astrin­ gent or metallic," "oily" and "flat" were used to describe some of the flavors that resulted. Associative action was apparently necessary to bring out flavors indicating that bacterial flavors - are chemically complex. 8 - The numbers of* organisms required to effect the certain flavors were not indicated. Brown, Smith and Buehle (5) found that tallowy flavors developed more frequently in raw cream butters than in pasteurized cream butters, but in a study of split churnings pastuerized cream butters developed faster in which cream was inoculated with fishy butter and treated with wash waters of varying acidities. Some of the organisms isolated from these butters were able to grow in agar containing 16 percent salt. LIPOLYTIC AMD OXLDIZING ORGANISMS Jensen and Grettie (36) found in work carried out on butterfat, shortening and lard that these fats serve as substrates for certain strains of organisms, which are responsible both for enzymes that liberate free fatty acids and oxidation products. They called the property "oxidative rancidity" and attribute it In part to the effect of light. Their analysis of the problem of fat hydrolysis or oxidation helps to eixylain the confusion that exists. They found that moisture-free fats did not support the growth of organisms but that 0.3 percent or more of moisture In an animal fat promoted growth. Stark and Scheib (59) named several species of micrococci as being prominent fat splitters. These were isolated from butter. In view of the explanation by Jensen and Grettie (36) that organisms may be both oxidative and lipolytic, and that Stark and Schieb found micrococci capable of growth at temperatures ranging from 5° to 45° C., it seems possible - 9 - that th© oxidative processes associated with oxidized flavor could originate in the normal cow's udder. A quantitative estimation of the relative oxidizing capacities or organisms has heen made hy G-ordon and McLeod (29). heated "blood agar plates with pathogens. They streaked After colony development, oxidizers were detected hy flooding the plates with p-phenylenediamine. When this was oxidized to indophenol, colors of the colonies varying from pink through red to "black resulted. Mixed cultures gave vivid color contrasts. The method had "been used formerly "by others to distinguish "between myeloid and lymphocytic leucoytes, on the assumption that the reaction in leucoytes was due to an oxidative ferment. When the anthrax bacillus was sus­ pended in the reagent in a hanging drop, blue granules developed in the cell. Ellingworth, McLeod and Gordon (18) could not successfully incorporate p-phenylenediamine into the agar medium. It detected more oxidizing organisms than did other diamines. These authors found also that when the colony turned black, indi­ cating the maximum detectable oxidizing capacity of the enzyme elaborated by the organism, the organism was then dead; no further reduction being possible. Fabian and Trout (22) de­ veloped a technique for the Isolation of lipolytic organisms. Ifile blue sulphate and sterile cream were incorporated into tryptose broth Just prior to pouring plates, in a study of frozen cream. Tryptose gave higher counts than standard agar - 10 - and lipolytic organisms were easily 6_etected "by the hlue color of the fatty acids released. Caste11 and Garrard (10) grouped several important genera according to their oxidizing abilities, us­ ing the technique of Gordon (29) except that they used p-aminodimethylaniline hydrochloride as the oxidation indicator. monas and Achromohacter were the most strongly oxidative. Pseudo­ Alkli- genes and Brucella followed, though still strongly positive. Other gram negative genera varied from weakly positive to nega­ tive. The "bacilli were variable and the' cocci and one anaerobe (Clostridium butyricum) were negative. All strong oxidizers were gram, negative; those classed as not being strong oxidizers were positive. Carpenter, Suhrland and Morrison (9) claimed to have improved the technique of Castell and Garrard (10) by employing the oxalate rather than the hydrochloride salt of p-aminodimethylaniline. The oxalate was more stable in crystalline than in liquid form. Xt showed less precipitate as it contacted the colonies and gave more distinct colors, ranging from pink to black. OXYGEN METABOLISM Keyes and Gillespie (41) made a study of the oxygen metabolism by Escherichia coli and Clostridium welehii. Wide differences were noted; the respiration by-products C0^ and Hg varied much more for Escherichia coli (facultative) than for Clostridium welchii (strict anaerobe) . That organisms general3.y - 11 - have a wide range of metabolic activity has been shown also by Mimdt and Fabian (47) who were studying the oxidation of corn oil. Some of their test organisms were udder-borne micrococci. The oat derivative Avenex, commonly used as an anti-oxidant in dairy products, was utilized by four species and did not prevent bacterial respiration in the presence of corn oil. Nunheimer and Fabian (48) studied the respiratory properties of the micrococci. Four of the 5 species studied are borne by the udder; these were found to be oxidizers of several sugars, alcohols and amino acids. Micrococcus aurantiacus was most active as a dehydrogenator on lactose and d-galactose. It also was more a,ctive on glycerol than Micrococcus flavus, Micrococcus luteus, Micrococcus cixmebareus and Micrococcus freimdenreichii in the order of 367-114-56-32-24 respectively, using the dehydrogenation value on glucose as 100. Substrates of particular interest in a cow's udder were not employed. Mattick (46) systematically eliminated bacteria as a possible cause of "oily11 flavors, arguing that by their own metabolism of molecular oxygen or by the production of acidity that would carry the electric potential outside the limiting pH, they retard oxidation; in so doing they are indirect­ ly invovlved in preventing ’’oily" flavor. BACTERIAL GROWTH AM) OXIDATION-REDUCTION While there is a difference of opinion regarding the ability of organisms to affect the oxidation-reduction potential of milk, most of the evidence points toward their being able to lower it. - 12 - Coulter (12) showed that a medium of deaerated sterile culture "bouillon drifted from an initial plus 0.250 volts to a minimum of minus 0.060, a marked increase in reducing intensity• the original value was restored upon the re—admission of air, indi­ cating that organisms were not necessary for the change. Using washed yeast suspensions as well as growing cultures of Pseudo­ monas fluorescens, Bacillus subtilis, Streptococcus lactis, Escherichia coli and Proteus vulgaris in the presence of methy­ lene "blue, succinate and glutathione, it was found by Cannon, Cohen and Clark (8) that progressively more negative potentials develop. Xn some of the earliest studies of the reducing capa­ city of organisms, Potter (52) measured the E.M.P. produced when platinum electrodes were immersed in two portions of a culture medium separated by a porous membrane, one half cell being in­ oculated and the other sterile. With yeast and Escherichia coli the inoculated portions were always more reducing than the uninoculated. Xn a similar study G-illespie (28) compared aerobes to anaerobes, particularly soil types, and found that while the former showed progressively increasing reduction potentials with lapse of time, the latter developed only to a uniform constant. That the rate of reduction of one organism is affected by another when associative action is involved was shown by Frazier and Whittier (26) . Streptococcus lactis for example, was re­ strained from making a rapid drop in Eh in the presence of gas forming fecal types of organisms. In another study the same - 13 - authors (25) concluded that of a wide variety of natural milk contaminate, Streptococcus mastitidis was the only one unable to produce a negative potential. It did not show a reduction in three days. Comparing pH and Eh measurements In this study it was found that between pH4 and 7 the relation to Eh is practically a straight line function, Eh Increasing about 0.06 volts for each one point decrease in pH, between 25° and 40° C . If oxygen is diffused into milk at a rate slower than its consumption by a single specie of organism, that organism, according to Hastings, Davenport and Wright (32), would appear to be a non-reducer. The authors felt, however, that in a mixed culture definitely showing reduction, the slow consumer would exert its effect, though slight. "There is no relationship between oxidized flavors and oxi­ dation reduction potentials in the milk from individual cows. The decreased susceptibility of summer milk to oxidized flavor does not appear to be due to bacteria." This statement by G-uthrie and Brueckner (31) was made in view of studies that showed milk with high potentials having oxidized flavor. Summer milk resisted oxidized flavor development even In the presence of a higher potential and its cause was attributed to oleinase not of bacterial origin. nucocYTEs All milk contains leucocytes (16). Normally a cow has about - 14 - 100,000 leucocytes per cubic centimeter (19). Horrall (34) examined mastitic udders that showed from 1,200,000 to 9,390,000 leucocytes per cubic centimeter. That high leucocyte counts regularly accompany streptococci infection and high pH, was shown by Plastridge, Anderson and Williams (51) . The reason for this according to Baker and Breed (3) is that the infection causes more unmodified blood to enter the udder; blood is alka­ line. Fewer leucocytes are held back by filtration. genized milk, leucocytes congregate in the cream (2). In unhomo­ Trout, Scheid, Peters and Mallmann (67) also observed that leucocytes as well as bacteria accompany fat to the cream layer in unhomogenized milk. In homogenized milk bacteria and fat remained distributed but leucocytes settled to the lower layers in great numbers. Yeasts also settle. Trout and Halloran (66) observed that an Increased gravitation of leucocytes occurs In unhomogenized milk that has been heated as high 70° C. This was ascribed to the re- tarted cream layer formation or to the possible change in charge on the fat globules or leucocytes. Skar (57) has shown that leucocytes may play an important part in bringing about a red^lcIng potential. Peters and Trout (49) (50) studying further the attrac­ tion, which they believed to be mutual, between fat and leucocytes, discovered that adding 7.5 grams per pint of washed leucocytes would deepen the cream layer and that 15 grams actually carried some of the fat to the bottom of the bottle. This was well - 15 - illustrated by coloring the fat -with Sudan III. This attrac­ tion was greatest at pH 4.3, the iso-electric point of the fatseruih. interface. Tarassuk and Palmer (62) did not specifically attribute the diminished tendency toward oxidized flavor to leucocyte removal in super centrifuged milk, as they were study­ ing phospholipids in remade milk samples. The work of Peters and Trout (49) would indicate, however, that leucocytes may have been involved. HECITHUT In 1940 Brown and Thurston (6) presented an exhaustive literature review on oxidized flavor and stated "the trend of the literature at the present time seems to point to the phos­ pholipid fraction as the source of oxidized flavors in milk and cream.” Dahle and Palmer (15) and Swanson and Sommer (61) both reported a reduction of the iodine number of the phospholipid fractions of milk exhibiting oxidized flavor; the latter authors further found no significant difference in the iodine numbers on butterfat from normal and oxidized flavored milks. Whitnah, Martin and Beck (70) concluded that milk, low in lecithin, developed oxidized flavor as readily as that high in lecithin. In order to disengage the phospholipid membrane, sometimes called the hull, from the globule of fat, Thurston, Brown and Dustman (64) employed various homogenizing pressures, violent and pro­ longed (two and one-half hours) agitation and alternate freezing and thawing on milk. All three practices reduced or eliminated - 16 oxidized flavor development. - While still engaged to the fat globules, the phospholipid membrane regularly caused oxidized flavor. This adds weight to another study by the same authors (63) showing that the development of oxidized flavor was a pro­ gressive reaction, initiated in the phospholipid fraction and continuing or completed on the butterfat. Whereas in prepared synthetic milks containing only pure butterfat, tallowy flavor was noticed, those containing lecithin had oxidized flavor. The work of Tarassuk and Palmer (62) is confirmatory. Roland and Trebler (55) and Holm, Wright and Deysher (33) partially confirm the work of Thurston, Brown and Dustman (64) by adding that in cream separation there is a change in the distribution of lecithin and related substances between the fat and aqueous phases of the milk system and that this may be responsible for the decreased sensitivity of milk to oxidized flavor. Kurtz and Jamieson (42) demonstrated that the lecithin-cephalin fraction of milk phospholipid is highly oxidizable. After separating true butterfat from phospholipid in 300 pounds of spray process sweet cream buttermilk by repeated acetone precipitations and ether ex­ tractions, they analyzed the fatty acid composition of each. The lecithin-cephalin fraction contained none of the lower fatty acids common to true neutral butterfat, but did contain, along with myristic, stearic and arachidic acids, the highly unsaturated dicostretinoic acid in the amount of 6.3 percent and 70.6 percent - of* oleic acid. 17 - Palmitic acid was not present. . According to Fetzer (23) milk from mastitic cows contains less lecithin than that from normal cows. opposite. Horrall (34) reports the However Dahlberg, Kucera and Eening (13) grouped mastitis into latent and active cases and found that in milk from the former there was no appreciable variation from normal values. In the latter, considerable variations were noted. ORIGIN CP THE PRESENT STUDY SPONTANEOUS OXIDIZED FLAVOR IN ASEPTICALLY DRAWN FRESH MILK During the fall and winter of 1938-39 a study was undertaken to determine if the salty flavored milk from individual quarters of certain cows could be more specifically expressed by chemical analysis than by taste. Individual quarter samples were collect­ ed aseptically from 63 cows, milked directly into sterile glass containers. The tasting and salt testing, of the warm milk follow­ ed immediately. On several occasions 2 or more of the 6 men who were judging flavors, noted that oxidized flavor was present in the milk of some quarters. Though It was known that milk from certain cows was susceptible to becoming oxidized by copper or or Iron contamination and by exposure to light, It was not known at the time that strictly fresh uncoiled raw milk would have the flavor. OCCURRENCE OF PIN POINT COLONIES IN AGAR PLATES, FROM SAMPLES OF ASEPTICALLY DRAWN MILK SHOWING AN OXIDIZED FLAVOR After observing oxidized flavor under the conditions set forth in the salt flavor experiment, it was thmight desirable to - 18 - study flavors of similar samples of milk before and after pasteur­ ization, and to note also the relationship, if any, between the bacterial counts of the raw and pasteurized samples. in the herd were known to have mastitis. Some cows In fact the salt flavor study had been undertaken partially to determine if a flavor test for salt could be used by the dairymen as an index to infections in the herd. Mastitis cows were excluded In the further study of the oxidized flavor problem and several series of platings were made on raw and pasteurized aseptically drawn samples from non-infected cows. Following the customary 48 hour incubation period for plate counts, one set of plates was permitted to stand at room temperature for one week. Pin point colonies had developed after this period of time In the plates of the raw and pasteurized samples from one cow whose milk had had an oxidized flavor both before and after pasteurization, and In the plates on pasteurized milk only from a cow whose raw milk was normal In flavor but whose pasteurized milk was oxidized. Pin points were not detected In any other plates. Milk from one of the College herds was pasteurized and bottled separately. It was In this herd that the salt flavor study shewed 10 of 63 cows giving milk with an oxidized flavor Immediately after milking. Inspection at the Creamery receiving room never showed oxidized flavor in the mixed raw milk but It was uniformly present after pasteurization and after storage at 19 - 4.4° C. for 24 to 48 hours. - For several years this situation had been casually observed to exist during November and continuing through March. SUMMARY OF PRELIMINARY OBSERVATIONS OXIDIZED FLAVOR IN ASEPTICALLY DRAWN FRESH RAW MILK Cows Quarters Examined Oxidized Cows 252 10 63 Flavor In Quarters Agreement Among Six Judges on Flavor Identity 18 2 or more PIN POINT COLONIES IN ASEPTICALLY DRAWN MILK Breed Jersey Oxidized Raw Past. + + Jersey - Jersey - + Holstein Holstein - Holstein __ _ Standard Count 48 Hours Raw Past. 320 210 280 200 1230 130 570 300 540 110 450 200 Showing Pin Points In both raw and pasteurized In pasteurized only THE PROBLEM A literature survey indicates that other workers do not attri­ bute oxidized flavor to bacterial activity, but rather in most cases to a lack of bacterial activity. The general conclusion of others is that bacteria are unimportant because the total counts in affected samples are uniformly low. When based on conventional standard methods of bacteria enumeration and identification this Is - apparently true. 20 - In 1939 a new bacteriological medium for the enumeration of bacteria in milk and dairy products became offi­ cial. This new medium, Tryptone glucose skimmilk extract agar, is responsible for more and larger colonies than is the previously official Standard Nutrient Agar. Even when adopted, however, it was admittedly not a medium adapted to the best development of all the possible types of organisms that might be found in milk and milk products. It has been shown in a number of Instances that the intensity of oxidized flavor increases during the normal storage period of market milk; that is, from one day to 4 or 5 days. Observation of pin point colonies in oxidized flavored milk and the intensifying of the flavor after pasteurization, together with a lack of pertinent data on the bacteriological phases of the problem seemed to justify further Investigation in this direction. SCOPE CO? THE INVESTIGATION The present study was designed to determine: 1. The numbers and types of organisms found In udders of cows whose milk becomes oxidized without the catalyzing effects of metals outside the udder, or by exposure to light. 2. If modifying standard bacteriological practices would result in better growth of pin point ^colonies. - 3. 21 - If oxidized flavor can be produced in milk other­ wise free "by inoculating Into normal and multipleclarified samples, certain udder borne organisms.r 4. The efficiency of the oxidase indicator p-phenylenediamine oxalate in classifying organisms associated with oxidized flavor. 5. Dehydrogenase studies of the udder flora. PROCEDURE The cows used in these studies were In the two herds main­ tained at the Agricultural and Mechanical College of Texas. In addition to their milk, mixed patron milk at the College Creamery was used in the work on clarification. between October 1946 and July 1948. Most of the work was done Some of the determinations on the heat resistance of udder borne organisms were made In the labora­ tories of the Department of Bacteriology and Public Health at Michigan State College during the Spring quarter of 1946. Samples collected from individual mastitis-free cows were taken after the hind quarters had been brushed and udders washed with cloths soaked In 250 ppm. chlorine solution. The ends of the teats were given a thorough washing with a separate chlorine soaked cloth. The assisting milker dipped his hands Into a 250 ppm. chlorine solution and allowed them to air dry before drawing the milk. The fourth stream drawn was placed directly Into 9 milli­ liters of sterile tryptose broth. The tubes were calibrated so that a near 1-10 dilution could be made. There was no practical - 22 difficulty because of foam. water away from light. - Tubes were placed instantly into ice Never more than 6 cows were sampled at one milking, and plating or other culture settings were completed with­ in two hours of railing time. Larger samples for laboratory pasteur­ ization; clarification and flavor studies were taken with the same precautions. Sterile one-half pint milk bottles and Erlenmeyer flasks were used to collect them. Laboratory pasteurization was done in the same containers in which the samples were collected. A Precision Scientific Company water bath; equipi:>ed with a motor driven agitator; was used. Plating and counting techniques were those recommended by the American Public Health Association in "Standard Methods For The Examination Of Dairy Products;" 1944 (58). Culture media used were those manufactured by the Digestive Ferments Company; Detroit; Michigan. The special media not available from this company were compounded according to their general directions. A Beckman potentiometer was used to adjust all media to pH 6.8. An udder unfusion broth containing no other nutrient; was pre­ pared by heating six pounds of trimmed udder tissue in distilled water. The filtered broth was brought to three liters volume. The udder used was from a Jersey cow that had passed two lacta­ tion periods and had shown no evidence of mastitis. Half of the broth was converted to 1.5 percent agar. Separator slime agar was prepared as a possible source of phospholipid nutrient. attempt was made to secure pure phospholipid nor to have the No - 23 - medium completely devoid of other milk constituents except free fat; free fat was removed by alternately washing, agitating and centrifuging hot dilute slime* Agitation was done in a Waring Blend or and centrifuging in an International centrifuge operated 1$ minutes at [+000 rpm* After repeating this sequence of treat­ ments four times, a 10 ml. sample of the diluted slime, taken from the top of a centrifuge tube, showed less than 0.0001 grams of fat or less than 0.001 percent, by the Mojonnier fat extraction tech­ nique. The resulting liquid, slightly turbid, was used as the sole source of nutrient and was converted to l.f> percent agar. One-tenth percent lecithin agar was prepared from soybean lecithin, using tryptose agar (Difco) as a base. The lecithin was first emulsified in one-fourth its weight of Tween 80, the oleic acid ester of a sorbitan derivative, made by the Atlas Powder Company, Wilmington, Delaware. This produced an agar hav­ ing a slight turbidity, but without a troublesome precipitate after aut oclavi ng * Oxygen demand of organisms was determined by the usual stab and slant techniques, together with parallel incubation in regular incubators and in carbon dioxide chambers. Preliminary tests show­ ed no difference in effect on growth between a 10 percent and a 2$ percent carbon dioxide; 10 percent was used thereafter. It was apparent after a few trials that with the organisms being studied and in consideration of the work of Nunheimer and Fabian (1+8) on the micrococci,. there was no good reason to determine the - 24 - need of free versus intramolecular oxygen. Manometric studies •were therefore discontinued, "but a series of Thunberg dehydro­ genation tests were run according to tie modification by Umbreit, Burris and Stauffer (69). Flavor determinations were made ordinarily hy three ex­ perienced judges. On occasions four or five participated. There was no discussion among the judges until all had made decisions and sample identity was never disclosed until all had finished. The presence and relative intensity or the absence of oxidized flavor was noted, and inidcated by plus or minus signs, according to the scheme of Trout and Sharp (68). By this, +++ indicated a strong to very strong oxidized flavor; -H-, distinct to pronounced; +, slight; ?, doubtful and no oxidized flavor. On larger batches of milk, homogenizing and clarifying were done in the Creamery, using a Manton Gaulin two-stage, 200 gallons an hour homogenizer and a DeLaval clarifier of the same capacity. The pasteurizing vat, all connections, all milk contact parts of the machines and the tubular cooler were of stainless steel. Lab­ oratory clarifying was carred out with the International centri­ fuge on milk pasteurized and homogenized in the Creamery. Organ­ isms inoculated into the homogenized and/or clarified milks were grown in flat bottles and harvested in distilled water. A saline wash was avoided to prevent confusing flavors caused by the organ­ isms; but to avoid the toxic effects of distilled water alone, the organisms were collected and placed into the milks as rapidly as possible. By handling a small number of samples it was uniformly - 25 - possible to limit the exposure of organisms to distilled water to less than ten minutes. Control milk samples, uninoculated, Indicated that this method was satisfactory. - 26 - R E S U L T S OCCURRENCE OF OXIDIZED FLAVOR IN MILK FROM INDIVIDUAL COWS To determine if the oxidized flavor of fresh milk from indi­ vidual cows Is a consistent property throughout a lactation period 6 cows were selected for study. Two of these (1039 and 1264) had previously shown it in their milk at once after milking, two (975 and 1083) had not shown it until the milk was held at 4.4° C. for 24 to 48 hours, and the milk of two other (1113 and 1131) did not show It at the end of one week, similarly held. The results shown in table 1 indicate that pasture feeding, previoiisly reported by Brown, Thurston and Dustman (7) Kende (39) and Whitnah, Martin and Beck (70),had an influence. However, the individuality of the cow is a prominent factor. In this connection it should he noted that these 6 cows were selected from a herd then numbering about 95, fewer than 10 percent of which were having oxidized flavored milk before pasteurization. The period of lactation exerts some influence; there were 13 negative judgments during the first half of the study and 9 during the second half. ported no Influence. Dahle (14) re­ - Table. 1. 27 - The Occurrence of Oxidized Flavor in Milk at Different __________ Stages of the Lactation Period._____________ __ _____ Fresh Paw Milk Tested Cow Apr. Oct. Dec. Feb. June Aug No. Breed 26 Freshened 28 18 21 12 23 Holstein April 7 - - - - + 1039 Jersey April 10 + + - - + 1083 Holstein April 9 - + - + 1113 Holstein March 26 - - - - - 1131 Jersey March 20 - - - - - 1264 Holstein March 8 - + -H- 975 ++ 4+ No sample Dry Preparatory to bacteriological studies of individual udders it was deemed advisable to determine the flavors of milk from the same 6 cows before and after pasteurization. The same periods were used as in Table 1 except that no tests were made in February. The test­ ing days were uniformly within one week of those reported in Table 1. Samples were collected directly into glass containers, were cooled at once to 15.5° C. and were protected from light. Pasteurization by the holding method (61.8° C. for 30 minutes) was begun within an hour of milking time. an ice bath. Prompt cooling at 1.7° to 4.4° C. followed in The results are shown in Tables 2a, 2b and 2c. 28 - ■s - t t t t t I I t t t l t t i l + + t s rH CD -P P •i —I d P i—1 P © rP -P p * o d P t> C D Ph •rH d d •H M m rP Poi © CD O O P 3 -H <-D c£> I i i I I I I i O’ I pq ■>* ra + + + + + i t + + + % + t + p C Xi I-1 - C75 g < + pq + O (X I O pq O 'j to I—1 o rH e> •rH o p p pq * I $ $ + + + + t $ * * t If t ex + co P Pt -p tn < O }> P d o -P p Ph 9 § pi < © P B © CQ DQ P •H -P pq CD + P A -p -P EH -p < c j > CQ CD CQ CQ p © CQ P 's P d P 's P d > s P d > s p d P s P d rQ o i— 1 C X I to d L O * - 23 ~ t nd O •H Pi i —cdI Ph nd + 0 H1 nd P? Cd iH + Pi P! O CD T d •i-l o ' DQ Ph Ph t t t % "rH CD IN! •H + ■f 1-1 'S' + -p o Ph Td •-i pq I ! I 1 I I o I I I I pq 1 t I 1 CQ Fh CO o I—1 pq -P •n 0 0 LO P CO Ph 0 P_iJ J hs 6 o 0 I I -p to l~) CO t" P £ cd CJ> £ cd P POh 0 CD 0 O P i O £ H1* o tt * 8 -d1 o 0 o P- -P cd £5 Td o' m /—I §H CO p o Pi o o CO ) * CO P O o I—1 o 0 Pi CO P CO p + + + + t 1 I + + + < T0d £ i —I 0 o •P cd p CO I —0I 0 o 8 r O cd Ed jj ■=ti CO 0 P 1— 1 I-'S cd P CV3 0 Pa cd P to CQ CQ r"a 0 P p & P LO !> £-1 CD CO o r— c - • PR -P 1 1 30 - + + + + + + m P © cd Ph o •rH -p cd o -P A CD ■S’ H n rM rH •rH Pi •H nd rH O w 0 © pq + pq > PR l 1 pq 1 1 t J 1 pq 1 1 pR t 1 pq 1 H © + + 03 & cd Ph c~t Q-. I I 1 p © -p . pR -P nd © to •H I 03 p © fp •H 0 M -p 0 cd Ph 0 0 £ © Ph 92 g 0 rH © IN) •rH u p © -p oq cd Ph nd s § Ph PS •H Ph O i— 1 CO 1— 1 1— 1 0 © pq -p O O i>j> r—1 0 fcd •H n zJ •rH £ ■s © o p 0 •1— ! > rd A O O rM rd *—'i © rP EH © * 03 * l 1 1 1 pR 1 1 I pq 1 i PR 1 1 I 03 cd Ph 92 Pi P t>- d) O rP £3; © © 03 te O O CT> OS rH pq pq 1 * PR -p CQ cd Ph i 1 1 1 1 1 o CV! I CV! © pi 0 O p 0

pq s Ph © *3 o o c i > © p Ph Pi Ph I 03 cd *N Ph -P pq pR ra H gD Pr P **\ to cd 1— ! Ph !-1 rH pq m }> I —cdI ^4 • pR -P 0 -P © rH •H PR P Ph © O <3 m cd Pi 1—1 r'S cd Pi CV) 03 P's © Pi to CQ co cd Pi © Pi LO Ph © © c * o oo -P Ph LO o Eri O O LO CO -p •LO •P LO CO ' stl CO -p P o I— [ -P 03 O 00 -p -p CO h ,i —1 -P The relation of Oxidized Flavor Intensity in Paw and Pasteurized Milk from Individual Cows, to the Lactation Period and to the Age of the Milk. niiiifiiifmfivm 1. LO Figure INTENSITY OF OXIDIZED FLAVOR - 31- jj;:;|j!jjii[iijs!!jSj[j[j!tiP CO - 32 - The data in Tables 2a, 2b, and 2c further emphasize the in­ fluence of oxidized flavor caused by pasture feeding and individu­ ality of the cov. Table 2a indicates that the influence owing to individuality was broken during the pasture season about two months after the cows had passed their peak production. Tables 2a and 2b show that the milk had consistently more intense oxidized flavor with age, both before and after pasteurization. The data in Table 2b strongly indicates the difference in protective action against flavor development, for whatever reason it may have been, between milks that oxidize spontaneously and those that do not show it until 24 to 48 hours after milking. Table 2c again shows the effect of individuality and perhaps a slight weakening (see December) of the protective action. Figure 1 is a summary of Tables 2a, 2b, and 2c. Another flavor study wan made to find the effect of holding raw for 24 hours, the milk of the same 6 cows, before pasteurizing. The results show the same general pattern for the three groups of cows as was shown by the more extensive study that covered an en­ tire lactation period, but emphasize that there is an increased tendency toward becoming oxidized, caused by the aging period of 24 hours, held at 4.4° C. Table 3 presents a typical trial of four such trials made early in May. It will be noted in Figure 1 that the occurrence of oxidized flavor was pronounced at this time for the first two groups of cows. - Table 3. 33 - The Effect on Oxidized Flavor Development in Pasteurized Milk Owing to a 24 Hours Aging Period of the Raw Milk. One of 4 Trials. Cows FlavorIntensity________________________ - Previsouly Fresh After Pasteurization Following 24 Aging at 4.4° C. Oxidized Spontane- Paw At once 1 day 2 days 5 days 4 days 5 days ously P P R P R P R P R P P P 1039 -I* 1264 + *4* + + Previously Oxidized after 24 to 4-8 hours 975 ? 1083 — - — " si. +- + + + + + + ++ - + ++ ++ *+• ~H* + Previously Rot Oxidi­ zed After 1 week 1113 - si. heat si. - heat - - - - - - * 1151_________________________________ *Not Oxidized After 10 Days * stale COLLECTION OF SAMPLES FOR BACTERIAL ANALYSIS It was noted earlier that in a plating trial, one series of plates stood for one week at room temperature, about 23° C., after the 48 hour incubation period at 37.5° C. Pin point colonies deep in the agar had developed by this time in the plates of both the raw and pasteurized samples from one cow and in the pasteurized sample from another cow. - 34 - Other plates were free of pin points and some were too crowded with other growth for accurate detection of pin points. The pasteur­ ized milk from both of these cows'was oxidized after holding at 4.4° C. for 2 days. Raw milk from the first cow was oxidized as soon as it was drawn. This chance observation suggested that an investiga­ tion into the relationship of bacteria and oxidized flavor might be valuable. These immediate problems were suggested: The possible need for drawing milk samples aseptically directly into a nutrient medium rather than into plain tubes, to permit an unbroken growth cycle; the possible need in the agar medium of nutrients or accessory substances to develop the pin points into larger colonies; incubation temperatures; frequency of occurrence of pin point colonies in milk subject to becoming oxidized; relation of the pin points to other udder borne organisms; identification, if possible, of the pin points. Owing to its well known ability to develop fastidious organisms, tryptose broth (Difco) plus 0.1 percent glucose was chosen for use in collecting samples. Samples collected into sterile tubes and into tubes of sterile broth were compared. In this way, exact duplication of samples was impossible from a single quarter at one milking but the experiment was repeated 24 hours later, reversing the order of placing the fourth and fifth streams drawn. The use of a broth collection tube did not increase the size of pin point colonies. It did, however, raise the total count; it also gave a higher percentage of the total count after the first 24 hours of the 48 hour incubation period. There was no relation between the - 35 - magnitude of* the total count and the percentage increase caused "by collecting the samples directly into "broth. (See Appendix, pp. i, ii) Subsequently all aseptically drawn samples from individual quarters collected for bacteriological examination were collected directly in­ to tryptose broth. As a preliminary to employing modifications of standard pro­ cedures, three plating trials were run, having all conditions stand­ ard, except that incubation was continued and plates were re-counted after 4 days and 6 days. Standard conditions require tryptone glucose skiirmilk extract agar and incubation at 37.5° C. for 48 hours only. The results in Table 4 confirm the previously held opinion that, except for pin point colonies, the milk from the cows under ob­ servation was of good quality, bacteriologically. The pin points that developed were noted after 96 hours incubation. The influence of holding the raw aseptically drawn milk, on the total standard count and upon the incidence of pin point colonies, was determined. The results are shown in Table 5. There is a general indication that pin points were more numerous in low count samples, though none of the counts could be considered high, and that they occurred in samples showing a declining count as the milk aged. Plating the samples after 1, 2 and 3 days had an inconsistent effect on the total count but it is significant that the two cows whose milk was consistently not oxidized (1113 and 1131), showed slight, steady increases in count as the milk aged. This agrees with the opinion held by Mattick (4-6) that all bacteria are probably capable of effecting a reducing action. Owing to the low magnitude - 36 - of all of the total counts and to the impracticability of counting pin point colonies accurately, it cannot be concluded from the data that there is a positive relationship between the two. - Table 4. 37 - Standard Bacteria Counts and the Incidence of Pin Point Colonies in Aseptically Drawn Milk and Counts After Ex­ tended Incubation. Bacteria Counts Cow Trial 48 hrs. 1039 1264 P.P. Standard Count After 96 hrs. P.P. 144 hrs. 1 70 - 180 2 110 - 230 3 30 - 1 1100 2 700 3 200 P.P. 210 ++ - 260 + 40 4- 60 ++ - 1400 - 1200 ? - - 1000 + 1600 + 370 _ 380 _ 410 - 190 + 900 - + Paw Milk Spontaneously Oxidized 975 1083 1 310 - 380 2 170 9 190 3 740 - 860 - 1 250 - 380 - Spreader - 2 900 - 1230 - 1210 ? - 3 1200 - 1600 - 2000 ~ - Paw Milk Oxidized After 24-48 Hrs. at 40° F. 1113 1131 1 1350 - 1380 - 1380 - 2 1600 - 2010 - 2040 - 3 860 - 860 - 910 - 1 760 - 790 - 830 - 2 1390 - 1420 - 1420 - 3 520 - 700 - 740 - Paw Milk Not oxidized After 7 days at 40^ F. - Table 5. 38 - The Effect of Holding Aseptically Drawn Eaw Milk at 4.4° C. Before Plating on the Standard Plate Count and on the Inci­ Cow dence of Pin Point Colonies Time Plated Within an Hr. of Milking After 24 Hrs. Count P.P. Count P.P. After 48 Hrs. Count P.P. After 72 Hrs. Count P.P. 1039 150 - 170 - 210 +■ 140 + 1264 2150 - 2020 - 2000 - 2300 975 370 - 30C - 330 4- 190 1083 600 520 •* 580 +- 560 1113 600 - 900 - 1280 - 1900 - 1131 1270 _ 930 _ 1400 _ 5180 — - COUNTS EMPLOYING MODIFIED AGARS An agar medium containing udder infusion as the only source of nutrient -was used in a plating trial on samples of milk which had previously contained pin point colonies in T.G.E.M. agar. This in­ fusion agar supported no bacterial and was given no further considera­ tion. Its use was attempted on the assumption that some nutrient sub­ stance that supported growth within the udder was lacking in'T.G.E.M. agar. If such a nutrient were in the Infusion It was appartnely destroyed by autoclaving or could not support growth alone. PHOSPHOLIPID ENRICHED AGARS Since the exhaustive literature review reported by Brown and Thurston (6) led them to conclude that oxidized flavor originated In the phospholipid fraction, agars fortified with this material w@2*e used for determining the effect on pin point growth. While, as shown previously, there was no definite relation between countable organisms - 39 - and "the incidence of oxidized flavor, there was a slight indication that, in "both raw and pasteurized milks from certain cows whose milk was oxidized, that pin point colonies involved. Throughout the course of the study colonies were picked from plates at various times until a collection of 4-0 cultures were collected. They were selected on the basis of color principally, according to Hucker (35), and Dorner (17) but particular attention was given to isolating and culturing the gray to white pin points that uniformly grew only deep in the agar plates. All of the organisms that were obviously aerobic grew well on stock culture agar (Difco) slants, some proved to be facultative in stabs but the pin point colonies would not grow under either condition. The use of COg Chambers (10 percent) was of no value in increasing the size of the pin points. This was determined with pour plates and streak plates of tryptose agar prepared in duplicate from broth cultures and incubated at 37.5° and 20° C. The broth cultures of all organisms except the pin points were ready for use after 18 to 24 hours. The pin point broth cul­ tures however did not show sufficient sediment until 48 to 72 hours. A short study of the effect of adding whole fresh egg yolk to tryptose agar proved that the pin points needed nutrient, or per­ haps a combination of nutrients, not supplied in tryptose agar alone. The yolk of one fresh egg was well washed, the yolk sac removed and the yolk only added to 500 ml. of tryptose agar. The heavy precipitate brought down by autoclaving was effectively re- - 40 - moved by filtering. Three pin point broth cultures were chosen for study, together with 3 cultures of representative micrococci, and an alkaline milk digesting gram variable rod that grew a pink surface colony on T.G.E.M. There was a definite increase in the size of all colonies except two. The yellow micrococcus colonies were no larger on the yolk-enriched medium. The gram, negative rod failed to grow on it. The egg yolk possibly furnished nutrient other than phospho­ lipid that promoted larger colonial growth of the micrococci. The results are shown in Table 6 . Another similar study using separator slime gave about the same results. The slime was washed, shaken and centrifuged several times until it was essentially freed of free fat. This, however, still left protein, lactose and mineral accompanying the phospholipid. Pure soybean lecithin was incorporated into tryptose agar by first emulsifying it in one-fourth its weight of Tween 80, the oleic acid ester of a polyoxyethylene derivative of sorbitan. Tween 80 was regarded as the best of the Tween series because the fatty acid content of lecithin is predominately oleic also, 70.6 percent according to Kurtz and Jamieson (42) . There was no apparent difference between agars containing 0.1 percent and 0.2 percent lecithin. One-tenth percent was used following this observation. After the selection of the original six cows used throughout this study, routine tests on a 4 year old Holstein cow, number 1279, showed that milk from her right front and left rear quarters was con­ sistently higher in total count than that in the other two quarters. - 41 - Pin point colonies had frequently "been noticed in the right rear and left front quarters and raw milk from these two quarters was oxidized after holding at 4.4° C. for 24 hours. Milk from the right front and left rear did not become oxidized when similarly held. (See Plate 3) ♦ Four plating trials on milk from the 4 quarters of this cow were run, using tryptose agar, tryptose agar plus 0.15 percent Tween 80, and tryptose agar plus 0.1 percent soybean lecithin emulsified with Tween 80. Table 6. The Influence of Enriching Tryptose Agar with Egg Yolk on the Colony Size of 7 Selected 42 U i M 0 —1 t*Q O < >H © td © tap o -p P fn i— I Eh Ph h © 0 a a -p PO *1 ft 00 IP l to 00 iO l ft ft LO 0 Ph ft ft ft Ph 0 0 0 0 ft ft 0 P •H o fD-j 0 P ft O ft Ph ft ft Pi Ph 0 ft ft © ft I ti o CO ft ra -p fPt o ft P Ph 0 LO Ph 0 Ph -P ?H O P h -P ft O P Ph 0 0 0 Ph 0 0 LO o ft Pi Sj 3 t£ O < © •P 00 ft 0 03 cJ 00 O - in i to P ft ft P 0 0 0 ft 1 to I l CO 00 0 O 0 0 o 0 M 0 O 0 © o CQ CQ CQ CQ d o 0 fPt ft 0 ft ft 0 rH CQ CQ 0 is 0 rd ft O etJ Ph O d CQ 0 Cfi ft 5 ft 0 -P •r-t ft o fCtQ < ft ft FQ O i—I rH O o o O u 0 -P 0 0 0 ^ 0 Pft © •P i— 1 •H rH £ 5 £ p ft ?H 0 -P 0 0 © f ft t ft fWt © f S 0 d j n o O ft ft fPt P 0 0 ?H Ph 0 ft O ft P F O a h d I E©H d ft 0 * ft O © P fot o ft 0 a ft e3 0 -P P ft O 0 Ph 0 c£3 O ft 0 O —1 ft s *h ft a oq f to CQ f t a* •H ft a co P O HD P Fh O d - 45 - Table 7. The Effect of Soybean Lecithin on the Development of Organisms in Aseptically Drawn rH CO fn M 0 p 0 rH > PiEh ! J> Q Ch jP § Pi o o 0 0 CO o o t o I jO o o CO o o to o oo ip rH o CO o roo o o 1 —1 LO o o fO to o o CO Hi o o C tO o O r- • O Hi r—i p o P £ > © •H P i 0 P O 00 00 CJJ rH LO -toP to © p rP© © ?H © P» + 53 O © IQ Pi PH Is ^0 0 p ra © -p £h Pi & ?o Eh rH co ■—i —» o t o 1.0 o o CD Hi o o to r— o o o rt ro H o o C7D ) -H O oo o Hi O o Hi o 00 O o o m rH 00 to Hi 00 o O Hi O LO to O o CVJ O o IP 00 o o o to to oo o to o o CD CO C0 O 0 o o oo o CO C rO H O o O i n to tO Hi 00 o Hi rH o o oo 00 <>* o oo o to PH ■r-i Eh rH P o Ph © m o u p gj B- |< If o oo rH O to g u JH g3 P, GP Ih K g Ph 0 PI <1 - 44 - // (• cd & ^ s © l MU - * y ••- ■■- •W••. •'* v- . » V \.O.k ■• • .7 • \.y ui o to 00 ujzO O ■ I■ I l - K ? . > - ^ I- •H t H oX cd +5 o C i - ££ I “ I- H ra to — - oo 5 1 1 §a * o^ ® -PH© a Ph h H 5 S h cd O CJ XI S3 cd © c3 - 45 - Lecithin provided better growth than Tween 80 in 14 of 16 cases. Tween 80 provided better growth than the basic Tryptose agar in 9 of 16 cases. Six of these 9 cases occurred in the right front and left rear quarters, both of which were free of pin points. It has been shown by Williams, Brcquist and Snell (71) that oleic acid favors cer­ tain organisms and may be toxic to others, depending somewhat on the concentration used. This suggests that organisms from the right rear and left front quarters may have been Inhibited by oleic acid but that this inhibition was overcome by some substance in the whole lecithin. Even though this possible explanation is based on results obtained from the milk of one cow, it is supported by the fact that In the herd as a whole, the variety of organisms from individual quarters was distinctly limited. Frequently not more than two different organisms, as could be Identified macroscopically, were seen. were seen. Occasionally as many as four The fariety from all four quarters of one cow, however, reached 6 - 1 0 frequently and 12 to 14 rarely. One particular cow not used in this study consistently showed very high counts of Micrococcus caseolyticus in all four quarters. Barely was any other organism seen. The work of Domer (17) substantiates this view, in part, though he grouped together the "white cocci", "yellow cocci", "streptococci of the long chain type" and "rods of the Bacterium lipolytlcum type". Differentiation of white and yellow cocci particularly can be made further by colonial characteristics other than color. Because of the careful selection of cows used in the present study streptococci were -46 not encountered. - Very few rods of any kind were ever encountered and none of these were fat splitting. Consistent with Domer's results, however, as well as those of Copeland and Olson (11) the micrococci predominated (see appendix p. iii). -47 - Plate 1. Comparing growth of an organism similar to Gaffkya tardissima on lecithin-fortifled tryptose agar (left) and on tryptone glucose skimmilk agar (right) . On the former some colonies are on the surface while on the latter all (pin points) are sub-surface. 48 - Plate 2. Showing the tetrad similar to the Bergey Manual description of* Micrococcus Gaffkya tardissima. The two hest focused are near the center. INOCULATION OF MI UK AND MILK FRACTIONS WITH AN ORGANISM SIMILAR TO GAFFKYA TARDISSIMA Bergey's Manual (4) was consulted for a description of the organism that had formed pin point colonies deep in T.G.E.M. medium and Tryptose agar culture dishes but which proved to he facultative when grown In lecithin agar (see Plate l) . Its properties did not completely match those of any described in the Manual hut were closest to those of Micrococcus Gaffkya tardissima. It resembled also Micrococcus Gaffkya anaerobia but was distinctly different In two outstand­ ing properties: It produced no gas In agar and grew better at 20° C . than at 37° C. Property Shape Arrangement Capsulated Gram Gelatin colonies Gelatin stab Agar colonies Broth Litmus milk Potato Indol Dextrose Aerobe or Anaerobe Temperature Habitat Gaffkya tardissima Isolated Pinpoint Small oval Small oval Grouped In fours Grouped In fours Yes Yes Positive Positive Very small, circular No growth on brownish by reflected gelatin light; coarsely gran­ ular under microscope Very slow and poor Very slow and development, no lique­ poor development faction. Very small, white, gran­ - Very small, white granular, erose ular circular, entire Fine, granular sediment Fine granular sedi­ ment Unchanged Unchanged No visible growth No visible growth Not formed Not formed Not fermented Not fermented Aerobic, facultative Anaerobic Optimum 57° C . Optimum 20° G . Natural Infection in Cow’s Udder guinea pigs - 50 - Large quantities of an aqueous suspension of the pin point organism were prepared from rinses in flat bottles. The organisms could not be harvested from the surface of the agar, but by shaking and subsequent aseptic filtering through a loose sterile cotton pad, the agar particles were removed. The turbidity of the filtrate was adjusted with sterile distilled water so that the addition of 5 ml. to one-half pint of product amounted to an increase of about 10,000,000 organisms per ml. of product. Samples of Winter and Summer milk were used to prepare raw and pasteurized 25 percent cream, 4 percent milk, skimmilk, 4 percent milk clarified once, 4 percent milk clarified twice and 4 percent milk clari­ fied three times. For each trial one set of samples was made from milk known to be susceptible to becoming oxidized in flavor; another set was made from milk known to be non-susceptible to becoming oxidized. Table 8 is a summary of the results of inoculating the various products and examining them for the occurrence and intensity of oxidized flavor after holding at 4.4° C. for 72 hours. The details for all trials, 5 on susceptible and 3 on non-susceptible milk are shown in pages XV and V of the Appendix. - Table 8. 51 - The Occurrence and Eelative Intensity of* Oxidized Flavor in Milk and Milk Fractions Inoculated with Organism Similar to Micrococcus Gaffkya tardissima; Summary of 8 Trials. Product Summer "Winter Susceptible Non-Suscep. Susceptible Non-Suscep Eaw Past. Haw Past. Eaw Past Eaw Past. 25$ Cream - 4# Milk - Skiramilk - - - 4ia Milk Clarifled Once Affo Milk Clarified Twice - 4^> Milk Clari­ fied 3 Times - - - + + - - -H* - - - - - - - + ++ - - - + ++ - - - + ++ - - These results indicate that the organism, though originally iso­ lated from quarters of cows whose milk "became oxidized, was unable to initiate the flavor in any of the samples inoculated. It did however Increase the Intensity of the flavor in those milks that were already oxidized or subject to "becoming so. The fact that clarified raw sus­ ceptible milk became somewhat more oxidized than the non-clarifled, can possibly be explained by the removal of leticocytes. There was no apparent advantage of multiple clarification over single clarification, if the removal of leucocytes was the controlling factor. - 52 - 11: was never possible to produce oxidized flavor in pasteurized homogenized milk as prepared commercially in the creamery* with inocu­ lations of the organism described above. Several trials on this milk* both clarified and non-clarified were made during the winter only. When inoculated and uninoculated samples were exposed to direct sun­ light for one hour* only the typical sunlight flavor developed. HEAT RESISTANCE OF UDDER BORNE ORGANISMS Page vi in the Appendix shows the numbers of organisms per ml. surviving pasteurization in 8 random aseptically drawn milk samples. While the average counts at both incubation temperatures 20° C . and o 37 C . were lower than for corresponding counts in raw milk samples similarly collected (appendix page vii) the percentage reduction of counts caused by pasteurization was very low. This suggested that udder borne organisms were thermoduric* that the milk medium served as a protectant against heat* or that high percentage reduction was not realized on samples kept free of contamination outside the udder. By heating separately several isolated species of organisms in broth and in milk it was found that none of the organisms was thermoduric and that milk did not serve as a protectant. Apparently* pasteuriza­ tion of the original milk samples within an hour of the time the cows were milked came at a time when the growth of organisms was more or less static. At this stage they are more difficult to kill. Table 9 shows the effect of heating separately several selected organisms in the presence of tryptose broth and in sterile ©kimmilk. Broth cultures - 53 - of each organism were transferred every other day while this study was in progress. Because of the slow growth of some of the micro- cocci in broth* cultures were not transferred to the tubes of milk and broth until after 36 to 48 hours of incubation. The holding temperature of pasteurization* 6l.8°C.* was used and inoculations made at 5 minute intervals into tryptose broth. Final judgment was not passed on the presence or absence of growth in the broth tubes inoc­ ulated from heated cultures of broth and milk until after 3 6 to 48 hours incubation. This extended incubation period was necessary be­ cause of the property of several of the micrococci to exhibit growth in the form of a viscous or granular sediment* while showing a clear broth above. All of the udder borne organisms shown in Table 9 were grampositive except the short rod that formed a pale white colony; it was gram-variable. No gram-negative organisms were found in the iso­ lations. These heating tests on single organisms are at variance with those on fresh milk. This was probably caused by the different stages of the life cycle at which the organisms in fresh milk and In broth cultures were heated. Even though the organisms were subjected to heat after a 3 6 to 48 hour incubation period* most of them were probably still in the logarithmic growth phase. sediment evidences this. The slow development of turbidity and/or - Table 9* 5 1* - Time Resis'bance "to 6l.8°C. of Udder Borne Organisms in Milk and Broth. Figures Indicate the Number of Instances __________Among 8 Trials that the Organism Survived._______________ Heated in Milk Heated in Tryptose Broth Organism Minutes Minutes 5 10 15 20 25 30 5 10 15 20 25 30 White ifticrococcus 8 2 0 0 0 0 8 3 l 0 0 0 White tetrad (sur­ face) 0 0 0 0 0 0 0 0 0 0 0 0 Gray tetrad (deep pin p.) 8 k 2 1 0 0 8 2 l 1 0 0 Pale white short rod 8 3 0 0 0 0 8 2 0 0 0 0 Buff micrococcus, ropy 5 3 1 0 0 0 6 2 0 0 0 0 Yellow micrococcus 0 0 0 0 0 0 2 0 0 0 0 0 Yellow short paired rods 0 0 0 0 0 0 1 0 0 0 0 0 Orange micrococcus large 0 0 0 0 0 0 0 0 0 0 0 0 Orange micrococcus small 3 0 0 0 0 0 2 1 0 0 0 0 Pink short rod 2 1 1 0 0 0 1 1 1 0 0 0 * To determine if any of the organisms heated (Table 9) and surviving 10 or more minutes were made more heat resistant, the longest heated broth tubes showing sediment and/or turbidity after 72 hours incubation were again heated similarly. Table 10 shows that the initial heating did not develop a thermal resistance in any of the organisms. All of them withstood 6l.8°C. for the same time or less during the second heating. Since, in the first experiment on heat resistance a milk - 55 - medium appeared to offer no protection, only tryptose broth was used in the second experiment. Table 10. Time Resistance to 6l.8°C. of Organisms Previously With­ standing the Same Temperature for 10 or More Minutes in ___________Milk or Tryptose Broth.________________________________ Organisms Which Previously Withstood Resistance to 6 l .8°C . on 6 l .8°C . for_______ Second Heating for_____ Minutes in Minutes in Tryptose Broth Tryp. Broth Milk 5 10 15 20 25 White Micrococcus 15 10 4 4 4 Gray tetrad (deep Pin P .) 20 20 4 4 4 4 - Pale white, short rod 10 10 4 - - - - Buff Micrococcus, ropy 10 15 4 4 4 - Orange Micrococcus, small 10 5 Pink, short rod 15 15 4 4 + . 30 - USE OF THE OXIDASE INDICATOR P -PEEWYLENEDIAMIHE OXALATE On the strength of the recommendations made for detecting oxidiz­ ing organisms by Castell and Garrard (10) and Carpenter, Suhrland and Morrison (9), tests on all isolated udder borne organisms were run. For reference, plates of E. coli were prepared, in order to contrast a typical gram negative organism with the gram positive and gram variable udder organisms. Pur© culture plates were flooded with a 1 percent aqueous solution of p-phenylenediamine oxalate and were observed over a period of several hours. This confirmed the opinion of Castell and Garrard (10) that gram positive organisms are poor oxidizers. Whereas -56the E. coli colonies quickly turned pink and progressively purple to black in 30 to 40 minutes, the udder borne organisms changed slightly or not at all during ^ to 5 hours. It was observed that the indicator solution did not wet the surface of some colonies. To overcome this, aqueous and saline suspensions were prepared and the indicator was added to the suspensions in test tubes. Slight de­ grees of variation in shades of pink color were noted immediately after mixing but within a few minutes the differences were too slight to de­ tect. When these tubes were compared in a Klett Summer son photoelectric colorimeter it was observed that the deeping of pink color with time, was more a function of the instability of the indicator than a measure of oxidation by the organisms. Since the indicator was prepared immedi­ ately before use, in double distilled water (glass), its use as a mea­ sure of the variation in oxidizing activities of udder borne organisms can not be recommended. DEHYDROGENATIOE BY UDDER BOEBE ORGANISMS Wunheimer and Fabian (M-8 ) have shown that the micrococci vary widely in their ability to dehydrogenate methylene blue when in the presence of different substrates. While their work included the use of specific carbohydrates, alcohols and amino acids, substrates of particular interest in this study were whole lecithin and oleic acid, owing to the increase in growth in agar plates on these materials. It seemed to be relatively unimportant whether any fractional part of lecithin other than oleic acid was the activating substance, particular­ ly for the organisms that were difficult to cultivate (pin points). In -57a series of tests on methylene blue reduction, however, choline was used as a substrate also to determine if* the rate of reduction were increased after its liberation from lecithin. The modified Thunberg technique recommended by TMbreit, Burris and Stauffer (6 9 ) was used. This differs from the original in that the cell suspension is added directly to the substrate and methylene blue mixture in an ordinary test tube. Oxygen diffusion is prevented except in the upper layers of the tube by the addition of a buffered 2 per cent agar. agar solidifies the entire mass just prior to incubation. The Reduction is noted in the lower parts of the tube. Table 11. A Comparison of the Methylene Blue Reducing Ability of Udder Borne Organisms in the Presence of Lecithin, Oleic Acid and ___________Choline at 35°C. Organism Average of 5 Trials_________________ Reduction Time Minutes No Substrate Lecithin Oleic Acid Choline 1 . White, Micrococcus 205 185 530 20 6 215 191 2 2 I4. 2 . White tetrad (surface) 3. Gray, tetrad (deep, pin pcfcit)1 ko 26 hi 72 k. Pale white, short rod 35 30 20 h3 5- Buff, Micrococcus, ropy lli-O 108 115 130 6 . Rose, Micrococcus, uncommon *4-32 k62 335 568 7. Yellow, short paired rods 260 135 121 278 8 . Orange, Micrococcus, large 39 36 33 hi 9. Orange Micrococcus, small 170 190 163 25if • 0 H h20 Pink, short rod 152 90 95 176 -58Of the 10 representative organisms studied, 1, including the unclassified tetrad (No, 3), reduced methylene blue in the presence of lecithin faster than they did with no substrate. Seven of them were more active on oleic acid than on lecithin; the unclassified tetrad was not in this group. All but one organism, reduced more slowly in the presence of choline than with no substrate, and in all cases the entire group was slower in the presence of choline than in lecithin or oleic acid. Hastings, Davenport and Wright (32) felt that all organisms in milk were reducing and that even a very slow reducer contributed its effect to the total reaction. It was point­ ed out earlier that there were very few species of organisms found in the separate quarters of the cows used in this study, and further (see Plate 3) that the pin point colony of the tetrad similar to Gaffyka tardissima was at times the only organism found in quarters showing milk with an oxidized flavor. Because of a lowered Eh in milk having a mixed flora end a high leucocyte count as well, it is conceivable that the effect of a single organism would be unnoticed. In this case, the ability of the tetrad to utilize lecithin, or a component of lecithin, would be inhibited or prevented. DISCUSSION Market milk without an oxidized flavor can be uniformly produced. Regardless of the breed of cows, their feed, the season, stage of lac­ tation, copper or iron content of the milk, the character or numbers of bacteria, enzymes of non-bacterial origin or the chemical composition -59 of the milk, the trade practice of homogenizing and in some cases the use of antioxidants or high heat treatment prevents oxidized flavor. Even so, there is every reason to believe that a market for raw milk and non-homogenized pasteurized milk will continue. As sanitary practices have improved and as the federal, state and city governmental control agencies have striven more intently to prevent the sale of milk from diseased cows, oxidized flavor has become more of a problem. For the distributor who is compelled to market all or part of his supply as raw or non-homogenized pasteur­ ized, or for whom the high temperature short time method of pasteuriza­ tion is impractical, or the use of antioxidants illegal, the solution to the oxidized flavor problem may be a matter of selecting milk on an individual cow basis. This has been necessary before in order to produce milk within certain bacteria count limits, to maintain a minimum fat or total solids percentage, to control color, and in some cases to avoid off flavors. SUMMARY The occurrence and intensity of oxidized flavor in the milk of 3 pairs of cows was established. The pair whose milk was oxidized immedi­ ately after being drawn exhibited the defect, as has been shown by several previous investigators, more prominently in winter than in summer, more prominently in pasteurized than in raw milk and more prominently when their milk was pasteurized after an aging period of 2 *1- hours. 6oThe second pair of cows, chosen for study because their milk did not become oxidized until it was held for 2 k hours or more, showed the same tendencies as the first pair, though the rate of flavor development was slower. The third pair of cows, a control pair, had not previously shown oxidized flavor in their milk after it was held for one week. This pair stood apart, distinctly, from the other two throughout the en­ tire study in that their milk was oxidized for a short period only during the winter and then, not as intensely so, before or after pasteurization. Of primary interest was the bacteriological picture of the milk from these 3 pairs of cows because standard plate counts on milk of three of them in the first two pairs had shown pin point colony de­ velopment when the plates were incubated beyond the standard k8 hour period. The pin points did not develop at all when an udder infusion was used as the sole source of nutrient. Agars fortified with egg yolk and with separator slime were used to provide a source of phos­ pholipid. Larger total counts resulted on these agars but the pin points did not grow appreciably larger. Finally, pure soybean leci­ thin was used as an enrichment in Difco Tryptose Agar. This not only developed the pin points into larger colonies but let them develop on the surface of the agar as well as in the deeper layers. None of them had ever developed on the surface of tryptose glucose skimmilk extract agar nor on any of the other enriched media. -6lWhen suspensions of the pin point colony organism were inoculated into milk and milk fractions in the amount of about 1 0 ,0 0 0 ,0 0 0 per ml. , oxidized flavor was not initiated in samples that were normally nonsusceptible. This was true in both winter and summer milk. However, when inoculated into susceptible milk and milk fractions, the flavor was more intense in the inoculated than in the uninoculated controls. There was a tendency toward a greater increase in intensity of flavor in raw clarified milks than in non-clarified whole milk, cream or skiramilk. Leucocyte removal in the clarified milks may have been responsible for this. Triple clarification effected no difference over double or single clarification. Inoculated homogenized pasteur­ ized milk was completely immune to a change in flavor as a result of similar inoculations. Organisms borne by the normal udder, under the conditions which they were studied, were not resistant to pasteurization by the holding method and did not develop a resistance during the first heating. The facts that pasteurization efficiencies on aseptic ally drawn milk were low and that the udder organisms in pure culture were never able to withstand over 2 0 minutes at 6 l .7 ° C . seem to be contradictory; this can be explained on the basis of the phase of growth that the organisms were in at the two different times. Since all total counts on the fresh milk were low, comparatively, the organisms were probably sub­ ject to the germicidal action within the udder and were not developing. If this had not been true, higher counts could have been expected. -62The fourth drawn stream was consistently taken for examination. According to Corner (17) such samples should have shown much higher counts. His samples were taken n&ar the end of the milking as an extra precaution against outside contamination, hut his counts were higher than those obtained in this study. The low heat resistance of the organisms in pure culture was probably owing to their being in the logarithmic growth stage when heating began. As Copeland and Olson (11), Hucker (35) Dorner (17) have shown, the micrococci are the most numerous of udder borne organisms. The one that showed an association with, but which was unable to initi­ ate oxidized flavor in milk, was tetrad meeting the Bergey Manual description of Gaffkya tardissima, except for differences with respect to growth on gelatin, colony description as grown in agar, and habitat. These differences are admittedly slight. Except for its ability to grow at 20°C . and that it did not produce gas in agar, the pin-point-tetrad found also resembles the Bergey Manual description of Gaffkya anaerobla . The micrococci of the udder showed a wide range of dehydrogenatinn power against methylene blue when in the presence of lecithin, oleic acid and choline. In most cases oleic acid was activated at a faster rate than was lecithin, but choline was activated more slowly. Seven of 10 selected udder borne organisms including the tetrad similar to Gaffkya tardisfeima, were able to activate lecithin faster than they did without an added substrate. This lends weight to the conclusion of Brown and Thurston (5) who in 19k0 said "the trend of the literature -6aat the present time seems to point to the phospholipid fraction as the source of oxidized flavors in milk and cream". CONCLUSIONS Standard procedure is inadequate for the determination of the organisms in aseptic ally drawn milk, particularly freshly drawn milk showing an oxidized flavor. An organism similar, if not identical to Micrococcus Gaffkya tardissima is associated with the development of oxidized flavor in­ side the udder of certain cows. While actively growing pure cultures of this organism in broth or skimmilk are destroyed in 20 minutes at 6 l.6 °C., it survives 30 minutes at the same temperature when heated in fresh aseptically drawn milk in which it occurs naturally. Udder borne organisms develop better in lecithin fortified agar media. -6hr BIBLIOGRAPHY (1) Anderson, J. A«, Hardenberg, J. K., and Wilson, L. T. Concerning the Cause of Rancid and Oxidized Flavor of Bovine Origin* (2) Babcock, C. J* Jour. Dairy Sci., 1 9 *i+8 5 -i48i4, 193&* The Effect of Homogenization on Certain Characteristics of Milk. U.S.D.A. Tech. Bull. 1|38 23 pp., 19324* (3) Baker, J. C. and Breed, R. S. The Reaction of Milk in Relation to the Presence of Blood Cells and Specific Bacterial Infections of the Udder. Exp. Sta. Tech. Bull. 80 N.Y. Geneva Agr. 19 pp, 1920. (24) Bergeyfs Manual of Determinative Bacteriology. 1 9 i48 . 6 fch. Ed. Williams and Wilkins Co., Baltimore, Md. (5) Brown, C. W., Smith, L. M- and Ruehle, G. L. A* A Bac- tiological and Biochemical Study of Experimental Butters. Jour. Dairy Sci. 3*3 7 5 -2405, 1920. (6 ) Brown, W. C. and Thurston, L. M. A Review of Oxidization in Milk and Milk Products as Related to Flavor. Sci. 23i629-^85, 19U0. Jour. Dairy -65(7) Brown, W. C., Thurston, L. M., and Dustman, R. B. Flavor in Milk 17 • Oxidized Studies of the Relation of the Feed of the C0w to Oxidized Flavor. Jour. Dairy Sci. 20#133- 1h5s 1937. M. (8) Cannan, R. K., Cohen, B., and Clark, dation-Reduction. X. pensions. Studies on Oxi- Reduction Potentials in Cell Sus­ Pub. Health Report. Sup. 55 Treas. Dept., TJ.S. Public Health Service, 1926. (9} Carpenter, C. M., Siahrland, L* G., and Morrison, M. Oxalate Salt of P-Amino dimethyl aniline, The An Improved Rea­ gent for the Oxidase Test. Sci. 105 Ho. 2738, & 49-6 5 O, 191*7 • (10) Castell, C. H., and Garrard, E. H. MOxidase Reaction” of Bacteria in Relation to Dairy Products. Food Res. 5*215- 222, 19i40. (11) Copeland, L., and Olson, T. M. Cows* TJdders. S. Dak. Agr. Exp. Sta. Bull. 218, 1926. (12) Coulter, C. B. Systems. Boullion. I. The Bacterial Flora of Normal Oxidation-Re duct ion Equilibria in Biological Reduction Potentials of Sterile Cui-fcure Jour. Gen. Physiol. 12fl39-lW>, 1928. -66(13) Dahlberg, A* C., Kucera, J. J*, Hening, J# C., and Hucker, C. J. Mastitis D T» The Composition of Milk as Affected by Latent Mastitis. N.Y • (Geneva) Agri. Exp. Sta. Tech. Bull. 239 1 6 pp. 1 9 3 6 . (lit) Dahle, C. D* Tallowy Flavor in Milk. Penn. Agr. Exp. Sta. Bull. 336 (ann. rpt. pp. 28), 193&* (15) Dahle, C. D., and Palmer, L. S. from Individual Cows* Oxidized Flavor in Milk Penn. Agr. Exp# Sta. Bull* 3^47- 30 pp., 1937* (16 ) Doane, C. F. Leucocytes in Milk and Their Significance. Md. Agri. Exp. Sta. Bull. 102. 8 pp., 1905* (17) Dorner, W. The Bacterial Flora of Aseptically Drawn Milk. N.Y. Agr. Exp. Sta. Tech. Bull. 1 6 5 * 39 PP*, 193^. (18) Ellingworth, S., McLeod, J. W., and Gordon, J. Further Observations on the Oxidization by Bacteria of Compounds of the Para—Phenylene Diamine Series. Jour. Path, and Bact. 32 1173, 1929* -67(19) Engelbreth-Holm, J. Leukaemia in Animals. 2l}5 PP* Oliver and Boyd, London, 19l*2. (20) Evans, A. C. The Large Numbers of Bact. Abortus var. Lip- olyticus Which May be Found in Milk. Jour. Bact. 2*185. 1917. (21) Evans, A. C. Udders. The Bacteria of Milk Freshly Drawn from Normal Jour. Inf. Dis. 18*107, 1917* (22) Fabian, F. W., and Trout, G. M. Influence of Various Treat­ ments on the Bacteria Content of Frozen Cream. Sci. 2 6 $9 5 9 -9 6 5 . (23) Fetzer, L. W. Conditions. Jour. Dairy I9h3 • The Lecithin Content of Milk Under Pathological Science 35*339 (n.s.) 1911 (PJ)) Frazier, W. C. A Defect in Milk Due to Light. Jour. Dairy Sci. Ilf375, 1928. (25) Frazier, W. C., and Whittier, E. 0. Studies on the Influence of Bacteria on Oxidation-Reduction Potentials of Milk I. fluence of Plate Count of Milk Organisms. 239-251, 1931a# Jour. Bact. 21 1 In­ —68— (26) Frazier, W. C., and Whittier, £• 0# Studies on the Influence of Bacteria on Oxidation-Reduction Potentials of Milk* Influence of Associated Cultures* II* Jour* Bact* 21*253“262. 195113• (27) Garrett, 0, F*, Bender, C, B., and Tucker, H* H* Relation of Grass Silage to the Color, Vitamin C and Flavor in Milk fr am Individual Cows# (28) Gillespie, L* J* Jour* Dairy Sci. 20*1^26, 1957* Reduction Potentials of Bacterial Cultures and Water Logged Soils. Soil Sci. 9»199-216 , 1920* (29) Gordon, J*, and McLeod, J. W. The Practical Application of the Direct Oxidase Reaction in Bacteriology. Jour. Path, and Bact. 31 1185, 1928. (30) Gould, I. A*, and Sommer, H. H. Effect of Heat on Milk With Especial Reference to the Cooked Flavor. Bull. 16L*. Mich. State Tech. i-jQ pp., 1939• (31) Guthrie, E. S., and Brueckner, H. J. Oxidized Flavor. 1 0 pp., 1933* The Cow as a Source of W. Y. (Cornell) Agr. Exp. Sta. Bull. 6 0 6 -69(32) Hastings, E. G., Davenport, A*, and Wright, W. H. Tne Influ­ ence of Certain Factors on the Methylene Blue Reduction Test for Determining the Humber of Bacteria in Milk. Jour. Dairy Sci. 5tk3Q-U5h9 1 9 2 0 . (33) Holm, G. E., Wright, P. A., and Deysher, E. F* lipids in Milk. I* Products* The Phospho­ Their Distribution Among Some Dairy Jour. Dairy Sci. l6*I|i+5“i+5^4t 1933* (3U) Horrall, B. E. Its Products. A Study of the Lecithin Content of Milk and Purdue Univ. Agr* Exp. Sta. Bull. i|01. 31 PP* 1935. (35) Hucker, G. J. Studies of the Coocaceae IV. of Genus Micrococcus Cohn. Bull. 102. U.Y. State Agr. Exp. Sta. Tech. I46 pp., 192U. (3 6 ) Jensen, L. B., and Grettie, D. P. on Fats. (37) Kende, S. The Classification Action of Microorganisms Food Research 2*97-120* 19J+0* it tt . Untersuchugen uber 1 olig-talgige1 *schmirgellige’ Veranderungen der Milch. 13il-lll, 1932a. Milchwirtschaftliche Forsuchengon. -70/ ” (38/ Kende, S. Ur sac hen und Bekampsung der woligen Schmirgeligen” n Milch und ahnlicher Milchfehler. Centbl. Bakteriologie Zweite Abt. 8 6 . Ban p. 91., 1932b# (39) Kendo, S. 29 11171* Combating the Emery Defect in Milk. Chem. Absts. Osten Milchwirtschaft Ztg. JL+ls277-280, 193^-a* (140)*Kende, S. Catalytic Oxidation of Butterfat in Milk. Brit. Chem. Abats. p 1080 (from Proc. World’s Dairy Congress) Milan. Sec. II. pp. 1-11, 193^+b. (141) Keyes, F. G., and Gillespie, L. J. A Contribution to Our Knowledge of the Gas Metabolism of Bacteria. The Absorption of Oxygen by Growing Cultures of B. Coli and of Bact. Welchii. Jour. Biol. Chem. 13*305-310* 1912. (J|P) Kurtz, F. E., Jamieson, G. S., and Holm, G. E. of Milk. Fraction. I. Tne Lipids The Fatty Acids of the Lecithin-Cephalin Jour. Biol. Chem. 106*717-72i|. 193U* (1*3) Leeder, J. G., and Herried, E. 0. The Relation of Bacteria and of Oxygen to the Flavor of Milk Susceptible to Becoming Oxidized. Vermont Agr. Exp. Sta. Bull. U57* 15 PP-* 19U0. / -71(1+1+) McFarland, G. C., and Burgwald, L* H. Prevention of Oxidized Flavor in Frozen Cream by Homogenization and High. Temperature Pasteurisation. Jour. Dairy Sci. 2 3 *1+9 ^ 496 , 191+0. (1+5) Manual of Dehydrated Culture Media and Reagents, 90-91+* pp. 88, Ed. Difco. Laboratories, Inc., Detroit, Mich. 191+1+. (1+6) Mattiok, A* T. R. Oilaness in Milk. Jour. Agr. Sci. 17*588-391, 1927. (1+7) Mundt, J. 0., and Fabian, F. W. The Bacterial Oxidation of Corn Oil. Jour. Bact. 1+8*1-11, 19l+l+» (1+8) Nunheimer, T. D., and Fabian, F. W. the Micrococoi. Respiratory Studies of Jour. Bact. 1+1+*215-232, 191+2. (1+9) Peters, I. I., and Trout, G. M. The Attraction Between the Fat Globules and the Leucocytes of Milk. Jour. Dairy Sci. 28 1277-283, 191*5. (50) Peters, I. I., and Trout, G. M. The Influence of pH on the Attraction Between the Fat Globules and Leucocytes of Milk* Jour. Dairy Sci. 28*283-289, 19l+5« -72(51) Plastridge, W. IT., Anderson B. 0., Williams, L. F., and Weirether, F. J. Infections Bovine Mastitis. Sta. Bull. 251. (52) Potter, M« C. Storrs Agr . Exp. 60 pp., 1959* The Rate of Fermentation as Measured by Differ­ ence of Potential. Ffoc. Univ. Durham Phil. Sec. 1+f230-231, After Jour. Chem. Soc. (London) 10l+*230-231, 1913* (55) Prewitt, E., and Parfitt, E. H. Flavors in Pasteurized Milk. (5 I4) R©od, 0* E. Effect of F©cds on Oxidized Jour. Dairy Sci. 1 8 *1468, 1935• Vitamin C Retards Development of 0xidized Flavor in Fresh Milk* Agricultural Research Administration. Report to the Chief of the Bureau of Dairy Industry, p. 23, 19^+7• (55) Roland, C. T., and Trebler, H. The Effect of Fat Content on Oxidized Flavor in Milk and Cream. Jour. Dairy Sci. 20* 3145-350, 1937* (5 6 ) Ruehle, G. L. A* Microorganisms. June 3 0 , 1921+• The Flavors produced in Milk by Butter Mich. State Coll. Rpt* of the Vet* Div. -73(57) Skar , Olav. The Behavior of Leucocytes in the Methylene Blue Reductase Test. Ztschr. Fleisch und Milchhyg. 23*14+2-1+1+7, 1913. (58) Standard Methods for the Examination of Dairy Products. American Public Health Assn. 7th. Ed., 191+1+. (59) Stark, C. N., and Scheib, B. J. A Study of Fat Splitting and Casein Digesting Bacteria Isolated from Butter. Jour. Dairy Sci. 19*191-213, 1936. (60) Supplee, G. C., and Beilis, B. Milk. The Copper Content of Cows’ Jour. Dairy Sci. 5*1+55, 1922. (6 1) Swanson, A. M., and Sommer, H. H. I. Oxidized Flavor in Milk. Effect of the Development of Oxidized Flavor on the Iodine Number of the Phospholipid Fraction of Milk. Jour. Dairy Sci. 23*201-208, 191+0. (62) Tarassuk, N. P., and Palmer, L. S. Study of the Riysical and Chemical Properties of the Adsorption Membrane Around the Fat Globules in Cream. I. The Effect of the Adsorption Membrane of Synthetic Creams on Curd Tension of Cows’ Milk Jour. Dairy Sci. 22f5l+3-558, 1939* -7i+(63) Thurston, L. M., Brown, C. W., and Dustman, R. B. Flavor in Milk. Oxidized Flavor* I* The Probable Relation of Lecithin to Jour* Dairy Sci. 18*301, 1935. (6 I4) Thurston, L. M., Brown, W* C., and Dustman, R. B. Flavor in Milk. Oxidized II. Oxidized The Effects of Homogenization, Agita­ tion and Freezing of Milk on its Subsequent Susceptibility to Oxidized Flavor Development. Jour. Dairy Sci. 19*671-681, 1936. (6 5 ) Tracy, P. H., Ramsey, R. J., and Reuhe, H. A. Certain Bio­ logical Factors Related to Tallowiness in Milk and Cream. 111. Agr. Bxp. Sta. Bull. 389* 17 PP«, 1933* (66) Trout, G. M., and H alloran, C. P. Milk. Sediment in Homogenized Mich. Agr • Exp. Sba. Quar. Bull. 15 (2)*107-110. 1932. (67) Trout, G. M., Scheid, M. V., Peters, I. I., and Mallmann, W. L. Influence of Bacteria, ^east and Leucocyte Distribution in Bottled Homogenized Milk and Sediment Formation. Mich. Agr. Exp. Sta. Quart. Bull. 26, number 1+, 285-296, 19i4+. -75(6 8 ) Trout, G. M», and Sharp, P. F. The Reliability of Flavor Judgments, with Special Reference to the Oxidized Flavor of Milk. Cornell Univ. Agr. Exp. Sta. Memoir 20l+. 60 pp., 1937- (6 9 ) Umbreit, W. W., Burris, R. H., and Stauffer, J. F. Manometric Techniques and Related Methods for the Study of Tissue Metabolism. Burgess Pub. Co. Minneapolis, Minn., 19U8* (70) Whitnah, C. H., Martin, W. H*, and Beck, G. H. Oxidized Milk Flavor as Related to Carotene, Lecithin and Vitamin C. Jour. Dairy Sci. 2 0 *l|.3 1 -i432 * 1937. (71) Williams, W. L., Broquist, H. P., and Snell, E. E. Oleic Acid and Relfeed Compounds as Growth Factors for Lactic Acid Bacteria. 19U7# Jour. Biol. Chem. Vol. 170 No. 2 i6L9-630, i A P P E N D I X A comparison of standard "bacteria counts in aseptically drawn milk from individual quarters of normal cows when samples were col­ lected Into sterile test tubes and Into tubes of sterile tryptose broth (Dlfco) plus .1 percent glucose. Series I and II collected and plated 24 hours apart. Series I Fourth Stream Drawn Into Tube; Fifth Stream Into Broth. __________ Standard Bacteria Count No Broth Broth Pin Points Pin Points 48 120 48 120 Sample 24 hrs. 48 hrs. hrs hrs. 24 hrs. 48 hrs. hrs. hrs. 1 0 0 2 0 0 3 0 4 0 0 30 40 10 20 20 20 50 30 40 5 80 100 90 90 6 120 170 220 220 7 200 220 210 230 8 400 460 450 490 9 460 500 490 500 10 620 710 660 690 11 1470 1620 1700 1740 12 2500 2530 2700 ? 2600 13 5300 6000. 5000 5900 14 7600 8500 8200 9100 15 12100 13800 13300 14600 17200 16 Average 3004 19000 3354 20000 3319 21500 3610 - + + - + + - + - + “ + - + ii Series II Fourth Stream Drawn Into Broth; Filth Stream Into Tube ___________ Standard Bacteria Count No Broth Broth Pin Points Pin Points 48 120 48 120 Sample 24 hrs. 48 hrs. hrs hrs. 24 hrs. 48 hrs. hrs. hrs, la 30 60 2a 0 0 3a 50 4a 0 5a 100 6a 200 320 7a 350 410 3a 100 100 9a 670 880 10a 1000 11a 100 110 40 60 90 200 200 20 10 20 150 190 240 280 480 500 160 160 990 1100 1300 1130 1200 920 1150 1200 1320 12a 4100 4900 4200 6000 13a 2000 2300 2000 2000 14a 10600 11100 12600 13100 15a 17000 17200 17000 17700 27000 4219 27300 4452 26800 16a Average 3995 - + 90 ? - - + - 25900 ? 4114 + _ + - + “ + - + - + Frequency of occurrence of pigmented colonies was recorded during the study. A total of 2150 plates made on asceptically drawn milk were examined. Since there was no apparent selec­ tivity for certain pigmented colonies by any particular agar medium, the percentage distribution shown below is for all agars used. Approximately 85 percent of the plates, however, were poured with Tryptose, Tryptose plus Tween 80, Tryptose plus lecithin and Tween 80, and T.G-.E.M. SUMMARY OH* OCCURRENCE OF PIGMENTED COLONIES ON 2150 )lony Color White PLATES OF ASEPTICALLY DRAWN MILK Plates Percentage 1554 72.3 Yellow 176 8.2 Orange 166 7.7 Buff 155 7.2 Pink 56 2.6 G-ray 37 1.75 Rose 6 .25 The occurrence of oxidized flavor, as a result of inoculating an organism similar iv CQ 0 § -P fcj 0 •H I -1to ■(— [ 51 s HR • P d Pi H> rH O O o 0 o • •H o h AJ J>. o rH Eh PI •r-l M 53 Pt• • HR ■—cd1 pj H1 o o o d rH O nd Pi c3 •H ^3 03 o o a I 0 Pi O S u HI« 'nH I--1 -p cd -p S’ 3 i —i roP 03 iH t Q •H 0 O S PH © -P P-t • o o Pi IH 0 •rH o 4 o a IHoPJ o o P CQ o o CQ O •d 0 i—i P o o PS 1— 1 PS •H M ll •H . Pi o M CQ o •H -8 •H 03 # o Pt 0) P 03 -p O 13 -P Pi CQ CQ *H PS O •rH rd -p -P Pi P i cd Cd # Ph s O -P (d 5> 0 CQ roO Pi o !=- + . p! o M o i— 1 a •H M s 4 r r PS H> o o ra * + Pi Hi i—as1 o f kHR CD o -P cd CO 0 t M t i I Hi lO * cd o 0 PS P H o * H R LOj CO PS o + o !>» Pi t> rH P *H Pi to H I LQ The occurrence of Oxidized flavor, as a result of inoculating an organism similar v ra § •rH -p i— 1 to •rH ai H 1 rH o © •H CD *S 1-1 o o © o > © 5 ^ *a* W NP. H 1 r—1 © i — I o r Q •H H< -P Hi © -P «3 •rH d i— 1 Fk rM 0 I—I [—1 ra * d 03 1 H* a o £3 i— I O o ^3 -p *H 03 © -P £3 © O to p © I p o S3 O rP CVJ d © -P <3 ■—1 d o o M£3 © -p o •p S3 •H %f •H to S3 o *H © r M i —i • rH H* s ra cd •H d -P c3 © t> S3 © U -P cJ J?

o i —1 EH p r^J •H ft © * Cd s , Pi • •rH 'e ft cd p H ft i— 1 o o o P © ft • ra © o o cd o ft *r~i 8 fpt ©A ft is: • f t p -H •eft cd p ft ft r —1 o P i o o © ft •M *H •H ft! to a •H Cy O -P © © rd i—1 P to P cd to cd 5s O t> cd ft ft) ft CD tsl •H ft *H 8 % C D O P CD P o o o CD £ P o © Cd © -P © cd o o ft © ft •M -P © i —! rO ft © cd o ft! O to p m CV3 o P i © -p ■iH © © -P cd & ft s O © P GO ft! O ft a nd © ft A P o o Hp •§ © P o ft ft cd {> Pi © © ft o + t t t + + + + i + I + + + + + + + t t + + • o f to p ■H ft f t! op CO •H o * o f t ! o i —1 p •H f t S3 TsP- p ft o o ft © nd Pi + O o © cd t t + + •f + + + • o £ o © © p P ft O tft p LO o CO o January o -p + + + + CO CO ft LO 1— 1 cd cd ft £ ft P *"3 ft Pi ft I I CO CO ft LO vii p 3 iH *rH 1 •H Pi CQ nd & 9 P O a *H Q {E q •H ra •H p £3 P 3 tD i— I P o o p •H 3 © P o O •d © P r— ! Pi C Q •H Pi P © C Q 3 «N P O S > 3 i-l C m *d CD *H nP •H M o pm 3 Pi © i—! CD O P CD P J3 o o o © 1 1 1 1 1 1 1 I 1 t 1 1 1 1 i t 1 I 1 1 1 1 1 g■ •rH •3 S • © p "eR- 3 ra d1 r— 1 O 3 Ph p • o o d* • d< -P 3 ra p o P -P Pi CO r- O P © ra P ra i P o p o -p 1 1 1 1 I 1 p © -P ra -p P © O p o ■tpLO to **\ CO CQ a £ o p P O o p M O P o © rd 3 gj ra CQ i CD ra 4 o 'g *d © »H p ll < A comparison of Tryptone glucose skimmilk agar and Tryptose agar, with standard Nutrient Agar (Official for milk before 1939) as reference, on aseptically drawn individual quarter pasteurized random samples. This study was made prior to the one employing Agars fortified with phospholipid. TRYPTOSE 20u 0. 37u C. SAMPLE T. Gr. E. M. 20° C. 37 °c . S. N. A. 20° C. 370 C. 1 80 590 10 210 70 2 700 900 1500 1700 210 160 ? 3 100 300 270 520 270 320 4 400 200 ? 260 490 10 60 5 ## 6 7 # ## 8 Average * Pin points 460 60 170 40 760 0 50 460* 580* 500 540 100 180 500* 800 340* 250 100 200 200* 270 350 200 313 476 409 584 90* 106 180 201 # Oxidized when drawn §§ Oxidized after pasteurization Colonies of apparently the same species were the largest on Tryptose agar, next largest on T.G-.E.M., and the smallest on S.N.A. This was true in "both the plating of raw and pasteurized random samples shown in the two tables above. It was also observed that proteolytic colonies were more easily distinguished in T.G.E.M. than In the two other agars. This has been a well recognized ob­ servation, however, owing to the clearing of turbidity of the skim­ milk. A comparison of Tryptone glucose skimmilk agar and tryptose agar, with standard Nutrient Agar (Official for milk before 1939) as reference, on aseptically drawn individual quarter raw milk random samples. This comparison and the one on pasteurized samples, preceding, was made preliminary to the study of Agars fortified with phospholipid. 1 50 # 2 1300 150* S. N. A. 37° C. 0 T. G. E. M. 20° C. 37° C. l i.o TRYPTOSE 20° C. 37° C. w :o SAMPLE 200 900 100 200 1400 1800 1300 1200 1200 3 600* 920 200 470 200 510 4 400 790* 100 300 0 200 5 2600 2900 1600 3300 500 1900 170 Spreader 6 # 300* 4900* 460 540 7 # 800* 6000* 300 1000* 8 650 900 400 9 590 2000 10 3000 1029' Average * Pinpoints # Oxidized Flavor 0 400 820 350 610 360 900 230 500 6500 2100 4500 1800 3000 2554 752 1356 455 801 X Leucocyte Counts of Six Mastitis-Free Cows During One Lactation Period. __________________________000_Omitted______________________ Cow Breed Freshened April dune Aug. Oct. Dec. Feh 1039 Jersey 4-10-47 72 105 127 166 200 175 1264 Holstein 3- 8-47 67 79 83 101 129 Dry Haw Milk Spontaneously Oxidized 975 Holstein 4- 7-47 276 181 232 250 400 600 1083 Holstein 4- 9-47 125 165 300 280 290 520 Eaw Milk Oxidized After 24-48 hrs. at 4.4° C. 1113 Holstein 3-26-47 164 300 210 400 526 700 1131 Jersey 3-20-47 324 262 360 700 721 890 Eaw Milk Hot Oxidized After 7 days at 4.4° C.