THE BACTERIOLOGY OF CREAM ated ITO eh Le, ad THESIS FOR DEGREE OF M. S. QRS ww tC): 1915 a. wo” ao £ f ‘ ee poo “ CLA i | AL me A at & ce cof ¥ f fo C € The author takes great pleasure in exnressing sincere appreciation for the valuable assistance and encourage~- ment given by Mr. C. ¥. Brown in the pureuit of this investigation. T hm me . 4 BA , ao 4 QO ahe ‘ N' | - f ) f } ( : Bi {ies RIPENING A Liat ND : ID CRURNING ’ a THESIS THE BACTERIOLOGY OF CREAM RIPENING AND CHURNING. CONTENTS. PART I. BACTERIA IN CREAM, THEIR NUMERICAL COUNT, TYPES, AND THEIR ITINERARY IN THE MANU- FACTURE OF BUTTER. I. Introduction. II. Previous Investigations. TII. Experimental Work. (a) Methods and conditions. {b) Tabulated data. (co) General table giving average data. TV. Discussion of Data. (e) Types and mumbers of organisms found in the raw material and finished product. (b>) Adaptation of casein agar ani litmus lactose agar for mumerical and differential oount. (c) Catalase and reductase in cream and butter. (d) Acidity of cream and butter. (e) MioroSrganisms. (1) Description, morphology, etc. (3) Possible source. (3) Spores of anaerobic gas producers. 103555 (4) Coli-aerogenes group. (5) Ocourrence in the different stages of manufacture. (6) General significance. Ve. Some Methods of Control. (a) Pasteurization. (dD) Acidity. (c) Judicious use of the starter. VI. Conolusions,. PART II. FACTORS INFLUENCING THE RESISTANCE OF THE LACTIC ACID BACTERIA TO PASTEURIZATION. I. Introduction. II. Variation in Temperature During Pasteurization. TII. Thermal Death point of Some of the Organisms Surviving Pasteurization. (a) Method. (1) Comparison of. (b) Determinations of. (co) Constancy of the thermal deathpoint. IV. Protective Agency or Factors in Milk. (a) Higher thermal deathpoint due to physical nature of milk. Ve VI. VII. IX. (>) Influence of cream. (c) Influence of acidity. Subsequent Infection, etc. General Discussion. Conclusions. Bibliography to Part I and II. THE BACTERIOLOGY OF CREAM RIPENING AND CHURNING, PART I. BACTERIA IN CREAM, THEIR NUMERICAL COUNT, TYPES AND THEIR ITINERARY IN THE MANUFACTURE OF BUTTER. INTRODUCTION, Milk produced under ordinary conditions always becomes contaminated with considerable, frequently with a very large number of bacteria. These organsime come from var- lous sources and comprise a large number of species. Many of the species gaining access to the milk find condi- tions very favorable for growth and multiply more or less rapidly from the very start. By far the larger part of the bacteria found in ordinary milk gain access from exter- nal sources and it is due to the presence and growth of these that milk changes in taste and odor, undergoes the process of souring and various decomposition processes. In the separation the bacterial clumps are broken up and a large mimber of each of the different types present in the milk are carried over with the cream. Many of these cause the hes lei fa 14 BY neé S< bi o-% TO the same changes in cream-as they do in milk while others have avery different action. The difficulties and prob- lems involved in the handling of the cream for the manvu- facturer of butter are so numerous and complex that it is little wonder there is lack of uniformity in the finished product. Dairymen have realized that butter of the best quality cannot be made from cream that is produced under unsanitary conditions. The destruction of the natural, sweet, rich flavor of cream caused by the carelessness and neglect of the producer can be noticed in the butter. Scientific research has taken the matter into consideration and many interesting facte which have helped to make the manufacture of butter a science as Well as a great commer- Cial enterprise have been discovered. The purvose of this paper is to make a study of the bacteria in cream, relative to the numerical count and prevailing types, and to trace these types throuzh the dif- ferent steps in the manufacture of butter; also to study their significance in cream and butter with some measures of control. PREVIOUS INVESTIGATIONS, "Dairying is an art the success of which depends almost entirely upon the extent to which we succeed in controlling the various fermentation processes" is a statement made by Ernst Kramer (27). The fermentation that takes pnlace in the dairy industry is largely due to bacteria. It is hoped that a knowledge of the nature and characteristics of these single celle belonging to the lowest type of plant life will help to throw light on some of those problems which are at present so difficult for the dairyman to solve. EH. W. Conn (5) states that fresh cream contains commonly a very large variety of bacteria. This is especially so in cases of creameries where cream is collected from a wide territory, the number found under these circumstances being dependent in a measure upon the initial contamination and the age of the oream. | The older the cream, the fewer are the types of bacteria found through bacteriological analy- Bis. The total number increases as the oream ripens, this inorease being due almost entirely to lactic acid bacteria which overshadow other grovps. After churning many bao- teria are removed from the butter with the buttermilk and many others with the Wash water. The butter, however, contains great numbers of bacteria that do not find condi- tions as favorable as in the oream and that die off more or less rapidly. Their death is generally attributed to the small amount of moisture and oxygen present and also to the action of salt. v. ¥. Esten and C. J. Mason found the dirt that comes from the surface of the cow enormously stocked with bacteria. The contamination of milk by cow feces is the most objectionable as germs that cause putre- faction, cas oreduction and various diseases are present. The number of bacteria that can be washed from hay stems or leaves is sometimes as hizh as 76,000,000 rer gram. Hay is the source of larze numbers of bacteria in the barn and cow stable. The liguefiers from hay are the worst kind. The number of acid organisms on grass is mich less in per=- centage and total numbers than in oured hay. Acid organ isms are high in percentage on all grain feeds. 1. A. Stocking, Jr. (34) in his study of the germicidal property of milk states that "Bact. lactis acidi and Bact. lactic aerogenes played an important part in milk or cream, the former occurring in much greater numbers than the latter". In the study of butter by Sayer, Rahn and Farrand (23), it was noted that the group of which Micrococcus lactis var= ians, a liquefying yellow cvocous, is a representative, ocourred most frequently in butter. Staph. pyogenes aureus belongs to this group. Members of the group are often found in the udder of the cow and on this account their frequency is easily explained. Micrococous lactis aureus and M. 1. albidus were alse found present in fairly large numbers, both resembling M. 1. varians in certain characteristics. The non-motile rod most frequently found was Bact. lactis lobatum, a slow, liquefying, orange bacterium. Bact. lectis album and Bact. lactis sorimi were rresent in varying numbers. The motile rods were present in small numbers. 3B. lactis cochleetus and B. 1. pruchii were fcund severcl timee. Both licuefying and non-liquefying yeasts vere frequently isolated. Three molds were fovnd frequently, these being Oidium lactis, Pencillium glaccum and Aspergillus glaucus. None of the cr- canisms were regularly present in brtter excert Pact. lactis acidi. §Roseneu, Frost end Bryant (20) in their study of market butter found that the number of bacteria in butter diminished sharply with ege. The averece number of bacter- ia found per gram wae five million, seven hundred thousand. They found no rarticular relation existing between the nur- ber of becteria and any other constituent such as selt, moist- ure, etc. They fount B. coli present in butter in six out of twenty-five samples. They found, however, that it soon dies ovt in the butter. Streptococci were found present in over half of their samples. 2B. welchii ( B. enteritides spororenes ) was not found in any semrle of butter. The tubercle bacilli was twice demonstreted. The origin of the bacteria in butter is not always the same. Several types of organisms, lactic acid bacteria, M. 1. varians, Oidium lactis, etc. are present in almost’ all milk. Other tacteria are derived from the washweter, B. fluorescens liquefaciens frequently coming from this source. ". W, Fsten end C, J. Mason (7) grouped their udder organisms in order of occurrence es follows: M. 1]. acidi, M. 1. elbi- dus, M. 1. variens, M. 1. aureus end P. subtilis. Usually there is no harm coming from these, yet from 2 diseesed udcer, scre threat, tonsilitis and scarlet fever may te contracted o> through the use of the milk. B. enteritides sroregenes, an anaerobic bacillus sometimes causing epidemio diarrhoea has ite source in feces. Acoording to W. L. favege (22) and C, =, Marshall (15) - Staphylocooci and Streptococci gain entrance through the orifice of the teats of healthy cows and multirly in tre milk cisterns; they aleo come from manurial contamination, from stale milk left in dirty cans end frem cows suffering from mastitis. The streptococci are of considerable imrortance as they are fount in larcve nunbers but as yet nothing very definite is known about them. B28. coli and allied orcanisme are of importance as indicators of pollution. The rresence of members of the colon grour mey be taken as signifying pollution from manurial source. The source of Beot. lactic acidi according to ¥, M. Feten (6) is the saliva of the cow. W. Le Savaze (22), H. A. Hopper (18), C. FE. Marshall (15), M. J. Rosenau (19) and many others agree that in general the following include the sources of contamination:— Int ramam- mary, introduction during the milking process, from the cow's Coat, the udder, the teat, milking shed, milker, feces, uten- Sils, strainers, coolers, vottlers, artificial cleansers, some in traneit and much during the final distribution. THE EXPERIMENTAL YORK. Methods and Conditions. This work was carried out as nearly as possible under commercial conditions so that our results should represent an average oount and inolude the most common types of or= ganisms present throuzhout the process cf butter manufacture inoluding the raw and finished product. Source of Cream.— Cream was obtained from dairymen in the vicinity of the Collere Dairy. Obtaining samples.—- The samples were obtained from the College Dairy where butter is menvufactured according to methods outlined by McKay and Larsen (16). Samples (about 50 coc. each of cream, pasteurized cream, starter, ripened cream and buttermilk), were obtained with sterile 10 cc. pipettes and placed into sterile 100 cc. Erlenmeyer flasks, Analysis of these samples were mede immediately after col- lection. Four samples of butter, one before washing and salting and three when the butter was ready to tub were taken from the churn with sterile triers, the middle third of each trierful being then placed into sterile deep culture dishes. The sample taken before salting and one of the samples taken after were analyzed at once, the other two were stored at 40° to 45°F. and examined after seven days and one month resrnectively. Bacterial Content and Isolation.-~ In all steps except those concerning butter, 1 cc. of the medium was added to $9 co. of sterile physiological salt solution. This was then shaken and other dilutions made by introducing 1 cc. of this into another 99 oc. of sterile salt solution. Dilutions of 1:1000, 1:10,000, 1:100,000 and 1:1,000,000 were then plated out in litmus lactose agar and duplicates in casein agar. Butter was introduced into a small Erlenmeyer flask and placed into a water bath at 35°C. One gram (1.15 cco.) was measured by use of a sterile pirette into 99 oc. of sterile salt solution which has a temperature of 35° to 40°C. Thies was Well shaken to a milky emulsion and other dilutions made as above. Samples were plated out in litmus lactose agar and casein egar, the same dilutions being used as for cream. After bacterial counts Were recorde, many of the organisms were isolated and transferred from the agar plated to sterile nutrient bouillon. Litmus lactose agar used for plating was mede according to the rules adopted by the Committee on Standard Methods (Journ. Inf. Dis. Suppl. No. 1, 1905) to which 1 % lactose and 0.05 % azolitmin was added. Casein agar Was made up according to the formula given by S. H. Ayers (3). Catalase.- The presence of catalase was determined through the use of the gum guaiac test. A tincture of gum guaiac was made by dissolving a little of the powdered resin in alcohol. About 10 cc. of the cream was placed into ea test tube, shaken with s few drops of hydrogen peroxide, then two drops of gum guaiao tincture were allowed to run down the sides of the tube coming in contact with the cream but not being mixed with it. A blue ring apvearing within a few minutes ie considered vositive for catalase. Reductasee- The presence of reductase was determined by adding l- cc. of Schardinger's solution (190 co. of dis- tilled water, 5 cc. formalin and 5 oc. of alk. meth. blue) to 10 oc. of the medium to be tested, shaken to mix the color and milk uniformly and placed in a waterbath at 37°C. for half an hour. Decolorization is reductase positive. Moisture.- The moisture was determined by heating 10 ems. of butter in an aluminum oup according to the Ames test (16). The sample was re-weighed and the percent of moisture obtained by multiplying the loss by 10. Salt.- The amount of salt was determined by a slight modification of the Shaw test (25), the silver nitrate being of such a strength as to have 1 co. represent .001 gem. of salt in butter or 0.1 % when 1 em. samples are used, potas— sium chromate being used as an indicator. Acidity. The acidity was determined by titrating 10 oc. of the medium diluted with distilled water by the addition of N/10 NaOH and recorded as peroent lactic acid. | Fat.- The percent of fat vresent was determined by the Baboock test (35). 10 Spores of anaerobic gas producers.— The anaerobic spore producers are tested for in every step of butter manufacture in the following way:- The medium to be exan- ined was placed in tubes; 1 co. in the first, 4 0c. in the second, 8 cc. in the third and 13 cc. in the fourth tube. . Enough sterile milk was added to the firet three tubes to make each approximately 12 oo. They were then heated in a water bath at 80°C. for ten minutes, cooled and the med- ium covered with one-half to three-quarters of an inch of sterile liquid paraffin to exclude the air. The tubes were incubated at 37°C. for two days. An abundant produc- tion of gas, the cream being torn and often thrown to the surface of the paraffin and coagulated masses of casein shows positive for spores of this group. Coli-aerogenes groupe- The presence of this group was determined by the inverted vial method (17), 1 oc., 0.5 co., dilutions of 1:10, 1:100, and 1:1,000 of the sample being ysed. The production of gas in dextrose broth was consid- ered positive. Peptonizers.- To determine the number of peptonizing colonies, the casein agar plates, after counts were record- ed, were flooded with N/10 laotic acid. The action of the lactic acid is to precipitate the casein in solution which produces an opadue white medium except about the peptoniz—- ing colonies where the casein has been dissolved by the pep- tonizing action of the bacteria. Colonies surrounded by a ll Clear zone and thus set off from the rest of the solid white medium are considered veptonizers. Acid Organisms.— The lactics were determined by direct count of litmus lactose agar plates. They form either a distinct boat-shaped or round colony, this being small and of a distinct reddish—-pink color. Inert and Indifferent Organismes.— These were estimated by difference, DISCUSSION OF DATA.» Types and Numbere of Organiems Found in the Raw Material and Finished Product. In our study of the bacteriology of cream ripening and churning, we have included in the following four groups of organisms:- First, those organisme which belong to the starter type; seoond, those that produce acid and gas; third, those that cause peptonization and putrefaction, and fourth, those that belong to the inert or indifferent group. By far the most important part pleved throughout the rivening and churning is taken by the starter and gas producing type of bacteria. The former is resvonsible for the ripening of the Cream, giving it the thick, even, clistening arrearance and 4s Te aly A pete Tht FS A ee een Fo nee OO OE rr are ne Oe 8 tm ne 18 good flavor which are essential for good butter. Some men bers of the gas producing type cause much. of the trouble that is often experienced in the manufacture of butter, as the formation of gas, bad odors, bad aroma and many other defects, causing the buttermaker moch concern. Next comes the peptonizing and putrefactive group which is a constant source of trouble and which is present in greater or smaller numbers in all cream. They are mostly rods, motile or non motile, capable of growing rather rapidly in cream and of performing marked changes, The casein in the cream is lique-— fied and digested with the formation of various products to Which are due the undesirable and offensive odors and the bitter, stale and rerhaps, the fishy flavors. Some products produced by members of this group are poisonous. The fourth groun, inert and indifferent organisms, does not affect the Oream or butter in any merked way, neither changing percepti- bly the chemical nor the physical condition of the medirn, but serves to swell the numerical count. According to many investizators, the number of becteria in "Top Milk® (cream layer) is moch greater per cc. (2) than the number found in "Bottom Milk" (skim milk). In separa- tion, the greater number of bacteria per unit volume leave the separator with the cream. The average number found in cream was 3,640,000. This great mumber was mede up of groups of all kinds, but members of the gas producing and peptonizing crours vredominated. In this first stave they T, GPAPEIC S¥PICE PFORFSFNTING THE PIFPECFNT TYPrs. ( % of total flora ) Qo & 8 ° 3 & o > s S So 9 8 o 8 Cream | ' i ' . ‘ Pins 7. Cream f / Cream @ 8 rm -\y QqQ Burter\mig\® a lo , > x a . bY en ry) ch rn h n 2 x Buller : y Butler apler et, age o) /ewk. 20 are more numerous than either the atarter tyne or the inert and indifferent zrovur. Pasteurization s~reatly reduced their numbers. It is in the nasteurized crerem that the starter grour of orcanismse is first ziven the leed and in consequence of the addition of the starter is able to keep its numbers far above thet of the other tyres rresent. According te the bacterie] ceunte given in ovr tables, it is seen that reste eurizetion kills $8 % of the bacterial flora. The ces rre- ducing and reptonizing tyres are esnecialily affected by reet- eurizstion, the relative rercentrere to tre total flora being recuced two-thirds. Tn tre addition of ~ure culture starter we heve added menv desireble starter orcanisms for every sur- vivins orranism in the ractevrized crerm and es a direct re- silt we heve € sreat percentere of the tctal flora of ripened Cream nece un of the starter type. It is also arrparent fron our table that © lerge number of the bacteria rresent in the rirerecd cream ere removed after churning with tre tuttermilk. Many more sere washed out of the tutter during the weshins erd workins rrecees. Eutter as a finished product ccntains a little rore than one-tenth ac many orzanisme ee did the ri~ened crean. It seems therefore entirely rlavsible to assume that the other nine-tenthe are removed with the tute termilk and the Weeh weter. Three-cquarters of the becter- Jel flora of fresh tyutter are the desirable eterter tvre of tecteria. Uron these, sc: it end lov tenreratures have a de- leterious effect end as ae reevlt they die off reridly at Ol FIC, II, GRAPHIC SZE™CE PPPPTETNmTNG TIP IePrP CF MICTOAPCANTONS IN TED MANUFACTURED CF PITmEn, { pverege of 13 exreriments ) ( Based on the loz of & ) 20 : Ster7er /00 . Aipened Cream Soo} urter milk a < Creams /0% Starkey Buller from churn 6 Salted Butter Buiter after 7deys in alovage Buiter after one »wenth 177" /6 elo roge to I - a) a? firet, then mere sradvelly until after some time in stcrace their numbers as compared with tke mumbers of the other tyres ile very much reduced. The relative nercentece of the Inert erd Indifferent type increases with the time in stor- age. The averese number of tecteria found elive in fresh butter was 56,00C,0CC. Storece and salt, however, seem to have co sreet eae deleterious effect on the butter flora that after storase of one montn, less than half of this mumber remain alive. Adartation of Casein Azar and Litmus Lactose Agar for Numerical end Tifferenticl Count. The two tyres of media which are of greet importance in Getcermining tne tyres of bacteria rresent are casein azer end litmus lactose acar. Throuzhout the wrole rrocess, the bacterial counts on the two media rere very clcee to- sether. The casein agar counts, nowcver, were cenerally lower than the litrus lectose counts. One cf the disedvan- tages of casein agar is thet it must te allowed six to eight days for incubation, wnerees litrus lactose azzr needs cnly three to five days vier the common temmereture of 20° to 35° C. is used. After total counts and isolations were mece, the casein ager rleted ere valuatle for the detection of reptonizers. The work with casein azar beers out in gener- al the work of Ayers (3) in thet the casein media fevored 23 the inert, elkeli forming and reptonizing vrours of bacterie, while acid forming orranisms were favored end mere readily distinguished on the litmus lactose acar. Catalase and Reductase in Cream and Butter. The action of the reducing end oxidizing enzyres was studlec in order to determine if rossitle the reason cf the diserrearance or inactivation of catalase and reductase found in fresh creer. In Table II it is to be noted that neither catalase nor reductase wes detected in rire starter, rinened crear and buttermilk from the churn. Poeitive re- sults were obtained for both oxidases and redvctases throvcre out the rest of the procece. It is known thet catalase is concentrated with the creem wren milk is centrifuged (14). It is also believes that ordinary lactic acid bacteria ere not able to produce catalase while meny other typical milk orzanisme are able to produce it in large quantities. The amount of catalase is measured in most cases by the amount of oxygen liberated in two hours, at 30° to 37°C. from a definite measure of standard hydrozen reroxide mixed with a definite volume of milk, but in our work the previously de- scribed color method for detection could be used with great satisfaction. we found that the cetalyvtic activity is in- fluenced by tie arount of hydrogen peroxice added. When either too much or too little is edded the test is either a4 retarded or necative. According to Koenig and Leon (13), Oatalese exists in milk rertly free end partly combined. The combined catalase does not act until after twenty-four hours have elapsed but upon addition of a weak alkeli the combined catalase becomes at once active. We know that the atarter does not ordinarily produce catalase but even if the crganism could rroduce catalese, the hich acidity develored by it would either temperarily or permenently in- activate it. Ripened cream and buttermilk are also very high in acid and there is little doubt that this factor is the exrlanation of the nevative results obtained in these steps of the ripening anc churning process. The catalace, however, is not permanently inectivated as a direct result of the acidity as positive tests for it are obtained in the butter which mav be explained by the low acidity of the bute ter. To further rrove what the direct action of acid would be on catalase, to tubes of fresh cream, which gave positive tests for catalase, wes added N/1 lectic ecid until the Cream was almost visibly curdled. On repeeting the tests for catalase, negative results were obteined. The cream was then brought down to a normel acidity ( 15° Fullerts scale ) by the addition of N/1 sodiwm hydroxide and then positive restlts were obtained, thus confirming our eaesump= tion that high and low acidity are factors in determining mecative or positive results in catalsse determinaticns. Reductase ie similarly affected by acidity, it being inectivated by the addition of lactic acid and its activity restored by the addition of a week alkali. This causes us to state that reductase as well as catalase is only tenpor- arily inactivated ty a hich acidity of the medium in which it is determined and rey eazvain be activated by the addition of weak alkalies. Acidity of Cream and Butter. as we note the different sters vaseed throuzh in the ripening cf the cream and the churning process, it is ap- rerent that great differences in the acidity exiet. The Original acidity does not differ zsrestly from that of past- eurized cream. The starter, having a high acidity, raises the acidity of the cream to a considerable heizht. Butter- milk is also possessed of a hish acidity. Butter, as a finished product, possesses a= comparatively low acidity. In storage the acidity of the butter seems to change but very little. Acidity not only plays a great part in the Quality of the butter but also shows its influence upon the keeping quality. Although investigators believe sweet Cream to possess a$good keeping quality as butter made from ‘ripened Cream, it is the orinion of many that butter made from sweet cream does not possess as good keering qualities as butter wece from ripened cresr. However, acidity ise by no means the only factor to be considered in the determina- 26 FIG. III. GRACEIC SKETCH REPEFSENTING CEANGES IN ACIDITY Tibleschic erd) 0 | Shorler Buttermilk +40 > Hipened Cream IO =m 2 e Creamel oJe Ongina/] Cream Sarler Pat. Cream 40% . v0 = Lited Gutter 020 cutter} yor Buthy no he Lor me 7s ehurr Bythhy i" v707 AGE 77doys e/e BT tion of the keeping quality of butter. Acidity also affects many other changes throughout the rrocess as enzymatic action, rhysical structure, chemical equilibrium, etc. Wicoroorganisns. Descrirtion, Morphology, etc. Table II gives counts of organieme obtained throughout the whole process of cream ripening and churning, showing also the percentage of the different groups of organisms present. In every step of the ripening and churning, organ- isms were isolated from plates and planted in bouillon. From the bouillon culture agar plates were made and pure cul- tures assured. These were then transferred to litmus milk bouilion, agar slants and gelatin stabs. The characteristic growth of each was then recorded and some of the more preva- lent organisms were identified. First, the bacteria were grouped according to the morphology into coocus, bacterium and bacillus, yeast and mold. Next they were subdivided into liquefying and non-liquefying and into acid and non-acid @rours. These were then divided into spore-forming or non- spore-forming organisms. In identification, "The Classifi- cation of Dairy Bacteria" by Conn, Esten and Stocking was used and it was comparatively easy to trace the more preva- 68 lent organisms to their proper group and name. In the following tables the morphology of the organisme is shown. The number of times that each of the organisms was found present is also shown in the tables. The organism present in greatest numbers and the only one present throughout the different steps of ripening and Churning was Bact. lactis acidi. It occurs in the smallest numbers in storage butter. Other organisms present in greater or smaller numbers were identified as Micrococcus lactie aureus (80), Micrococous lactis variane (21), Micro- coccus lactis albidus (19). These three belong to the cocous group and have very slight differences by which one may be distinguished from the other. M.i1.varians lique- fies gelatin while M. 1. aureus does not liquefy it. Their cclor varies from orange yellow to white. M. 1. albidus may be designated as a white variety of M. 1. varians. Very few rod-shaped bacteria were found. 3B. coli (126) was perhaps the most prevelent throughout but was not found in butter. B. fluorescens Jiquefaciene (122) was present in all steps except in pasteruized cream and butter- — milk, Only twice out of the twelve experiments did it occur in fresh cream. 8B. subtilis (103) was found present in all stages except in the finished product. Yeast and torula, both of a liquefying and non-liquefy- ing type, were found present and as we have at hand no good 33 guide for their classifications, they were grouped only as to their ability to liquefy geletin and according to their color. Yeasts and torulae were especially prevalent in the original cream. The part they play in the manufacture of butter is minimized by pasteurization since their num bers are greatly reduced, A few molds were found but these with one exception, Oidium lactis, were discarded as they were thousht to have little influence in butter. Possible Source. Under the subject of "Previous Investigations" the poseible source of organisms in general has been fairly Well covered. It is, however, necessary to give a few of the possible sources of the specific organisms occurring throughout the process. Bact. lactis acidi is generally believed to be present in the saliva or on the tongue of the coQwe It has also been found present on plants but only where some bovine has had access. The organism is spread about the dairy in insufficiently cleaned or unster- ilized utensils. Hair from sides of the cow and other ma- terial which at some time has been moistened with saliva is responsible for its dessemination. The gae formers, B. coli communis and Bact. lactis aerogenes have as their source usually fecal matter, dust or dirt from the animal's 34 Coat or the upper layer of the soil. The microococol are commonly found in the udder. 5B. fluorescens liquefaciens is found in the soil and water, gaining entrance to butter by washing the utensils and the butter with an impure water. The bacterial content of the washwater is a subject to be given consideration since many harmful organisms are from this source. Filtering and heating of water has been sug gested to avoid this source of contamination. 2B. subtilis is added to the milk from hay, dust of barn and soil. In general the outside contamination would be due to the milk- er, air and dust of the steble, the milk pail, water supply, cooler, cans, transportation, ripener, churn, buttermaker, package and distribution. Spores of Anaerobic Gas Producers. Throughout our work, teste for B. enteritidi sporogenes spores ( B. welchii ) were made by the method previously given. This group of anaerobio spore producers is frequent- ly found in milk and cream. This would lead one to suppose it would be found in butter also. In all the steps of rip- ening and churning, this organism was found present with the exception of butter. It is indeed surprising that it is not found in the finished product yet the fact remains that many investigators who have worked on the same subject have 35 hed negative results, for which no reason has been given. Coli - aerogenes Group. This group includes a considerable variety of organ- isms. According to Hastings (10) they are classed among the facultative anaerobes and differ greatly in morphology, cultural characteristics and amounte of their by-products. He also states that they are to be found in every sample of market milk in varying proportions. Tests for this group were made during each of the different steps of butter manu- facture. The same results were obtained in these tests that were obtained in the B. enteritidi sporogenes teste, the organism being present in every stase except in butter. For this no reason could be given other than that the but- ter contains less than ten percent of the milk serum present in the cream, unless the agitation during churning or plas- molysis of the cell due to the added salt kills the organ- isms. Reinman (18), Jensen (13), Rosenau, Frost and Bryant (20) and a number of other investigators have found this group in butter though always in rare instances whereas some authors have made the group conspicuous by stating that they have never found it present in butter. The only reason, therefore, that may be ascribed to the rare appearance of the group must be the one given above. QS ae IIe XI a n It IX Ve ICH AN ORGANISM WA II aa 39 TABLE Churnin FOUND IN THE DIFFERENT CHURNING EXCERIMEUTS. Organ- T Il tr 1 iem. No. of | THE NUMEER OF PLATINGS IN ONNMNONMNONOHONKNHODOADAAINMNEROOON OO ONNOOMANRNRANOTAOCDNOTARNOCOONS Le OTVRNRNNOOOTFOODOWVDONOOMAdACTW WKNMNOONM ONO ODO ONMNOMOONOOANANNOWDMOAANRNIDIOIOON NVONRVOVDVVCONO DO OCOCKOCOCWMDNMN AVON’ NMC x ONVHARVANRVNYASTUNODODOOMWORNOOCOMIOO OMNRNVOOFHOUNUVIONRCARDOOVIVOHAMOOIOHASY ODO ACFANMOONMRNONNANVHOWDANONRNORVIODONO ww OOSOSORKRRVOMOAAIRIMMOMMDARIMHAMOOOO >HOORNNONMNNOYRUNRVMADOONIDVIIVOIDNANOONM OM RNOONANNNODCODOWVCOVODMOMNDIOIOMNON SS AMNAONMOHAMOCODWODOVOOCOWOOONO Or Frio PACINO MO DAO OD APO OD ALO LOUD AO UN O ANMUMRAEMMNODOKRE DOMOOrtiriVNNtN aAncdtcicteiet Just before the final typewriting that part of Note:- Table IV which contained the results of Churnings Nos. Iv, V and VI wes irretrievably lost. 40 The Occurrence of Microorganisms. In the preceding tables the occurrence of microorgan- isms with the excertion of yeasts and molds have been re=- corded. Yeasts were present in every churning in varying numbers. They are most prevalent in the cream before it has undergone pasteurization. Molds were found in but few cases with the exception of Oidium lactis. Most of those found were in butter which had been kept in storgae for one month and undoubtedly had some relation with the condition of the butter at that particular time. Members of both the Penecillium and Aspergillus types were rerresented but neither a morphological nor physiological study was made of them. Table V also shows the number of platings in which en organism was found in the different churning experiments. General Significance. In the first rart of the disoussion we have taken up the influence of certain tyres of bacteria found throughout the process of ripening and churning. It remains, however, to discuss the general influence of all types found present throughout the work, laying erecial emphasis on their effect on the finished product. Sweet cream butter as a rule soon undergoes deleterious changes especially when made from un- 41 pasteurized cream. On the other hand, butter mede from oream which has been properly ripened differs greatly in flavor from that prerared from sweet creem and is believed by most investigators to possess better keeping quality than sweet cream butter. Butter meade from cream which is fresh and sweet when it is delivered to any creamery is of higher craie end has better keerine quality than butter mace by the same buttermaker from cream which is old and BCure The acid fermentation of the rirening process gives intensity and character to the flavoring compounds. It should be our sim to eliminate to a great extent undesirable organisms as those of the coli-aerogenes group, etc. and to allow the desirable Bact. lactis acidi tyre to predominate at all times. This of course means that we should use the pasteurization process to help eliminate many of the undesir- ables and employ a good starter which is responsible for the desirable flevor and uniformity in quality of the finished butter. According to Herter (11), the coli-aerogenes group sometimes induces excessive fermentations of lectose and other sugars with the sroduction of irritating acide ( es- pecially lactic and acetic ) and at the same time liberates an excessive amount of gas. They cause a sharp taste in the finished product very noticeable especially when butter is made from unpasteurized cream. Members of the sroup are present frequently in large numters due to fecal contan- ination. The peptonizing bacteria or casein digesting 42 ‘group digest vasein either with or without coagulation. They often develop strong putrefactive odors and are in many cases the cause of abnormal flavors in butter, but are in general kept down to euch small numbers that their influence is generally not heavily felt. The yeasts were present in all churnings and although we know very little as yet of their influence on the keeping quality of putter, yet we do know that "oily" butter and certain abnormal flavors are due to them. Moldy butter due to the growth of certain molds on the lining of the butter tube or the paper in which the butter is wrapped, is also a factor to be taken into consideration. Oidium lactis is always present in butter and seems to decrease in number with the age of the butter. Hastings (10) states that itself is not a good substratum for mold growth. The remedy recom mended or rather a preventative is the paraffining of but- ter tubs and the heating in hot water of the paper linings and Wranpers. SOME METEODE OF CONTROL. Pasteurization. In any discussion rertaining to butter manufacture it is almost a necessity as well es a matter of great interest to fullv discuss rasteurizaticn and and its real influence on the finished product. About a quarter of a century azo, Storch, the notei Danish scientist isolated certain bacter- ia that were necessary in the rirenins of cream. The best ' results in ripening were not elvays obtained when these i- solated cultures were edded to cream already crovded with a great number of bacteria uiu On tuls acccunt he started to work to obtain a cleaner field for his cultures by destroy- ing by means of heat the organisms that already existed in Cream. This process in later years, vreatly modified, re- ceived the name of pasteurization. Ye now define rnasteur— ization as the yrocess in which milk or cream is heated to a sufficiently high temperature to destroy a portion of the bacteria and then is rapidly cooled to prevent growth of the surviving organisms. The two main purposes of the process are to destroy all pathogens end to imnrove the quality of the product. In the process of buttermaking it enables the buttermaker to eliminate undesirable taints due to many bdac- teria in cream and to produce a more uniform flavor and qual- ity by allowing the starter organisms a good, clean field. It also enhances the keeping quality of butter. As a con- trol, efficient pasteurization first of all removes all dangers of pathogenic organisms. Fitgze (9) lays partic- ular stress upon peptonizing bacteria, most of which form spores. The spores survive pasteurization and would then have a field free for growth and activity. Ayers and Johnson (4) state that many inert and alkali forming organ- 44 isms resist the heat of pasteurization. They have also found strains of Bact. lactis acidi, B. coli, B. lactis aerogenes and tyres of strevtococoi which survive pastesur- ization. It seems, therefore, that pasteurization when used alone es a control would not accomplish the desired eni. Acidity. Acidity is taken up as a factor whicn may control the type of orcenisr. The tolerance of several organisms for | acid, on agar ana in milk was tried and it was found that all of the organisms used in the experiments could grow on a medium containine .4% lactic acid. The organisms used were M. 1. varians, M. 1. albus, M. 1. albidus, B. subtilis, B. fluorescens liguefaciens, Oldium lactis,a liquefying and a non-liquefying yeast, ( 59, 37). The results were sim ilar in both milk and agar. It is thus apparent that an acidity of .4% is not sufficient to prevent growth of the common types of organisms existing in butter. “ 45 Judicious Use of the Starter. It has been previously mentioned that when a consider- able number of species of bacteria are present in milk or cream, not all of them are able to multiply at the same rate. Most species of bacteria thrive best on a neutral or slightly alkaline medium. In nearly all milk, however, as it comes from the milk pail, there are a number of lac- tic acid orsanisme which soon produce sufficient acidity in the milk to check the srovth of a number of species of bec- teria which do not thrive in the presence of acid. On the basis of antagonism of mixed cultures, it is hoped we can establish a control of the organisms that ocour in butter. Firet, in considering tne starter organism, let it be re- membered that a culture of Bact. lactis acidi noted for its characteristics should be used. This should have good vi- tality and fermentative power, self preservation, flavor producing properties and be able to form a curd of uniform consistency with no presence of ges. We know that of the creat number of organisms present in the original crean, only two percent or less are alive after pasteurization, 1. e. ninety-eizht rercent are killed by the heat of efficient pasteurization. In the addition of ten percent starter we add 5,000 or more of the beneficial starter type of bac- teria for every organism in the pasteurized cream. This pleurality gives the starter organism so zreat an advantage 46 that under rroper conditions they will develop in the ripen- ing cream the desired flavor and aroma. At the sane time the etarter orzanism also exerts an action strongly antagon- istic to the members of the group of peptonizers and rutre- fiers crowing in the same cream. When the ripening is well under way, tnis group is almost vonpletely held in check and their harmful changes ere arrested, The inert or in- different group are really of little importance except in so far es they increase the total count of bacteria. It is easily seen that a judicious use of the starter can do much along lines of control. There is, however, a danger of allowing the starter to use its influence for too long @ duration, giving as a result over-ripened cream which im- parts to the butter a sour cream flavor and impairs the keeping qualities. In the majority of cases the starter cannot do ite full work of control if the cream which is to be rivened is not vreviously rasteurized for it usually con- tains so large a number of injurious bacteria that the fav- Orable influence of the pure culture is greatly diminished. It is thus seen that a good starter and pasteurization rep- resent the most efficient control of mioroorzanisme in cream and in butter. a7 CONCLUSIONS. 1. The orzanisme in butter are for the greater part those which are present in cream. | &. Neither the buttermilk nor the freshly churned butter taken from the churn before the buttermilk was drawn contains per unit volume as many living bacteria as the ripened cream. The average from our data shows that about 30 % of the organisms were killed in the churn, doubtless due to mechanical agitation during churning. 3. The process of washing and salting removes fifty percent of the microorganisms. 4. Plates made with casein azar give a total count a trifle lower than litmus lactose agar and such plates aid in the detection of the veptonizing organisms. Litmus lactose azar is well adapted for the detection of acid pro- ducers. 5. Positive tests for catalase and reductase were ob- teined in the oream and butter but not in the starter, ripened Cream or buttermilk. Negative tests are due to a temporary inactivation by high acidity. 6. Spores of anaerobic gas producers and members of the coli-aerovenes group were not detected in butter although they were present in greater or less numbers in crear. 7e Bact. lactis acidi is the predominant type through- out the manufacture of butter. Other organisms appearing frequently in our samples are M. 1. varians (21), M. 1. aureus (90), M. 1. albidus (119), S. 1. fulvus (19), Bact. lactis album (2), B. coli (12€), liquefying and non-liquefy-— ing yeasts, Oidium lactis. 8. The growth in milk of orzanisme frequently found in dairy products was not entirely prohibited by the pres-— ence of 0.4 % lactic acid, 49 REFERFICES, (1) Aikman, C. I. 1909 Milk, Ite Nature and Comrosition. (3) Anderson, J. F, 1908 Milk and Ite Relation to the Public Health in Bul. 56, Publ. Health and Marine Hosp. Service. P. 739. (3) Ayers, S. He 1911 Casein Media Adapted to the Bacterial Exam ination of Milk. Bur. of Animal Industry. 28th Annual Report. (4) Ayers, ©. H. and Johnson, W. T. 1911 The Bacteriology of Commercially Pasteurized and Raw Market Milk. Bul. 126 .of Bur. of Animal Industry. (5) Conn, H. 1905 Bacteria in Milk and its Productea, (6) Esten, ¥. M. 1909 Bact. lactis acidi and its Sources. Bul. 59, ppe 5-27. Storra Sta., Conn. (7) Esten, ". M. and Mason, C. J, 1908 Sources of Bacteria. Bul. 51. Storre Agr'l Exp'te Sta, Conn. (8) Fettiok, 0. 1999 Quantitative und Qualitative Untersuchungen tiber die Bacterien, Hefen und Pilze der But- ter und tber den Einfluss des Kochsalzes auf dieselben. Centr. ff. Bakt. II. Pd. 322, (1909) Pe 386 (9) Flugge, C. 1894 Die Aufgaben und Leistungen der Milch Steri- lisierung gegeniber den Darmkrankcheiten der Saigling. Zeit. f. Hyge Vol. 17,p. 273. (10) Hastings, E. G. 1911 Marshall's Microbiology, P. 310. (11) Herter, G. 1908 Common Bacterial Infections of the Digestive Tract and the Intoxications Arising from Theme {12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) 50 1908 The Cream Suprly. Bul. c09. Univ. of Cal. Jensen, 0. 1902 Studien ueber das Ranzigwerden der Butter. Centr. f. Bakt. II. Bd. 8, pe ll. Klein, L. Ae 1912 A paper on Catalase and Reductase presented at the Conference for Veterinarians ab Ithaca, N. Y. Cornell Vet. Vol. II. Marshall, C. Fe 1911 Microbiology. McKay and Larsen. | | 1913 Principles and Practice of Butter-laking. Muir and Ritchie. 1899 Manual of Bacteriology. pe 86. Reinman, R. 1910 Untersuchungen ueber die Ureachen des Ran- ziewerdens der Butter. Centr. f. Bakt. II. Bd. 6. Rosenau, M. J. 1912 The Milk Question. Rosenau, Frost and Bryant. 1913 A Study of Market Butter of Boston Jour. of Med. Res. Vol. XXX. No. l. Rahn, O., Brown, C. ®. and Smith, L. M. 1909 Keeping Qualities of Butter. Tech. Bul. No. 4, E. Lansing, Michigan. Le Savage, *¥, Milk and the Public Health. 19138 Sayer, Rahn and Farrand. 1908 The Keeping Qualities of Butter. No. l. Miche Exp't. Sta. Tech. Bul. Stocking, W. Re Jre 1907 The So-called Germicidal Property of Milke Bul. 37. Storrs, Conn. Van Slyke, L. 1912 Modern Methods of Testing Milk and Milk Products. 51 (26) Wing, HE. E. 1904 Milk and Its Products, (27) Woll, F. ™. ( Quotation by Frnst Kramer ). 1904 Principles of Modern Dairy Practice from a Bacterlological Point of View. p.1. PART II. FAOTORS INFLUENCING TEP RESTSETANCH OF LACTIC ACID BACTERIA TO PASTEURIZATION. The ready acceptance of the fact that orgenisms survive the pasteurization process because of the great bulk of cream and milk used or because of their high thermal death- point, has deterred dairy investigators from discovering the real cause or protective factors which allow their survival. In 1913 J, J. Kinyoun (8) found in commercial pasteurized milk of Washington, D. C., high counts of colon bacilli and streptococci, and contributed their presence to dirty milk, inefficiently pasteurized. C.F, Marshall (10), in 1897, found that a large percentaze of the samples of milk pasteur- ized at 68°C. for 20 minutes, lonpered with the production of acid, althouch no true lactics were found on plating. Among those isolated were four non-snore-forming bacteria, three of which vere able to survive 80°C. for 20 minutes in bouillon and the fourth rithstood 70°C. for the same lenecth of time. H. BD. Pease (11) found that members of the ooli group are more difficult to kill by pasteurization than Bact. tubercu- losis. A number of streptococci and members of the colon group, whose thermal deathpoint is hizh, were isoleted by o3 Harrison (6) in 1905. NN. L. Russell (12) says - that to a large extent the lactic acid bacteria are destroyed by the pasteurization process. Gace and Stoughton (5) in 1906 found a strain of B. coli whose thermal deathpoint was 80°. and attempted to increase its resistance to heat by subcul- turing from the surviving few, but no increase in the resis-— tanoe was obtained. Many investigators have found strepto- cocci amd members of the Bact. lactis acidi end colon zroups that can survive at least 70°C. for 20 minutes. In 1910 Ayers end Johnson in their work on "The Bacteriology of Con- mercially Pasteurized and Raw Market Milk" (3) say that "tem- perature ( 62.8°C,. kept for 30 minutes ) would be sufficiently high to afford protection azainst pathogenic bacteria and yet would leave in the milk the maximum prorortion of lactic acid bacteria and the group proportions would be very similar to those of all grades of market milk", In their conclusions they also essume the fact that the souring of pasteurized milk is due to the development of lactic acid bacteria which on account of their high thermal deathpoint survive resteur- ization or which come from subsequent infection during cooling or bottlins. They elso claim that the thermal deathpoint of one lactic acid orvsnism which was isolated from milk was 74.4°C. in broth and 75.6°C. in milk when heated in Sternberg bulbs for thirty minutes. Theobald Smith (15) in 1899, working with Bect. tuvercu- losis states that the organism susrended in water, normal O4 salt solution, bovillon and milk ere destroyed et SO°C. while the pellicle which forms on milk during the exrost:re at 60°C. ney contain livinz orvanisms even after 60 minutes. The work of FE, lL. Pussell and F. G. Hastings (13) in 1902 con- cludes that the destruction of bacteria in milk by means of heat, derends upon the conditions under which the exposure was mace; the formation of a pellicle rrotects any orzanism within the pellicle. The resistance of bacteria in the sur- face membrane ie thovr-rt to be due not entirely to the lower- ins of the terperature, but is affected by the nature of the enolosinz memorane itself. In closed vessels as Sternbers bulbs they state trat milk offered no protection creater than whey or bouillon. ‘The opinion was expressed by A.®ollf (17) in 1908 that the lactic acid bacteria ( B. Guntheri ) past- evrized et 70°C. for thirty minutes ere protected by the heat forming arovnd the cell an acid coagulum of albumin. Heat resisting strains of B. coli were found in milk by T. Zelen- sky (18) who states that their thermal deathpoint is higher in milk then in broth and suczests that the rrotein and fat in milk act as a protecting factor. Nearly every investigator working rith pasteurized milk agrees that members of the B. coli and Bact. lectis acidi rrouns survive pasteurization and too that there exist members of these xrours whose thermal deathroint is sufficiently high to carry them throvzh commercial restevrization. weny ere of the orinion that the trermal deathvoint is the same whether determined in bouillon or in milk, especially if the milk is sealed from the air to vrevent tre formation of a surfece pellicle or membrane; while others have found that the ther- mal deathpoint of a number of organisms when determined in milk ile a little hiszher than in bouillon. The object of this work is to study: 1. The varistion in terperature durinz commercial rast- eurization ( holding vrocess ). &e The thermal deathpoint of some orzanisms surviving vasteurization. 3. Subsequent infection during cooling and bottling. 4. The protective agencies in milk and how they effect different types and strains of bacteria. The Variation in Temperature During Cormercial Pasteuri- zation by the Holdinzs Process. This work was carried on in coOperation vith the Colleze Dairy Department, a "Perfection" Pasteurizer ( 300 sal. cere- city ) being used. The Holdinz process of rastevrization ( 145°F, for 30 minutes ) has been used by the department with good results. The determinations of temperature in this ex- periment was mace in oream vesteurized by this process. The temmerature at the four corners end in the center of the rast- eurizer was taken at one minute intervals. The resulte in Table I show that the temperature did not vary over two de- grees durins the entire process. The slight variation in temperature may be to the credit of the special tyve of past- eurizer used as the srirals stirred the cream thoroughly. It is evident from the data obtained that the duestion of bulk does not anewer for the rrotection of the organisms throush uneven distribution of the pesteurization tempera- tures in a restevrizer of this tyre. TABLF Te Variation of Temperature During Pasteurization. Min. Corner I. Corner II. Corner III. Corner IV. Center. 0 145° 145° 144° 144° 145° 1 144° 144° 145° 145° 145° 2 145° 1459 144° 145° 144° 3 145° 145° 145° 144° 1450 4 144° 145° 144° 146° 145° 5 146° 145° 145° 144° 145° 6 145° 144° 146° 145° 1469 ” 145° 145° 145° 145° 145° 8 1459 145° 145° 145° 144° 9 144° 145° 145° 145° 145 10 145° 145° 144° 144 144° 11 144° 146° 145° 146° 1456 12 145° 145° 144° 144 144° 13 146° 144° 144° 144° 145° 14 146° 1459 145° 1448 146° 15 1450 145° 146° 145 144° 16 145° 145° 145 145° 145 17 145° 145° 145° 144° 146° 18 145° 145° 1450 145° 1450 19 145° 145° 145° 145 144 20 145° 145° 146° 145° 146° — -_-_—S- oe => oo The Thermal Teathrcint of ferme cf the Organisms Surviving Pasteurization, Meti.od. By thermel déathroint we meen the lowest terpereture which caused Ceeth to vegetative forms on exrosure fer ten minutes (4). The bacteria rere srown in nutrient broth fer twenty-four Lours and then one cc. of the culture wee rlaced inte tubes of 10 cc. sterile nutrient troth. Duplicate tuces were heated in e waterbath for 10 minutes at definite temperatures renzing from 50°C. to 78°C, each wet being ex-— rosed to a tempereture 83°C, nicher than the one preceding. The temperature vas net ellowed to vary one-half degree C. The tubes were then quickly cooled below 30°C., one oc. incc- ulated into 10 cc. of sterile litmus milk and placed at crti- mum temperatures ( 30°C. for Pect. lectis acidi and 37°C, for B. colie ). The organism wes not considered killed until it head teen incvtetec for 5 dave and neither tube showed crowth. Other tubes vere treeted by the sene method as abceve enly tie time of heating to which they were svtjectea was lengthened to 30 min. so that it would correerond to the length of time used in commercial pastevrization. Another tyve of technic on the effect of a lack of oxygen curing determinaticns of a thermal ceati;oint wes tried throuch the eddition of a layer of rareffin one-half inch in depth on tor of tie mecie in which the therrel deathroint was to be determined. Cream: 58 ana skirmed rnilk were used ae the mecia, inoculeted with the different cultures and covered with paraffin; then the ther- mal deathroint was determined. Later parallel tests vere run by the use of the Cternberg glass bulb method (7). The tube wes Filled by werming the tulb slightly to expand the air and then the stem "ce at once inserted into the bacter- lal susrension which ie drawn by suction into the bulb as it cocle. The neck was then sealed es it cooled. The ineat— ins wes cone witn the bulb lLelc suspended by «& wire and com pletely immersed in a weterbath. After teing cuickly cccoled below 30°C. the contents were emptied into a tube of sterile Litmus milk and this set aside et optimum temveratures. Comparison of Methods. The three methods described above, namely, the open tube, closed ture and paraffin tube method, were studied for the relative merits of each and to see whether one would heve eny greater or less influence on the thermel deathpoint of microorganisms. To show the influence exerted ty the differ- ent technic, cream and skimmed milk, inoculated with the lec- tice Nos. II A, IV A and Knop, B. coli V. and Bact. tulgaricrr were oorparede In the following table it is seen that no difference which would influence results was found in any of the trials. TAELF II. Comparison of Cpen and Closed Tubes. — oe oe C oO -RD =—_— ee ee —— Names of Orgenisms Open Tutes Closed Tubes Pereffin Tubes Crean Wilk Crean Wilk Creem Milk oo. aS a ee inn Bact.l.eacidi No.IZ A 69°C 60°C 89°C 609°C 89°C ec% Pact.l.eacidi No.IV A 71 63 721 63 71 63 Bact.l.acidi Knop 721 63 63 63 69 63 E. coli No. V. 73 «65 73 «85 73 «63 - -— @ ae = Determination of Thernal Deethroint. In this phese of the work non-srpore forming becteria were used. The thermel deathroint of twelve cultures of Bact. lectis acici isolated from pasteurized cream and butter, four cultures of B. coli isolated from pasteurized cream end one of Pact. lactis citronis isolated from butter were studiec. The variation in the thermsl cesthroint of the different crgem isms proved to be grest, verying from 56°C. to 78°C. Name TAPLE III. Thermal Veathroint. Organism __. LO mine «0 min. | Bact. lectis ecidi I 58°C 56°C, n " " II 72°C 66°C. " n rn III 76ec 74°C, fn n " Iv 6250 5600. n n n v 56°C 52°C, " fn n WT 760 7200, " n n vTI 78°C 74°C. " n n VIII 68°C 66°C. n fn n I” 66°C 6200. n " " X 72°C 6evc, n " " vI 74°C 760C " " " XII 64°C 620C. Pacillus coli T 74ec 70°C. " rn It 60°C 58°C " n ITI 70°C 66°C " " Iv §60C 52°C . Ract. lectis citronis 64°C 62°C 60 We observe that some bacteria whose thermal death point is below the pasteurization temperature, survive ypssteurization; due to some agency or factor which milk brings into play and which the nutrient broth is unable to menifest. TABLE IV. Constancy of Thermal Death Points. i 2 5 4 2 Name of Organism min. min. mine min. min. 10 20 10 60 10 20 10 «80 10 «0 Bact. lactis acici I 56 56 58 56 58 S56 5&6 54 58 54 " " n II 7868 76 66 78 66 70 68 786 68 III 76 78 74 78 76 74 74 74 76 74 IV 60 56 68 56 60 56 60 56 60 5&6 V 56 53 56 50 56 538 S56 Se 56 Fé VI 76 70 78 70 74 78 76 78 76 7@ VII 78 76 7€ 74 76 74 76 74 78 76 VIII 68 66 70 66 70 66 70 66 68 66 IX 64 62 64 60 64 60 66 60 66 60 X 70 68 70 68 70 68 70 66 70 66 XI 74 70 74 78 74 70 74 76 74 70 " XII 626 66 64 62 64 68 68 628 68 60 Bacillus Celi I 7470 74 70 74 70 74 70 74 70 " " II 60 56 60 56 58 56 58 56 60 56 " " III 70 68 70 68 70 68 70 68 70 68 " " IV 56 50 56 50 56 50 56 58 S6 50 Bact. lactis citronis 64 62 62 60 68 60 5654 60 64 60 sazsa332 32322323 3 2 2a3aa3ags3as3232 22 3 sz 323 2223232 23232 3 Constancy of Thermal Deeth Points. The constsency of the thermal death points was determined by obtaining five times at three day intervals the thermal death points of each organism formerly studied by the test tube method heating for both ten and twenty minutes. The results which are tabulated in Table IV show a variation for the same orcanism heated for the same length of time of not 61 more than two dezrees which is only one step in our determina- tion, Subsequent Infection During Cooling and Bottling of Milk. Three sterile liter flasks were used, the first being used for a sample of the milk as soon as pasteurization by the "Holding" process was completed; the second for a sample which had been cooled in the pasteurizer and the third for the milk that had been bottled. As soon as possible after Collecting, each sample was pleted. From the results record- TABLE V. Subsequent Infection Table. Average of 20 trials ). Lact. Alkali Milk after Total Acid and inert Peptonizers Bacteria per 00. per Co. Pasteurization 46,000 34,000 3,900 8,000 Cooling 69,000 39,000 14,000 11,000 Bottling 86,000 63,000 17,000 6,000 ed in Table V we assume that Ayers and Johnson in their assurp- tion that the sourins of resteurized milk is due to subsequent infection with acid-forming bacteria during cooling and botéling are only partly right as the increase in number found immed lately after pasteurization may be accounted for partly by ‘multiplication. The process of cooling and bottling takes about half an hour. The remeining increase in numbers is through subsequent infection from utensile during the rrocess of cooling and bottling. 52 The Protective Agency or Factors in Milk. Bact. lactis acidi and B. coli were grown in nutrient broth for twenty-four hours at their respective optimum tem peratures. One cc. was then introduced into each of the following: 10 cc. of sterile distilled water; 10 cc. of sterile broth and 10 cc. of sterile litmus milk ( skim milk plus 0.05 % azolitmin ). Then the thermal death roint was determined in each of the three by the methods previously described. It becomes evident from this data, the average of TABLE VI. Thermal Death Point Average of 18 Determinations. | Organism Water Broth Milk Bact. lactis acidi XIT 63°C, 64°C. 68° C, Bacillus coli ITI 686°C. 694°C. 72° C. Which is recorded in Table VI, that milk exerts some influence which proteots the bacteria from heat, an influence which is not so marked or is absent in water and broth. The thermal death point of the cultures showed a difference of 33° to 4°. between broth and milk and about 4)° between water and milk. Higher Thermal Death Point Due to the Physical Nature of Milk. The different constituents of milk were next respectively dissolved or suspended in distilled water in the proportion found in fresh milk, placed in test tubes, sterilized in steam and then inoovlated with one cc. of a twenty-four hour broth 63 culture. The thermal death point of the organiem introduced into the solution or suspension of each constituent ( water, fat, lactose, ash, cesein and albumin )* was determined. It The constituents were present in the following proporticne (14). Constituents Percent Range Percent Water 87.00 82.4 to 89.6 Fat 3250 2.5 * 6.0 Casein 3225 2.5 4% 4.0 Albumin 0.50 O.5 °* 0.8 Milk sugar 5.00 4.5 " 6.0 Ash 0.75 0.6 * 0.8 Solids 13.00 10.4 to 17.6 was found that in the nresence of casein, albumin and fat, the thermal death voint was higher than with the other constituents. TAPLF VII. Thermal Death Points. ( Average ). —_— oo ae @- a Orzanism __—* Water Fat Casein Albumin Sugar Ash Bact. lactis 0 acidi No. XII. 633°C. 65°C. 68°C. seeC. 64°C. 642°C. Bacillus coll ° No. III. 68i°c, 71 c. 738°C. 73°C. 70°C. 70 %. *Much difficulty was encountered in trying to dissolve and suspend the different constituents. Fat wos emuleified bv constant shaking in a shaker machine, Casein was dissolved in a weak solution of sodium hydroxide _ and then neutralized. Albumin was mace slichtly alkaline ena boiled. 64 The results recoried in Table VII indicate that it is both the nitrozenous matter and fat which act as the protector of the bacteria when these are subiected to pasteurization. Fat does not seem to exert as much protection as the albumin and casein. To determine how each of the nitrogenous compounds would act in combination with fat, mixtures of casein and fat, of albumin and fat, and of casein, albumin and fat were mede in the proportions in which they are present in milk and ster- ilized. In these the thermal death point of B. coli and Bact. lsctis ecidi were determined. From the results in Teble TABLE VIII. Organism Casein Albumin Albumin, Casein and Fat and Fat and Fat. Bact. lactis acidi No. XII. 68°C, es°c. eamc, B. coli III. 93°C, 72°, 72°, VITI we would infer that a combination of nitrogenous matter and the fat of milk does not tend to materially increase the ability of organisms to resist heat nor reise the thermal death point over that of either of the notrogenous constituents used alone. Thus far the constituents of milk which had been used were commercial products. [t was thoveht that there might be a difference in the effects of the constituents of milk as they exist in natural milk and those which are commercially prepared. To determine any difference in the action of the 65 natural and commercial constituents the following method was used: Thole milk was tubed ( 10 cc. per tube ). Then another portion was skimmed and tr.e skimmed milk was tubed. To a yart of the skimmed milk rennet was added and the casein coagulated and separated by filtration and the milk serum tubed. All of these tutes were sterilized in steam, then the thermal death noints of both B. coli and Bact. lactis acidi were determined in each of the solutions. The differ enoe in the thermal death point in the wnole milk and in the skimmed milk shows the rrotection exerted by fat, while the difference between the skimred milk and milk serum shows the amount of protection given orsanisms bv casein and albumin. Then, since the organisms are the same strains as those used in the previous experiments with broth, water and milk, still further comparison as to protection of the other substances besides fat and casein that are rresent in milk and not pres- ent in broth and vater can be made. From the results obteined TABLE IX Whole Skim Mik — Orcanism Milk Milk Serum Broth Fater Bact. lactis -acidi No. xtt. 70°C 67% 641°C 64°C Ssh 00 B. coli 74°C 712°c esi®c _—gg°c ge2°C in Table IX it is evident that the casein and fat are the main protective factors which exist in milk. There is also some rrotective action in milk serum as the thermal death roint is hisher in it than it is in either broth or water. o 66 TABLE X. Thermal Death Point in Acid Media. ( Average of 20 trials ). Acid 0.25 % (Lactic Acid ° 0.40 % (Lactic Acid 57°C ,# 61°C. 61°C. Acidity of Broth 0.25 % (Lactic Acid 53°C, 56°C. §5°C, 0,40 % (Lactic Acid 53°C 55°C, 55°C, *Except in four cases when the acidity of the medium was higher than 40° the milk curdled and the thermal death point was higher than any temperature used during the trials. TABLE XI, * 2 _ Thermal Death Points, Medium o1 55.55 57 59 61 63 65 67 69 71 73 75 77 79°C, Bact. lactis acidj No. II A. + + Cream + + + + + + ¢ e8&© © #© «@© @ «@ Whole Milk + + + #¢ + + = o o o o o @ o o Skim Milk + + + + - - eo oe e - e e ~ o o Milk Serum + + + 2 2 - = os = = = = - e = Boujlion + + = oe o o e@ e @ oe os oe 2 o o _. Bact. lactis acidi No, IVA, | —s_s___T Cresm + + + + + + + + + + - e o « - Whole Milk + ¢ ¢ + + + + &© © © © 2@ © «© «o@ Milk Serum + + + + + @ = = oo e a @ o o a Bouillon a _ Bac lactis acidi "Knop"* Cream + + ¢+ + ¢ $+ ¢ + + w@ wm» @ © «@ Whole Milk + + ¢ + ¢ + ¢ &@ 2@ @ © «© @ © «o Skim Milk + ¢ ¢ + ¢ ¢ @© @ &© «@ «© «@ «@ «@ «@ Milk Serum + + ¢ + ¢ 38 2 © @ © «© « «@ @« e@ Bouillon + + + « «2 «© © «© «a w@ w@ w@ «© @w Cf 67 ( Table 11 ), Concluded. Thermal Death Points. _ Mediwmn 51 53,55 57 59 61 65 65 67 69 71 73.75 77 79°C —_ Be coli No. Vs _ Cream + + + ¢ + + + + + + € © e= © & Whole Milk + + + + + + € + € © © © © © © Skim Milk + + + + + + + - oe o a o o a o Milk Serum + + + + + + o o o o - - o o o Boujlion + + + + + o o o o o o o a o oe Non-acid B, No. 16. Creem + ¢ ¢ ¢ ¢€ + € + € + $+ + $+ © @ Whole Milk + + + + + + ¢ + + ¢ @&© w@ w@ @ © Skim Milk ¢ ¢ + + + + + + ¢€¢ e@ e& & @ - e Milk Serum + + + ¢ ¢ ¢€ + @ © e@ e& 2 © © -@ Bouillon + + + + + + 2@ © © © = = = = = . Non-acid B, No. 14 A, Cream + +¢ + ¢ + &¢ ¢+ w@ © 8e& e& ewe e@ & @ Whole Milk + + + + + - oe o a o o o o oe o Skim Milk + + + «© @ »@ © = 2 © © © © @ @ Milk Serum + + = = e - -= « o = eo oe a oa - Boujlion + © «© 2 e@ © © © 2 © © © © 2 = t. bulgaricium, _ 69 71 735 75 77 79 81 83 85 87 89 90 91 93_ 95°C Cream + +¢ + + + + ¢+ ¢ ¢ + €+ + + +2 @ Whole Milk + + + ¢ + + ¢€ + + + $$ w@ 2@ @& @ Skim Milk + + ¢ ¢ $+ + + + + + e® e2# a@ & -@ Milk Serum + + + + + + + + o o o o o o oe Bouillon + + + + + + + aed = e “a @ a 2s 2 It is also evident that the commercial constituents sand the constituents as they exist in milk act differently in protect- ing an organism from the pasteurization temperature, i.e., the natural constituents exert a greater protective influence. ee Influence of Crear. The fact that the thermal death roint of twe organisms is found to be higher in wrole milk than in serareted milk led to the determination of the thermal death roints of Bact. lactis acidi Nos. II.A, IV.A.and Knop, RP. coli No. V. and Bact. bulgaricum and of two inert bacilli Nos. 16 and 14 A. in cream, whole milk, ekim milk, milk serum and bouillon. The resulte in Teble X show that the presence of fat enables the various organisms to vithstend higher temperatures. In the case of B. 14 A. the thermal death roint shown in bouillon was over ten devrrees lover tran tre terrerature of resteuri-~ zation by the "holdine" nrocess, vet thie orranism was one iscleted from rasteurized cream, its thermal death roint, throuzh the influence of the fat in the cream being raised ebove thet of the pasteurization temperature. Prv. F. Smith (14) in his book on "Bacteria in the Re- lation to Plant Tiserses" Vol. I, states in s footnote tuat an acid medium protects crzanisme frem neat which would be Gestreyed in a medium containing less acid, nevtral or alke- line. To verify tnis staterernt, tubes conteining litmus milk which nad an acidity of 25° and 20° Fuller's scale (mace acid by the addition of N/l lectic acid) and tubes of bouillon of like acidity were inceculated with lactics II A., IV A. ene vnon and their thermal death roints determined. Pesults as shown in Teble ¥ lead us to the helief that acid, when used in srall quantities kas no er*act on the thermal desth rcoint. 59 “hen it is used in lerve enovzh quantities to curdle milk an increase of tne thermal deeth noint results which is prob- ably due to the coazuletion of the casein in milk. General Discussion of Possible Influencing Factors. Organisms tnrat are non-srore formers are found to survive cesteurization, Arong these are the lectic acid bacteria. Meany theories and assumptions have been presented and each ne One ig asserted sy the author to be the true anc most feasible explanation. For the average layman, the most easily believ- able theory is that becteria survive pasteurization on account of the huge amount of milk or cream nasteurized at a time, vbe- cause the temperature is not evenly distributed througnout. This assumption might be true if we were to use the Flash system for pasteurization or pesteurized in a -lain tank oon- taining an inefficient etirrinzs device, thus not causing dis- tributicn of heat evenly throushout the medium which is being pasteurized. But our modern commercial rnasteurizer contains spiral coils wiioh stir the liquid constantly and there is little reason why the temoerature should vary greatly at any part of the tank. This theory, so easily believable must therefore only be taken into consideration in those establish- ments which rasteurize their milk or cream by the use of the Flash system or have inefficient cocteurizers. The assumntion that some orgeniems found in pasteurized milk do not survive pasteurization but gain access during the process of cooling, bottling and capyvins was found to be true, contamination from apparatus, tne room eir etc. being the imperfectly sterilized Cause. This source when proverly controlled adds so few bacteria, most of tuese being of the harmless type, that it is almost negligible, yet it is a factor which allows bacteria to enter milk after it is properly resteurized. This means of entrance of becteria could be avoided by thorough cleansing ani sterilization of utensils and cveneral cleanliness in the dairy. Organisms which have a hich thermal death noint have been found by many authors and have been ziven as one cause of bacteria surviving pasteurization. Ayers and Johnson as Well as Zelonki found members of the colon group which varied sreatly in thermal death reints. The same authors found streptococci wnoich showed great variation in their ability to resist heat. This then is one of the reasons whereby bacter-— Ja of the lactic acid tyre are found in milk after rroper pasteruizaticn,. Russell and Hastings found that a pellicle formed dvring heating of the milk would exert a protective influence. The pellicle not being observed to form during any of the trials mede, their proof is accented as a factor Which influences the death of microorgenisms in milk by heat. Toe shutting off of air avring thermal death point determins- tions as tried in tre closed and reraffin covered tube in con- parison with tre onen tube has little if any influence. This 71 theory may almost be entirsly neglected in attemoting to group the true reason of bacteria surviving pasteurization. The nitrogenous constituents and the fat in milk or cream do protect organisms, the fat having a greater influence than the casein. The action of casein and albumin is probably one which was assumed by A. Folff , namely, the heat forming around the cell an acid coagulum of albumin. Fat must act in a Way very similar to that of the nitrogenous constituents of milk or cream in protecting the organisms. A high acidi- ty ( 0.4% lactic acid ) has no effect on the thermal death point. An acidity hish enough to cause the curdling of tne milk during thermal death roint determinations raises the thermal death point several desrees. The breed of cattle from which the milk is drawn, the general health of the cow and the condition of the udder, the products resulting from bacterial growth previous to pasteurization, heating under reduced or increased nressure and the relative specific heat of milk or cream may also be factors that protect organisms during heating. SIP TAPY AND CONCLUSION, 1. The thermal death roints of seventeen different strains of bacteria isolated from pasteurized milk varied zreatly wnen heatea for 10 minutes or for 20 minutes. Se Fifty-three percent of the cultures which were used in this experiment hai a thermal death roint below the tem perature commonly used for rasteurization ( 628°C, ) 3. The variation of temperature during commercial past- eurization ( Holder rrocess ) is very slight. This shows conclusively tnat the variation in the bulk of cream pasteur- ized does not account for the survival of the bacteria. 4, The thermal death point of some of the organisms used was very high, this accounting for their survival of rasteur- ization. Se Contamination during cooling and bottling is very slight, since the count obtained directly after the pasteuri- zation is more than half as mch as that obtained after bott- ling. This increase is accounted for, partly by multiplica- tion and rartly by contamination from utensils. 6. The protective azency in milk is the nitrogenous matter ( casein and albumin ) and fat. Thie is true even when commercial constituents of milk are used. 7. When constithents of milk in its natural state are used, the fat and casein offer creater protection from heat. 8. Commercial constituents of the milk in its natural stqte differ in their protective influence. With commercial 73 constituents tne fat, the casein and the albumin exert al- most an equal protection. But with the fat anc casein in the natural state the nresence of the fat adds additional protection to the suspended orzaniems. 9. A sealed tube or a tube closed with liquid pareaf- fin does not protect organisms to greater extent that an ordinary open tube. 10. A low acid medium has no effect on the thermal death point while a hich acid medium may raise it several déprees. ll. Factors which eccount for the presence of organisms in pasteurized milk are the contamination during cooling, and bottling, pellicle formation during heating, a high acidity of the medium, the nigh thermal death voint of some organisms and the protection given by the fat and nitrogenous contents of the milk or cream. 74 RPPURENCES. (1) Ayers, ©. EH. and Johnson, ¥, T, Jr. 1914 The Ability of Streptococci to Survive Past- eurization. Journal of Agr'l. Research, Vol. II. NO. IV. 1911 A Study of the Bacteria Which Survive Past-— eurization. Dept. of Agri. Bur. of Ani- mal Industry. Bul. 161. (3) 1911 The Bacteriology of Commercially Pasteurized and Rav Market Milk. (4) Eyre, J. 7. 1914 Bacteriological Technique. (5) Gage, S. and Stouchton, Grace. 1906 A Study of the Laws Governing the Resistance of B. coli to Heat. Technology Quarterly, Vole 15, Ppe 41-54, (6) Harrison, F. C, | 1905 A Study of Sixty-six Varieties of Gas-producing Bacteria Found in Milk. Centr. f. Bakt. II 14. pe 359-373, (7) Jordan, E. O. 1914 General Bacteriology. (8) Kinyoun, J, J. and Dieter, L. V. 1912 A Bacteriological Study of Milk Supply of Washington. Amer. Jour. Publ. Health IIe Ppe 263=-274.- (9) Lipman, J. Ge 1911 Bacteria in Relation to Country Life. (10) Marshall, Cc. E. 1897 Pasteurization of Milk. Bul. 147, Mich. Agr'l. Exp't. Stae (11) Prase, H. D. 1913 Problems in Sanitary Milk Pasteurization. Reprot of Soc. of Amer. Bact. oe (12) Russell, N. L. 1895 A Biological Study of Pasteurized Milk and Cream Under Commercial Condition. Centr. f. Bakt. II. Pe 741. (13) Russell, H. L. and Hastings. 1902 Pasteurized Wilk and Cream, Centr. f. Bakt. Il, Be pe 339. (14) Snyder, F.T. 1911 Dairy Chemistry. (15) Smith, Theobald. 1899 Studies on the Tubercle bacillus. Jour, of Ned. Vol. 4, Pe B17. (16) Smith, E. Te 1905 Bacteria in Relation to Plant Diseases. Vol. I. (17) Yolff, A. 1908 Zur Kentnis der Versanderungen in der Bact-~ erienflora der frischen Milch. Centr. f. Bakt. II, 60, -—. 651. (18) Zelenki, T. 1907 Zur Frage der Pasteurisation der Langlings- milch. Centr. f. Bakt. II, 16, De 175. ‘ ‘ , e ee | “4 “oe, c Wa ? ‘ 1 tate a To | Ml) 03142 8547