THE UNIVERSITY OF ILLINOIS cop. "2. AfiRIGULTURAL UUUBY ING \ BULLETIN No. 230 ELIMINATION OF GERMS FROM DAIRY UTENSILS I. BY RINSING II. BY DRYING IN SUN AND AIR BY M. J. PRUCHA AND H. A. HARDING URBANA, ILLINOIS, NOVEMBER, 1920 CONTENTS OF BULLETIN No. 230 PACK INTRODUCTION 139 METHODS OF STUDY 140 How the Utensils Were Washed 140 Method of Counting Bacteria in the Utensils 141 EXPERIMENTAL DATA: PART I. ELIMINATION OF GERMS FROM CANS BY EINSING WITH HOT WATER. .142 Cooling Effect of Cans on Rinse Water 143 Effect of Einse Water on Germ Life in Cans 144 PART II. ELIMINATION OF GERMS PROM CANS BY DRYING IN SUN AND AIR.. 157 General Condition of the Cans and Pails After Being Kept on the Eack. .157 Number of Bacteria in Untreated Cans 163 Bacteria in Cans and Pails After Eight Hours of Exposure to Sun and Air . 163 Bacteria in Cans After Twenty Hours ' Exposure to Sun and Air 164 Bacteria in Dry Cans and in Moist Cans 165 Importance of Dryness in Controlling Germ Life in Utensils 167 GENERAL DISCUSSION . . . 168 ELIMINATION OF GERMS FROM DAIRY UTENSILS I. BY RINSING II. BY DRYING IN SUN AND AIR Br M. J. PRUCHA, CHIEF IN DAIRY BACTERIOLOGY AND . H. A. HARDING, CHIEF IN DAIRY BACTERIOLOGY INTRODUCTION Before saying that a milk is good one wishes to know that it is rich, safe, clean, and sweet. Accordingly, quality in milk is said to depend upon the four items: (1) food value, (2) healthfulness, (3) cleanliness, and (4) keeping quality. 1 Of these four elements, keeping quality, or the ability to remain sweet, is the most difficult to protect successfully during the produc- tion and delivery of the milk. If it were not for the action of germ life on the milk it would remain sweet indefinitely. However, every time milk is exposed to dust or is changed from one container to an- other it receives germ life. This germ life living and growing in the milk breaks the milk sugar into acid and sours the milk. Accordingly the first step in protecting the keeping quality of milk is to reduce as much as practicable the number of germs which get into it. In Bulletin 204 of this station 2 it was pointed out that the uten- sils in which milk is handled are an extremely important source of germ life; hence, it is good dairy practice to reduce as far as prac- ticable the number of utensils coming into contact with the milk. As there is a limit to such reduction it is also important to know how the necessary utensils may be handled so as to add the smallest num- ber of germs to the milk. As a part of the washing process, dairy utensils are practically always rinsed, hot water being commonly used for this purpose. This rinsing, in addition to removing traces of the washing powder, me- chanically removes some of the remaining germ life and if the water is hot enough it also destroys some of the germs. A study of the effect of rinsing with water upon the germ life in the cans, is reported in Part I of this bulletin. This study was confined to cans because Warding, H. A., Breed, E. S., Stocking, W. A., Jr., and Hasting, E. G., What is Meant by "Quality" in Milk. 111. Agr. Exp. Sta. Circ. 205. 1917. Trucha, M. J., Weeter, H. M., and Chambers, W. H., Germ Content of Milk: II As Influenced by the Utensils. 111. Agr. Exp. Sta. Bui. 204. 1918. 139 140 BULLETIN No. 230 [November, among the common dairy utensils cans 1 are the outstanding factor in adding germ life to the milk. Perhaps the simplest and most universal treatment given utensils is, after washing, to invert them to dry on a rack, preferably in the sun. The effect of this drying in the air, upon the germ life in cans and pails, has been studied and the results are given in Part II of this bulletin. In stating the amount of germ life in milk it is customary to give the number present in a cubic centimeter (about 20 drops), or more briefly "per cc." The public has become familiar with the fact that certified milk ordinarily contains less than 10,000 bacteria per cc. and that the presence of 1,000,000 bacteria per cc. indicates that the keeping quality of the milk has been seriously impaired. Accord- ingly it has seemed helpful to state the germ life which would be contributed by any given utensil in terms of the number of germs per cc. it would add if filled with sterile milk. It is believed that this form of expressing the results will not only assist in comparing the results of handling utensils in different ways, but it will also in- dicate whether the contamination arising from any particular uten- sil is heavy enough to furnish any considerable part of the final germ content of the milk. It is a regrettable fact that, particularly in hot weather, much of the milk as it is delivered at the shipping station or the bottling plant is heavily seeded with germ life. In the hope of finding practicable means by which this seeding could be reduced, the present studies were directed primarily to operations which could be carried out on the farm, tho the results are equally applicable in other places. The larger part of the data reported in this bulletin was obtained during 1915-1917. Messrs. H. M. Weeter and W. H. Chambers, then members of this department, took an active part not only in the rou- tine conduct of experiments but also in developing the plans for this study. The faithful service and intelligent interest in the earlier por- tions of these studies of both of these men is gratefully acknowledged. METHODS OF STUDY These studies were made upon the eight-gallon cans and the four- teen-quart covered milking pails used in caring for the milk from the University dairy herds. The handling of these pails and cans was in close accord with good commercial practice. How THE UTENSILS WERE WASHED The cans were washed at the University creamery in a vat contain- ing about 40 gallons of water at about 110 F. to which was added 'See pages 222-230 and 246-247 of Bui. 204, HI. Agr. Exp. Sta., noted on preceding page. ELIMINATION OF GERMS FROM DAIRY UTENSILS 141 one percent sodium-carbonate washing powder. Each can was scrubbed with a brush, and after draining for about five seconds was rinsed in another vat containing plain water at about 110 to 120 F. The pails were washed in a similar manner except that the washing took place at the farm. METHOD OF COUNTING BACTERIA IN THE UTENSILS After the utensils had received the desired treatment, one liter (approximately one quart) of sterile cool water was poured into each can or pail, and after a thoro shaking the water was poured out and the number of bacteria removed by this water taken as the number present in the utensil. The extended tests of this method of determi- nation which are reported in Bulletin 204 (pages 222-226) indicate that the numbers of bacteria found in this way represent about 75 percent of the total which would be obtained by repeated rinsings. This method of testing was employed, not because it was considered ideal, but because it seemed the best available method. The plate method was used for counting the bacteria in this water. In all cases three plates were prepared from each of two dilutions. The counts recorded in this bulletin are in each case the average of the counts of the three plates from the dilution in which the number of colonies approached the closer to 200 per plate. All the plates were incubated for five days at 20 C. and for two days at 37 C. before they were counted. Nutrient agar of the following composition was used for plating: Agar shreds 15 grams Liebig 's meat extract 3 grams Witte 's peptone 10 grams Lactose 10 grams Distilled water 1 liter The resulting medium had a reaction varying between 6 cc. and 9 cc. normal acid to the liter, phenolphthalein being used as an indicator. 142 BULLETIN No. 230 [November, EXPERIMENTAL DATA PART I. ELIMINATION OF GERMS FROM CANS BY RINSING WITH HOT WATER The rinsing of utensils is commonly employed to complete the cleaning process and to remove traces of washing powder. The pres- ent study is concerned primarily with the effect of rinsing upon the germ life in the utensils. According to common commercial practices in moderate sized plants, cans as they come from the washing vat are rinsed in a vat of warm water. This warm water is so efficient in removing germs from the cans that the vat of rinse water quickly becomes loaded with germ life. Samples of typical rinse water were employed on three different days in seeding the cans reported upon in Table 6, and these samples carried 4,220,000, 7,650,000, and 11,500,000 bacteria per cc. respectively. In considering the germ content of cans rinsed under these conditions it should be remembered that about 10 cc. of this rinse water adheres to the inside of even well-drained cans. The present study is concerned primarily with conditions as they exist on farms, and here the rinse water commonly comes into con- tact with one or at most only a few utensils. The tests here described were made during June, July, and October, 1915, and April, May, and June, 1916. In these experiments 80 cans were rinsed with water at 70 F., 103 were rinsed with water at 150 F., and 266 were rinsed with water at about 205 F., making a total of 449 cans examined. The cans which were to be studied were first washed, and after standing in the creamery from one to four hours were rinsed in the following manner : A measured amount of rinse water was poured slowly into each can in such a way that the stream of water came into contact first with the upper edge of the neck of the can and then ran down the inner surface. A portion of the water was also poured over that part of the lid which comes in contact with the milk, and this water was also allowed to run into the can. The cover was replaced and the can was then shaken for 30 seconds in order that the inner surface of the can might come in contact with the rinse water. After this the water was poured from the can. The effectiveness of hot water was tested at two temperatures at 150 F. and at about 205 F. At each temperature different amounts of water were tested: at 150 F., 1 quart, iy 2 quarts, 2 quarts, and 4 quarts ; and at 205 F., 1 quart, iy 2 quarts, 2, 3, 4, 6, and 9 quarts. Rinse water at 70 F. was tested on the same days that similar cans were rinsed with hot water, the only difference in the treatment of the cans being in the temperature of the water used. 19*0] ELIMINATION OF GERMS FROM DAIRY UTENSILS 143 It will be noted that the application of this water to the cans con- stituted in reality a second rinsing inasmuch as the cans had been rinsed in connection with the regular washing process. However, in view of the high germ content of the wash water employed in con- nection with the regular washing process and the great variability of the germ content of ordinary cans, it is thought that using cans washed in the regular way gave more uniform material upon which to test the effects of the hot-water treatment. COOLING EFFECT OF CANS ON RINSE WATER Everyone knows that when cold hands are placed in hot water the hands are warmed and the water cooled. Likewise everyone under- stands that when a cool can is rinsed with hot water the can is warmed and the water cooled tho few realize the extent of this temperature change. The decrease in the temperature of the various amounts of rinse water used in these studies was determined by taking the temperature of the water just before it was poured into the can at 72 F. and again immediately after it was poured out, an interval of about sixty seconds. The observations made upon the reduction in temperature of the rinse water as used in these studies are summarized in Tables 1 and 2. TABLE 1. EFFECT OF CANS IN REDUCING TEMPERATURE OF RINSE WATER When the cans were treated singly Amount of water Temperature of water Drop in temperature Before rinsing After rinsing quarts F. F, F. 1 150 110 40 2 150 122 28 4 150 131 19 1 210 140 70 2 210 154 56 4 210 170 40 6 210 180 30 9 210 187 23 TABLE 2. EFFECT OF CANS ON TEMPERATURE OF RINSE WATER When four cans were rinsed in succession by the same lot of rinse water Amount of water Before rinsing Temperature of water Total drop in temperature After 1st can After 2d can After 3d can After 4th can quarts 2 6 Of, 210 210 F. 158 178 F. 131 160 F. 113 150 F. 100 138 F. 110 72 As shown in Table 1, the decrease in the temperature of the hot water during the process of scalding was very marked and depended both on the amount of water used and on its initial temperature. For example, one quart of water at 150 fell in temperature to 110, a drop 144 BULLETIN No. 230 [November, of 40 degrees; while four quarts decreased to 131, a drop of 19 de- grees. When the cans were scalded with water at 210, one quart of water decreased in temperature from 210 to 140, and four quarts dropped to 170, a loss of 70 and 40 degrees respectively. The results in Table 2 show that when six quarts of boiling water was applied successively to four cans at 72 F. the temperature of the water fell from 210 to 138, a drop of 72 degrees, while after a simi- lar application of two quarts of boiling water the temperature of the water fell to 100 F., which is a drop of 110 degrees. It is thus seen that when hot water is poured into utensils for the purpose of scalding them the heat passes quickly from the water to the walls of the utensils. In considering the temperatures found in the rinse water as it came from the cans it should be remembered that hot water below 140 F. has but little killing effect when the time of exposure to it is short. 1 Accordingly rinse water at 150 F., in the quantities which are avail- able on any ordinary farm, will be so promptly cooled as to have little killing effect upon the germ life in the utensils. Even boiling water is so promptly cooled by the cans that unless two or more quarts are applied directly to each can the germ-killing effect is much less than is commonly believed. That the cooling effect of the utensils on the scalding water is prob- ably not fully appreciated by many dairy operators, is shown by the following observation made in a large city milk plant. It was the custom at this plant to treat the pasteurizing vat, 50 feet of sanitary pipe, the tubular cooler, and the tank under the cooler, with hot water for the purpose of "sterilizing" these utensils. Three hundred gal- lons of boiling water were pumped from the vats thru the pipe, and were allowed to trickle down over the cooler into the tank. It took about twenty minutes to pump this water. When all of the water had reached the tank the temperature of the water had fallen from 210 to 120 F., a drop of 90 degrees. Long before this water had reached the end of its appointed journey its temperature had fallen below the point where it would be destructive to germ life, and any further reduction of germ life resulting from its use depended solely upon its mechanical removal of germs from the utensils. EFFECT OF RINSE WATER ON GERM LIFE IN CANS Cans at the farm are scalded or rinsed as a final step in removing the germ life and preparing the cans for receiving milk. Accordingly the amount of germ life remaining in the cans after such treatment a Smith, Theobald. The Thermal Death-point of Tubercle Bacilli in Milk and Other Fluids. Jour. Exp. Med. 4:217-233. 1899. Russell, H. L. and Hasting, E. G. Thermal Death-point of Tubercle Bacilli under Commercial Conditions. Wis. Agr. Exp. Sta. Ann. Ept. 17(1900) :147- 170. 1900. ELIMINATION OF GERMS FROM DAIRY UTENSILS 145 is a matter of first importance. In this study such a determination was made in the case of each of the 449 cans tested. As the available supply of hot water for rinsing utensils at the farm is limited, the question of how much rinse water per can is really needed is likewise important. In these studies the use of varying amounts of water was tested, the range of these tests being especially wide in the case of boiling water. In the application of rinse water to milk cans, the object usually in mind is the destruction of germ life. While such destruction is a natural result, particularly where boiling water is used, the rapid accumulation of germ life in rinsing vats makes it evident that the mechanical removal of germ life is also an important function of rinse water. As a means of getting information regarding the importance of this mechanical removal of germ life from cans by rinse water, the number of living germs in the rinse water as it came from the cans was also determined. The results of these studies in connection with the rinsing of 449 cans are given in Table 3, in which the successive columns show the number of the can, the number of living germs found in the rinse water from the can, the number of living germs recovered from the rinsed can according to the method given on page 141, and the num- ber of germs per cubic centimeter which the can would have con- tributed if it had been filled with milk. As is ordinarily the case in studies of germ life in cans the results given in Table 3 show wide variations in the findings from appar- ently similar cans. In order to bring this large amount of data to- gether so that they may be more readily compared the results obtained from each group of cans treated alike have been averaged. However, in considering these averages the varying number of cans which they include and the wide variations in the data which they represent should be kept in mind constantly. . The average number of living germs removed by each different amount of rinse water at each temperature and the corresponding av- erage number of germs found in the rinsed cans are given in Table 4. Bacteriological Condition of Cans Rinsed with Water at 70 F. The results as given opposite this temperature in Table 4 show marked irregularities. By referring to the number of cans in each group it is seen that a comparatively small number of cans are rep- resented in each of the averages except those where 1.5 and 6 quarts of water were used. Taking the averages as a whole, but remembering that those rep- resenting cans rinsed with 1.5 and 6 quarts of water are the more representative, it appears that cans rinsed with water at 70 F. and immediately filled with milk will ordinarily add to such milk about 146 BULLETIN No. 230 [November, 10,000 bacteria per cc. The use of large volumes of rinse water some- what reduces the number of bacteria remaining in the cans. TABLE 3. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 1 Quart of Water at 70 F. 1 8 000 000 263.2 2 46 300 000 1 523.2 3 88 000 000 2 894.7 4 1 210 000 000 39 802.6 5 4 630 000 000 152 315.8 Cans Rinsed with 1^ Quarts of Water at 70 F. 6 17 498 000 530 000 17.4 7 17 332 000 547 000 17.9 8 15 393 000 680 000 22.4 9 16 950 000 812 000 26.7 10 22 125 000 1 102 000 36.2 11 64 950 000 1 500 000 49.3 12 64 837 000 1 550 000 50.9 13 53 325 000 2 412 000 79.3 14 54 937 000 2 467 000 81.2 15 41 040 000 3 370 000 110.8 16 242 775 000 11 320 000 372.4 17 42 562 000 12 075 000 397.2 18 88 275 000 19 500 000 641.4 19 178 350 000 30 000 000 986.8 20 330 000 000 33 500 000 1 101.9 21 637 500 000 122 000 000 4 013.3 22 720 000 000 308 600 000 10 151.3 23 9 675 000 000 810 000 000 26 644.7 24 13 950 000 000 1 877 000 000 61 743.4 25 21 750 000 000 : 3 150 000 000 103 618.4 Cans Rinsed with 3 Quarts of Water at 70 F. 26 35 175 000 3 130 000 103.0 27 85 200 000 11 125 000 366.0 28 154 200 000 30 775 000 1 012.3 Cans Rinsed with 3 Quarts of Water at 70 F. 29 1 282 000 000 57 700 000 1 898.0 30 130 500 000 58 750 000 1 932.6 31 110 400 000 60 200 000 1 980.3 32 2 550 000 000 213 000 000 7 006.6 33 4 212 000 000 317 000 000 10 427.6 34 4 545 000 000 362 000 000 11 907.9 35 5 625 000 000 370 000 000 12 171.0 36 5 175 000 000 485 000 000 15 953.9 37 8 062 500 000 1 120 000 000 36 842.1 ELIMINATION OF GERMS FROM DAIRY UTENSILS 147 TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS Cans Rinsed with 6 Quarts of Water at 70 F. No. of can Number of germs removed by rinse water No. of can Number of germs removed by rinse water No. of can Number of germs removed by rinse water 38 606 000 50 48 900 000 62 132 450 000 39 606 000 51 50 700 000 63 183 600 000 40 906 000 52 54 900 000 64 371 700 000 41 1 212 000 53 59 400 000 65 456 498 000 42 6 900 000 54 62 400 000 66 535 398 000 43 7 800 000 55 66 450 000 67 614 196 000 44 18 300 000 56 72 750 000 68 922 200 000 45 30 450 000 57 83 550 000 69 1 794 396 000 46 31 998 000 58 84 600 000 70 2 059 500 000 47 42 900 000 59 84 600 000 71 10 722 000 000 48 43 350 000 60 87 600 000 72 32 580 000 000 49 45 150 000 61 119 100 000 No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Four Cans Rinsed in Succession by Same Lot of 9 Quarts of Water at 70 F. 73 549 000 000 150 000 000 4 934.2 74 1 395 000 000 170 000 000 5 592.1 75 1 350 000 000 23 500 000 773.0 76 1 530 000 000 6 270 000 206.2 77 15 300 000 000 1 050 000 000 34 539.5 78 16 200 000 000 415 000 000 13 651.3 79 25 200 000 000 910 000 000 29 934.2 80 46 800 000 000 1 000 000 000 32 894.7 Cans Rinsed with 1 Quart of Water at 150 F. 81 15 900 000 230 000 7.6 82 110 500 000 520 000 17.1 83 69 000 000 3 830 000 126.0 84 300 000 000 8 510 000 279.9 85 605 200 000 11 500 000 378.3 86 3 000 000 000 44 500 000 1 463.8 87 24 600 000 000 1 840 000 000 60 526.3 Cans Rinsed with 1^ Quarts of Water at 150 F. 88 16 162 000 243 000 8.0 89 15 055 000 302 000 9.9 90 17 460 000 310 000 10.2 91 19 522 000 335 000 11.0 92 16 650 000 401 000 13.2 93 21 543 000 417 000 13-. 7 94 9 120 000 470 000 15.5 95 17 812 000 475 000 15.6 96 18 195 000 490 000 16.0 97 19 980 000 502 000 16.5 98 10 850 000 810 000 26.6 99 15 532 000 1 295 000 42.6 100 207 000 000 6 500 000 213.8 101 450 000 000 28 010 000 921.4 102 1 050 000 000 37 507 000 1 235.1 103 9 975 000 000 265 000 000 8 717.1 148 BULLETIN No. 230 [November, TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 2 Quarts of Water at 150 F. 104 800 000 40 000 1.3 105 1 200 000 80 000 2.6 106 3 600 000 100 000 3.3 107 2 600 000 100 000 3.3 108 2 200 000 120 000 3.9 109 2 000 000 140 000 4.6 110 2 000 000 150 000 4.9 111 2 600 000 210 000 6.9 112 210 000 6.9 113 1 000 000 220 000 7.2 114 3 800 000 230 000 7.6 115 3 000 000 240 000 7.9 116 1 200 000 240 000 7.9 117 4 500 000 270 000 8.9 118 2 400 000 460 000 15.1 119 1 400 000 460 000 15.1 120 4 600 000 560 000 18.4 121 5 800 000 1 040 000 34.2 122 5 000 000 1 510 000 49.7 123 12 600 000 3 390 000 111.5 124 5 400 000 4 200 000 138.2 125 10 800 000 4 700 000 154.6 126 19 800 000 7 510 000 247.0 127 77 600 000 24 400 000 802.6 128 56 200 000 38 590 000 1 269.4 129 347 600 000 50 700 000 1 667.8 130 1 976 000 000 67 500 000 2 220.4 131 200 000 000 77 000 000 2 532.9 132 163 400 000 95 620 000 3 145.4 133 2 166 000 000 113 000 000 3 717.1 134 151 800 000 116 520 000 3 832.9 135 964 800 000 252 500 000 8 305.9 136 683 400 000 294 000 000 9 671.1 137 2 000 000 000 313 000 000 10 296.5 138 2 800 000 000 403 000 000 13 256.6 139 1 734 400 000 420 000 000 13 815.8 140 1 332 000 000 560 000 000 18 421.5 141 2 800 000 000 613 000 000 20 164.5 142 1 580 000 000 1 040 000 000 34 210.5 143 4 460 000 000 1 183 000 000 38 914.5 144 22 000 000 000 3 130 000 000 102 960.5 Cans Rinsed with 4 Quarts of Water at 150 F. 145 2 520 000 30 000 1.0 146 1 200 000 70 000 2.3 147 5 600 000 70 000 2.3 148 2 400 000 90 000 2.9 149 800 000 110 000 3.6 150 10 800 000 120 000 3.9 151 3 200 000 120 000 3.9 152 5 200 000 170 000 5.6 153 6 800 000 490 000 16.1 1920] ELIMINATION OP GERMS FROM DAIRY UTENSILS 149 TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 4 Quarts of Water at 150 F. (Cont'd.) 154 4 400 000 520 000 17.1 155 14 800 000 540 000 17.8 156 6 800 000 1 100 000 36.1 157 2 120 000 1 260 000 41.4 158 36 800 000 3 440 000 113.2 159 16 000 000 4 250 000 139.8 160 27 200 000 4 950 000 162.8 161 56 400 000 8 320 000 273.7 162 57 200 000 13 000 000 427.6 163 67 200 000 13 400 000 440.8 164 107 600 000 15 000 000 493.4 165 126 800 000 22 920 000 753.9 166 61 200 000 23 300 000 766.4 167 304 400 000 36 250 000 1 192.4 168 235 600 000 58 000 000 1 907.9 169 296 000 000 80 300 000 2 641.4 170 322 800 000 103 000 000 3 388.2 171 163 600 000 114 860 000 3 778.3 172 482 800 000 123 400 000 4 059.2 173 1 172 000 000 164 500 000 5 411.2 174 4 547 600 000 243 200 000 8 000.0 175 40 800 000 000 366 000 000 12 039.5 176 3 664 000 000 440 000 000 14 473.7 177 1 724 000 000 526 000 000 17 302.6 178 6 400 000 000 673 000 000 22 138.2 179 4 544 000 000 700 000 000 23 026.3 180 4 584 000 000 790-000 000 25 986.8 181 ' 8 880 000 000 922 000 000 30 328.9 182 10 564 000 000 970 000 000 31 907.9 183 10 000 000 000 1 385 000 000 - 45 559.2 Cans Rinsed with 1 Quart of Water at 200-208 F. 184 2 400 000 150 000 4.9 185 1 000 000 190 000 6.2 186 3 700 000 190 000 6.2 187 1 300 000 200 000 6.6 188 1 700 000 460 000 15.1 189 7 300 000 550 000 18.1 190 191 3 900 000 5 300 OdO 1 100 000 1 700 000 36.2 55.9 192 6 900 000 2 000 000 65.8 193 14 500 000 2 010 000 66.1 194 6 100 000 2 260 000 74.3 195 5 000 000 2 310 000 75.9 196 9 300 000 2 600 000 85.5 197 11 400 000 2 710 000 89.1 198 3 900 000 3 010 000 99.0 199 4 400 000 3 200 000 105.3 200 72 400 000 5 900 000 194.1 201 25 400 000 6 600 000 217.1 202 60 000 000 9 400 000 309.2 203 71 600 000 11 000 000 361.8 204 370 000 000 20 000 000 657.9 205 140 000 000 20 000 000 657.9 150 BULLETIN No. 230 [November, TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 1 Quart of Water at 200-208 F. (Cont'd.) 206 37 000 000 21 000 000 690.8 207 24 800 000 22 000 000 723.7 208 20 300 000 24 000 000 789.5 209 34 400 000 32 000 000 1 052.6 210 44 900 000 44 000 000 1 447.3 211 95 600 000 47 660 000 1 567.8 212 234 300 000 56 000 000 1 842.2 213 345 300 000 67 300 000 2 213.8 214 210 000 000 76 000 000 2 500.0 215 i 620 000 000 83 000 000 2 730.3 216 50 300 000 100 000 000 3 289.5 217 530 000 000 110 000 000 3 618.4 218 1 100 000 000 120 000 000 3 947.4 219 5 000 000 000 730 000 000 24 013.2 220 9 000 000 000 1 060 000 000 34 868.4 Cans Rinsed with 1> Quarts of Water at 200-206 F. 221 3 093 000 222 3 600 000 30 000 0.9 223 7 380 000 65 000 2.1 224 5 844 000 70 000 2.3 225 1 087 500 78 000 2.6 226 1 705 500 100 000 3.3 227 1 462 500 111 000 3.7 228 1 200 000 136 000 4.5 229 1 855 500 144 000 4.7 230 6 474 000 145 000 4.8 231 20 535 000 165 000 5.4 232 12 070 500 180 000 5.9 233 579 000 231 000 7.6 234 1 263 000 284 000 9.3 235 11 177 500 377 000 12.4 236 7 957 000 445 000 14.6 237 8 845 000 520 000 17.1 238 11 310 000 541 000 17.8 239 3 075 000 560 000 18.4 240 3 543 000 653 000 21.5 241 1 222 500 762 000 25.1 242 6 738 000 932 000 30.7 243 12 513 000 1 127 000 37.1 244 44 775 000 1 465 000 48.2 245 1Q 462 500 1 815 000 59.7 246 12 607 500 2 276 000 74.9 247 83 025 000 3 347 000 110.1 248 77 362 500 4 357 000 143.3 249 32 512 500 4 975 000 163.7 250 34 725 000 6 640 000 218.4 251 94 575 000 18 300 000 601.9 252 359 850 000 22 632 000 744.5 253 456 000 000 49 000 000 1 611.8 254 360 000 000 64 150 000 2 110.2 255 1 800 000 000 129 000 000 4 243.4 256 1 905 000 000 348 000 000 11 447.4 1920} ELIMINATION OP GERMS FROM DAIRY UTENSILS 151 TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 2 Quarts of Water at 200-208 F. 257 200 000 258 4 200 000 259 200 000 10 000 .3 260 400 000 10 000 .3 261 600 000 10 000 .3 262 400 000 10 000 .3 263 400 000 20 000 .7 264 400 000 30 000 1.0 265 600 000 50 000 1.6 266 200 000 50 000 1.6 267 1 000 000 60 000 2.0 268 400 000 60 000 2.0 269 200 000 60 000 2.0 270 200 000 80 000 2. 271 400 000 80 000 2.6 272 . 200 000 100 000 3.3 273 200 000 150 000 5.0 274 600 000 160 000 5.3 275 200 000 160 000 5.3 276 200 000 220 000 7.2 277 600 000 230 000 7.6 278 600 000 280 000 9.2 279 200 000 300 000 9.9 280 600 000 300 000 9.9 281 1 200 000 320 000 10.5 282 2 000 000 390 000 12.8 283 3 400 000 460 000 15.1 284 800 000 530 000 17.4 285 1 200 000 530 000 17.4 286 800 000 610 000 20.1 287 960 000 690 000 22.9 288 1 000 000 700 000 23.0 289 1 800 000 710 000 23.4 290 600 000 760 000 25.0 291 400 000 1 160 000 38.2 292 6 400 000 1 170 000 38.5 293 1 000 000 1 290 000 42.4 294 4 000 000 1 320 000 43.4 295 200 000 1 500 000 49.3 296 5 400 000 1 570 000 51.6 297 8 800 000 1 600 000 52.6 298 2 800 000 1 710 000 56.2 299 2 000 000 2 300 000 75.7 300 34 200 000 2 320 000 76.3 301 22 800 000 . 2 710 000 89.1 302 3 800 000 2 850 000 93.7 303 25 200 000 2 900 000 95.4 304 11 600 000 3 420 000 112.5 305 60 000 000 3 620 000 119.1 306 144 000 000 5 300 000 174.3 307 23 200 000 10 780 000 354.6 308 80 000 000 11 200 000 368.4 309 77 000 000 15 920 000 523.7 152 BULLETIN No. 230 [November, TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS No. of can Number of germs removed by rinse water Number of germs remaining in can after rinsing Germs per cc. of milk due to can Cans Rinsed with 3 Quarts of Water at 204-208 F. 310 210 000 46 000 1.5 311 381 000 74 000 2.4 312 960 000 101 000 3.3 313 306 000 215 000 7.1 314 7 260 000 275 000 9.0 315 2 955 000 553 000 18.2 316 2 676 000 627 000 20.6 317 2 106 000 660 000 21.7 318 6 231 000 766 000 25.2 319 7 371 000 771 000 25.4 320 9 900 000 785 000 25.8 321 15 636 000 994 000 32.7 322 5 496 000 1 232 000 40.5 323 17 106 000 2 849 000 93.7 324 69 750 000 3 183 000 104.7 325 2 040 000 4 337 000 142.7 326 444 750 000 5 225 000 171.9 327 243 000 000 10 100 000 332.2 328 118 800 000 12 250 000 403.0 329 777 000 000 48 700 000 1 602.0 330 1 387 500 000 63 525 000 2 089.6 331 1 014 000 000 72 500 000 2 384.9 332 997 500 000 73 000 000 2 401.3 333 5 880 000 000 258 750 000 8 511.5 Cans Rinsed with 4 Quarts of Water at 200-208 F. 334 10 000 .3 335 30 000 1 336 337 338 339 340 341 1 200 000 1 200 000 1 200 000 400 000 800 000 30 000 30 000 30 000 40 000 40 000 50 000 1.0 1.0 1.0 1.3 1.3 1 6 342 60 000 2.0 343 344 2 800 000 70 000 80 000 2.3 2 6 345 346 347 1 200 000 4 400 000 80 000 100 000 120 000 2.6 3.3 3.9 348 349 350 1 200 000 7 600 000 130 000 340 000 750 000 4.4 11.1 24.7 351 352 353 354 1 600 000 2 000 000 60 800 000 6 000 000 1 050 000 1 060 000 1 430 000 1 600 000 34.5 34.9 47.0 52.6 ELIMINATION OF GERMS FROM DAIRY UTENSILS 153 TABLE 3. Continued. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF RINSE WATER ON GERM LIFE IN CANS Cans Rinsed with 6 Quarts of Water at 190-208 F. No. of can Number of germs removed by rinse water No. of can Number of germs removed by rinse water No. of can Number of germs removed by rinse water 355 378 36 000 401 294 000 356 379 42 000 402 300 000 357 6 000 380 48 000 403 300 000 358 6 000 381 60 000 404 360 000 359 12 000 382 60 000 405 444 000 360 12 000 383 60 000 406 534 000 361 12 000 384 66 000 407 600 000 362 12 000 385 72 000 408 600 000 363 12 000 386 84 000 409 600 000 364 12 000 387 90 000 410 648 000 365 12 000 388 102 000 411 678 000 366 18 000 389 120 000 412 678 000 367 18 000 390 120 000 413 1 332 000 368 18 000 391 120 000 414 1 500 000 369 18 000 392 120 000 415 1 524 000 370 24 000 393 150 000 416 1 716 000 371 24 000 394 150 000 417 2 154 000 372 24 000 395 174 000 418 2 424 000 373 24 000 396 180 000 419 3 120 000 374 30 000 397 246 000 420 9 612 000 375 36 000 398 258 000 421 19 600 000 376 36 000 399 264 000 377 36 000 400 264 000 No. of can Number of germs removed by rinse water Number of germs remaining in cans after rinsing Germs per cc. of milk due to can Four Cans Rinsed in Succession by Same Lot of 6 Quarts of Water at 190 F. 422 10 000 3 423 25 000 8 424 45 000 1 5 425 85 000 2 8 426 30 000 1 427 50 000 1 6 428 5 000 2 429 105 000 3.5 Four Cans Rinsed in Succession by Same Lot of 6 Quarts of Water at 205 F. 430 20 000 7 431 170 000 5 6 432 200 000 6 6 433 765 000 25 2 434 55 000 1 8 435 18 825 000 619 2 436 200 000 6.6 437 1 680 000 55.3 154 BULLETIN No. 230 [November, TABLE 3. Concluded. EFFECT OF VARYING AMOUNTS AND TEMPERATURES OF , RINSE WATER ON GERM LIFE IN CANS No. of Can Number of germs removed by rinse water Number of germs remaining in cans after rinsing Germs per cc. of milk due to can Four Cans Rinsed in Succession by Same Lot of 9 Quarts of Water at 206-208 F. 438 1 233 000 144 000 4.7 439 1 593 000 22 000 .7 440 29 385 000 2 892 000 95.1 441 460 800 000 35 000 000 1 151.3 442 351 000 50 000 1.6 443 1 710 000 2 750 000 90.5 444 3 015 000 24 750 000 814.1 445 1 665 000 000 107 000 000 3 519.7 446 315 000 280 000 9.2 447 1 530 000 000 31 500 000 1 036.2 448 1 305 000 000 4 655 000 153.1 449 2 790 000 000 132 500 000 4 358.6 TABLE 4. AVERAGE NUMBER OF BACTERIA IN CANS AFTER BEING RINSED Number of cans Amount of rinse water used Temper- ature of rinse water Average num- ber of germs removed by rinse water Average germ content of cans after rinsing Germ content per cc. of milk due to can qts. F. per can per can 5 1 70 1 196 460 000 39 357 20 1.5 70 2 399 142 450 319 448 250 10 508 12 3 70 2 663 914 583 257 390 000 8 467 35 6 70 1 470 773 314 7 1 150 4 100 086 000 1 909 090 000 8 971 16 1.5 150 742 492 560 21 441 750 705 41 2 150 1 111 987 802 215 073 414 7 075 39 4 150 2 546 354 870 200 225 129 6 586 37 1 200-208 491 200 000 72 716 216 2 392 36 1.5 200-206 150 400 736 18 433 694 606 53 2 200-208 10 184 151 1 561 698 51 24 3 204-208 458 872 250 23 396 583 770 21 4 200-208 4 400 000 339 524 11 67 6 190-208 780 686 Bacteriological Condition of Cans Rinsed with Water at 150 F. Too close comparisons of the results with different amounts of rinse water cannot be made because the testing of the effect of 1.5 quarts was done in June, 1915, and the remaining tests were made in April, 1916. Again only seven cans were tested after being rinsed with one quart of water at 150 F. and these seven cans were all examined on the same day. 1920'] ELIMINATION OF GERMS FROM DAIRY UTENSILS 155 Bearing these facts in mind it will be seen from Table 4 that rinsing cans with water at 150 F. leaves them in somewhat better condition than when rinse water at 70 F. is used. Bacteriological Condition of Cans Rinsed with Water at 190 -208 F. Rinsing dairy utensils on the farm is commonly referred to as scald- ing them. Accordingly there is an unusual interest connected with the results from the use of boiling water. On this account the num- ber of cans in each of these high-temperature groups is large and the range in amounts of rinse water tested is wider than in the case of the other temperatures. The results from the two groups of cans rinsed with one quart and three quarts of water, seem unduly high as compared with those from the other groups. An inspection of the detailed results as given in Table 3 shows that 66 percent of the bacteria found in the 37 cans rinsed with one quart of water came from two of the cans. Had the remaining 35 cans been filled with milk they would have increased the germ content of the milk but 835 bacteria per cc. Among the 24 cans rinsed with three quarts of water, three cans contributed 70 percent of the germs. The remaining 21 cans would have increased the germ content of milk by only 231 bacteria per cc. Considering these results as a whole it is seen that these cans were in much better condition than those rinsed with cooler water. When a can is rinsed with more than one quart of boiling water, it will rarely add 1,000 bacteria per cc. when filled with milk. "When the amount of rinse water becomes large, the effect of the can on the milk would usually be below 100 per cc. In considering the relation of these results to farm practice it should be remembered that the water available for rinsing at the farm is frequently not fully up to the boiling point and the amount avail- able rarely permits the use per can of the larger amounts tested in these studies. Mechanical Removal and Destruction of Bacteria by Rinse Water References have already been made (page 142) to the high germ content of rinse water in commercial plants. This suggests that rinse water mechanically removes from the cans a large amount of germ life. Again the fact that rinsing with a liter of sterile water gives a usable measure of the germ life in the cans is further evidence of the ease with which water loosens and removes germs. By using sterile rinse water at a temperature too low to destroy the germs, and determining the germ content of the rinse water as it comes from the can, the mechanical removal may be accurately meas- ured. When the temperature of the rinse water is increased, its effi- ciency in removing germ life is probably also increased. However, a 156 BULLETIN No. 230 [November, count of the living germs in the rinse water as it comes from the can under such conditions gives, not the total number of germs removed, but rather the total number removed less the number which at the same time have been destroyed by the heat of the rinse water. When the cans were rinsed with water at 70 F., no killing effect occurred ; and from Table 4 it is seen that as an average of seventy- two cans this rinse water mechanically removed more than 2 billion living germs per can. When rinse water was applied at 150 F., the average number of living germs found in the rinse water from the 103 cans was slightly under 2 billion per can. When but one quart of rinse water was ap- plied, it was promptly cooled below the temperature at which germs are destroyed, as shown in Table 1. The average number of living germs found in the rinse water from 7 cans, each rinsed with one quart at 150 F., was practically double the average of the rinse water at 70 F., but 85 percent of these germs came from a single can. The average germ content of the rinse water from the other six cans was 683,433,333. The temperatures given in Table 1 indicate that when larger amounts of rinse water were used the rinse water remained for only a few seconds at a temperature sufficiently high to destroy germ life. Allowing for the variation noted in the cans rinsed with one quart of water at 150 F., the measurements of germ life as given in Table 4 indicate the removal of an increasing number of bacteria by the use of increasing amounts of rinse water. The small number of living germs in the rinse water at 190-208 F. makes it quite clear that here the effect of mechanical removal is over- shadowed by the destructive effect of the high temperature. Even where but one quart was applied the average germ content of the rinse water from 37 cans was only one-half billion per can as contrasted with about 2 billion per can where cooler water was applied. With the use of increasing quantities, the water remains at destructive tem- peratures for a longer time, and fewer germs survive in the rinse water. The result from the use of three quarts of rinse water is an ap- parent exception, but an inspection of the detailed results in Table 3 shows that these high averages were due to the results from a few cans. From the data here presented it would appear that the use of one quart of rinse water per can, at a temperature of 150 F,, gives good results in the mechanical removal of germ life but has only a small destructive effect upon the germs present. With the use of larger amounts of water at 150 F., or the same amount at higher tempera- tures, the removal or destruction of germ life is constantly increased. 1920] ELIMINATION OF GERMS FROM DAIRY UTENSILS 157 PART II. ELIMINATION OF GERMS FROM CANS BY DRYING IN SUN AND AIR Between May 22 and June 9, 1917, two hundred and thirteen cans and fifty-eight pails, after having been washed and rinsed as described on page 141, were inverted on a rack and exposed to sun and air. This rack was located on the south side of a farm building so that the utensils might have the fullest exposure to the sun. It should be noted, however, that their inner surfaces were not exposed to the di- rect rays of the sun, and hence the data in this study have no rela- tion to the disinfecting action of direct sunlight. Each day the pails and cans were placed on the rack at 8 a. m. At 4 p. m., after having been exposed for eight hours, all the pails and half of the cans were examined. The remaining half of the cans were examined at 4 a. m. the following morning. The examination consisted first, of noting whether the utensils were dry, clean, and free from odors, and second, of determining the number of bacteria in each according to the method already described on page 141. The fourteen check cans used in connection with these experiments were examined for bacteria immediately after being washed and rinsed. The results of the bacteriological examinations are given in Table 5. The number of bacteria found in each utensil at the time it was re- moved from the drying rack is recorded, and the numbers so obtained are also stated in terms of the number of bacteria which would have been added per cc. of milk had each utensil been filled with sterile milk. As far as could be judged by ordinary inspection, the cans and pails which had been inverted on the rack for eight and twenty hours respectively were in good condition for receiving milk. They were uniformly clean and free from any odor. All of the cans held for eight hours appeared dry at the time of final examination except those of May 22. On this day rain had fallen fairly continuously and the humidity had evidently prevented drying. Similarly moisture was evident in the cans held for twenty hours, on May 22 and 28, rain having fallen on the latter night as well. It rained on the nights of May 23, 29, 31, June 4, and 5, and was cloudy on May 31 and June 1. No moisture was found in these cans at the close of the twenty-hour period but the effect of the weather condi- tions upon their germ content is discussed on page 164. 158 BULLETIN No. 230 [November, "G^ fl J G O o ^ 03 W 0_ijj u St-rz3 O o> 3 -^ 3 ^3 ^< ffi O 8 1 o'E G 1 g'ijj CO ' Q . ^ a a IM -H r-i (N (M CO id >oo :O O >oo 1C O 00 CD CD CD CO 01 00 ^ 10 rHTfO 00 O3O T-< OOOO OOOO COCOO5O i-H OCO 1-1 CO * (N >, 09 S ooooo O O GO O O 1 1 1 < 1 < co 10 CDt- 00 050 I-H (N CO * 10 COCCCDO3COlOOOi-HO CDCOCDOSCOCDM^CO OOOOi-HOCOOO OOOOi-ii-HCDCDOO l-H CO i-H Tt< 00 00 i-H CN ^ _ ~~\ rf> It ,v-) (-.. 8 S 8 CO O cc jg. i 1 rH H-H i-i(NCOTtOO CD CD CD CD CD CD CD CD O "T3 a; ^if>Ul> O OO COOGOCOCOOOO O CO rH CO O O ") O5 1C rH t^ O _ rH rH O5 rH l> If3 rH (M CO >O lOrHOOOOOO lOOOOOOOOO * Ol O rH rH rH 1C s: i O O O O O < i O O O O O < 00 lO O C O O < CD O N- CO O O < O5 r)H O5 CO 00< ^ * 00 O5 i-t 160 BULLETIN No. 230 [November, 8 3 o JS O g ls fl 82^ gS3|5 O C5 coooio-^oip OJ CO i I i I U3 i I rH CDlOl IOOOO -; O O 'OO iSi o o o o o 00 Tf O O O O5 O "J 1C -l> 3Tti O 00000 N O O O O 1C CO t^CO t CO CO CO CO CO id it>-t-r} cs 6-g ^ p^ c3 f3 O S ^^3 r 3 fe aS O ^ o3 m O'C C . 03 O O O O O O O O O O ooooo O O O O O (M O O O O * IMO I-H CO i-H CO Tf* CO Oi O5 O r-lO} CO i 1 i ( i 1 O5 W CO i-H CO t^ t^ Oi OS i-l 1 1C O OOI> IN O Ci-HN. CO CO CO CO i CO ^O ^^ Oi CO t^OOOSO ( OO- T-H (N CO >O CO O OO TJH t^ OC35|- M* rt< IOCO 3 ^ G 8 v 1 i S fc|S O -c I a r x o'C c . a/ oj y fcj.s G plj "" CO o |i 49 c '5 7 S 8 g^ a ssl 5 0) &i o B"j 6| a 1 J-S 00 -o 0) iri a X W 03 "S 2 "i? -f-3 G ^j ^ .. O C S fe 7 ^ 5 CP P^ G O *S'C G 6-S IM o a I 1 t^ CD O 00 O (M CO id GO i i CO CD t^ i 1 CD O5 t^ i i O O O i-H (M (M l> O> O O O O O O O OO 00 (M CO lO CD C5 I-H t^ (M d Tt< !>. OO i I CD CO id J^ (M (M CO CO * 00 00 OJ G 3 SO O O O oooo i O O OO o G 1-5 G 1-5 CO dO d TjH id CD t^ OO id >d id c id CO d CD 10 CO CO id id CO O (M t- CO CO t^ 00 co id i- co >o C3 (N 1-1 CO CO * id OS IIIIIIH OOO OO O oooo oooooooo i i oo >d -*i i i >d >d oooo oooo I-H id O O 0000000 o o o o o o o COfNt^-OOCOCOCD (M t^ 00 rt< 00 IM 00 CO CO * id ^ id co d CD -^ o m Id CO t^ 00 03 rH (M 6 gl 1920] ELIMINATION OF GERMS FROM DAIRY UTENSILS 163 NUMBER OF BACTERIA IN UNTREATED CANS The results from the examinations of the check cans were quite in accord with the results given in Bulletin 204 of this station (pages 222-239) in that the number of bacteria present in apparently similar cans varied to an astonishing degree. It is not uncommon to find a can which has twenty times more bacteria than other cans which have had apparently identical treatment. Accordingly the number of bac- teria found in the untreated cans is only a rough measure of the bacteria that may have been present in the treated cans before they were exposed to the sun and air. Of the fourteen check cans, eight had more than one hundred mil- lion bacteria each, and the smallest number was 7,100,000 bacteria. The average for the fourteen cans was 133,314,111 bacteria. If these cans had been filled with milk at the time they were examined they would have added to it an average of 4,385 bacteria per cc. of milk. In connection with other studies on utensils, several hundred in- dividual utensils, mostly cans, have been examined. All the utensils were washed in a similar manner and by the same operator as those used in these studies, so that they may be taken to represent in a measure the condition of the utensils in the present study before they were placed on the rack. The number of bacteria found in the cans first referred to, which were examined soon after they were washed, was invariably much larger than the number found in the check cans in this study: for example, a set of fifty cans would have added 87,059 bacteria per cc. of milk; and another set of thirty-two cans would have added 47,863 bacteria per cc. of milk (Bulletin 204, pages 222-224). All of these examinations point to the conclusion that the cans selected as checks in this study contained much smaller numbers of bacteria than the average freshly washed can. Accordingly this treat- ment of the utensils, namely, to invert them on the rack so that they are exposed to the air and the sun, undoubtedly brings about a more decided reduction in the germ life in the utensils than is indicated by comparison with the numbers found in these check cans. BACTERIA IN CANS AND PAILS AFTER EIGHT HOURS OF EXPOSURE TO SUN AND AIR The number of bacteria in the cans and pails after they had been exposed to sun and air from 8 a. m. to 4 p. m. varied widely. The smallest number found in a can was 10,000 and the largest was 103,000,000. Seventeen percent of the cans and 10 percent of the pails had less than 100,000 bacteria; 19 percent of the cans and 14 per cent of the pails had more than 100,000 and less than one mil- lion bacteria ; and 64 percent of the cans and 76 percent of the pails had more than one million bacteria. The average for all the cans was about 24,000,000 and for the pails about 10,000,000 bacteria. 164 BULLETIN No. 230 [November, The importance of these cans and pails in milk contamination may be shown by calculating how many bacteria they would have added to the milk poured into them. Such calculation shows that 26 percent of the cans and 14 percent of the pails would have added less than 10 bacteria per cc. of milk ; 25 percent of the cans and 15 percent of the pails would have added more than 10 and less than 100 bacteria; 30 percent of the cans and 50 percent of the pails would have added more than 100 but less than 1,000 bacteria ; and 13 percent of the cans and 21 percent of the pails would have added more than 1,000 bac- teria per cc. The average contamination by these cans would have been 385 bacteria per cc. of milk and by the pails 848 bacteria. These calculations are based on the assumption that each utensil is filled with milk but once. In actual operations each pail is commonly used in milking several cows, and thus is filled a number of times, so that the number of bacteria added to the milk by the pails in practice would be smaller than the above calculated number. A comparison of these results with those from the check cans shows that there were eleven times as many germs in the check cans as in those cans which were kept on the rack for eight hours. Assuming that the cans which were exposed to the sun and air had approxi- mately the same germ life before they were placed on the rack as the check cans, it is evident that a decided reduction in the germ life in the cans was brought about by this treatment. BACTERIA IN CANS AFTER TWENTY HOURS' EXPOSURE TO SUN AND AIR The data given in Table 5 show that in the cans of May 22, 28, 29, 31, and June 1, 4, 5, and 6 more bacteria -were found after twenty hours than in the corresponding cans after being held for eight hours. It is of significance in this connection that rain fell 1 on each of these nights except June 1 and on this night the relative humidity was 80 at 7 p. m., with the practical certainty that this increased as the tem- perature fell during the night, it being 90 at 7 a.m. the following morning. Under such meteorological conditions a deposition of mois- ture on the surface of the cans would readily occur and thus produce conditions favorable to germ growth. During the nights of May 23, 24, 25, and June 7 and 8, when there was no rain and the relative humidity was much lower, the germ content of the cans held for twenty hours, as compared with those held eight hours, remained fairly con- stant or continued to decrease. Of the 101 cans kept on the rack from 8 a. m. to 4 a. m. the fol- lowing morning, 19 percent would have added less than 10 bacteria per cc. of milk, 26 percent more than 10 and less than 100, 28 percent more than 100 and less than 1,000, and 27 percent more than 1,000. *Data furnished from the record sheets of the Local Volunteer Weather Observation Station thru the kindness of Prof. J. G. Hosier. 1920} ELIMINATION OF GERMS FROM DAIRY UTENSILS 165 The average contaminations per cc. of milk, if all the cans had been filled, would have been 1,303 bacteria. BACTERIA IN DRY CANS AND IN MOIST CANS While the observations summarized in Table 5 were made upon cans held under conditions identical with those to which cans are exposed in practice, it was somewhat difficult to interpret the results because the humidity of the air varied so widely. Likewise the amount of germ life present in the cans before treatment could not be sat- isfactorily determined. For the purpose of supplementing this data, a test was made dur- ing January and February, 1919, with eight-gallon cans which had been so thoroly steamed as to render them free of germ life. After they were cool and dry there was added to each can 10 cc. of wash water or rinse water which was taken from vats in which milk uten- sils had just been cleaned and the germ content of which had been carefully determined. After adding this liquid to the cans they were covered and shaken vigorously to distribute the material over the inner surface. This volume of liquid was chosen because observation had shown that about this quantity of liquid usually remains in well- drained cans. Six eight-gallon cans were treated in this way on each of eight days. On each day, immediately after the liquid had been thoroly distributed in the cans, the covers were removed from three cans, which were left lying on their side, but left on the other three cans. The cans were then held for twenty-four hours in a room with a tem- perature of approximately 70 F. and a relatively low humidity, ordi- narily between 40 and 60. The bacterial life found in these two groups of cans at the end of twenty-four hours, determined according to the methods described on page 141, is recorded in Table 6. To facilitate comparisons of the results, the cans are grouped in the order of the increasing amount of original inoculation added to the cans. Perhaps the most evident point in the data in Table 6 is the lack of any apparent relationship between the extent of the original in- oculation placed in the cans and the amount of germ life found at the end of twenty-four hours. This is equally evident in the cans from which the covers had been removed and in those on which the covers had been left. This suggests that the final germ content of cans held for a period of twenty-four hours during warm weather depends more upon the conditions under which the cans are held than upon their germ content at the close of the washing process. In all but three of the cans from which the covers were removed the germ life fell quite sharply during the twenty-four hours. On the other hand, in all the covered cans the germ life increased, the extent of the increase varying from 20 to 3,000 fold. 166 BULLETIN No. 230 [November, Of the twenty-four cans from which the covers were removed six- teen, if filled with milk, would have added to each cc. a germ content of between 10 and 100; four would have added between 100 and 1,000; three between 1,000 and 10,000; and one can 21,381 per cc. Of the twenty-four cans which stood with their covers on there were none which would have added a germ content below 10,000 per TABLE 6. CHANGES IN BACTERIAL LIFE IN COVERED AND IN UNCOVERED CANS DURING TWENTY -FOUR HOURS 10 cc. of rinse water added to each can at beginning of period No. of can Cans Not Covered No. of can Cans Covered Germs remaining in can Germs per cc. of milk due to can Germs remaining in can Germs per cc. of milk due to can Each Can Inoculated with 12,300,000 Bacteria 1 400 000 13.1 4 24 375 000 000 801 809.2 2 500 000 16.4 5 31 200 000 000 1 026 315.8 3 85 000 000 2 796.0 6 35 350 000 000 1 162 829.0 Each Can Inoculated with 42,200,000 Bacteria 7 8 9 1 9 25 800 000 000 000 59 2 8 2 10 11 12 1 610 000 000 1 700 000 000 3 500 000 000 52 55 115 960.5 921.0 131.5 Each Can Inoculated with 43,900,000 Bacteria 13 14 15 1 35 81 400 300 750 000 000 000 46 1 161 2 689 1 1 16 17 18 2 600 000 000 2 680 000 000 2 700 000 000 85 . 88 88 526.3 157.9 815.7 Each Can Inoculated with 46,900,000 Bacteria 19 20 21 1 4 5 150 875 050 000 000 000 37 160 166 8 3 1 22 23 24 7 000 000 000 9 300 000 000 11 800 000 000 230 305 388 263.1 921.0 157.9 Each Can Inoculated with 55,600,000 Bacteria 25 26 27 400 550 900 000 000 000 13 18 29 1 6 28 29 30 3 740 000 000 4 920 000 000 5 220 000 000 123 161 171 026.3 842.1 710.5 Each Can Inoculated with 76,500,000 Bacteria 31 32 33 1 2 500 000 000 000 000 000 16 32 05 4 8 7 34 35 36 1 500 000 000 2 300 000 000 2 700 000 000 49 75 88 342.1 657.9 815.7 Each Can Inoculated with 82,000,000 Bacteria 37 38 39 1 280 405 700 000 000 000 9 13 55 2 3 9 40 41 42 3 350 000 000 4 750 000 000 7 150 000 000 110 156 235 197.3 250.0 197.3 Each Can Inoculated with 115,500.000 Bacteria 43 5 600 000 184.2 46 2 200 000 000 72 368.4 44 44 000 000 1 447.3 47 2 350 000 000 77 302.6 45 650 000 000 21 381.5 48 4 400 000 000 144 736.8 ELIMINATION OF GERMS FROM DAIRY UTENSILS 167 cc. of milk; seventeen cans would have added a germ content of be- tween 10,000 and 100,000 ; five cans a germ content of between 100,000 and 1,000,000; and two cans a germ content of over 1,000,000 per cc. These results may be summarized by saying that the worst of the open cans showed less than one-half the germ content of the best of the cans which were covered. Had all the cans been filled with sterile milk, that in the covered cans would have had an average germ con- tent of 247,772 per cc. ; while similar milk in the open cans would have had an average germ content of 1,284 per cc., and two-thirds of these germs would have come from a single one of the twenty-four open cans. Since the six cans used each day were practically identical except in the matter of moisture the marked differences in final germ con- tent may be attributed to differences in moisture. In the closed cans the moisture could not escape, the air promptly became saturated, and the conditions for the growth of bacteria became good over the entire inner surface of the can. The amount of germ life present in these moist cans at the end of twenty-four hours seemed to depend mainly upon the vigor of the germs present and upon the amount and char- acter of food available to them. In the open cans evaporation began at once. Ordinarily the cans became apparently dry within a few hours. However, in a few cans, probably because of the uneven distribution of the moisture, the drying was materially retarded and at least a few drops of water remained in the cans at the end of twenty-four hours. For example, when Cans 13, 14, and 15 were tested at the end of twenty-four hours, Can 13 seemed quite dry, while moisture was evident in Cans 14 and 15. The examination of these cans indicated a germ content in Can 13 of 46 per cc., while Cans 14 and 15 had a germ content of 1,161 and 2,689 per cc. respectively. Again, the day on which Cans 43, 44, and 45 were tested was damp and rainy, and at the end of twenty-four hours moisture was evident in all of the cans but was most pro- nounced in No. 45. The germ content found in these cans was 184, 1,447, and 21,381 per cc. respectively. These observations of the presence of moisture in certain cans account for all the comparatively high numbers found in the cans from which the covers had been removed, except in the case of Can 3. The notes do not show that any moisture was present at the end of twenty-four hours in this can, but the whole trend of the data makes it highly probable that the can, for some reason, dried very slowly. IMPORTANCE OF DRYNESS IN CONTROLLING (TERM LIFE IN UTENSILS The main fact which stands out distinctly as the result of the studies reported in Part II is the overshadowing importance of dry- ness as a means of reducing and keeping down germ life in utensils. 168 BULLETIN No. 230 Pails and cans given such a washing as is practicable even on the farm, where steam is not available, will add to the milk later put into them less than 100 bacteria per cubic centimeter, provided they are promptly and thoroly dried and kept dry until used. "When the sun is hot and the air dry, the exposure of the utensils, with the covers off, to the heat of the sun is a satisfactory treatment. However, when the weather is rainy, exposure to the damp air does not lead to quick and thoro drying and under such conditions the germ life in the utensils may hold its own or even increase in number. GENERAL DISCUSSION From information kindly furnished by milk companies in various cities it is evident that the morning's milk as it reaches their bottling plants or shipping stations in warm weather rarely contains less than 50,000 bacteria per cc. and occasionally exceeds 1,000,000 per cc. It should be remembered that the interval between milking and deliv- ery in these cases is so short that little growth has occurred. Accord- ingly the above large numbers of bacteria indicate the amount of seed- ing to which the milk is normally exposed under present conditions. The studies described in Bulletin 204 make it evident that the seed- ing of the milk under normal conditions comes principally from the utensils in which the milk is handled. It further points out that among the utensils coming into contact with the milk up to the time of its delivery to the milk plant, the cans are ordinarily the principal source of the bacteria added to the milk. The present publication points out that the amount of germ life in milk cans in warm weather twenty-four hours after they have been fairly well washed is controlled principally by the moisture which re- mains in the washed cans. Observations made upon the milk cans as sent out by a consider- able number of the leading milk companies show that a considerable proportion of these cans are moist as returned to the producers. Ac- cordingly in these cans the conditions are favorable for the develop- ment of an amount of germ life which will fully account for the seed- ing which the milk ordinarily receives before it reaches the milk plant. The present publication further points out that when these high germ content cans reach the farm a rinsing with liberal amounts of water at or near the boiling point will so reduce the number of germs in them that if used immediately they will ordinarily add only about 100 bacteria per cc. to the milk. It also points out that if such of these cans as are not needed immediately are promptly and thoroly dried and kept dry, the germs in them will not grow but will con- tinue to decrease, and the cans when used will have little effect upon the germ content of the milk. UNIVERSITY OF ILLINOIS-URBANA