T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
862
column into t h e scale portion of the neck with more hot water a n d centrifuge I minute. Reduce t h e meniscus with glymol and multiply t h e reading by 4. DRIED mLK-Add concentrated sulfuric acid, j t o 6 cc. a t a time, until a dark brown, almost black, color develops. Prepare a mixture of sulfuric and nitric and add acids ( I O cc. of H2S04 t o 2 . 5 CC. of ” 0 3 ) , 0 . j cc. a t a time, shaking thoroughly after each addition, until a clear, light red color results (this requires ordinarily j t o 6 cc. of t h e mixture). Immerse in boiling water until t h e dark brown color returns, add more of t h e mixture until t h e light red color is produced, and again immerse in t h e boiling water until t h e dark brown color appears. Centrifuge j minutes a n d proceed as for malted milk. DISCUSSIOK O F RESCLTS
The results obtained b y this method on different products have been compared with those obtained b y t h e Roese-Gottlieb method, as indicated in t h e accompanying table. COMPARATIVE TESTS-PERCENTAGES OF FAT ROESB-GOTTLIEB AUTHORS’ METHOD METHOD
SAMPLE
Ice C r e a m . . ..................... Ice Cream. . . . . . . . . . . . . . . . . . . . . . . Unsweetened Evaporated M i l k . , . , , Sweetened Condensed Milk., . , , , , , Malted Milk . . . . . . . . . . . . . . . . . . . . . Dried Skim M i l k . , . . , . , . , , , , , , , , ,
7.7
7.58 7 .93 9.35 7.26 1.36
...
7.68 7 .90 9.35 7.14 1.29
...
7.7 8.0 10.0 7.4 1.2
..
7.7 8.0 10.0 7.4 1.2
I t appears t h a t our method gives uniformly higher results t h a n t h e Roese-Gottlieb method. I n every case a good separation of fat was obtained free from particles of carbon and bubbles. I t is important t h a t t h e nitric acid be added with care and a n excess avoided; too much nitric acid may cause t h e formation of a large quantity of gas, t h e gas bubbles continuing t o rise for some time and producing a froth which interferes with t h e reading of t h e fat column. The action of t h e acid is more rapid when added t o t h e warm mixture, a n d i t is suggested t h a t t h e bottle be immersed in warm water while t h e acid is being added. S L-M LIAR Y
A method for determining f a t , including treatment of t h e samples, in such dairy products as ice cream, evaporated milk, malted, dried skim milk and similar milk products, is described. The procedure is similar t o t h a t followed when using t h e well known Babcock test, b u t in place of sulfuric acid, mixtures of glacial acetic, sulfuric and nitric acids are prescribed. The fat is separated and read in a Babcock bottle. CHEMICALDEPARTMENT OKLAHOMA AGRICULTURAL EXPERIMENT STATION STILLWATER, OKLAHOMA
APPLICATION OF THE CRYOSCOPIC METHOD FOR DETERMINING ADDED WATER IN MILK’ B y J. T. KEISTER Received M a y 3 1 , 19 17
Later J. Winter’ reported results of similar work and found t h e freezing point t o vary from - 0 . j 4 t o -0. j 7 ’. The freezing points determined by Bordas a n d Genin2 showed for normal milk as great variation as -0.44 t o -0.56’ C. Hamburger3 found t h a t t h e freezing point of “fore” milk a n d “strippings” differed. Pins4 states t h a t evening’s milk showed a slightly higher freezing point t h a n morning’s milk, although his findings were not confirmed by Abati and Sohn5 who found no difference. \‘an Eck6 showed t h a t too low temperature of t h e freezing b a t h a n d excessive supercooling of t h e milk raised t h e freezing point. Stroecklin’ found t h a t t h e acid content and the-freezing point of t h e sour milk could not serve as a means of accurately calculating t h e freezing point of t h e fresh milk. Koeppe,s Van R a ~ u l t Reicher,Io ,~ Henderson a n d hlestonll and others have published results of freezing-point determinations of milk where results are reported quite similar t o those recorded in t h e work mentioned here. Monier-Williamsl* showed t h a t t h e presence of fat is without influence in t h e determination of t h e freezing point. He concludes further t h a t neither t h e proportion of total milk solids nor t h e non-fatty solids exerts any influence. The most striking fact found by a n examination of t h e results of t h e freezing-point determinations of milk in t h e literature is t h e slight variation in readings made by t h e different investigators. S o determination of any constituent of milk has given such closely agreeing values as t h e freezing points recorded in t h e literature by t h e different experimenters. The range in temperature is from-0. j 4 t o - 0 . j j ’ C., in general. Hence, t h e value of this determination for detecting added water in milk is evident. One of t h e chief difficulties encountered in t h e -interpretation of t h e results of milk analysis is t o distinguish between rich milk containing a small amount of added water a n d naturally thin milk. The freezing-point determinations appear to serve as a n index for making this distinction in almost every case. The freezing point of milk is evidently controlled by substances in solution. Substances like fat exert no influence on t h e freezing point a n d as fat is t h e most variable constituent of milk t h e most widely varying factor is removed. On this point the majority of observers have agreed, though there are some conflicting s t a t e m e n u . I t has been shown t h a t t h e freezing point is independent of breed, age of cow, period of lactation or quantity of milk. I t is also claimed t h a t substances which are in colloidal condition, such as albuminoids, affect t h e freezing point either not a t all or t o a very slight extent. I n a n y 1 Compt. rend., 121 (1895), 696; 123 (1896). 1298; 124 (1897), 777. ZIhid., 123 (1896), 425. 3 .Won. sci., 666 (18971, 462.
Dissertation, Leipzig. 1910. Milchzeitung, 28 (1899), 177. 6 Chem. Weekblad., 12 (1914). 108. 7 Ann. fais., 4 (1911), 232. 8 Jahr. Kinderheilk., 47 (18981, 398. 9 Chem. Weekblad., 11 (1914), 206. 10 Ihid., 11 (1914), 323. 1 1 Proc. Roy, SOL.Queensland, 24 (19131, 168. 12 Report Local Govt. Board Pub. Health (Great Britain), Food Reporf 22 (1914). 4
6
Beckmann a n d Jordis2 determined t h e freezing point of different samples of normal milk a n d found t h a t t h e freezing points varied from -0.j; to- 0.56’ C. 1 Published b y permission of t h e Secretary of Agriculture; read a t t h e meeting of t h e American Chemical Society, New York City, September 25 t o 30, 1916. Forschungsherichl uher Lehensmittel, 2 (1895). 367.
Vol. 9, No. 9
T H E JOC‘R.VAL O F I N D C S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
Sept., 1917
T A B L E I-ANALYSIS
Age of BREED Holstein., . , ,
Cow No. , . . , 1
,
cow Yrs. 5
Time since Calving 11 mos.
AND
FREEZING POINT
Milking Time h-oon
Milk Lbs.
P.M.
2
3
9 mos.
3
9
6 mos.
4
7
......
A.M. Noon P.M.
Grade Holstein.. Holstein., . ,
,,
,
...
,,,,
Grade Holstein., , .
.
A.M.
.........
P.M.
5
3
5 mos.
6
4
5 wks.
P.M.
i
,
21/2 mos.
A.M. P.M.
A.M.
8
7 9
Holstein..
P.M. A.M.
A.M. P.M. A.M.
9 10
Grade Holstein. . . . . I 1 Holstein.. . . . . . . . . . 12
1 mo. l l j l mos. 2 mos.
A.M. P.H.
P.M. A.M.
3
-
4 mos.
P.Y. A . %I.
3 mos. 21/2 mos.
P.M.
Noon P.M. h.M. P.M. P.M. P.M.
5 weeks P a r t Tersev.. . . . . . . . 13 8 or 9 5 mos. P a r t Durh-am . . . . . . . 14 6 or 7 8 or 10 mos. P a r t Jersey.. . . . . . . . 1 5 2% 1 mo. P.M. P a r t Jersey. . . . . . . . . 16 18 or 20 Herd M i l k . . . . . . . . . . . . . . . . . . . . . . . . . . . . ................... Herd Milk with 5 per cent added water . . . . . . . . . . . . . . . . . . Herd Milk a i t h 8 per cent added water. . . . . . . . . . . . . . . . . . .................. hIarket Milk . . . . . . . .
case, as t h e molecular weight of these substances is 1-ery high their relative effect upon t h e freezing point would be very small. T h e only constituents, therefore, t h a t exert a n y appreciable influence upon t h e freezing point of milk are lactose a n d t h e soluble salts. which later are composed largely of t h e chlorides of t h e alkali metals. I n this connection it should be noted t h a t there appears t o be a more or less constant relation between t h e amounts of lactose and sodium chloride present in t h e milk. When there is a n y increase of lactose a decrease of sodium chloride is found. and v i c e w r s a . Therefore, there is a tendency for a n y influence exerted on t h e freezing point b y these substances t o be balanced. This is shown b y t h e work of L. IT. Ferris of this laboratory in a separate paper under t h e title “ T h e Detection of Added Water in Milk b y hIeans of a Simplified hlolecular Concentration Constant.”’ t h e in\-estigation being made a t t h e same time a n d upon t h e same samples as were used in t h e freezing-point measurements here reported. ,Another factor which exerts a n influence upon t h e freezing point of milk is t h e per cent of acid in t h e sample v h i c h , if present in considerable a m o u n t , might lead t o erroneous conclusions. as with increased acidity of t h e milk a n increased depression or lowering of t h e freezing point takes place. Therefore, i t is necessary t o make t h e test on samples before a n y considerable amount of acid has developed if confidence is t o be placed in t h e results. I n the work recorded here t h e samples of milk were authentic and represent portions from complete milkings, both morning and evening, in every case except cows 1-0s. I and 2 . which were milked morning, noon and evening. The freezing points of t h e separate samples were determined as recorded in t h e tables a n d the maximum variations of t h e freezing points of milks from individual cows are shown. T h e freezing point figures in Table I represent t h e mean of two and sometimes of three readings, t h e usual 1
T o appear in next issue
of
THISJOURNAL.
863
O F MILK O F I N D I V I D U A L C O W S P e r cent Per cent F a t Per cent Per T o t a l Roese-Gottlieb Solidscent Solids Method not-Fat Ash 3.49 8.67 0.79 0 .86 3.31 8.54 3.10 8.39 0.80 3.95 8.77 0.67 3.26 8.95 0.72 0.71 3.18 8.98 0.73 3.54 i.63 0.69 3.63 7.83 0.71 3.24 9.08 0.67 3.41 9.14 2.32 0.6i 8.04 3.4i 0.72 8.03 4.20 0,io 9.20 4. l i 8.51 0.71 0.75 3.46 8.00 3.40 i.i2 0.i2 i.71 0.i2 3.14 7.51 0.69 3.11 0.73 8.85 4.99 4.05 8.84 0.72 3.31 9.00 0.69 3.62 8.70 0.74 0.63 4.12 8.84 8.57 0.il 4.01 3.46 8.49 0.69 3.0i 8.,52 0 72 4.43 8.18 0.72 4.65 9.52 0.76 6.80 9 . 10 0.74 0.69 7.80 3.87 0.73 3.89 8.73 8.41 3.66 0.il 3.46 0.68 8.20
FREEZING POIKTI N
O C. Milk with Added D L , ~ Milk Water ( 5 % ) erence -0.553 .....
-0.551
-0.544 -0.558 -0.544 -0.547 -0.552 -0.558 -0.548 -0.563 -0.548 -0.560 -0.570 -0.559 -0.563 -0.550 -0.559 -0.555 -0.559 -0 . 5 5 9 -0.ji3 -0.562 -0.541
-0.574 -0,566 -0.576 -0.546 -0.557 -0.552 -0.jSO -0.563 -0.534 -0.514 -0.549
...
.....
... ... ...
.....
.....
...
.....
.....
...
..... -0.530 -0.522 -0.537 -0.517
0:029 0.021 0.025 0.029
-0.534 -0.541 -0.52’ -0.524
0:032 0.035 0.024 0.033
.....
.....
..... .....
...
.....
-0.52
I
range in variation of readings on t h e same sample being t o 0.003. I t should be noted t h a t Samples 3 , j. 7 , 8 and 16 represent t h e lowest grade or thinnest milk t h a t could be found, having in mind when obtaining t h e m t h e difficulty encountered in ordinary milk analysis in distinguishing between naturally thin milk and rich milk containing small amounts of added water. T h e results of these five samples are shown not t o give abnormal freezing-point figures. T o obtain accurate results, it is very necessary t o control t h e amount of supercooling. .Is a result of some experiments on the effect produced b y different degrees of supercooling, it was found t h a t b y reducing the supercooling t o the extent of approximately 0 . j o t h e freezing point vias lowered approximately and uniformly 0.01’. The results are recorded in Table 11. The samples used in these experiments were fresh market milk. 0.001
TABLE11-THE
EFFECT O N THE FREEZING POIKT O F FRESH~ I A R K E T O F DIFFEREKT DEGREES O F SUPERCOOLIKG Freezing Freezing Test Degrees of Point Degrees of Point Di.feiPvence h-0. Supercoolinc ‘C Supercoolinr O C c. l . . . . . 0.9 to1.05 -0,553 0.5 t o 0 . 6 -0.54’ 0.011 -0,553 -0.545 0.008 -0.552 -0,545 0.007 2 . . . . . 1.0 t o l . l -0.552 0.5 toO.7 -0.56.3 n.011 -0.552 -0.564 0.012 3 . . . . . 1.0 -0.553 0.5 to0.6 -0.563 0.010 -0.564 -0.553 0.009 4. . . . . 1.3 to1.4 -0.559 0.5 t o 0 . 6 -0.568 0,009 -0.556 -0.570 0.014 -0.570 5 . . . . . l . 0 0 t o 1.1 -0.551 0.5 to0.6 -0.556 0.005 -0.55 1 -0.556 0.005 -0,550 3fIl.K
As t h e amount of acidity developed in milk has been found t o influence its freezing point, a few determinations were made on milks in which different degrees of souring had developed. T h e results of these determinations are recorded in Table 111. The samples used in these experiments mere pasteurized market milk of presumably normal composition. except in t h e case of Sample 3 which shows t h e presence of added water. These results shorn t h a t t h e freezing point is, on a n average, lowered approximately 0 . 0 0 3 ’ for each 0.01per cent increase of acidity. The figures
864
T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
Vol. 9 , No. g
here recorded represent t h e mean of two a n d sometimes of three readings, t h e range of variation in readings of t h e same samples being 0.001t o 0.003'.
raised I or 2' a n d all ice crystals were melted. These operations were repeated and t h e freezing point taken two or three times. T h e freezing point of recently TABLE 111-EFFECTOF ACIDITY UPON THE FREEZING POIXTOF PASTEURIZEDboiled distilled water is determined in exactly t h e MARKETMILK same manner as in t h e case of milk; t h e difference Per cent Freezing Lowering of Piverage Lowering between t h e reading obtained on distilled water a n d Sample Acid Point Freezing Point for each No. as Lactic C. Due to Acid O . O l ~ oof Acid t h a t on milk represents t h e freezing point of t h e sample. 1. . . . . . 0.15 -0.545 0.18 -0.548 It is necessary t o check u p t h e freezing point of dis0.42 -0.637 0.092 0.0030 2 . . . . . . 0.15 tilled water frequently as slight variations occur due t o -0.539 0.18 -0.548 changes in t h e glass. 0.34 -0.602 0.063 0.0037 3...... 0.18 0.21 0.24 0.27 4......0.15 0.17 0.20 0.46 S...... 0.16 0.18 0.22
-0.496 -0.515 -0.522 -0.536 -0.552 -0.555 -0.558 -0.636 -0.541 -0.546 -0.564
D I S C U S SI 0 N 0.040
0.0044
0.084
0.0027
0.023
0.0038
DESCRIPTION O F APPARATUS
The apparatus used consisted of a n ordinary Beckmann thermometer graduated t o 0.01' divisions a n d a sample t u b e made from tubing about 2 2 mm. in diameter a n d 1.5 mm. thick. The entire t u b e is about 19 cm. long a n d has t h e upper length of about 3.5 cm. widened t o 30 mm. diameter t o carry a stopper which serves t o hold t h e Beckmann thermometer a n d t h e usual wire stirrer. This t u b e below t h e collar was fitted into a large test t u b e t o allow t h e smallest amount of air space between t h e walls. A piece of t h i n walled rubber tubing was fitted over t h e top of t h e outer tube, making a tight joint a t t h e shoulder of t h e inner tube. A glass jar of approximately two quarts capacity held t h e freezing mixture, t h e jar being wrapped in cotton or other material t o assist in maintaining low temperature and a box of suitable dimensions with a n opening in t h e t o p through which t h e thermometer passed, provided with a sliding shelf in t h e box at proper point for t h e easy removal of jar. A s t a n d held t h e thermometer in position a n d a lens aided accurate readings. PROCEDURE
Fifteen t o twenty cc. of t h e sample of milk previously cooled nearly t o t h e freezing point were placed C., a in t h e tube. T o obtain a temperature of -4' freezing mixture of shaved ice and salt was prepared i n t h e proportions of 4 0 t o 50 g. of salt : 900 t o 1000 g. of ice mixed with sufficient water t o fill all air spaces. T h e thermometer was placed in t h e sample, and t h e t u b e lowered into t h e freezing mixture so t h a t all of t h e milk was below t h e surface of t h e freezing bath. When t h e mercury column was I O below t h e t r u e freezing point t h e sample was stirred gently a t t h e rate of two strokes per second t o break u p a n y ice particles formed. Too rapid stirring must be avoided t o prevent t h e generation of heat mechanically. A rapid rise of t h e mercury column was noted. T h e highest point i t reached was slightly above t h e freezing point. T h e n slow uniform stirring was continued until t h e mercury remained stationary for one minute when t h e reading was taken. T h e thermometer a n d t u b e were t h e n removed a n d warmed with t h e heat of t h e hand or in water a t 40' C. until t h e temperature of t h e sample
SUPERCOOLING-ASalreay stated, i t is very necessary t o control t h e supercooling in order t o obtain concordant results. T h e results of experiments t o determine this demonstrate t h e necessity of keeping t h e amount of supercooling uniform a n d within narrow limits a n d further emphasize t h e importance of uniformity a n d proper control of conditions in all work of this character. This was also proved b y Bordas a n d Genin in their report of work on t h e freezing point of milk. It should be stated t h a t experience with t h e form of apparatus used b y t h e writer showed t h a t a supercooling of about I t o 1.2' was t h e most satisfactory a n d t h e results recorded in this paper represent readings made with a n average supercooling of approximately I. I '. Readings made with milk when t h e supercooling was about 0 . 5 t o 0.6' were not satisfactory because of t h e tendency of t h e mercury column t o d a r t u p a n d down a n d not remain stationary, as when a greater amount of supercooling was allowed. It may be safely stated, therefore, t h a t it is not necessary in case of milk t o keep t h e a m o u n t of supercooling down t o t h e least possible amount so long as uniformi t y is maintained. An amount of supercooling of I t o 1.3' was used in this work since it seemed t o give t h e best working conditions a n d also concordant results. AcIDITY-It was found t h a t a n increase of acidity from 0.15 per cent, which is about t h e average figure for fresh milk, t o 0.24 per cent, which increase is not a t all unusual, lowered t h e freezing point about 0.z 5 t o 0.3 per cent which approaches t h e amount of lowering produced b y t h e presence of 5 per cent of added water (see Table I). This same effect was found t o be produced with samples preserved with formaldehyde. Therefore, a sample of milk in which a considerable amount of acid has developed or which is preserved with formaldehyde a n d containing a small percentage of added water might easily pass as normal milk b y t h e freezing-point test, A too rapid stirring of t h e sample while freezing must be avoided t o prevent t h e generation of heat. Our experience is in accordance with t h a t of other observers on this point, v i z . , t h e stirring should be so regulated as t o reduce t h e amount of heat generated t o a minimum a n d a t t h e same time be sufficient t o break up any ice crystals as t h e y are formed. By rapid stirring a rise of 0.01' or more in t h e freezing point may easily be effected. A rate of one or two strokes of t h e stirrer per second is most satisfactory, which condition is determined largely by t h e behavior of t h e mercury column.
Sept., 1917
T H E J O U R N A L OF IA'DUSTRIAL A N D ENGINEERlNG CHEMISTRY CONCLUSIONS
I-The freezing-point figure of milk is t h e most constant one yet obtained a n d t h e safest basis upon which t o draw conclusions as t o t h e presence or absence of added water. 11-The freezing-point figures on t h e milks of sixteen individual cows examined show t h a t t h e presence of 5 per cent added water can be detected in t h e majority of cases b y this method a n d in a n y case place t h e sample in t h e suspicious class. 111-The presence .of water added t o fresh milk in excess of 5 per cent can be detected with certainty b y t h e freezing-point measurement. T h e use of sufficient formaldehyde for preservation was found t o lower t h e freezing point. IV-It is essential t h a t t h e test be applied only t o reasonably fresh milk as t h e presence of acidity t o t h e extent of 0.1 per cent beyond t h e normal for fresh milk (0.15 per cent) counteracts t h e amount of decrease in t h e freezing point depression produced b y t h e presence of approximately 5 per cent of added water. V-The method is practical in milk control work i n t h a t t h e test need be applied only t o samples of doubtful character. DAIRYLABORATORY AND LABORATORY OF FOODCONTROL BUREAUOF CHEMISTRY. WASHINGTON, D . c.
INSOLUBLE PHOSPHORIC ACID IN ORGANIC BASE
GOODS By E, 0. THOMAS Received June 2, 1917
865
sults given in Table I on insoluble phosphoric acid when deviating from t h e Official hIethod. 500 Ibs. Base Goods (1.56% Insol. PzO6) 900 Ibs. Acid Phosphate No. 2 (0.12% Insol. PPOS) 520 lbs. Kanona Nitrogene 80 lbs. Muriate of Potash
T h e calculated amount of insoluble phosphoric acid in t h e complete fertilizer mixture is 0 . 4 4 per cent, or 0.09 per cent more t h a n t h a t found. As t h e Official Method is shown t o t a k e care of all t h e avaiIable phosphoric acid in acid phosphate i t is fair t o TABLEI-RESULTS ON 2-GRAM SAMPLES WITH VARYING AMOUNTS OF NEUTRAL AMMONIUM CITRATE SOLUTION, GRAVITY, I .09 (100 Cc. Neutral Ammonium Citrate Solution Used in Official Method) Cc. Neutral Per cent Insoluble Phosphoric Acid (PzO,) Ammonium Citrate Base Acid Phosphates Complete Goods No. 1 No. 2 Solution Fertilizer 1.56 1.41 0 . 1 2 0.35 100 1.44 0.09 0.28 133 1.44 1.35 0.10 0.26 200 1.38 1.42 0.08 0.28 400 1.05 1.37 0.08 0.20 800 0.88 1200 0.80 .... .... .... 1600 0.70 ....
....
....
assume t h a t this gain of 0.09 per cent available is derived from t h e 500 lbs. of base goods, or a gain t o t h e buyer of 0 . 3 6 unit of available phosphoric acid per ton. Leaving out of consideration t h e other materials in this mixture, of t h e 2. o g. taken for analysis, 0 . 5 g. is base goods which is acted on by IOO cc. of citrate solution; i. e . , t h e ratio of 2 . 0 g. base goods t o 400 cc. citrate solution. T h e insoluble phosphoric acid found in t h e b?.se goods using 4 0 0 cc. citrate solution (Table I ) was r.oj Der cent, a gain in available of 0 . 5 1 per cent over that found b y t h e Official Method. By following all details of t h e Official Method, except t h e using of I O O cc. of N / I O normal citric acid in place of t h e I O O cc. ammonium citrate as suggested b y Rudnick,' this sample of base showed 1.78 per cent insoluble phosphoric acid. Using I O O cc. z o per cent citric acid t o replace t h e ammonium citrate this same base showed I. 1 5 per cent insoluble. This latter figure is comparable with t h a t obtained when using 400 cc. citrate solution in t h e digestion, namely, I o j per cent, and is also comparable with t h e figures obtained when this garbage base is batched (viz., in complete fertilizer, Table I) not t o exceed 500 lbs. t o the ton. As shown in Table I, there is a gain of 0 . 3 6 per cent available phosphoric acid per ton of base goods when this material is batched at t h e rate of 500 lbs. per t o n of complete fertilizer and t h e Official Method is used. This gain of 0 . 3 6 per cent plus t h e I. 1 5 per cent found when using t h e 2 . o per cent. citric acid solution gives I . j~ per cent insoluble phosphoric acid, very close t o t h e amount found by t h e Official Method, v i z . , I . 56 per cent.
T h e Official Method' for t h e determination of t h e citrate-insoluble phosphoric acid in fertilizers while giving all t h e available phosphoric acid i n acid phosphate does not give all of t h e available in certain classes of organic materials. However, when this same material is batched into a complete fertilizer more available phosphoric acid (or what is t h e same thing, less insoluble phosphoric acid) is obtained, due t o a relatively larger amount of citrate solution acting on t h e material. Particular reference is here made t o base goods made b y acidulating garbage tankage without a n y addition of phosphate rock. By t h e Official Method this material ranges from 0 . 9 5 per cent t o 1.80 per cent insoluble phosphoric acid. T h a t t h e Official hfethod does not give t h e t r u e available phosphoric acid in this material is shown by t h e following experiments. A sample of this base goods containing 1 7 . 8 2 per cent moisture showed b y t h e Official Method 1 . 5 6 per cent insoluble phosphoric acid; b y increasing t h e amount of citrate soluCONCLUSION tion used per 2 . 0 g. of base t h e results in Table I The Official Method for t h e determination of citratewere obtained. T h a t t h e Official Method gives all of t h e available insoluble phosphoric acid gives the t r u e value of acid phosphate, but some modification should be used phosphoric acid in acid phosphate is shown by t h e for t h e analysis of materials of the character of acidtests (Table I) on two samples of acid phosphate. A complete fertilizer made on t h e following formula ulated garbage tankage. 8 14 PAUL-GALE-GREENWOOD BUILDINQ using no phosphatic material b u t base goods (Table NORFOLK, VIRGINIA I) and acid phosphate No: 2 (Table I ) gave t h e re1
Jour. A . 0.A . C., 1, No. 4, PEL?2, p. 4.
1
THISJOURNAL, 6 (1914). 486.
