RETENTION OF PHOSPHATE BY PEAT SAMPLES Armi Kaila University of Helsinki, Department of Agricultural Chemistry Received June 17, 1959 Since Way in 1850 (ref. 15) demonstrated that soils are able to retain soluble phosphate, the problem of the sorption of phosphate by soils and soil constituents has been dealt by a great number of scientific papers. Generally, these works confirm the early results. They give, however, a somewhat confusing picture regarding the reactions involved. Most of the studies have been made on mineral soils. Among the few works concerning peat soils the papers by Doughty (1, 2) and Kasakow (6) may be mentioned. These authors conclude that precipitation and physical adsorption are both functioning in the removal of phosphate from solution by peat samples. Doughty states that in the material studied the formation of iron, aluminum and calcium phosphates will account for the fixation of phosphorus under field conditions. According to Kasakow, the maximum retention of phosphate by peat occurs at about pH 2—3; owing to the higher content of iron, aluminum and calcium in the fen peat, the fixation of phosphate by this group is higher than thatby the bog peats. McCool (9) noted that the capacity of peat and muck soils to take up phosphorus increased with the mineral content and the degree of decomposition. Verhoeven (14) maintains that the retention of phosphate by irreversibly shrinking peat soils primarily depends on the mobile iron content of the soils. Larsen (8) found a positive correlation between the sesquioxide content of organic soils and their phosphate fixing capacity. Obviously, the retention of phosphate by peat soils may be connected with several factors. In the present paper an attempt is made to study the capacity of Finnish virgin peat soils to retain phosphate and the factors on which this capacity depends. There is no generally accepted way to determine the phosphate -sorption capa- city of a soil. It is a quantity which varies with the ratio of soil to phosphate solu- tion and with the phosphorus concentration of the solution. Other ions present also exert their effect, likewise the acidity, temperature, and time of connection. https://www.c-info.fi/en/info/?token=nY_KrAnJpbhGtBdc.yGgGu9DYLfxjTseUaTkAlA.IBxRIphNE3oAbU7hzstucqTxkxN3_n-QJzkSM1nBeQkK1tkIoKjTYHr1ddBaqEkyB1gyNiW_CgBVFANoV_dIp1TA-kJRRklzZ40_ZqlcBFF3b2gYj9xDwzLGiJUW9QnzmtGVt0HV_2s7GRbT7m4eca7F78CcVCnb 216 Therefore, one is compelled to choose a more or less conventional method which gives values that, at least, are correlated with the phosphate fixing capacity of the soil. Russel and Prescott (13) already found that the relationship between the amount of phosphate sorbed and the corresponding equilibrium concentration of phosphate in experiments generally comply with the Freundlich adsorption isotherm. Recently Olsen and Watanabe (10) and Rennie and McKercher (12) have emphasized the applicability of the Langmuir isotherm to the estimation of phosphate retention by soils. In this study the coefficient k in the Freundlich adsorption equation y =k X cn was taken to characterize the retention of phosphate by the peat samples (y = the amount of P adsorbed, mg/kg of soil, c = the equilibrium concentration of phos- phorus in the solution mg/1, n = a constant). According to Russell and Prescott (13) k »represents the tenacity with which the soil keeps its phosphate or the reluc- tance with which the soil parts with its phosphate under the conditions of the experiment». The factors the effect of which on the value of k was studied in the present work were the degree of humification and the acidity of the peat, the samp- ling depth, the amount of extractable calcium, and the amounts of iron and alu- minum soluble in diluted hydrochloric acid. Material and methods The material of the present study consisted of 134 peat samples collected from different layers of virgin peat lands. The samples were air-dried and ground in a Wiley mill. The methods used for the determination of the degree of decomposition (H), the weight of volume, pH, and the content of extractable calcium are described in previous papers (4, 5). The »exchangeable» phosphorus was extracted by an alkali solution 0.1 N with respect to potassium hydroxide and potassium carbonate. The ratio of soil to solution was 1 : 100, and the shaking time was 2 + 4 hours in two consecutive days. The inorganic phosphorus in the solution was determined after the preci- pitation of the organic matter by sulphuric acid. The coefficient k in the Freundlich adsorption equation was calculated on the basis of data for removal of phosphorus by 5 grams of peat from 100 ml of KH 2P04 - solution containing phosphorus 15.5 mg/1 and 155.0 mg/1, respectively. The suspen- sions were heated on a boiling water-bath for two hours in two consecutive days. The values of »exchangeable» phosphorus were used in the calculations to represent the native sorbed phosphate. Iron and aluminum were extracted from 2 g-samples with 100 ml of 0.1 N HCI by shaking for one hour. The iron dissolved was determined by the method of Kumin (7). The aluminum content of the extract was estimated by the Aluminon method in which the disturbing effect of iron was eliminated by hydroxylamine hydrochloride. 217 Table 1. Analyses of peat samples. Volume Exchange- Extract- 0.1 NHCI-soluble Sample Depth H pH able P k able Ca dm weight ppm PP m I' e Ppm AI ppm Sphagnum peat 65 o—2 1 3.7 0.05 68 0 1800 110 120 K SI o—2 1 4.2 0.08 58 0 4100 170 170 K 32 4—6 1 4.4 0.08 28 0 4700 280 270 K 21 o—2 1 4.3 0.09 71 0 62(H) 50 180 K 34 o—2 1 4.2 0.09 72 112 3400 2840 3250 A 4 3—5 1 4.7 0.10 15 0 2800 110 180 K 37 o—2 1 4.5 0.11 96 0 4900 490 470 K 6 I—2 1 4.5 0.11 144 8 7000 1220 1700 A 27 o—2 1 4.5 0.12 31 209 2000 210 3600 36 o—2 1 4.0 0.12 70 36 2000 610 2300 A 37 o—2 1 4.4 0.28 53 40 3200 2000 1250 A 31 o—2 1 3.9 0.29 64 0 2100 410 1050 66 2—5 1 3.6 0.09 34 0 280 230 K 22 I—l 2 5.0 0.14 72 0 10600 140 100 K 7 2—3 2 4.6 0.17 154 126 6800 1440 2730 A 5 5—7 3 4.9 0.23 19 0 3500 170 440 A 1 2—3 3 3.7 0.29 35 0 3200 300 970 A 32 3—5 3 4.0 0.34 35 3 1500 140 1150 A 6 12—14 4 4.7 0.21 19 0 7100 140 440 105 2—4 4 4.4 0.20 27 0 5800 80 110 A 2 3—4 5 3.8 0.49 40 0 3600 440 850 A 3 7—lo 7 4.4 0.38 36 137 5600 2160 3130 Carex-Sphagnum peat 69 o—3 2 4.2 0.09 135 0 2300 610 930 K 38 2—4 2 4.6 0.23 95 0 4300 610 560 107 I—3 3 4.4 0.16 83 51 2000 560 1820 28 4—6 3 4.2 0.23 27 9 2100 2110 950 37 o—3 3 4.7 0.33 44 22 2800 220 1910 A 52 I—3 3 3.8 0.33 54 0 1530 860 550 K 39 4—6 4 4.6 0.25 53 0 3500 380 1100 A 28 5—7 4 4.3 0.30 19 123 1600 250 3270 70 3—5 4 4.4 0.25 103 22 1730 460 2230 34 o—3 4 4.5 0.34 60 399 1300 2720 3650 35 (I :S 4 4.5 0.38 75 318 2600 2670 2550 K 8 3—4 5 4.6 0.39 96 299 3900 1060 4500 K 33 6—B 5 4.5 0.25 62 0 4400 550 690 29 15—20 5 5.1 0.33 18 123 4400 5670 2450 A 29 B—lo 5 4.1 0.35 27 50 1800 260 2150 106 2—4 6 4.7 0.26 82 36 5600 210 1340 71 7—lo 6 4.3 0.36 54 139 1180 260 3350 K 42 2—4 7 3.9 0.39 71 66 3000 260 2140 A 46 4—6 7 4.1 0.49 52 78 2200 280 2100 218 Volume Exchange- Extract- 0.1 N HCI-soluble Sample Depth able P k able Ca dm H pH weight ppm ppm Fe ppm AI ppm Sphagnum-Carex peat K 28 o—2 1 4.5 0.20 115 26 4500 2080 1300 A 19 I—3 2 5.0 0.25 46 13 6000 830 740 A 12 3—5 2 4.8 0.25 44 16 4600 1500 1250 A 23 I—3 2 4.7 0.38 99 63 4400 2170 1250 59 o—2 3 3.6 0.23 98 106 1600 1220 1800 K 12 o—l 3 4.4 0.27 101 240 5100 1000 2650 B 14 I—3 3 4.3 0.30 31 5 4800 1890 900 A 53 I—3 3 4.2 0.42 31 43 1100 220 1800 A 47 2—4 3 4.6 0.23 62 519 2400 6120 3900 A 11 o—2 3 4.9 0.26 46 18 2400 830 820 A 16 6—B 3 4.5 0.30 28 57 1800 120 2200 A 35 o—4 3 5.5 0.31 40 0 13200 40 80 K 18 o—2 3 5.5 0.27 67 74 3300 1000 1600 B 6 I—3 4 4.3 0.21 74 303 2160 3780 2200 K 24 6—B 4 5.1 0.35 25 0 6600 750 330 A 49 I—3 4 4.3 0.34 95 30 2700 2840 1700 89 o—2 4 4.5 0.22 155 0 2280 1050 33 o—3 4 4.7 0.30 100 0 2700 1330 680 A 20 4—6 4 4.2 0.30 31 0 2900 390 560 B 2 I—3 4 4.1 0.33 26 145 1840 430 3100 A 15 2—4 4 4.3 0.34 52 19 2100 1000 1540 B 7 4—6 4 4.3 0.33 69 272 2360 4170 2650 B 8 B—l 2 5 4.4 0.30 71 329 2500 4950 3450 A 33 B—lo 5 4.1 0.30 25 20 1300 100 1910 60 3—9 5 3.5 0.32 53 7 1700 1000 850 K 35 4—6 5 4.9 0.34 99 170 3400 1250 2830 K 23 4—6 5 5.0 0.32 42 0 7900 420 210 B 3 4—6 6 4.0 0.36 22 250 790 680 4300 K 59 o—s 6 5.3 0.45 30 83 3000 780 1980 61 10—13 7 4.5 0.40 52 268 1900 3490 3550 76 60—62 9 4.8 0.71 60 71 4400 1120 2300 Carex-peat K 29 5—7 2 4.6 0.20 94 91 3700 2060 1500 B 10 I—3 3 4.3 0.20 26 68 1570 1780 1800 24 I—3 3 5.0 0.22 33 30 6470 260 1270 A 40 o—2 3 4.7 0.30 39 30 3550 2610 1000 A 41 3—5 3 4.8 0.25 18 13 2400 2330 900 A 8 o—3 3 4.7 0.32 36 126 3970 6340 2350 A 38 3—5 3 4.5 0.29 35 29 2100 950 1150 A 24 5—6 3 5.0 0.34 30 25 3700 2110 1300 A 43 2—5 3 4.5 0.24 43 66 3000 1260 3300 45 o—2 4 4.7 0.28 46 163 5000 1330 2600 38 o—2 4 4.9 0.36 152 38 6800 2840 1050 K 13 I—3 4 4.9 0.28 65 138 3900 610 2220 K 14 5—7 4 5.1 0.26 53 87 3400 780 2010 K 25 I—3 4 4.6 0.21 65 159 3400 940 2380 K 26 5—7 4 4.2 0.20 58 176 3500 1220 3180 219 Volume Exchange- Extract- 0.1 N CI- soluble Sample Depth able P k able Ca dm H pH weight ppm PP m Fe PP m AI ppm 103 2—4 4 4.8 0.28 130 276 3200 2670 1950 109 I—3 4 4.6 0.24 104 120 1300 2110 850 111 I—3 4 4.7 0.26 61 112 2400 1550 2550 K 27 11—14 4 4.4 0.23 59 367 2900 1670 5400 A 44 6—B 4 4.3 0.27 21 468 3300 1670 4950 110 2—4 5 4.6 0.24 45 142 2800 1660 2700 B 15 4—6 5 4.3 0.34 11 3 4600 1820 750 K 36 4—6 5 4.9 0.34 72 426 3700 3000 5400 A 17 B—lo 5 4.1 0.31 27 39 2i)0() 180 2500 K 30 2—5 5 4.8 0.38 70 343 3600 3670 3300 K 41 2—6 S 4.2 0.28 59 37 3900 1280 1150 23 2—4 6 5.0 0.34 141 484 9800 1330 4850 26 2—4 6 6.1 0.34 58 20 18900 280 230 K 20 6—B 6 5.4 0.30 52 83 3800 760 2550 104 6 4.6 0.29 85 135 2300 2860 1750 A 25 B—9 6 5.1 0.37 30 240 4100 2280 3400 B 4 B—l 2 7 4.3 0.33 26 309 850 830 5050 B 16 B—l 2 7 4.2 0.35 11 12 5000 2270 1100 B 12 B—l 2 7 4.4 0.37 15 63 2400 2210 1970 B 11 4—6 7 4.1 0.50 23 99 1570 2480 2000 A 21 B—lo 7 5.8 0.37 20 20 9000 110 1550 K 19 3—5 7 5.4 0.46 59 81 3400 840 1680 A 50 3—5 7 4.6 0.31 39 83 2100 1940 1900 31 3—6 8 4.9 0.39 28 105 13400 800 2400 30 o—3 8 4.6 0.54 44 98 13200 1030 2050 A 9 o—3 8 4.6 0.58 62 292 6340 2450 A 45 12—14 8 4.9 0.52 213 716 2900 3390 5250 32 o—s 8 4.7 0.69 40 623 7700 1310 6150 Bryales-Carex peat K 9 o—2 1 4.9 0.16 87 48 5200 1330 1800 K 1 o—2 1 5.5 0.14 32 0 10500 110 360 K 10 2—3 2 5.2 0.24 44 85 3700 560 1650 74 o—3 2 6.2 0.20 45 0 13300 110 70 39 o—2 2 4.7 0.23 34 0 4300 830 540 139 o—2 2 3.9 0.48 205 0 4400 1400 1450 K 2 3—5 3 5.2 0.28 26 58 8600 120 2420 97 o—2 3 3.9 0.26 70 0 23500 720 920 73 o—2 3 8.0 0.44 33 219 23500 330 1540 75 o—2 4 5.4 0.24 70 23 10200 2050 1100 125 4—7 4 4.9 0.30 24 19 7300 2330 800 134 4—lo 6 5.7 0.43 51 552 11500 720 4950 142 2—4 5 4.8 0.53 44 0 11700 330 120 140 2—4 6 5.1 0.58 56 138 14700 2560 1400 K 11 5—7 7 5.0 0.37 42 116 4000 610 2380 K 3 7—9 7 5.3 0.34 26 68 6800 450 2550 Eutrophic Sphagnum-Carex peat 91 o—2 2 5.8 0.18 108 0 0 60 40 o—3 3 5.6 0.25 54 0 14000 190 70 63 3—7 5 4.7 0.32 42 0 9550 330 480 220 Results The analytical results of the peat samples are reported in Table 1. There are 22 samples of Sphagnum peat (Sp), 19 samples of Carex-Sphagnum peat (CSp), 31 samples of Sphagnum-Car ex peat (SCp), 43 samples of Carex peat, 16 samples of Bryales-Carex peat (BCp), and 3 samples of eutrophic Sphagnum-Carex peat (EuSCp). On account of the low number of samples in some of the peat groups, the samples of Sp and CSp are treated as one group, and so are also the samples of Cp, BCp and EuSCp. In order to get a general survey of the material the mean values for the analyti- cal data of each group are calculated. They are recorded in Table 2 which also contains the corresponding standard deviations. Owing to one sample (number 76) taken from the depth of 60—62 dm, the average sampling depth is somewhat higher in the SCp-group than in the other ones. Without sample 76, the mean sampling depth in this group is 3.8 dm. Both the average degree of decomposition, H, and the weight of volume appear to be lower in the group of Sp and CSp than in the other groups. Also the members of the former group tend to be more acid and contain less samples with a high content of extractable calcium than the other ones do. Table 2. Mean values of the analytical results in different peat groups Peat group Sp and CSp SCp Cp and BCp Number of samples 41 31 62 mean s mean s mean s Sampling depth, dm 3.9 3.4 5.6 11.1 4.2 3.0 H 3.1 1.9 4.0 1.6 4.6 1.9 Weight of volume 0.24 0.12 0.32 0.09 0.32 0.11 pH 4.3 1.8 4.5 1.8 4.9 0.6 Extractable Ca, ppm 3550 2000 3470 2500 6200 5500 »Exchangeable» P, ppm 60 34 61 33 56 41 k 59 304 101 7,30 135 165 Acid-soluble Fe, ppm 880 1200 1620 1490 1520 2.520 » AI, ppm 1540 1210 1900 1160 2090 2220 The data for the »exchangeable» phosphorus are low in all the samples which, of course, is connected with the low content of inorganic phosphorus in Finnish uncultivated peat soils. In a previous work the author (4) found that the average content of inorganic phosphorus in the different peat groups varied from 120 to 180 ppm. The variation in the values of the coefficient, k seems to be very large. There are numerous samples, particularly in the Sp and BCp groups, which apparently do not retain phosphate under the conditions of the experiment. On the other hand. 221 there are samples which have values of k higher than 300. For the sake of com- parison it may be mentioned that for a clayed and cultivated peat soil which is known to have a high capacity to fix phosphate (according to the method by Piper (11) its phosphate exchange capacity is about 13000 ppm of P) the k value obtained by the present method was about 300. The highest k-values of the present material are 623 and 716 for the Cp-samples 32 and A 45, respectively. The average k-values for the different groups are, however, markedly lower: 59, 101, and 135 for the Sp and CSp, SCp, and Cp and BCp, respectively. Owing to the large variation, no significant difference exists between the peat groups. The amounts of iron and aluminum extracted by diluted hydrochloric acid also vary in the present material quite markedly. The lowest iron content is 40 ppm and the highest one 6340 ppm. The corresponding limits for the aluminum content are 70 ppm and 5250 ppm, respectively. The average contents of soluble iron and aluminum tend to be lower in the Sp and CSp group than in the other groups, although the differences are not significant. It is of interest to notice that a large part of the samples contains more acid soluble aluminum than acid soluble iron. In order to study the association between the phosphate retention and the different other factors in these samples, the total correlation coefficients between the values of k and the sampling depth, the degree of decomposition, the weight of volume, pH, the content of extractable calcium, and the contents of soluble iron and aluminum, respectively, were calculated. The following total correlation coefficients were obtained between k and depth H w./v. pH Ca Fe A1 Sp and CSp —O.Ol 0.24 0.33* 0.07 —0.19 o.49*** o.Bl*** SCp 0.15 0.16 —O.lO —0.17 —0.25 o.7s*** o.B4*** Cp and BCp 0.35** 0.39** 0.39** 0.01 0.06 0.30** o.BB*** All samples 0,108 0.324*** 0.319*** 0.018 —0.098 0.465*** 0.855*** The phosphate retention capacity of these peat samples, as characterized by the coefficient k, appears to be most closely connected with the content of acid- soluble aluminum. Also the association with the acid-soluble iron is quite marked, but the extractable calcium or the acidity apparently do not play any important role in the phosphate retention under the conditions of the experiment. There may be some connection with the degree of humification, represented by the H- values and the weights of volume, but the sampling depth is probably only of a minor importance. The association of the k values with the amounts of soluble aluminum and iron, and the degree of decomposition (H) was further studied by calculating the partial correlation coefficients in which the effect of each variable was isolated from the effects of the other variables. The elimination of the effect of the soluble iron or the degree of decomposition does not to any noteworthy degree change the correlation between k and soluble aluminum: the respective partial correlation coefficients are 222 rkA i;Fe = o.B3B*** and rkAI;H = 0.837*** The elimination of the effect of soluble aluminum from the correlation between k and soluble iron decreases the degree of reliability of the association, but the elimination of the effect of the degree of decomposition has less influence. The partial correlation coefficients are r kFe;Ai = 0.362*** and rkFe;H = 0.442*** The elimination of the effect of soluble aluminum reduces the correlation between k and the degree of decomposition to a nothingness whereas the elimination of the effect of soluble iron causes a far less decrease: r kH;Ai 0.036 and rkH;Fe 0.285** The partial correlation coefficients between k and soluble aluminum, or soluble iron, or the degree of decomposition are after the elimination of the effect of the two other variables are the following; rkA I;FeH = 0.823*** rkFe;AIH = 0.364*** rkH;AIFe = 0.051 Similar results are obtained, if instead of the degree of decomposition the weight of volume is used to indicate the degree of humification. The multiple correlation coefficient is rk(AIFeH) = 0.876*** which proves that the linear regression technique employed is fairly well suited for this material. The regression equation for estimating k (y) for any particular values of soluble aluminum (xx), soluble iron (x 2), and H (x 3) is thus y = o.oBs*** xx + 0.022*** x 2 1.68 x 3 75.2 and the standard deviation of any estimate will be 69.0 Owing to the fact that the elimination of the effect of the soluble aluminum content results in the disappearance of the association between k and the degree of decomposition, k may be fairly reliably estimated on the basis of the contents of soluble aluminum and iron, only. This regression equation is y = o.oB3*** Xj + o.o2l*** x 2 79.6 and the standard deviation of any estimate will be 68.7. The multiple correlation coefficient is rk(AlFe) = 0.875***. Discussion As far as the coefficient k of the present work actually represents the sorption of phosphorus by the peat samples, the results obtained contain some interesting facts. Under the conditions of the experiment the retention of phosphorus varied quite markedly, even within the same kind of peat, and no significant differences 223 could be detected between the peat groups, although the Sphagnum peats tended to retain phosphorus to a lower degree than did the fen peats. This sorption was most closely connected with acid-soluble aluminum whereas acid-soluble iron appeared to play a minor role. The present literature shows an essentially universal agreement that phosphate fixation in acid soils is primarily due to iron and aluminum, but there is very little information as to their mutual importance. Recently Williams et al. (16) found that in mineral soils aluminum extracted by different acid solutions gave highly significant correlations with the phosphate sorption, and in no case did the addition of a term for iron significantly improve the estimate of sorption given by aluminum alone. Although in these peat samples the degree of association of acid-soluble iron with the phosphate sorption was markedly lower than that of acid-soluble aluminum, also the effect of iron had to be counted. There is, of course, no reason to suppose that just the amounts soluble in diluted hydrochloric acid would be equal to the active fractions of aluminum and iron in peat soils. Williams et al. (16) maintain that aluminum extracted by the Tamm acid-oxalate method is the best single criterion of phosphate sorption in the mineral soils studied. Attention must also be paid to the fact that the peat samples were air-dried and ground which may exert some effect on the retention of phosphate and on the solubility of aluminum and iron. The results of the present study do not explain the mechanism of the phosphate retention by the peat samples. The Freundlich adsorption isotherm was used, but it does not imply an adsorption process. According to Fisher (3) the equation also applies to many reactions that are not adsorption. It must be emphasized that the values of k by no means are equal to the phosphate sorption capacity, they only are supposed to be correlated with this quantity the determination of which is difficult, or practically impossible if absolute figures are demanded. Summary The factors on which the retention of phosphorus by peat depends were studied on the basis of a material of 134 virgin peat samples. The coefficient k in the Freund- lich adsorption isotherm y = kc n was used as an indicator of the phosphate sorp- tion. The association of k with the sampling depth, the degree of decomposition, weight of volume, pH, extractable calcium, and the iron and aluminum dissolved by 0.1 N hydrochloric acid was treated. The acid-soluble aluminum gave with k a highly significant correlation which did not decrease when the effects of acid-soluble iron and the degree of humifi- cation were eliminated. The correlation between k and the acid-soluble iron was also highly significant although less close than the former association, and it was to some extent lowered by the elimination of the effect of aluminum. The fairly low, although statistically significant correlation between k and the degree of humification as expressed by the degree of decomposition or by the volume 224 weight disappeared when the effect ofaluminum was isolated. In the present material the sampling depth, pH and the content of extractable calcium did not give any significant correlation with k. As far as these results are valid under natural conditions, aluminum appears to play a more important role in the phosphorus sorption of peat soils than iron does. REFERENCES (1) Doughty, J. L. 1930. The fixation of phosphate by a peat soil. Soil Sci. 29: 23—35. (2) —»— 1935. Phosphate fixation in soils, particularly as influenced by organic matter. Ibid 40: 191—202. (3) Fisher, E. A. 1922. The phenomena of absorption in soils: a critical discussion of hypotheses put forward. Trans. Faraday Soc. 17: 305—316. (4) Kaila, A. 1956. Phosphorus in virgin peat soils. J. Sci. Agr. Soc. Finland 28: 142—167. (5) Kaila, A. & Kivekäs, J. 1957. Extractable calcium, magnesium, potassium and sodium in dif- ferent peat types. Ibid. 29: 41—55. (6) Kasakow, E. 1934. Adsorption der Phosphate durch Moorböden. Pedology 29: 439 (Ref. Z. Pflanzenern., Diing. Bodenk. 42: 105, 1936). (7) Kunin, R. 1943. Microdetermination of iron by the mercurous nitrate method. Soil Sci. 55: 457. (8) Larsen, J. E. 1957. Phosphorus availability in organic soils. Dissert. Abstr. 17: 1880—1881 (Ref. J. Food Agric. 9: i i—92). (9) McCool, M. M. 1921. Peat and muck soils. Fixation of fertilizers. Mich. Quart. Bui. 3: 127. (10) Olsen, S. R, & Watanabe, F. S. 1957. A method to determinea phosphorusadsorption maximum of soils as measured by the Langmuir isotherm. Soil Sei. Soc. Amer. Proc. 21: 144—149 (11) Piper, C. S. 1944. Soil and plant analysis. New York, 368 p. (12) Rennie, D. A. & McKercher, R. B. 1959. Adsorption of phosphorus by four Saskatchewan soils. Canad. J. Soil Sci. 39: 64—75. (13) Russell, E. J. & Prescott, J. A. 1916. The reaction between dilute acids and the phosphorus compounds of the soil. J. Agr. Sci. 8: 65—110. (14) Verhoeven, B. 1946. Fosfaatvastlegging aam indrogende veengronden. Landbowk. Tidschr. 58: 237—242. (15) Wild, A. 1950. The retention of phosphate by soil. A review. J. Soil Sci. 1: 227—238. (16) Williams, E. G. & Scott, N. M. & McDonald, M. J. 1958. Soil properties and phosphate sorp- tion. J. Sci. Food Agric. 9: 551—559. SELOSTUS: TURVENÄYTTEIDEN FOSFORIN PIDÄTYKSESTÄ Armi Kaila Yliopiston maanviljelyskemian laitos, Helsinki Tutkimuksessa on yritetty selvittää turpeiden fosforin pidätyskykyyn vaikuttavia tekijöitä. Aineistona oli 134 luonnontilaisilta soilta kerättyä näytettä, jotka edustivat eri turvelajeja. Fosforin pidätyskyvyn indikaattorina käytettiin Freundlichin adsorptioyhtälön, y = ke11 , kerrointa k. Tutkittiin k:n riippuvuutta maatumisasteesta, näytteenottosyvyydestä, tilavuuspainosta. pH:sta, uuttuvasta kalsiumista sekä 0.1 n suolahappoon liukenevasta raudasta ja aluminiumista. 225 Todettiin, että aluminiumin ja k:n välillä vallitsi erittäin voimakas korrelaatio, joka ei heikenty- nyt, kun raudan tai maatumisasteen vaikutus eliminoitiin. Myös k:n ja raudan välinen korrelaatio oli verraten voimakas, joskin huomattavasti matalampi kuin edellinen, Aluminiumin vaikutuksen elimi- nointi heikensi jonkin verran k:n ja raudan välistä riippuvuutta ja hävitti kokonaan heikohkon korre- laation k:n ja maatumisasteen väliltä. Näytteenottosyvyys, pH ja kalsiumin pitoisuus eivät tämän aineiston perusteella vaikuta turpeen fosforin pidätyskykyyn. Mikäli näitä ilmakuivia näytteitä käyttäen saatuja tuloksia voidaan soveltaa luonnon olosuhtei- siin, näyttää aluminium yhdisteineen olevan myös turvemaissa tärkeämpi fosforin pidättäjäkuin rauta.