Maataloustieteellinen A ikakauskirja Vol. 61: 89—97, 1989 Extractable aluminium, iron and manganese in mineral soils 111 Comparison of extraction methods RAINA NISKANEN University of Helsinki, Department of Agricultural Chemistry, SF-00710 Helsinki, Finland Abstract. The extractability of soil Al, Fe and Mn were studied in 102 mineral soil samples. The extractants were 0.05 M oxalate (pH 2.9), 0.05 M K 4P 20 7 (pH 10), 0.02 M EDTA (pH 5.3) and 1 M CH3 COONH4 (pH 4.8). In the group of clay and silt soils (n = 51), the Al extracted by the four extractants correlated closely; the r values ranged from o.9l*** to o.96***; in coarser soils (n = 51) the r values ranged from 0.42* to o.B2***. In clay and silt soils, the organic carbon content and soil pH together explained 50 % of the variation in oxalate- extractable Al, 70 % of the variation in pyrophosphate-extractable Al, 53 % of the variation in pyrophosphate-extractableFe and 56 % of the variation in acetate-extractable Al. The clay and organic carbon contents together with soil pH explained 77 % of the variation in EDTA- extractable Al in clay and silt soils. In coarse soils, the extractable metals were not closely related to the soil characteristics. Index words: acetate-extractable Al, EDTA-extractable metals, oxalate-extractable metals, pyrophosphate-extractable metals, organic carbon content, clay content, soil pH Inlroductiun Non-crystalline inorganic components of soil are often termed “active” to indicate their importance in soil processes (Mitchell et al. 1964). The principal forms of these compo- nents are oxides and hydrous oxides of Al, Fe, Mn, Si and Ti which may be adsorbed on mineral surfaces, or they exist as discrete par- ticles (Huang and Lee 1969, Huang and Kozak 1970, Wada and Harward 1974). Ex- traction methods have been used to differen- tiate vark s forms of soil oxides. Ammoni- um oxalate (Tamm 1922,Schwertmann 1964, McKeague and Day 1966) and pyrophosphate (Aleksandrova 1960, McKeague 1967), for instance, have been used to extract iron and aluminium in oxides and complexed by or- ganic matter. The purpose of this paper was to compare the release of soil aluminium, iron and manganese by different extraction methods and to study the relationship of metals to soil particle-size distribution, organic carbon content and pH. 89 JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND https://www.c-info.fi/en/info/?token=GhQhgGjWFZkjPwdi.S3AM28Tc5B1_ZEd5CzUd0w.MOp_cHdqL19XZXhjxkxwhfemJ3IUAw6gTvaPmjMT2u3Rf1Ow0kdLXQlLtGt1cLE9JkNw-zbHdI3oveAX0NSZf4HkvYolRSgYeYSEN0U9cCfgkYIlf6LeghiGMINxopZ9H-Km3-xRXlAQB_2sKeEegqXprKrIJbFYH2D14w Material and methods The material of 102 mineral soil samples was collected in 15 localities mainly in south- ern Finland (Table 1). The 53 sampling sites represented principally agricultural soils (9 of them were virgin). The surface layer was sampled from 0 to the depth of 20—30 cm; at 48 sites the deeper layer was sampled to the depth of 40—60 cm. The material included 36 clay soil samples with a clay fraction content (0 < 2 urn) of 30—72 %, and 15 were silt soil samples. Because of the high clay content in the silt soils, they were examined in the same group with clay soils. The samples were air-dried at room tem- perature and ground to pass through a 2-mm sieve. The pH of the soil was measured in a soil-0.01 M CaCl2 suspension (1:2.5 v/v) (Ry- ti 1965). The particle-size distribution of in- organic matter was determined by the pipette method (Elonen 1971), the organic carbon content by the Alten wet combustion meth- od (Graham 1948). Soil aluminium, iron and manganese were extracted by the methods presented in Table 2. The extractions were carried out in dupli- cate or triplicate. Aluminium, iron and man- ganese in filtrates were determined by atomic absorption spectrophotometry, Fe and Mn with air-acetylene flame and Al with N2O- -flame. Results Of the extractants studied, 0.05 M oxalate (Table 3) released metals most effectively. In the group of clay and silt soils, the oxalate- extractable aluminium correlated closely with the aluminium extracted by 0.05 M K4P,0 7 , 0.02 M EDTA and 1 M ammonium acetate (Tables 4—7). In coarse soils, the oxalate- extractable aluminium did not correlate very closely with the aluminium extracted by the other methods (Table 7). The oxalate-extract- able iron and manganese (Table 3) did not correlate closely with the metals extracted by pyrophosphate (Table 4) and EDTA (Table 5). Table I. Soil characteristics. Particle-size distribution (urn), % pH(CaCI,) Org.C, "7o < 2 2—20 > 20 n x s range x s range x s range x s range x s range Clay and silt soils: Surface soils 27 5.4 0.9 3.9—7.2 4.6 3.0 1.1 14.6 37 14 15—70 33 12 8—55 30 15 9—61 Subsoils 24 5.4 0.6 4.1—6.9 3.2 2.3 0.8—11.5 43 13 19—72 32 15 6—61 25 13 7—55 All 51 5.4 0.8 3.9—7.2 3.7 2.8 0.8—14.6 40 14 15—72 33 13 6—61 28 14 7—61 Coarse soils: Surface soils 26 5.2 1.0 3.5—7.3 4.0 2.5 1.5—12.3 9 7 2—28 12 6 2—23 80 11 56—95 Subsoils 25 5.3 0.9 4.1—7.2 1.6 0.9 0.6—3.7 7 8 I—2B 11 8 1—29 82 13 51—98 All 51 5.2 1.0 3.5—7.3 2.8 2.2 0.6—12.3 8 8 I—2B 12 7 1—29 81 12 51—98 All soils: Surface soils 53 5.3 1.0 3.5—7.3 4.3 2.7 1.1 14.6 23 18 2—70 23 14 2—55 55 28 9—95 Subsoils 49 5.3 0.8 4.1—7.2 2.1 1.8 0.6—11.5 25 21 1—72 22 16 I—6l 54 31 7—98 90 Table 2. Extraction methods. Extractant pH Extraction Shaking Reference ratio, w/v time, h 0.05 M oxalate 2.9 1:20 2 Niskani n 1989 a (0.026 M ammonium oxalate, 0.024 M oxalic acid) 0.05 M K 4P,O, 10 1:100 3 Niskanen 1989 a 0.02 M Na,-EDTA 5.3 1:50 3 1 M ammonium acetate 4.8 1:10 2 McLiANetal. 1958 Table 3. 0.05 M oxalate-extractable Al, Fe and Mn. AI, mmol/kg soil Fe, mmol/kg soil Mn, umol/kg soil n x s range x s range x s range Clay and silt soils: Surface soils 27 78 41 32—210 96 38 43—191 2 610 1 730 230—6 240 Subsoils 24 64 39 29—222 108 46 32—202 2 020 I 520 160—4 870 All 51 71 41 29—222 102 42 32—202 2 330 1650 160—6 240 Coarse soils: Surface soils 26 79 49 16—186 60 29 7—144 1460 1070 120—4 640 Subsoils 25 82 63 11—249 44 32 3—135 730 900 20—3 990 All 51 81 55 11—249 53 31 3—144 1 100 1050 20—4 640 All soils: Surface soils 53 78 45 16—210 79 38 7—191 2 050 1540 120—6 240 Subsoils 49 73 52 11—249 76 50 3—202 1360 1 400 20—4 870 Table 4. 0.05 M K 4 P 207 -extractable Al, Fe and Mn. AI, mmol/kg soil Fe, mmol/kg soil Mn, umol/kg soil n x s range x s range x s range Clay and silt soils: Surface soils 27 46 51 6—243 33 22 7—lll 790 480 240—1910 Subsoils 24 29 47 4—233 24 21 5—76 430 290 120—1 050 All 51 38 50 4—243 29 22 s—lll 620 440 120—1910 ( oarsc ■-oils: Surface soils 26 46 28 12—104 24 15 7—77 800 500 120—1940 Subsoils 25 39 28 4—93 14 15 4—53 300 480 o—l 800 All 51 43 28 4—104 19 15 4—77 550 550 o—l 940 All soils: Surface soils 53 46 41 6—243 28 19 7—lll 800 490 120—1940 Subsoils 49 34 39 4—243 19 19 4—76 360 400 o—l 800 The pyrophosphate-extractable aluminium (Table 4) correlated closely (Table 7) with the EDTA-extractable (Table 5) and acetate-ex- tractable aluminium (Table 6). The correlation between the EDTA-extractable and acetate-ex- tractable aluminium was also close (Table 7). The pyrophosphate-extractable iron (Table 4) correlated with the EDTA-extractable iron (Table 5). In clay and silt soils, the correla- tion between the pyrophosphate-extractable and EDTA-extractable manganese was weaker than in coarse soils (Table 7). In clay and silt soils, the organic carbon content and soil pH together explained (P = 0.001) 50 % of the variation in the oxalate- extractable aluminium, the regression equa- 91 Table 5. 0.02 M Na 2-EDTA-extractable Al, Fe and Mn. Al, mmol/kg soil Fe, mmol/kg soil n x s range x s range Clay and silt soils: Surface soils Subsoils All Coarse soils: Surface soils Subsoils All All soils: Surface soils Subsoils 27 21 18 7—Bl 24 17 18 4—85 51 19 18 4—85 26 17 9 5—34 25 14 13 2—61 51 16 11 2—61 53 19 14 s—Bl 49 15 15 2—85 9 7 3—32 4 5 1—23 6 7 1—32 Table 6. 1 M ammonium acetate-extractable Al. Al, mmol/kg soil n x s range Clay and silt soils: Surface soils Subsoils All Coarse soils: Surface soils Subsoils 27 8.8 11.0 0.2—48.2 24 8.0 10.1 1.4—48.9 51 8.4 10.5 0.2—48.9 26 7.8 5.5 1.2—22,5 25 9.4 7.6 0.9—34.8 51 8.6 6.6 0.9—34.8All All soils: Surface soils Subsoils 53 8.3 8.7 0.2—48.2 49 8.7 8.8 0.9—48.9 tionbeing: Al(oxal.) (mmol/kg) = 118.75 + 10.350rg.C-% —15.96pH. The partial corre- lationcoefficients for the relation between the r 14.5 o.67*** —0.36*r i5.4 Table 7. Linear correlation coefficients between extractable metals. Clay and silt soils (n = 51) AI(oxal.) o.96*** o,94*** o.9l*** o.9s*** o.9s***A1(K4 P 20 7 ) AI(EDTA) o.92*** Fe(K 4P 2O 7) Fe(EDTA) Fe(oxal.) 0.43** o.63*** o.72***Fe(K 4 P 2O 7 ) Mn(K 4 P 2O7 ) Mn(EDTA) Mn(oxal.) o.s3*** o.69*** Mn(K 4P,O 7 ) 0.32* Mn, |xmol/kg soil x s range 830 530 110—1 590 230 250 20—1 110 540 510 20—1 590 13 10 3—46 1040 720 110—3 720 10 10 1—42 630 780 20—3 380 oxalate-extractable AI (1), the organic carbon content (4) and soil pH (5) were: Soil properties did not significantly explain the variation in the oxalate-extractable alumini- um in coarse soils and the oxalate-extractable iron and manganese in clay and silt soils. In coarse soils, the clay content weakly corre- lated with iron (r = 0.40*) and manganese (r = o.s4***). The organic carbon content and soil pH together explained (P = 0.001) 70 % of the variation in the pyrophosphate-extractable aluminium in clay and silt soils, the regression equation being; A1(K4P,0 7) (mmol/kg) = 79.56 + 13.590rg.C-% —17.06pH. The par- Coarse soils (n = 51) A1(K4P,0 7) AI(EDTA) Al(acet.) AI(K 4 P 2O v) AI(EDTA) Al(acet.) o.76*** 0.42* 0.68»** o.74*** o.Bo*** Fe(K 4P 2Q 7) Fe(EDTA) o.B2*** o.6B*** o.ss*** o.72*** Mn(K 4 P2 O 7 ) Mn(EDTA) o.7o*** o.7s*** o.l2*** 18 II 4—46 1230 820 210—3 720 16 II 2—42 1040 920 130—3 380 17 II 2—46 1140 870 130—3 720 92 tial correlation coefficients for the relationship between the pyrophosphate-extractable Al (1), the organic carbon content (4) and soil pH (5) were: r 14.5 o.B2*** —o.44**r ,5.4 Soil properties did not significantly explain the variation in the pyrophosphate-extractable aluminium in coarse soils. The organic carbon content and soil pH together explained (P = 0.001) 53 °7o of the variation in the pyrophosphate-extractable iron in clay and silt soils, the regression equa- tion being as follows: Fe(K 4P 2 O 7 ) (mmol/kg) = 77.87 + 4.220rg.C-% —12.07pH. The partial correlation coefficients for the relation- ship between the pyrophosphate-extractable iron (2), the organic carbon content (4) and soil pH (5) were: 24.5 o.62*** r25. —o.s4*** In coarse soils, the organic carbon content correlated weakly with the pyrophosphate- extractable iron (r = 0.45**). In clay and silt soils, the pyrophosphate-extractable man- ganese correlated weakly with the soil silt con- tent (r = o.47***). The pyrophosphate-extractable metals are expressed as the percentages of the oxalate- extractable metals in Table 8. In clay and silt soils, the organic carbon content correlated closely with the percentages of aluminium(r = o.Bo***), iron (r = o.Bo***) and manganese (r = o.77***). When the whole coarse soil group was examined, the correlation for man- ganese was weak (r = o.s3***), and that for aluminium and iron was insignificant. In coarse surface soils (n = 26), the organic car- bon content correlated with the percentages of aluminium (r = 0.59*), iron (r = o.73***) and manganese (r = o.77***). The clay content, the organic carbon content and soil pH explained (P = 0.001) 77 % of the variation in the EDTA-extractable aluminium in clay and silt soils, the regression equation being: AI(EDTA) (mmol/kg) = 28.87 + 0.26c1ay-% + 4.490rg.C-<% —6.85pH. The partial correlation coefficients for the relation between the EDTA-extractable Al (1), the clay content (3), the organic carbon content (4) and soil pH (5) were: r 13.45 r I4,35 r !5.34 0.39** 0 83*** ■o.s3*** In coarse soils, only soil pH correlated signifi- cantly with theEDTA-extractable aluminium (r = —o.49***). The organic carbon content correlated weakly with the EDTA-extractable iron in clay and silt soils (r = 0.46**). In coarse soils, the clay and organic carbon content together ex- plained (P = 0.001) 40 % of the variation in the EDTA-extractable iron. The partial corre- Table 8. 0.05 M K 4P 207 -extractable Al, Fe and Mn as the percentage of oxalate-extractable. AI, % Fe, % Mn, % n x s range x s range x s range Clay and silt soils: Surface soils 27 48 25 13—116 35 19 12—74 50 39 6—157 SubsoiK 24 34 24 9—105 22 18 5—86 36 37 6—169 AH 51 42 25 9—116 29 19 5—86 43 38 6—169 Coarse soils: Surface soils 26 62 22 28—105 44 22 17—97 73 36 12—194 Subsoils 25 53 28 17—110 42 32 6—113 51 48 o—lBo All 51 58 25 17—110 43 27 6—113 62 43 0—194 All soils: Surface soils 53 55 25 13—116 39 21 12—97 61 39 6—194 Subsoils 49 44 28 9—llo 32 28 5—113 44 43 o—lBo 93 Table 9. 0.02 M Na 2EDTA-extractable Al, Fe and Mn as the percentage of oxalate-extractable. AI, % Fe, % Mn, % tl x s range x s range x s range Clay and silt soils: Surface soils 27 25 7 11—39 19 12 5—47 61 35 12—189 Subsoils 24 23 8 13—45 14 8 2—36 58 30 11 138 All 51 24 8 11—45 17 11 2—47 60 32 11 189 Coarse soils: Surface soils 26 27 14 1—59 17 16 4—71 72 41 16—222 Subsoils 25 22 16 2—72 16 19 1—59 52 43 6—190 All 51 24 15 1—72 17 17 I—7l 62 43 6—222 All soils: Surface soils 53 26 11 1—59 18 14 4—71 66 38 12—222 Subsoils 49 23 13 2—72 15 14 1—59 55 37 6 190 lation coefficients for the relation between the EDTA-extractable Fe (2), the clay content (3) and the organic carbon content (4) were; r23.4 r24.3 0.37** o.6o*** In clay and silt soils, the EDTA-extractable manganese correlated weakly with soil pH (r = 0.39*) and in coarse soils with the clay content (r = o.4s***). The EDTA-extractable metals are expressed as percentages of the oxalate-extractable metals in Table 9. In clay and silt soils, the organic carbon content correlated with the percentages of aluminium (r = o.6l***) and iron (r = o.64***). In coarse surface soils (n = 26), the organic carbon content cor- related weakly with the percentages of aluminium (r = 0.42*) and iron (r = 0.43*). Table 10. 1 M ammonium acetate-extractable Al as the percentage of oxalate-extractable Al. Al, % Clay and silt soils; Surface soils Subsoils All n x s range Coarse soils; Surface soils Subsoils 27 9 6 0.4—23 24 10 6 3 —22 51 10 6 0.4—23 All All soils: Surface soils Subsoils 26 11 6 4 —27 25 13 8 4 —23 51 12 7 4 —27 The organic carbon content and soil pH ex- plained (P = 0.001) 56 % of the variation in the ammoniumacetate-extractable aluminium in clay and silt soils, the regression equation being: Al(acet.) (mmol/kg) = 26.40 + 2.310rg.C-% —4.95pH. The partial correla- tion coefficients for the relation between the acetate-extractable Al (1), the organic carbon content (4) and soil pH (5) were: r 14.5 r 15.4 o.6B*** In coarse soils, the acetate-extractable aluminium correlated weakly with soil pH (r = 0.44*). q 49*** In clay and silt soils, the acetate-extractable aluminium, expressed as the percentage of the oxalate-extractable aluminium (Table 10), cor- related weakly with the organic carbon con- tent in soil (r = 0.43**). Discussion The four extractants to some degree dis- solved metals complexed with organic matter in soil. An expression of this was that the organic carbon content of soil together with soil pH explained the variation in the contents of extractable metals. The conventional oxalate extractant (Tamm 1922, Schwertmann 1964, McKeague and Day 1966) dissolves iron and aluminium from poorly crystallized oxides as well as from organic matter complexes (Schnitzer and 53 10 6 0.4—27 49 II 7 3 —23 94 Skinner 1964, Schwertmann 1964, 1973, McKeague and Day 1966). The iron oxide extracted in darkness is mainly ferrihydrite (Schwertmann 1959); in light and by prolonged extraction, all iron compounds are obviously solubilized (McKeague et al. 1971, Pawluk 1972, Schwertmann 1973). In the study of Huang et al. (1977), the oxalate-extractable aluminium and iron were related to the content of organic matter in soil, but not to acidity or clay content. In the present material, the content of organic car- bon in soil together with soil pH explained only the variation in the oxalate-extractable aluminium in clay and silt soils, whereas in coarse soils, the oxalate-extractable iron and manganese were slightly related to the clay content in soil. The pH of the pyrophosphate solution used as the extractant was about 10. Alkaline pyrophosphate is reasonably specific for the organic iron complexes and somewhat less specific for aluminium complexed with or- ganic matter (McKeaoue 1967, McKeague et al. 1971). Aluminium may partly be derived from oxides. Little iron is extracted from poorly or well crystallized iron oxides by alka- line pyrophosphate (Bascomb 1968, Bascomb and Thaningasalam 1978, McKeague et al. 1971, Arshad et al. 1972), whereas iron is solubilized from oxides as well as from silicate minerals by neutral pyrophosphate (Titova 1962, Kononova et al. 1964). The manganese extracted by alkaline pyrophosphate is mainly divalent and complexed by organic matter (Heintze and Mann 1949). Hydrated man- ganese oxides are fully soluble in neutral pyrophosphate but are poorly soluble in alka- line extractant (Heintze and Mann 1949). Pyrophosphate extracts can also contain manganese derived from oxides of higher oxidation states reduced by organic matter (Heintze 1957). In the study of Shoji and Fujiwara (1984), soil total carbon correlated with the sum of the pyrophosphate-extractable aluminium and iron. In the present material, both the alumi- nium and iron extracted by pyrophosphate correlated with the soil organic carbon, espe- cially in clay and silt soils. If the concentra- tion of metals was expressed as percentages of the oxalate-extractable metal, also man- ganese correlated with the organic carbon con- tent. EDTA is a widely used extractant for or- ganically-bound metal ions. At pHs from 4 to 7, crystalline iron oxides are not soluble in EDTA (Aguilera and Jackson 1953, Rubio and Matijevic 1979). Very long extraction (90 days) by EDTA at pH 4.4—6.0, proposed for the extraction of ferrihydrite (Borggaard 1976, 1979), does not dissolve more iron than does acid oxalate (pH 3.0) during 4—5 hours (Borggaard 1976). Instead, 0.05 M EDTA at pH 9 is an effective extractant of aluminium and iron (Viro 1955). In Finland, micro- nutrients are extracted by acid ammonium acetate-EDTA (0.5 M CH 3 COONH4, 0.5 M CHjCOOH, 0.02 M Na2-EDTA, pH 4.65) (Lakanen and Erviö 1971), the EDTA con- centration of which was used in the present study. However, the pH was adjusted to 5.3 because the solubilitiesof aluminium and iron seemed to be relatively constant at pH 5.3— 6.8 in a previous paper (Niskanen 1989b). In clay and silt soils, the clay content, together with organic carbon content and soil pH, explained the variation in the content of EDTA-extractable aluminium. This result may mean that EDTA extracts aluminium hydroxide polymers from clay surfaces. Aluminium hydroxide polymers have a tendency to form films over larger surface areas than iron hydroxide, which appears to precipitate at specific hydroxyl sites on the clay surface (El Swaify and Emerson 1975, Rengasamy and Oades 1977). The acetate-extractable aluminium, which consisted of exchangeable aluminium ions, hydroxy aluminium polymers as well as or- ganic aluminium complexes (Jackson 1963, Mclean et al. 1964), aluminium phosphates (Yuan and Fiskell 1959) and newly preci- pitated aluminium hydroxide (Pratt and Bair 1961), correlated closely with the EDTA-extractable aluminium. The correla- 95 tion between the acetate-extractable alumi- nium in clay and silt soils and the soil organic carbon content is in agreement with the opin- ion of Pionke and Corey (1967), who con- tended that acetate-extractable aluminium is largely complexed by organic matter. In coarse soils, the acetate-extractable aluminium was not related to the organic carbon content. In coarse soils, the view of Hargrove and References Aguilera, N.H. & Jackson, M.L. 1953. Iron oxide removal from soils and clays. Soil Sei. Soc. Amer. Proc. 17: 359—364. Aleksandrova, L.N. 1960. The use of sodium pyrophosphate for isolating free humic substances and their organic-mineral compounds from the soil. Soviet Soil Sci. 2: 190—197. Arshad, M.A., Arnaud, R.J.S. & Huang, P.M. 1972. Dissolution of trioctahedral layer silicates by ammo- nium oxalate, sodium dithionite-citrate-bicarbonate, and potassium pyrophosphate. Can. J. Soil Sci. 52: 19—26. Bascomb, C.L. 1968. Distribution of pyrophosphate-ex- tractable iron and organic carbon in soils of various groups. J. Soil Sci. 19: 251—268. & Thaningasalam, K. 1978. Comparison of aqueous acetylacetone and potassium pyrophosphate solutions for selective extraction of organic-bound Fe from soils. J. Soil Sci. 29: 382—387. Borggaard, O.K. 1976. Selective extraction of amor- phous iron oxide by EDTA from a mixture of amor- phous iron oxide, goethite, and hematite. J. Soil Sci. 27: 478—486. 1979. Selective extraction of amorphous iron oxide by EDTA from a Danish sandy loam. J. Soil Sci. 30: 727—734. Elonen, P. 1971. Particle-size analysis of soil. Acta Agr. Fenn. 122: 1 122. El Swaify, S.A. & Emerson, W.W. 1975. Changes in physical properties of soil clays due to precipitated aluminium and iron hydroxides: I. Swelling and aggregate stability after drying. Soil Sei. Soc. Amer. Proc. 39: 1056—1063. Graham, E.R. 1948. Determination of soil organic matter by means of a photoelectric colorimeter. Soil Sci. 65: 181—183. Hargrove, W.L. & Thomas, G.W. 1984. Extraction of aluminium from aluminium-organic matter in rela- tion to titratable acidity. Soil Sci. Soc. Am. J. 48: 1458—1460. Heintze, S.G. 1957. Studies on soil manganese. J. Soil Sci. 8: 287—300. Thomas (1984), who proposed that acetate poorly identifies reactive aluminium asso- ciated with organic matter, may be valid. In any case, acetate extracts a rather small frac- tion of the soil aluminium. It seems to be more advantageous to use pyrophosphate to extract metals complexed by organic matter. Another disadvantage of acetate is that it is a poor iron extractant (Paasikallio and Häkkinen 1977). & Mann, P.J.G. 1949. Studies on soil manganese. J. Agric. Sci. 39: 80—95. Huang, P.M. & Kozak, L.M. 1970. Adsorption of hydroxy-aluminum polymers by muscovite and bio- tite. Nature 228; 1084—1085. & Lee, S.Y. 1969. Effect of drainage on weathering transformations of mineral colloids of some Canadian Prairie soils. Proc. Int. Clay Conf. 1: 541 —551. —, Wang, M.K., Wu, M.H., Liu, C.L. & Wang, T.S.C. 1977. Sesquioxidic components of selected Taiwan soils. Geoderma 18: 251—263. Jackson, M.L. 1963. Aluminum bonding in soils: A unifying principle in soil science. Soil Sei. Soc. Amer. Proc. 27: I—lo. Kononova, M.M., Aleksandrova, TV. & Titova, N.A. 1964. Decomposition of silicates by organic substances in the soil. Soviet Soil Sci. 1005—1014. Takanen, E. & Ervio, R. 1971. A comparison of eight ex- tractants for the determination of plant available micronutrients in soils. Acta Agr, Fenn. 123: 223—232. McKeague, J.A. 1967. An evaluation of 0.1 M pyrophosphate and pyrophosphate- dithionite in com- parison with oxalate as extractants of the accumula- tion products in Podzols and some other soils. Can. J. Soil Sci. 47: 95—99. & Day, J.H. 1966. Dithionite- and oxalate-extracta- ble Fe and A 1 as aids in differentiating various classes of soils. Can. J. Soil Sci. 46: 13—22. —, Brydon, J.E. & Miles, N.M. 1971. Differentiation of forms of extractable iron and aluminum in soils. Soil Sei. Soc. Amer. Proc, 35: 33—38. McLean, E. 0., Heddleson, M.R., Bartlett, R.J. & Holowayshuk, N. 1958. Aluminum in soils: I. Extrac- tion methods and magnitudes in clays and Ohio soils. Soil Sci. Soc. Proc. 22; 382—387. —, Hourigan, W.R., Shoemaker, H.E. & Bhumbla, D.R. 1964. Aluminum in soils: V. Form of aluminum as a cause of soil acidity and a complication in its mea- surement. Soil Sci. 97: 119—126. Mitchell, 8.D., Farmer, V.C. & McHardy, W.J. 1964. Amorphous inorganic materials in soils. Adv. Agron. 96 16: 327—383. Niskanen, R. 1989a. Extractable aluminium, iron and manganese in mineral soils. II Extractability by oxalate and pyrophosphate. J. Agric. Sci. Finl. 61: 79—87. -1989b. Extractable aluminium, iron and manganese in mineral soils. I Dependence of extractabilityon the pH of oxalate, pyrophosphate and EDTA extractants. J. Agric. Sci. Finl. 61: 73—78. Paasikallio, A. & Kakkinen, U. 1977. Acid ammonium acetate and acid ammonium acetate/EDTA as extrac- tants for phosphorus-32, aluminum and iron in soils. Ann. Agr. Fenn. 16: 227—237. Pawluk-, S. 1972. Measurement of crystalline and amor- phous iron removal in soils. Can, J. Soil Sci. 52: 119—123. Pionke, H.B. & Corey, R.B. 1967. Relations between acidic aluminum and soil pH, clay and organic matter. Soil Sei. Soc. Amer. Proc. 31: 749—752. Pratt, P.F. & Bair, F.L. 1961. A comparison of three reagents for the extraction of aluminum from soils. Soil Sci. 91: 357—359. Rengasamy, P. & Oades, J.M. 1977. Interaction of mono- meric and polymeric species of metal ions with clay surfaces. I. Adsorption of iron (III) species. Aust. J. Soil Res. 15; 221—233. Rubio, J. & Matijevic, E. 1979. Interactions of metal hydrous oxides with chelating agents. 1. P-FeOOH- EDTA. J. Colloid Interface Sci. 68: 408—421. Ryti, R. 1965, On the determination of soil pH. J. Scient. Agric. Soc. Finl. 37: 51 —60. Schnitzer, M. & Skinner, S.I.M. 1964. Organo-metallic interactions in soils: 3. Properties of iron- and alu- minum-organic-matter complexes, prepared in the SELOSTUS Kivennäismaiden uuttava alumiini, rauta ja mangaani 111 Uuttomcnetelmien vertailu Raina Niskanen Helsingin yliopisto, Maanviljelyskemian laitos, 00710 Helsinki Kivennäismaiden (n = 102) alumiinia, rautaa ja man- gaania uutettiin 0,05 M oksalaatilla (pH 2,9), 0,05 M K4P,C) 7 :lla, 0,02 M EDTAdIa (pH 5,3) ja 1 M ammo- niumasetaatilla (pH 4,8). Savi- ja hiesumaissa (n = 51) eri menetelmiilä uuttavan alumiinin pitoisuudet olivat kiinteässä korrelaatiossa keskenään, karkeammissa maissa (n = 51) korrelaatio ei ollut kovin kiinteä. Savi- ja hie- sumaissa orgaanisen hiilen pitoisuus ja maan pH yhdes- laboratory and extracted from a soil. Soil Sci. 98: 197—203. Schwertmann, U. 1959. Die fraktionierte Extraktion der freien Eisenoxyde in Boden, ihre mineralogischenFor- men und ihre Entstehungsweisen.Z. Pflanzenernähr. Diing. Bodenk. 84: 194—204. 1964. Differenzierung der Eisenoxide des Bodens durch photochemische Extraktion mit saurer Am- moniumoxalat-Lösung. Z. Pflanzenernähr. Diing. Bodenk. 105: 194—202. 1973. Use of oxalate forFe extraction from soils. Can. J. Soil Sci. 53: 244—246. Shoji, S. & Fujiwara, Y. 1984. Active aluminum and iron in the humus horizons of Andosols from Northeastern Japan: their forms, properties, and significance in clay weathering. Soil Sci. 137: 216—226. Tamm, O. 1922. Eine Methode zur Bestimmung der anorganischen Komponente des Gelkomplexes im Boden. Statens Skogsförsöksanstalt, Medd. 19: 387—404. Titova, N.A. 1962. Iron humus complexes in certain soils. Soviet Soil Sci. 1351—1356. Wada, K. & Harward, M.E. 1974. Amorphous clay con- stituents of soils. Adv. Agron. 27: 211—260. Viro, P.J. 1955. Use of ethylenediaminetetraacetic acid in soil analysis: I. Experimental. Soil Sci. 79: 459—465. Yuan, T.L. & Fiskell, J.G.A. 1959. Aluminum studies: 11. The extraction of aluminum from some Florida soils. Soil Sci, Soc. Proc. 23: 202—205. Ms received January 12, 1988 sä selittivät 50 % oksalaattiuuttoisen alumiinin, 70 % pyrofosfaattiuuttoisen alumiinin, 53 % pyrofosfaatti- uuttoisen raudan ja 56 % asetaattiuuttoisen alumiinin vaihtelusta. Saveksen ja orgaanisen hiilen pitoisuudet yhdessä maan pH:n kanssa selittivät 77 % EDTA- uuttoisen alumiinin vaihtelusta savi- ja hiesumaissa. Karkeammissa maissa maan ominaisuudet selittivät hei- kosti uuttavien metallien pitoisuuksien vaihtelua. 97