THE FATE OF WATER-SOLUBLE PHOSPHATE APPLIED TO SOME MINERAL SOILS Armi Kaila University of Helsinki, Department of Agricultural Chemistry Received January 13, 1965 In a previous paper (9) results were reported on the distribution of applied water-soluble phosphate in the various fractions of inorganic phosphorus in 180 samples of mineral soils. Samples suspended in KH2P04-solution in the ratio of 1 to 50 of soil to solution retained, on the average, about one half of the applied 250 mg P/kg of soil during a period of contact of 24 hours. On the average, about 56 per cent of the retained phosphorus was found in the fluoride soluble fraction, and about 40 per cent in the alkali soluble fraction when the treated samples were analysed by the method of Chang and Jackson (3) after standing moist for three days since the removal of the phosphate solution by centrifuging. It was emphasized that the results may be significantly different, if the period of contact would be longer, and also, if smaller or larger amounts of phosphate were applied. In order to get more information about the effect of the rate of the application of the soluble phosphate on the forms in which it will be retained, some further studies were performed. An incubation experiment under the laboratory conditions was carried out, and also the distribution of varying amounts of added phosphate shortly after the application was studied. The results are compared with some data from a couple of field experiments. Material and methods The material of the present stud}* consists of twelve samples chosen to represent mineral soils of a largely varying phosphate sorption capacity and of different patterns of phosphate retention. The samples are listed in Table 1 with some of the characteristics important for the present study. Ten of the samples are from the plough layer of arable lands, the sample 4 is from the surface layer of a virgin soil, and the sample 12 originates from the layer of the depth of 20 to 30 cm of a https://www.c-info.fi/en/info/?token=31887wxutqq6w8D_.5xsUGD_GRHWSTBYE_W-nKg.bGAkHFo3M7KL7Lova4jGIYscPmP9UIrb1-24CqhpkSv03jLhtkGRa3WDKt-B8Fsh0tZkY0O7yo-7lE_pp_LUSJjqjDrFVWfPE_w_q2kfqd0f7ilVykjBFUFWIO50HftYDsShMheaFY6TuVq-C5d1ezh8gine0YkA 105 Table 1. Soil samples Clay Org.C AI Fe Inorg. P pm. extracted by % % pH PPm ppm k NH 4 CI NH,F NaOH H2 S0 4 1. Vi 3a Clay loam 47 4.3 4.4 2. VN 1 » 43 5.8 4.5 3. VN 2 • 40 4.6 5.1 4. Vi 2a Sandy clay 45 1.0 4.8 5. K 107 Silt 18 2.1 4.7 6. K 100 » Il 4.8 4.9 7. K 5 » 18 2.4 5.5 8. C 7 Sandy clay 47 3.6 6.1 9. C 1 Fine sand 8 1.9 5.1 10. C 2 » O 2.9 5.4 II K 101 Silt 22 3.5 5.3 12. To lb Silty clay 40 0.5 6.4 7420 22080 1000 0 17 312 99 6990 17010 820 1 80 512 168 4300 11170 796 1 154 492 210 2650 6650 252 0 5 116 200 2330 4250 232 2 37 145 288 2950 4700 201 1 39 137 253 1490 3030 142 2 69 133 390 2260 4740 135 3 53 186 438 2140 2780 130 1 34 61 263 1790 990 93 0 29 12 27 1910 3670 90 1 18 76 351 990 1520 55 0 6 22 526 virgin soil. Samples 1 and 2 represent the so-called Litorina soils, postglacial marine sediments characterized by a fairly high acidity and high contents of soluble salts and sesquioxides. The soil pH was measured in 1: 2.5 suspension in 0.01 M CaCl 2 by the glass electrode. The content of organic carbon was estimated by the procedure of Walkley, using the iodometric titration, and the content of clay was determined by the common areometer method. Aluminium and iron were extracted by Tamm’s acid ammonium oxalate solution. Aluminium was determined by the Aluminon method and iron by the sulfosalicylic acid procedure after the destruction of the organic matter by ignition. The indicator of the inorganic-P-sorption capacity was estimated by a procedure based on theFreundlich adsorption isotherm (cf. 7). The fractionation of inorganic phosphorus was performed by the method of Chang and J ackson (3), instead of the neutral NH,F-solution a slightly alkaline extractant was used. The occluded forms were not determined. The samples are listed in the order of the decreasing values of the indicator of the phosphorus sorption capacity, k. In this material it ranges from 1000 to 55. The average value of k in a larger material of our soils was found to be 290 ±l7 in sand and fine sand soils, 201 ± 24 in loam and silt soils, and 308 ± 20 in clay soils (7). It does not appear to be closely connected with the clay content, or with the acidity. On the other hand, the amounts of acid oxalate soluble aluminium and iron seem to be associated with the values of k also in the present material. The results of the fractionation of inorganic phosphorus reveal some features typical of our soils (8). The content of the NH4CI-soluble phosphorus is very low. An extremely low content of NH 4F-soluble phosphorus is found in the samples 4 and 12 from the virgin soils, and this fraction is smaller than that soluble in alkali in all the other soils except in the fine sand sample C 2. The relatively high content of alkali-soluble phosphorus characterizes the samples of Litorina soils, numbers 1 and 2. The occurence of a fairly large part of the inorganic phosphorus in the 106 other samples as the acid-soluble form, may be attributed to the fact that our soils are rather young, and the weathering processes appear to have only sparingly mobilized the original apatite phosphorus to forms bound by iron and aluminium. The treatments in the incubation experiment were0, 100, 200, or 500 mg P/kg of soil. 100 g samples of air dry and ground soil were weighed in glass jars, and 50 ml of distilled water, or 50 ml of KH 2P0 4 solutions containing 200, 400, or 1000 mg P/l, respectively, were added. The soils were dried to about the field capacity, mixed well, and incubated under a loose cover for three months at 18 —2O°C. The samples were air-dried and ground before analysing. In order to preparate the material for the study of the distribution of the applied phosphorus immediately after the application, 1 g samples of air-dry and ground soil were weighed in a centrifuge tube, and 50 ml of distilled water, or solutions of KH 2P04 containing2,4, or 10 mg P/l, resp., were added. Thus the applications of phosphorus correspond to those in the incubation experiment. In addition, one sample was treated with phosphate solutions corresponding to applications of 1000, 2000, 5000, or 10000 mg P per kilogram of soil. The suspensions were shaken for two hours, centrifuged, and the fractionation of inorganic phosphorus was carried out. The increase in the phosphorus content of the various fractions in the samples treated with phos- phate as compared with the values obtained for the samples treated with distilled water is taken to indicate the accumulation of the applied phosphate. Mineralization of soil organic phosphorus, or microbiological immobilization of inorganic phosphorus is supposed to be equal in all the treatments, and relatively insignificant. In the incubation experiment it was found that the totalamount of phos- phorus extracted by the fractionation procedure from the samples incubated for three months without any application of phosphate were 0 to 25 ppm higher than those from the original samples. Results The results of the incubation experiment are reported in Table 2. The pH values in the incubated samples differ from the orginal ones only by —0.3 to -)-0.1 pH units. The changes seem to be independent on the treatments. Thus, differences in the acidity of the variously treated samples are not likely to play any significant role in the distribution of the applied phosphorus. The recovery of the phosphorus is usually somewhat less than 100 per cent of the amount added. There are no reasons to suppose that this deficit would be caused by an accumulation of phosphorus in the occluded forms. It is probable that losses during the fractionation procedure and a possibly heterogeneous distribu- tion of the added phosphorus in spite of the thorough mixing account for the in- complete recovery. As could be expected, the largest part of the applied soluble phosphate is recovered in the fluoride-soluble and alkali-soluble forms. Only in a few samples it has remained soluble in ammonium chloride to any marked extent. On the other hand, the acid-soluble fractions has also accumulated very little or not at all of the applied soluble phosphate, except in some samples with a fairly high pH and a low content of iron and aluminium. It is likely that the small increases in this fraction in acid samples may be due to analytical errors. There are typical differences in the pattern of the distribution of the applied phosphorus between the fluoride-soluble and alkali-soluble forms. The clay loam sample Vi 3a has retained 80 to 90 per cent of the recovered phosphorus in the Table 2. Distribution of the applied P in various fractions in soil samples incubated for three months P P c , tj Per cent of recovered P extracted bySample ppm pH ppm applied recovered NH,CI NH,F NaOH H 2SO, 1. Vi 3a 100 4.5 95 0 11 88 1 200 4.5 185 0 15 84 I 500 4.5 492 0 17 80 3 2. VN 1 100 4 3 102 0 27 73 0 200 43 199 0 32 68 0 600 4.3 483 1 36 63 0 3. VN 2 100 4.7 104 (I 45 55 0 200 4.7 201 0 51 41» 0 50(1 4.7 491 2 52 45 1 4. Vi 2a 100 4.9 83 0 29 71 0 200 4.9 172 0 34 66 0 500 5.0 461 2 42 54 2 B. K 107 100 4 4 98 2 52 43 3 200 4.4 178 4 59 35 2 500 4.4 477 5 63 29 3 6. K 100 100 4.5 93 3 59 35 3 200 4.5 179 4 64 28 4 50(i 4.5 481 6 67 25 2 7. K 5 100 5.2 104 7 56 36 1 200 5.2 204 8 55 32 5 500 5.2 489 12 60 25 3 8. C 7 100 5.8 90 3 45 48 4 200 5.8 189 4 40 43 13 500 5.8 484 7 48 40 4 9. C 1 100 5.0 88 1 52 40 7 200 5.0 192 3 61 31 5 500 5.2 469 3 59 35 3 10. C 2 100 5.3 90 2 85 12 1 200 5.3 179 3 88 8 1 500 5.3 486 6 86 8 0 11. K 101 100 5.0 94 5 48 36 11 200 5.1 192 7 53 34 6 500 5.0 492 11 58 26 5 12. To lb 100 6 5 93 6 47 36 11 200 6.5 176 15 51 23 11 500 6.3 487 22 54 17 7 107 108 latter fraction supposed to be bound by iron whereas the fine sand sample C 2 has accumulated 85 to 88 per cent of it in the former fraction which is assumed to be mainly connected with aluminium. In the other samples the applied phosphorus appears to have been divided more equally between these fractions. Usually, the fluoride-soluble part is larger, but in the samples 1, 2 and 4 the alkali-soluble fraction dominates. The most interesting question of this experiment, the effect of the rate of the application on the distribution is answered by these results fairly distinctly. The larger the amount of phosphate applied, the higher part of it may be found in the fluoride soluble form and the less in the alkali-soluble form. Also the part which remains soluble in NH 4CI increases with an increase in the application. When the soil samples were submitted to the fractionation procedure immediately after the treatment with the phosphate solutions, results recorded in Table 3 were obtained. The part of applied phosphorus retained during the contact of two hours tends to decrease from the samples with a high value of k to those with a low k, although not quite regularly. It is of particular interest to note that the sample Vi 3a, with a value of k = 1000, is able toretain within two hours the lowest applica- tion of phosphate completely, and even about two thirds of the application of 500 ppm. The percentile retention decreases, of course, when the application is increased, but the absolute amounts increase. From the 10000 ppm soluble phosphorus added this sample sorbs 2200 ppm. Even in this case more than one half of the retained phosphorus is found in the alkali soluble fraction which fact indicates the dominance of iron in the pattern of phosphorus fixation in this soil. The samples K 5 and K 101, on the other hand have retained only 20 ppm P or less from the application of 100 ppm, and 16 or 15 per cent from the application of 500 ppm. The figures in Table 3 showing the percentage of the recovered phosphorus in the various fractions differ surprisingly little from the corresponding values in Table 2 for the samples incubated for three months. The part which has been remained soluble in ammonium chloride is slightly higher in the samples analysed immedi- atety after the application of phosphate, but this fraction, apparently, represents in this case only an indefinite portion of the phosphate not sorbed by the soil con- stituents during the two hours of contact. The percentage of recovered phosphorus in the fluoride-soluble fraction is higher and that in the alkali-soluble fraction correspondingly lower than in the incubated samples, but the differences may be rather small in some samples. Only in few cases, some of the applied phosphorus has been recovered in the acid-soluble form. In accordance with the results obtained for the incubated samples, the part of the applied phosphate found in the fluoride- soluble form tends to increase with an increase in the application whereas the contrary is true in regard to the alkali-soluble form. The pattern of the retention appears to be characteristic of the soil from the period of contact of two hours to the period of three months. In the sample Vi 3a 79 to 82 per cent of the recovered phosphorus in the unincubated samples was in the alkali-soluble fraction, in the incubated samples the corresponding part was from 80 to 88 per cent. In the sample C 2, on the other hand, 70 to 80 per cent of the recovered phosphorus was sorbed in the fluoride-soluble form during the two 109 Table 3. Distribution of the applied P in various fractions in soil samples without incubation Per cent Per cent of recovered P extracted by P ppm of P Sample applied retained NH 4 CI NH 4F NaOH H 2SO, 1. Vi 3a 100 100 2 16 82 0 200 75 2 17 81 0 500 67 2 19 79 0 1000 61 1 22 73 4 2000 47 2 26 70 3 5000 31 3 31 64 2 10000 22 5 35 55 5 2. VN 1 100 82 2 37 61 0 200 69 2 38 60 0 500 58 3 40 57 0 3. VN 2 100 80 3 46 51 0 200 54 2 50 48 0 500 45 3 55 42 0 4. Vi 2a 100 55 2 38 60 0 200 48 3 38 59 0 500 33 3 45 52 0 5. K 107 100 57 5 48 47 0 200 39 6 53 41 0 500 26 9 57 34 0 7- K 5 100 20 20 65 15 0 200 18 11 64 25 0 500 16 16 59 25 0 8. C 7 100 37 8 51 41 0 200 29 8 54 38 0 500 20 8 59 33 0 9. Cl 100 28 11 64 25 0 200 25 6 67 27 0 500 16 10 74 16 0 10. C 2 100 30 0 70 6 24 200 26 2 77 6 15 500 20 5 80 6 9 11- K 101 100 19 6 76 18 0 200 19 7 70 23 0 500 15 13 64 23 0 12. To lb 100 29 4 48 24 24 200 22 5 53 30 12 5OO 13 12 49 21 18 Table 4. Ratio of the applied P in the fluoride-soluble fraction to that in the alkali-soluble fraction, and the absolute differences in these fractions in the unincubated (a) and incubated (b) samples Sorbed »Al-P/Fe-P» b-a Sample Al/Fe »Al-P/Fe-P» ppm a b »Al-P» »Fe-P» applied PP m PPm 1. Vi 3a 0.7 0.05 100 0.2 0.1 -6 2 200 0.2 0.2 2 33 500 0.2 0.2 20 129 2. VN 1 0.8 0.16 100 0.6 0.4 -3 24 200 0.7 0.5 12 52 500 0.7 0.6 58 139 3. VN 2 0.8 0.31 100 0.9 0.8 10 56 200 1.1 1.0 49 46 500 1.3 1.2 131 126 4. Vi 2a 0.8 0.04 100 0.6 0.4 3 26 200 0.6 0.6 22 57 500 0.9 0.8 120 163 5. K 107 1.1 0.26 100 1.0 1.2 27 15 200 1.3 1.7 64 30 500 1.8 2.2 227 94 7. K 5 1.0 0.52 100 4.3 1.6 45 34 200 2.6 1.7 89 56 500 2.4 2.4 246 102 8. C 7 1.0 0.28 100 1.3 0.9 22 28 200 1.5 0.9 45 59 500 1.8 1.2 173 161 9. C 1 1.6 0.56 100 2.6 1.3 28 28 200 2.4 2.0 83 46 500 4.5 1.7 218 151 10. C 2 3.7 2.41 100 10.5 6.9 56 9 200 13 3 11.3 118 11 500 13.2 10.4 338 33 11. K 101 1.1 0.24 100 3.0 1.3 31 31 200 3.0 1.5 75 56 500 2.7 2.2 237 111 12. To lb 1.4 0.27 100 2.0 1.3 30 26 200 1.8 2.2 67 27 500 2.3 3.1 230 69 110 111 hours of contact, whereas the values found for the incubated samples were 85 to 88 per cent. The patterns of the other soils are less distinct, but almost without an exception the samples which in the incubation experiment had accumulated more applied phosphorus in the alkali-soluble than in the fluoride-soluble form, or vice versa, did so also when the fractionation was performed after two hours of contact. This is particularly noteworthy, since in some soils the immediate retention has been only from 13 to 30 per cent of that in the incubation experiment. Since in our soils the retention of soluble phosphorus seems to be almost com- pletely depending on the factors determining its extractability by the fluoride or the alkali solutions of the fractionation procedure, a more detailed study of these frac- tions in the results of the present experiments may be illuminating. Table 4 contains data which show the ratio between the increases in the so-called »Al-P», or the fluo- ride-soluble fraction, and in the »Fe-P», or the alkali-soluble fraction both in the results obtained immediately and in those from the incubation experiment. Also the ratios of the acid oxalate soluble aluminium to iron (in moles) and the ratios of the fluoride-soluble and alkali-soluble phosphorus in the original samples are recorded. These data show that the distribution of the applied phosphorus in these two fractions in the incubated and unincubated soils is fairly similar in the first four samples. All of them are characterized by a higher increase in the alkali-soluble than in the fluoride-soluble form, at least in connection with the lowest application. This is in accordance with the molar ratio of oxalate-soluble aluminium to iron being less than 1 in these soils. In samples having a higher ratio of Al/Fe, the applied phosphorus has been accumulated in the fluoride-soluble form in larger amounts than in the alkali-soluble form. Although also in these samples the ratio between the sorbed amounts of »Al-P» and »Fe-P» tends to increase with the rate of applica- tion, this seems to be less regular than in the first samples. In these soils, also the ratios in the incubated samples partly appear to be higher than those in the unin- cubated ones. It has been claimed (2) that while the newly retained phosphorus is mainly fixed as the aluminium bound form, the amount of the iron bound form will increase in time and usually surpass the other forms. The data in Table 4 showing the differ- ence in the fluoride-soluble phosphorus and in the alkali-soluble phosphorus in the incubated and unincubated samples indicate that this is not always the case. In the iron dominated samples Vi 3a, VN 1, and Vi 2a the increase in the alkali-soluble fraction during the incubation has been markedly higher than that in the fluoride- soluble fraction. Yet, in most of the other soils the increase in the latter fraction has been equal to, or even significantly higher than that in the iron bound form. In all samples the relative increase in the fluoride-soluble fraction as compared to that in the alkali-soluble fraction tends to be the higher the higher the application of phosphate. According to these laboratory experiments, the part of the applied phosphate sorbed as the fluoride-soluble form usually seems to be the higher and the part sorbed as the alkali-soluble form the lower the larger the amount of phosphate added. Under the field conditions this pattern may be less distinct, particularly on account of the effect of the uptake of phosphorus by plants, and also because of the less thorough mixing of the fertilizers with the soil. 112 In a field experiment on a silty clay soil with the pH 4.9, the annual applications of 400 or 800 kg of superphosphate (8.5 % P) per hectare for 12yearshave increased the amount of the phosphorus in the various fractions as compared to the results obtained for the samples from the untreated plots with the following quantities: Superphosphate 400 kg/ha NH 4CI-soluble P NH 4-F-soluble P NaOH-soluble P 2 ppm 32 ppm 43 ppm 92 * 109 *800 » 6 * No increase in the acid-soluble fraction could be detected. In this soil somewhat more of the applied fertilizer phosphorus appears to be accumulated in the alkali- soluble than in the fluoride-soluble fraction. Yet, the higher application has resulted in a relatively higher increase in the latter form. Also a slightly higher part of the applied phosphorus is remained soluble in ammonium chloride in the samples which got the heavier application. In a field experiment on an acid peat soil (pH 4.2) reported in a previous paper (5), the annual applications of superphosphate during 34 years were 100, 200, and 300 kg/ha, respectively. The differences in the various phosphorus fractions between the samples from these treated plots and the untreated ones were the following: Superphosphate 100 kg/ha NHj-F-soluble P 4 ppm 20 * 66 * NaOH-solube P 16 ppm 200 » 300 » 54 » 148 » In this soil rich in iron a far larger part of the applied phosphorus is found in the alkali-soluble than in the fluoride-soluble fraction. However, even in this case the ratio between these increases in the former and the latter fractions tends to be the higher the higher the application of super-phosphate. These ratios are 0.25, 0.37, and 0.45, for the 100, 200, and 300 kg of superphosphate, respectively. In this soil from 44 to 53 per cent of the increase in phosphorus content caused by the fertiliza- tion is found in the fraction of organic phosphorus. Also the acid-soluble fraction in the treated samples was larger than that in the untreated plots. This is likely to be partly connected with the possibility that in this soil, very poor in available phosphorus, the crops of the untreated plots have been compelled to utilize also the apatite-phosphorus in the mineral fraction of the clayed peat. In any case, in this old experiment other factors than the sorption of phosphate by the soil constit- uents have played their role. Discussion The results of the present study show that the soils tend to have a characteristic pattern of phosphate retention. In the extreme cases it means that in one soil the applied water-soluble phosphate is almost completely sorbed as the alkali-soluble form supposed to be bound by iron and its compounds, whereas in an other soil by far the largest part of the sorbed phosphorus is found in the fluoride soluble fraction 113 which is assumed to be bound mainly by aluminium and its compounds. In such extreme cases the pattern appears to be distinct very soon after the phosphate has come into contact with the soil, and this pattern will in time grow more evident. In most of our soils, however, the distribution of the applied phosphate between the fractions supposed to be bound by aluminium or iron appears to be more equal. This could be proved in a previous work (9) on the basis of a larger material as to the newly retained phosphorus. The present results also indicate that these fractions tend to be of the same order, and usually the part of the fluoride-soluble forms is somewhat higher than that of the alkali-soluble forms. The retention of applied soluble phosphorus by calcium as an apatite-like compound seems to occur in our soils only seldom to any marked degree. Another tendency revealed by the present results is the relative increase in the fluoride-soluble fraction as compared with the alkali-soluble fraction when the rate of the application of the soluble phosphate increases. This is apparent even in those samples in which the pattern of phosphate retention is dominated by iron. The few data from the field experiments are in accordance with the results of the laboratory studies. It is likely that in the fractionation procedure used, the ammonium fluoride solution will extract not only aluminium bound phosphorus but also e.g. dicalcium- phosphate (5, 6, 10, etc.). The importance of this fraction as a source of phosphorus for plants has been emphasized (4, 11). In alkaline soils the uptake of phosphorus has been reported to be correlated with the iron bound form (1). There are no in- formations of the availability of phosphorus in the various fractions in our soils. Probably, there are differences even within the same fraction in the intensity of the sorption. It is possible that e.g. the newly retained phosphorus in the fluoride- soluble or alkali-solublefraction is more easily available for the plants than phospho- rus in the same fractions in the samples incubated for three months. If fhe fluoride- soluble fraction would represent phosphorus which the plants are able to utilize more easily than the alkali-soluble phosphorus, the present results would, for their part, explain why a fairly large application of phosphate fertilizers is necessary to secure the phosphorus nutrition of the crops in most of our soils: the part of the applied phosphorus sorbed in the fluoride-soluble forms is the lower the lower the application of fertilizer phosphorus. Attention must be paid to the fact that the lowest amount of phosphorus added in the present laboratory experiments was 100 ppm which corresponds to about 2200 to 2800 kg superphosphate per hectare. From this addition a slightly higher part was sorbed as the fluoride-soluble form than as the alkali-soluble form. When only about 200 to 400 kg superphosphate is added per hectare, the part which will be retained as the fluoride-soluble fraction may be markedly lower than that retained in the alkali-soluble fraction. Summary The distribution ot applied water-soluble phosphorus in the various fractions of soil inorganic phosphorus was studied in an incubation experiment under the laboratory conditions. Samples of twelve soils were incubated for three months 114 at room temperature with the applications of KH2P0 4 in amounts corresponding to 100, 200, or 500 mg P/kg of soil. The results of the fractionation showed that in most samples the applied phosphorus could be found almost completely in the fluoride- soluble and alkali-soluble fractions, the part of the former being the higher and that of the latter the lower the higher the rate of the phosphate application. Similar results were obtained also when the fractionation was performed after the samples had been in contact with the phosphate solutions only for two hours. Analyses of samples from two field trials were in accordance with these results. The soils tended to have a characteristic pattern of phosphate retention which in extreme cases means an almost complete sorption of the applied phosphate either as the fluoride-soluble form or as the alkali-soluble form. In most soils, however, the distribution was more equal. Usually the fluoride-soluble part of the recovered phosphorus tended to be somewhat higher than the alkali-soluble part. This tendency was more distinct in regard to the newly retained phosphorus. The effect of phosphate fertilizers in our soils is discussed on the basis of the results. REFERENCES (1) Al-Abbas, A. H. & Barber, S. A, 1964. A soil test for phosphorus based upon fractionation of soil phosphorus. Soil Sei. Soc. Amer. Proc. 28: 218 224. (2) Chang, S. C. & Chu, W. K. 1961. The fate of soluble phosphate applied to soils. J, Soil Sci. 12: 286-293. (3) —»— & Jackson, M. L. 1957. Fractionation of soil phosphorus. Soil Sci. 84: 133 144. (4) Hanley, K. 1962. Soil phosphorus forms and their availability to plants. Irish J. Agr. Res. 1: 192-193. (5) Kaila, A. 1961. Fertilizer phosphorus in some Finnish soils. J. Sci. Agr. Soc. Finland 33: 131 139. (6) —»— 1963. Fertilizer phosphorus in various fractions of soil phosphorus. Ibid. 35: 36 46. (7) —»— 1963. Dependence of the phosphate sorption capacity on the aluminium and iron in Finnish soils. Ibid. 35: 165 177. (8) — » 1964. Fractions of inorganic phosphorus in Finnish mineral soils. Ibid. 36: 1 13. (9) —» 1964. Forms of newly retained phosphorus in mineral soils. Ibid. 36: 65 76. (10) Laverty, J. C. & McLean, E. O. 1961. Factors affecting yields and uptake of phosphorus by different crops: 3. Soil Sci. 91: 166 171. (11) Mackenzie, A. F. 1962. Inorganic soil phosphorus fractions of some Ontario soils as studied using isotopic exchange and solubility criteria. Canad. J. Soil Sci. 42: 150.156. SELOSTUS: KIVENNÄISMAIHIN LISÄTYN VETEEN LIUKENEVAN FOSFORIN PIDÄTTYMISESTÄ Armi Kaila Yliopiston maanviljelyskemian laitos, Pihlajamäki Kahdentoista kivennäismaan näytteisiin lisättiin 100, 200 tai 500 mg P/kg KH,PC) 4:na ja näytteitä muhitettiin kolme kuukautta huoneen lämpötilassa. Toinen näytesarja analysoitiin heti, kun niitä oli kahden tunnin ajan huiskutettu vastaavat fosforin määrät sisältävissä liuoksissa. Fraktioinnin tulokset 115 osoittivat, että useimmissa maissa lisätty fosfori oli joutunut miltei kokonaan fluoridiin ja emäkseen liukeneviin fraktioihin. Edellisen osuus oli tavallisesti sitä suurempi ja jälkimmäisen sitä pienempi, mitä enemmän fosforia oli lisätty. Vastaavanlaisia tuloksia saatiin myös analysoimalla eräitten kenttä- kokeitten maanäytteitä. Maat näyttivät edustavan tiettyjä fosforin pidätyksen tyyppejä, joiden äärimmäisyyksinä olivat maat, jotka pidättivät lisätyn fosforin miltei kokonaan joko fluoridiin tai emäkseen liukenevaan muo- toon. Useimmissa maissa lisätty fosfori jakautui tasaisemmin näiden fraktioiden kesken. Myös muhi- tetuissa näytteissä fluoridiin liukeneva osa oli tavallisesti hiukan suurempi kuin emäkseen liukeneva. Todettiin, että mikäli emäkseen liukeneva fraktio todella edustaisi huomattavasti vaikeammin kasveille käyttökelpoista fosforia kuin fluoridiin liukeneva osa, olisi syytä käyttää etenkin runsaasti rautaa sisältävissä maissa voimakasta fosforilannoitusta satojen fosforin saannin turvaamiseksi.