p 1 1 » <^ Issued August 24, 1914. HAWAII AGRICULTURAL EXPERIMENT STATION, E. V. WILCOX, Special Agent in Charge. Bulletin No. 35. ABSORPTION OF FERTILIZER SALTS BY HAWAIIAN SOILS. BY Wm. McGEORGE, ASSISTANT CHEMIST. UNDER THE SUPERVISION OF OTTTCE OF EXPERIMENT STATIONS, U. 8. DEPARTMENT OF AGRICULTURE. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1914. Issued August 24, 1914. HAWAII AGRICULTURAL EXPERIMENT STATION, E. V. WILCOX, Special Agent in Charge. Bulletin No. 35. ABSORPTION OF FERTILIZER SALTS BY HAWAIIAN SOILS. BY Wm. McGEORGE, ASSISTANT CHEMIST. UNDER THE SUPERVISION OF OFFICE OF EXPERIMENT STATIONS, U. 8. DEPARTMENT OF AGRICULTURE. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1914. HAWAII AGRICULTURAL EXPERIMENT STATION, HONOLULU. [Under the supervision of A. C. True, Director of the Office of Experiment Stations, United States Department of Agriculture.] Walter H. Evans, Chief of Division of Insular Stations, Office of Experiment Stations. STATION STAFF. E. V. Wilcox, Special Agent in Charge. J. Edgar Higgins, Horticulturist. W. P. Kelley, Chemist. C. K. McClelland, Agronomist. D. T. Fullaway, Entomologist. Wm. McGeorge, Assistant Chemist. Alice R. Thompson, Assistant Chemist. C.J. Hunn, Assistant Horticulturist. V. S. Holt, Assistant in Horticulture. C. A. Sahr, Assistant in Agronomy. (2) LETTER OF TRANSMITTAL. Honolulu, Hawaii, September 29, 1913. Sik: I have the honor to submit herewith and recommend for publication as Bulletin No. 35 of the Hawaii Agricultural Experiment Station, a paper on the Absorption of Fertilizer Salts by Hawaiian Soils, by William McGeorge, assistant chemist. In order to be in position to recommend a rational program for the management of Hawaiian soils it has been found necessary to make a study of all the properties of these soils. In the present paper many interesting points are brought out upon the subject of the fixing power of these soils for different fertilizer salts. It appears that the concentration of a soil solution depends perhaps more upon the fixing power of the soils than upon the solubility of the salt. Respectfully, E. V. Wilcox, Special Agent in Charge. Dr. A. C. True, Director Office of Experiment Stations, U. S. Department of Agriculture, Washington, D. C. Recommended for publication. A. C. True, Director. Publication authorized. D. F. Houston, Secretary of Agriculture. (3) CONTENTS. Object of work 5 Soil types used 5 Method 6 Absorption of phosphoric acid 6 Absorption of potash 10 Absorption of nitrogen 12 Ammonium sulphate 12 Sodium nitrate 14 Absorption of fertilizer salts by fresh and air-dried soils 16 Absorption of phosphoric acid 16 Absorption of potash 19 Absorption of nitrogen 20 Ammonium sulphate 20 Sodium nitrate 21 Absorption of fertilizer salts when applied in mixtures, and the effect of heat and antiseptics 22 Absorption of phosphoric acid 22 Absorption of potash 24 Absorption of nitrogen 26 Ammonium sulphate 26 Sodium nitrate 27 Removal of absorbed salts 27 Removal of absorbed phosphate 28 Removal of absorbed potash 28 Removal of absorbed nitrogen 29 Summary 29 Acknowledgments 32 (4) ABSORPTION OF FERTILIZER SALTS BY HAWAIIAN SOILS. In undertaking investigations on soil fertility it is very necessary to have some knowledge of the absorptive or fixing power of a soil, since this factor is one of prime importance in the successful use of fertilizers and varies greatly with the physical structure, the organic matter content, and other factors of a chemical and biological nature. OBJECT OF WORK. The object of the work here presented was to give some under- standing of the absorptive power of Hawaiian soils for fertilizer salts. These soils contain an abnormally high percentage of iron and alu- minum compounds, and from their physical condition would be expected to have a high fixing power. Many of the soil types of the islands also contain large amounts of organic matter and humus. J. T. Crawley x carried on some experiments with Hawaiian soils to determine the effect of irrigation upon added fertilizer salts. He found phosphoric acid to be firmly fixed, while ammonium sulphate and potassium sulphate were not so strongly fixed. SOIL TYPES USED. Soils representing in a general way the important types of the islands were selected for the work. The following table shows the chemical composition of the soils, as determined by digestion in hydrochloric acid of specific gravity 1.115: Composition of soils used in the experiments. Constituents. Moisture Volatile matter Insoluble matter Ferric oxid (Fe 2 3 ; Alumina (A1 2 3 ) Titanium oxid (Ti0 2 ). . . . Manganese oxid (Mn 3 4 ;. Lime(CaO) Magnesia (MgO) Potash (K 2 0) Soda(Na 2 0) Sulphur trioxid (SO a ). .. . Phosphoric acid (P^ >») - - Soil No. Soil No. 292. 448. Per cent. 7.65 8.42 38.49 16.63 12.85 2.00 .24 1.84 8.71 .39 1.36 .08 .57 Per cent. 15.00 25.58 15.10 19.20 16.64 4.20 .06 .50 1.80 .15 .68 .53 .29 Soil No. 428. Per cent. 14.95 22.24 34.99 8.24 10.73 3.20 .20 1.91 2.24 .24 1.40 .45 .22 Soil No. 474. Per cent. 13.59 20.01 33. 77 7.00 1G.79 1.80 .07 3.80 .85 .72 .10 .45 2.18 Soil No. 517. Per cent. 3.54 13.71 41.99 21.76 17.23 i Jour. Amer. Chem. Soc., 24 (1902), p. 1114; 25 (1903), p. 47. (5) Soil No. 518. Per cent. 3.97 13.56 41.53 21.46 18.21 .12 .36 .32 .54 .23 .58 .13 .04 .20 .24 .66 .46 .52 .16 Soil No. 292. This type of soil occurs in the lowlands in and about Honolulu, now being used for growing bananas, rice, and for truck farming. It has a sandy texture, being partly derived from black or volcanic ash. It has a grayish-brown color, abnormally high mag- nesia content, and low content of organic matter. No. 448 represents the type of yellow clay scattered throughout the islands, this sample being taken near Hilo, Hawaii. No. 428 is a dark colored, highly organic soil from Glenwood, Hawaii. It has a very sandy texture, is subject to heavy rainfall, and is rather unproductive. No. 474 is a sample of soil from Parker ranch, Waimea, Hawaii. It is a brown-colored soil of floury texture and very productive. No. 517 represents the type of soil which is most abundant in the islands, namely, the heavy red clay, a highly ferruginous type. METHOD. The method of treatment adopted in this investigation was as fol- lows: 100 grams of air-dry soil was placed in glass tubes, 1 inch in diameter, and fitted with rubber stoppers and pinchcock to regulate the passage of the solution through the soil. The percolation was regu- lated to flow at a rate of 100 cubic centimeters in 24 hours, and each successive 100 cubic centimeters of percolate was analyzed. The salts used were sodium nitrate, potassium phosphate, and calcium phos- phate, separately and as a mixture. One series was also heated to 230° C. and another treated with chloroform to determine the effect of these agents upon absorption. All determinations were made- by colorimetric methods, except those of potash, which was precipitated and weighed as potassium chloroplatinate. ABSORPTION OF PHOSPHORIC ACID. In this series the percolation was carried on for nearly two months, 5 liters of the solution of potassium phosphate passing through the soil. The solution used contained about 200 parts phosphoric acid (P0 4 ) per million, and each time a new solution was made up the strength was determined by analysis. Owing to the fact that percolation through a column of the soil was found to be impossible, due to the strong deflocculating effect of this salt, the percolation in this series was carried on in funnels. Even then several of the samples filtered very slowly. The filtrate from the clay soil was very cloudy, and the percolates became slightly stagnant in several instances after the percolations had been carried on for about one and a half months. In order to get a clear conception of the fixation of phosphates it is necessary to have some idea of the solubility of phosphoric acid already present in the soil when treated in the same way as in the experiments. For this purpose the glass tubes were filled with 100 grams of soil, covered with distilled water, and each 100 cubic centi- meters of filtrate analyzed. Phosphoric acid removed from the soils by distilled water. [Expressed in parts per million of PO< in the percolate.] Percolates of 100 cc. each. Soil No. 292. Soil No. 448. SoU No. 428. SoU No. 474. Percolates of Soil No. 100 cc. each. 292. Soil No. 448. Soil No. 428. Soil No. 474. 100 6.4 3.2 3.8 4.4 3.8 4.4 500 5.6 600 11.2 700 12.0 800 10.8 2.8 2.0 2.0 3.6 4.6 4.4 6.0 20.0 7.0 200 10.8 300 8.8 3.8 5.2 400... 3.2 5.0 11.2 The general tendency of these soils is to yield a solution of fairly constant concentration. This is in direct harmony with what should be expected, namely that the phosphoric acid is so firmly retained by Hawaiian soils that the first leachings should not yield a more concentrated solution than those following. The following table shows the absorbing power of the soil for phos- phoric acid in monopotassium phosphate (KH 2 P0 4 ) : Absorption of phosphoric acid from a solution of monopotassium phosphate (KH 2 P0 4 ). [Expressed in parts per mUlion of PO* in the percolate.] SOLUTION CONTAINED 175 PARTS PER MILLION PO,. Percolates of SoU No. 100 cc. each. 292. Soil No. 448. Soil No. 428. Soil No. 474. Percolates of 100 cc. each. Soil No. Soil No. Soil No. 292. 448. ; 428. Soil No. 474. 100 45.6 13.6 11.2 17.2 '. 2,100 200 35.2 29.0 13.2 40.0 2,200 300 52.0 10.4 17.2 44.0 2,300 400 38.0 13.6 i 34.0 49.0 2,400 , 500 39.0 9.6 19.4 36.0 2,500 600T 48.0 11.2 15.6 I 39.0 2,600 700 57.0 15.6 16.8 I 42.0 2,700 800 ! 27.0 36.0 36.0 55.0 2,800 900 20.0 5.8 20.8 35.6 2,900 1,000 17.8 5.8 27.8 41.6 i 3,000 1,100 1 71.2 5.2 12.0 24.8 3,100 1,200 ! 37.2 6.8 13.2 40.0 3,200 1,300 1 72.0 6.4 I 14.0 34.4 j 3,300 1,400 , 60.0 10.0 i 16.8 20.8 3,400 1,500 ' 76.0 .7.2 I 12.0 52.0 3,600 1,600 72.0 5.6 | 11.6 56.0 3,800 1,700 i 42.0 13.6 j 12.0 24.0 4,000 1,800 64.0 9.6 8.0 20.0 4,200 1,900 ! 66.4 4.4 | 4.4 18.0 4,400 2,000 54.4 4.0 4.0 14.8 SOLUTION CONTAINED 140 PARTS PER MILLION P0 4 . I j j j j 4,600 40.0 6.0 7.6 24.0 5,000 34.4 I 4.0 4,800 24.8 6.8 8.4 23.2 62.8 4.0 5.2 60.0 5.2 4.8 60.0 4.0 4.0 66.4 5.6 6.0 62.8 4.8 4.8 56.0 4.8 4.8 44.0 4.0 4.8 28.8 4.0 4.8 31.2 3.6 4.0 39.2 4.4 4.4 21.6 4.0 4.0 6.8 5.6 5.6 33.6 5.6 5.6 25.6 10.0 29.6 20.8 8.0 8.8 46.6 10.0 9.6 29.6 4.4 6.4 46.4 5.2 12.4 34.4 4.8 8.0 25.6 38.4 22.4 22.4 31.2 16.8 32.0 28.8 35.2 32.0 24.0 26.4 44.0 36.8 29.6 31.2 34.4 48.0 48.0 4.0 24.0 Summary of above table. SoU No. PO, added to 100 gm. soU. PO< fixed by 100 gm. soU. Per cent ofP0 4 fixed. 292 Gram. 0.8540 .8540 .8540 .8540 Gram. 0.6872 .8146 .7977 .6882 80.6 448 95.5 428 93.3 474 80.7 8 The amount of phosphoric acid fixed from a solution of mono- calcium phosphate (CaH 4 (P0 4 ) 2 ) is shown in the following table: Absorption of phosphoric acid from a solution of monocalcium phosphate (CaB^ (P0 4 ) 2 ). [Expressed in parts per million of PO4 in the percolate.] SOLUTION CONTAINED 232 PARTS PER MILLION PO*. Percolates of '100 cc. each. Soil No. 292. Soil No. 448. Soil No. 428. Soil No. 474. Percolates of 100 cc. each. Soil No. 292. Soil No. 448. Soil No. 428. Soil No. 474. 100 40.0 24.0 23.2 50.0 10.4 7.2 10.8 14.0 9.6 9.2 8.0 14.4 24.8 18.4 22.4 41.0 500....' 33.0 17.2 13.6 16.0 11.6 11.6 15.2 11.2 10.4 39.0 200 600 19.2 300 700 16.8 400 SOLUTION CONTAINED 220 PARTS PER MILLION PO*. 800.. 900.. 1,000. 1,100. 11.6 11.6 11.2 14.4 15.6 4.4 6.4 11.2 24.0 4.0 4.8 21.2 30.4 4.8 6.0 21.6 1,200. 1,300. 1,400. 1,500. 14.4 6.0 6.4 18.4 4.0 8.4 28.8 5.2 10.0 17.6 4.4 8.0 SOLUTION CONTAINED 132 PARTS PER MILLION PO< 11.6 13.6 17.2 21.6 1,600 44.0 22.4 17.6 17.6 4.4 4.0 4.0 4.8 4.0 5.6 8.0 7.6 21.6 14.4 15.2 14.4 2,000 22. 4 21.6 35.2 4.0 5.2 5.6 4.0 7.2 6.4* 16.8 1,700 2,100... . 16.8 1,800 2,200... 17.6 1,900 SOLUTION CONTAINED 200 PARTS PER MILLION P0 4 Soil No. PO* added to 100 gm. soil. 2,400 36.0 22.4 12.4 4.0 17.2 7.2 25.6 12.8 2,800 2,900 22.4 20.8 12.4 5.2 10.0 5.6 18.4 2,600 16.4 SOLUTION CONTAINED 240 PARTS PER MILLION PO<. • 3,100 39.2 76.0 4.0 4.8 8.4 8.0 24.8 36.8 3,500 16.8 5.2 5.6 18.4 3,300 SOLUTION CONTAINED 240 PARTS PER MILLION PO*. 3,700 28.0 10.0 12.0 14.0 3,900 24.0 4.0 6.4 13.6 Summary of above table. P0 4 fixed by 100 em. soil. Per cent of PO< fixed. 292 448 428 474 Oram. 0. 8308 .8308 .8308 Gram. 0. 7190 .8043 .7966 .7516 86.4 96.7 95.8 90.4 9 The series reported in the above table was started in glass tubes, 100 grams of soil being used in each instance, but it was found neces- sary to transfer the soils to funnels, as there was no percolation at all through soil No. 474, and it was extremely slow in Nos. 292, 448, and 428. The extracts all came through clear for about one month, after which they began coming through cloudy, and when the series was stopped the percolation was very slow even in the funnels. Phosphoric acid being the constituent of phosphates which forms insoluble compounds with the bases always present in soils, such as iron, aluminum, titanium, lime, and magnesium, it is not very difficult to understand the retention of soluble phosphoric acid by soils. In the presence of sufficient calcium carbonate the application of soluble phosphoric acid will result in a " reversion" of the phosphate, i. e., the formation of the less soluble dicalcium phosphate which, however, is quite readily available, and hence there results a gain rather than a loss. But in case the soil is deficient in lime and contains an excess of iron and aluminum hydrates and silicates, similar to Hawaiian soils, an entirely different problem is encountered. In this case the phosphoric acid will be fixed by the iron and aluminum compounds, thus being rendered not only practically insoluble in water, but also in weak organic acid solvents. For such conditions various investi- gators recommend the application of lime preceding that of the super- phosphate, the theory being that the lime will revert the phosphoric acid. This theory has been put in practice in the red clay soils of the Wahiawa district of Oahu, but has failed to produce any bene- ficial results. This is probably due to the excessive amounts of iron and aluminum hydrates in these soils. As indicated in the preceding tables, there is considerable difference in the absorption of the potassium and calcium phosphates. Since they were not carried to the saturation point, we can only compare the rates of absorption, and here the fixation of calcium phosphate is strikingly faster. It will be seen that more phosphoric acid was fixed from calcium phosphate in two of the soils and practically the same in the other two, even though 1 liter more of the potassium phosphate solution was passed through. On the other hand, nearly the same weight of the salt has passed through, and the general prop- erty of absorption is similar. In both cases soil No. 292 fixed the least phosphoric acid, No. 474 next least, No. 428 next, and No. 448 the most. Both of the soils that fixed the least phosphoric acid contained a high percentage of phosphoric acid, a sufficiency of lime, and a high percent- age of organic matter. It is probable that reversion takes place more quickly with the calcium salt, which accounts for the higher rate of fixation in this case. There appears to be little correlation between the rate of fixation and the mechanical composition of the soil in cases where the size of the particles is offset by the organic matter, 48303°— 14— 2* 10 the highest and the lowest in fixing power being both sandy soils but differing in organic-matter content. The fact that the fixation of phosphoric acid from the calcium salt was not excessively greater than that from the potassium salt was probably due to the fixation being largely a result of the action of iron and aluminum compounds and only a partial reversion of the calcium salt. Crawley * found that upon irrigating Hawaiian soils immediately after application of water-soluble phosphate one-half of the phosphoric acid remained in the first inch of soil, nine-tenths in 3 inches, and practically all in 6 inches of the surface soil. These results indicate the absolute neces- sity of turning all applications of phosphate under by deep plowing in order to get the best results. Otherwise the rain is not able to wash it down to the roots, and consequently the dissemination of this fertilizer is incomplete. At the point where these series were stopped the soils had appar- ently lost none of their fixing power. This fact lends very strong proof to the theory that the concentration of the soil solution with regard to phosphoric acid is not increased by the addition of this element in moderate quantities either as a soluble or insoluble salt; also, that while there are differences in the concentration of the solu- 'tion in different soils, they are due to factors other than the solubility of the salt in water. ABSORPTION OF POTASH. For the study of the absorption of potash a solution of potassium sulphate, containing about 200 parts per million of potassium (K) was used. The soils were the same as used in the phosphate series, and the method of percolation was through a column of 100 grams of the soil placed in glass tubes, as already described. At the outset the solution percolated quite rapidly, but after five days much more slowly in soils Nos. 292 and 428, and extremely slowly in soil No. 448. A precipitate, apparently of ferric hydrate, formed upon stand- ing overnight in the extract from soil No. 292. After about one month the percolation from soil No. 448 (yellow clay soil) became so slow as to be several hundred cubic centimeters behind the rest of the series. However, strange to say, about one week following the date of above conditions, the percolation in soil No. 448 was faster than with the other soils, and when the experiments were stopped soil No. 474 was percolating the most slowly of all. In order to get a clear conception regarding the absorption of pot- ash, it is of some value to know the effect of leaching the soils with water upon the solubility of this element. The table following throws some light upon this. i Jour. Amer. Chem. Soc, 24 (1902), p. 1114. 11 Potash removed from the soils by distilled water. [Expressed in parts per million of K in the percolate. 1 Percolates of 100 cc. each. SoU No. 292. SoU No. 448. Soil No. 428. Soil No. 474. 100 52 44 20 28 8 20 44 44 28 16 16 108 68 52 56 44 200 300 44 40 20 400 500 Thus it is shown that the general tendency of the soils was to yield a solution of fairly constant concentration. However, attention should be called to the fact that these figures do not represent parts per million in the soil, but simply in the solution obtained through percolation. The following table shows the absorbing power of the soils for potash, using a solution containing 214 parts per million of potassium sulphate. Absorption of potash from a solution of KzSO^. [Expressed in parts per mnlion of K in the percolate.] Percolates of of 100 cc. each. 100.. 200.. 300.. 400.. 500.. 600.. 700.. 800.. 900.. 1,000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 Soil No. Soil No. Soil No. Soil No. 292. 448. 428. 474. 60 52 48 100 52 92 56 80 40 80 40 76 64 100 52 84 76 140 124 104 56 148 152 88 60 160 156 96 72 164 188 84 76 188 192 88 76 168 192 76 64 168 212 72 84 196 192 84 136 208 200 84 96 204 204 104 120 172 200 116 128 160 204 140 124 160 196 160 Percolates of of 100 cc. each. 1,800. 1,900 2,000 2,100 2,200 2,300 2,400 2,500 2,700 2,900 3,100 3,300 3,500 3,700 3,900 4,100 4,300 Soil No. Soil No. Soil No. 292. 448. 428. 140 164 184 132 148 188 128 164 192 120 188 180 100 172 184 148 172 188 132 200 172 116 200 180 136 200 200 152 204 216 152 224 224 184 212 232 152 220 216 160 204 224 148 216 204 164 228 200 164 220 228 Soil No, 474. 172 160 176 168 156 180 156 168 188 168 184 204 208 212 168 200 212 Summary of above table. SoU No. K added tolOOgm. soU. K fixed by 100 gm. soil. Per cent ofK fixed. 292 Gram. 0.9030 .9030 .9030 .9030 Gram. 0. 4030 .1496 .2380 .2782 45 448 ; 17 428 26 474 31 In order more easily to explain the absorption of potash by soils it is of considerable importance to know the effect of the addition of potash upon the solubility of the other bases commonly occurring in soils. For this reason several determinations were made to ascer- tain the concentration of lime and magnesia in the filtrate. The table followiug gives the results of these determinations. 12 Effect of the potassium sulphate solution upon the solubility of lime and magnesia in the soils. [Expressed in parts per million in the percolate.] Percolates of 100 cc. each. Lime. Magnesia. Soil No. 292. Soil No. 448. SoU No. 428. SoU No. 474. SoU No. 292. SoU No. 448. SoU No. 428. SoU No. 474. 100 104 56 66 50 68 36 26 44 28 22 20 36 24 14 40 10 24 24 24 12 8' 514 146 150 158 164 70 48 102 70 94 72 68 32 54 24 34 32 32 26 24 26 34 28 26 18 22 22 34 82 300 46 500 40 700 38 900 38 2,700 24 3,300 34 The data presented in the preceding tables throw considerable light upon the retaining power which Hawaiian soils possess for potash. In the absorption of potash the salts undergo a decomposition, the result of which is a replacement of calcium or magnesium by potassium. The two former elements combine with the acid constituent of the potash salt and pass off in the drainage water. It has been found that potassium sulphate is more firmly fixed than the chlorid. In general the reaction taking place is a replacement of the calcium in the zeolitic silicates, but humus and the iron and aluminum hydrates also fix potash to a certain extent. It may be seen from the above tables that the soil highest in lime and magnesia had the highest fixing power for potash, and the other three soils in proportion. This is in agreement with the findings of other investigators. Crawley * found that Hawaiian soils fixed potash quite firmly, but the fixation was not nearly so lasting as that of phosphoric acid. The results given herewith indicate this to be true and also the saturation point for potash to be far below that of phosphoric acid, even in the soils high in lime and magnesia. In the preceding table there are some very striking results showing the decrease in concentration of lime and magnesia in the filtrate, with decrease in amount of potash fixed by the soil. The fixation of this element in the soils highest in lime and magnesia is almost constant for the first liter of solution passing through the soil column. On the other hand, the fixing power of the other soils decreases more rapidly and they are more easily saturated, while the soil containing 8 per cent of magnesia had not reached a state of saturation at the close of the experiments. ABSORPTION OF NITROGEN. AMMONIUM SULPHATE. This series was carried out in a manner similar to the previous one — namely, 100 grams of soil was placed in glass tubes, with percola- tion at the rate of 100 cubic centimeters per 24 hours. The percolate 1 Jour. Amer. Chem. Soc, 25 (1903), p. 47. 13 remained clear through the series, except for a flocculent precipitate which appeared to be ferric hydrate, and which was deposited from soil No. 428. The following table shows the amount of ammonia nitrogen removed from the original soils by distilled water: Ammonia nitrogen removed from the soils by distilled water. [Expressed in parts per million nitrogen in the percolate.] Percolates of 100 cc. each. SoU No. 292. SoU No. 448. Soil .No. 428. Soil No. 474. 100 6.5 5.7 2.2 2.9 11.4 8.4 5.7 5.7 5.6 13.4 8.8 5.4 6.4 7.3 4.2 4.4 2.3 3.0 5.1 200 300 400 500 From these data it may be seen that these soils possess the same general tendency to produce a solution of constant nitrogen content. In the following table may be observed the absorbing power of the soils for nitrogen in ammonium sulphate: Absorption of nitrogen from a solution of (NH 4 ) 2 S0 4 . [Expressed in parts per million nitrogen in the percolate.] SOLUTION USED CONTAINED 171 PARTS PER MILLION NITROGEN. Percolates of 100 cc. each. Soil No. 292. SoU No. 448. SoU No. 428. SoU No. 474. 100 3.6 2.6 3.6 5.6 12.8 14.7 22.8 15.8 17.1 18.7 46.8 64.4 36.8 39.6 36.8 51.5 51.5 39.6 36.8 42.8 39.6 64.4 34.2 39,6 39.6 51.5 44.8 36.8 36.8 39.6 2.6 2.6 8.6 7.4 2.6 4.5 5.4 12.1 .12.8 15.8 200 300 400 500 600 700 800 900 1,000 Percolates of 100 cc. each. 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 1,900 Soil No. SoU No. SoU No. 292. 448. 428. 21.4 42.8 39.6 51.5 51.5 51.5 51.5 46.8 51.5 51.5 57.2 64.4 51.5 57.2 62.9 68.4 73.6 68.4 64.4 68.4 75.2 64.4 87.1 87.1 86.0 94.4 86 Soil No. 474. 17.1 46.8 51.5 46.8 51.5 57.2 64.4 73.6 80.8 SOLUTION USED CONTAINED 168 PARTS PER MILLION NITROGEN. 2,000 2,100 2,200 2,300 2,400 2,500 2,700 2,900 70.8 73.6 78.8 73.6 76.4 91.6 73.6 73.6 123.6 128 117.9 96.6 105.2 93.3 117.1 73.2 114.1 128.8 128.8 128.8 121.2 121.2 128.8 121.2 117 156.6 174.2 143 156.4 158.4 167.8 140.7 3,100 3,300 3,500 3,700 3,900 4,100 4,300 4,500 119 126 134 117.6 135.2 156.8 186 163 138 156.5 148.9 115.6 152.4 149.9 88.8 152 152 120.8 147.2 137.6 120 164.8 171.2 164.8 112 124.8 148 139.8 137.6 141.6 120 164.8 Summary of above table. Soil No. Nitrogen added to 100 gm. soil. Nitrogen fixed by 100 gm. soU. Per cent of nitro- gen fixed. 2y2 Gram. 0.6811 .6811 .6811 .6811 Gram. 0. 2782 .2290 .2753 .3015 41 448... 34 428 40 474 44 14 The nature of the reaction accompanying the absorption of am- monium compounds is very similar to that of potash salts ; namely, the replacing of calcium in humus, double silicates, and in some cases calcium carbonate. Hence the application of ammonium salts as fertilizer tends to deplete the soil of its basic constituents. It may be seen from a comparison of the preceding tables that the fixation of nitrogen is far in excess of that of potash in every instance except soil No. 292, which is the highest in magnesia content. The fixing power of the four soils in the series agrees more closely than in the potash series, but in each instance the clay soil fixed the least. Attention is called to soils Nos. 428, 448, and 474, which absorb much more nitrogen than potash. In case of two of the soils (428 and 474) this may be accounted for by the high content of organic matter. In the last two, fractions of percolate nitrates and nitrites were de- termined and both were found to be present in one case to the extent of 14.4 parts per million N as N0 3 and 3.1 parts per million N as N0 2 . This indicates the rate at which nitrification was going on at the close of the experiments. As in the potash series, the highly basic soils fixed much more nitrogen at the beginning of the experiments and a much larger total amount than the less basic. On the other hand, the decrease in fixing power was much slower and more gradual in the other soils. SODIUM NITRATE. Of the salts commonly used as fertilizing materials all are strongly fixed by the soil except nitrates. However, nature has made a wise provision for retaining nitrogen in an insoluble form, which becomes slowly available for growing plants. Determinations of the amount of nitrate nitrogen removed from the original soils gave the following results : Nitrate nitrogen removed from the soils by distilled water. [Expressed in parts per million nitrogen in the percolate.] Percolates of 100 cc. each. Soil No. 292. Soil No. 448. Soil No. 428. Soil No. 474. 100 4.2 2.4 .0 8.6 .0 .0 5.9 .0 .0 106 200 2 300 .4 These data indicate a condition found to be true in all soils, namely, the readiness with which nitrates are leached from the soil by rains. Soil No. 474 is a very porous, floury soil, containing a high percentage of organic matter, and under the existing climatic conditions would be expected to have a high nitrate content. 15 The following table shows the absorbing power of these soils for nitrate nitrogen, using a solution of sodium nitrate which contained 250 parts per million of nitrogen: Absorption of nitrogen from a solution of NaN0 3 . [Expressed in parts per million of nitrogen in the percolate.] Percolates of 100 cc. each. SoU No. 292. SoU No. 448. Soil No. 428. Soil No. 474. Percolates of 100 cc. each. SoU No. 292. Soil No. 448. SoU No. 428. Soil No. 474. 100 147 184 215 245 240 225 205 230 230 157 162 190 240 245 220 205 240 225 142 180 180 205 225 220 215 215 225 290 170 200 235 235 200 195 220 175 1,000 240 240 230 240 245 250 250 250 250 225 230 235 235 240 245 250 250 250 230 230 235 240 240 250 250 250 195 200... 1,100 185 300... 1,200 215 400... 1,300 215 500... 1,400 215 COO... 1,500 220 700 1,600 225 800 1,700 900 1,800 Summary of above table. Soil No. Nitrogen added to 100 gm. sorl. Nitrogen fixed by 100 gm. soil. Per cent of nitro- gen fixed. 292 Gram. 0.4500 .4500 .4250 . 4000 Gram. 0.0384 .0456 .0518 .0610 8.5 10 448 428 12 474 15 The above table presents some very interesting data. It is quite generally conceded that soils have no fixing power for nitrates and for this reason it is difficult to explain the action of soil No. 474 toward this salt. The percolation was very slow in this instance and the rate decreased to such an extent that the series had to be stopped after 1,600 cubic centimeters had passed through, as the solution would no longer filter through the column. This condition exists in spite of the fact that the soil contained only an extremely small percentage of clay. Soil No. 428 acted somewhat similarly, but percolation did not stop completely as in the case of No. 474. This condition is undoubtedly brought about by the action of sodium nitrate upon the organic matter, as both of these soils were high in this constituent. Soil No. 474 was apparently still fixing nitrogen at the close of the experiment, as in no case except with the first 100 cubic centimeters did the percolate reach a concentration of 250 parts per million. These figures indicate that while soils are unable to retain nitrates against the action of nitrate-free water, they are able to retain limited amounts against the action of water with a high nitrate content. It is possible that considerable denitrification took place in soil No. 474. The sluggish movement of the solution through this soil indicates the existence of just the conditions which are conducive tp denitrification. The same is true of No. 428. 16 Denrtrification refers, of course, to any transformation which nitrates may undergo, such as its conversion into nitrate, ammonia, free nitro- gen, or protein. ABSORPTION OF FERTILIZER SALTS BY FRESH AND AIR-DRIED SOILS. The type of soil occurring in greatest abundance on the islands is a highly ferruginous red clay (No. 517). For this reason it was de- cided to make a series of percolations using both soil and subsoil of this type in the fresh and air-dry condition, using sodium nitrate, ammonium sulphate, potassium phosphate, and calcium phosphate. The fresh soil contained 19.7 per cent moisture; the fresh subsoil, 24.4 per cent moisture. The method employed was essentially the same as that used in the previous series except that it was found to be necessary to use only 50 grams of soil with the phosphates in order to effect a passage of the solution through the soil column. Also the concentration of the solu- tion was increased in an attempt to saturate the soil with phosphates. Determinations were made of the solubility in distilled water of the phosphate in the saturated soil, and it was found to be negligible. On passing distilled water through a column of 50 grams of soil and determining the percentage of phosphoric acid in each 100 cubic centimeters passing through, only a faint trace was detected. ABSORPTION OF PHOSPHORIC ACID. The following table shows the absorbing power of the red clay soil for phosphoric acid when applied as monopotassium phosphate: Absorption of phosphoric acid from a solution of KH 2 PO^. [Expressed in parts per million of PO4 in the percolate.] P0 4 IN SOLUTION, 800 PARTS PER MILLION. Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 44 38 124 72 128 165 Trace. 21 22 Trace. 27 29 400 180 220 290 260 340 340 24 32 200 500 300 600 P0 4 IN SOLUTION, 1,400 PARTS PER MILLION. 500.. 600.. 700.. 800.. 900.. 1,000. 1,100. 1,200. 410 390 400 430 530 620 460 460 500 500 400 560 150 325 350 560 675 825 290 325 360 665 675 825 1,300. 1,400. 1,500. 1,600. 1,700. 1,800. 1,900. 2,000. 950 900 750 750 725 875 675 875 850 825 750 750 600 850 675 825 P0 4 IN SOLUTION, 1,025 PARTS PER MILLION. 1,500 700 675 600 700 675 600 2,800 950 950 1,750 3,250.... 600 600 2,250 3,300... 1,025 1,025 2.500 750 675 675 675 3,800 4,300 950 1,025 950 1,025 2,725 17 Summary of preceding table. Sou. P0 4 added to 100 gm. soil. PO, fixed by 100 gm. soil. Per cent 01PO4 fixed. Fresh soil Fresh subsoil.. Air-dry soil Air-dry subsoil Grams. 9. 5950 9. 5950 6.8350 6.8350 Grams. 3.8062 3. 85 1 1 2. 7372 2. 6820 39.6 40.2 40.1 39.3 The absorption of phosphoric acid from monocalcium phosphate was as follows : Absorption of phosphoric acid from a solution of Cal^ (PO i ) 2 ' [Expressed in parts per million of PO< in the percolate.] SOLUTION CONTAINED 1,300 PARTS PER MILLION PO4. Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 210 470 615 210 490 585 203 750 625 203 700 800 400 700 1,012 925 650 1,012 1,200 700 703 200 500 300 600 SOLUTION CONTAINED 1,700 PARTS PER MILLION P0 4 . 500. 775 850 1,250 1,000 1,100 1,025 850 1,300 775 850 1,100 1,000 1,250 1,000 975 1,275 1,300 1,400 1,500 1,600 1,700 1,800 1,900 1J00 1,200 1,350 1,350 1,350 1,250 925 1,100 1,150 1,400 1,250 1,250 1,325 925 600 700 1,300 825 900 1,100 1,100 950 1,275 950 950 1,100 1,100 950 800 900 1,000 1,100 1,200 SOLUTION CONTAINED 2,812 PARTS PER MILLION PO4. 1,700 1,300 1,250 1,250 2, 750 3,000 1,250 1,925 2,812 2,812 1,950 1,300 3,250 3,500 1,600 1,550 2,000 1,350 1,350 2,812 2,812 2,450 1,600 3,800 4,300 2,812 2,812 2,812 2,812 2,500 2,725 1,450 1,350 1,400 1,600 Summary of above table. Soil. PO4 add- ed to 100 gm. soil. PO< fixed by 100 gm. soil. Per cent of PO4 fixed. Grams. 8. 3328 8.3328 6.9416 6.9416 Grams. 5.9110 5. 9880 5. 5232 5. 4732 70.9 71.8 79.6 78.8 18 The results in the above tables can be compared with those of the previous series only relatively, due to the fact that the solution in this case was so much more concentrated. They indicate the prac- tical impossibility of saturating Hawaiian soils with phosphoric acid or adding an excess in a practical way. It will be noted that this type of soil is able to absorb nearly 4 per cent of its weight of phos- phoric acid (P0 4 ) in the fresh soil and nearly 3 per cent in the air-dry soil from the potassium salt;, also, that from the calcium salt the soil absorbed nearly 6 per cent of its own weight of phosphoric acid in the fresh soil and 5.5 per cent in the air-dry soil. It is difficult to explain the higher absorptive power of the fresh soil over the air dry, but it is probably due to the physical properties, and is related to the soil films. This soil is composed of very fine particles, exposing relatively enormous surface to the action of the soil solution or any added salt solution. In the fresh soils of this type these particles are in a high state of deflocculation and the effect of drying in the air tends to flocculate them to a great extent, thereby reducing the area of the exposed surface. Drying would also tend to modify the film sur- rounding each particle. Even with only 50 grams of soil it was found impossible, due to the strong deflocculating action of the phosphate salts, to make the percolations in tubes, but funnels had to be used. The samples previously dried in the air percolated more slowly than the fresh soils. This is probably due to the fact that the soil swelled more in the tube after the addition of the solu- tion, thus packing more closely and closing up the pore spaces. There was apparently very little difference between the absorbing power of the soil and subsoil, but considerable variation between the fresh and air-dry soils. The rate of fixation in the early part of the experiment was considerably faster in the latter than in the former, and hence the air-dry soils were more quickly saturated by the salts. Another interesting fact is the difference in the absorp- tive power of this type of soil for phosphoric acid in the two forms. The data are sufficient to justify the statement that this difference is due to the reversion of the calcium salt, although due also in great part to the state of the iron and aluminum compounds which exist in this type of soil. The absorption from the potash salt was more complete at the first application, but thereafter decreased quite rapidly and regularly. It should also be noted that at the outset the air-dry soil absorbed the potash salt more completely than the fresh soil. This is thought to be due to the partial elimination of the film surrounding the soil particles, thus allowing the solution to penetrate more thoroughly. 19 ABSORPTION OF POTASH. The strength of solution used in the potash series was the same as in the first series. One hundred-gram portions of soil were used. The results of extraction of the original soils are given in the follow- ing tables: Removal of potash from soil by distilled water. [Expressed in parts per million of K ih the percolate.] Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 40 17.4 56 32 14.8 32 64 52 48 32 36 32 400 52 68 32 84 16 27 23 200 500 300 600 The results of determinations of the absorption of potash from potassium sulphate are given in the following table : Absorption of potash from a solution containing 204 parts per million Kfrom K 2 S0 4 . [Expressed in parts per million in the percolate.] Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. Percolates of , 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 64 68 212 216 164 172 192 180 184 52 44 136 160 168 180 180 180 196 52 80 104 180 180 180 172 184 192 68 24 60 120 120 120 140 180 180 1,000 1,100 1 ,200. . . 172 152 176 152 180 180 196 192 172 196 204 188 212 176 200 192 300 184 400 1,300. 200 500 1,400 1,500 1,600 1,700 | 1,800. 200 200 216 20S 200 192 212 204 180 600 176 700 188 800 192 900 200 1 Summary of above table. Sofl. K added to 100 gm. soil. K fixed by 100 gm. soil. Per cent ofK fixed. Fresh soil Grams. 0. 34G8 .3468 .3672 .3672 Grams. 0.0468 .0636 .0528 .0972 13.5 Fresh subsoil 18.3 Air-dry soil 14.4 Air-dry subsoil 26.5 The effect of the potassium sulphate solution on the solubility of lime and magnesia is shown in the following table : Effect of potassium sulphate solution upon the solubility of lime and magnesia. [Expressed in parts per million in the percolate.] Percolates of 100 cc. each. Lime (CaO). Magnesia (MgO). Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 300 600 1,700 60 50 44 u 38 64 34 16 62 38 62 20 50 54 52 24 34 56 20 18 28 56 20 12 34 18 20 14 30 20 20 12 20 These tables indicate that the potash in this type of soil is quite soluble. The fixing power of this soil is far below that of the four soils used in the previous series; that is, the red clay soil of the islands is more easily saturated with potash than the other types. This is partly due to the low lime and magnesia content of this soil. The two series illustrate quite well the effect of these bases upon the fixation of potash. The figures in the table on page 19 indicate the subsoil to have the power of fixing more potash than the soil, and that drying in the air tends to increase this power. ABSORPTION OF NITROGEN. AMMONIUM SULPHATE. This series was carried through similarly to the previous ammonium sulphate series. A table showing the solubility in distilled water of the ammonia nitrogen in the original soil is given herewith: Ammonia nitrogen removed from the soil by distilled water. [Expressed in parts per million nitrogen in the percolate.] Percolates of 100 cc. each. Fresh soU. Fresh subsoil. Air-dry soU. Air-dry subsoU. 100 5.1 Trace. Trace. Trace. Trace. Trace 7.47 11.16 Trace. 5.04 200 6.1 300 7.2 This type of soil is shown to contain only small amounts of ammonia nitrogen soluble in water, the amounts being slightly lower than those found in the previous series. The following table shows the absorbing power of this soil for ammonium nitrogen: Absorption of nitrogen from a solution of (NH 4 ) 2 S0 4 . [Expressed in parts per mUlion in the percolate.] SOLUTION CONTAINED 246 PARTS PER MILLION NITROGEN FROM (NH^SO*. Percolates of 100 cc. each. Fresh soU. Fresh subsoU. Air-dry soU. Air-dry subsoU. Percolates of 100 cc. each. Fresh soU. Fresh subsoU. Air-dry soU. Air-dry subsoU. 100 26.5 65.2 71.6 185 181.3 211.5 17.8 54.9 66.6 143 183.3 167.4 12.5 113.2 178.2 162.3 165.1 172 25.2 111 145.6 149. 6 168.9 160 700 151.3 192.9 178.6 239 224 157.1 178.6 152.3 204 242 188 180 206 188 224 172 200 800 172 300 900 184 400 1,000 184 500 1,100.... 214 600 SOLUTION CONTAINED 204 PARTS PER MILLION NITROGEN FROM (NH 4 ) 2 SO*. 1 200 181.4 211.6 182.6 200 224 206 214 206 1,400 212 212 206 206 1,300 21 Summary of preceding table. Soil. Fresh soil Fresh subsoil . . Air-dry soil Air-dry subsoil Nitrogen added to 100 gm. soil. Gram. 0.3318 .3318 .2706 .2706 Nitrogen fixed Dy 100 gin. soil. Gram. 0.1000 .1164 .0916 .1019 Per cent of nitrogen fixed. 30.1 35 33.9 37.6 Since ammonium salts are retained by the soil in most respects by the same reactions which govern the absorption of potash, we would expect the red clay soil to have the low absorptive power shown in the above table, which is less than one-half that of the soils used in the previous series. The subsoil showed a slightly higher fixing power than the soil, while the effect of drying in the air was to reduce the fixing power. This latter rinding is just the reverse of that obtained in case of potash. SODIUM NITRATE. The absorbing power of this soil for sodium nitrate is very much lower than that of the other types, as may be seen from the following tables : Removal of nitrate nitrogen from soil by distilled water. [Expressed in parts per million of nitrogen in the percolate.] Percolates of 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry subsoil. 100 19.2 8.8 12.8 7.2 200 1 Absorption of nitrogen from a solution of 250 parts per million nitrogen from NaNO z . [Expressed in parts per million of nitrogen in the filtrate.] Percolates of ! 100 cc. each. Fresh soil. Fresh subsoil. Air-dry soil. Air-dry 1 subsoil. Percolates of 100 cc. each. Fresh sofl. Fresh subsoil. Air-dry ! Air-dry soil. subsoil. 100 187.5 250 180.0 255 215.0 240 215.0 240 300 250.0 250 250.0 250 250.0 250.0 200 1 400 2.50 250 Summary of above table. Soil. Nitrogen Nitrogen added to fixed bv 100 gm. 100 gm". soil. soil. Per cent of nitrogen fixed. Gram. Gram. Fresh soil 0. 1000 0. 0062 Fresh subsoil 1000 . 0065 Air-dry soil 1000 . 0045 Air-dry subsoil 1000 . 0045 6.2 6.5 4.5 4.5 22 The above results show the low fixing power of this type of soil for nitrates. This fact strongly indicates the r61e of organic matter in the absorption of this salt. The organic matter content of the previous series of soils was much higher than that of the red clay. There was apparently no difference between the fixing power of the soil and the subsoil, but it was stronger in the fresh than in the air- dried samples. ABSORPTION OF FERTILIZER SALTS WHEN APPLIED IN MIX- TURES, AND THE EFFECT OF HEAT AND ANTISEPTICS. A third series of experiments was made with the idea in mind of applying a solution containing a mixture of fertilizer salts and at the same time determining the effect of heat and volatile antiseptics upon the absorbing power. The soils chosen for this series were No. 428, a highly organic soil used in the first series, and No. 517, the red clay soil used in the second series. Three fertilizer mixtures were used and applied to the soil in series of three, namely, untreated, heated (230° C. in air bath), and partially sterilized (5 cubic centi- meters chloroform to 100 grams soil kept in a closed fruit jar 48 hours, then spread out in the air 24 hours before placing in the glass tubes). The mixtures were as follows: (1) ammonium sulphate, potassium phosphate, and potassium sulphate; (2) ammonium sulphate, calcium phosphate, and potassium sulphate; and (3) sodium nitrate, calcium phosphate, and potassium sulphate. The solutions were allowed to percolate through the soil at the rate of 100 cubic centimeters in 24 hours, and the percolates were analyzed. ABSORPTION OF PHOSPHORIC ACID. The table following shows the fixing power of these soils for phos- phoric acid when applied in mixtures. 23 Absorption of calcium and potassium phosphate in solutions of fertilizer mixtures. [Expressed in parts per million of PO< in the percolate.] Percolates of 100 cc. 100.. 200.. 300.. 500.. 700.. 900.. 1,100 1,300. 1,500. 1,700. 1,900. 2,100. Soil No. 517. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Untreat- ed. Heated, j Trace. 46 86 34 140 22 200 70 480 360 168 Trace. 480 420 480 460 540 540 480 500 580 640 560 560 form. 26 34 100 240 460 200 540 540 680 500 640 660 Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Sodium nitrate, calcium phosphate, and potas- sium sulphate. Untreat- ed. Heated. 26 38 50 120 392 512 550 650 1,700 1,550 1,600 1,700 1,750 1,750 1,600 1,650 1,700 1,750 1,500 1,800 1,950 1,950 1,950 2,000 Chloro- form. Untreat ed. 44 380 448 i 700 1,750 : 1,650 2,000 1,850 1,750 1,700 2,000 2,000 224 360 232 600 1,050 1,400 1,400 1,350 1,250 1,500 Heated. 50 56 328 550 1,050 1,350 1,050 1,100 1,500 1,150 Chloro- form. 280 400 256 750 1,050 1,350 1,450 1,400 1,550 1,150 SUMMARY. P0 4 added to 100 grams soil, grams. POi fixed by 100 grams soil, grams. Per cent of PO* fixed 1.5750 1.5750 1.5750 4.3050 4.3050 4.3050 2.8050 2.8050 .7588 48.2 .8548 54.3 .6670 42.3 1. 3982 32.5 1.2780 29.7 1.0678 24.7 1.6134 57.3 1.2116 43.2 2.8050 1.5014 53. 5 Soil No. 428. Percolates of 100 cc. 100.. 200.. 300.. 500.. 700.. 900.. 1,100 1,300 1,500 1,700 1,900 2,100 Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Untreat- ed. Trace. 16 16 12 33 19 19 18 7 11 22 31 20 13 13 12 16 9 90 236 236 264 240 320 Chloro- form. Trace. 20 20 12 15 14 19 20 6 21 21 16 Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Untreat- ed. Heated. 19 15 16 15 16 15 11 10 16 8 21 36 6 6 8 19 6 15 9 33 17 36 13 70 Chloro- form. Trace. 15 15 11 15 Sodium nitrate, calcium phosphate, and potas- sium sulphate. Untreat- ed. Heated. Chloro- form. SUMMARY. P0 4 added to 100 grams soil, grams. P0 4 fixed bv 100 1.4700 1.4700 1.4700 0. 8265 0.8265 0.8265 0.6375 0.6375 0.6375 grams soil, grams. Per cent of PO< fixed 1.4296 1. 1104 1.4298 .7995 . 7754 .8033 .61.78 .6191 .6161 97.1 75.5 97.1 9R.8 93.9 97.3 96.9 97.3 96.7 24 The solution used in the first series of three, columns 1, 2, and 3 contained 750 parts per million P0 4 from potassium phosphate; 4, 5, and 6, 2,050 parts per million P0 4 from calcium phosphate; 7, 8, and 9, 1,650 parts per million P0 4 from calcium phosphate; 10, 11, and 12, 700 parts per million P0 4 from potassium phosphate; 13, 14, and 15, 435 parts per million P0 4 from calcium phosphate; 16, 17, and 18, 425 parts per million P0 4 from calcium phosphate. The solution used with soil No. 428 was made up to a much weaker strength for the reason that it would be more comparable with the results obtained on this soil given in the first series. The absorbing power of the red clay soil was appreciably less for phosphates in mixtures, but that of the highly organic soil is very much the same, regardless of method of application. The effect of heat or antiseptics was not striking, but in most instances caused a decrease in the fixing power. In one instance, namely, with the highly organic soil, the heat caused a decided decrease in fixing power. ABSORPTION OP POTASH. The results obtained with the application of potash in mixtures are shown in the following table : Absorption of potash from a solution of fertilizer mixtures. [Expressed in parts per million K in the percolate.] Percolates 100 cc. each. SoU No. 517 Ammonium sulphate, po- Ammonium sulphate, cal- tassium phosphate, and I cium phosphate, and potassium sulphate. potassium sulphate. Untreat- ed. Heated. Chloro- form. Untreat- ed. Heated, Chloro- form. Sodium nitrate, calcium phosphate, and potas- sium sulphate. Untreat- ed. Heated. Chloro- form. 100.. 200.. 300.. 500.. 700.. 900.. 1,100. 1,300. 1,500. 1,700. 1,900. 2,100. 196 348 440 484 544 540 540 564 544 544 488 428 180 316 440 460 508 580 584 576 560 576 564 528 268 300 376 480 552 524 572 588 504 528 544 496 172 152 44 164 124 184 156 152 196 164 120 132 80 84 224 260 216 288 224 244 188 192 156 88 72 64 228 276 200 204 176 140 216 184 108 104 176 84 216 204 180 256 228 156 284 212 112 132 104 120 232 228 188 180 236 360 144 108 68 108 140 208 232 132 172 280 288 248 SUMMARY. K added to 100 grams soil grams.. K fixed by 100 grams soil grams. 1.0038 1.0038 1.0038 0. 3570 0. 3570 0. 3570 0.4536 0. 4536 0. 4536 Per cent of K fixed . 25 Absorption of potash from a solution of fertilizer mixtures — Continued. Soil No. 428. Percolates of 100 cc. each. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Sodium nitrate, calcium phosphate, and potas- sium sulphate. Untreat- ed. Heated. Chloro- form. Untreat- ed. Heated. Chloro- form. Untreat- ed. Heated. 268 192 224 240 • 228 220 228 200 292 Chloro- form. 100 96 168 432 432 520 516 552 572 608 620 580 620 600 524 112 192 292 400 428 432 412 424 408 460 448 448 96 104 144 212 216 184 224 212 232 212 228 224 272 192 188 220 268 224 224 180 200 220 212 236 118 104 140 192 248 196 216 148 76 112 192 204 192 188 104 200 72 300 500 700 900 1,100 300 360 416 420 368 416 416 112 180 188 216 188 1,300 1,500 180 188 192 240 192 252 1,700 456 424 432 224 200 236 1,900 2,100 SUMMARY. K added to lOOgrams soil grams. . K fixed bv lOOgrams 1.2264 . 9468 77.6 1. 2264 1. 2264 1. 0224 83.6 : 0.3822 0. 3822 0. 3822 0.3090 0.3090 0.3090 Per cent of K fixed. 1 The above table presents some striking results, and indicates that Hawaiian soils possess a very low fixing power for potash when applied with phosphates, especially calcium phosphate. In every instance, except two, the amount of potash found in the filtrate was greater than the weight added to the soil. This is undoubtedly due partly to a replacement of the potash by lime. The effect of heat in case of the highly organic soil was to considerably reduce the fixing power, but chloroform reduced it only slightly. With the red clay soil there was very little variation, due to sterilization either with heat or antiseptics. This was contrary to the results obtained when potash was used alone. Drying in the air increased the fixing power. The solutions used on samples reported in columns 1, 2, and 3 contained 478 parts per million K from K 2 S0 4 ; 4, 5, and 6, 170 parts per million; 7, 8, and 9, 216 parts per million; 10, 11, and 12, 584 parts per million; 13, 14, and 15, 182 parts per million; 16, 17, and 18, 206 parts per million. 26 ABSORPTION OF NITROGEN. AMMONIUM SULPHATE. The following table shows the results obtained by the application of ammonium sulphate in mixtures: Absorption of nitrogen from a solution of ammonium sulphate in a mixed fertilizer. [Expressed in parts per million nitrogen in the percolate.] Soil No. 517. Percolates of 100 cc. each. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. 100 81,5 128.8 99 130 131 137 149 178 178 172 188.4 148.6 138 133 130 174 141 166 178 172 88.9 64.9 125.4 135 148 168 156 153 175 159 172 133.9 167.2 114 143 128 159 151 170 159 172 91.3 200 168.7 30 111 128 129 136 151 178 178 172 114 500 153 700 136 900 156 1,100 164 1,300 163 1,500 172 1,700 172 1,900 SUMMARY. Nitrogen added to 100 grams soil gram . Nitrogen fixed by 100 grams soil do . . . Per cent of nitrogen fixed 1892 0. 1892 0. 1892 0. 1892 0. 1892 0342 .0152 .0350 .0268 .0194 IS 1 8.03 18.5 14.2 10.2 0. 1892 .0231 12.2 SoU No. 428. Percolates of 100 cc. each. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. 100 66.2 88.2 89 109 120 103 90 118 147 147 172 154.5 116.9 112 109 136 116 139 157 172 172 187 44.9 103.6 93 103 107 110 114 147 159 147 172 86.1 108 110 123 118 149 148 176 187 172 187 140.7 .128 122 154 133 146 134 162 172 172 187 81.5 200 112 300 107 500 116 700 112 900 122 1,100 ... 144 1,300 145 1,500 178 1,700 172 1,900 187 SUMMARY. Nitrogen added to 100 grams soil. Nitrogen fixed by 100 grams soil. Per cent of nitrogen fixed .do. 0. 2064 0.2064 0. 2064 0. 2064 0.2064 .0668 .0322 .0618 .0358 .0257 32.4 15.6 29.9 17.3 12. 6 0.2064 . 0437 21.2 27 The veiy concordant results in the above table add proof to the theory that the fixation of ammonium nitrogen and potash are strikingly similar. The fixing power of the soils was far less for the nitrogen of ammonium sulphate in mixtures than when used alone. It was found that the heat decreased the fixing power of the soil greatly, while chloroform had a very slight effect. All solutions used in this series contained 172 parts per million nitrogen from ammonium sulphate. SODIUM NITRATE. The following table gives the results of applying sodium nitrate in mixtures : Absorption of nitrogen from a solution of sodium nitrate in a mixtd fertilizer . [Expressed in parts per million nitrogen in the percolate.] Soil No. 517. Soil No. 428. Percolates of 100 cc. each. Sodium nitrate, calcium phos- phate, and potassium sul- phate. Sodium nitrate, calcium phos- phate, and potassium sul- phate. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. 100 225 210 210 165 215 220 210 215 210 215 245 200 200 170 215 190 160 175 190 215 190 215 220 225 215 185 200 no 300 145 500 220 700 215 SUMMARY. Nitrogen added to 100 grams soil. . .gram. Nitrogen fixed by 100 grams soil do. . . Per cent of nitrogen fixed 0. 1075 0.1075 0.1075 0.1075 0. 1075 .0060 .0010 .0085 .0145 .0010 5.6 0.9 7.9 13.5 0.9 0. 1075 .0200 18.6 The solutions used contained 215 parts per million nitrogen from nitrates, and, as was to be expected, the soils absorbed only extremely small amounts. The fixing power was shown to be very much less when this salt was applied in mixtures than when applied alone, the effect of heat was to decrease the fixing power, while the effect of chloroform was to produce a decided increase in fixing power. The latter is probably due to the sterilizing effect of the antiseptic upon the organisms present. REMOVAL OF ABSORBED SALTS. At the conclusion of the preceding series distilled water was allowed to percolate through the tubes at the rate of 100 cubic centi- meters in 24 hours. In eveiy 100 cubic centimeters of the solution after the first thus obtained phosphoric acid, potash, and nitrogen were determined. 28 REMOVAL OF ABSORBED PHOSPHATE. In the following table will be found the results showing removal of absorbed phosphoric acid by distilled water from soil No. 517: Absorbed phosphoric acid removed from soil. [Expressed in parts per million PO< in the percolate.] Percolates of 100 cc. each. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Sodium nitrate, calcium phosphate, and potas- sium sulphate. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. 200 425 350 ^550 425 325 300 145 115 140 44 36 425 350 425 350 400 375 125 100 110 96 64 375 250 450 475 400 425 100 100 100 96 64 625 525 700 500 325 650 625 400 700 825 200 350 140 120 120 96 82 425 300 525 475 300 300 390 120 120 96 82 500 350 500 525 300 300 160 145 100 96 82 550 300 325 400 650 500 475 475 600 325 450 150 135 110 96 88 325 350 135 120 - 120 96 88 300 700 300 800 190 900 125 1,000 110 1,100 96 1,200 82 SUMMARY. 0. 7588 .2855 0.8548 .2820 0. 6670 .2835 1. 3982 .3679 1.278 .3184 1.0678 .3658 1.6134 .3133 1.2116 .3058 37.7 33.1 42.4 26.4 25. 34.1 19.4 25.2 P0 4 fixed gm.. P0 4 removed..gm.. Per cent of P0 4 re- moved 37.7 1.5014 20.0 The above results show that the concentration of phosphate in the percolate decreased quite rapidly, approaching a constant. Ap- parently the potash salt was less strongly fixed as the precentage re- moved is greater than the calcium salt. REMOVAL OF ABSORBED POTASH. In the following table will be found the results showing removal of absorbed potash by distilled water from soil No. 517: Removal of absorbed potash. [Expressed in parts per million K in the percolate.] Percolates of 100 cc. each. Ammonium sulphate, po- tassium phosphate, and potassium sulphate. Ammonium sulphate, cal- cium phosphate, and potassium sulphate. Sodium nitrate, calcium phosphate, and potas- sium sulphate. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. Un- treated. Heated. Chloro- form. 200 204 115 112 108 96 96 72 76 76 48 68 56 84 80 72 52 84 52 72 48 44 68 108 96 104 96 68 92 68 64 56 60 68 48 36 40 40 16 52 28 40 32 20 24 44 32 44 32 16 40 48 56 40 40 32 32 48 40 48 16 32 44 36 40 68 20 36 60 56 56 44 32 44 44 300 40 400 36 500 52 600 20 700 36 800. . , 28 900 44 16 16 24 36 12 20 32 44 28 24 32 32 1,000 20 1,100 20 1,200 20 SUMMARY. K removed gm. 0. 1072 0.0712 0. 0S80 0. 0376 0. 0424 0. 0316 0.0344 0. 0420 0. 0348 29 The above table adds further proof toward indicating the small amounts of potash absorbed by this type of soil when added in mix- tures. (See also p. 25.) There is little decrease in concentration of the percolate with regard to this element. REMOVAL OF ABSORBED NITROGEN. In the following table will be found the results showing removal by distilled water of nitrogen absorbed from ammonium sulphate from soil No. 517: Removal of absorbed nitrogen. [Expressed in parts per million nitrogen in the percolate.] Percolates of 100 cc. each. Ammonium sulphate, potas- sium phosphate, and potas- sium sulphate. Ammonium sulphate, calcium phosphate, and potassium sul- phate. Untreated. Heated. Chloro- form. Untreated. Heated. Chloro- form. ■ 200 73 51 46 42 31 25 21 16 21 21 13 68 44 39 30 20 17 16 16 16 16 16 66 44 42 33 22 22 18 16 21 18 11 84 46 28 18 7 7 4 3 11 3 2 73 38 25 13 9 7 6 5 9 4 4 73 300 40 400 22 500 10 600 700 .. 3 800 3 900 3 1,000 6 1,100 2 1,200 2 SUMMARY. Nitrogen fixed gm. . Nitrogen removed gm. . Per cent of nitrogen removed 0.0342 .0360 0. 0152 .0298 0. 0350 . 0313 39.5 0. 0268 . 0213 79.6 0. 0194 .0193 99.4 0. 0231 .0164 71.0 The above table discloses the peculiar fact that practically all the nitrogen fixed by the soil from ammonium sulphate was removed by passing a liter of water through it. The concentration of the solu- tion tended to decrease toward a constant value, as was the case with all the other elements of plant food. SUMMARY. The data presented in the foregoing pages throw considerable light upon the behavior of fertilizer salts in Hawaiian soils. They show the variation in absorbing power with the variation in soil types and composition of fertilizer added. Hawaiian soils have resulted from the degradation of lava rocks, some of which have subsequently been changed through the addition of coral limestone or submergence by the sea. Therefore they would naturally be expected to be of a highly basic nature, and to yield a highly basic soil solution, depending upon the absorptive power of the soil. Some of the soils have been 30 subjected to dense tropical plant growth, resulting in the accumula- tion of high percentages of humus, which has been shown in the previous tables to affect materially the absorbing power. Further- more, the data indicate that the concentration of the soil solution does not depend primarily upon the solubility of the mineral con- stituents, nor the amount of fertilizer added, but upon the absorbing power of the soil. As was expected, the fixation of phosphoric acid was much higher than the other elements. This is due to the highly basic character of the soils, and especially to the large amounts of iron, aluminum, and titanium present. It has been found in recent pot experiments with this type of soil that crops respond most readily to soluble phosphates — namely, sodium phosphate and acid phosphate. There was considerable difference in the physical action of calcium and po- tassium phosphates, the latter having a decided deflocculating action upon the clay, while the calcium salt filtered through the soil column perfectly clear. This, coupled with the results of the pot experi- ments cited above, indicates that absorbed sodium and potassium phosphates are not insoluble, but diffuse more readily and are more easily available for the growing plants. This indicates that phos- phate should be applied to Hawaiian soils in the soluble form, and the best time for application is just before planting, not on account of any danger of loss through drainage, but through the danger of a slight decrease in availability, due to reversion. Apparently the controlling factors in the fixation of potash are the amounts of lime and magnesia present. This is very clearly shown in the above tables, and the soils used in the experiments were good examples with which to illustrate this point. The fixing power for this element, while not so strong as for the phosphoric acid, is quite marked. However, it should not be applied in too large quantities, nor too often, as it is quite readily leached from the soil by rains and irrigation. The fixation of ammonium nitrogen, as already mentioned, is con- trolled by the same general factors which govern the absorption of potash. But the point of saturation is in most cases above that of the potash. However, it is not so strongly fixed and is leached out quite readily by the rains and drainage water. Some investigators claim that ammonia replaces the bases combined with the complex "hu- mates," and, if so, this accounts for the soils in the first series having such a high fixing power both for potash and ammonium nitrogen, while the red clay soil was strikingly lower. The power of the soil for fixing nitrate nitrogen is almost negligible, except in case of the highly organic soils. Apparently the organic 31 matter reacted with the nitrate solution, as the effect of this solution on the soil was quite marked. The series showing the relation of the fixing power of soil and sub- soil, and the effect of drying in the air, gave only slight differences. It was found, however, that phosphoric acid was fixed more strongly by the fresh soil, but there was scarcely any difference between the soil and subsoil. This is probably due to the fact that there is little, if any, difference in the mechanical condition of soil and subsoil in this red clay type, and also very little difference in chemical compo- sition. The fixation of potash was higher in the air-dried soil, as previously explained, and higher in the subsoil than the soil. The ammonium nitrogen, strange to say, unlike the potash, was more strongly fixed by the fresh soil, which indicates the possibility of cer- tain organisms affecting the fixation. The subsoil had a higher fixing power than the soil. There probably are also organisms acting as fixing agents for the nitrates, as the fresh samples had a higher fixing power than the air dry, while there was no difference in that of the soil and subsoil. The most striking results are those obtained from the series in which a solution of mixed fertilizer was used. From the data at hand the conclusion is thought justified that the least waste is to be had by application of fertilizer salts singly rather than in mixtures. When the salts were applied singly there was a marked loss of potash, a decrease in amount of ammonium nitrogen fixed, a decrease in nitrate nitrogen, and a decrease in phosphates in case of the red clay, but scarcely any difference with the organic soil. However, there was no deflocculation of the soil when the salts were added in mixtures, except to a small extent in the mixtures which contained potassium phos- phate. In this instance the percolates came through cloudy — that is, they contained deflocculated clay. On the other hand, the extracts in which the calcium salt was used were perfectly clear and colorless. Again, all the percolations proceeded quite rapidly, while several of the salts, the phosphates in particular, when used alone, would not allow a solution to pass through a column of soil. Solutions contain- ing potassium phosphate percolated more slowly than those contain- ing calcium phosphate. The effect of heat and antiseptics was not very striking and the results were not very consistent. In one instance, a highly organic soil, heat decreased the fixing power for phosphoric acid, while in general it decreased the fixing power for potash, ammonium nitrogen, and nitrate nitrogen. The effect of chloroform on the fixation of the first three elements was negligible, while it increased the fixing power for nitrates. 32 The removal of the absorbed elements approached quite rapidly a constant in the case of the potash and ammonium salts, but more slowly in that of the phosphates. This was due to the excessive amounts of this constituent which had been added. By reference to tables on pages 5 and 8 it will be seen that when phosphates were added to the soil in light applications the concentration of the solution remained practically unchanged for an indefinite period. ACKNOWLEDGMENTS. Acknowledgments are due and thanks are hereby extended to Dr. W. P. Kelley for valuable suggestions and for interest shown through- out this investigation. ADDITIONAL COPIES OF THIS PUBLICATION MAY BE PROCURED FROM THE SUPERINTENDENT OF DOCUMENTS GOVERNMENT PRINTING OFFICE WASHINGTON, D. C. AT 5 CENTS PER COPY