BULLETIN No. 270 FEBRUARY 1950 RESPONSE of CROPS to VARIOUS PHOSPHATE FERTILIZERS CEMENT ADVAN OFSCIENCE S ZM AGRICULTURAL EXPERIMENT olie ALABAMA M. J. Funchess, Director POLYTECHNIC STATION INSTITUTE Auburn, Alabama FOREWORD "Response of Crops to Various Phosphate Fertilizers," a presentation of significant results from 88 years of research involving over 1,800 experiments by the Alabama Agricultural Experiment Station, was prepared by L. E. Ensminger, Associate Soil Chemist, who since joining the staff in late 1944 has been concerned with phosphorus research. In preparing this report, Dr. Ensminger has compiled the results of many workers of this Station. Summarized are results of phosphorus experiments at the Main Station, Auburn, Alabama, carried on by R. Y. Bailey*, F. E. Bertram, E. F. Cauthen**, J. F. Duggar**, Franklin Fudge*, M. J. Funchess, W. H. Pierre*, F. W. Parker*, G. D. Scarseth*, D. G. Sturkie, J. W. Tidmore**, H. B. Tisdale, and G. W. Volk*. Phosphorus response studies at the Experiment Fields reported herein were conducted by H. R. Benford*, F. E. Bertram, C. L. Breedlove*, J. W. Richardson, J. F. Segrest, Jr., J. R. Taylor*, R. W. Taylor*, and J. T. Williamson. Phosphorus studies at the Substations were done by K. G. Baker, R. C. Christopher*, S. E. Gissendanner, Fred Stewart, and J. P. Wilson**. Cooperative cotton, peanut, and pasture tests with farmers were conducted by E. L. Mayton, H. T. Rogers*, J. M. Scholl*, E. H. Stewart*, and J. T. Williamson. - Editor. * Resigned ** Deceased CONTENTS DESCRIPTION OF MATERIALS--------------- - ----- Page 8 3 RESULTS OF EXPERIMENTS EXPERIMENTS OF LONG DURATION ------- 6-------------------6 EXPERIMENTS OF SHORT DURATION.------------------------RESPONSE PHOROUS OF CLOVER PASTURES TO SOURCES --------------- 19 OF PHOS- ---------------------------------- 29 34 DISCUSSION----------------------------------SUMMARY ---------------------------------------------------- -------- 36 LITERATURE CITED --------------------------------------------------------- 389 FIRST PRINTING 5M RESPONSE of CROPS to VARIOUS PHOSPHATE FERTILIZERS L. E. ENSMINGER, Associate Soil Chemist* MANY OF THE SOILS of Alabama are low in available phosphorous; consequently, phosphate fertilizers must be used to obtain satisfactory crop yields. Most of these soils were deficient in available phosphorous when first cleared; unless they have been given large applications of an available source of phosphorus during the period of cultivation, they have remained deficient in available phosphorus. As will be pointed out later, loss of phosphorus by erosion is one reason why rather large applications of phosphorus may be required over a period of years to build up an appreciable supply of phosphorus in soils. It should be pointed out in this connection that, since removal of phosphorus by the harvested portion of most crops is small, it should not preclude the possibility of considerable accumulation of applied phosphorus. The importance of phosphorus in Alabama agriculture is shown by the volume of consumption. During the 1945-46 crop year, Alabama farmers spent approximately 11.5 million dollars for phosphorus in mixed fertilizers and as separate material. The State ranked third in the United States in consumption of phosphate fertilizers during that crop year, with a total of 91,6380 tons of P2 0 5 being applied. The trend has been upward, as shown by an annual average of 55,306 tons of P205 used during the period 1935-44 (6). DESCRIPTION of MATERIALS There are several kinds of phosphate materials available for distribution and there are others that may be placed on the market if found satisfactory as sources of phosphorus for plant growth. Sources on the market at present include such materials as raw rock phosphate, treated phosphate (superphosphate), and by-product phosphates. Solubility of these materials varies * Author-compiler, see FOREWORD. 4 ALABAMA AGRICULTURAL EXPERIMENT STATION considerably; usually the more soluble the material, the more quickly its phosphorus is released for plant growth. However, the comparative availability of different sources of phosphorus may depend to a considerable extent on soil conditions and the crop being grown. The sources of phosphorus used in the experiments reported herein are briefly discussed with respect to preparation and properties. Rocx PHOSPHATE. Rock phosphate occurs in natural deposits throughout the world. The ones of greatest commercial importance in the United States are located in Florida and Tennessee in the East, and in an area comprising Idaho, Montana, Utah, and Wyoming in the West. The principal constituent of the American phosphate rock is fluorapatite, which may be represented by the formula CaloF 2 (PO4 ) 6. Raw rock phosphate is not used extensively for direct application to the soil because of its insolubility. To be effective it must be finely ground. A good grade of rock phosphate will contain about 32 per cent total P 20 5. Only a trace of this is water-soluble. COLLOIDAL PHOSPHATE. The USDA Yearbook of Agriculture, Soils and Men, 1938 (10), gives the following description of colloidal phosphate: "'Colloidal phosphate' is a trade name applied to finely divided, comparatively low-grade rock phosphate or phosphatic clay. It is also designated 'waste pond phosphate' for the reason that in the hydraulic operation involved in mining rock phosphate in Florida a considerable quantity of fine phosphatic material, virtually colloidal from a mechanical standpoint, is washed into ponds and settles out. When removed, following drainage and evaporation of water, it contains a relatively high proportion of clay, so that the Colloidal phosphate usually contains only from 18 to 23 per cent of phosphoric acid. On account of the presence of so much foreign material, principally clay, Colloidal phosphate is considered to be unsatisfactory for treatment with sulfuric acid. The claim is made for this material and others of a similar nature that not only is the phosphoric acid more quickly available than that of mechanical ground rock phosphate, but also that the content of minor elements in Colloidal phosphate makes it superior to its close relative, rock phosphate. These claims, while highly interesting, have failed of substantiation in a number of states, particularly so when a comparison of such materials with superphosphate is taken into consideration." According to the above discussion, colloidal phos- CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 5 phate may be considered as rock phosphate diluted with colloidal material. SUPERPHOSPHATE. Superphosphate (16 to 20 per cent avail- able P2 0 5 ) is made by treating ground rock phosphate with sulfuric acid in about equal proportions by weight. In this process all of the sulfuric acid is consumed. This results in a mixture of more soluble phosphates and calcium sulfate (gypsum). Ordinary superphosphate usually contains 18 to 20 per cent available P20 5. Its P2 0 5 content depends largely on the grade of rock phosphate used. Practically all of the phosphorus in superphosphate is water-soluble. It is used as a source of phosphorus in mixed fertilizers as well as for direct application, and it ranks first among phosphatic materials in quantity consumed for fertilizer use. TRIPLE SUPERPHOSPHATE. Triple superphosphate is made by treating rock phosphate with phosphoric acid instead of sulfuric acid. This results in a product higher in P20 5 than ordinary superphosphate. Triple superphosphate usually contains 42 to 52 per cent P205 and it may contain up to 10 per cent gypsum, depending upon method of manufacture. It is largely monocalcium phosphate, and most of the phosphorus is water-soluble. It should be pointed out here that the TVA triple superphosphate used in the experiments reported herein did not contain any appreciable amount of gypsum. AMMO-PHOS A. Ammo-phos A is a trade name used to designate a grade of mono-ammonium phosphate used as a fertilizer material. It contains about 11 per cent N and 45 to 48 per cent P205. It is produced by partially neutralizing phosphoric acid with ammonia. DI-AMMONIUM PHOSPHATE. Di-ammonium phosphate is made by adding ammonia in the proper proportion to a solution of mono-ammonium phosphate. The di-ammonium phosphate, being less soluble, separates as crystals. It contains about 21 per cent N and 53 per cent P2 05s. FUSED TRICALCIUM PHOSPHATE. Fused tricalcium phosphate is made by heating rock phosphate to the fusion point in the presence of water vapor and silica. This causes most of the fluorine to volatilize, resulting in the formation of a tricalcium phosphate containing about 80 per cent P20 5, most of which is soluble in ammonium citrate. 6 ALABAMA AGRICULTURAL EXPERIMENT STATION PRECIPITATED TRICALCIUM PHOSPHATE. Tricalcium phosphate may be formed by ammoniation of mono-calcium phosphate. It contains approximately 42 per cent total P 20 5 . CALCIUM METAPHOSPHATE. Calcium metaphosphate is pro- duced by bringing hot gaseous phosphorus in contact with phosphate rock at high temperature. After the mass melts, it is withdrawn from the furnace and allowed to cool. When ground it is ready for use as a fertilizer. It contains 62 per cent available P 20 5 (soluble in ammonium citrate solution). POTASSIUM METAPHOSPHATE. It is the potassium salt of metaphosphoric acid and contains about 60 per cent P 2 0 5 and 40 per cent K20. BASIC SLAG. Basic slag is produced as a by-product of the steel industry. Phosphorus occurs in some iron ores, and steel made from them is brittle unless the phosphorus is largely removed. In removing the phosphorus and other impurities, a blast of air is blown through the molten iron in a converter containing lime. The phosphorus oxidizes and unites with the lime. The resulting mass is lighter than iron and thus rises to the surface where it is drawn off, cooled, and ground to a certain fineness. Basic slag usually contains 8 to 10 per cent total P20 5 , most of which is soluble in citric acid. Basic slag should not be confused with blast furnace slag which is calcium silicate. RESULTS of EXPERIMENTS The importance of phosphorus in Alabama agriculture was recognized years ago. Some of the experimental work reported in this bulletin was started as early as 1911. Experiments of short and of long duration were conducted in order to test the immediate relative efficiencies of various sources of phosphorus, as well as relative efficiencies involving cumulative effects. Since the relative efficiency of sources may depend on conditions, a number of crops were tested at a large number of locations in the State. For convenience in discussing the results, the work reported herein is divided into experiments of long duration and of short duration dealing with crops other than pastures, and experiments with permanent pastures. EXPERIMENTS OF LONG DURATION STUDY OF SOURCES OF PHOSPHOROUS IN A ROTATION OF CORN AND COTTON WITH WINTER LEGUMES. In 1930 a sources-of-phosphorus study was started at the Tennessee Valley, Sand Mountain and CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 7 Wiregrass substations, and at the Prattville Experiment Field on Decatur, Hartsells, Norfolk, and Greenville soils, respectively. The different sources of phosphorus used in these tests were superphosphate, basic slag, triple superphosphate, precipitated tricalcium phosphate, Ammo-phos A, rock phosphate, and colloidal phosphate. The phosphorus was applied to the cotton and winter legumes. The rate per acre of P 20 5 used, and yield data are given in Table 1. Response of corn to phosphates was rather low at all locations except the Sand Mountain Substation, and as a result the differences in yield for the various phosphates were rather small. In the case of cotton, there was a rather large response to phosphorus at the Tennessee Valley and Sand Mountain substations, while at the two other locations the response was rather low. Analysis of soil samples collected at the beginning of the experiment showed that the soils at the Tennessee Valley and Sand Mountain substations were lower in soluble phosphorus than the soils from the Wiregrass Substation and Prattville Field. Winter legumes responded well to phosphorus at all locations; however, the greatest response was obtained at the Tennessee Valley and Sand Mountain substations, as was in the case of cotton. The average yield increases of cotton, corn, and winter legumes at the four locations are given in Table 2. In relative terms the increases of cotton may be expressed as follows: superphosphate, 100; triple superphosphate, 90; basic slag, 88; precipitated tricalcium phosphate, 87; and rock phosphate, 49. The increases of corn expressed in a like manner are as follows: superphosphate, 100; basic slag, 110; triple superphosphate, 106; precipitate tricalcium phosphate, 105; and rock phosphate, 64. The winter legumes show the following relative increases: superphosphate, 100; basic slag, 89; precipitated tricalcium phosphate, 89; triple superphosphate, 89; and rock phosphate, 51. It is evident that the exact efficiency of the phosphates depends somewhat on the test crop. However, rock phosphate was low in all cases. The foregoing efficiencies were from the use of 48 pounds of P20 5 from each source of phosphorus. The use of 24 pounds of P20 5 shows relative efficiencies for cotton, corn, and legumes of 82, 80, and 67, respectively, as compared to 48 pounds of P2 0 5. Colloidal phosphate, Ammo-phos A, and the double rate of rock phosphate (96 pounds P 20 5 ) were not included in the test at the Sand Mountain Substation because of the lack of space. TABLE 1. YIELD OF CROPS ON CHECK PLOTS AND INCREASED YIELDS OVER CALCULATED CHECK FROM VARIOUS SOURCES PHOSPHORUS IN A 2-YEAR ROTATION OF COTTON AND CORN WITH WINTER LEGUMES, 1930-1945 OF Plot Source of no. phosphorus Amount per acre Tennessee Valley Substation of P 2 0 5 to cotton Green wt. of no. to Seed Corn iand phsphrus winter winter legumes cotton legumes Lb. Lb. Bu. Lb. 2 Yield increase per acre over calculated check Sand Mountain Wiregrass Substation Substation Prattville Field Seed Corn Green wt. of winter legumes Bu. Lb. 32.7 5.2 7.0 4.1 32.7 4.9 5.4 7.0 29.9 7.7 6.0 7.1 33.1 7.1 6.7 5.1 33.1 6,007 3,999 5,571 2,408 6,749 3,066 3,183 4,123 5,920 4,275 3,294 4,686 6,830 5,916 5,728 3,675 6,208 r Green Seed Corn wt. of winter cotton legumes Lb. Bu. Lb. 875 484 546 383 948 - Seed cotton Lb. 1,178 176 213 50 954 100 120 134 883 117 -87 230 976 179 226 155 1,177 Corn Green wt. of legumes Lb. Bu. 24.7 4.9 4.9 1.6 17.4 3.4 2.7 4.0 15.4 3.7 3.5 5.4 18.3 4.7 6.0 3.1 23.3 Lb. 1,469 100 114 44 1,390 88 78 128 1,305 174 -177 138 1,350 181 187 101 1,477 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 None (check) Basic slag Superphosphate Rock phosphate None (check) 2 Rock phosphate Colloidal phos. Ppt. trical. phos. None (check)2 Triple superphosphate Ammo-phos A Superphosphate2 None (check) 2 Superphosphate Superphosphate 4 Superphosphate None (check) 2 0 48 48 48 0 96 48 48 0 48 48 24 0 48 24 24 0 1,025 421 460 172 1,151 320 217 368 1,051 390 383 437 1,145 392 428 350 892 28.2 7.9 5.7 3.5 30.5 6.1 3.5 6.0 26.9 6.0 6.5 7.1 29.7 6.4 6.2 5.2 25.0 1,870 7,539 8,279 3,548 2,588 5,585 3,441 7,323 1,928 7,483 7,557 8,448 2,443 7,452 7,423 4,602 1,506 15.3 19.9 16.9 1.2.8 17.7 18.9 . 18.9 18.9 22.3 15.9 22.4 14.2 15.3 1,201 3,760 3,745 3,549 1,524 5,051 4,897 5,174 2,138 4,236 5,135 3,081 1,230 4,947 3,857 2,839 869 > 4,170 1,377 1,143 1,807 3,719 1,526 1,088 2,291 4,421 2,776 4,346 2,341 4,836 E G c r o r 522 517 459 1,107 489 481 489 886 ' x -o m Z ... 1 In addition to phosphorous, cotton received 36 pounds of nitrogen and 24 pounds of potash per acre at all locations except at the Tennessee Valley Substation, where nitrogen was discontinued in 1937. Winter legumes received the above rates of P2 0 5 beginning in 1935 at the Tennessee Valley and Sand Mountain Substations, and in 1936 at Wiregrass Substation and Prattville Experiment Field. Corn received 600 pounds per acre of an 0-5-2 from 1930-34 at the Tennessee Valley and Sand Mountain Substations and Prattville Field, and from 1930-35 at the Wiregrass Substation but received no fertilizer after these dates. Check yields (in italic) are total yields per acre; all others are calculated increases over check (based on assumed uniform soil variation between check plots).-4 In addition to superphosphate, rock phosphate applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. In addition to superphosphate, basic slag applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. 2 4 > O0 Z 0 TABLE 2. AVERAGE INCREASED YIELSS OF SEED COTTON BY 4-YEAR PERIODS AND AVERAGE INCREASED YIELDS OF SEED COTTON, CORN, AND WINTER LEGUMES, 1930 TO 1945 (Summary of Data Reported in Table 1) Amount of Average increased yields per acre of seed cotton by P 205 per acre t ttownternd_ ctontad4-ereeiosSed 4-year___periods__SeedGreen legumes 0 phosphorus Source of 1930-33 1934-37 1938-41 1942-45 Lb. 92 150 48 Average yields per acre Relative efficiency of phosphates (48 lb. P205 as 1930-45 superphosphate =100) Corn of winter z -r. wt. 0 cotton legumes Cotton Corn Legumes Lb. Basic slag Superphos Rock phos; Ppt. trical cium phos. Triple superphosphate 1 Superphos 3phate Superphos ,phate' Superphos 3phate Lb. 225 248 100 Lb. 438 472 232 Lb. 421 463 268 Lb. 294 333 162 Bu. 9.5 8.6 5.5 Lb. 4,789 5,101 2,593 0 87 105 89 90 106 89 95 106 101 99 120 111 82 80 67 TENNESSEE VALLEY AND WIREGRASS SUBSTATIONS AND PRATTYILLE EXPERIMENT FIELD 48 130 187 356 324 256 6.0 5,471 -100 100 100 Superphosphate 48 37 60 126 131 89 3.1 2,275 35 S 52 42 Rock phosphate 66 S 169 4.8 3,344 92 135 223 227 96 80 61 Rock phosphate 54 3.9 2,589 190 207 138 48 41 117 65 47 Colloidal phosphate 40 5.3 3,980 16 3 88 48 115 80 102 -138 Ammo-phos A 73 3,539 79 288 202 4.5 87 149 285 24 75 67 Superphosphate SIn addition to superphosphate, rock phosphate applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. zIn addition to superphosphate, basic slag applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. 3phate phate 48 48 48 C TENNESSEE VALLEY, SAND MOUNTAIN, AND WIREGRASS SUBSTATIONS AND PRATTYILLE. EXPERIMENT FIELD N' 88 100 49 110 100 64 94 100 51 0 0 48 48 24 24 24 125 169 163 162 101 222 232 230 236 188 397 406 422 465 385 407 392 449 461 420 288 300 316 331 273 9.0 9.1 9.6 10.3 6.9 4,551 4,545 5,150 5,658 3,425 'O a~ mf m1 M~ mI N 10 ALABAMA AGRICULTURAL EXPERIMENT STATION As reported in Table 2, these phosphates show the following relative efficiencies for cotton at the three locations: superphosphate, 100; colloidal phosphate, 54; double rate of rock phosphate, 66; rock phosphate, 35; and Ammo-phos A, 16. For corn the relative increases are in the following order: superphosphate, 100; Ammo-phos A, 88; double rate of rock phosphate, 80; colloidal phosphate, 65; rock phosphate, 52. The order of effciency for winter legumes is the same as for corn. The double rate of rock phosphate, which is about equivalent in cost to 48 pounds of P20 5 from superphosphate, did not give as large an increase in yields as superphosphate. Colloidal phosphate produced greater yields than the same amount of P20 5 in the form of rock phosphate. However, colloidal phosphate did not give as large increases in yields as 24 pounds of P20 5 from superphosphate, which would be about equivalent in cost to the colloidal phosphate. It has been suggested that the relative efficiency of the insoluble phosphates may increase with successive applications. Increased yields of seed cotton from 1930 to 1945 by 4-year periods are reported in Table 2. These data show that there is a tendency for the relative efficiency of the insoluble phosphates to increase with time. However, with time the actual increases in yield show a wider spread between superphosphate and the insoluble phosphates. Studies in Alabama (5, 9, 11) have shown that considerable added phosphorus may be lost by erosion. Such loss of phosphorus would tend to prevent a cumulative effect of added phosphates. Certain plots received 24 pounds of P 20 5 as superphosphate plus periodic applications of rock phosphate or basic slag. As indicated by data in Table 2, the addition of 2,000 pounds of rock phosphate or basic slag every 6 years increased the yields of cotton, corn, and winter legumes appreciably. The addition of basic slag gave somewhat higher yields than did the addition of rock phosphate. STUDY OF SOURCES OF PHOSPHORUS IN A ROTATION OF CORN AND COTrTON WITHOUT WINTER LEGUMES. A source-of-phosphorus test was started in 1930 at the Tennessee Valley, Sand Mountain and Wiregrass substations, and at the Monroeville and Alexandria experiment fields using corn and cotton in rotation as test crops. The sources of phosphorus (superphosphate, basic slag, triple superphosphate, precipitated tricalcium phosphate, Ammo-phos TABLE 3. INCREASED YIELDS FROM VARIOUS SOURCES OF PHOSPHORUS OVER CALCULATED CHECK PLOT YIELDS IN A 2-YEAR ROTATION OF COTTON AND CORN, 1930-45 Amount of PO, Tenn. Valley 2 per Substation acre to Seed Corn cottoncotton Lb. 828 387 455 204 973 328 233 382 861 399 340 433 978 426 458 378 805 Bu. 81.9 7.1 7.0 2.1 31.5 4.8 3.2 5.4 30.1 4.7 3.0 4.9 32.3 4.4 5.7 5.0 28.3 Yield increase per acre over calculated check Monroeville Wiregrass Sand Mountain Substation Substation Field Seed Corn Seed Corn Seed Corn cotton cotton cotton Lb. 831 452 485 390 779 . 494 563 . 567 724 528 575 500 698 Bu. 28.1 9.1 7.9 7.7 27.6 Lb. 1,141 115 123 23 1,104 73 110 136 1,079 142 -217 146 1,139 147 160 88 1,155 Bu. 22.9 2.3 1.8 1.8 21.5 1.8 1.9 3.5 23.1 1.4 -2.4 -1.9 22.9 0.4 2.4 0.7 24.0 Lb. 858 227 320 148 867 170 136 221 872 189 -6 288 906 330 305 291 789 Bu. 31.2 2.1 2.0 0.9 29.0 2.2 1.6 1.8 29.5 1.6 1.5 1.0 29.7 1.1 0.8 2.3 29.8 2 A O o Plot Plot Source of Sourceuofs Alexandria Field Seed Corn cotton Lb. 732 343 441 215 863 407 Bu. 31.4 2.5 3.2 2.9 36.5 3.0 m 0 in O 0 7v Z 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Lb. 0 None (check) 48 Basic slag 48 Superphosphate 48 Rock phosphate 2 0 None (check) 96 Rock phosphate 48 Colloidal phosphate 48 Ppt. tricalcium phos. 2 0 None (check) Triple superphosphate 48 48 Ammo-phos A 24 Superphosphate' 2 0 None (check) 48 Superphosphate 24 Superphosphate' 24 Superphosphate 2 0 None (check) 2 z m C H 'O -- 0 z *w 8.4 8.8 9.3 25.7 9.3 12.3 7.9 25.2 fl 329 440 835 404 387 328 678 1.7 2.6 35.0 2.1 2.8 2.1 30.0 -4 m -r m N In SCotton received 600 pounds per acre of 6-10-4 from 1930 to 1934, inclusive, and 600 pounds of 6-8-4 thereafter at all locations except at the Wiregrass Substation where the fertilizer was changed to a 6-8-4 in 1936. Corn received 600 pounds per acre of 6-5-2 from 1930 to 1934 and 36 pounds of nitrogen only thereafter at all locations except at the Wiregrass Substation where the fertilizer was changed to 36 pounds of nitrogen only in 1936. 2 Check yields (in italic) are total yields per acre; all others are calculated increases over check (based on assumed uniform soil variation between check plots). 2 In addition to superphosphate, rock phosphate applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. 'In addition to superphosphate, basic slag applied at the rate of 2,000 pounds per acre in 1930, 1936, and 1942. _m r_ 12 ALABAMA AGRICULTURAL EXPERIMENT STATION A, rock phosphate, and colloidal phosphate) were the same as those used in the study previously discussed. Yield data for each of the five locations are presented in Table 3. The response of cotton to phosphorus was rather large at all locations except at the Wiregrass Substation. In most cases corn gave little response to phosphorus and as a result is not a very satisfactory crop to use in evaluating sources of phosphorus. Since each source of phosphorus was not tested at every location, Table 4 has been divided into two sections so that comparisons may be made between phosphates tested at the same places. The relative efficiencies of materials tested at five locations are as follows for cotton: superphosphate, 100; basic slag, 83; and rock phosphate, 58. A half rate of superphosphate (24 pounds P20 5) gave 87 per cent as large an increase in yield as 48 pounds of P205 from the same source. The use of a ton of rock phosphate or basic slag every 6 years in addition to the half rate of superphosphate increased the yield of seed cotton by 56 and 58 pounds, respectively. The relative efficiencies of sources tested at the Tennessee Valley and Wiregrass substations and at the Monroeville Field are: superphosphate, 100; double rate of rock phosphate, 61; colloidal phosphate, 51; rock phosphate, 40; and Ammo-phos A, 22. The double rate of rock phosphate did not increase the yield of cotton as much as superphosphate. Colloidal phosphate (48 pounds P20 5 ) did not produce as large an increase in yield as superphosphate applied at the rate of 24 pounds of P20 5 per acre. According to these data, the insoluble phosphates did not increase the yield of cotton as much as superphosphate when applied on an approximately equal cost basis. The cotton yields by 4-year periods are given in Table 4. As was the case for the rotation with legumes, the efficiency of the insoluble phosphates tends to increase with time. The actual spread between superphosphate and the insoluble phosphates, however, becomes larger with time. LIME PHOSPHATE EXPERIMENT. In 1922 an experiment was started to test the value of basic slag, superphosphate, and two rates of rock phosphate with and without lime. This experiment was conducted at Atmore, Prattville, Cusseta, Sylacauga, and Hackleburg on Greenville, Red Bay, Cecil, Decatur, and Atwood soils, respectively. The phosphates were applied at the rate of 600 pounds per acre with a second rate of rock phosphate ap- n 0 m TABLE 4. AVERAGE INCREASED YIELs OF SEED COTTON BY 4-YEAR PERIODS AND AVERAGE INCREASED YIELDS OF SEED COTTON AND CoRN, 1930-45 (Summary of Data Reported in Table 3) 0 z m Source of phosphorus P2 05 per acreapplied acre applied to cotton Average increased yield per acre of seed cotton by 4-year periods 1930-33 1934-37 1938-41 1942-45 Average acre, 1930-45 Seed cotton yields per Corn Relative efficiency (48 lb. as super2 phosphate = P05 100) Cotton Corn 0 c Lb. Basic slag Superphosphate 48 48 Lb. 118 210 Lb. 233 293 Lb. 397 454 Lb. 465 504 Lb. 303 Bu. TENNESSEE VALLEY, SAND MOUNTAIN AND WIREGRASS SUBSTATIONS AND MONROEVILLE AND ALEXANDRIA EXPERIMENT FIELDS Rock phosphate 1 Superphosphate 2 Superphosphate Superphosphate Superphosphate 48 24 24 24 48 97 232 220 162 259 159 295 276 230 251 235 462 489 407 367 352 504 517 468 352 365 211 373 375 317 307 4.6 4.5 83 102 0 in in 3.1 4.0 4.8 3.6 3.6 100 58 102 103 87 100 100 69 89 107 80 100 N TENNESSEE VALLEY AND WIREGRASS SUBSTATIONS AND MONROEVILLE EXPERIMENT FIELD 40 44 1.6 190 123 138 61 81 2.9 253 188 206 51 61 157 2.2 245 188 22 28 67 1.0 163 -162 82 75 252 2.7 324 355 of 2,000 pounds per acre in 1930, 1936, and 1942. the rate rate of 2,000 pounds per acre in 1930, 1936, and 1942. m n 88 77 48 Rock phosphate 135 157 96 Rock phosphate 118 76 48 Colloidal phosphate 123 77 48 Ammo-phos A 161 170 24 Superphosphate 1In addition to superphosphate, rock phosphate applied at 2In addition to superphosphate, basic slag applied at the (a) TABLE 5. RESPONSE' OF CROPS TO PHOSPHATE WITH AND WITHOUT LIME AT ATMORE, PRATTVILLE, CUSSETA, SYLACAUGA, AND HACKLEBURG, 1923-30 Weighted yield averages per acre of Lime locations from 1923-303 fper P Suracre S applied per acre Cotton, in fall applied Oats, Semier C or umes, 26 yerleu annually2 23 years 13 years 1922 2 20years per acre Increases in yield per acre due to phosphorus orf phosphorus phosphorus'1 Sorce of Summer Cotton Lb. 245 )IfI- Oats Bu. 8.4 Lb. 1,088 s Lb. None Basic slag 0 0 0 0 4000 4000 4000 Superphosphate Rock phosphate Rock phosphate None Basic slag Superphosphate 0 Lb. 0 96 Lb. 584 829 Bu. 27.2 35.6 Lb. 4,166 5,254 Bu. Bu. 4.8 14.6 19.4 96 192 768 0 96 96 880 736 774 654 879 914 4000 192 763 Rock phosphate 4000 768 777 29.4 Rock phosphate 1 Each plot received 200 pounds of sodium nitrate 33.2 30.5 31.6 26.6 33.7 35.7 81.4 4,454 4,160 4,279 5,261 6,402 6,524 5,806 6,036 to oats and 1 18.0 15.7 19.1 19.6 20.2 20.4 18.9 18.0 296 152 190 225 260 109 123 6.0 3.3 4.4 - -4 113 1,141 1,263 288 3.6 1.1 4.5 - A C L..v 7.1 9.1 4.8 2.8 ('/2N 545 775 -0.7 -1.6 0.6 0.8 C r- m m i.z 2N in drill to cotton) and 100 pounds of muriate of :potash (%/ to winter legume preceding cotton and %/ in drill to cotton). 2 Two-thirds of phosphorus to winter legume preceding cotton and %/ in drill to cotton. 'Weighted averages based on number of crops harvested; experiment did not run the full time at all locations. z 0 CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 15 plied at 2,400 pounds per acre. Yields and treatments for this experiment are reported in Table 5. The increased yields of cotton were in the following decreasing order for the phosphates with and without lime: superphosphate, basic slag, high rate of rock phosphate, and rock phosphate. For oats the positions of superphosphate and basic slag were reversed. In the case of summer legumes without lime, basic slag gave the best results, while with lime superphosphate resulted in slightly higher yields than those from slag. Differences between treatments for corn were small, especially for the limed plots. At these five locations an exceptionally high rate of rock phosphate failed to produce as much cotton as 600 pounds of superphosphate. Slag also gave somewhat lower yields than superphosphate. ROCK PHOSPHATE Vs. SUPERPHOSPHATE, AUBURN. An experiment designed to evaluate superphosphate and rock phosphate for production of cotton was located on a Chesterfield soil. The experiment was begun in 1920 and was continued through 1931. The phosphates were applied annually at the rate of 320 pounds per acre, which means that twice as much P20 5 in the form of rock phosphate was used as was applied in the form of superphosphate. Results of the experiment are presented in Table 6. TABLE 6. RESPONSE OF COTTON TO ROCK PHOSPHATE AND SUPERPHOSPHATE, MAIN STATION, 1920-31 Source of phosphorus Phosphate applied acre Lb. None Superphosphate Average yield Increase per per acre of seed acre due to cotton 1920-31 phosphorus Lb. Lb. 297 100 85 254 pounds per acre of per acre of sodium acre thereafter. Relative efficiency pper 0 597 320 894 851 Rock phosphate 320 In addition to phosphate each plot received 160 kainit containing 12.5 per cent K 2 O and 100 pounds nitrate from 1920-23, inclusive, and 160 pounds per 1 Superphosphate resulted in an average increase of 297 pounds per acre of seed cotton as compared to a 254-pound average increase from rock phosphate. Thus, rock phosphate had a relative efficiency of 85 in the 12-year experiment. FERTILIZER ROTATION EXPERIMENT. Fertilizer rotation experi- ments, started in 1916, were located on a Hartsells soil near Albertville and on a Greenville soil near Jackson. Yield data TABLE 7. RESPONSE OF COTTON AND CORN TO PHOSPHATES IN AN EXPERIMENT ON HARTSELLS AND GREENVILLE SOILS Average yields and increased yields per acre Treatments' Source of phosphorus None Superphosphate (16% P2 0 5 ) Basic slag (16% P0Os) Rock phosphate Superphosphate Basic slag Rock phosphate Albertville', average 1920-26 Phosphate Lime per Seed per acre acre cotton Lb. Lb. Lb. 0 0 223 240 0 494 240 0 515 480 0 392 240 4,000 602 240 4,000 573 480 4,000 592 Corn Bu. 21.8 25.5 29.5 27.5 29.2 27.0 32.0 Jackson3 , average 1916-29 Seed cotton Lb. 498 736 728 709 875 837 737 Corn Bu. 19.2 22.3 22.7 22.2 24.1 22.9 21.5 Weighted average of two locations Seed cotton Corn Average Increase Average Increase yields yields Lb. Lb. Bu. Bu. 406 20.0 655 249 23.4 3.4 657 251 25.0 5.0 603 197 24.0 4.0 784 378 25.8 5.8 749 343 24.3 4.3 689 283 25.0 5.0 0% w I0 ]> W I- 'All plots received 100 pounds of sodium nitrate and 50 pounds of muriate of potash. 2 Hartsells soil. * Greenville soil. TABLE 8. RESPONSE OF CROPS TO SUPERPHOSPHATE AND ROCK PHOSPHATE IN CULLARS ROTATION, MAIN STATION c C Total P 2 0 5 Plot Source of per acre 1911-31 Average yields per acre, 1911-31 Seed 19 av. Corn 17-yr. av. Oats 15-yr. av. Vetch, 4-yr. av. Average yields per acre, 1932-47 Seed 15-yr. av. Corn 16-yr. av. Oats 16-yr. av. Vetch, av. c *" F n c Lb. Lb. Bu. Bu. Lb. Lb. Bu. Bu. Lb. None 0 772 38.8 36.1 2,410 473 27.7 29.3 764 Superphosphate 914 1,243 49.2 48.6 7,369 995 45.2 46.8 2,856 Rock phosphate 3,656 1,161 46.1 50.8 9,088 1,169 51.3 51.5 7,385 'Two-thirds of phosphorus was applied to cotton and % to corn. From 1911-23 corn received 790 pounds per acre of dried blood and 532 pounds of kainit (12.5% K 2 0); cotton, 900 pounds of dried blood and 614 pounds of kainit; and oats, 468 pounds of sodium nitrate and 510 pounds of kainit. From 1924-31 corn received 700 pounds of sodium nitrate and 132 pounds of muriate of potash; cotton, 800 pounds of sodium nitrate and 282 pounds of muriate of potash; and oats, 468 pounds of sodium nitrate. From 1932 on corn, cotton, and oats received 240 pounds of sodium nitrate. 2 3 5 m x -4 m m z -I -z I z CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 17 from the Albertville test for the first 4 years were lost by fire. According to the data in Table 7, the average cotton yields for the two areas show that 240 pounds per acre of slag (16 per cent P20 5 ) produced as much cotton as 240 pounds of superphosphate on the unlimed plots. However, on the limed plots superphosphate was superior to basic slag. An application of 480 pounds of rock phosphate (four times as much P20s) did not increase the yield of cotton as much as superphosphate on either the limed or unlimed plots. There was little response of corn to phosphorus; consequently, differences in yields due to sources were small. ROCK PHOSPHATE Vs. SUPERPHOSPHATE, CULLARS ROTATION. The Cullars rotation on Norfolk loamy sand was started in 1911. One objective of the experiment was to study the value of superphosphate and rock phosphate for the production of crops in a rotation. These plots were converted to a residual phosphorus study after 21 years of phosphate applications. The yields obtained from the different phosphates are given in Table 8. The yield data from 1911-31 show that superphosphate produced a little more cotton and corn than rock phosphate but less oats and vetch. It should be pointed out, however, that the rock phosphate applied contained 4 times as much total P2 0 5 as the superphosphate. In 1932 the phosphate treatments on plots 3 and 5 were discontinued in order to study residual phosphorus. The average yields from 1932 to 1947 for all crops grown were higher where rock phosphate had been applied than where superphosphate had been used. In Figure 1 the cotton yields are plotted by years for the residual period. This graph shows that there was not much difference in value of the two phosphates the first 9 years; however, rock phosphate resulted in consistently higher TABLE 9. SOLUBLE AND TOTAL PHOSPHORUS CONTENT OF SOILS OF PLOTS, CULLARS ROTATION, MAIN STATION Plot No. 2 3 5 Source of phosphorus None Superphosphate Rock phosphate 2 (Samples collected June 3, 1947) NH 4 F 2 soluble P 20 5 p.p.m. 23 40 71 Total Total P 05 Dilute acid soluble applied from 1911-31 P2 0 5 Lb. 0 914 3,656 p.p.m. 9 18 265 P205 p.p.m. 249 322 721 1 Two gin. of soil extracted with 200 ml. of .002N H2 SO4 buffered pH 3.0 with ammonium sulfate. 2 Two gmi. of soil extracted with 100 ml. of neutral 0.5N NH4 F. 18 ALABAMA AGRICULTURAL EXPERIMENT STATION yields than superphosphate the last 7 years. The total and soluble phosphorus contents of samples taken in June, 1947, are given in Table 9. It is evident from these data that the high rate of rock phosphate caused a greater accumulation of total phosphorus as well as soluble phosphorus than superphosphate. The superphosphate plot contained little more total or soluble phosphorus Lb. per acre of seed F 2000 FIGURE 1. The effect of residual phosphate on the yield of cotton grown in Cullars rotation at Auburn. The superphosphate plots received a total of 914 pounds per acre of PsO5 from 1911 to 1931. The rock phosphate plots received a total of 3,656 pounds per acre of P20 5 during the same period. CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 19 than the check plot. The fact that most of the added phosphorus had been removed from the superphosphate plot by crops and erosion helps to explain the lower yields from superphosphate in the later years of the residual study. EXPERIMENTS OF SHORT DURATION SUPERPHOSPHATE, COLLOIDAL PHOSPHATE, AND BASIC SLAG FOR PEANUTS. A fertilizer experiment with peanuts was con- ducted at 50 locations in 1988 and 1989. The number of loca- tions by soil groups are as follows: Norfolk, 38; Greenville, 10; and Lufkin, 2. The yield data for the sources of phosphorus treatments in this experiment are given in Table 10. The rate TABLE 10. AVERAGE YIELDS AND INCREASES OBTAINED FROM FERTILIZER EXPERIMENTS WITH PEANUTS AT 50 LOCATIONS, 1938-39 Treatment Phosphate applied per acre Lb. 0 300 300 300 150 150 Average yields per acre from 50 locations Marketable Lb. 998 1,167 1,151 1,094 1,180 1,164 Increase Lb. 169 153 96 182 166 None Basic slag Basic slag Colloidal phosphate Superphosphate Superphosphate' 1 Received 25 pounds muriate of potash. of P2 0 5 applied per acre according to source was: superphosphate, 24 pounds; basic slag, 24 pounds; and colloidal phosphate, 54 pounds. Yield increases in the 2-year period averaged 182, 153, and 96 pounds per acre of peanuts, respectively. At 1949 prices for these phosphates, the rate of P2 0 5 used in this experiment would be about the same in cost. Based on these data, it is obvious that superphosphate would be the most economical of the three phosphates for production of peanuts. Basic slag and superphosphate were tested when applied with 25 pounds of muriate of potash. Addition of potash to basic slag increased the yield 16 pounds over basic slag alone, while addition of 25 pounds of potash to superphosphate decreased the yield 16 pounds over superphosphate alone. SUPERPHOSPHATE, ROCK PHOSPHATE, AND BASIC SLAG FOR COTTON AND CORN. Several cooperative tests were carried on from 1920-22 to study the value of three sources of phosphorus for 20 TABLE 11. ALABAMA AGRICULTURAL EXPERIMENT STATION RESPONSE OF COTTON AND CORN TO SOURCES OF PHOSPHORUS, 1920-22 Fertilizers 1 ___________________________acre, Sources of Socspofadded phosphorus Phosphate per acre Cotton yields per Corn yields per acre, Conyldprac, acre, Av.eih Av. eight Av. three locations locations Seed Increase Crn cotton Increase Lb. 885 1,023 969 1,001 Lb. 138 84 116 Bu. 32.5 29.9 33.3 32.3 Bu. -2.6 0.8 -0.2 Lb. 0 None Superphosphate, 16% P 2 01 240 240 Basic slag, 16% P2 0 5 480 Rock phosphate 1All plots received 200 pounds per acre of cottonseed meal and 200 pounds of kainit, which contained 12.5 per cent potash. production of cotton and corn. The data for these tests are given in Table 11. The average increases in yield of seed cotton for the phosphates at eight locations were as follows: superphosphate, 138 pounds; rock phosphate, 116 pounds; and basicslag, 84 pounds. The order of response was probably influenced to a considerable extent by the quantity of P2 0 5 supplied by each source. Basic slag, superphosphate, and rock phosphate were applied at rates equivalent to 38, 38, and 152 pounds of P2 0 5 per acre, respectively. In these tests superphosphate was superior to basic slag and rock phosphate, even though the amount of P2 0 5 in the rockphosphate was four times greater than that of the superphosphate. There was no response of corn to phosphorus. This is in line with data from experiments presented earlier in the report. Since superSUPERPHOSPHATE Vs. BASIC SLAG FOR COTTON. phosphate and basic slag are used in large quantities. in the State, they have been tested at a number of locations for production of cotton. Results obtained from these two materials are given in Table 12. Where, superphosphate and basic slag were used with ammonium sulfate as the source of nitrogen, superphosphate produced 914 pounds per acre of seed cotton and basic slag 852 pounds. These are averages obtained from 222 experiments conducted from 1927 to 1931. The data presented by soil groups show that in most cases superphosphate was superior to basic slag. Presented. also in Table 12 are data from 106 experiments comparing superphosphate and basic slag when used with sodium nitrate as the source of nitrogen. The average yield from super- A *0 m 0 TABLE 12. YIELD OF COTTON IN POUNDS PER ACRE FROM SUPERPHOSPHATE AND BASIC SAG N, z m 045 Source of phosphorus' Clarksville soil group Average yield of seed cotton, pounds per acre, by soil groups Cecil Oktibbeha Greenville Holston Hartsells Decatur soil soil soil soil soil soil group group group group group group COOPERATIVE TESTS, 1927-31 0 Norfolk soil General average group 0 C, 10 expt. Superphosphate' 932 Basic Slag 2 787 7 expt. None3 434 Superphosphate 3 937 3 Basic Slag 751 2Received 32 expt. 804 739 19 expt. 721 899 789 21 expt. 999 970 17 expt. 892 800 COOPERATIVE TESTS, 1926-30 8 expt. 18 expt. 3 expt. 899 914 852 1,112 1,003 951 1,072 904 890 36 expt. 1,036 926 15 expt. 791 796 9 expt. 599 676 709 34 expt. 936 872 19 expt. 815 932 892 57 expt. 891 855 23 expt. 773 928 869 222 expt. 914 852 106 expt. 762 930 859 H -I m On1 m -4 N m 'Received Phosphates applied at rate of 64 pounds P2 0 5 per acre; all plots received 25 pounds K 20 as muriate of potash. ammonium sulfate at rate of 30 pounds per acre of N. sodium nitrate at rate of 30 pounds per acre of N. N. 22 ALABAMA AGRICULTURAL' EXPERIMENT STATION phosphate was 930 pounds per acre of seed cotton as compared with 859 pounds from basic slag. This is a difference of 71 pounds in favor of superphosphate. The difference in yield when the phosphates were used with ammonium sulfate was 62 pounds. These differences indicate that the source of nitrogen had little effect on the efficiency of these phosphates. PHOSPHATES PRODUCED BY TENNESSEE VALLEY AUTHORITY. The Alabama Agricultural Experiment Station has cooperated with the Tennessee Valley Authority in testing new sources of phosphorus under field conditions. The data obtained as a result of this work are reported in Tables 13, 14, and 15. In most cases superphosphate has been used as the standard of comparison. Work reported by Bartholomew (1) in 1985 showed that the availability of several rock phosphates was correlated with their fluorine content. Studies in the laboratories of the United States Department of Agriculture indicated the conditions under which rock phosphate could be defluorinated and the effect of defluorination on availability. Curtis et al (3) and Elmore et al (4) developed a process of defluorination for the purpose of producing a product to be used as a fertilizer. Since the cost of the products depends on the degree of defluorination, it became necessary to determine under field conditions the effect of degree of defluorination on availability to plants. The degree of fineness of the quenched material also is an important factor in connection with availability. The Alabama Station has tested fused tricalcium phosphates of varying degrees of defluorination and fineness, as well as other phosphates that the Tennessee Valley Authority has produced. Data in Tables 18 and 14 show that within limits the availability of fused tricalcium phosphate increases with increasing fineness and decreasing fluorine content. These and other data from states in the Tennessee Valley area indicate that the efficiency of fused tricalcium phosphate does not increase with decreasing fluorine content below 0.4 per cent fluorine or with increasing fineness beyond 40 mesh. Average yields of the superphosphate plots at the three locations listed in Table 13 are as follows: vetch, 8,453 pounds; oats, 45.8 bushels; and sorghum, 44,267 pounds. Average yields of plots treated with fused tricalcium phosphate of less than 0.4 per cent fluorine and finer than 40 mesh are: vetch, 7,099 pounds; oats, 41.8 bushels; and sorghum, 38,917 pounds. The fused tri- TABLE 13. RESULTS OF TESTS WITH FUSED TRICALCIUM PHOSPHATE, CALCIUM METAPHOSPHATE, THREE LOCATIONS, 1941-42 AND SUPERPHOSPHATE AT M Phosphorus' Plot no. Source Alexandria Field Fineness Fluorine all in fused Green weight through rock Vetch Sorghum Oats Mesh -tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium tricalcium phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. 6 10 40 80 6 10 40 80 6 10 40 80 20 35 60 60 Pet. _ 1.29 1.22 1.36 1.30 0.98 1.03 0.92 0.99 0.10 0.10 0.10 0.13 Lb. 7,507 7,384 8,389 8,860 9,247 8,137 8,338 8,621 8,381 8,586 8,910 9,522 9,465 9,627 9,678 10,250 11,076 10,427 Lb. 56,000 55,800 57,500 61,700 66,100 61,600 62,300 59,800 57,100 58,800 58,300 60,800 63,800 63,900 63,000 67,300 69,600 70,500 Bu. 26.3 25.2 32.4 29.3 30.9 29.9 29.4 30.1 29.9 31.8 29.2 34.1 26.9 32.4 30.7 34.8 37.5 36.6 Yields per acre Prattville Field Green weight Vetch Sorghum Lb. 2,820 3,470 3,550 3,640 4,740 3,050 2,760 3,530 4,750 3,150 3,510 5,310 7,470 8,420 10,240 12,560 11,650 11,270 Lb. 28,650 37,150 35,700 35,200 39,450 32,300 30,750 26,850 34,900 31,950 31,600 34,600 33,100 37,950 41,000 43,750 42,950 45,700 Bu. 36.4 40.0 42.1 43.5 41.3 36.7 32.6 35.6 40.3 43.1 45.6 49.4 56.8 58.8 65.4 58.7 58.2 55.1 Monroeville Field Vetch Sorghum Vetch 1941 1941 1942 Lb. 2,070 1,815 1,960 1,805 2,520 1,685 1,585 1,990 2,145 2,440 2,126 2,010 3,560 2,615 2,315 2,775 2,400 2,355 Lb. 17,850 17,900 18,350 18,850 19,150 18,800 17,360 18,300 19,750 19,256 19,800 20,500 20,700 19,000 19,550 19,900 20,500 20,600 Lb. 5,910 4,590 6,110 6,520 7,985 6,015 6,035 6,175 7,075 8,320 8,150 9,745 9,710 10,085 9,235 12,155 11,180 9,760 'o z m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 None Fused Fused Fused Fused Fused Fused Fused Fused Fused Fused Fused Fused ou "C z -4 .4 m an Im N Calcium Calcium Calcium Calcium metaphos. metaphos. metaphos. metaphos.2 - Superphosphate' The following fertilizers applied to vetch and sorghum: 48 pounds P20 5 and 25 pounds K20 to each crop; 36 pounds of nitrogen as sodium nitrate to sorghum; 50 pounds per acre of gypsum applied to all plots except 17 and 18. Fertilizer for oats: 36 pounds of nitrogen as sodium nitrate applied as top dressing in March. 2 No gypsum. W TABLE 14. RESPONSE OF CROPS TO FUSED TRICALCIUM PHOSPHATE, CALCIUM METAPHOSPHATE, AND SUPERPHOSPHATE Phosphorus1 Plot no. Source of phosphorus phosphorus No phosphate Fused Fused Fused Fused tricalcium tricalcium tricalcium tricalcium phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. phos. 0.55 0.55 0.55 0.55 0.36 0.36 0.36 0.36 0.20 0.20 0.20 0.20 Fluorine in fused rock Pet. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 Green weight per acre, 1943 Fineness Mesh ___ - 60+ 80 - 80+100 -100+150 -150 - 60+ 80 - 80+-100 -100+150 -150 - 60+ 80 - 80+100 -100+150 -150 -60+ 60+ 80 80 Prattville Vetch Lb. 11,880 11,430 1.2,360 12,720 12,675 12,420 11,070 10,380 10,930 9,810 9,720 10,280 10,580 11,880 13,410 Monroeville Field Vetch Lb. 14,335 27,130 27,230 28,275 28,290 27,260 26,615 29,165 28,260 28,595 25,855 28,640 29,020 28,605 27,760 Soybeans Lb. 18,721 18,914 19,493 19,879 18,721 16,984 18,914 18,721 16,791 18,142 17,177 19,686 20,265 18,528 19,493 Green weight of sudan grass per acre, 1942 Sylacauga Prattville Monroeville area Field area Lb. 6,570 6,690 6,660 6,885 7,050 7,050 6,990 6,765 6,615 6,495 5,760 6,495 6,435 7,395 6,945 Lb. 5,505 6,548 7,590 7,815 5,873 6,143 6,255 6,533 6,270 6,383 6,495 6,683 5,715 6,135 6,060 Lb. 13,600 14,100 14,300 14,300 14,300 14,750 14,400 14,000 14,400 14,250 14,500 14,500 14,650 15,100 13,500 Av. 3 expt. Lb. 8,558 9,113 9,517 9,667 9,074 9,314 9,215 9,099 9,095 9,043 8,918 9,226 8,933 9,543 8,835 rF a W Fused tricalcium Fused tricalcium Fused tricalcium Fused tricalcium Fused tricalcium Fused tricalcium Fused tricalcium Fused tricalcium a a IC -I C c F" m Calcium metaphosphate Calcium metaphosphate' x m z *o -I -I m 11,730 25,205 19,300 6,690 8,820 14,700 10,070 Superphosphate_ -_ Fertilizers applied to vetch (none to soybeans) as follows: All plots received 48 pounds P 205 (calcium metaphosphate and superphosphate used on available basis and fused rock phosphate on total basis) and 25 pounds K 20 per acre; all plots except plots 15 and 16 received 50 pounds per acre of gypsum. 2 No gypsum. a -I H z CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 25 calcium phosphate plots received 50 pounds per acre of gypsum to eliminate sulfur as a variable. Average yields of the superphosphate plots for the experiments listed in Table 14 are: vetch, 18,467 pounds; and sudan grass, 10,070. For fused tricalcium phosphate containing less than 0.4 per cent fluorine and finer than 60 mesh, the yields are: vetch, 19,287 pounds; and sudan grass, 9,105. According to the data in Table 15, the yields of cotton, corn, oats, winter legumes, and sudan grass were about the same from the use of fused tricalcium phosphate as from triple superphosphate. There was little or no relationship between the fineness of fused tricalcium phosphate and yields within the range of fineness tested in this experiment. The data reported in Tables 13, 14, and 15 show that calcium metaphosphate was equal to superphosphate and triple superphosphate as a source of phosphorous. The data in Table 15 show that potassium metaphosphate resulted in yields of cotton, corn, oats, winter legumes, and sudan grass equal to or slightly higher than those produced by triple superphosphate. TABLE 15. CALCULATED YIELDS' OF CROPS FROM FERTILIZERS CONTAINING PHOSPHATE OF DIFFERENT SOURCES AND FUSED TRICALCIUM PHOSPHATE OF DIFFERENT FINENESSES AT EIGHT LOCATIONS, 1944-46 Yields per acre Seed cotton, 1944 Lb. Triple superphosphate 1,120 Fused tricalcium phos., 6 mesh 5 1,073 Fused tricalcium phos., 20 mesh 5 1,089 Fused tricalcium phos., 20 mesh 1,109 Fused tricalcium phos., 40 mesh 1,096 Fused tricalcium phos., 80 mesh 1,095 Calcium metaphosphate Potassium metaphosphate None 1,118 1,167 1,074 Source of phosphorus 2 Oats 1945 Bu. 34.9 34.7 32.7 37.0 34.2 33.4 33.6 35.6 28.0 Sudan WinterCorn grass hay, legumes 4 Corn, 1945 (green wt.) Lb. Lb. Bu. 3,206 17,145 42.5 3,000 17,465 42.8 3,095 17,693 43.4 3,207 17,853 42.0 3,089 18,438 42.8 3,181 18,559 43.9 3,151 3,120 2,624 17,123 17,891 11,733 41.2 42.5 35.7 on an assumed uniform soil variation between check plots (triple superphosphate plots used as check plots). 2 The crops received the following fertilizers: Cotton, 600 pounds 6-8-4; oats, 300 pounds 0-12-9 under and 36 pounds N about March 1; Sudan grass, 300 pounds 6-8-6 under and 36 pounds about June 20; vetch, 600 pounds 0-8-6; corn, 300 pounds 0-8-6. The sudan grass, vetch, and corn plots received gypsum at the rate of 50 pounds per acre. SAverage yield of seven experiments. Blue lupines used in one test, vetch on all others. Larger size particles screened out; all others fused rock phosphate ground to pass indicated screen. 'Based 26 ALABAMA AGRICULTURAL EXPERIMENT STATION EFFECT OF SUPPLEMENTS ON RESPONSE OF CROPS TO SOURCES OF PHOSPHORUS. Phosphorus in various forms may be equally available, but the materials may give different results when tested for growth of plants. Some phosphorus carriers may contain other essential elements in sufficient quantities to increase yields when applied to soils deficient in these particular elements. The data presented in Tables 16, 17, and 18 show that the efficiency of phosphates often changes as a result of the addition of supplements to the phosphorus carrier. In Table 16 are given the relative efficiencies of several phosphates for cotton when used with ammonium sulfate as the source of nitrogen. The order of efficiencies are: superphosphate, 100; di-calcium phosphate, 99; triple superphosphate, 92; and tricalcium phosphate, 81. The sulfate supplied by the ammonium sulfate should be sufficient to supply the sulfur needed by the crops. It appears, therefore, that the relative response may be explained either on the basis of solubility differences or differences in calcium content. The low efficiency of tricalcium phosphate is probably due to its slow solubility. Phosphorus in TABLE 16. RESPONSE OF COTTON TO SOURCES OF PHOSPHORUS AND SUPPLEMENTS AT 358 LOCATIONS, 1934-38 Fertilizers Yield of Increase Relative over no effiiency no. Source of nitrogen Source of phosphorus and supplement seed cotton phorus sponse to superper acre plots, 100) per acre hos Lb. 982 Superphosphate 1 (NH 4 ) 2S0 4 928 Tri-calcium phosphate 2 (NH 4 ) 2SO 972 3 (NH 4 ) 2 S0 4 Di-calcium phosphate 727 None 4 (NH 4 ) 2SO4 956 Triple superphosphate 5 (NH 4 )2SO 4 2 6 (NH 4 ) 2SO Triple superphos. + dolomite 1,013 Superphosphate 966 7 (NH 4 ) 2SO, 2 997 8 NaNO, Triple superphos. +- dolomite Triple superphosphate 997 9 NaNO 3 728 10 (NH 4 )2 S0 4 None Triple superphos. + gypsum' 1,050 11 NaNO3 2 1,020 12 (NH 4 ) 2SO 4 Superphosphate + dolomite Superphosphate 975 13 (NH 4 )2SO 4 974 Average, plots 1, 7, 13 Lb. 254 200 244 227 285 238 269 269 322 292 247 246 81 99 92 116 ... 109 109 131 119 1 Basis of 600 pounds 6-10-4 per acre. All potash from muriate. Monocalcium phosphate used in 1934 on all triple superhoshate plots. 2 Dolomite (212 pounds per acre) sufficient to correct the acidity from ammonium sulfate. 3 Gypsum (164 pounds per acre) equal to the amount supplied on the superphoshate plots. CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 27 TABLE 17. YIELD PER ACRE OF SEED COTTON FROM THREE SOURCES OF PHOSPHORUS WITH AND WITHOUT DOLOMITE, 1932-35 Fertilizer treatments 1 Av. yield seed cotton per acre by soil groups Av. all Clarks- De- Green- Norsoil 2 199 soil soil soil soil mite expt. per acre group group group group 41 40 34 84 expt. expt. expt. expt. Lb. Lb. Lb. Lb. Lb. Lb. 952 925 Superphosphate 0 800 995 926 931 1,037 928 1,028 993 Superphosphate 212 988 1,044 935 1,025 1,0503 899 Superphosphate Di-ammonium phosphate 0 731 980 868 846 853 Di-ammonium phosphate 212 884 1,013 906 962 947 1,000 966 Di-ammonium phosphate 212 888 1,014 920 0 729 945 872 849 847 Mono-ammonium phosphate 882 1,025 950 978 963 Mono-ammonium phosphate 212 1 Fertilizers applied on basis of 600 pounds per acre of 6-10-4; nitrogen from ammonium sulfate for superphosphate plots and remainder of nitrogen from ammonium sulfate for ammonium phosphate plots; potash from muriate of potash. and applied with NPK. 3 Dolomite broadcast in 1932 and none applied thereafter. Received 50 pounds per acre of gypsum. Dolo- ville soil catur soi ville ol folk si groups Source of phosphorus 'Mixed superphosphate and triple superphosphate is mainly in the form of mono-calcium phosphate. Sixty pounds of P2 0 5 as superphosphate would supply 84 pounds of CaO, while the same amount of P2 0 5 from triple superphosphate would supply only 27 pounds of CaO. When enough dolomite was used with these two phosphates to correct the acidity of the ammonium sulfate, superphosphate produced 1,020 pounds per acre of seed cotton and triple superphosphate 1,013 pounds. Without dolomite superphosphate produced 974 pounds per acre of seed cotton and triple superphosphate 956 pounds. The data in Table 16 show that when triple superphosphate was used with sodium nitrate as the source of nitrogen, it produced 997 pounds per acre of seed cotton. Triple superphosphate and sodium nitrate plus 164 pounds per acre of gypsum resulted in 1,050 pounds, or an increase of 53 pounds for the gypsum. Presented in Table 17 are the average yields of seed cotton from 199 experiments on four soil groups involving use of superphosphate, mono-ammonium phosphate, and di-ammonium phosphate with and without lime. Without lime the following average acre yields of seed cotton were obtained: superphosphate, 925 pounds; di-ammonium phosphate, 853; and mono-ammonium IN TABLE 18. Fertilizer AVERAGE YIEDS OF SEED COTTON FROM COMPLETE FERTILIZERS WITH AND WITHOUT SULFATE Increase 100% r- Average yield of seed cotton per acre treatments' Source of nitrogen 100% urea 100% urea 50 % urea and 50% (NH 4 ) 2S0 4 0 100%1 urea 100% urea Clarksville Decatur Gypsum soil group soil group per acre 41 expt. 76 expt. Lb. Lb. Lb. 0 991 1,019 Holston soil group 27 expt. Lb. 1,339 Hartsells Cecil Greenville Norfolk All soil group soil group soil group soil group soil groups 57 expt. 53 expt. 64 expt. 102 expt. 420 expt. Lb. Lb. Lb. Lb. Lb. 1,048 738 659 664 864 over 420 urea, av. of expt. Lb. 62 77 80 97 0 22.5 45.0 90.0 1,016 1,037 1,048 1,047 1,049 1,055 1,037 1,051 1,357 1,379 1,351 1,396 1,136 1,157 1,180 1,210 776 787 784 812 722 753 762 772 771 778 792 801 926 941 944 961 M c ' Fertilizer applied on basis of 600 pounds per acre of 6-8-6; fertilizers made from triple superphosphate, muriate of potash, nitrogen source as indicated above, and enough dolomite to make mixture nonacid-forming. m x z -I -v m -I -4 0 z CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 29 phosphate, 847. With 212 pounds of lime per acre, the average yields were: superphosphate, 993 pounds; mono-ammonium phosphate, 963; and di-ammonium phosphate, 947. Although the yields from superphosphate were highest in both cases, the addition of lime brought the yields closer together. Ammonium sulfate was used with the phosphates to bring the total nitrogen applied to 36 pounds per acre. Each plot, therefore, received a different quantity of sulfate supplement because of the varying nitrogen content of the phosphorus carriers. The di-ammonium phosphate received 62 pounds of ammonium sulfate as compared with 109 pounds for the mono-ammonium phosphate plots. Sixty-two pounds of ammonium sulfate evidently did not furnish sufficient sulfur for maximum yields, since the addition of 50 pounds of gypsum to di-ammonium phosphate increased the average yield of seed cotton 19 pounds. Di-ammonium and mono-ammonium phosphates produced 966 and 963 pounds of seed cotton, respectively, when each received about the same sulfur supplement. RESPONSE OF COTTON TO GYPSUM. An experiment was started in 1939 to determine the response of cotton to sulfur under a wide range of conditions. This experiment was conducted from 1939 to 19438 at 420 locations involving a number of soil types. The fertilizer was applied at the rate of 600 pounds per acre of 6-8-6 made from triple superphosphate, urea, and muriate of potash. Results of this experiment are reported in Table 18. Addition of 22.5, 45, and 90 pounds per acre of gypsum to the fertilizer mixture increased cotton yields 62, 77, and 80 pounds, respectively. According to these data, ordinary superphosphate applied at 48 to 60 pounds of P205O per acre would supply more than enough sulfur for cotton. RESPONSE OF CLOVER PASTURES TO SOURCES OF PHOSPHORUS The development of a livestock industry in Alabama depends to a large extent on production of an adequate supply of feed. Many pastures will not produce a satisfactory quantity of good quality forage without fertilization. Since phosphorus is likely to be one of the deficient elements, use of the proper source of phosphorus is of considerable importance. A number of cooperative tests have been conducted over the State to evaluate sources of phosphorus for pasture production under a wide range of conditions. Clipping yields were obtained as a measure of the availability of the phosphates used. O TABLE 19. CLIPPING YIELDS FROM PASTURE FERTILIZER EXPERIMENT'S IN NORTHERN ALABAMA, 1944 Treatment every 3 years' Source of phosphorus, 144 pounds P5 peracre None Superphosphate 2 Colloidal phosphate Basic slag Clipping yields per acre of green material Braly farm Limestone County Lb. 4,918 18,502 16,063 20,938 Rosch farm Lauderdale County Lb. 3,959 7,688 6,438 6,943 Average of five locations Lb. 5,368 14,314 10,151> 14,539 Muriate Craig farm Stafford farm Thompson farm of potash Lauderdale Madison Limestone per acre County County County Lb. 0 75 75 75 Lb. 4,376 8,313 9,190 11,563 Lb. 5,105 15,252 7,439 19,876 Lb. 8,481 21,813 11,626 13,375 ' All plots except no fertilizer plots limed to pH 6.5. 2 Received gypsum at rate of 100 pounds per acre. a C rc zm m TABLE 20. CLIPPING YIELDS FROM SIX PASTURE FERTILIZER EXPERIMENTS IN SOUTHEASTERN ALABAMA, 1942-44 Treatments' every 3 years Source of phosphorus Clipping yields Baxley Dees farm, Dees farm, new land, old land, farm, SMuriate Dale Coffee Dale Phosphate of potash County, per acre per acre County, County, 1 2-yr. av. 2-yr. av. yr. result Lb. 0 0 900 1,620 2,000 Lb. 0 150 150 150 150 Lb. 282 563 4,313 2,032 3,282 Lb. 313 500 7,063 4,688 4,625 Lb. 1,469 3,125 8,907 3,063 4,969 per acre of green material Hixon farm, Pike County, 2-yr. av. Lb. 3,375 4,125 12,813 6,313 10,094 Womack Wood farm, Henry farm, County, Houston , County, 1-C yr. results 2-yr. av. Lb. 250 2,750 7,188 5,188 4,375 Lb. 1,500 1,750 11,313 719 6,344 Average of 6 loca tions Lb. 1,198 2,135 8,599 3,667 5,615 x m None None Superphosphate Colloidal phosphate Basic slag m .. 4. -4 z m z SAll plots except no fertilizer plots and basic slag plots received one ton of lime. z CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 31 In Table 19 are given clipping yield data from pasture fertilizer experiments in northern Alabama in 1944. The averages of five locations are as follows: basic slag, 14,539 pounds per acre; superphosphate, 14,314; and colloidal phosphate, 10,151. According to field notes, most of the herbage was grass with very little clover. The phosphates were applied at the rate of 144 pounds per acre of P 20 5, which means that the colloidal phosphate treatment would cost only about half as much as the superphosphate or basic slag. The yield data reported in Table 20 show the response of six clover pastures in southeastern Alabama to superphosphate, colloidal phosphate, and basic slag when applied on an approximate cost basis. The average yields of six locations are 8,599, 5,615, and 3,667 pounds per acre for superphosphate, basic slag, and colloidal phosphate, respectively. The unphosphated plots averaged 2,135 pounds. Thus, 900 pounds of superphosphate was superior to 2,000 pounds of basic slag or 1,620 pounds of colTABLE 21. CLIPPING YIELDS SHOWING RESPONSE OF CLOVER PASTURES TO VARIOUS SOURCES OF PHOSPHORUS, FOUR LOCATIONS treatments Fertilizer acreatments per acre Clipping yields per acre of green material Gulf Sand Vail Coast stountain Moss MFertilizer Farm, Av. of Armfour Substa- Farm, Substation, tion, None LK LK and superphosphate LK and Colloidal phosphate K and basic slag 2-yr. av., 5-yr. av., 1944 and 1943-47 1946 Lb. Lb. 4,625 10,346 10,879 13,746 3,976 9,049 11,211 13,766 Lamison, results, re1941 Lb. 2,403 5,712 2,816 6,316 1-year strong, 1 yr., 194 Lb. 313 3,313 375 locations Lb. 8,370 4,537 Coast Substation: L = 2,000 pounds of lime; K = 100 pounds of muriate of potash; superphosphate = 600 pounds; colloidal phosphate = 1,200 pounds; and basic slag = 2,400 pounds. Phosphorus and potash applied annually. Moss Farm: L = 4,000 pounds of lime; K = 150 pounds of muriate of potash; superphosphate -= 900 pounds; colloidal phosphate -= 1,620 pounds; and basic slag = 2,000 pounds. Phoshorus and potash repeated every 3 years. Sand Mountain Substation: L =4,000 pounds of Lime; K =75 pounds of muriate of potash; superphosphate = 600 pounds; colloidal phosphate = 516 pounds; and basic slag = 2,000 pounds. Phosphorus and potash applied every 3 years. Vail Farm: L = 4,000 pounds of lime; K = 150 pounds of muriate of potash; superphosphate = 900 pounds; and colloidal phosphate = 1,620 pounds. Phosphorus and potash applied every 3 years. 'Gulf 32 ALABAMA AGRICULTURAL EXPERIMENT STATION TABLE 22. CLIPPING YIELDS SHOWING RESPONSE OF CLOVER PASTURES VARIOUS SOURCES OF PHOSPHORUS AND TO RATES OF COLLOIDAL PHOSPHATE, FOUR LOCATIONS TO Fertilizer treatment 1 per acre Clipping yields per acre of green material Upper GulfMCoastal Av. of Piedmont tal Coast McCreary loo Sub, station, Farm , Substation, B-rw station,av.,2yr., 1 tions 1947 2-yr. av., av., -yr.av2-yr. 1946 1946-47 1946 and 1948 Lb. Lb. 8,265 3,185 5,067 8,432 Lb. 1,776 12,607 6,109 7,271 9,561 Lb. 2,097 6,757 1,631 7,223 Lb. 9,380 3,628 8,025 None LK and superphosphate LK and colloidal phosphate LK and double rate colloidal phosphate K and basic slag 9,890 3,587 2,603 6,886 1Gulf Coast Substation and McCreary Farm: L = 2,000 pounds of lime; K = 150 pounds of muriate of potash annually; superphosphate = 600 pounds; colloidal phosphate = 1,000 pounds; double rate of colloidal phosphate = 2,000 pounds; and basic slag = 1,000 pounds. Phosphorus applied every 2 years. Upper Coastal Plain Substation: L = 3,000 uounds of lime; K = 100 pounds of muriate of potash; superphosphate = 800 pounds; colloidal phosphate = 1,500 pounds; double rate of colloidal phosphate = 3,000 pounds; and basic slag = 1,500 pounds. Phosphorus and potash applied every 2 years. Piedmont Substation: L = 2,000 pounds of lime; K= 100 pounds of muriate of potash; superphosphate = 800 pounds; colloidal phosphate 1,500 pounds; and basic slag = 1,500 pounds. Phosphorus and potash applied every 2 years. loidal phosphate. The basic slag plots did not receive lime in these experiments, and the low yield from basic slag may have been due to inadequate lime. In Table 21 are given the results from four other locations in the State, showing response of clover pastures to sources of phosphorus. Superphosphate produced an average 8,370 pounds of green material as compared with 4,537 pounds from colloidal phosphate. The two phosphates were applied on an approxi- mately equal cost basis except at the Sand Mountain Substation. There was not much difference between the yields from superphosphate and colloidal phosphate on the Moss farm. This. is to be expected, since the plot getting no phosphate yielded rather high. At the three locations where basic slag was used, it produced 9,710 pounds of herbage as compared with 10,112 from superphosphate. The clipping yields in Table 22 show the response of clover CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 33 pastures to three sources of phosphorus and to two rates of colloidal phosphate. The averages of four locations show clipping yields of 9,380, 8,025, and 3,628 pounds for superphosphate, basic slag, and colloidal phosphate, respectively, when applied on an approximately equal cost basis. At three locations a double rate of colloidal phosphate also was tested. The average yields for these locations are 10,321, 4,294, and 4,980 pounds per acre for superphosphate, single rate of colloidal phosphate, and double rate of colloidal phosphate, respectively. A rate of colloidal phosphate equivalent to twice the cost of superphosphate produced only half as much herbage as superphosphate. In the fall of 1947, two tests were established on lime land at the Black Belt Substation. One was designed to test the value of several sources of phosphorus and the other was designed to test the residual value of some phosphates. The two experiments were located within a few hundred feet of each other. The results of these tests are presented in Tables 23 and 24. Using 80 pounds of P 20 5, the relative increases in yield from the phosphates listed in Table 23 are: superphosphate, 100; basic slag, 103; triple superphosphate, 94; calcium metaphosphate, 85; colloidal phosphate, 8; and rock phosphate, 1. Doubling the rate of colloidal phosphate had no appreciable effect on yields. The yield data reported in Table 24 show that 144 pounds of P20 5 from superphosphate increased the yield of herbage 8,185 pounds but that colloidal phosphate applications ranging from 288 to 1,152 pounds of P 20 5 per acre had little or no effect on yields. The increased yield from 576 pounds of P20 5 applied as rock TABLE 23. RESPONSE OF CLOVER TO VARIOUS SOURCES OF PHOSPHORUS ON LIME LAND, BLACK BELT SUBSTATION Relative Fertilizer treatment' Source of phosphorus Green weight Increased yield due to per acre, phosphorus 1948 P2 O applied per acre Lb. None Superphosphate Triple superphosphate Calcium metaphosphate Basic slag Rock phosphate Colloidal phosphate Colloidal phosphate 0 80 80 80 80 80 80 160 Lb. 480 9,510 8,935 8,135 9,805 585 1,175 300 Lb. 9,030 8,455 7,655 9,325 105 695 -180 100 94 85 103 1 8 -2 efficiency (superphosphate 100) All plots received 60 pounds of KO from muriate of potash. 20 34 TABLE 24. ALABAMA AGRICULTURAL EXPERIMENT STATION RESPONSE OF CLOVER TO HIGH RATES OF COLLOIDAL AND ROCK PHOSPHATES ON LIME LAND, BLACK BELT SUBSTATION Fertilizer treatmentRelative Fertilizer treatment1 Source of phosphorus Green weight Increased per acre, yield due to phosphorus 1948 P 2O5 applied per acre Lb. 0 144 288 576 1,152 576 Lb. 1,595 9,780 2,865 1,730 2,260 2,345 Lb. 8,185 1,270 135 665 750 efficiency (superphosphate 100) None Superphosphate Colloidal phosphate Colloidal phosphate Colloidal phosphate Rock phosphate 2 years. 'Al 100 16 2 8 9 plots to receive 90 pounds of K20 from muriate of potash every phosphate was slight. One year's results on lime land show that colloidal and rock phosphates are inferior to the processed phosphates for the growth of clover pastures. DISCUSSION Results of studies presented herein show that different phosphates vary as to their efficiency for crop production. Relative values of these materials vary somewhat depending upon soil conditions and the crop grown. Some of the raw phosphates, such as rock and colloidal, were not satisfactory sources of phosphorus for the crops studied, but were relatively more efficient for cotton than they were for pastures. Certain soil conditions may affect the response obtained from phosphates. The more insoluble phosphates are usually better sources of phosphorus for acid soils than for alkaline soils. Ammonium phosphates give relatively better response when applied to soils containing considerable calcium than when applied to light sandy soils low in calcium. This was especially evident in the experiments of long duration. The lack of sulfur may be another reason for the low response obtained from ammonium phosphates. Since a number of variables affect the efficiency of phosphates, it is impossible to give ratings that will apply under all conditions. However, certain phosphates give consistently high response, while others are consistently low. For example, results of these studies show that the superphosphate plots always ranked at the top or near the top in yields. It appears that superphosphate would be a satisfactory source of phosphorus under a wide range of conditions. CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 35 From a practical standpoint, the best source of phosphorous for the farmer to use is the one that gives the greatest returns over cost of material. A cheap source of phosphorus may be 80 per cent as efficient for cotton as another and yet give a lower return over cost of material. At present the cost of an average application of superphosphate (48 pounds P20 5) would be about $1.60 more than the cheapest source of phosphorus. It is evident, therefore, that the efficiency of the cheap source must be very near that of superphosphate to be considered a satisfactory source of phosphorus. It is often suggested that the cheaper insoluble phosphates may have a greater cumulative effect or residual value than processed phosphates. Results from experiments of long duration do not indicate any appreciable cumulative effect. The data reported herein are in agreement with those obtained by other experiment stations. Salter and Barnes (8) working with Ohio soils found that, for both grain crops and clover on unlimed land, rock phosphate was 40 per cent as efficient as superphosphate. Under the same conditions basic slag showed an efficiency of 85 per cent for cereals and 140 per cent for clover. Efficiency of rock phosphate and basic slag decreased with liming. Noll and Irvin (7) studied the response of a number of crops to sources of phosphorus when grown on a Hagerstown silt loam in Pennsylvania. They found that on unlimed land with manure, basic slag and superphosphate yielded nearly alike and exceeded the yields from rock phosphate applied at twice the equivalent rate of P2 0 5 . On limed land, rock phosphate gave lower yields than superphosphate when applied at two and three times the equivalent rate of P20 5. Indiana (12) results show that 715 pounds of superphosphate and 1,000 pounds of rock phosphate ($5.00 worth in each case) produced average annual crop increases of $5.19 for superphosphate and $2.80 for rock phosphate when each was used alone. When nitrogen and potash were applied, the increase was $8.11 for superphosphate and $3.61 for rock phosphate. Results reported by Bauer et al (2) of the Illinois Station show that on limed land at five locations superphosphate produced greater increases in yield on the average than rock phosphate. On unlimed land at the same location, rock phosphate was more effective than superphosphate. In 1929, the Carthage and Lebanon fields were modified for the purpose of comparing superphosphate and rock phosphate. Rock phosphate was applied 36 ALABAMA AGRICULTURAL EXPERIMENT STATION at the rate of 400 pounds in the drill with wheat, and 125 pounds was hill-dropped with corn. Superphosphate was applied at the rate of 200 pounds to wheat and 125 pounds to corn. The data at these two locations show that superphosphate produced larger increases at first but that the results have been somewhat similar in recent years. SUMMARY A number of studies have been conducted over a period of years by the Alabama Agricultural Experiment Station to test the relative value of different sources of phosphorus. The data obtained from these studies are reported in this publication. The relative yield values for the various phosphates given below are based on the increased yield from superphosphate as 100. Results of the experiments of long duration may be summarized as follows: 1. Triple superphosphate showed relative yield increases of 91, 102, and 89 for cotton, corn, and winter legumes, respectively. 2. Using equivalent quantities of P2 0 5 (48 pounds) basic slag showed the following efficiencies: cotton, 85; corn, 114; and winter legumes, 94. 3. Relative values of yields from rock phosphate (48 pounds P2 0 5) on cotton, corn, and winter legumes were 54, 67, and 51, respectively. When used at double the P2 0 5 rate of superphosphate, the relative yield values of rock phosphate on cotton, corn, and legumes were 57, 43, and 61, respectively. Use of four times as much phosphorus from rock phosphate as from superphosphate did not increase the yields of cotton and corn as much as superphosphate. 4. The relative yield values of colloidal phosphate ranged from 47 for legumes to 65 for corn. 5. The relative yield values of tricalcium phosphate, which were similar to those of basic slag, were 87 for cotton, 109 for corn, and 89 for legumes. 6. Ammo-phos A was not a satisfactory material for cotton, corn, or winter legumes. On sandy soils it reduced the yield of cotton below check yields after several years of continued application. The relative efficiencies of Ammo-phos A were 26, 57, and 73 for cotton, corn, and legumes, respectively. CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 37 The following is a summary of the results of experiments of short duration: 1. Fertilizer experiments conducted at 50 locations show the following increases in yield of peanuts: superphosphate, 182 pounds per acre; basic slag, 153; and colloidal phosphate, 96. 2. Results of experiments conducted at eight locations show that superphosphate, rock phosphate, and basic slag increased the yield of seed cotton 138, 116, and 84 pounds per acre, respectively. Superphosphate and basic slag were applied at the rate of 240 pounds per acre, while rock phosphate was aplied at the rate of 480 pounds. 3. An average of 222 experiments shows that where ammonium sulfate was used as the source of nitrogen, 914 pounds of seed cotton was obtained from superphosphate, while the same amount of P2 0 5 from slag yielded 852 pounds. Where sodium nitrate was used as the source of nitrogen, 930 pounds was obtained from superphosphate and 859 pounds from slag. 4. Fused tricalcium phosphate of 40-mesh fineness or less and of less than 0.4 per cent fluorine content produced relative yield increases of 84, 72, 40, and 34 for vetch, oats, sorghum, and sudan grass, respectively. In another series of tests using triple superphosphate as the standard, fused tricalcium phosphate produced practically the same yields of cotton, corn, oats, winter legumes, and sudan grass as triple superphosphate. 5. Calcium and potassium metaphosphates were as effective as the superphosphate and triple superphosphate in increasing yields of most of the crops tested. 6. Results of 858 tests in which ammonium sulfate was used as the source of nitrogen show the following relative yield values of seed cotton, with that of superphosphate representing 100: triple superphosphate, 92; di-calcium phosphate, 99; and tricalcium phosphate, 81. The use of enough dolomite to correct acidity from ammonium sulfate gave relative yields of 119 for superphosphate and 109 for triple superphosphate. The addition of gypsum to triple superphosphate increased the yield of seed cotton 58 pounds. 7. Mono-ammonium and di-ammonium phosphates were about as satisfactory sources of phosphorus for cotton as superphosphate when used with dolomite. The use of gypsum and dolomite with di-ammonium phosphate increased the yield of seed cotton 19 pounds over dolomite alone. 38 ALABAMA AGRICULTURAL EXPERIMENT STATION 8. Many Alabama soils responded to sulfur. An average of 420 experiments shows that 22.5, 45, and 90 pounds per acre of gypsum increased the yield of seed cotton 62, 77, and 80 pounds per acre respectively. Therefore, the response of certain crops to superphosphate may be due partially to the sulfate supplied by the phosphate. The results from pasture experiments dealing with the value of various sources of phosphorus are briefly summarized: 1. Clipping yields from eight experiments on acid land show that colloidal phosphate produced less than half as much herbage as superphosphate when the two phosphates were applied on an equivalent cost basis. 2. In three experiments on acid land, colloidal phosphate applied at double the cost of superphosphate produced only i5per cent as much herbage as superphosphate. 3. One year's results on lime land of the Black Belt showed that colloidal and rock phosphates produced little more herbage than the plots getting no phosphate. The plots receiving superphosphate produced a satisfactory growth of clover. 4. On acid land basic slag produced 88 per cent as much herbage as superphosphate. The slag plots were not limed; this may account for the lower response of slag in some cases. CROP RESPONSE to VARIOUS PHOSPHATE FERTILIZERS 39 LITERATURE CITED (1) BARTHOLOMEW, R. P. Fluorine, its effect on plant growth and its relation to the availability to plants of phosphorus in phosphate rocks. Soil Sci. 40: 203-217. 1935. F. C., LANG, A. L., BADGER, C. J., MILLER, L. B., FARNHAM, C. H., JOHNSON, P. E., MARRIOTT, L. F., AND NELSON, M. H. The BAUER, (2) effects of soil treatment on soil productivity. Bul. 516. 1945. (3) Ill. Agr. Expt. Sta. CURTIs, H. A., COPSON, R. L., BROWN, E. H., AND POLE, G. R. Fer- tilizer from rock phosphate. 1987. (4) Indus. and Engin. Chem. 29: 766-70. Defluorination ELMORE, K. L., HUFFMAN, E. O., AND WOLF, W. W. of phosphate rock in the molten state. Indus. and Engin. Chem. 34: 40-48. 1942. (5) ENSMINGER, L. E., AND COPE, J. T., JR. Effect of soil reaction on the efficiency of various phosphates for cotton and on loss of phosphorus by erosion. Jour. Amer. Soc. Agron. 39: 1-11. 1947. MEHRING, A. L., WALLACE, H. M., AND DRAIN, MILDRED. Nitrogen, phosphoric acid and potash consumption in the United States, by years and by states, with preliminary figures for 1944. Jour. Amer. Soc. Agron. 37: 595-609. 1945. NOLL, C. F., AND IRVIN, C. J. Field test of phosphate fertilizers. Penn. Agr. Expt. Sta. Bul. 428. 1942. SALTER, R. M., AND BARNES, E. E. The efficiency of soil and fertilizer phosphorus as affected by soil reaction. Ohio Agr. Expt. Sta. Bul. 53388. 1985. SCARSETH, GEORGE D., AND CHANDLER, W. V. Losses of phosphate from a light-textured soil in Alabama and its relation to some aspects (6) (7) (8) (9) of soil conservation. Jour. Amer. Soc. Agron. 80: 361-374. 1988. Fer- (10) (11) OSWALD, MERZ, ALBERT R., AND BROWN, B. E. tilizer materials. U.S.D.A. Yearbook 514. 1988. SCHREINER, VOLK, GARTH W. Response to residual phosphorus of cotton in continuous culture. Jour. Amer. Soc. Agron. 87: 880-840. 1945. WIANCKO, A. G., AND CONNER, S. D. Acid phosphate vs. raw rock (12) phosphate as fertilizer. Ind. Agr. Expt. Sta. Bul. 187. 1916.