R", Q01”; {a CAFWFJUS. E89-6M-L18O TEXAS AEEICI/L Tl/EAL EXPERIMENT STAHDN A. B. CONNER, DIRECTOR _ COLLEGE STATION, BRAZOS COUNTY. TEXAS BULLETIN NO. 490 MARCH, 1934 j Alriculturam i’: R A R Y DIVISION OF AGR 3l8li0n, 1am The Effect of Time and Rate 0f Application of Nitrate of Soda 0n the Yield of Cotton AGRICULTURAL AND MECHANICAL COLLEGE OF TEXAS T. O. WALTON, President hanica! Calla ‘ l? U’ T31a$ STATION STAFFT Administration : A. B. Conner, M. S., Director R, E. Karper, M. S., Vice Director Clarice Mixson, B. A., Secretary M. P. Holleman, Chief Clerk J. K. Francklow, Asst. Chief Clerk Chester Higgs, Executive Assistant Howard Berry, B. S., Technical Asst. Chemistry: G. S. Fraps, Ph. D., Chief; State Chemist S. E. Asbury, M. S., Chemist J. F. Fudge, Ph. D., Chemist E, C. Carlyle, M. S., Asst. Chemist T. L. Ogier, B. S., Asst. Chemist A. J. Sterges, M. S., Asst. Chemist Ray Treichler, M. S., Asst. Chemist W. H. Walker. Asst. Chemist Velma Graham, Asst. Chemist Jeanne F. DeMottier, Asst. Chemist R. L. Schwartz, B. S., Asst. Chemist C. M. Pounders, B. S., Asst. Chemist Horticulture : S. H, Yarnell, Sc. D., Chief Range Animal Husbandry: J. M. Jones, A. M. Chief B. L. Warwick, Ph. D., Breeding Investiga. S, P. Davis, Wool Grader J. H. Jones, B. S., Animal Husb. Entomology: F. L. Thomas, Ph.D., Chief; State Entomologist H. J, Reinhard, B.S., Entomologist R. K. Fletcher, Ph. D., Entomologist W. L. Owen, Jr., M. S., Entomologist J. N. Roney, M. S., Entomologist J. C. Gaines, Jr., M. S., Entomologist S. E. Jones, M. S., Entomologist F. F. Bibby, B. S., Entomologist **E. W. Dunnam, Ph. D., Entomologist **R. W’. Moreland, B. S., Asst. Entomologist C, E, Heard, B. S., Chief Inspector C. J. Burgin, B. S., Foulbrood Inspector Veterinary Science: “Vi. Francis, D. V. M., Chief H. Schmidt, D. V. M., Veterinarian **F. P. Mathews, D. V. M., M. S., Veterinarian’ J. B. Mims, D. V. M., Asst. Veterinarian Plant Pathology and Physiology: J. J. Taubenhaus, Ph. D., Chief W. N. Ezekiel, Ph. D., Plant Pathologist Farm and Ranch Economics: L. P. Gabbard, M. S., Chief W, E. Paulson, Ph. D., Marketing C. A. Bonnen, M. S., Farm Management ;t**W. R. Nisbet, B. S., Ranch Management **A. C. Magee, M. S., Ranch Management Rural Home Research: Jessie Whitacre, Ph. D., Chief Mary Anna Grimes, M. S., Textiles Sylvia Cover, Ph.D., Foods ' Soil Survey: **W. T. Carter, B. S., Chief E. H. Templin, B. S., Soil Surveyor A. H. Bean, B. S., Soil Surveyor R, M. Marshall, B. S., Soil Surveyor Botany: Y: ' z f ww-fln, V. L. Cory, M. S., Acting Chief Swine Husbandry: Fred Hale, M. S., Chief Dairy Husbandry: O. C. ‘Copeland. M. S., Dairy Husbandman Poultry Husbandry: R, M. Sherwood, M. S., Chief J. R. Couch, B. s., Asst. Poultry Husbandmau Agricultural Engineering: H. P. Smith, M. S., Chief Main Station Farm: G. T, McNess, Superintendent Apiculture (San Antonio): H. B, Parks, B. S., Chief A. H. Alex, B. S., Queen Breeder Feed Control Service: F. D, Fuller, M. S., Chief James Sullivan, Asst. Chief Agronomy: S. D. Pearce, Secretary E. B. Reynolds, Ph. D., Chief J. H. Rogers, Feed Inspector R. E. Karper, M. S., Agronomist K. L. Kirkland, B. S., Feed Inspector P. C. Mangelsdorf, Sc. D., Agronomist S, D. Reynolds, Jr., Feed Inspector D. T. Killough, M. S., Agronomist P. A. Moore, Feed Inspector Publications: E. J. Wilson, B. S., Feed Inspector A, D. Jackson, Chief H. G. Wickes, D. V. M., Feed Inspector SUBSTATION S No. 1, Beeville, Bee County: R. A. Hall, B. S., Superintendent No. 2, Lindale, Smith County: P, R. Johnson, M. S., Superintendent "B. H. Hendrickson, B. S., Sci. in Soil Erosion **R. W. Baird, M. S., Assoc. Agr, Engineer No. 3, Angleton, Brazoria County: R. H. Stansel, M. S., Superintendent H. M. Reed, M. S., Horticulturist No. 4, Beaumont, Jefferson County: R. H. Wyche, B. S., Superintendent **H. M. Beachell, B. S., Junior Agronomist No. 5, Temple, Bell County: Henry Dunlavy, M. S., Superintendent C. H. Rogers, Ph. D., Plant Pathologist H. E. Rea, B. S., Agronomist S. E. Wolff, M. S., Botanist **E. B. Deeter, B. S., Soil Erosion "P. L. Hopkins, B. S., Junior Civil Engineer No. 6, Denton, Denton County: P. B. Dunkle, B. S., Superintendent **I. M. Atkins, B, S., Junior Agronomist No. 7, Spur, Dickens County: R. E. Dickson, B. S., Superintendent B. C. Langley, M. S., Agronomist No. 8, Lubbock, Lubbock County: ' D. L. Jones, Superintendent Frank Gaines, Irrig. and Forest Nurs. Members of Teaching Staff Carrying G. W. Adriance, Ph. D., Horticulture S. W, Bilsing, Ph. D., Entomology D. Scoates, A. E., Agricultural Engineering A. K. Mackey, M. S., Animal Husbandry R. G. Reeves, Ph.D., Biology *Dean, School of Veterinary Medicine. No. 9, Balmorhea, Reeves County: J. J, Bayles, B. S., Superintendent No. 10, College Station, Brazos County: R. M. Sherwood, M. S., In Charge L. J. McCall, Farm Superintendent No. 11, Nacogdoches, Nacogdoches County: H. F, Morris, M. S., Superintendent **No. 12, Chillicothe, Hardeman County: **J. R. Quinby, B. S., Superintendent "‘*J. C. Stephens, M. A., Asst. Agronomist No. 14, Sonora, Sutton-Edwards Counties: W. H, Dameron, B. S., Superintendent I. B. Boughton, D. V. M., Veterinarian W. T. Hardy, D. V. M., Veterinarian O. L. Carpenter, Shepherd “O. G. Babcock, B. S., Asst. Entomologist No. 15, Weslaco, Hidalgo County: W. H. Friend, B. S., Superintendent S. W, Clark, B. S., Entomologist W. J. Bach, M. S., Plant Pathologist J. F. .Wood, B. S., Horticulturist No. 16, Iowa Park, Wichita County: C. H. McDowell, B. S., Superintendent L. E. Brooks, B. S., Horticulturist No. 19, Winterhaven, Dimmit County: E. Mortensen, B. S., Superintendent **L. R, Hawthorn, M. S., Horticulturist Cooperative Projects on the Station: J. S. Mogford, M. S., Agronomy F. R. Brison, M. S., Horticulture W. R. Horlacher, Ph.D., Genetics J. H. Knox, M. S., Animal Husbandry A. L. Darnell, M. A., Dairy Husbandry TAs of March 1, 1984‘ "In cooperation with U. S. Department of Agriculture. $In cooperation with Texas Extension Service. Fertilizers increased the yield 0f cotton by producing larger plants which provided more space for more bolls and larger bolls and not by increasing the length of lint or percentage 0f lint, as shown in experiments at Nacogdoches and Troup, during four years ending 1930, to determine the best time and rate of applica- tion of nitrate of soda for cotton. The application of 200 pounds of nitrate of soda furnished the optimum amount of nitrogen for cotton at Troup and 100 pounds gave the best results at Nacogdoches. These amounts of nitrate of soda (with superphosphate and muriate of potash) are roughly equal to 400 pounds of 8-12-4 and 4-12-4 fertilizer, respectively. Side dressings of nitrate of soda on the Ruston and Orangeburg soils at Nacogdoches made an average yield of 203 pounds of lint per acre, which was 24 pounds, or 13.4 per cent, more than the yield where all of the nitrogen was applied before planting. However, on the Kirvin fine sandy loam at Troup, applications of nitrate of soda before planting produced slightly larger yields _ of cotton than side dressings. These differences in yield are probably caused by differences in the nature of the subsoils, the Ruston and Orangeburg soils having open friable clay subsoils and the Kirvin soil a less permeable subsoil. These results indi- cate that the practice of side dressing is satisfactory on soils with friable, permeable subsoils, especially where large amounts of nitrate of soda are used. On soils with less permeable subsoils all of the nitrogen may be applied before planting. All of the fertilizers used increased yield, size of boll, number of _ ‘bolls per plant, percentage of 5-lock bolls, size of plant, number of fruiting branches, and earliness. Fertilizers, however, did not increase the length of lint, nor the percentage of lint, nor did they reduce the amount of shedding. CONTENTS Page Introduction _______________________________________________________________________________________________________ _o 5 Plan of the Experiment _______________________________________ 1 5 Description of Soils Used ................................... 5 Varieties of Cotton Used ........................................................................ __ 6 Size of Plat 6 Fertilizer Treatments _ 6 Effect of Fertilizer on Yield of Cotton, _________________ 1 7 Rates of Application of Nitrate of Sodao_____i.,i_1_,._o._ , 7 Side Dressings of" Nitrate of Soda '7 __ 9 Comparison of Different Nitrogenous Fertilizers ,,,, 1 111 Effect of Fertilizers on the Character of Lint,.i.____..,_ . 12 Length of Lint _ . . . _ . A . A _ . . . . . _ . . . _ _ _ _ _ _ _ , _ _ . _, 112 Percentage of Lint ............. s12 Size of Boll ____________________________________________________________________________________________ 113 Effect of Fertilizers on Size of Plants..,.__.i-. Height of Plants ______________ .. _, _____________________ .15 Number of Fruiting Branches per Plant 1 a .1115 Number of Bolls per Plant ________________________________________ Percentage of 5-Lock Bolls _________________ r’ 116 Effect of Fertilizers on Shedding . _ _ . _ _ i o _ , , , _, 17 Earliness __________________________________ -1 18 Summary and Conclusions _ 19 BULLETIN NO. 490 MARCH, 1934 THE EFFECT OF TIME AND RATE OF APPLICATION OF NITRATE OF SODA ON THE YIELD OF COTTON E. B. REYNOLDS, P. R. JOHNSON, AND B. C. LANGLEY The main object of the fertilizerwork reported in this Bulletin was (1) to determine the most profitable amounts of nitrate of soda in combina- tion with suitable amounts of phosphoric acid and potash for cotton and (2) to study the effect of time and rate of application of nitrate of soda on yield, length and percentage of lint, size of boll, shedding, and other characters of the cotton plant on the sandy soils of eastern Texas, as a basis for developing a more intelligent and profitable fertilizer practice for cotton in the region‘ More fertilizer is used for cotton on the sandy soils in eastern and northeastern Texas than in any other section of the State. Many farmers in the region follow the practice of applying a complete fertilizer on the land before the cotton is planted and then applying a readily available nitrogenous fertilizer, such as nitrate of soda or sulphate of ammonia, as a side dressing soon after the cotton has been thinned to a proper stand. The practice of side dressing is based on the assumption (1) that part of the readily available nitrogen will. be washed out of the soil by rains before it can be taken up by the young growing plants and (2) that if part of the nitrogen is applied to the soil after the plant has made some growth and can take up plant food more rapidly, extensive losses of nitrogen by leaching will be avoided. The losses of nitrogen by leaching are influenced largely by the kind of soil; the amount, intensity, and distribution of rainfall; and kind and age of plant. The work reported here was planned to secure information that could be used in developing a better fertilizer practice, especially as related to best time of applying nitrogenous fertilizer as side dressings to cotton on different kinds of soil. . PLAN OF EXPERIMENT Description of Soils Used The fertilizer work was conducted on Kirvin fine sandy loam at Sub- station No. 2, Troup’, and on land consisting of Ruston and Orangeburg fine sandyloams at Substation No. 11, Nacogdoches. The Kirvin soils have grayish-brown or reddish-brown surface soils underlain by a heavy, red clay subsoil. The surface relief ranges from undulating to rolling or even hilly in places. These soils occur extensively in northeastern Texas and extend into Arkansas and Louisiana. The ‘ Kirvin soils are moderately productive and respond well to good methods of 1This work with nitrate of soda was made possible through a fellowship established at the Agricultural Experiment Station, A. & M. College of Texas, by the Educational Bureau, Chilean Nitrate of Soda, New York, N. Y. "Substation No. 2 was removed from Troup to Tyler, Smith County, in 1981. £;~ 6 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION improvement. They are well suited to the general farm crops, fruits and vegetables grown in the region. The Ruston and Orangeburg soils have grayish-brown or light-browr surface soils. The Ruston soils have friable, crumbly yellowish-red oi reddish-yellow sandy clay subsoils. The subsoils of the Orangeburg soils are similar, except that they are deeper red in color. These soils are not so extensive as the Kirvin soils. They are moderately productive respond well to good soil-improving practices, and are well adapted te the general farm crops, fruits, and vegetables. Varieties of Cotton Used The Mebane variety of cotton was grown in the experiment at Trou; and the Acala variety at Nacogdoches because the results of the variet) tests of cotton have shown that these varieties are well adapted te these parts of the State. The time and method of preparing the land planting, and cultivating which have been found best by experience were adhered to in this work. The cotton was thinned to a stand of one plant every 12 inches in the row when the plants were four to six inches high. The test was conducted on the same land each year at both Nacogdo- ches and Troup, the cotton being grown continuously without a winter covei crop. The several treatments occurred on the same plats each year. Size of Plat The plats consisted of four rows 3 feet wide and 132 feet long, compris ing an area 2/55 of an acre in size. The two inside rows of eacl plat were used as the test part of the plat. The two outside row: served as guard rows to prevent the fertilizer treatment on one plai from influencing the yield on adjoining plats and were dise \rded a1 picking time. Fertilizer Treatments There were two series of fertilizer treatments. In one series an e11 the fertilizer was applied about two weeks before the cotton was planted In the other series, all of the phosphoric acid and potash and one-half oi the nitrogen was applied about two Weeks before planting and the othei half of nitrogen was applied as a side dressing immediately after the cotton was thinned to the desired stand. Sixteen per cent superphos- phate or equivalent amounts of higher grades of superphosphate were used in this work. The following treatments were included in the experi- ment at both Nacogdoches and Troup: Fertilizer treatment Approximately equal to Pounds Material 400 pounds of- No treatment ..................................................................... ...... .. Check 10o nitrate of soda .............................................................. ............ .. 4-12-0 300 superphosphate THE EFFECT OF NITRATE OF SODA ON YIELD OF COTTON 7 100 nitrate of soda 300 superphosphate ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, e- 4-12-4 30 muriate of potash J 80 sulphate of ammonia] 300 superphosphate ‘g ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, V, 4-12-4 30 muriate 0f potash J 50 nitrate of soda‘ 100 cottonseed meal 300 superphosphate 4-12-4 30 muriate of potash 67 nitrate of soda 70 cottonseed meal 300 superphosphate - g t, V, g 4-12-4 30 muriate of potash 150 nitrate of soda 1 300 superphosphate 1 6-12-4 30 muriate of potash J 200 nitrate of soda 1 300 superphosphate 1 _ , V , _ _ . . _ , , . _, 8-12-4 30 muriate of potash J 160 I sulphate of ammonia 300 superphosphate . _ .... ., 8-12-4 30 muriate of potash J 250 nitrate of soda 1 300 superphosphate t. ., H 10-12-4 30 muriate of potash J 300 nitrate of soda 300 superphosphate ........ .. 12-12-4 30 muriate of potash 300 nitrate of soda 1 300 superphosphate ................... .. , ........................................................... .. 12-12-8 60 muriate of potash 300 nitrate of soda 450 superphosphate ., ...................................................... l. 8-12-4‘ 45 muriate of potash J *This analysis was used at the rate of 600 pounds per acre. EFFECT OF FERTILIZER ON YIELD OF COTTON Yields were obtained from cotton that received (1) different rates of. application of nitrate of soda, (2) side dressings of nitrate of soda, and (3) different nitrogenous fertilizers, at Nacogdoches and Troup, Texas. Rates of Application of Nitrate of Soda Nitrate of soda was used at the rates of 100, 150, 200, 250, and 300 pounds per acre in combination with superphosphate and muriate of potash. The several combinations of these materials are given in the :'irst three‘ columns of the tables and the approximate equivalents of ;hese treatments, expressed as fertilizer analyses, are given in column :'our. This means, for example, that the treatment consisting of 100 pounds >f nitrate of soda, 300 pounds of superphosphate, and 30 pounds of muriate >f potash is approximately equal to 400 pounds of a 4-12-4 fertilizer. Results Obtained at Troup: The average yield of cotton for the four rears 1927-1930 gradually increased from 206 pounds of lint per acre from ahe use of 100 pounds of nitrate of soda to 228 pounds of lint where 300 8 BULLETIN -NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION pounds of nitrate of soda was used (Table 1). The first increment of 50 pounds of nitrate of soda—an increase from 100 to 150 pounds—increased the yield only 6 pounds of lint per acre; the second increment of 50 pounds of nitrate of soda, 14 pounds per acre; and the third and fourth increments, only 1 pound per acre. The application of 200 pounds of nitrate of soda, however, furnished the optimum amount of nitrogen, since larger applications produced only slight additional increases in yield. The treatment consisting of 300 pounds of nitrate of soda, 450 pounds of superphosphate, and 45 pounds of muriate of potash, (which contained the same ratio of nitrogen, phosphoric acid, and potash but 50 per cent more plant food than the treatment of 200 pounds of nitrate of soda, 300 pounds of superphosphate, and 30 pounds of muriate of potash) made an average yield of 256 pounds of lint per acre, or 30 pounds of lint more than the latter treatment. Potash produced only small increases in yield during the four years of the Work. The use of 30 pounds of muriate of potash (with nitrate of soda and superphosphate) made an average yield of 206 pounds of lint per acre, or only 6 pounds more than the plats that received nitrogen and phosphate but no potash (Table 1). The application of 60 pounds of muriate of potash made a similar increase in yield. The use of 160 pounds of sulphate of ammonia produced an average yield of 246 pounds of lint, or 20 pounds more than 200 pounds of nitrate of soda (Table 1). Table 1. Effect of rate and time of application of nitrate of soda on the yieldof cotton at Troup Fertilizer treatments per acre Pounds of lint per acre Nitrate Super- Muriate Approximately of phosphate of equal to 400 1927 1928 1929 1930 Average soda 16% potash pounds of— Pounds Pounds Pounds none none none 152 178 84 90 126 All nitrogen applied before planting 100 800 0 4-12-0 161 270 207 160 200 100 300 30 4-12-4 215 236 211 163 206 150 300 30 6-12-4 201 251 219 178 212 200 300 30 8-12-4 209 287 238 170 226 250 300 30 10-12-4 206 286 253 163 227 300 300 30 12-12-4 204 288 246 1'73 228 300 300 60 12-12-8 240 294 ‘ 240 164 235 300 450 45 8-12-4" 231 325 270 198 256 160" 300 30 8-12-4 261 . 274 " 273 176 246 One-half of nitrogen as side dressing 150 ' 300 30 6-12-4 209 '" 251 228 163 213 200 $00 30 8-12-4 22K) 258 224 161 216 250 300 30 10-12-4 194 269 223 142 207 300 300 30 12-12-4 233' 4 265 264 153 229 300 300 60 12-12-8 237 266 259 v 190 238 300 450 45 8-12-4" 237 267 258 160 231 160“ I 300 30 8-12-4 204 268 241 160 218 “Sulphate of ammonia equivalent to 200 pounds of nitrate of soda "600 pounds per acre. THE EFFECT_ OF NITRATE OF SODA ON YIELD OF COTTON 9 The results obtained at Troup indicate that the use of about 200 pounds of nitrate of soda, 300 pounds of superphosphate, and 30 pounds of muriate of potash, a treatment that is roughly equivalent to 400 pounds of an 8-12-4 fertilizer, is good fertilizer practice. These results are in general agreement with the results obtained at Troup during the same years, and published in Bulletin 469 (1932). Results Obtained at Nacogdoches: The application of 100 pounds of nitrate of soda made an average yield of 1'70 pounds of lint for the two years, 1927 and 1929, while the application of 150, 200, 250, and 300 pounds of nitrate of soda produced average yields of 172, 175, 168, and 169 pounds of lint per acre, respectively (Table 2). The application of 100 pounds of nitrate of soda apparently supplied enough nitrogen where 30 Table 2. Yield of cotton fertilized with nitrate of soda at different rates of application at Nacogdoches Fertilizer treatments per acre Pounds of lint per acre Nitrate Super- Muriate Approximately 4 of phosphate of equal to 400 l 1927 1929 Average soda 16% potash pounds of— 1 I Pounds I Pounds Pounds none i none none 98 162 130 All nitrogen applied before planting 100 300 0 4-12-0 141 212 177 100 300 30 4-12-4 158 181 170 150 300 30 6-12-4 126 218 172 200 300 30 8-12-4 149 201 175 250 300 30 10-12-4 140 195 168 300 300 30 12-12-4 142 196 169 300 300 60 12-12-8 168 213 191 300 450 45 8-12-41’ 164 200 182 160" 300 30 8-12-4 167 220 194 One-half of nitrogen as side dressing 150 300 30 6-12-4 229 179 204 200 300 30 8-12-4 218 147 183 250 300 30 10-12-4 220 174 197 300 300 30 12-12-4 203 161 182 300 300 6O 12-12-8 272 152 212 300 450 45 8-12-41’ 264 1,85 225 160“ 300 30 8-12-4 251 188 220 “Sulphate of ammonia equivalent to 200 pounds of nitrate of soda "600 pounds per acre pounds of muriate of potash was used, for larger applications did not produce further increase in yield of cotton. Increasing the amount of muriate of potash from 30 to 60 pounds per acre produced an average gain of 22 pounds of lint per acre, which apparently is a significant gain (Table 2). Side Dressings of Nitrate of Soda Nitrate of soda was applied at the rates of 150, 200, 250, and 300 pounds per acre (a) before planting and (b) one-half before planting 10 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION the cotton and one-half soon after the cotton was thinned to a stand. This work, as stated previously, was conducted on Kirvin fine sandy loam at Troup and on Ruston and Orangeburg fine sandy loams at N acogdoches. The results obtained in individual years at Troup are given in Table 1, and at Nacogdoches in Table 2. The average yields obtained during the four years of the experiment at both points are given in Table 3 for direct comparison. At Nacogdoches on Ruston and Orangeburg fine sandy loams, all of the treatments applied as side dressing produced somewhat larger yields than the same rates of nitrate of soda applied before planting. The largest average gain from side dressing, 43 pounds of lint per acre, was produced by 300 pounds of nitrate of soda. This fact probably indicates that a division of the nitrogenous fertilizer is more desirable with heavy applications than with lighter applications. Somewhat different results were obtained at Troup on Kirvin fine sandy loam soil, which has a stiff plastic clay subsoil. Side dressing reduced the yield considerably in four of the treatments. The average decreases for the four years ranged from 10 pounds to 48 pounds of lint per acre, as shown in Table 3. Where nitrate of soda was used at the rates of 150 and 300 pounds per acre, there was practically no difference in yield where the material was applied before planting or as a side dressing. The differences in results obtained at Nacogdoches and Troup are probably due to differences in the structure of the subsoil at the two points. The Ruston and Orangeburg soils have friable sandy, clay sub- soils, while the Kirvin fine sandy loam has a more compact and less Table 3. Yield of lint cotton per acre where all of the nitrate of soda was applied before planting and where one-half was applied as side dressing Troup Nacogdoches Pounds of . One-half _ _ One-half _ nitrate of soda NltrQgen nitrogen Gam or Nltlw?“ nitrogen Gal“ or per acre applied applied loss for applied applied loss for before as side side before as side side planting dressing dressing planting dressing dressing Pounds l Pounds Pounds Pounds Pounds Pounds 150 212- 213 1 172 204 32 200 226 216 -—10 175 183 8 250 227 I 207 -—2O 168 197 29 300 228 229 1 169 182 13 300“ 235 238 3 191 212 21 300" 279 231 ——48 182 225 43 160° 246 218 ——28 194 220 26 “With 60 pounds muriate of potash “With 45 pounds muriate of potash °Su1phate of ammonia equivalent to 200 pounds of nitrate of soda permeable subsoil. The results of this work indicate that side dressing may be good farm practice on soils with friable, permeable subsoils. On soils with more compact and less permeable subsoils all of the nitrogen may be applied before the cotton is planted. THE EFFECT OF NITRATE OF SODA ON YIELD OF COTTON 11 Comparison of Different Nitrogenous Fertilizers Nitrate of soda, sulphate of ammonia, and cottonseed meal were com- pared on the basis of equivalent amounts of nitrogen as sources of nitrogen for cotton. One hundred pounds of nitrate of soda contains about 15 pounds of nitrogen, which is approximately equal to the nitrogen con- tained in 200 pounds of cottonseed meal or in 80 pounds of sulphate of ammonia. These nitrogenous fertilizers were used in combination with 30 pounds of muriate of potash and 300 pounds of 16 per cent superphosphate or an equal amount of phosphoric acid in other grades of superphosphate. Results Obtained at Troup: The application of 100 pounds of nitrate of soda produced an average yield of 221 pounds of lint per acre, which was 27 pounds, or 14 per cent, more than the yield produced by 200 pounds of cottonseed meal (Table 4). On the other hand, 80 pounds of sulphate of ammonia made an average yield of 255 pounds of lint per acre, which was Table 4. Yield of cotton fertilized with nitrate of soda, sulphate of ammonia, and cottonseed meal, applied at rates to supply approximately equal amounts of nitrogen Pounds of lint per acre at Amount and kind of Troup Nacogdoches nitrogenous fertilizer 1927 1928 1929 1930 Average 1927 1929 Average 1927-29 None 152 178 84 90 138 98 162 130 50 lbs. nitrate of soda 100 lbs. cottonseed meal 245 247 II 224 .... .. 289 143 211 177 67 lbs. nitrate of soda 70 lbs. cottonseed meal 234 235 211 ---- _. 227 145 193 169 200 Lbs. cottonseed meal 170 220 193 ____ .. 194 205 ____ __ 100 lbs. nitrate of soda 215 236 211 163 221 158 181 170 80 lbs. sulphate of ammonia 268 256 240 179 255 143 197 170 160* lbs. sulphate of ammonia 261 274 2773 176 269 167 220 194 ‘This treatment supplies twice as much nitrogen as the otheritreatments, 15 per cent more than the yield resulting from the nitrate of soda and 31 per cent more than the yield from 200 pounds of cottonseed meal. Stated in another way, the yield of the cottonseed meal was 100 per cent, nitrate of soda 114 per cent, and sulphate of ammonia 131 per cent. The mixtures of (a) 50 pounds of nitrate of soda and 100 pounds of cottonseed meal and (b) 67 pounds of nitrate of soda and 70 pounds of cottonseed meal were slightly superior to the nitrate of soda alone. The first~named mixture produced an average yield of 239 pounds of lint, which was 18 pounds, or 8 per cent, more than the yield produced by 100 pounds of nitrate of soda. 12 BULLETIN NO. 490. TEXAS AGRICULTURAL EXPERIMENT STATION Results Obtained at Nacogdoches: The work was conducted at Nacog- doches in 1927 and 1929, the results of which are reported in Table 4 There were no significant differences in the yields of cotton that receivet the different nitrogenous fertilizers and the mixtures composed of varying amounts of nitrate of soda and cottonseed meal. A comparison of the results obtained at Nacogdoches and Troup indi- cates that soils may respond differently to various forms of nitrogen. Or the Ruston and Orangeburg soils at N acogdoches nitrate of soda, sulphate of ammonia, and mixtures of nitrate of soda and cottonseed meal gave almost identical average yields. On the Kirvin fine sandy loam at Troup however, sulphate of ammonia was decidedly the best source of nitrogen, and the varying mixtures of nitrate of soda and cottonseed meal were some- what better than either alone. EFFECT OF FERTILIZERS ON THE CHARACTER OF LINT The length and percentage of lint are important characters of cotton. The length of lint is involved in the quality of cotton and governs, to a large extent, the price of cotton. Consequently these characters are usually emphasized by cotton breeders. Length of Lint It is sometimes thought that application of fertilizers to soil will in- crease the length of lint. In order to show whether or not fertilizers affect this character, the cotton obtained from the variously treated plats was graded and the length of lint determined by a State licensed cotton grader for each year of the test at both Nacogdoches and Troup. The average length of lint of the Mebane variety from each of the fertilizer treatments was 30/32 of an inch at Troup, as shown in Table 5. The length, however, varied somewhat from year to year, probably on account of differences in environmental conditions, although the seasonal differences in length are not shown in the table. There was a slight variation of 1/ 32 of an inch in the length of lint among the several treatments at Nacogdoches (Table 5). Six of the eleven treatments produced lint 31/32 of an inch long, while the other five treatments yielded 30/ 32-inch lint. This variation apparently is the nor- mal range in variation in the length of lint of the variety, or it may be caused by differences in soil moisture, since the unfertilized soil produced as long lint as any of the fertilizer treatments. The results obtained at both Nacogdoches and Troup indicate that ferti- lizers had no effect on the length of lint. The Arkansas Agricultural Experiment Station (Bulletin 273) also found that fertilizers were not effective in increasing the length of lint when determined by the usual methods. i Percentage of Lint The percentage of’ lint is largely a varietal character and consequently is inherited. The percentage of lint, however, may be influenced to a THE EFFECT OF NITRATE OF SODA ON YIELD OF COTTON 13 small extent by seasonal conditions. The cotton grower usually considers the percentage of lint as a fixed character and has not been concerned with any effect the soil or fertilizers may have on this character. In this fertilizer work, however, it was deemed advisable to determine the percentage of lint at both Nacogdoches and Troup. As Was mentioned previously, the Mebane variety of cotton was used in the work at Troup and the Acala variety at N acogdoches. The average percentage of lint of the Mebane variety was practically the same for each fertilizer treatment at Troup, as shown in Table 5. Similar results were obtained with the Acala variety at Nacogdoches (Table 5). It will be noted that the Mebane variety has a somewhat higher percentage of lint than the Acala variety. There were, of course, some slight differences in the percentages of lint, which were probably due to slight Table 5. Length and percentage of lint grown on soil which received different amounts of nitrate of soda Average length of lint in thirty- Average percentage of 1i Pounds of seconds of an inch nt nitrate of d $0811.28!‘ Troup Nacogdoches Troup Nacogdoches 1927-1930 1927 and 1929 1927-1930 1927 and 1929 None 30 31 36.2 34.4 100“ 30 31 36.6 34.3 100 30 31 36.9 34.7 100" 30 30 26.9 34.3 150 _ 30 31 7.1 34.0 200D g0 31 37.0 34.9 200 0 30 36.9 33.4 250 30 31 37.0 34.3 300 30 30 36.9 33.8 300° 30 30 36.6 34.7 300d 30 30 36.6 34.6 “Nitrate of soda used with 300 pounds of super-phosphate but no potash “Sulphate of ammonia. equivalent to amount of nitrate indicated "With 60 pounds of muriate of potash “With 450 pounds of superphosphate and 45 pounds of muriate of potash. errors in weighing and ginning the seed cotton in the process of determin- ing the percentage of lint and not to differences that could be ascribed to fertilizers. These results indicate that fertilizers had no appreciable effect on the percentage of lint. Since the fertilizers had no appreciable effect on the length or per- centage of lint, it will be necessary to select varieties of cotton that naturally have these desirable characters if one wishes to produce longer lint or higher percentage of lint. SIZE OF BOLL The size of boll is important only as it may influence the amount of cotton that may be picked in a day or other given time. can pick more cotton in a day from large bolls than from small bolls, and, consequently, the larger bolls are of economic importance to that extent. A laborer usually \ 14 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION The size of boll in this work is measured, or expressed, by the number of well-opened bolls required to make a pound of seed cotton. It is obvious that the larger the bolls the fewer will be required to make a pound. In general all of the fertilized soil at Troup produced larger bolls than the unfertilized soil. During the four years, 1927-1930, 92 bolls were required to make a pound of seed cotton on the unfertilized soil and 80 to 87 bolls on the fertilized soil (Table 6). Apparently there was no consistent relation between the size of boll and the rate or time of applica- tion of nitrate of soda. At Nacogdoches the fertilized soil produced larger bolls than the un- fertilized soil. During the two years, 1927 and 1929, 103 bolls from the ‘ unfertilized soil and 88 to 97 bolls from the fertilized soil were required to make a pound of seed cotton (Table 6). Apparently the amounts of nitrate of soda had no significant effect 0n size of boll. In 1929, how- ever, the size of boll was reduced markedly by the side dressings of Table 6. Number of well-opened bolls in a pound of seed cotton at Troup and Nacogdoches Fertilizer treatments Troup Nacogdoches per acre Nitrate Super- Muriate of phosphate of 1927 1928 1929 - 1930 Average 1927 1929 Average soda 16% potash Pounds Pounds Pounds None None None 93 87 105 83 92 97 108 103 A11 nitrogen applied before planting 100 300 0 91 80 92 79 86 82 95 89 100 300 30 86 82 85 80 83 _ 84 98 91 150 300 30 87 83 82 79 83 91 96 94 200 300 30 90 77 85 83 84 85 95 90 250 300 30 90 76 89 81 84 87 93 90 300 300 30 89 80 84 84 84 84 98 91 300 300 60 87 82 88 87 86 84 96 90 300 450 45 89 77 89 80 84 80 97 89 160“ 300 30 87 74 81 77 80 80 95 88 80“ 300 30 93 80 80 80 83 89 104 97 One-half of nitrogen as side dressing 150 300 30 85 75 86 79 81 82 98 90 200 300 30 85 77 82 80 81 86 103 95 250 300 ' 30 92 82 88 85 S7 82 103 93 300 300 30 85 80 85 86 84 80 106 93 300 300 60 82 82 84 84 83 80 102 91 300 450 45 84 74 84 84 82 82 101 92 160“ 300 30 90 79 87 85 85 80 96 88 “Sulphate of ammonia containing same amount of nitrogen as 200 pounds of nitrate of soda "Sulphate of ammonia containing same amount ofnitrogen as 100 pounds of nitrate of soda nitrate of soda. As an average of the two years, where the cotton received a side dressing of nitrogen 92 bolls were required to make a pound of seed cotton, while with the corresponding treatments where all of the nitrogen was applied before planting, 90 bolls were required to make a pound. THE EFFECT OF NITRATE OF SODA ON YIELD OF COTTON 15 The soil that received nitrate of soda at various rates also produced a larger percentage of 5-lock bolls than the unfertilized soil, as shown later in Table 7. This fact accounts in a large measure for the larger bolls on the fertilized soil, since in a given variety of cotton, 5-lock bolls are usually larger than 4-lock bolls. EFFECT OF FERTILIZERS ON SIZE OF PLANTS In areas where fertilizers are used profitably, well-fertilized soil usually produces larger plants than unfertilized soil. In this work, data on height of plants, number of fruiting branches, and number of bolls per plant were obtained from each fertilizer treatment in order to show any relation that these characters may have to the yield of cotton. Height of Plants The relative size of plants on fertilized and unfertilized soil is probably of importance only as size may be related to the quantity and quality of lint produced. The average height of plants at Troup and Nacogdoches was about the same, as shown in Table 7. The unfertilized soil at Troup produced plants 21 inches high and the fertilized soil, plants 25 to 28 inches high. The largest application of fertilizer (300 pounds of nitrate of soda, 450 pounds of superphosphate, and 45 pounds of muriate of potash) produced the tallest plants, which were 28 inches high as an average for the four years. Somewhat similar results were obtained at Nacogdoches (Table 7). Number of Fruiting Branches per Plant \ The larger plants on the fertilized soil contained more fruiting branches and more bolls per plant than the plants on the unfertilized soil (Table 7). The number of fruiting branches per plant includes the fruiting branches on the main stem of the plant and the fruiting branches on the vegeta- tive branches. At Troup the number of fruiting branches ranged from 10.9 per plant on the unfertilized soil to 16.5 per plant on the soil which received 300 pounds of nitrate of soda, 450 pounds of superphosphate, and 45 pounds of muriate of potash (Table '7). In general, the number of fruiting branches increased as the amount of nitrate of soda was increased from 100 to 300 pounds per acre. As a rule the plants on fertilized soil produced more fruiting branches than plants on the untreated soil at Nacogdoches, but there was no consistent relation between the number of fruiting branches and the amount of nitrate of soda used. Number of Bolls per Plant The number of bolls per plant is a good indicator of yield and in fact is sometimes used as a basis of estimating yield of cotton before harvesting. The number of bolls per plant, together with the height of plant and number 16 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION of fruiting branches was obtained to show what relation these characters may have to yield. At Troup the number of bolls per plant ranged from 3.2 on the untreated soil to’ 5.9 0n the soil which received 160 pounds of sulphate of ammonia (equivalent to 200 pounds of nitrate of soda), Table 7. This is a difference of 2.5 bolls per plant, which is equal to 120 pounds of lint per acre, as may be seen by referring to Table 1. At Nacogdoches the plants on fertilized soil also produced more bolls than the plants on the unfertilized soil, but there appeared to be no correlation between the number of bolls and the amount of nitrate of soda (Table 7). Percentage of 5-Lock Bolls The percentage of 5-lock bolls frequently is emphasized because in a given variety of cotton the 5-lock bolls are usually larger than the 4-lock bolls. This character, however, is of importance mainly because a higher Table 7. Height of plant, number of fruiting branches, number of bolls per plant, and percentage of 5-lock bolls on soil which received different amounts of nitrate of soda Average height of ‘ti-zigzag; Efiegf Average number of Percentage of Pounds plant in inches per plant bolls per plant 5-lock bolls of ' $1223: Troup Nacogdoches Troup NacogdochesITroup Nacogdoches Troup Nacogdoches per acre 119537; 192g23nd 11927- 1927 and I 1927- 1927 and 1927- 1927 and 930 1929 i 1930 1929 1930 1929 None 21 23 I 10.9 I 11.1 I 3.2 4.2 45 s7 100° 26 25 13.7 11.3 I 4.7 4.6 54 44 10Gb 25 24 12.9 I 10.6 I 4.7 4.7 59 47 100 26 25 14.5 5.4 55 150 25 26 I 13.9 I 12.0 I 4.7 4.3 59 47 20Gb 27 25 14.5 I 11.7 I 5.1 5.2 60 48 200 27 27 14.0 5.9 58 250 27 26 I 14.8 I 12.0 I 5.4 4.4 57 52 300 27 26 14.8 I 12.2 5.4 4.3 59 51 300° 27 26 15.5 I 12.2 5.5 5.1 56 53 300d 28 27 II 16.5 J 12.2 5.7 4.9 59 49 I "Nitrate of soda used with 300 pounds of superphosphate but no potash "Sulphate of ammonia equivalent to amount of nitrate of soda indicated “With 60 pounds of muriate of potash “With 450 pounds of superphosphate and 45 pounds of muriate of potash percentage of 5-lock bolls enables a laborer to pick slightly more cotton. The percentage of 5-lock bolls is a varietal character and is inherited, although it is influenced to some extent by environmental conditions. All of -the fertilized soil at both Troup and Nacogdoches produced a considerably higher percentage of 5-lock-bolls than the unfertilized soil. Considering all of the treatments as a whole, the fertilized soil produced 28 per cent more 5-lock bolls than the unfertilized soil at Troup and 32 per cent more at Nacogdoches (Table 7 ). I The results obtained on the size of plants show that fertilizer increased the yield of cotton by producing larger plants, which had more fruiting branches and more bolls, a higher percentage of- 5-lock bolls and conse- quently larger bolls than plants grown on unfertilized soil. THE EFFECT‘ OF NITRATE OF SODA ON YIELD OF COTTON 17 EFFECT OF FERTILIZERS ON SHEDDING The percentage of shedding was obtained in these experiments because it was desired to ascertain if various rates and time of application of nitrate of soda would reduce shedding. By percentage of shedding we mean the percentage of the total number of blooms produced that do not form well-matured bolls. For example, if 1000 blooms were produced and 600 of these blooms did not-result in mature bolls, there would be 60 per cent of shedding. This may be stated in another way by saying that of the 1000 blooms, only 400, or 40 per cent, developed into mature bolls. The amount of shedding was obtained by counting the total number of blooms produced during the season and the total number of bolls picked on 50 average plants receiving the different fertilizer treatments. The Table 8. Percentage of shedding as influenced by various fertilizer treatments at Troup and Nacogdoches Percentage of shedding Fertilizer treatments pounds per acre Troup Nacogdoches Nitrate Super- Muriate l of phosphate v of 1927 1928 1929 I 1930 Average 1927 1928 1929 Average soda 16% potash l Pounds Pounds I Pounds None None None 61 38 38 20 39 61 34 62 52 All nitrogen applied before planting 100 800 0 64 48 50 32 49 64 39 59 54 100 ‘x300 30 63 44 52 39 50 59 39 63 ' 54 150 ‘ 300 30 67 45 54 3 51 66 41 62 56 200 300 30 70 52 56 39 54 59 40 60 53 250 300 30 65 50 58 35 52 65 32 63 53 300 300 30 68 54 54 36 53 64 48 62 58 300 300 60 67 56 55 34 53 63 30 61 51 300 450 45 67 52 57 47 56 61 24 62 49 160° 300 30 63 49 54 44 53 80" 300 30 56 48 52 36 48 One-half of nitrogen as side dressing 150 300 30 64 48 53 33 50 56 30 68 51 200 300 30 68 45 63 39 54 53 33 69 52 250 300 30 67 51 56 39 53 52 28 66 49 p 300 300 30 68 52 57 43 55 49 28 _ 67 48 , 300 300 60 68 48 60 37 53 53 29 69 50 § 300 450 45 68 53 59 45 56 58 30' 69 52 E 160“ 300 30 64 50 59 44 54 s “Sulphate of ammonia equivalent to 200 pounds of nitrate of soda I “Sulphate of ammonia equivalent to 100 pounds of nitrate of soda i‘ percentage of shedding was found by dividing the number of bolls picked by K the total number of blossoms produced and multiplying by 100. It Will be observed in Table 8 that all of the fertilizers increased the ‘dpercentage of shedding at Troup. The average amount of shedding was '39 per cent on the unfertilized soil and more than 50 per cent on almost 18 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION all of the fertilized soil. At Nacogdoches, however, the fertilizers did not appreciably affect shedding. There was a slight tendency for the amount of shedding to increase with the amount of fertilizer used at Troup, but no such tendency was apparent at Nacogdoches. The side dressing apparently had a slight tendency to reduce shedding at Nacogdoches but had no appreciable effect at Troup. EARLINESS The length of the growing season is not a limiting factor in cotton production in eastern Texas unless planting is delayed far beyond the normal planting dates. Earliness, however, is important for several reasons: (1) the plant puts on more fruit during the usual favorable weather before the drouth periods that often occur during late summer, (2) insect damage, especially by boll weevils, almost invariably becomes greater as the season advances, and (8) an earlier crop is usually harvested during more favorable weather and escapes weather damage, resulting in a better grade of cotton. Under these conditions, any practice, such as the Table 9. Effect of fertilizer on earliness of cotton as measured by the percentage of total yield of seed cotton harvested in the first three pickings at Troup and Nacogdoches Fertilizer treatments Troup Nacogdoches per acre Nitrate Super- Muriate I I I I I of phosphate of I 1927 I 1928"I 1929 I 1930 IAverage 1927 1929 Average soda 16% potash I I I I I I I Pounds Pounds Pounds P.ct. P.ct, P.ct, P.ct,I P.ct P.ct, P.ct,I P.ct. I None None None 46 48 30 22 37 47 66 57 All nitrogen applied before planting 100 300 0 58 57 57 45 54 69 69 69 100 800 30 53 62 65 54 59 69 64 67 150 300 30 57 63 59 56 59 72 68 70 200 300 30 52 61 62 68 61 75 65 70 250 300 30 45 61 66 64 59 73 66 70 300 300 30 42 65 65 62 59 76 65 71 300 300 60 48 59 60 59 57 76 68 72‘ 800 450 45 44 54 60 64 56 82 69 76 160“ 300 30 58 59 73 63 63 82 77 80 80" 300 30 65 58 71 59 63 82 71 77 50d 300 30 62 56 60 59 77 73 75 67d 300 30 63 57 56 59 68 71 70 800° ‘ 300 30 44 68 7O 61 One-half of nitrogen as side dressing 150 300 30 52 65 67 52 59 65 79 72 200 300 30 50 65 69 53 59 64 74 69 250 300 30 46 69 70 66 63 64 77 71 300 300 30 49 65 75 64 63 55 87' 71 300 300 60 39 69 72 67 62 69 78 74 300 450 45 41 68 75 66 63 76 83 80 160° 300 30 48 68 76 78 68 75 82 79 “Sulphate of ammonia equivalent to 200 lbs. of nitrate of soda bSulphate of ammonia equivalent to 100 lbs. of nitrate of soda ”First two pickings in 1928 “With 100 lbs. and 70 lbs. of cottonseed meal, respectively "Cottonseed meal THE EFFECT OF NITRATE OF SODA ON YIELD OF COTTON 19 use of fertilizers, that will allow the cotton plant to put on more fruit early in the season and consequently produce an earlier and larger crop is good farm practice. Data were obtained to measure the effect of fertilizers on earliness. Earliness in cotton may be measured or indicated by (1) the first bloom, (2) the peak of the blooming period, and (3) by the percentage of the total yield harvested in the first two, three, or four pickings. In these studies the cotton was picked at weekly intervals or as nearly at weekly intervals as conditions would permit, after the cotton began to open actively. The earliness was measured by the percentage of the total yield that was harvested in the first three pickings in 1927, 1929, and 1930, and in the first two pickings in 1928. i On the average all of the fertilized soil produced an earlier crop than the unfertilized soil. During the four years, 1927-1930, the unfertilized soil at Troup produced 37 per cent of its total yield at the first three pickings while the soil which received 150 pounds of nitrate of soda pro- duced 59 per cent of the total yield at the first three pickings (Table 9). There appeared to be no significant difference in the effect of different amounts of nitrate of soda on earliness. Sulphate of ammonia, however, hastened maturity more than nitrate of soda at both Nacogdoches and Troup (Table 9). As a whole, side dressing was conducive to earliness at Troup, but probably was not significant. The results indicate that the use of a properly balanced fertilizer hastens the maturity of the cotton crop on the sandy soils of the region. Further, sulphate of ammonia made larger yields and a slightly earlier crop of cotton than nitrate of soda or varying mixtures of nitrate of soda and cottonseed meal on the Kirvin soils at Troup. SUMMARY AND CONCLUSIONS p The main object of this fertilizer work was to secure information on éthe effect of the time and rate of application of nitrate of soda on the iyield, length and percentage of lint, size of boll, shedding, and other lcharacters of cotton that may be used in developing a more profitable gpfertilizer practice. The nitrate of soda was applied at the rates of 100, i150, 200, 250, and 300 pounds per acre in combination with suitable Earnounts of superphosphate and muriate of potash. Applications of nitrate ,sof soda before planting were compared with side dressings applied when lathe cotton was thinned to a stand. The experiment was conducted on Kirvin fine sandy loam soil at Troup and on Ruston and Orangeburg fine (sandy loams at Nacogdoches during the four years, 1927 to 1930. A The average yield of cotton increased as the rate of application of 'itrate of soda was increased from 100 to 300 pounds per acre at Troup. The use of 200 pounds ‘of nitrate of soda per acre, however, apparently urnished enough nitrogen, since larger applications produced only small p dditional increases in yield. At Nacogdoches the application of 100 ounds of nitrate of soda supplied enough nitrogen, for larger applications K ere only slightly more effective. 20 BULLETIN NO. 490, TEXAS AGRICULTURAL EXPERIMENT STATION On the soils at Nacogdoches side dressings produced larger average yields of cotton than applications of all the nitrogen before planting. On Kirvin fine sandy loam at Troup, however, larger yields were obtained from the applications of l-nitrogen~-before planting. These differences in yields“are probablydue toatheialifferences in structure of the soils at the two points. Ruston and Orangeburg soils have friable sandy clay sub- soils while Kirvin fine sandy loam has a dense, compace subsoil Which prob- ably does not permit as much leaching of fertilizer as the more permeable subsoils of Ruston and Orangeburg soils. These results indicate that side dres- sing is good farm practice on soils with friable, permeable subsoils, especially where large amounts of fertilizer are usd. On soils with more compact and less permeable subsoils, all of the nitrogen may be applied before the cotton is planted. Nitrate of soda, mixtures of nitrate of soda and cottonseed meal, and sulphate of ammonia were compared as sources of nitrogen. These ma- terials produced practically the same yields of cotton on Ruston and Orangeburg soils at Nacogdoches. Sulphate of ammonia, however, was the best source of nitrogen on Kirvin fine sandy loam at Troup and produced a larger and earlier crop of cotton than nitrate of soda or mixtures of nitrate of soda and cottonseed meal. In general, applications of fertilizers hastened maturity of cotton. For example, at Troup soil which received 150 pounds of nitrate of soda pro- duced 59 per cent and the unfertilized soil, 37 per cent of its total yield, at the first three pickings. Apparently there was no consistent or sig- nificant difference in the effect of different amounts of nitrate of soda on earliness. Sulphate of ammonia, however, hastened maturity more than nitrate of soda. The time and rate of application of nitrate of soda had no significant effect on the percentage of shedding at Nacogdoches. At Troup, how- ever, there was a considerably higher percentage of shedding on the fertil- ized soil than on the unfertilized soil. The various rates of application of nitrate of soda with phosphoric acid and potash had no effect on the length of lint, since the unfertilized soil produced just as long lint as the fertilized soil. In general fertilizers did not influence the percentage of lint, although at Troup the percentage of lint on the unfertilized soil was slightly lower than on the fertilized soil. The fertilized soil produced larger plants with more fruiting branches and more bolls and larger bolls than the unfertilized soil. The fertilized soil also produced a higher percentage of 5-lock bolls than the unfertilized soil. This fact probably accounts in a large measure for the larger bolls on the fertilized soil, for in a given variety of cotton 5-lock bolls are generally larger (heavier) than 4-lock bolls. ~ From these results it is concluded that fertilizers increase the yield of cotton by producing larger plants which have more bolls and larger bolls than are produced on unfertilized soil and not by increasing the length of lint or the percentage of lint.