TEXAS AGRICULTURAL EXPERIMENT STATION R. D. LEWIS, Director. College Station. Texas Mm 758 Cotton Fertilizer Tests in the El Peso Area, 1943-51 w!» 1195a , _¢, The TEXAS AGRICULTURAL AND MECHANICAL COLLEGE SYSTEM I b I GIBB GILCHRIST. Chancellor CONTENTS Page Introduction .................................................................................................... ._ 3 Many Treatments on Field Plots Included ............................................. _. 3 Yield Increases from Nitrogen ................................................................... .. 5 Little Yield Increase from Phosphorus ................................................... .. 7 Higher Yields Following Alfalfa ............................................................. .. 8 Potassium Fertilizer Not Required ........................................................... .. 9 No Yield Increases from Trace Elements ................................................ ._ 9 Discussion of Results ................................................................................... .. 9 Conclusions and Recommendations ........................................................... .. 11 Nitrogen .................................................................................................. .. 11 Phosphorus ............................................................................................. .- 11 Potassium ............................................................................................... .. 11 Trace Elements ...................................................................................... _. 12 General Recommendations .......................................................................... .. 12 Acknowledgments ......................................................................................... _. 12 Literature Cited ............................................................................................ .. 13 Appendix ........................................................................................................ _. 14 Fertilizer Guarantees ........................................................................... .. 14 Factors Governing Choice of Fertilizer .......................................... .. 14 Cotton Fertilizer Tests in the El Paso Area, 1943-51 Paul D. Christensen and Paul J. Lgerlg* COTTON IS THE PRINCIPAL CROP grown in El Paso Val- ley. It was planted on 60 to 85 percent 0f the cultivated land from 1943 through 1951. The maintenance of high yields in- cludes a consideration of the fertilizer requirements of the crop on the soils where it is grown. Work before 1943 on Mesilla Valley soils in New Mexico (7, 11)1, which are similar in nature and productivity to those in El Paso Valley, showed that yield increases in cotton re- sulting from commercial fertilizer applications were generally limited to certain types of soils. In many cases, the yield in- creases would not pay the cost of the fertilizer and its appli- cation. The advent of higher cotton prices and the likelihood of soil depletion as a result of continuous cropping made ad- visable an examination of the fertilizer needs of the soils in the El Paso area. The soils in El Paso Valley are composed of stratified river sediments varying in texture from medium sands to heavy clays. They have been classified by the Soil Conser- vation Service in the Gila and closely related soil series. The soils are variable in physical properties and salt content. MANY TREATMENTS ON FIELD PLOTS INCLUDED Early fertilizer tests (1943-45) at the El Paso Valley Ex- periment Station were aimed at determining ‘the fertilizer nutrients required to produce the highest yields of cotton (8). Later experiments (1, 3, 4) were directed toward determin- ing the types of soils which require fertilizer, the amounts of fertilizer required and the influence of alfalfa in rotation on the need of cotton for fertilizer. Until 1950, all fertilizer ex- periments were conducted on the station at Ysleta. Tests were established in 1950 and 1951 on private farms in the El Paso Valley. In all the fertilizer experiments reported in this *Respectively, assistant agronomist and superintendent, El Paso Valley Experiment Station, Ysleta, Texas. lNumbers in parentheses refer to the literature cited. __4_ bulletin, Acala cotton was used as the test crop. The experi- ments included approximately 75 different fertilizer treat- ments in which nitrogen, phosphorus, potassium and the trace elements were applied to more than 2,000 field plots. The fertilizers were applied to the soil as a side-dressing after the cotton was up, usually between May 10 and June 10. Applications were made 4 to 10 inches on each side of the row and 4 to 6 inches deep. Each fertilizer treatment in each test was replicated several times in randomized blocks. The ferti- lizer plots were either two or four rows wide and usually 50 feet long. Prior to 1950, fertilizers were applied with a tractor- mounted field fertilizer distributor. In 1950 and 1951, appli- cations were made with a tractor-mounted experimental plot fertilizer distributor shown in Figure 1. For purposes of discussion, the results of the nitrogen and phosphorus fertilizer experiments are divided into two parts: first, tests on soils wnich had been planted to cotton Figure 1. Experimental plot fertilizer distributor mounted on tractor. _5_ Table 1. Yield of seed cotton as influenced by application of nitrogen fertilizer, 1947-51 Year Soils in continuous cotton Soil planted to cotton Light Medium Heavy following alfalfa Check Nitro. Check Nitro. Check Nitro. Check N itro. Pounds of seed cotton per acre 1947 2269 2139 1735 21341 2181 24271 3891 3845 1948 2696 30681 3388 3318 2215 2380 3618 3556 1949 1630 22881 2055 25621 2753 30411 3178 2986 1950 2833 2876 2243 1998 2802 3015 5381 5541 1951 1261 1139 3655 43001 774 1031* 3683 3895 Av. 2138 2302 2615 2862 2145 2379 3950 3965 1The odds are 19 to 1 that the difference in yield between the check and the nitrogen- treated plots is real and not the possible result of plant and soil variations in the field. for 3 to more than 10 years, and second, tests on soils Where cotton followed immediately after alfalfa. YIELD INCREASES FROM NITROGEN Preliminary tests before 1947 indicated yield increases from nitrogen and combinations of nitrogen and phosphorus on the lighter soils. Table 1 shows the yield resulting from nitrogen applications in the tests conducted from 1947 through 1951. Where nitrogen was applied alone on land in contin- uous cotton, average yields were above the check in 11 out of 15 tests. In 8 of these tests, the odds were 19 to 1 that the yield increases were real and not the possible result of plant and soil variations in the field. Considering all treatments in the 9-year period, 1943-51, the average yield increase from nitrogen fertilizer applied alone was 2.3 pounds of seed cotton per pound of available nitrogen (N) applied (Table 2).) This Was equivalent to ap- proximately 190 pounds of seed cotton per acre (Table 2 and Figure 2) for each treatment. The treatments varied from 50 to 130 pounds of nitrogen (N) per acre. Table 2. Increased yield of seed cotton, pounds per acre, over check as influenced by application of nitrogen (N), available phosphoric acid (P105), and combinations of both, 1943-51 Fertilizer __ Type of soil Average Average nutrient Light Medium Heavy all soils all soils Pounds of cotton per pound of available nutrient Pounds cotton per treatment1 N 2.0 2.6 2.4 2.3 190 P205 -—0.2 0.2 0.2 0.1 l0 N -|- P205 1.6 1.5 0.8 1.3 210 1The average increase resulting from the various fertilizer treatments. Nitrogen applied at rates of 50 to 130 pounds per acre. Available phosphoric acid applied at rates of 50 to 320 pounds per acre. 2600 2400- 2200 - 2000- l8OC)- 200- $3 5Q eaten n3" t. muciva | 32> 3.3 .3352 ‘gage Fertifizer f s Figure 2. Yield of seed cotton as influenced by application nitrogen, phosphorus, and combinations of both. Average of 19 t in 9 years. _7__ Figure 3, which Was adapted from recent fertilizer tests, indicates that the lower rates of nitrogen resulted in more cotton per pound of fertilizer than the higher rates. Sixty pounds of nitrogen increased the yield 200 pounds over the unfertilized plots. An additional 60 pounds of nitrogen (120 pounds N) further increased the yield only 100 lbs. Similar results were obtained where the nitrogen plus phosphorus treatments Were applied. LITTLE YIELD INCREASE FROM PHOSPHORUS Yield increases from phosphorus resulted in only 2 out of 19 tests. The average yield increase per pound of avail- able phosphoric acid (P2O5) applied alone Was 0.03 pound of seed cotton per acre. This was equivalent to only 10 pounds of cotton per fertilizer treatment (Table 2), With treatments varying from 50 to more than 200 pounds of available phos- phoric acid per acre. There appears to be little need to apply phosphorus in combination with nitrogen, since the combina- 2.600 - 2400 - 2200 2000 I800 Yield — pounds of seed cotton per acre 200 . _ g‘. ‘w; . 5 t * I . "v ‘k . ~ - ._ 0 60+6060tl20lg%+ 192%; O 6O . l - Nitrogen Phosphorus(P,0,) Nitrogen + phosphorus Potassium (K20) (N + P2 05) Rates of Fertilizer Applications -— Pounds Per Acre Figure 3. Yield of seed cotton as influenced by rates of application of nitrogen and phosphorus (applied singly and in combination), and potassium. Average of 1949-51 tests. .—8_. Table 3. Yields of seed cotton as influenced by rotation with alfalfa, compared with yields from continuous cotton‘. r- Fertilized continuous cotton Cotton following Year Average? Highest treatment3 alfalfa‘ Pounds per acre 1947 2170 2383 3926 1948 2911 3152 3596 1949 2563 2845 3085 1950 2591 2735 5433 1951 2090 2368 3768 Av. 2496 2697 3939 lCotton grown continuously for 3 or more years. 2Average yield of all fertilized plots on each soil. “Average yield of highest-yielding treatments on each soil. ‘Average yield of all plots on each soil. tion gives little or no more yield than nitrogen alone (Table 2 and Figure 3). Results of the 1945 test, the 1949 test on heavy soil and unpublished data in 1951 indicate, however, that previously unfertilized soils or soils not fertilized for 5 to 10 years may show yield increases from phosphorus appli- cations. HIGHER YIELDS FOLLOWING ALFALFA Two outstanding results of the fertilizer tests on soil planted to cotton following alfalfa are: first, the application of nitrogen and phosphorus fertilizers to soil planted to cot- ton the first year following alfalfa was of little or no value in increasing cotton yields; second, the alfalfa-cotton rotation appeared to be generally more effective in increasing cotton yields than the application of commercial fertilizer. Table 3 shows the average cotton yields during the 5-year period, 1947-51, on fertilized soils in continuous cotton and on soils in cotton following alfalfa. In all comparisons, yields were higher where cotton followed alfalfa. The differences in yield varied from approximately one-fifth to over two bales per acre and averaged approximately one bale per acre in favor of the alfalfa-cotton rotation. This was true even where the highest-yielding fertilized plots were singled out for com- parison. (See the section, Discussion 0f Results, for further discussion on these data.) Table 4. Yields of seed cotton as influenced by applications of potash (K20) fertilizer, 1951‘ Treatment Average yield Lbs. K20 per acre Lbs. per acre None 2634 s0 ' 2636 lAverage of three tests. _ 9 __ POTASSIUM FERTILIZER NOT REQUIRED Several tests were conducted at the El Paso Valley Ex- periment Station during the 9-year period in which potassium fertilizers were applied to cotton land. In no case was the yield increased by these applications. In 1951, three fertilizer tests were established on privately-owned farms in the El Paso Valley in which potassium was applied singly and in com- bination with four levels of nitrogen and two levels of phos- phorus. Figure 2, showing the effects of potassium applied alone, and Table 4, showing the overall effects of potassium, indicate that applications of potassium had little or no influ- ence on the cotton yield. NO YIELD INCREASES FROM TRACE ELEMENTS Preliminary tests using the trace elements, copper, boron, manganese, zinc and iron, applied singly or in various combi- nations were conducted in 1945, 1947 and 1951. In no in- stance did the application of any of the trace elements result in higher cotton yields. DISCUSSION OF RESULTS The data presented in this bulletin and elsewhere (3, 4, 8) show a wide variation in results obtained each year and from year to year. These variations are due in part at least to the extreme variability in the soils in the El Paso Valley. Varia- tions in the soils result from wide differences in soil texture and salt content, both in the surface and in the underlying layers, and soil structural conditions stemming in part from various cultural practices. Soil variations which occur in short distances in the field are reflected in wide variations in crop growth and often mask yield differences due to ferti- lizer treatments. For these reasons, conclusions drawn from fertilizer experiments should come from an overall consider- ation of the results of several years’ work. On most soils which have been in continuous cotton for a number of years, the application of nitrogen usually has re- sulted in yield increases. Results in 1945, 1949' and unpub- lished data in 1951 also indicate that, where soils have not been fertilized with phosphorus within the previous 5 to 10 years, cotton yields may be increased by applications of super- phosphate. The lack of response to nitrogen found on some soils may be due to the fact that other factors affecting plant growth are more critical. In many instances, certain physical and __.1()__.. chemical conditions of the soil, such as restricted water pene- tration, inadequate soil aeration, low water holding capacity, hard-pan formation or excessive salt concentration inhibit crop growth even-when ample nutrients are present in the soil. In a similar manner, the lack of yield increases from ni- ‘ trogen applied on certain soils may be attributed to the gen- erally high native fertility of the soils, which are composed of layers of sediment deposited by the Rio Grande. In other words, assuming that all the cotton stalks and leaves are re- turned to the soil at the end of each growing season, the fer- tilizer nutrients removed from the land through cotton farm- ing are practically all contained in the cotton seed. The amounts of the fertilizer nutrients removed from the land in a bale of seed cotton (500 pounds of lint and 870 pounds of seed) are approximately 32 pounds of nitrogen (N), 11 pounds of available phosphoric acid (P205) and 10 pounds of potash (K20). In contrast with one bale of cotton, 4 tons of alfalfa remove approximately 6 times as much nitrogen (presumably partially replaced by root nodule bacteria), 3 times as much phosphoric acid and 20 times as much potash. In similar man- ner, 10 tons of corn fodder removes 8 times as much nitrogen, 6 times as much phosphoric acid, and 20 times as much pot- ash. When it is recognized that varying amounts of nitrogen are added to the soil each year through fixation by soil organ- isms not associated with legumes, it is reasonable to assume that in some soils sufficient nitrogen is present in some sea- sons to take care of the needs of cotton. When the fertilizer tests were established for a compari- son of fertilizer response on light, medium and heavy soils, and soils the first year following alfalfa, no attempt was made to design the experiments so that specific comparisons be- tween soils could be made. The physical requirements of a field plot test in which specific comparisons between soils could be made were beyond practical limits under conditions prevailing in the area. The comparisons between the yields on soils in continu- ous cotton and those in alfalfa-cotton rotation are, therefore, not direct, and conclusions drawn from the data should be in- terpreted in a general way. It is believed, however, that the soils in the El Paso Valley are sufficiently similar in produc- tivity to justify the general comparison made between cot- ton yields following cotton with yields following alfalfa. _ 11 _ CONCLUSIONS AND RECOMMENDATIONS Results of fertilizer experiments at the El Paso Valley Experiment Station (1, 3, 4, 8) and similar findings at other stations (2, 7, 9, 10, 11, 12) in the arid West lead to the fol- lowing conclusions regarding the fertilization of short staple cotton: Nitrogen Nitrogen is the nutrient most lacking in the soils of the El Paso Valley. Applications of nitrogen will result in yield increases on most soils where cotton does not follow alfalfa. The increases occur often enough to warrant the annual ap- plication of the fertilizer. There is evidence that yield in- creases resulting from nitrogen applications are more likely to occur on the lighter soils than on the heavier types, but re- sults at the El Paso Valley Experiment Station indicate that past cropping history is more important than soil texture in determining fertilizer needs. Fertilizer applications equivalent to 60 to 80 pounds of available nitrogen (N) per acre are adequate for high yields of cotton. Higher applications of fertilizer generally do not appreciably further increase the yield. Where cotton is planted immediately following alfalfa, nitrogen applications are not recommended. Phosphorus Increases in yield of cotton resulting from the application of superphosphate are infrequent and are generally confined to soils which have not received any phosphate for 5 to 10 years. The application of superphosphate probably once every 5 years is adequate to meet the need of cotton grown on soils - in the El Paso Valley. Before applying the fertilizer, it is recommended that samples of the soil be analyzed chemically to determine whether sufficient phosphorus is available to meet the needs of the crop. Where alfalfa is grown in rotation with cotton, applica- tions of phosphorus to the alfalfa are probably sufficient to meet the needs of both alfalfa and the cotton which follows. Potassium Results of several experiments at the El Paso Valley Ex- periment ‘Station show that soils in this area contain sufficent potassium. In addition, the irrigation water applied at pres- __12_ ent rates supplies approximately 75 pounds of potassium per acre annually (5). This amount is far in excess of that re- moved by cotton. Trace Elements Experimental Work at this station indicates that applica- tions of trace elements are of little value in increasing the yield of cotton. A GENERAL RECOMMENDATIONS For highest cotton production, rotate cotton with alfalfa. Under an alfalfa-cotton rotation, apply phosphorus to al- falfa and eliminate phosphorus applications to cotton. A lab- oratory test for available soil phosphorus is suggested as an aid in determining Whether to apply the fertilizer. Apply nitrogen fertilizers to cotton annually after the second year following alfalfa. Sixty to 80 pounds of nitrogen per acre are suggested. Applications of potassium or the trace elements are not recommended. It is highly recommended that each farmer make a prac- tice of limited fertilizer testing on his farm, as all soils do not respond in the same manner to fertilization. To test a ferti- lizer, the farmer should apply it to one border and leave the- adjacent border unfertilized as a check. By repeating this procedure several times in the field, reliable information can be obtained. ' ACKNOWLEDGMENTS The authors Wish to acknowledge the very fine coopera- tion of the following farmers on whose farms fertilizer tests were conducted: Ord Gary, Erich Brandes, George Q, Payne, Jess M. Taylor, Richard E. Sparks, Paul C. Davidson, Roy Davidson Farms, R. L. Skov, S. R. Davis, R. T. Hoover Farms, Howard Surratt Farms and Ralph’s Farms. (1) (2) (3) (4) (5) (6) (7) (3) (9) (10) (11) (12) __13__. LITERATURE CITED Anderson, Olan E. and P. J. Lyerly. Cotton Fertilizer Tests in the El Paso Valley, 1949. Tex. Agric. Expt. Sta. Prog. Rept. 1230. 1950. Brimhal, L. L. and W. T. McGeorge. Agric. Expt. Sta. Bul. 234. 1951. Christensen, P. D. and P. J. Lyerly. Cotton Yields in the El Paso Valley as Influenced by Applications of Ammonium Nitrate and Superphosphate. Tex. Agric. Expt. Sta. Prog. Rept. 1395. 1951. Christensen, P. D. and P. J. Lyerly. Cotton Yields in the El Paso Valley in 1951 as Influenced by Applications of Ammonium Ni- trate and Superphosphate. Tex. Agric. Expt. Sta. Prog. Rept. 1490. 1952. Christensen, Paul D. and Paul J. Lyerly. Water Quality . . . As It Influences Irrigation Practices and Crop Production. Tex. Agric. Expt. Sta. Cir. 132. 1952. Fudge, J. F. and T. L. lOgier. Analysis of Commercial Fertilizers Sold during 1951-52. Tex. Agric. Expt. Sta. Bul. 755. 1952. Hinkle, D. A. and Glen Staten. Fertilizer Experiments with Acala Cotton on Irrigated Soils. N.M. Agric. Expt. Sta. Bul. 280. 1941. Lyerly, P. J. Fertilizer Tests with Cotton and Corn at the El Paso Valley Substation, 1943-1945. Tex-. Agric. Expt. Sta. Prog. Rept. 993. 1946. Mikkelsen, D. S. and Marvin Hoover. Calif. Agric. Vol. 6, No. 8:3-4. 1952. Serviss, G. H. and R. S. Hawkins. Cotton Fertilizer Experiments in the Salt River Valley. Ariz. Agric. Expt. Sta. Bul. 129. 1929. Staten, Glen and D. A. Hinkle. Fertilizer Experiments with Cot- ton on Heavy Irrigated Soils N. M. Agric. Expt. Sta. Bul. 248. 1937. Staten, Glen and D. A. Hinkle. Maintaining Cotton Yields through Fertilizer and Crop Rotation. N. M. Agric. Expt. Sta. Bul. 340. 1947. Cotton Fertilization. Ariz. Cotton Fertilization Trials. _ 14 ._ APPENDIX Fertilizer Guarantees Dealers in fertilizer are required by law to state the guaranteed analysis of the fertilizer on the fertilizer bag (6). This guarantee is shown in three numbers, of which the following are examples: 33-0-0,‘ 16-20-0, 0-45-0 and 5-10-5. The three numbers in each guarantee show, in order from left to right, the percent of nitrogen, available phosphoric acid and potash. Nitrogen is expressed as total nitrogen (N). Terms such as “am- monia” or “equivalent to ammonia” are not allowed. Phosphorus is expressed as the percent of available phosphoric acid (P205). The term “available” is defined as the phosphate which will dissolve in a neutral solution of ammonium citrate. In products such as _ raw rock phosphate, bone meal or tankage, “total” phosphoric acid may be guaranteed; but where this is done, no “available” phosphoric acid can be guaranteed. Potassium is expressed as the percent of potash (K10) which will dissolve in distilled water. For a detailed discussion on commercial fertilizers, see Texas Agri- cultural Experiment Station Bulletin 755, “Analysis of Commercial Fer- tilizers Sold During 1951-52” (6). Factors Governing Choice of Fertilizers The choice of fertilizer to apply involves consideration of the ferti- lizer nutrients lacking in the soil, the method of fertlizer application, the acid or alkaline reaction produced in the soil by the fertilizer and the cost of the fertilizer. The following general rules should be follow- ed in selecting fertilizer: Apply only the fertilizer nutrients which are in short supply in the soil. Select the fertilizer or fertilizers which will supply only the re- quired nutrients. The application of nitrogen and phosphorus where only nitrogen is required, or the application of nitrogen, phosphorus and potassium where only nitrogen and phosphorus are required, involves both a waste of fertilizer and an unnecessary increase in the cost of fer- tilization. On arid soils, apply fertilizers which produce an acid or neutral re- action in the soil. Nitrate of soda, for example, is not recommended be- cause it tends to produce alkali soil conditions. Purchase fertilizers on the basis of the cost per pound of available plant nutrient, other factors being equal. Where liquid fertilizers are applied in irrigation water, the irriga- tion system should be adequate to provde an even dstribution of water over the land. Where anhydrous ammona is applied directly to the soil, applica- tions should be at least four inches deep; otherwise, appreciable am- monia may be lost to the atmosphere.