@article{cox_barnes_2002, title={Peanut, corn, and cotton critical levels for phosphorus and potassium on Goldsboro soil}, volume={33}, ISSN={["0010-3624"]}, DOI={10.1081/CSS-120003880}, abstractNote={Fertilizer applications of phosphorus (P) and potassium (K) are most economic if based upon local soil test information. The objective of this study was to determine the soil test (Mehlich 3-extractable) and plant tissue critical levels of P and K for corn (Zea mays L.), peanut (Arachis hypogaea L.), and cotton (Gossypium hirsutum L.) grown on a Goldsboro (fine-loamy, siliceous, subactive, thermic Aquic Paleudult) sandy loam for 17 years. Four soil test levels of P were created and maintained by applying triple superphosphate at about 6-year intervals in a factorial design, with three rates of K applied annually prior to planting crops other than peanut. Soil and tissue samples were taken in midseason, and yield and grade determined at harvest. Yields from the first two years and one drought year were not included. Critical levels were determined by a linear response and plateau method on data restricted to that portion of the crop being grown at a sufficient level of the other nutrient. The Mehlich 3 P critical levels (mg dm−3) were 22 for corn, 21 for peanut, and 16 for cotton, whereas the plant tissue critical levels for these three crops were 0.23, 0.26, and 0.21%, respectively. It was calculated that an annual application of 37 kg P2O5 ha−1 (33 lb acre−1) would maintain the Mehlich 3 P level at 33 mg dm−3, a practical maximum for the recommendation of P fertilizer, on this soil. The Mehlich 3 K critical levels (cmol dm−3) were 0.08 for corn, 0.10 for peanut, and 0.10 for cotton on this soil which has a low cation exchange capacity, whereas the plant tissue critical levels for these crops were 1.0, 1.6, and 0.9%, respectively. Cotton required a greater application of K than corn.}, number={7-8}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Cox, FR and Barnes, JS}, year={2002}, pages={1173–1186} }
 @article{byers_mikkelsen_cox_2001, title={Greenhouse evaluation of four boron fertilizer materials}, volume={24}, ISSN={["1532-4087"]}, DOI={10.1081/PLN-100103665}, abstractNote={Management of boron (B) in soils can be difficult since a narrow range exists between plant deficiency and toxicity. A greenhouse study was conducted to evaluate the effectiveness of four soil-applied B fertilizers varying in solubility (Granubor, Hydroboracite, Ulexite, and Colemanite) in supplying B to alfalfa (Medicago sativa L.). Boron was incorporated at rates of 0, 1, or 2 mg B kg−1 into Norfolk (fine-loamy, kaolinitic, thermic, Typic Kandiudults) or a Cecil (fine, kaolinitic, thermic Typic Kanhapludults) soil and planted with alfalfa. After establishment, shoots were harvested four times at approximately five-week intervals and analyzed for dry matter and B. Following the last harvest, soil samples were taken and analyzed for Mehlich-1 extractable B. There was no effect of fertilizer material, application rate, or their interaction on the dry weight of alfalfa. Boron concentrations in alfalfa did not significantly decline with each successive cutting, indicating that a sufficient amount was applied for the duration of the study. In the first cutting, plants grown with Granubor and Ulexite had higher concentrations of B than those with Hydroboracite, which was in turn higher than Colemanite, reflecting differences in solubility. However, by the third and fourth cuttings, there were no significant differences in tissue B concentrations among the four fertilizer sources. This indicates that given adequate time, even less-soluble fertilizer sources may provide sufficient B to growing plants.}, number={4-5}, journal={JOURNAL OF PLANT NUTRITION}, author={Byers, DE and Mikkelsen, RL and Cox, FR}, year={2001}, pages={717–725} }
 @article{george_magbanua_tubana_quiton_almendras_khatib_cox_yost_2000, title={Estimating buffer coefficients for the phosphorus decision support system: Preliminary comparisons among field and laboratory estimates from three Asian upland soils}, volume={31}, ISSN={["0010-3624"]}, DOI={10.1080/00103620009370567}, abstractNote={Abstract A phosphorus decision support system (PDSS) has been developed to diagnose P deficiency and predict P fertilizer requirements for Ultisols and Oxisols. Uncertainty analysis has indicated that a major source of uncertainty in prediction was the value of the buffer coefficient (BC, the increase in extractable P per unit of applied P). In the current implementation of PDSS, BC is predicted using an equation that relates BC with soil clay content. We compared PDSS estimates of BC with those obtained by regressing extracted Mehlich 1 P values against amounts of P applied to soil, (a) in field experiments at crop harvest and one year after P addition, and (b) in laboratory wet‐to‐dry incubations with 1:1 soil:P solution mixture. In the field, P amounts were added to upland rice‐soybean rotations at 0 to 236 mg P dm‐3 at Matalom and 0 to 166 at Siniloan, Philippines, and at 0 to 275 at Sitiung, Indonesia, sites in the IRRI Long Term P Experiment (LTPE) network. In the laboratory, P rates of 0 to 272 mg P dm‐3 were added and the soil dried at 30 C for 10 days. The PDSS estimates of BC were 0.22 at Matalom (35% clay), 0.10 at Siniloan (54% clay) and 0.07 at Sitiung (63% clay). The field BC values decreased over time as expected from 0.25 at crop harvest to 0.14 after one year at Matalom and from 0.21 to 0.11 at Siniloan but remained unchanged at 0.09 and 0.11 at Sitiung. The laboratory estimates were similar for soil sampled at different times and were close to the one‐year field estimates with values averaging 0.16 at Matalom, 0.07 at Siniloan and 0.12 at Sitiung. These estimates, however, differed somewhat from both field estimates at crop harvest and PDSS predictions. While PDSS predictions were generally in the range of the field and laboratory estimates, there was a lack of consistency in the rankings in this limited range of sites and clay content. A much larger set of sites and comparisons would be needed to further refine the prediction of BC. Notes Joint contribution of the International Rice Research Institute (IRRI) and the University of Hawaii/Soil Management CRSP.}, number={11-14}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={George, T and Magbanua, R and Tubana, B and Quiton, J and Almendras, A and Khatib, W and Cox, F and Yost, R}, year={2000}, pages={2101–2110} }
 @article{cox_hendricks_2000, title={Soil test phosphorus and clay content effects on runoff water quality}, volume={29}, ISSN={["0047-2425"]}, DOI={10.2134/jeq2000.00472425002900050027x}, abstractNote={Abstract Dissolved phosphorus (P) in runoff increases with an increase in soil test P, but slopes found for this relationship vary by an order of magnitude. A difference in clay content contributes to this variation. Experiments were conducted on Ultisols of 5 and 32% clay by incorporating poultry ( Gallus gallus domesticus ) litter and triple superphosphate (TSP) to create a wide range in Mehlich 3‐extractable phosphorus (M3P). Mehlich 3 P increased similarly for the two sources with rate of P applied, but the increase was greater as clay content decreased. Total P in runoff was measured and it was related primarily to the quantity of sediment. Dissolved P (mg L −1 ) was measured with simulated rainfall and it increased with an increase in M3P (mg dm −3 in the surface 15–20 cm of soil). The slope of dissolved P versus M3P was 0.0040 for the 5% clay soil and 0.0014 for the 32% clay soil. With these relationships, M3P would need to be 250 mg dm −3 at 5% clay and 714 mg dm −3 at 32% clay to result in 1.0 mg L −1 dissolved P in runoff. Dissolved P in samples of runoff collected from natural rainfall with automatic samplers was substantially less than that with simulated rainfall. Thus, there was a marked difference due to method of measurement as well as the difference noted due to day content. Both factors should be considered if attempting to predict the level of dissolved P in runoff from soil test data.}, number={5}, journal={JOURNAL OF ENVIRONMENTAL QUALITY}, author={Cox, FR and Hendricks, SE}, year={2000}, pages={1582–1586} }
 @article{borkert_cox_1999, title={Effects of acidity at high soil zinc, copper, and manganese on peanut, rice, and soybean}, volume={30}, ISSN={["0010-3624"]}, DOI={10.1080/00103629909370293}, abstractNote={Abstract Plant availability of zinc (Zn), copper (Cu), and manganese (Mn) are affected not only by the extractable concentration, but also by soil pH. However, the effect of acidity on uptake at high soil test levels of these nutrient elements is not clear. Three greenhouse experiments were conducted to evaluate the influence of pH on soil and plant tissue concentrations of Zn, Cu, and Mn and on the growth of peanut (Arachis hypogaea L.), rice (Oryza sativa L.), and soybean [Glycine max (L.) Merr.]. Two Inceptisols and five Ultisols that had received large amounts of Zn and/or Cu were used. Treatments included three replications of five pH levels from 5.0 to 7.0. Mehlich‐3 extractable soil Zn and Cu were affected little between pH 5.0 and 7.0. Likewise, soil Mn was affected little between pH 6.0 and 7.0, but there was an acidity by soil interaction between pH 5.0 and 6.0. Plant Zn increased with increasing soil Zn and decreased with increasing soil pH. Zinc in peanut and soybean also appeared to decrease with higher levels of soil Cu. Plant Mn also decreased with increasing soil pH. Plant Cu was not affected consistently by soil pH. Toxicity was expressed by a decrease in dry weight with increasing soil Zn. Peanut dry weight decreased at soil Zn levels between 60 and 300 mg dm‐3, but it was not clear from these data if this limit would be altered by a change in soil pH. A similar effect was shown for soybean. Soybean dry weight decreased with decreasing pH, but the opposite effect was noted for rice, apparently because of a pH effect on iron (Fe) utilization.}, number={9-10}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Borkert, CM and Cox, FR}, year={1999}, month={May}, pages={1371–1384} }
 @article{fontes_cox_1998, title={Iron deficiency and zinc toxicity in soybean grown in nutrient solution with different levels of sulfur}, volume={21}, ISSN={["0190-4167"]}, DOI={10.1080/01904169809365516}, abstractNote={Abstract A typical symptom of iron (Fe) deficiency in plants is yellowing or chlorosis of leaves. Heavy metal toxicity, including that of zinc (Zn), is often also expressed by chlorosis and may be called Fe chlorosis. Iron deficiency and Zn toxicity were evaluated in soybean (Glycine max [L.] Merr.) at two levels each of Zn (0.8 and 40 μM), Fe (0 and 20 μM), and sulfur (S) (0.02 and 20 mM). Reduction in dry matter yield and leaf chlorosis were observed in plants grown under the high level of Zn (toxic level), as well as in the absence of Fe. Zinc toxicity, lack of Fe, and the combination of these conditions reduced dry matter yield to the same extent when compared to the yield of the control plants. The symptoms of Zn toxicity were chlorosis in the trifoliate leaves and a lack of change in the orientation of unifoliate leaves when exposed to light. The main symptoms of Fe deficiency were chlorosis in the whole shoot and brown spots and flaccid areas in the leaves. The latter symptom did not appear in plants grown with Fe but under Zn toxicity. It seems that Fe deficiency is a major factor impairing the growth of plants exposed to high levels of Zn. Under Zn toxicity, Fe and Zn translocation from roots to shoots increased as the S supply to the plants was increased.}, number={8}, journal={JOURNAL OF PLANT NUTRITION}, author={Fontes, RLF and Cox, FR}, year={1998}, pages={1715–1722} }
 @article{sobral_aquino_cox_1998, title={Mehlich-3 phosphorus buffer coefficients}, volume={29}, ISSN={["0010-3624"]}, DOI={10.1080/00103629809370065}, abstractNote={Abstract Phosphorus (P) fertilizer recommendations can be improved if the amount of P required to raise the soil test to a desired value is known. The objective of this study was to determine the increase in soil test P per unit of applied P, the P buffer coefficient, in the laboratory by various means and to compare those values with P buffer coefficients from field data. Phosphorus was applied to soils with predominately kaolinitic mineralogy. The mixtures were then incubated for either 180 and 120 d with numerous drying cycles, or for 7 d and 16 h with either one drying cycle or maintained moist. The samples were extracted with the Mehlich‐3 solution and the change in extractable P per unit of applied P was calculated and termed the Mehlich‐3 P buffer coefficient (M3PBC). The M3PBC varied widely among soils, but was linear within the range of 0 to 300 μg P cm‐3 applied for each. The M3PBC decreased with an increase in time, markedly between 16 h and 7 d, then more slowly from 7 d to 180 d. Drying also reduced the M3PBC, especially on the coarser textured soils. Mehlich‐3 P buffer coefficients from ten field studies were related to the laboratory M3PBC values from dried samples. The prediction of Field M3PBC from the 180‐d M3PBC was linear, while those for the 7‐d and 16‐h M3PBC were quadratic. As time is critical in routine analysis, the 16‐h, dried method was selected as most practical. A prediction equation was calculated to estimate the Field M3PBC from the results of the 16‐h, dried method.}, number={11-14}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Sobral, LF and Aquino, BF and Cox, FR}, year={1998}, pages={1751–1761} }
 @article{aquino_sobral_cox_1998, title={Properties of ultisols and oxisols related to Mehlich-3 phosphorus buffer coefficients}, volume={29}, ISSN={["0010-3624"]}, DOI={10.1080/00103629809370016}, abstractNote={Abstract The accuracy of phosphorus (P) fertilizer recommendations can be improved by knowing the soil buffering capacity. Field estimates of the change in Mehlich‐3 P per unit of P applied after one year, termed the Field Mehlich‐3 P Buffer Coefficient (M3PBC), were available from five Ultisols and five Oxisols. Specific surface area (SSA), clay content, and iron (Fe) and aluminum (Al) extractable by both citrate‐bicarbonate‐dithionite (CBD) and oxalate (Oxa) were determined on these soils and the results related to the Field M3PBC. The Field M3PBC was highly correlated linearly with SSA and clay content. Also, Field M3PBC was related well with CBDFe, CBDAl, and their summation. Neither OxaFe, OxaAl, nor their summation was correlated with Field M3PBC. The relationship between Field M3PBC and these factors was exponential. Specific surface area and clay gave the best relationships, but Field M3PBC could also be predicted from the CBD summation of Fe and Al, CBDFe alone, or CBDAl alone, in that order. Knowledge of CBD‐extractable Fe and/or Al can be used to establish the rate of fertilizer required to increase the Mehlich‐3 P to a sufficient level on Ultisols and Oxisols. An estimate of CBDFe is available from soil survey data and it should be applicable within a region.}, number={9-10}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Aquino, BF and Sobral, LF and Cox, FR}, year={1998}, pages={1155–1166} }
 @article{borkert_cox_tucker_1998, title={Zinc and copper toxicity in peanut, soybean, rice, and corn in soil mixtures}, volume={29}, ISSN={["0010-3624"]}, DOI={10.1080/00103629809370171}, abstractNote={Abstract Applications of zinc (Zn) and copper (Cu) at excessive rates may result in phytotoxicity. Experiments were conducted with mixtures of soils that were similar except for their Zn and Cu levels. The critical toxicity levels (CTL) in the soils and plants for these elements were determined. Peanut (Arachis hypogaea L.), soybean [Glycine max (L.) Merr.], corn (Zea mays L.), and rice (Oryza sativa L.) were the crops grown. One soil mixture had Mehlich 3‐extractable Zn concentrations up to 300 mg dm‐3 with no corresponding increase in soil Cu; two soil mixtures had soil Zn concentrations up to 400 and 800 mg dm‐3 with a corresponding increase in soil Cu up to 20 and 25 mg dm‐3, respectively; and four soil mixtures had no increase in soil Zn, but had Mehlich 1‐extractable Cu concentrations from 6 to 286 mg kg‐1. Under a given set of greenhouse conditions, the estimated Mehlich 3‐extractable Zn CTL was 36 mg dm‐3 for peanut, 70 mg dm‐3 for soybean, between 160 and 320 mg dm‐3 for rice, and >300 mg dm‐3 for corn. No soil Cu CTL was apparent for peanut or soybean, but for corn it was 17 mg dm‐3 and for rice 13 mg dm‐3. With different greenhouse procedures and the Mehlich 1 extractant, the soil CTL for rice was only 4.4 mg kg‐1. Therefore, peanut and soybean were more sensitive to Zn toxicity, whereas corn and rice were more sensitive to Cu toxicity. Plant Zn CTL for peanut was 230 mg kg‐1, while that for soybean was 140 mg kg‐1. Copper appeared to be toxic to corn and rice at plant concentrations exceeding 20 mg kg‐1.}, number={19-20}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Borkert, CM and Cox, FR and Tucker, MR}, year={1998}, pages={2991–3005} }
 @article{fontes_cox_1998, title={Zinc toxicity in soybean grown at high iron concentration in nutrient solution}, volume={21}, ISSN={["0190-4167"]}, DOI={10.1080/01904169809365517}, abstractNote={Abstract Abstract Iron (Fe) deficiency in plants may be caused by heavy metal toxicity and is expressed mainly by chlorosis in young leaves. Zinc (Zn) is often the heavy metal involved. The growth of soybean (Glycine max [L.] Merr.) in a 40 μM Zn nutrient solution was studied using a factorial 2x2x2 experiment; two Fe (20 and 100 μM), two S (0.02 and 20 mM), and two Fe foliar fertilization (with and without Fe‐DTPA) levels. Lower dry matter yield, chlorosis in young leaves, and change in the periodic movement (light/dark) of unifoliate leaves were the main symptoms of Zn toxicity. Plants supplied with 100 μM Fe produced more dry matter than those supplied with 20 μM and did not show leaf chlorosis, but still showed a change in the periodic movement of the leaves. A high Fe supply prevented most of the detrimental effects of toxic Zn. There was no effect of sulfur (S) on plant symptoms, and foliar fertilization with Fe‐DTPA did not result in regreening of chlorotic leaves.}, number={8}, journal={JOURNAL OF PLANT NUTRITION}, author={Fontes, RLF and Cox, FR}, year={1998}, pages={1723–1730} }
 @article{fontes_cox_1997, title={Rhythmic movement of leaves in soybean plants grown under Zn toxicity at different doses of Fe and S}, volume={9}, number={2}, journal={Revista Brasileira De Fisiologia Vegetal}, author={Fontes, R. L. F. and Cox, F. R.}, year={1997}, pages={131–134} }
 @article{chen_yost_li_wang_cox_1997, title={Uncertainty analysis for knowledge-based decision aids: Application to PDSS (phosphorus decision support system)}, volume={55}, ISSN={["0308-521X"]}, DOI={10.1016/S0308-521X(97)00021-8}, abstractNote={The uncertainty associated with predictions generated by computer decision support systems is often not quantified. First order uncertainty analysis was adapted to quantify the uncertainty in the prediction of phosphorus requirements by the phosphorus decision support system (PDSS). Uncertainties were evaluated on the basis of input values of each variable rather than using the population mean and standard deviation as in standard first order uncertainty analysis. The comparison between first order uncertainty analysis and Monte Carlo simulation showed a good agreement when the coefficient of variation of the buffering coefficient was small. Components of overall uncertainty in PDSS were identified and compared. Crop critical level and buffer coefficient contributed the most uncertainty —á more than variation in extractable P or rate of the P slow reaction. The identification and ranking of the source of variation helps prioritize research to improve recommendation precision.}, number={3}, journal={AGRICULTURAL SYSTEMS}, author={Chen, G and Yost, RS and Li, ZC and Wang, X and Cox, FR}, year={1997}, month={Nov}, pages={461–471} }
 @article{cox_1996, title={Economic phosphorus fertilization using a linear response and plateau function}, volume={27}, ISSN={["0010-3624"]}, DOI={10.1080/00103629609369575}, abstractNote={Abstract The linear response and plateau model is now used extensively to depict the relationship between crop yield and soil test level. This function has not been subjected to economic analysis extensively as the variation in critical level is seldom known. Yield data for corn, soybean, and wheat grown on a sandy Typic Umbraquult, which has a low phosphrous (P) sorption capacity, for 13 years were available from a residual P experiment. The study had six replications with 20 observations each, so critical levels of Mehlich‐3‐extractable P (M3P) were calculated on each replication. The critical level mean and standard deviation per crop were used to determine the probability of the critical level being within a 2.5 mg/L increment of M3P. Expected profit was calculated by summation of the gross income minus P fertilizer and other production costs for each increment at selected M3P recommendations. The yield critical levels for the three crops were between 25 and 32 mg/L and the standard deviations 23 to 31% of the mean. With current crop prices and fertilizer costs the economic critical levels were between 31 and 35 mg/L, so a mean of 33 mg/L seems applicable for these three crops. This value is about 14% greater than the average yield critical level when using the linear response and plateau model, indicating that for most field crops the soil test level beyond which no fertilization is suggested should be about 1.14 times the average critical level for yield. Certainly, this value should not be greater than 1.5 times the yield critical level. Utilization of this information would result in more economic production of these field crops and less over‐fertilization of soils.}, number={3-4}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={Cox, FR}, year={1996}, pages={531–543} }
 @inbook{cox_1992, title={Current research on organic phosphorus}, number={92-02}, booktitle={Soil water efficiency of sorghum and cowpea as affected by nutrient input and tillage}, publisher={Raleigh, NC : Soil Management Collaborative Research Support Program, North Carolina State University}, author={Cox, F. R.}, editor={Z. Kouyate and Wendt, C. W.Editors}, year={1992}, pages={141} }
 @article{cox_uribe_1992, title={MANAGEMENT AND DYNAMICS OF POTASSIUM IN A HUMID TROPICAL ULTISOL UNDER A RICE-COWPEA ROTATION}, volume={84}, ISSN={["0002-1962"]}, DOI={10.2134/agronj1992.00021962008400040024x}, abstractNote={Abstract Little is known about the role of K fertilization, stover management, and tillage methods on soil K availability as they affect rice ( Oryza sativa L.) and cowpea [ Vigina unguiculata (L.) Walp.] productivity on Ultisols of the humid tropics. The effects of five K rates (0‐120 kg K ha −1 ), returning or removing stover, and three tillage methods (no‐till, strip, and conventional) were evaluated during 12 crops of rice and cowpea grown for a 4‐yr period. Fertilizer K was applied to the first seven crops. The site was a recently cleared, 18‐yr‐old secondary forest in the Peruvian Amazon Basin. The soil was a fine‐loamy, siliceous, isohyperthermic Typic Paleudult. Soils samples were collected at each crop harvest to 90 em in 15‐cm increments. Potassium fertilizer always increased grain yields when stover was removed. Conversely, responses to K additions were seldom obtained when the stover was returned. The extractable K (Modified Olsen) critical level for both upland rice and cowpeas was calculated to be 0.10 cmol L −1 . Returning stover with no K fertilization maintained soil K concentrations above critical levels for both species up to the last crop of the rotation. Residual effects of fertilizer K were prolonged by returning the stover. When stover was returned, subsoil exchangeable K increased with increasing rate of K fertilization. Removal of stover resulted in greater increases in subsoil exchangeable K at the 40 kg K ha −1 rate than at 120 kg K ha −1 , apparently because the higher rate resulted in K fixation. Tillage methods did not affect crop yields.}, number={4}, journal={AGRONOMY JOURNAL}, author={COX, FR and URIBE, E}, year={1992}, pages={655–660} }
 @article{cox_uribe_1992, title={POTASSIUM IN 2 HUMID TROPICAL ULTISOLS UNDER A CORN AND SOYBEAN CROPPING SYSTEM .1. MANAGEMENT}, volume={84}, ISSN={["0002-1962"]}, DOI={10.2134/agronj1992.00021962008400030024x}, abstractNote={Abstract Sub‐optimum soil K levels often limit corn ( Zea mays L.) and soybean [ Glycine max (L.) Merr.] production in humid tropical Ultisols. The objectives of this study were to determine soil and plant critical levels of K and optimum rates of K fertilization for corn and soybean grown in the Amazon Basin of Peru. Two field experiments were conducted on Typic Paleudults, a loam and a sandy loam. Five K rates, ranging from 0 to 120 kg ha −1 , were broadcast and incorporated prior to planting the first crop of a corn, corn, soybean rotation at both sites. At the loam site, K treatments were reapplied at the end of this rotation and three more corn crops grown. The corn stover was returned while the soybean stover was removed. Critical exchangeable K levels for corn were 110 kg ha −1 on the loam and 90 kg ha −1 on the sandy loam, while for soybean it was 75 kg ha −1 for both soils. The critical levels of K in plant tissue at flowering were 13 g kg −1 in corn and 12 g kg −1 in soybean. Soybean seemed to lower exchangeable K to where K was released from non‐exchangeable sources. An annual cropping system of corn and soybean with an intervening cover crop during the rainy season is proposed for the region. Potassium fertilization is recommended only prior to growing corn at 90 kg ha −1 on a loam and 60 kg ha −1 on a sandy loam.}, number={3}, journal={AGRONOMY JOURNAL}, author={COX, FR and URIBE, E}, year={1992}, pages={480–484} }
 @article{cox_1992, title={RESIDUAL VALUE OF COPPER FERTILIZATION}, volume={23}, ISSN={["1532-2416"]}, DOI={10.1080/00103629209368573}, abstractNote={Abstract The residual effect of Cu fertilization is not well quantified. The change in extractable soil Cu with time after a single fertilization was determined in five soils (four Ultisols and one Inceptisol) with humic matter (HM) content ranging from 3.8 to 21.4 g/L. Rates of Cu to 4.48 kg/ha were broadcast in solution and incorporated to 20 cm on each soil. The soils were sampled for six years after application and extracted with Mehlich‐3 solution. The data were fitted to a model based on the fraction of fertilizer Cu predicted to be extractable immediately after application and an exponential decrease in extractable Cu with time. Considerable variation in extractable Cu concentration was found, both among soils and among years within a soil. The model predicted that 47% to 74% of the applied Cu could be extracted from the soil immediately after fertilization. Extractable Cu decreased exponentially with time, but at a very slow rate. Neither the percentage found immediately after fertilization nor the rate of decrease in extractable Cu with time was related to HM, but the HM content of the soils studied was quite low. An application of 4.48 kg Cu/ha to these soils would increase extractable Cu by two standard deviations, a significant measurable increase, for 9 to 27 years, with a mean of about 16 years.}, number={1-2}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={COX, FR}, year={1992}, pages={101–112} }
 @article{cox_1991, title={Interpretive summary of Part 1: factors affecting the availability of phosphorus in soils of the semi-arid tropics}, ISBN={929066200X}, journal={Phosphorous nutrition of grain legumes in the semi-arid tropics}, publisher={Patancheru, Andhra Pradesh, India : International Crops Research Institute for the Semi-Arid Tropics}, author={Cox, F. R.}, editor={C. Johansen, K. K. Lee and Sahrawat, K. L.Editors}, year={1991}, pages={43} }
 @article{cox_1990, title={A note on the effect of soil reaction and zinc concentration on peanut tissue zinc}, volume={17}, DOI={10.3146/i0095-3679-17-1-6}, abstractNote={Abstract Zinc uptake by peanuts (Arachis hypogaea L.) is affected by both soil pH and extractable Zn concentration, but the combined effect of these two factors is not well defined. An experiment with lime rates was conducted using NC7 peanuts that showed an exponential decrease in leaf Zn as the soil pH increased from 4.3 to 6.1. The decrease was very rapid when the soil was more acid, and less rapid as acidity decreased. Plant Zn was also shown to increase quadratically with increasing soil Zn with a data set from Georgia. These two relationships were combined, assuming no interaction exists, to be able to predict peanut tissue Zn as a function of both soil pH and extractable Zn with either the Mehlich-1 or Mehlich-3 solutions. Equations are presented that conform closely with currently assumed values of critical deficient and toxic concentrations in the tissue and soil for peanuts. These should be especially helpful in predicting potential toxicities over a range of pH and soil Zn levels.}, number={1}, journal={Peanut Science}, author={Cox, F. R.}, year={1990}, pages={15} }
 @article{cox_espejo_1990, title={READILY MEASURABLE SOIL PROPERTIES THAT AFFECT THE PHOSPHORUS-FERTILIZER REQUIREMENT OF ULTISOLS}, volume={21}, ISSN={["0010-3624"]}, DOI={10.1080/00103629009368360}, abstractNote={Abstract Phosphorus fertilizer recommendations could be improved mark‐edly if, in addition to the extractable P concentration, some estimate of the P buffering capacity were considered. Measuring the P buffering capacity or its estimate from clay content or surface area, however, is too time consuming for most soil testing laboratories. Soil samples from the southeastern USA and western Spain were analyzed for several properties that either are or could be measured readily. These properties were cation exchange capacity (CEC), sample density (SD), humic matter (HM), and a color index (CHL). Topsoil samples with HM > 20 g/L were eliminated from the set. The P buffering capacity was estimated with a sorption index. This index correlated well with clay content, so clay content was used for comparison with CEC, SD, and CHL. Of these parameters, clay content was best correlated with CHL (r2 = 0.81). Since CHL can be determined very quickly under routine laboratory conditions, it could be utilized for refining the P soil test interpretation and improving subsequent fertilizer P recommendations.}, number={17-18}, journal={COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS}, author={COX, FR and ESPEJO, R}, year={1990}, pages={2079–2088} }
 @article{cox_martin_1974, title={Effect of temperature on time from planting to flowering in Virginia type peanuts (Arachis hypogaea L.)}, volume={1}, DOI={10.3146/i0095-3679-1-2-12}, abstractNote={Abstract Field studies with a planting date variable were utilized to determine an empirical relation between time from planting to first flowering of NC2, NC5, and Florigiant peanuts and minimum and maximum daily temperatures. Two basic types of curvilinear response functions and two heat unit systems, which used linear functions, were compared on the basis of days missed by each prediction. The mathematical expression of the data that gave the least days missed was the daily fraction of time to flowering being the sum of quadratic functions for minimum and maximum temperature. The rate of slope change was greater at the higher end of the temperature range. The relation between time to flowering and minimum temperature was more curvilinear that that for maximum temperature except at higher temperatures. Minimum temperatures below 43° F lengthened the time to flowering for the three varieties. Varietal differences appeared to be expressed more by the relation with daily maximum than with daily minimum temperatures. The expressions calculated should be more accurate for prediction purposes than a linear heat unit system, plus they tend to describe the individual responses to changes in minimum and maximum temperatures. A certain lack of fit for the relation still exists, indicating perhaps some other measure, such as solar radiation, should also be included.}, number={2}, journal={Peanut Science}, author={Cox, F. R. and Martin, C. K.}, year={1974}, pages={86} }
 @inbook{cox_kamprath_1972, title={Micronutrient soil tests}, booktitle={Micronutrients in agriculture:  proceedings of a symposium held at Muscle Shoals, Alabama, April 20-22, 1971}, publisher={Madison:   Soil Science Society of America}, author={Cox, F. R. and Kamprath, E. J.}, year={1972}, pages={289} }