SciELO - Scientific Electronic Library Online

vol.75 issue2Electrophysiological and behavioral responses of pea weevil Bruchus pisorum L. (Coleóptera: Bruchidae) to volatiles collected from its host Pisum sativum LEffects of supplemental irrigation on water consumption characteristics and grain yield in different wheat cultivars author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Chilean journal of agricultural research

On-line version ISSN 0718-5839

Chilean J. Agric. Res. vol.75 no.2 Chillán June 2015 


Phosphorus use efficiency in pima cotton (Gossypium barbadense L.) genotypes

Elcio Santos1*, Nericlenes Marcante2, Takashi Muraoka1, and Marcos Camacho3

1Universidade de São Paulo, Centro de Energía Nuclear na Agricultura, Av. Centenario, 303, Sao Dimas, Piracicaba, São Paulo, Brasil. *Corresponding author (
Universidade de São Paulo, Escola Superior de Agricultura "Luiz de Queiroz", Av. Pádua Dias 11, Piracicaba, São Paulo, Brasil.
Universidade Estadual de Mato Grosso do Sul (UEMS), Rodovia Aquidauana/UEMS km 12, CEP 79200-000, Aquidauana, Mato Grosso do Sul, Brasil.

In the Brazilian Cerrado, P deficiency restricts cotton production, which requires large amounts of phosphate fertilizer. To improve the yield of cotton crops, genotypes with high P use efficiency must be identified and used. The present study evaluated P uptake and use efficiency of different Gossypium barbadense L. genotypes grown in the Cerrado. The experiment was carried out in a greenhouse with a completely randomized design, 15 x 2 factorial treatment structure (15 genotypes x 2 P levels), and four replicates. The genotypes were MT 69, MT 70, MT 87, MT 91, MT 92, MT 94, MT 101, MT 102, MT 103, MT 105, MT 106, MT 110, MT 112, MT 124, and MT 125; P levels were sufficient (1000 mg pot-1, PS treatment) or deficient (PD treatment). Dry matter (DM) and P levels were determined in cotton plant parts and used to calculate plant P content and use efficiency. In general, DM and P content were higher in the PS than in the PD treatment, with the exception of root DM and total DM in some genotypes. Genotypes also differed in terms of P uptake and use capacity. In the PS treatment, genotypes MT 92 and MT 102 had the highest response to phosphate fertilization. Genotype MT 69 exhibited the most efficient P uptake in the PD treatment. Genotype MT 124 showed the best shoot physiological efficiency, apparent recovery efficiency, and utilization efficiency, whereas MT 110 exhibited the highest root physiological efficiency.

Key words: Nutritional efficiency, phosphorus supply, pima cotton.


Pima cotton (Gossypium barbadense L.) is widely distributed in Brazil and is found in all regions. This species originated in northern Peru and southern Ecuador and is endemic to South America (Zhang et al., 2002). The use of G. barbadense as a source of genetic rusticity in interspecific crossing with G. hirsutum L. is capable of providing features, such as tolerance to soils with low P content, which is characteristic of tropical soils. In the second half of the 1990s, cotton crops in Brazil migrated from traditional croplands to the Cerrado because farmers saw an excellent business opportunity for growing cotton in rotation with soybean. Currently, 84% of cotton production in Brazil is located in the Cerrado. Nevertheless, soil in the Cerrado exhibits certain restrictions, such as high acidity, low fertility, and low P content and availability (Pavinato et al., 2009).

In Brazil, P is one of the main nutrients used in fertilization (Pavinato et al., 2010) because plants generally exhibit low P recovery from phosphate fertilizers. Soil P availability is low due to high soil iron and aluminum oxide content (Dorahy et al., 2007), which can be positively charged, thereby adsorbing phosphate anions from particle surfaces (Pavinato et al., 2009). Phosphorus fertilizers are mainly produced from phosphate rocks (Corrêa and Sharma, 2004), which is a limited natural resource. A number of studies have sought sustainable procedures for P use that preserve natural sources. One solution is to combine the use of plants with good P use efficiency and good farming practices (Shujie and Yunfa, 2011). However, there are only a few studies that have screened cotton for P efficiency.

The selection of genotypes that tolerate a wide range of nutritional stress conditions has been one of the main contributors to increasing crop yield (Zambrosi et al., 2012). Phosphorus-efficient genotypes within a species are able to grow well and produce high yields in soils with low available P; they are therefore tolerant to P deficiency (Fageria et al., 2010). Considerable intraspecific variations in P efficiency have been identified in a variety of plant species, including cowpea (Araújo et al., 2012), rice (Fageria and Moraes, 2013), and citrus (Zambrosi et al., 2012). The evolution of pima cotton has led to a highly developed root system to penetrate deep soil layers in its natural habitat (Zhang et al., 2002). Thus, we believe that a vigorous root system can give pima cotton enhanced P efficiency and consequently, provide a source of genetic variation to improve cotton P efficiency.

The present study evaluated P uptake and use efficiency of different Gossypium barbadense genotypes grown in the Cerrado.


The experiment was carried out in arched greenhouses in the Aquidauana unit of the State University of Mato Grosso do Sul (20°27' S, 55°40' W; 174 m a.s.l.), Brazil. According to Koppen-Geiger's classification, climate in the area is Aw (wet tropical) and characterized as tropical wet with a rainy season in the summer and dry in the winter. The 6 dm3 pots were filled with 5 kg of sandy textured Argissolo Vermelho-Amarelo (Oxisol) soil from the top layer (020 cm), (Embrapa, 2006), which was sieved in a 2-mm mesh. Before the experiment, soil chemical properties were determined according to methods proposed by van Raij et al. (2001): 5.6 pH, 15 g dm-3 organic matter, 4.1 mg dm-3 available P (resin-extractable), 1.5 mmolc K dm-3, 30 mmolc Ca dm-3, 14 mmolc Mg dm-3, 18 mmolc H + Al dm-3, 63.5 mmolc dm-3 cation exchange capacity (CEC), 45.5 mmolc dm-3 sum of bases, and 72% base cation saturation. Soil textural analysis was 340 g clay, 100 g silt, and 660 g sand per kg of soil.

The experiment was completely randomized, arranged in a 15 x 2 factorial design with four replicates, and totaled 120 pots. The factors under consideration were G. barbadense genotypes (15 levels) and soil P levels, which were sufficient (PS) or deficient (PD). The genotypes, obtained from the Embrapa germplasm bank, were identified as MT 69, MT 70, MT 87, MT 91, MT 92, MT 94, MT 101, MT 102, MT 103, MT 105, MT 106, MT 110, MT 112, MT 124, and MT 125. All the genotypes were collected in the state of Mato Grosso. Three cotton seeds were sown in each pot. Seven days after emergence, plants were thinned to one per pot. An adaptation of Johanson's solution (Epstein and Bloom, 2006) was applied to fertilize the treatments. It contained 4 mL KNO3 solution 1 mol L-1, 6 mL NH4NO3 solution 1 mol L-1, 2 mL MgSO4 7H2O solution 1 mol L-1, 4 mL CaCh2 solution 1 mol L-1, 2 mL KCL solution 1 mol L-1, 1.0 mL MnSO4 H2O solution 0.002 mol L-1, 1 mL ZnSO4 7H2O solution 0.002 mol L-1, and 1 mL CuSO4 5H2O solution 0.0005 mol L-1. Phosphorus was applied to plants in the PS treatment at 1000 mg pot-1 (200 mg P kg-1 soil). The P source was monobasic ammonium phosphate, which was diluted in deionized water before being administered. The PD plants did not receive any additional P, but N addition at 1000 mg pot-1 (200 mg P kg-1 soil) in plating. The N source used in the PD plants was urea. Both treatments were watered when necessary.

At 195 d after emergence (DAE), plants were collected, separated into shoots and roots, washed in deionized water, and oven-dried at 65 °C for 72 h to determine root dry matter (RoDM), shoot dry matter (ShDM) and total dry matter (ToDM). The material was then ground in a Wiley-type stainless steel grinding mill. Samples were analyzed to determine P levels by the colorimetric method (Malavolta et al., 1997). Plant DM production and P level data were used to calculate P content in root (RoP), shoot (ShP), and the entire plant (ToP, total P). The following efficiency parameters were calculated: uptake efficiency (UpE, mg g-1) = ToP/RoDM, transport efficiency (TrE, mg mg-1) = ShP/ToP x 100, biological utilization efficiency (BUE, g2 mg-1) = ToDM2/ToP, shoot physiological efficiency (ShPE, g mg-1) = (ShDMPS - ShDMPD)/(ShPPS - ShPPD) root physiological efficiency (RoPE, g mg-1) = (RoDMPS - RoDMPD)/(RoPPS - RoPPD) apparent recovery efficiency (ARE, percentage) = (ShPPS - ShPPD)/Papplied x 100, and utilization efficiency (UtE, g g-1) = ShPE x ARE where ToP is total plant P content (mg), ShP is shoot P content (mg), ShDMPAdeq is shoot DM in plants under PS, ShDMPD is shoot DM in plants under PD, ShPPS is shoot P content (mg) in plants under PS, ShPPD is shoot P content (mg) in plants under PD, RoDMPS is root DM in plants under PS, RoDMPD is shoot DM in plants under PD, RoPPS is root P content (mg) in plants under PS, RoPPD is root P content (mg) in plants under PD, and Pappiied is the amount of applied P (mg).

Data were analyzed by descriptive statistics and factorial ANOVA as follows:

Yijk = μ + αi+ βj + (αβ)ij + εijk

where μ is the overall mean, αi is the effect of the ith phosphate fertilization level, βj is the effect of the jth genotype, (αβ)ij is the combined effect of the ith phosphate fertilization level and jth genotype, and εijk is the random error effect.

Means were compared according to the F-test (soil P level) and Student's t-test (genotypes) with the SAS package (SAS Institute, Cary, North Carolina, USA).


Dry matter and phosphorus content
In general, all genotypes showed significantly reduced growth under limited P supply. However, an exception was observed for the genotypes MT 69, MT 103, MT 105, MT 110, and MT 124 in which RoDM production did not change as a function of P application (Table 1). The potential for root system development of these genotypes is therefore independent of soil P levels. This characteristic is desirable because a well-developed root system can increase water and nutrient uptake (Katsvairo et al., 2007). According to Fageria and Moraes (2013), nutrient uptake is usually proportional to root development. Only in genotype MT 103 was ToDM production not affected by P application (Table 1); this is probably because it is less responsive to phosphate fertilization. A decrease in ToDM as a result of P application is also a useful parameter to evaluate the relative tolerance of cotton varieties to soil P deficiency (Dorahy et al., 2004). Cotton genotypes that do not change DM production in response to P supply (sufficient or deficient) are best grown in soils with low P content (Wang et al., 2008).

Table 1. P-value of F-test for dry matter production, phosphorus uptake, and efficiency parameters of Gossypium barbadense genotypes grown under sufficient and deficient phosphate fertilization.

RoDM: Root dry matter, ToDM: total dry matter, RoP: root P content, ToP: total P content, UpE: uptake efficiency, TrE: transport efficiency, BUE: biological utilization efficiency.
Statistical significance was set at 0.05 error probability.

Genotypes produced different RoDM and ToDM in the PS and PD treatments (Table 2). In the PS treatment, RoDM varied significantly from other genotypes, ranging from 4.89 (MT 110) to 13.28 g pot-1 (MT 92). In the PD treatment, the highest RoDM was obtained in MT 105 with 6.43 g pot-1. Genotypes capable of producing high RoDM under nutritional stress are likely to achieve high agronomic yield given that the development of an extensive root system allows the exploitation of a larger soil volume (Fageria and Moraes, 2013). In the PS treatment, the range of variation in ToDM production was higher (24.32 g pot-1) than in the PD treatment (8.15 g pot-1) (Table 2). The large variation in the PS treatment indicates that G. barbadense genotypes are more responsive to P application. Responsive genotypes are more productive and can thus be used as a criterion for selecting and evaluating varieties according to their nutritional efficiency (Fageria et al., 2010).

Table 2. Mean values for root dry matter (RoDM) and total dry mater (ToDM) of Gossypium barbadense genotypes grown under sufficient (PS) and deficient (PD) phosphate fertilization.

CV: Coefficient of variance, MSD: minimum significant difference.

Root P (RoP) and entire plant (ToP, total P) content was affected by P application in most evaluated genotypes. Genotypes that did not change RoP in the PS treatment were MT 69, MT 103, MT 110, and MT 124, and those that did not change ToP were MT 69 and MT 103 (Table 1). The nature of the RoP and ToP variation was specific to the different genotypes (Table 3). In the PS treatment, accumulated RoP ranged from 21.93 to 6.36 mg among genotypes, and ToP was higher in genotypes MT 92 (96.29 mg), MT 101 (81.66 mg), and MT 124 (70.60 mg). In the PD treatment, some genotypes showed significant ToP loss with 91% reduction for MT 92 and 87% for MT 101 (Table 3). This significant decrease indicates the inability of these genotypes to accumulate P under P-deficiency stress conditions (Fageria et al., 2010).

Table 3. Mean phosphorus content in roots (RoP) and entire plant (ToP) of Gossypium barbadense genotypes grown under sufficient (PS) and deficient (PD) phosphate fertilization.

CV: Coefficient of variance, MSD: minimum significant difference.

P uptake, transport, and biological utilization efficiency
The current study allowed the determination of the genotypic variation in P efficiency and the observation of the nutritional features of pima cotton under sufficient and deficient P supply conditions. No similar study about pima cotton has yet to be reported. Uptake efficiency, TrE, and BUE varied according to P application (Table 1) and genotypes (Table 4). Genotypes MT 102, MT 105, and MT 110 improved UpE when cropped in the PS treatment (Table 1). Only the remaining genotypes exhibited efficient P uptake because their results were similar in both the PD and PS treatments. In the PS treatment, genotypes MT 69, MT 91, MT 92, MT 101, MT 102, and MT 110 showed the most efficient P uptake. The P supply resulted in increased UpE for these genotypes (Table 1), suggesting higher nutrient concentrations in the soil that promote better efficiency in roots in the absorption process, as already observed for other crops (Zambrosi et al., 2012). Uptake efficiency mostly increased because of the positive effect of P supply in RoP and ToP (Table 3) under sufficient phosphate fertilization. In the PD treatment, only MT 69, MT 91, MT 92, and MT 94 had a greater capacity to extract soil nutrients (Table 4). Different UpE in genotypes observed in the present study can result from diversiied plant morphology, particularly the ability to develop secondary roots (Baligar et al., 2001; Lynch and Ho, 2005). These peculiarities can be used to indicate genotypes that can cope with P-deficiency stress and P uptake.

Table 4. Phosphorus uptake and use efficiency (mean valúes) of Gossypium barbadense genotypes grown under sufficient (PS) and deficient (PD) phosphate fertilization.

UpE: uptake efficiency, TrE: transport efficiency, BUE: biological utilization efficiency,
CV: coefficient of variance, MSD: minimum significant difference.

Comparing treatments for each genotype (MT 70, MT 94, MT 103, and MT 110) showed no changes in TrE in response to P application (Table 1). Genotypes MT 70, MT 94, and MT 103 increased TrE (Tables 1 and 4) in the PD treatment, indicating that nutrient transport from root to shoot is more eficient under P-deiciency stress conditions (Baligar et al., 2001). On the other hand, genotype MT 110 drastically reduced TrE when it was cropped in the PD treatment, suggesting that P transport in this variety is promoted by an increase in soil P. Genotypes MT 69, MT 70, MT 91, MT 92, MT 101, and MT 102 exhibited high TrE in both the PS and PD treatments (Table 4), reflecting high productive potential. These genotypes are responsive to P application and eficient in transporting this nutrient. A study with beans showed a significant relationship between crop production and P transport from root to shoot (Fageria et al., 2010).

Only genotypes MT 101, MT 105, and MT 110 exhibited the same BUE between the PS and PD treatments, that is, the same capacity for converting absorbed nutrients into DM (Table 1). In the PS treatment, BUE ranged from 16.27 (MT 124) to 5.99 (MT 110) g g-1, while in the PD treatment, it ranged from 9.67 to 1.59 g g-1. In another study, significant BUE differences were observed in six cotton genotypes cropped under both low and sufficient P conditions (Wang et al., 2008). In the present study, it was higher in MT 102, MT 105, and MT 112 for the PS and PD treatments (Table 4). High BUE genotypes are recommended for cropping in soil with low P availability because they can better redistribute minerals from senescent organs by reutilizing these nutrients in metabolic processes such as plant growth (Fageria et al., 2010). Under P-deiciency conditions, BUE is increased as a result of changes in activity of glycolysis enzymes on the route, which will start operating in alternative pathways of carbohydrate catabolism during disability, contribute to more P use by eficient recycling reduction consumption, and use other fractions of the nutrient (Mengel and Kirkby, 2001; Marschner, 2012).

Plant properties and characteristics related to P uptake and utilization are heritable and will, no doubt, be exploited more vigorously as fertilizer costs continue to increase for cotton cultivation. Inter- and intraspeciic variations for plant growth and mineral nutrient use eficiency are known to be under genetic and physiological control and modiied by plant interactions with environmental variables (Baligar et al., 2001). Baligar et al. (2001) also reported that identifying traits, such as nutrient absorption, transport, utilization, and mobilization, in plant cultivars should greatly enhance fertilizer use eficiency.

Physiological efficiency, apparent recovery, and utilization efficiency of P
Data on ShPE, ARE, UtE, and RoPE are shown in Table 5. These eficiency parameters indicate the relative production of the genotypes in soils that are deicient compared with production in soils containing optimal nutrient levels. The highest ShPE values were found in MT 91, MT 94, MT 103, and MT 124 with ShDM of 0.89, 1.14, 1.55, and 0.81 g (per g applied P), respectively (Table 5). Genotype MT 110 showed the highest root production (RoPE = 1.47 g g-1) and the lowest shoot production. Physiological eficiency indicates higher DM production capacity with suficient nutritional supply compared with a nutritional stress situation (Fageria et al., 2009). Thus, MT 110 was more efficient in producing roots than shoots. Genotypes with this characteristic can be suggested for cropping under low P supply conditions where plants must be more eficient in producing roots that achieve higher soil exploitation. When P limits plant growth, roots become a strong carbohydrate sink, limiting shoot rather than root growth (Wang et al., 2010). Bean roots grown under P-deiciency conditions show higher sugar levels than those grown under suficient P supply because of higher photoassimilate transport from shoot to root (Wang et al., 2008).

Table 5. Physiological phosphorus use efficiency (mean values) of Gossypium barbadense grown under sufficient and deficient phosphate fertilization.

ShPE: shoot physiological efficiency, RoPE: root physiological efficiency, ARE: apparent recovery efficiency,
UtE: utilization efficiency, CV: coefficient of variance, MSD: minimum significant difference.

With regards to ARE, it was higher in MT 101, MT 92, and MT 124 than in other genotypes, and it exhibited, on the average, 6% of accumulated P in the plant per applied P unit. This result shows the best P recovery from fertilizer by these genotypes (Wang et al., 2008). Cotton response to soil P application in soil in the Cerrado is higher than that of rice cropped under similar conditions in the highlands. Santos and Fageria (2007) found that crop yield is positively related to nutrient recovery. For UtE, MT 92 and MT 124 were the most eficient genotypes with 22.41 and 20.24 g g-1 , respectively. Utilization eficiency is the result of a positive interaction between uptake, transport, and nutrient redistribution. The high UtE of these genotypes was mainly expressed by the corresponding ARE means, which were also high. Utilization eficiency is usually associated with productivity, that is, the higher the UtE, the higher the grain production (Shujie and Yunfa, 2011).


Signiicant differences in shoot dry weight, shoot P concentration, P uptake, and P use index were found among the 15 pima cotton genotypes under suficient and deicient P supply. Pima cotton genotypes differed regarding uptake capacity and P use: MT 92 and MT 102 were the most responsive to P application, MT 69 was the most eficient in P uptake in crops grown under low P levels, MT 124 produced the highest shoot physiological eficiency, apparent recovery eficiency, and utilization eficiency, and MT 110 showed the highest root physiological eficiency.


We thank the Fundação de Apoio ao Desenvolvimento do Ensino, Ciencia e Tecnologia do Estado de Mato Grosso do Sul (Fundect) for inancial support, and to EMBRAPA for providing the genotypes tested in the study.


Araújo, E.O., E.F. Santos, G.Q. Oliveira, M.A., Camacho, and D.M. Dresch. 2012. Nutritional efficiency of cowpea varieties in the absorption of phosphorus. Agronomía Colombiana 30:419-424.         [ Links ]

Baligar, V.C., N.K. Fageria, and Z.L. He. 2001. Nutrient use efficiency in plants. Communications in Soil Science and Plant Analysis 32:921-950.         [ Links ]

Corrêa, J.C., e R.D. Sharma. 2004. Produtividade do algodoeiro herbáceo em plantio direto no cerrado com rotação de culturas. Pesquisa Agropecuária Brasileira 39:41-46.         [ Links ]

Dorahy, C.G., G.J. Blair, I.J. Rochester, and A.R. Till. 2007. Availability of P from 32P-labelled endogenous soil P and 33P-labelled fertilizer in an alkaline soil producing cotton in Australia. Soil Use and Management 23:192-199.         [ Links ]

Dorahy, C.D., I.J. Rochester, and G.J. Blair. 2004. Response of field grown cotton (Gossypium hirsutum L.) to phosphorus fertilization on alkaline soils in eastern Australia. Australian Journal of Soil Research 42:913-920.         [ Links ]

Embrapa. 2006. Centro Nacional de Pesquisa de Solos. Sistema brasileiro de classificação de solos. 2a ed. 306 p. Empresa Brasileira de Pesquisa Agropecuária (Embrapa), Rio de Janeiro, Brasil.         [ Links ]

Epstein, E., e A.J. Bloom. 2006. Nutrição mineral de plantas: Principios e perspectivas. 2a ed. 399 p. Editora Planta, Londrina, Paraná, Brasil.

Fageria, N.K., V.C. Baligar, A. Moreira, and T.A. Portes. 2010. Dry bean genotypes evaluation for growth, yield components and phosphorus use efficiency. Journal of Plant Nutrition 33:2167-2181.         [ Links ]

Fageria, N.K., and M.F. Moraes. 2013. Phosphorus nutrition of lowland rice in tropical lowland soil. Communications in Soil Science and Plant Analysis 44:2932-2940.         [ Links ]

Fageria, N.K., A.B. Santos, and V.A. Cutrim. 2009. Nitrogen uptake and its association with grain yield in lowland rice genotypes. Journal of Plant Nutrition 32:1965-1974.         [ Links ]

Katsvairo, T.W., D.L. Wright, J.I. Marois, D.L. Hartzog, K.B. Balkcom, PP. Wiatrak, et al. 2007. Cotton roots, earthworms, and infiltration characteristics in sod-peanut-cotton cropping systems. Agronomy Journal 99:390-398.         [ Links ]

Lynch J.P, and M.D. Ho. 2005. Rhizoeconomics: carbon costs of phosphorus acquisition. Plant and Soil 69:45-56.         [ Links ]

Malavolta, E., G.C. Vitti, e S.A. Oliveira. 1997. Avaliação do estado nutricional das plantas: principios e aplicações. 2a ed. 319 p. Potafos, Piracicaba, São Paulo, Brasil.         [ Links ]

Marschner, P 2012. Marschner's mineral nutrition of higher plants. 3rd ed. 651 p. Academic Press, London, UK.         [ Links ]

Mengel, K., and E. Kirkby. 2001. Principles of plant nutrition. 5th ed. 849 p. Kluwer Academic Publishers, Dordrecht, The Netherlands.         [ Links ]

Pavinato, P.S., T. Dao, and C.A. Rosolem. 2010. Tillage and phosphorus management effects on enzyme-labile bioactive phosphorus availability in Cerrado Oxisols. Geoderma 156:207-215.         [ Links ]

Pavinato, P.S., A. Merlin, and C.A. Rosolem. 2009. Phosphorus fractions in Brazilian Cerrado soils as affected by tillage. Soil & Tillage Research 105:149-155.         [ Links ]

Santos, A.B., e N.K. Fageria. 2007. Manejo do nitrogênio para eficiência de uso por cultivares de feijoeiro em várzea tropical. Pesquisa Agropecuária Brasileira 42:1237-1248.         [ Links ]

Shujie, M., and Q. Yunfa. 2011. Effects of phosphorus concentration on adaptive mechanisms of high- and low-P eficiency soybean genotypes when grown in solution. Plant, Soil and Environment 57:61-66.         [ Links ]

van Raij, B., J.C. Andrade, H. Cantarella, e J.A. Quaggio. 2001. Análise química para avaliação da fertilidade de solos tropicais. 285 p. Instituto Agronômico, Campinas, São Paulo, Brasil.         [ Links ]

Wang, X.R., J. Shen, and H. Liao. 2010. Acquisition or utilization, which is more critical for enhancing phosphorus eficiency in modern crops. Plant Science 179:302-306.         [ Links ]

Wang, X., C. Tang, C.N. Guppy, and P.W.G. Sale. 2008. Phosphorus acquisition characteristics of cotton (Gossypium hirsutum L.), wheat (Triticum aestivum L.) and white lupin (Lupinus albus L.) under P deficient conditions. Plant and Soil 312:117-128.         [ Links ]

Zambrosi, F.C., D. Mattos Junior, R. Furlani, J.A. Quaggio, and R.M. Boaretto. 2012. Eficiency of phosphorus uptake and utilization in citrus rootstocks. Revista Brasileira de Ciência do Solo 36:486-496.         [ Links ]

Zhang, J., W. Guo, and T. Zhang. 2002. Molecular linkage map of allotetraploid cotton (Gossypium hirsutum L. x Gossypium barbadense L.) with a haploid population. Theoretical and Applied Genetics 105:1166-1174.         [ Links ]

Received: 4 June 2014.
Accepted: 24 February 2015.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License