SciELO - Scientific Electronic Library Online

vol.38 issue4Boron deficiencies in the northern and central interior of PortugalOcurrence and efficiency of Sesbaniavirgata (Cav.) Persmicrosymbiont in function of the soil properties author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO


Revista de Ciências Agrárias

Print version ISSN 0871-018X

Rev. de Ciências Agrárias vol.38 no.4 Lisboa Dec. 2015 


Manganese toxicity in Portuguese Cambisols derived from granitic rocks: causes, limitations of soil analyses and possible solutions

Toxicidade de manganês nos Cambissolos derivados de rochas graníticas em Portugal: causas, limitações das análises de solo e soluções possíveis

Mário Carvalho1*, Michael J. Goss2 and Dora Teixeira3


1 Universidade de Évora, Instituto de Ciências Agrárias e Ambientais Mediterrânicas (ICAAM), Largo dos Colegiais, 7000 Évora, Portugal (*corresponding author. E-mail:

2 University of Guelph, School of Environmental Sciences. E-mail:

3 Universidade de Évora, Laboratório HERCULES, Largo dos Colegiais, 7000 Évora, Portugal. E-mail:



Cambisols are the major Reference Soil Group in Portugal. The yield of annual crops in these soils is generally poor, and the situation is aggravated in wet winters. In the south of Portugal, manganese toxicity has been identified as the major cause of poor growth and leaching as the main reason for the negative effect of rainfall observed in Cambisols developed on granitic formations. Manganese toxicity also appears to be present in the Cambisols in other regions of Portugal. Manganese toxicity is cross-related to the magnesium concentration, either in the soil solution or in plant shoots. Therefore, soil amendment using dolomitic limestone is needed to overcome the problem. Current soil test methods are unable to predict the level of Mn toxicity. However, new approach using the extraction of soil solution is proposed, although further work is needed to fully implement the method.

Key words: leaching, magnesium, Mg/Mn ratio, Soil, soil amendment



Os Cambissolos são o Grupo de Referência de Solos mais representativo em Portugal. As produções das culturas anuais são geralmente baixas e a situação agrava-se em anos de Inverno húmido. A toxicidade de manganês foi identificada como a principal causa das produções baixas e a lixiviação o principal efeito negativo dos Invernos húmidos nos Cambissolos derivados de granito no Sul de Portugal. A toxicidade de Mn está relacionada com um desequilíbrio com o magnésio, quer na solução do solo quer na concentração dos dois iões na parte aérea da planta. A solução do problema exige, assim, a aplicação de calcário dolomítico. Os actuais métodos de análise de solo não parecem capazes de prever a ocorrência da toxicidade de manganês. Uma nova abordagem utilizando a extracção da solução do solo é proposta. No entanto, a utilização desta abordagem exige investigação suplementar. A toxicidade de Mn pode não ser restrita aos Cambissolos do sul do país.

Palavras-chave: lixiviação, magnésio, razão Mg/Mn, Solo, correcção do solo



The WRB Reference Soil Group Cambisols (WRB, 2006), mostly derived from granites, quartz-diorites and sandstones, occurs in large areas in Portugal. Cambisols developed on granites, designated by Pg and Pgm in the Portuguese Soil Classification (Cardoso, 1974), are generally characterized by their coarse texture (mostly sandy loam), small cation exchange capacity (CEC) (around 5 cmolc per kg of soil), organic matter content (≤ 1%) and water holding capacity, and strong acidity (pH ≤ 5.5 in water); extractable P is mostly less than 15 mg kg-1, and exchangeable K is between 0.15 and 0.25 cmolc kg-1.

The most significant agricultural system practiced on these soils is the agro-forestry-pastoral Montado system. The most important trees present in the system are the holm oak (Quercus rotundifolia Lam.) and the cork oak (Quercus suber L.). Both provide fruit for animal feed during the winter and the latter yields cork. Animal production is an important component of the Montado system, being based on temporary or permanent pastures and annual forages (usual cereals). Typical yields of these crops are poor and there is a significant and negative relationship between the production of the annual crop, wheat (Trititcum aestivum L.), and winter rainfall, which is an unexpected result given the small water holding capacity of these soils (Carvalho, 1987). In contrast, this same relationship has a positive value for soils with vertic properties of the region, which have a much smaller saturated hydraulic conductivity ( Figure 1). The negative effect of winter rainfall on the growth of the wheat in the Cambisol is more related to leaching than waterlogging, with the growth of the crop being greater under any soil water regime influenced by the tree canopy (Q. rotundifolia in the example; see Figure 2). Outside the influence of the trees, the values of soil pH, organic matter and magnesium contents are less but extractable Mn is greater (Table 1). The growth of the plants was significantly and negatively correlated with the Mn concentration in the shoots (Figure 3). However, the negative effect of leaching was not related to any change of soil pH (Table 2) although leaching increased the Mn and reduced the Mg concentration in the shoots of the wheat (Figure 4).

Manganese toxicity has long been recognized as an important factor limiting plant growth on acid and waterlogged soils. The prediction of Mn toxicity is very complex and the specification of a critical concentration is hardly possible (Horst, 1988). The solubility of Mn is greatly affected by the pH (Leeper, 1970) that can rapidly change in the soil with a low CEC due to mineralization of organic matter and release of root exudates. Carvalho (1987) recorded a reduction of the pH (in water) of a Cambisol in the South of Portugal from 5.5 during the summer to 4.5 in November. The redox potential of the soil, which can rapidly change under field conditions, also affects the solubility of Mn (Ponnamperuma, 1984). Plants may differ considerably within and between species in manganese tolerance due both to genetic characteristics and environmental factors such as nutrient availability in the soil. For example, the presence of other ions such as Fe2+, Ca2+ and Mg2+ can modify the uptake of manganese from solution (Löhnis, 1960; Chinnery and Harding, 1980). Within the plant, Horst and Marschner (1978) showed that expression of toxicity symptoms at a given concentration of manganese depended on the silicon content of the old leaves of Phaseolus vulgaris L.


The significance of Mg to the Mn toxicity

The role of Mn toxicity and the possible implication of effects on Mg in the poor growth of the wheat on the Cambisol in areas away from the influence of trees were investigated in a controlled environment experiment using containers, 7 cm internal diameter and 30 cm deep. The treatments were soil leaching (2 levels: with a bed volume of glass-distilled water or without), application of lime (2 levels: 0 and 2 g of CaCO3 kg‑1 of dry soil) and application of Mg (2 levels: 0 and half a bed volume of a solution containing 1.5 mM magnesium as MgSO4). Three pre-germinated seeds of T. aestivum cv. Mara were sown in each column and allow to growth for three weeks. For more detailed information on experimental procedures see Goss and Carvalho (1992). There was no unique relationship between growth of the wheat and the concentration of manganese in the shoots (Figure 5). An empirical curve could be fitted to results for plants grown in soil unamended with calcium carbonate. However, in the limed soil a wide range of plant growth was observed without any major variation of Mn content in shoot tissue. Nevertheless, when growth (Y) was expressed as a function of the ratio of magnesium to manganese concentrations in the tissue (Rp), a relationship of the form Y = A + B e-kRp accounted for 89% of the variance (Figure 6). The relationship indicated that growth was significantly reduced when the ratio of magnesium and manganese fell below 20:1 on a weight basis or 45.8 on a molar basis. Although liming reduced manganese in shoot tissue, the plants grown on soil amended with lime had less magnesium than those from the comparable treatments on the unamended soil. Consequently liming did not necessarily increase the ratio of the two ions in the shoots. In fact, using the same treatments in a pot experiment, Teixeira (1997) found that the application of CaCO3 to the soil reduced both the concentration of Mg and Mn in the soil solution resulting only in a marginal increase of the ratio of the two ions in soil solution (Figure 7). It is a matter of speculation why Mg is affecting plant tolerance to excessive levels of Mn. The ratio of Mg/Mn in plant cells may affect intracellular Mn distribution. When the ratio is high, Mg2+ may replace Mn2+ from physiologically active sites in the cytoplasm and Mn2+ may be sequestered in cell walls and vacuoles (Le Bot et al., 1990).

To understand the relationship between the concentration on Mg and Mn in the soil solution, the absorption of the two ions by the wheat plants and the level of Mn toxicity, was investigated in a solution culture experiment using 4 concentrations of Mg and 6 of Mn in factorial combination (Table 3). For more details of the experiment see Goss and Carvalho (1992). The maximum concentration of manganese in the shoots of the wheat decreased with an increasing concentration of magnesium in the nutrient solution (Figure 8). This effect is, at least, partially due a reduction in the proportion of manganese translocated from the roots to the shoots, when the concentration of magnesium increases (Figure 9). According to Marschner (1986), Mn2+ not only competes more effectively but also blocks binding sites for Mg2+ uptake. Therefore, the relative concentration of the two ions in the solution might affect both the uptake by the plant and the translocation to the shoots. This competition between the two ions was also observed in chestnut (Castanea sativa Mill.) growing in the north of Portugal in soils derived from Mn-rich schists and greywackes of the Ordovician and the Silurian (Portela et al., 2011).


Limitations of soil analyses

To investigate the possibility of using soil analyses to predict manganese toxicity in the Cambisol used in the experiments referred above, different extraction methods were tested (Table 4). The detailed information of the extraction procedures are referred in Teixeira (1997). The differences in the magnesium and manganese concentration in the soil, as well the ratio of the two ions, given by the different extraction conditions were considerable (Table 5). To compare the values of these parameters in the soil with the respective concentrations in the shoot, wheat plants were grown for four weeks in a pot experiment (1.5 kg of soil per pot). N, P and K were added to the soil at wheat planting (18 mg of NH3NO3 and 7 mg of KH2PO4 per pot). The relationship of the concentration of the two ions in the soil to that in the plant was very poor (data not shown). In Table 6 the significant regressions obtained between the ratio of the two ions in the soil and in the shoots are identified. The coefficients of determination were very poor and the best result was obtained by extracting with KCl 1.0 mol L-1 (extraction time 24 hours) with r2 = 35%. However, for most of the situations the values for r2 were below 20%. Cleary, none of the studied methods seems appropriate to predict Mn toxicity in this soil type.

Goss et al (1992), combining data from pot and solution culture experiments (Figure 10), fitted a single linear regression, that accounted for 93.8% of the variance of the natural logarithm of the ratio of the two ions (on a weight basis) both in the soil solution around the root system or in the solution culture (Rs) and in the plants (Rp):

ln Rp = 0.42 (±0.046) + 0.48 (±0.013) ln Rs.



Therefore, an alternative method to predict manganese toxicity would be to extract the soil solution and directly measure the concentration of the two ions. Teixeira (1997), using this approach, was able to predict the ratio of the concentration of two ions in wheat from the ratio in the soil solution after two weeks growth (Rp = 0.44 Rs + 2.5) and the equation accounted for 99% of the variation. However the method needs further developments. On one hand, it will be necessary to investigate the relationship between Rp and Rs for different plant species and their critical Rp value. On the other hand, the values of Rs are changing with the growth of the crop (Figure 11). This effect of plant growth on the Rs of the soil solution might depend on the plant species and the soil, especially the cation exchange capacity.



Discussion and conclusions

Manganese toxicity might be a major limitation for the pasture and forage production in the Montado system in Cambisols derived from granites in the south of Portugal, although the problem might also affect Cambisols derived from other rocks and in other regions of the country. The Mn toxicity is related to the ratio of the concentration of Mg/Mn in shoot tissue. The severity of the toxicity increases in wet winters due waterlogging and leaching, but mainly to the latter, apparently because the Mn concentration increases in the soil. The same effect could be detected just by growing of plants, probably due to acidification of the soil by root exudates and the use of N fertilizer. The prediction of the Mn toxicity using soil test methods is very difficult, as extraction procedures are not able to mimic the effects of leaching and soil acidification on the ratio of Mg and Mn in the soil solution. An alternative approach is to measure the ratio of Mg/Mn concentrations in the soil solution and to obtain an effective correlation as for the one found for wheat plants. However, further knowledge is needed, particularly the critical concentration of the two ions Mg/Mn ratio in the cells for different plant species. In addition, we need to understand the effect of plant species on the concentration of the two ions in the soil solution and identify the best soil solution extraction procedures that mimic these effects.

Mn toxicity in the Cambisols cannot be alleviated by the addition of CaCO3, because Ca depressed both the Mg and the Mn present in soil solution and, therefore, had only a marginal effect on the ratio between the two ions. Magnesium has to be added simultaneously to complete remove the stress and the application of dolomitic limestone is the obvious solution. Because of the real difficulty in predicting manganese toxicity in Portuguese Cambisols based on soil analyses, we need to understand how pedogenetic processes, together with the nature of the parent material, influence the severity of the problem. We could then establish how the magnitude of the problem varies across the country and thereby advise farmers.



Cardoso, J.V.C. (1974) - A classificação dos solos de Portugal – Nova versão. Boletim de Solos do S.R.O.A., n. 17, p. 14-46.         [ Links ]

Carvalho, M. (1987) - Factores limitantes e técnicas culturais da produção de trigo mole no Alentejo. Tese de doutoramento. Évora, Universidade de Évora, 214 p.         [ Links ]

Chinnery, L.E. and Harding, C.P. (1980) - The effect of ferrous ion on the uptake of manganese by Juncus effuses. Annales of Botany, vol. 46, n. 4, p. 409-412.         [ Links ]

Goss, M.J. and Carvalho, M. (1988) - Causes of variation in yields of wheat under dryland farming in the Alentejo region of Portugal and some future prospects. In: Unger, P.W.; Jordan, W.R. and Sneed, I.V. (Eds.) - Proceedings of the International Conference on Dryland Farming. Texas A&M University, College Station, USA, p. 445-448.         [ Links ]

Goss, M.J. and Carvalho, M. (1992) - Manganese toxicity: The significance of magnesium for the sensitivity of wheat plants. Plant and Soil, vol. 139, n. 1, p. 91-98.        [ Links ]

Goss, M.J.; Carvalho, M.; Cosimini, V. and Fearnhead, M.L. (1992) - An approach to the identification of potentially toxic concentrations of manganese in soils. Soil Use and Management, vol. 8, n. 1, p. 40-44.        [ Links ]

Horst, W.J. (1988) - The physiology of manganese toxicity. In: Graham, R.D.; Hannam, R.J. and Uren, N.C. (Eds.) - Manganese in soils and plants. Dordrecht, The Netherlands, Kluwer Academic Publishers, p. 175-188.         [ Links ]

Horst, W.J. and Marschner, H. (1978) - Effect of silicon on manganese tolerance of bean plants (Phaseolus vulgaris L.). Plant and Soil, vol. 50, n. 1, p. 287-304.        [ Links ]

Le Bot, J.F.; Goss, M.J.; Carvalho, M.; Van Beusichem, M.L. and Kirkby, E.A. (1990) - The significance of magnesium to manganese ratio in plant tissues for growth and alleviatium of manganese toxicity in tomato (Lycopersicon esculentum) and wheat (Triticum aestivum) plants. Plant and Soil, vol. 124, n. 2, p. 205-210.        [ Links ]

Leeper, G.M. (1970) - Six trace elements in soils. Melbourne, Melbourne University press, 59 p.         [ Links ]

Löhnis, M.P. (1960) - Effect of magnesium and calcium supply on the uptake of manganese by various crop plants. Plant and Soil, vol. 12, n. 4, p. 339-376.        [ Links ]

Marschner, H. (1986) - Mineral nutrition of higher plants. London, Academic Press, 674 p.         [ Links ]

Ponnamperuma, F.N. (1984) - Effect of flooding on soils. In: Kozowski, T.T. (Ed.) - Flooding and plant growth. London, Academic press, p. 10-45.         [ Links ]

Portela, E.; Coelho-Pires, C. and Louzada, J. (2011) - Carência de magnésio em castanheiro: Influência do manganês do solo. Revista de Ciências Agrárias, vol. 34, n. 2, p. 154-162.         [ Links ]

Teixeira, D. (1997) - Métodos de determinação de magnésio e manganês no sistema solo-água-planta para a previsão da toxicidade de Mn num solo ácido. Tese de Mestrado. Évora, Universidade de Évora, 159 p.         [ Links ]

WRB (2006) - World reference base for soil resources, 2nd edition. World Soil Resources Reports 103. Rome, FAO, 128 p.         [ Links ]


Recebido/Received: 2015.09.24

Aceite/Accepted: 2015.10.15

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