Abstract
• Background and aims Natural and semi-natural, non-fertilized calcareous soils are consistently low in soluble and easily exchangeable phosphate. An over-utilization, or possibly an immobilization, of inorganic P in the tissues of calcifuge plants may take place, if such plants are forced to grow on a calcareous soil, though this has not been experimentally demonstrated. The objectives of this study are, therefore, to elucidate if calcifuge plants, when forced to develop on a calcareous soil, not only have lower total P (Ptot) concentrations in their leaves than calcicole plants grown on such soil, but also a lower proportion of Ptot as water-soluble, inorganic phosphate. Such differences may be of importance in understanding the calcicole–calcifuge behaviour of plants.
• Materials and methods Plants of five calcicole and five calcifuge herbs and three calcicole and three calcifuge grasses were cultivated in a glasshouse on a moderately acid Cambisol and a calcareous Rendzic Leptosol using seeds of wild populations from southern Sweden. The calcifuges were: Corynephorus canescens, Deschampsia flexuosa, Holcus mollis, Digitalis purpurea, Lychnis viscaria, Rumex acetosella, Scleranthus annuus and Silene rupestris. The calcicoles were: Melica ciliata, Phleum phleoides, Sesleria caerulea, Arabis hirsuta, Sanguisorba minor, Scabiosa columbaria, Silene uniflora ssp. petraea and Veronica spicata.
• Key results At harvest, calcifuges had much lower leaf tissue concentrations of Ptot and Pi than calcicoles when grown on the calcareous soil, and biomass production of the calcifuges was poor on this soil. Moreover, the calcifuge herbs had, on average, a lower proportion of their Ptot as Pi than had the calcicole herbs. The calcifuge herbs were also unable to avoid excessive uptake of Ca from the calcareous soil. The calcifuge grasses maintained a similar proportion of Ptot as Pi as the calcicole grasses, but their growth was still poor on the calcareous soil.
• Conclusions On calcareous soil, very little Pi in the tissues of calcifuge herbs is, at any time, available for use in various physiological functions. This is of importance to their photosynthesis, growth, competition and final survival on such soils.
Key words: Calcifuge, calcicole, calcium, inorganic phosphorus, phosphate, plant, tissue
INTRODUCTION
Natural and semi-natural, non-fertilized calcareous soils are consistently low in soluble and easily exchangeable phosphate. Inorganic P of such soils is mainly present as insoluble apatite or apatite-like Ca compounds, which plants or plant–microbial systems need to transform and solubilize to render the P constituent available for uptake. Previous studies have demonstrated that calcifuge plants (which do not grow on calcareous soils) have a low capacity to solubilize such forms of soil P (Tyler, 1992). This is related to low root exudation rates of di- and tricarboxylic organic acids (Tyler and Ström, 1995). Oxalic acid, in particular, is a powerful solubilizer of mineral Ca phosphate due to the very low solubility of Ca oxalate. Calcicole species (which grow on calcareous soil) exude much oxalic acid/oxalate, whereas calcifuges are much less able to do so, if exposed to calcareous soil or otherwise nutrient-poor conditions.
Phosphorus in plant tissues is, to a large extent, present in organic compounds. Only a fraction of total P is supposed to occur as free orthophosphate ions (Bieleski, 1973; Bollons and Barraclough, 1997). Theoretically, phosphate in plant tissues may also be immobilized as inorganic precipitates, though this has not been experimentally demonstrated in the calcicole–calcifuge context. An immobilization of P in the tissues of calcifuge plants may occur, if such plants are forced to grow on a calcareous soil. Under such conditions, an excessive uptake of Ca may take place in calcifuges, causing precipitation of Ca phosphate in their tissues. It is also likely that plants suffering from shortage of P may, at any time, have a lower proportion of free inorganic phosphate, as a larger proportion of their total P has to be utilized in organic compounds. Inability of calcifuges to solubilize Ca phosphate from calcareous soils would therefore cause a more instantaneous and complete utilization of the phosphate actually released and taken up from such soils and a lower proportion of inorganic tissue phosphorus would remain for various physiological demands and reactions.
The objectives of this study are to test the following hypotheses:
Calcifuge plants, when forced to develop on a calcareous soil, have lower total P (Ptot) concentrations in their leaves than calcicole plants grown on the same soil.
A lower proportion of leaf Ptot is present as water-soluble, inorganic phosphate (Pi) in calcifuges than in calcicoles grown on calcareous soil.
When grown on non-calcareous soil with naturally higher contents of soluble/easily exchangeable phosphate, leaf concentrations of Ptot do not differ between calcicoles and calcifuges.
Such differences may be of importance in understanding the calcicole–calcifuge behaviour of plants.
MATERIALS AND METHODS
Soils
Soils used were the top 10 cm of (a) a non-calcareous, moderately acid sandy-silty Cambisol, and (b) a calcareous Rendzic Leptosol, both from southern Sweden. Each soil was sampled a few days before the start of the experiment, sieved (mesh 6 mm) and thoroughly mixed. The pH-H2O, pH-KCl, loss on ignition, soluble phosphate, and phosphate bound to Al, Fe, and Ca were determined (Table 1). The difference in pH between the acid and the calcareous soil was about 3 units. Organic matter content (loss on ignition) was somewhat higher in the acid soil, whereas easily exchangeable phosphate was about six times lower in the calcareous soil. Aluminium and Fe phosphate concentrations were much higher in the acid than in the calcareous soil, whereas Ca phosphate was about three-fold higher in the calcareous soil.
Table 1.
Properties of the soils used in the experiment
| Acid soil |
Calcareous soil |
|
|---|---|---|
| pH-KCl | 4·40 ± 0·01 | 7·76 ± 0·01 |
| pH-H2O | 5·42 ± 0·02 | 7·94 ± 0·03 |
| Loss on ignition, % d. wt. | 10·2 ± 0·1 | 6·45 ± 0·03 |
| Soluble phosphate | 0·069 ± 0·003 | 0·011 ± 0·001 |
| Al phosphate | 1·20 ± 0·03 | 0·12 ± 0·01 |
| Fe phosphate | 2·02 ± 0·01 | 0·04 ± 0·01 |
| Ca phosphate | 2·4 ± 0·03 | 7·5 ± 0·4 |
pH was determined electrometrically after extraction (1 h) of 10 g fresh soil with 50 mL 0·2 m KCl or H2O; loss on ignition at 600 °C (2 h); soluble (easily exchangeable) phosphate according to Tyler (1992); Al, Fe and Ca phosphate according to Kuo (1996).
Values are means ± s.e. (n = 4 of the sieved soil).
Phosphorus concentrations given as μmol g−1 d. wt.
Concentrations of elements in the soil solution were not determined in this experiment. However, almost identical soils from the same sites have been used in other studies of soil solution conditions. According to these studies, soil solution expelled by high-speed centrifugation technique contained 1·2 mm Ca and 0·8 μm inorganic phosphate in the calcareous soil compared with 0·25 mm Ca and 5 μm inorganic phosphate in the acid soil (Tyler, 2000b; Tyler and Olsson, 2002).
Glasshouse experiment
A glasshouse pot experiment using the two different soils was conducted for 70 d in November–January. Plants of 16 species (three calcicole and three calcifuge grasses; five calcicole and five calcifuge herbs) were cultivated from seeds of wild populations from southern Sweden. Originating from field evidence, including extensive studies of soil–plant relationships in southern Sweden (e.g. Tyler, 1996, 2000a), the species were selected and categorized as calcicoles or calcifuges. The calcifuges were: Corynephorus canescens (L.) P. Beauv., Deschampsia flexuosa (L.) Trin., Holcus mollis L., Digitalis purpurea L., Lychnis viscaria L., Rumex acetosella L., Scleranthus annuus L. and Silene rupestris L. The calcicoles were: Melica ciliata L., Phleum phleoides (L.) H. Karst., Sesleria caerulea (L.) Ard., Arabis hirsuta (L.) Scop., Sanguisorba minor Scop., Scabiosa columbaria L., Silene uniflora ssp. petraea (Fr.) Jonsell & H.C. Prent. and Veronica spicata L.
The seeds were sawn as monocultures in 2-L plastic pots with bottom drainage (four replicates of each species and soil) arranged at random in the glasshouse. Soil moisture was adjusted to 60 % of the water-holding capacity and maintained at 55–60 % throughout the experiment by daily addition of H2O to the soil surface. Air temperature at night was kept at 15 ± 1°C, by day at 20 ± 2°C. Additional light, approx. 160 μE s−1 m−2, was supplied by high pressure sodium lamps for 12 h d−1. After germination, seedlings were thinned to 20 individuals per pot.
Plant tests and analyses
All equipment used was washed in freshly prepared 1 % H2SO4 and H2O. Acid/H2O-washed rubber gloves were used during handling of the leaves to avoid contamination.
To find the best suited method to determine soluble inorganic P (Pi) in tissues, leaves of Digitalis purpurea were extracted with H2O, 0·004 m HCl (pH 2·4), or 0·1 m HCl (pH 1). The number of replicates of each leaf treatment was four. One share of each leaf sample was dried immediately at harvest in a preheated oven at 140 °C. Another share of each leaf sample was washed, cut into 10–20 mm pieces with scissors, wrapped into Al foil and immersed in liquid N2. Of these samples 0·20 g were ground in a porcelain mortar with 1 mL extractant, 9 mL cold extractant were added and the material collected in a 50-mL polythene flask. After washing the mortar three times with 10 mL extractant and adding to the former extraction, all was shaken for 60 min at room temperature. Previous tests on H2O-extracted leaves had revealed no significant differences in the amounts of phosphate released between extractions for 15, 60 and 180 min. The volume was made to 50 mL with extractant and filtered. The dried leaves (0·050 g) were treated with 40 mL extractant, shaken for 1 h, diluted as above and filtered. Molybdate-reactive phosphate was analysed in a flow injection system (FIA), as described below.
Differences among replicates of fresh samples were greater than differences among dried samples, which indicated that it was more difficult to get homogeneous replicates of fresh than of dried material. The HCl-extractions of fresh materials tended to give lower phosphate values than the water extractions, whereas the three extractants of dried samples did not differ. The lower Pi concentration after HCl-extraction of fresh material might be caused by protonation of phosphate groups, which become less soluble and stay inside the cell (T. Drakenberg, pers. comm.). A problem with fresh material is also a possible hydrolysis of organic P compounds in the extract with time, e.g. of phosphoesters contained in chloroplast membranes and proteins. Therefore, extracts of fresh and dried samples were analysed with 31P-NMR (Nord et al., 2001) at the Department of Chemistry, Lund University, to assess possible phosphatase activity. Such activity was detected in several samples of fresh material, but not in the dried materials. A risk of bacterial degradation in fresh material also exists. It is possible that hydrolysis cannot be avoided completely even if the sample is heated to 140 °C, but the time was very short (<5 min) until this temperature was reached, and it is not likely that hydrolysis had much influence on the results.
Further, to test if the molybdate method for Pi analysis measures any organic phosphate, 20 μm inositoltriphosphate solutions were analysed, but no phosphate could be detected in the solutions. Furthermore, oxalate and citrate were added to one series of phosphate standards, because organic acids might be found at high concentrations in leaves and can disturb the phosphate analysis. Addition of 500 μm citrate or 50–200 μm oxalate to a 20 μm phosphate standard did not, however, influence the phosphate readings.
Based on these method tests we decided to use plant material dried at 140 °C and H2O as ‘extractant’ of soluble inorganic phosphate (Pi).
At harvest, total plant (leaf) biomass produced in each pot was cut approx. 5 mm above the soil surface, immediately transferred to a thermostat drying oven, preheated at 140 °C to minimize possible risks of organic P hydrolysis, and dried at that temperature to constant weight. Dry weights were determined. The plant materials were subsequently finely ground in a ball mill and redried at 140 °C prior to further analysis. Of each sample, 25 mg were shaken with 25 mL H2O (suprapure) for 60 min, part of the extract filtered, and directly analysed after appropriate dilution for Mo-reactive phosphate with SnCl2 as reductant, using a flow injection (FIA) system. For analysis of total P, another 25 mg of each sample were digested with 15 mL conc. HNO3, excess acid evaporated to 2 mL, residues diluted with H2O to 50 mL, and phosphate analysed as above after appropriate dilution and pH adjustment. As uptake of excess Ca may be of importance to P uptake and metabolism, Ca was also determined on these plant digests using flame atomic absorption spectrophotometry with 1 % La as LaCl3 in standards and samples to suppress phosphate and sulphate interferences.
Concentrations of P and Ca were calculated as µmol g−1 d. wt of plants and soils. Results are given as means ± standard error (s.e.; n = 4); significance of difference between means were calculated using the t-test.
RESULTS
The calcareous soil had lower Pi and, according to previous studies, higher Ca concentrations than the acid soil. Growth was poor in the calcifuges on the calcareous soil (Table 2). Shoot biomass production of the calcifuges on the acid soil was between two and five times higher than on the calcareous soil. The maximum percentage difference in growth was measured in Lychnis viscaria. The calcicole species produced more, or similar amounts of, shoot biomass on the calcareous as on the acid soil (Table 2). Calcium was probably sufficient for the calcicoles also on the acid soil, though calcicoles have been observed not to grow well with little available Ca, which might be due to high Ca fluxes to the vacuoles of calcicoles (White and Broadley, 2003).
Table 2.
Above ground biomass produced (harvested)
| Acid soil |
Calcareous soil |
|||
|---|---|---|---|---|
| Calcicole grasses | ||||
| Melica ciliata | 142 ± 10 | 315 ± 25 | ||
| Phleum phleoides | 279 ± 8 | 229 ± 9 | ||
| Sesleria caerulea | 161 ± 7 | 145 ± 12 | ||
| Calcicole herbs | ||||
| Arabis hirsuta | 111 ± 12 | 302 ± 35 | ||
| Sanguisorba minor | 298 ± 6 | 259 ± 27 | ||
| Scabiosa columbaria | 118 ± 6 | 208 ± 15 | ||
| Silene uniflora | 154 ± 3 | 171 ± 8 | ||
| Veronica spicata | 227 ± 24 | 319 ± 16 | ||
| Calcifuge grasses | ||||
| Corynephorus canescens | 294 ± 33 | 103 ± 12 | ||
| Deschampsia flexuosa | 756 ± 28 | 204 ± 6 | ||
| Holcus mollis | 567 ± 55 | 225 ± 4 | ||
| Calcifuge herbs | ||||
| Digitalis purpurea | 328 ± 18 | 173 ± 15 | ||
| Lychnis viscaria | 366 ± 12 | 66 ± 7 | ||
| Rumex acetosella | 112 ± 8 | 69 ± 5 | ||
| Scleranthus annuus | 351 ± 40 | 246 ± 9 | ||
| Silene rupestris | 210 ± 16 | 56 ± 2 | ||
Values (mg d. wt per pot) are means ± s.e.
The calcicole herbs had more Pi and Ptot in their leaves than the calcifuges, when grown on the calcareous soil (Table 3). The Pi concentrations in shoots of the calcifuge herbs were, on average, only one-third of the concentrations in the calcicole herbs when grown on the calcareous soil. Grown on the acid soil, the Pi concentrations of the herbs were generally higher and there was no consistent difference in Pi concentration between the two categories. There was only little difference in Pi concentrations of the grasses as related to category or soil type, though calcifuge grasses had somewhat lower Pi concentrations than the calcicole grasses on both soils.
Table 3.
Total phosphorus (Ptot) and water-soluble inorganic phosphate (Pi) concentrations in plant tissues
| Soil type |
||||||||
|---|---|---|---|---|---|---|---|---|
| Total P (Ptot) |
Inorganic P (Pi) |
|||||||
| A |
C |
A |
C |
|||||
| Calcicole grasses | ||||||||
| Melica ciliata | 44·4 ± 5·2 | 39·4 ± 4·7 | 15·1 ± 0·2 | 15·6 ± 0·7 | ||||
| Phleum phleoides | 80·6 ± 0·6 | 41·3 ± 0·6* | 16·5 ± 0·5 | 14·0 ± 0·3* | ||||
| Sesleria caerulea | 56·8 ± 1·4 | 36·1 ± 0·6* | 16·4 ± 0·2 | 13·1 ± 0·3* | ||||
| Calcicole herbs | ||||||||
| Arabis hirsuta | 37·2 ± 3·4 | 37·7 ± 4·5 | 16·7 ± 0·5 | 19·9 ± 1·5 | ||||
| Sanguisorba minor | 56·6 ± 2·0 | 40·6 ± 2·6* | 14·1 ± 0·2 | 9·2 ± 0·2* | ||||
| Scabiosa columbaria | 89·0 ± 1·1 | 49·1 ± 3·3* | 28·9 ± 0·7 | 19·4 ± 0·6* | ||||
| Silene uniflora | 58·4 ± 5·5 | 59·0 ± 1·8 | 17·5 ± 0·8 | 17·2 ± 0·9 | ||||
| Veronica spicata | 84·6 ± 11 | 34·4 ± 0·7* | 21·3 ± 1·0 | 13·9 ± 0·1* | ||||
| Calcifuge grasses | ||||||||
| Corynephorus canesc. | 56·3 ± 2·5 | 35·3 ± 0·1* | 14·7 ± 0·4 | 11·0 ± 1·0* | ||||
| Deschampsia flexuosa | 46·9 ± 5·7 | 34·4 ± 2·3 | 14·0 ± 0·4 | 12·8 ± 0·8 | ||||
| Holcus mollis | 30·9 ± 5·5 | 27·3 ± 0·4 | 10·4 ± 0·4 | 11·9 ± 0·3 | ||||
| Calcifuge herbs | ||||||||
| Digitalis purpurea | 54·7 ± 1·0 | 35·9 ± 0·4* | 16·7 ± 0·4 | 9·2 ± 0·6* | ||||
| Lychnis viscaria | 62·9 ± 5·4 | 21·4 ± 0·1* | 19·3 ± 0·9 | 6·6 ± 0·3* | ||||
| Rumex acetosella | 37·4 ± 0·1 | 15·0 ± 0·7* | 4·0 ± 0·1 | 0·9 ± 0·1* | ||||
| Scleranthus annuus | 76·7 ± 9·5 | 24·2 ± 1·1* | 24·5 ± 0·3 | 7·9 ± 0·5* | ||||
| Silene rupestris | 99·9 ± 0·6 | 12·8 ± 0·3* | 16·4 ± 0·3 | 3·2 ± 0·1* | ||||
Values (μmol g−1 d. wt) are means ± s.e. (n = 4).
Soil type: A = acid soil, C = calcareous soil.
Differ (P < 0·05) from concentration in plants of the same species grown on the acid soil.
The calcifuge herbs had about three times higher concentrations of Ptot on the acid than on the calcareous soil, whereas the calcicole herbs differed less between soils, as did the grasses of both categories (Tables 3 and 5). The Ptot concentrations were, on average, twice as high in the calcicole as in the calcifuge herbs on the calcareous soil, whereas there were no consistent differences between the herb categories on the acid soil (Table 4). The slightly lower Pi concentrations of the calcicoles on the calcareous than on the acid soil did apparently not influence their growth, though their total P concentrations were also lower on the calcareous soil (Table 4). According to Broadley et al. (2004) the accumulation of P by P-sufficient plants is phylogenetically independent, which seems consistent with our observations.
Table 5.
Water-soluble inorganic P (Pi), total P (P tot), Pi as percentage of total P (Pi % P tot) and Ca in calcicole and calcifuge grasses, on acid and calcareous soil
| Calcicole grasses |
Calcifuge grasses |
|||||
|---|---|---|---|---|---|---|
| Acid soil |
Calc. soil |
Acid soil |
Calc. soil |
|||
| Pi | 16 ± 0·3* | 14·2 ± 0·4* | 13 ± 0·6 | 11·9 ± 0·5 | ||
| P tot | 60·6 ± 4·6* | 38·9 ± 1·0* | 44·7 ± 3·6 | 32·3 ± 1·1 | ||
| Pi % Ptot | 28 ± 2 | 37 ± 1 | 31 ± 2 | 37 ± 2 | ||
| Ca | 120 ± 16 | 222 ± 17 | 98 ± 11 | 262 ± 24 | ||
| Ca : Pi | 8 ± 1 | 16 ± 1 | 8 ± 1 | 24 ± 4 | ||
| Ca : Ptot | 2 ± 0 | 6 ± 0 | 3 ± 0 | 8 ± 1 | ||
Concentration values (μmol g−1 d. wt) are means ± s.e.
Differ (P < 0·05) from calcifuges grown in the same soil.
Table 4.
Water-soluble inorganic P (Pi), total P (Ptot), Pi as percentage of total P (Pi % P tot) and Ca in calcicole and calcifuge herbs, on acid and calcareous soil
| Calcicole herbs |
Calcifuge herbs |
|||||
|---|---|---|---|---|---|---|
| Acid soil |
Calc. soil |
Acid soil |
Calc. soil |
|||
| Pi | 19·7 ± 1·3 | 15·9 ± 1·0* | 16·2 ± 1·6 | 5·6 ± 0·7 | ||
| Ptot | 65·1 ± 4·7 | 44·1 ± 2·2* | 66·3 ± 5 | 21·8 ± 1·9 | ||
| Pi % Ptot | 32 ± 2* | 37 ± 3* | 24 ± 2 | 24 ± 2 | ||
| Ca | 342 ± 23* | 743 ± 48* | 237 ± 12 | 1021 ± 51 | ||
| Ca : Pi | 19 ± 2 | 51 ± 4* | 17 ± 3 | 437 ± 171 | ||
| Ca : Ptot | 6 ± 1 | 18 ± 2* | 4 ± 0 | 51 ± 6 | ||
Concentration values (μmol g−1 d. wt) are means ± s.e.
Differ (P < 0·05) from calcifuges grown in the same soil.
The calcicole herbs had a significantly higher proportion of total P in their leaves in the form of Pi than the calcifuges on both soils (Table 4; acid soil P = 0·012, calcareous soil P < 0·001). The calcicole herbs were able to maintain or even increase the proportion of Pi in their leaves on the calcareous soil, whereas the calcifuge herbs were not able to do so. There was no difference (P > 0·05) in Pi % Ptot between the calcifuge and the calcicole grasses on any of the soils (Table 5).
Calcifuge herbs are apparently also less able to counteract excessive uptake of Ca, as they had, on average, higher concentrations of Ca on the calcareous soil than had the calcicole herbs (Tables 4 and 6). When grown on the acid soil, the opposite condition tended to be valid. Again, calcifuge grasses were relatively less affected, with no significant differences between the categories on the same soil (Tables 5 and 6). The Ca : Pi ratio in the calcifuge herbs was, on average, 25-fold higher on the calcareous soil than on the acid soil, whereas this difference was less than three-fold in the calcicole herbs (Table 4). In the calcifuge grasses the Ca : Pi ratio differed three times between the soils, in the calcicole grasses about two times (Table 5). The ratio of Ca to total P was about three times higher in the calcifuge than in the calcicole herbs on the calcareous soil (Table 4), whereas grasses of both categories did not differ much in this respect (Table 5).
Table 6.
Calcium concentrations in plant tissues
| Soil type |
|||
|---|---|---|---|
| A |
C |
||
| Calcicole grasses | |||
| Melica ciliata | 99 ± 1 | 201 ± 4 | |
| Phleum phleoides | 191 ± 13 | 297 ± 13 | |
| Sesleria caerulea | 71 ± 6 | 170 ± 8 | |
| Calcicole herbs | |||
| Arabis hirsuta | 488 ± 50 | 1100 ± 56 | |
| Sanguisorba minor | 374 ± 3 | 657 ± 9 | |
| Scabiosa columbaria | 239 ± 21 | 601 ± 14 | |
| Silene uniflora | 364 ± 17 | 627 ± 1 | |
| Veronica spicata | 246 ± 22 | 713 | |
| Calcifuge grasses | |||
| Corynephorus canescens | 78 ± 7 | 342 ± 15 | |
| Deschampsia flexuosa | 74 ± 16 | 233 | |
| Holcus mollis | 142 ± 25 | 196 ± 4 | |
| Calcifuge herbs | |||
| Digitalis purpurea | 285 ± 27 | 763 ± 12 | |
| Lychnis viscaria | 197 ± 1 | 1330 | |
| Rumex acetosella | 179 | 1300 | |
| Scleranthus annuus | 299 ± 17 | 966 ± 7 | |
| Silene rupestris | 199 ± 10 | 1070 | |
Values (μmol g−1 d. wt) are means ± s.e.
Soil type: A = acid soil, C = calcareous soil.
Where no s.e. is given, too few samples were analysed for calculation.
DISCUSSION
Decreased production of leaf biomass is typical for P deficiency (Tyler, 1992). Some of the calcifuges also tended to have a slightly violet colour, which could be a sign of P deficiency. Inorganic phosphate controls several important enzyme reactions and photosynthesis. Especially growth is affected at P deficiency and insufficient levels of Pi are suggested to reduce the expansion of epidermal cells, resulting in reduced leaf area and shoot dry weight (Fredeen et al., 1989). Leaf area was not measured in our experiment, but leaves of the calcifuges on the calcareous soil were visually much smaller than on the acid soil, whereas most calcicoles grew, on the contrary, better on the calcareous than on this moderately acid soil. They produced, on average, more biomass (up to 2·7 times) on the calcareous than on the acid soil (Table 2). The calcicoles possibly took up more P on the acid soil than they really needed for growth, because leaf P concentrations normally do not differ in the field between plant species that are able to grow on both acid and calcareous soils (Tyler and Zohlen, 1998). Several authors (Chapin and Bieleski, 1982; Bollons and Barraclough, 1997) reported that, if plants receive additional P, they store the main part of this surplus as Pi.
Calcium requires a low cytoplasmic concentration to prevent, for example, precipitation as Ca phosphate (Webb et al., 1996). An inability of calcifuges to transport Ca out of the cytoplasm seems to be crucial to compete successfully on calcareous soil (Lee, 1999). It is, therefore, likely that the excessive uptake of Ca in the calcifuges on the calcareous soil may have caused precipitation of P as Ca phosphate in the calcifuge herbs. Inorganic P (Pi) was lower at high Ca concentrations compared with other P fractions. Many calcifuges may increase their oxalate production to precipitate Ca in their leaves, whereas calcicoles increase their malate production and maintain Ca in a soluble form (Kinzel, 1982). However, if the oxalate production of the calcifuges was sufficient to lower the soluble Ca level or if Ca phosphate precipitated in the biomass, was not determined in our experiment. Monocotyledons of the Poaceae take up less Ca than dicotyledons for adequate growth (Loneragan and Snowball, 1969; Broadley et al., 2003) which was confirmed by our experiment. These findings might indicate that the calcifuge–calcicole behaviour has developed later during evolution than the separation of monocots and dicots.
High HCO3− concentrations of the soil may increase the solubility of HPO42−. This can lead to an increased P uptake (Ao et al., 1987), though obviously not in the calcifuges according to our results. Their P uptake by the roots seems to be impeded, or P is immobilized in the roots or during transport to the leaves. Smith et al. (1990) reported a retention of P in the roots and a net translocation of P from the shoot to the root under P deficiency. This was accompanied by an increase of carbohydrate partitioning towards the root, especially of starch and sucrose, because photosynthates produced could not be used for growth (Fredeen et al., 1989). It seems most likely that an inability to solubilize Ca phosphate is a main limitation for calcifuges on calcareous soils and not an insufficient uptake mechanism of Pi actually available. Tissue phosphorus concentrations were actually similar in calcicoles and calcifuges grown on the acid soil.
According to previous studies there are at least some calcifuges which are primarily limited by Fe, not by P, when forced to grow on a calcareous soil (Tyler 1994; Zohlen and Tyler, 1997, 2000). On calcareous soils much of the leaf Fe is immobilized in physiologically less active forms in these calcifuges, even when sufficient amounts of Fe are actually taken up by the plant. It seems, however, that primary P-limitation is much more widespread in calcifuges than are the Fe limitations, but it is important not to generalize the limiting mechanisms. Also P-limited plants may easily display Fe chlorosis if supplied with adequate amounts of soluble phosphate (Tyler, 1992).
Soil organic P may be an important P source of plants. On average, half of the total soil P was present as organic phosphates in calcareous grassland topsoils studied by Tyler (2002). At the high pH prevailing in calcareous soils, plants have to acidify their immediate rhizosphere to be able to use their main phosphatases, which presupposes a high exudation rate of acids by their roots to solubilize the main constituent of a limestone soil, CaCO3. Alkaline phosphatase is not at all, or only in small amounts present in root exudates of vascular plants. However, it is produced by the soil microflora and the presence of vascular plants may influence its release and activity (Joner and Jakobsen, 1995). However, in a study by van Aarle (2002) it was concluded that no active release of phosphatase activity by arbuscular mycorrhiza could be detected, which supports previous observations that arbuscular mycorrhizal fungi do not mineralize organic P in soil through the release of phosphatases. Moreover, the way plants mobilize inorganic soil phosphate for root uptake is not likely to influence its chemical state in the leaf tissues and therefore not really pertinent to the current problem.
CONCLUSIONS
The three hypotheses stated in the Introduction were shown to be valid for the calcicole and the calcifuge herbs, but only partly or less valid for the grasses studied. The calcifuge herbs were also unable to avoid excessive uptake of Ca from the calcareous soil. Ecophysiological implications of the results concerning the calcicole–calcifuge behaviour of plants could be the finding that calcifuge herbs not only have lower concentrations of soluble Pi and of Ptot in their leaves compared with calcicoles but also a lower relative proportion of soluble Pi when grown on a calcareous soil. This means that, on calcareous soil, relatively little soluble Pi in the tissues of calcifuge herbs is, at any time, available for use in various physiological functions. This is of importance to their photosynthesis, growth, competition and final survival on such soils. The calcifuge grasses, however, maintained a similar proportion of P in water-soluble inorganic form as the calcicole grasses, but still their growth was poor on the calcareous soil.
Supplementary Material
Acknowledgments
We thank Prof. Torbjörn Drakenberg at the Department of Chemistry, Lund University, for the NMR analysis and his comments on the method tests, also to Mrs Maj-Britt Larsson for her technical assistance.
LITERATURE CITED
- Ao TY, Chaney RL, Korcak RF, Fan F, Faust M. 1987. Influence of soil moisture level on apple iron chlorosis development in a calcareous soil. Plant and Soil 104: 85–92. [Google Scholar]
- Bieleski RL. 1973. Phosphate pools, phosphate transport and phosphate availability. Annual Review of Plant Physiology 24: 225–252. [Google Scholar]
- Bollons HM, Barraclough PB. 1997. Inorganic orthophosphate for diagnosing the phosphorus status of wheat plants. Journal of Plant Nutrition 20: 641–655. [Google Scholar]
- Broadley MR, Bowen HC, Cotterill HL, Hammond JP, Meacham MC, Mead A, White, PJ. 2003. Variation in the shoot calcium content of angiosperms. Journal of Experimental Botany 54: 1431–1446. [DOI] [PubMed] [Google Scholar]
- Broadley MR, Bowen HC, Cotterill HL, Hammond JP, Meacham MC, Mead A, White PJ. 2004. Phylogenetic variation in the shoot mineral concentration of angiosperms. Journal of Experimental Botany 396: 321–336. [DOI] [PubMed] [Google Scholar]
- Chapin FS, Bieleski RL. 1982. Mild phosphorus stress in barley and a related low-phosphorus-adapted barleygrass: phosphorus fractions and phosphate absorption in relation to growth. Physiologia Plantarum 54: 309–317. [Google Scholar]
- Fredeen AL, Rao IM, Terry N. 1989. Influence of phosphorus nutrition on growth and carbon partitioning in Glycine max. Plant Physiology 89: 225–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Joner EJ, Jakobsen I. 1995. Growth and extracellular phosphatase activity of arbuscular mycorrhizal hyphae as influenced by soil organic matter. Soil Biology and Biochemistry 27: 1153–1159. [Google Scholar]
- Kinzel H. 1982.Pflanzenökologie und Mineralstoffwechsel. Eugen Ulmer, Stuttgart, Germany. [Google Scholar]
- Kuo S. 1996. Phosphorus. In: Bartels JM, Bigham JM, eds. Methods of soil analysis. 3. Chemical methods. Madison, WI: Soil Science Society of America, 869–919. [Google Scholar]
- Lee JA. 1999. The calcicole–calcifuge problem revisited. Advances in Botanical Research 29: 1–30. [Google Scholar]
- Loneragan JF, Snowball K. 1969. Calcium requirements of plants. Australian Journal of Agricultural Research 20: 465–478. [Google Scholar]
- Nord J, Nyman P-O, Larsson G, Drakenberg T. 2001. The C-terminus of dUTPase: observation on flexibility using NMR. FEBS Letters 492: 228–232. [DOI] [PubMed] [Google Scholar]
- Smith FW, Jackson WA, van den Berg PJ. 1990. Internal phosphorus flows during development of phosphorus stress in Stylosanthes hamata Australian Journal of Plant Physiology 17: 451–464. [Google Scholar]
- Tyler G. 1992. Inability to solubilize phosphate in limestone soils – key factor controlling calcifuge habit of plants. Plant and Soil 145: 65–70. [Google Scholar]
- Tyler G. 1994. A new approach to understanding the calcifuge habit of plants. Annals of Botany 73: 327–330. [Google Scholar]
- Tyler G. 1996. Soil chemistry and plant distributions in rock habitats of southern Sweden. Nordic Journal of Botany 16: 609–635. [Google Scholar]
- Tyler G. 2000. Integrated analysis of conditions accounting for intersite distribution of grassland plants. Nordic Journal of Botany 20: 485–500. [Google Scholar]
- Tyler G. 2000. Effects of sample pretreatment and sequential fractionation by centrifuge drainage on concentrations of minerals in a calcareous soil solution. Geoderma 94: 59–70. [Google Scholar]
- Tyler G. 2002. Phosphorus fractions in grassland soils. Chemosphere 48: 343–349. [DOI] [PubMed] [Google Scholar]
- Tyler G, Olsson T. 2002. Conditions related to solubility of rare and minor elements in forest soils. Journal of Plant Nutrition and Soil Science 165: 594–601. [Google Scholar]
- Tyler G, Ström L. 1995. Differing organic acid exudation pattern explains calcifuge and acidifuge behaviour of plants. Annals of Botany 75: 75–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tyler G, Zohlen A. 1998. Plant seeds as mineral nutrient resource for seedlings: a comparison of plants from calcareous and silicate soils. Annals of Botany 81: 455–459. [Google Scholar]
- van Aarle IM. 2002.The ecophysiology of arbuscular mycorrhizal fungi: phosphatase activity associated with extraradical and intraradical mycelium. PhD Thesis, Lund University, Sweden. [Google Scholar]
- Webb AAR, McAinsh MR, Mansfield TA, Hetherington AM. 1996. Carbon dioxide induces increases in guard cell cytosolic free calcium. Plant Journal 9: 297–304. [Google Scholar]
- White PJ, Broadley MR. 2003. Calcium in plants. Annals of Botany 92: 487–511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zohlen A, Tyler G. 1997. Differences in iron nutrition strategies of two calcifuges, Carex pilulifera L. and Veronica officinalis L. Annals of Botany 80: 553–559. [Google Scholar]
- Zohlen A, Tyler G. 2000. Immobilization of tissue iron on calcareous soil—differences between calcicole and calcifuge plants. Oikos 89: 95–106. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
