|Home | About | Journals | Submit | Contact Us | Français|
Soil salinity is often heterogeneous, yet the physiology of halophytes has typically been studied with uniform salinity treatments. An evaluation was made of the growth, net photosynthesis, water use, water relations and tissue ions in the halophytic shrub Atriplex nummularia in response to non-uniform NaCl concentrations in a split-root system.
Atriplex nummularia was grown in a split-root system for 21 d, with either the same or two different NaCl concentrations (ranging from 10 to 670 mm), in aerated nutrient solution bathing each root half.
Non-uniform salinity, with high NaCl in one root half (up to 670 mm) and 10 mm in the other half, had no effect on shoot ethanol-insoluble dry mass, net photosynthesis or shoot pre-dawn water potential. In contrast, a modest effect occurred for leaf osmotic potential (up to 30 % more solutes compared with uniform 10 mm NaCl treatment). With non-uniform NaCl concentrations (10/670 mm), 90 % of water was absorbed from the low salinity side, and the reduction in water use from the high salinity side caused whole-plant water use to decrease by about 30 %; there was no compensatory water uptake from the low salinity side. Leaf Na+ and Cl− concentrations were 1·9- to 2·3-fold higher in the uniform 670 mm treatment than in the 10/670 mm treatment, whereas leaf K+ concentrations were 1·2- to 2·0-fold higher in the non-uniform treatment.
Atriplex nummularia with one root half in 10 mm NaCl maintained net photosynthesis, shoot growth and shoot water potential even when the other root half was exposed to 670 mm NaCl, a concentration that inhibits growth by 65 % when uniform in the root zone. Given the likelihood of non-uniform salinity in many field situations, this situation would presumably benefit halophyte growth and physiology in saline environments.
Saline environments can be highly variable in soil salinity in both space and time (e.g. Davidson et al., 1996; Mensforth and Walker, 1996; Bleby et al., 1997; Barrett-Lennard and Malcolm, 1999; Slavich et al., 1999). The factors contributing to this variability are the complex interactions of climate, topography, soil properties (e.g. texture, surface mulches) and the presence of fluctuating groundwater. Halophytes in field situations with distinct sources of water differing in salinity are able to take up water mainly from the least saline source (e.g. Suaeda fruticosa, Anabasis articulata, Atriplex halimus and Tamarix nilotica; Yakir and Yechieli, 1995).
Split-root experiments, where a root system is divided into two or more portions that are exposed to different conditions, have been useful for studies of how plants respond to heterogeneous soil conditions, e.g. partial root drying (Lawlor, 1973; Sobeih et al., 2004) and heterogeneous nutrient distribution (Drew and Saker, 1978; Arredondo and Johnson, 1999; Paterson et al., 2006). A split-root approach has also been used to study physiological responses (i.e. growth, water relations, water use) of some glycophytes to non-uniform root zone salinity (e.g. Kirkham et al., 1969; Zekri and Parsons, 1990; Shani et al., 1993; Lycoskoufis et al., 2005), with responses of growth being most commonly intermediate to the uniform low and uniform high salinity treatments. For halophytes, there have only been two studies that have used split-root systems with non-uniform salinity, and the purpose of these was to study nutrient uptake (Messedi et al., 2004; Hamed et al., 2008). However, as these studies used a salt-free solution as the uniform low salt ‘control’, growth enhancement in non-uniform salinity (zero on one side and 300–800 mm NaCl on the other side) was probably due to ion requirements in these dicotyledonous halophytes for maximal growth (Yeo and Flowers, 1980). For halophytes, the absence of salt results in sub-optimal growth owing to ion deficiency, and so use of NaCl-free solutions as controls has been criticized (Yeo and Flowers, 1980; Flowers and Colmer, 2008).
To study physiological responses of a halophytic species when exposed to non-uniform salinity, experiments were conducted on Atriplex nummularia (old man saltbush). This is a deep-rooted perennial C4 shrub that occurs naturally on saline lands in the semi-arid zone of Australia, and has been established on salt-affected agricultural lands for grazing livestock (Barrett-Lennard et al., 2003). Atriplex spp. occur in habitats that are often characterized by seasonally and spatially variable soil salinities (Sharma and Tongway, 1973; Barrett-Lennard and Malcolm, 1999; Slavich et al., 1999), and water sources can change within the soil profile depending on seasonal changes in soil water availability (Slavich et al., 1999). In the present experiments, A. nummularia had its root system simultaneously exposed to a range of uniform NaCl (10–670 mm) or non-uniform NaCl concentrations, with one root half exposed to 10 mm NaCl and the other root half to higher concentrations up to 670 mm. Control plants were grown in 10 mm NaCl, so as to avoid ion deficiencies (cf. Yeo and Flowers, 1980).
Experiments were conducted with A. nummularia in a split-root system to test four hypotheses, that with high salinity in one root half: (1) there will be little or no adverse effect on shoot ethanol-insoluble dry mass accumulation or net photosynthetic rate; (2) most of the water will be drawn from the low salinity side; (3) plants with high salinity only in one root half will have similar shoot water potential, leaf osmotic potential and stomatal conductance to those of plants exposed to uniform low salinity; and (4) plants will have ion (Na+, K+, Cl−) concentrations in shoots intermediate to those in the low salinity controls and plants at uniform high salinity.
A commercial clone of Atriplex nummularia Lindl., ‘Eyres Green’ (Tamlin's Nursery, South Australia), was used. Stem cuttings (10 cm long) with about five nodes and leaves on the upper two nodes were taken from a mother plant. Cuttings were propagated in a glasshouse with day/night temperatures of 25/15 °C. The stem cuttings were moistened at the base, dipped in a hormonal rooting powder (‘Richgro Root Strike’, containing 3 g kg−1 indole-butyric acid) and planted into drained containers filled with washed white sand. The containers were flushed daily with water until small roots were visible at the shoot base, and then were irrigated with 0·1-strength nutrient solution for 4 d, followed by 0·5-strength solution for 7 d, and thereafter full-strength solution was used. The full-strength nutrient solution consisted of (mm): 4·7 K2SO4, 9·3 CaCl2, 5·0 Na2SO4, 1·0 MgSO4, 0·8 Fe EDDHA (‘Sequestrene 138’), 0·7 Ca(NO3)2, 0·3 K2HPO4, 0·2 NH4H2PO4; and (μm): 23 H3BO3, 2 MnSO4, 2 ZnSO4, 0·5 CuSO4, 0·5 Na2MoO4. The nutrient solution was adjusted to pH 6, using KOH.
After 4–6 weeks, when roots were about 2 cm long, established cuttings were transferred to a naturally sunlit phytotron with day/night temperatures of 20/15 °C. Cuttings were washed free of sand and transferred to 4·5 L plastic pots containing aerated full-strength nutrient solution. This solution contained the same nutrient concentrations reported above, except that 0·1 mm Na2O3Si and 1·0 mm MES were added; pH was again adjusted to 6, using KOH. There were four plants per 4·5 L pot, and nutrient solutions were topped up with deionized water as required and renewed weekly.
In experiment 1, 2 weeks after transferring the cuttings to nutrient solution culture, plants were selected for shoot and root uniformity, and transferred into split-root pots (one plant per pot, with 1·2 L per side; the split-root pots are described below). After a further 4 d, NaCl was increased in both compartments of all the split-root pots in increments of 55 mm every 12 h, until NaCl concentrations reached 670 mm. Three days after reaching this concentration, all treatments were imposed with a single step down from 670 mm NaCl to the required level in each compartment. Plants were all exposed to 670 mm NaCl before applying treatments in order to mimic seasonal dynamics in soil salinity in the field, where there is salt accumulation after periods of high evapotranspiration demands (summer) and, in autumn and winter, rainfall can then leach salts out of the upper soil (Mensforth and Walker, 1996). In experiment 2, plants were grown in the same way, except that they were provided with full-strength nutrient solution for 2 weeks longer before being transferred into the split-root pots. This was done in order to have larger plants to enhance measurements of water uptake over 12-h periods.
Roots were divided into two approximately equal halves, with each positioned in a split-root pot, so that the two root halves could be exposed, at the same time, to different NaCl concentrations. To prevent mixing of solution, roots were laid in an inverted, lengthwise-cut, plastic T-piece (length, 6 cm; height, 6 cm; diameter, 3 cm). Inverted T-pieces were sealed and then placed over two cylindrical plastic containers with a notch into the top, cut to fit the T-piece, so that each root half was in a cylindrical container. Each container was filled with 1·2 L of nutrient solution. A similar split-root pot system was also designed to measure water use in each container without removing the roots from the pots. For water use measurements, each pot of the split-root system had two electric wires glued on the inside that allowed re-filling to a precise and constant level, with a precision of 10 µL, indicated by the presence or absence of electrical conductivity between the wires.
The experiment was conducted to assess growth, ion concentrations and water relations of A. nummularia when exposed to uniform or non-uniform NaCl concentrations, over a range of concentrations, in a split-root system. The experiment was conducted in a naturally lit phytotron [20/15 °C day/night with an average photosynthetically active radiation (PAR) at midday during the experimental period of 870 µmol m−2 s−1]. The experiment tested seven treatments (Table 1) with five replicates in a completely randomized block design. In four treatments, the two halves of the root systems were both exposed to the same NaCl concentrations (mm): 10, 230, 450 or 670. The remaining three treatments had the two halves of the root system exposed to two different NaCl concentrations, in each case with one side at 10 mm NaCl, and the other at 230, 450 or 670 mm NaCl. Osmotic potentials of the solutions were determined with a freezing point osmometer (Fiske Associates, Needham Heights, MA, USA).
Leaf gas exchange measurements were taken on day 19 of treatments on three randomly chosen plants in each treatment. Leaves on each side of the shoot, directly above each root side in treatments 10/230, 10/450 and 10/670, were measured separately. Measurements of net photosynthetic rate and stomatal conductance were determined on the youngest fully expanded leaves using a LI-COR 6400 Photosynthesis System (LI-COR, Inc., Lincoln, NE, USA) at ambient relative humidity (50–60 %), reference CO2 of 380 µmol mol−1, flow rate of 500 µmol s−1 and PAR of 1500 µmol m−2 s−1.
Plants were harvested on days 0 and 21 after commencement of treatments. Leaf area, shoot and root fresh mass were determined. Leaf area was measured with a portable leaf area meter (LI-COR LI-3100). In order to assess for any differences between sides, the two halves of the root system and each side of the shoot above each of the root halves were harvested separately (Fig. 1). The portion of the stem that was central was also sampled separately. Each shoot side directly above each root half was then sub-divided into expanding leaves, expanded leaves and side branches with leaves removed. Roots were washed for 2 min in three changes of iso-osmotic mannitol solution, also containing 9 mm CaSO4, and blotted dry. Fresh mass was recorded. All samples were snap-frozen in liquid N2, stored at −80 °C and then freeze-dried.
Pre-dawn shoot water potential was measured on excised shoot segments using a Scholander pressure chamber. Leaf tissue osmotic potential was measured on expanding leaves (at the third or fourth node, leaf area varied from 20 to 40 % of the size of fully expanded leaves) and on fully expanded leaves. These parameters were measured on three randomly chosen plants from each treatment, after 21 d of treatments. Shoot segments and leaves from both sides of the shoot, directly above each root half, were measured. As no differences were found between the two shoot sides, data presented for each parameter are the average for the data pooled from the two shoot sides for each replicate.
To determine ethanol-insoluble dry mass, ground plant tissues were extracted twice with 80 % ethanol, refluxed twice for 20 min, centrifuged for 10 min at 10 000 r.p.m., and the insoluble fraction was dried at 70 °C for 24 h and weighed.
Ground tissue samples were extracted with 0·5 m HNO3 by shaking in vials for 48 h. Diluted extracts were analysed for Na+, K+ (Jenway PFP7 Flame Photometer; Essex, UK) and Cl− (Buchler-Cotlove Chloridometer 662201; Fort-Lee, NJ, USA). Reliability of the methods was confirmed by analyses of a reference tissue taken through the same procedures.
The experiment was conducted to assess water use patterns of A. nummularia when exposed to non-uniform salinity in the root zone. The plants were raised in a naturally lit phytotron (20/15 °C day/night with an average PAR at midday during the experimental period of 790 µmol m−2 s−1). The experiment consisted of three treatments with four replicates, in a randomized block design. There were 2 harvests, after 8 and 22 d. The treatments were (mm NaCl): 10/10, 10/670 and 670/670. The first measurement of water use was performed 7 d after treatments commenced, as root halves were expected to be approximately equal in surface area early after the commencement of treatments. The morning after the water use measurements, plants were harvested (H1). Water use was also measured for a second complete set of replicate plants 21 d after imposing the treatments. Plants were again harvested the morning after the water use measurements (H2). Midday shoot water potentials were measured for the 10/10 and 10/670 mm NaCl treatments as described for experiment 1.
At 48 h prior to water use measurements, plants were transferred to a split-root system designed for water use measurements (described above) in a controlled-environment room (20/15 °C day/night, 12 h day/night, average relative humidity 70 %, with an average PAR at shoot height of 310 µmol m−2 s−1). All containers were bubbled with pre-humidified air, and three blank pots (i.e. without plants) were used to determine any background evaporative losses. To measure water use, each pot containing nutrient solution was topped up with deionized water at 0600 h and then again at 1800 h (commencement and end of the 12 h light period) to the point where both electrical wires (described above) were in contact with the nutrient solution, and volumes added were recorded.
Root and shoot fresh mass and leaf areas were measured as described in experiment 1. The two halves of the root system were harvested separately as in experiment 1, but no separations were made between shoot sides. Total leaf area was measured with a portable leaf area meter (LI-COR LI-3100), and stems and leaves were oven-dried at 60 °C to determine dry masses. Roots were blotted to remove excess surface moisture, sealed in plastic bags and stored at 4 °C for 12 h. Root systems were scanned for surface area using a WinRhizo root scanner (Regent Instruments Inc., Quebec, Canada). Roots were oven-dried at 60 °C to determine dry mass.
Statistical analyses were conducted using Genstat for Windows 10th Edition (Genstat software, VSN International, Hemel Hempsted, UK). Analysis of variance (ANOVA) was used to identify overall significant differences between treatments and between sides within treatments, depending on the data set. When significant differences were found, mean separations were calculated using Duncan's multiple range test. Unless otherwise stated, the significance level was P ≤ 0·05.
Growth was measured as ethanol-insoluble dry mass, as in dicotyledonous halophytes ions may contribute up to 30–50 % of the dry mass (Flowers et al., 1986). The ratio of ethanol-insoluble dry mass to dry mass ranged from 0·92 (central stem at 230 mm NaCl) to 0·55 (expanding leaves at 670 mm NaCl; data not shown).
Under uniform conditions, no differences in shoot ethanol-insoluble dry mass were found at 10 and 230 mm NaCl, but at 450 and 670 mm NaCl shoot dry mass declined by 42 and 65 %, respectively, compared with the uniform treatment at 10 mm NaCl (Fig. 2A). When only one root half was exposed to these NaCl concentrations, shoot growth was equal to that in the uniform 10/10 mm NaCl treatment. The difference between uniform and non-uniform salt treatments was most apparent at the highest NaCl concentration, where shoot ethanol-insoluble dry mass of plants at 10/670 mm NaCl was 2·3-fold higher than in plants grown in uniform 670 mm NaCl.
In contrast to shoot ethanol-insoluble dry matter accumulation, uniform and non-uniform treatments had no significant effects on total root ethanol-insoluble dry mass (Fig 2B). There were no significant differences in root ethanol-insoluble dry mass between the low and high NaCl sides in any of the non-uniform treatments.
All gas exchange parameters in non-uniform treatments were measured for both shoot sides directly above each root half. A significant difference (P < 0·05) was observed in the treatment 10/450 mm NaCl, where all gas exchange parameters in the shoot side above 450 mm NaCl were 20–26 % lower compared with those measured in the shoot side above the 10 mm NaCl (data not shown). However, as no other differences between shoot sides were observed in the other treatments, the data presented are the average values of the measurements from the two shoot sides.
In uniform treatments, net photosynthetic rate (Fig. 3A) and stomatal conductance (Fig. 3B) declined as salinity in the medium increased, and at 670 mm NaCl net photosynthetic rate and stomatal conductance were decreased by 72 and 77 %, respectively, compared with those at 10 mm NaCl. In plants subject to non-uniform salinities, the net photosynthetic rate was similar to that of plants uniformly subject to 10 mm NaCl. However, applying 450 and 670 mm NaCl to one root half caused a 21–24 % decline of stomatal conductance compared with the uniform 10 mm treatment (Fig. 3B). Uniform high NaCl was, however, more inhibitory compared with non-uniform NaCl concentrations. The difference between uniform and non-uniform salt treatments was most apparent at the highest NaCl treatment, where net photosynthetic rate and stomatal conductance in plants growing with only one root half at 670 mm NaCl were 3·2- and 3·5-fold, respectively, those of plants with both root halves at 670 mm NaCl (Fig. 3).
Independently of the NaCl concentration in the high salt compartment, pre-dawn shoot water potential of plants exposed to non-uniform NaCl was equal to that of plants exposed to uniform 10 mm NaCl (Fig. 4A). In uniform treatments, shoot water potential decreased as NaCl concentrations in the medium increased, with the value at 450 mm NaCl almost 2-fold lower (i.e. more negative) than that for plants at 10 mm. Values for plants growing at 670 mm NaCl were not obtained as shoots were succulent and pressures applied were particularly high, and plants burst out of the Scholander chamber gasket at around −2·7 MPa.
In uniform treatments, the osmotic potential of expressed leaf sap declined significantly as NaCl concentrations in the root medium increased (Fig. 4B, C). For expanding leaves, osmotic potential at uniform 670 mm NaCl was 2-fold lower (i.e. more negative) than that of plants in uniform 10 mm NaCl. Similar results were found for expanded leaves of plants in uniform 670 mm NaCl. Non-uniform salinity had small effects on sap osmotic potential in both expanding and expanded leaves, compared with the effect of uniform salinity (Fig. 4B, C); as examples, in expanding leaves, the osmotic potential was 1·36- and 1·20-fold lower (i.e. more negative) at 10/450 and 10/670, respectively, compared with that in the uniform 10 mm control (Fig. 4B).
The contributions of Na+, K+ and Cl− to the sap osmotic potential of the leaves were calculated from the tissue ion data (discussed below); these ions were estimated to contribute 73–89 % of the sap osmotic potential in leaves of plants grown under uniform and non-uniform salinities. Estimates of ‘bulk turgor’ in the leaves could have been calculated from osmotic and water potential data; however, this approach does not take into account the partitioning of water and solutes between the apoplastic and symplastic compartments (Wardlaw, 2005). This indirect calculation of turgor may therefore lead to erroneous estimates and is further compromised in the present study by ions in salt bladders contributing to leaf sap osmotic potential. The role of leaf bladders in the storage of salt in the leaves of Atriplex species is well known: depending on leaf age, bladders may account for more than half of leaf Na+ (Aslam et al., 1986).
Ion concentrations were measured in roots, expanding and expanded leaves, and are expressed on a tissue water basis (Fig. 5). In plants grown in non-uniform NaCl concentrations, root ion concentrations differed between the two root halves and so data for each side are presented. Differences in K+ concentrations were also found in leaves above each root half in non-uniform treatments; K+ concentrations in the side above 10 mm NaCl were 23–52 % higher compared with those in leaves above the high salinity side. As Na+ and Cl− concentrations in non-uniform treatments were not significantly different between shoot sides, the data presented are the average values for the data pooled from the two shoot sides for each replicate.
In uniform treatments, Na+ and Cl− concentrations in root tissues increased with increasing NaCl concentrations in the medium, with tissue Na+ and Cl− concentrations at 670 mm being 5·1- and 5·8-fold higher than those at 10 mm, respectively (Fig. 5A, G). In non-uniform treatments, Na+ and Cl− concentrations in roots in each side were similar to the values measured in roots in uniform treatments at the same NaCl concentrations (Fig. 5A, G). There were little, or no, differences in K+ concentrations on an ethanol-insoluble dry mass basis in roots (data not shown), but as tissue water content declined at higher NaCl concentrations in the root zone (data not shown), K+ concentrations on a tissue water basis increased at 450 and 670 mm (Fig. 5D).
In uniform treatments, the concentration of Na+ in both expanding (Fig. 5B) and expanded leaves (Fig. 5C) increased with the external NaCl concentration. Compared with the uniform 10 mm control, plants at uniform 670 mm NaCl had Na+ concentrations 3·3-fold higher in expanding leaves and 2·8-fold higher in expanded leaves. However, with non-uniform treatments, Na+ concentrations in leaves were relatively constant across the range of non-uniform NaCl treatments and were only 30–75 % higher than in the uniform 10 mm control. For leaf Cl−, the general trends were similar to those of Na+ (Fig. 5H, I). Potassium concentrations in leaves were affected by salinity, uniform/non-uniform salt treatment and in the non-uniform treatments by the differences in salinity of each root half. In leaves of plants exposed to uniform NaCl concentrations, tissue K+ concentrations were highest at 10 mm NaCl and declined by 55–60 % at 230 mm and then remained relatively constant at NaCl concentrations up to 670 mm (Fig. 5E, F). In all non-uniform treatments, leaf K+ concentrations were generally higher than in uniform treatments with the same high salt concentration.
Ion concentrations in side branches and central stems followed similar trends to expanding and expanded leaves (data not shown).
The patterns of water use across treatments were similar after 7 and 21 d of treatment; the data for day 21 are presented here (Table 2). In all treatments, whole plant water use after 21 d was 1·4- to 1·5-fold higher than that measured after 7 d of treatment. In comparison with the 10 mm control, whole plant water use decreased by 77 % in the uniform 670 mm NaCl treatment, and by 31 % in the non-uniform 10/670 mm treatment. Under non-uniform conditions, at both times, most water uptake was from the compartment with 10 mm NaCl, but there was still some water uptake (7–13 % of total water use) from the compartment at 670 mm. There was no compensatory increase in water uptake from the low salinity side. The differences in water uptake between treatments were not mediated by changes in root or shoot growth; similar patterns in water use between treatments also occurred when the data were expressed on a root surface area (Table 2) or leaf area (see Table 2 footnotes) basis.
Midday shoot water potential was also measured in experiment 2, and for treatments 10/10 and 10/670 mm NaCl, average values were −1·43 ± 0·7 and −1·67 ± 0·3 (s.e.) MPa, respectively.
The present study shows that A. nummularia with a non-uniform salt distribution in the root zone (i.e. 10 mm NaCl in one half, up to 670 mm in the other half) maintained shoot growth and photosynthesis relative to the uniform low salt control (Figs 2A and 3A). This occurred even with NaCl in one half of the root system of 670 mm, a concentration sufficient to decrease shoot ethanol-insoluble dry mass by 65 % in a uniform treatment. The present results contrast with previous studies in which halophytes exposed to non-uniform salinity had up to approx. 100 % increases in shoot dry mass compared with control plants growing in NaCl-free conditions (Messedi et al., 2004; Hamed et al., 2008). The present data for A. nummularia would seem more appropriate for understanding halophyte physiology as, unlike the two previous split-root experiments with halophytes, we avoided use of the NaCl-free solutions that can cause ion deficiency in dicotyledonous halophytes (cf. Yeo and Flowers, 1980).
In accord with our second hypothesis, under non-uniform salinity most of the water (90 %) came from the low salt side (Table 2). However, compared with the low salt control, there was no increase in water uptake from the low salt side to compensate for the decrease in water uptake from the high salt side, as had been previously reported for some glycophytes exposed to non-uniform salinity in the root zone (West, 1978; Zekri and Parsons, 1990). Non-uniform salinity had no effect on the partitioning of root growth between the high and low salt sides during the 21 d of treatment. Previous experiments with Atriplex spp. have reported no effect of NaCl at concentrations up to 360 mm applied for 1 month on root dry mass, whereas shoot growth declined (e.g. A. nummularia, Greenway, 1968; A. griffithii var. stocksii, Khan et al., 2000). Thus, the effects on water use are not caused by changes in the partitioning of root growth between the two sides.
The 31 % reduction in water use by plants exposed to non-uniform 10/670 mm NaCl (calculated from Table 2, experiment 2) was generally consistent with the declines of 21–24 % in stomatal conductance in leaves of plants exposed to 10/450 and 10/670 mm non-uniform NaCl treatments, as compared with the uniform low (i.e. 10 mm) NaCl control (Fig. 3B, experiment 1). For some halophytes at uniform high salinity (700 mm), root signals have been suggested as the cause for stomatal closure (e.g. Atriplex portulacoides and Sarcocornia fructicosa; Redondo-Gómez et al., 2006, 2007). The slight decrease in stomatal conductance between plants exposed to non-uniform salinities (10/450, 10/670 mm) and uniform low salt controls referred to above in A. nummularia (present study) provides some evidence for root-to-shoot signalling from the root portion at high salinity. Such a signal could have conceivably been transported out of the high salinity root half via the xylem as some water uptake occurred from this side (Table 2).
Irrespective of the salinity on the high salt side, all plants with non-uniform salinities had pre-dawn shoot water potentials similar to the uniform low (i.e. 10 mm) NaCl control, suggesting that there was equilibration overnight of plant water potentials with the low salinity side. This has also been shown for a range of glycophytic species with non-uniform soil moisture (e.g. Bouteloua gracilis, Sala et al., 1981; Quercus spp., Bréda et al., 1995; Betula pendula, Fort et al., 1998) provided that there are enough roots in the high water potential side to enable the equilibration to occur (Améglio et al., 1998). Similarly, in experiment 2, plants subject to non-uniform salinity (10/670 mm NaCl) had similar midday water potentials to the uniform low (i.e. 10 mm) NaCl control. Despite the limitation that stomatal conductance and midday water potential were not measured on the same day, it is highly probable that the 21–24 % decline in stomatal conductance in plants subject to non-uniform salinity treatments, compared with the uniform low NaCl controls (Fig. 3), was sufficient to maintain the similar midday water potentials in plants subject to non-uniform salinity.
Interestingly, leaf osmotic potentials under non-uniform NaCl concentrations were not substantially affected by the level of NaCl on the high salt side; even with 670 mm NaCl in one root side, leaf osmotic potential was only 20 % more negative than that of the uniform low (i.e. 10 mm) NaCl control, and was approx. 20 % less negative than the osmotic potential of the nutrient solution on the high NaCl side (670 mm NaCl = −3·07 MPa). Given these data on osmotic potential, and that midday shoot water potential was also less negative than that of the 670 mm NaCl treatment solution, it is unclear how these plants maintained water uptake from the high salt side (up to 13 % of the total whole plant water use). The similarity in leaf osmotic and water potential of plants exposed to non-uniform salinity and zero salt controls has also been reported for some glycophytes, but in these cases leaf osmotic and water potentials were always more negative than those of the saline solutions (Kirkham et al., 1969; Zekri and Parsons, 1990).
Under non-uniform conditions, the root half exposed to high salinity had high concentrations of Na+ and Cl−, but there were not large increases in these ions in the shoot tissues, even when one root half was exposed to 670 mm NaCl (Fig. 5). In non-uniform salinity treatments, leaf Na+ and Cl− concentrations remained substantially below those in uniform treatments at 450 and 670 mm. Across all non-uniform treatments, plants maintained relatively constant Na+ and Cl− concentrations in leaves; this shows that tissue ion concentrations in A. nummularia are well regulated under non-uniform salinity in the root zone. Regulation of leaf ions is essential for salt tolerance, as adverse effects can result if concentrations become too high, even in halophytes (Flowers and Colmer, 2008).
In conclusion, the present study shows that non-uniform salinity is not damaging to the halophytic shrub A. nummularia, even when the NaCl concentration in the high salinity side does impede growth when applied to the whole root system. The present study adds to knowledge on halophyte physiology (reviewed by Flowers and Colmer, 2008), which has previously been studied at a range of uniform salinities in the root zone. Given the temporal and spatial salinity variations in most field situations (see Introduction), the responses described here to non-uniform salinity would undoubtedly contribute positively to the growth and physiology of halophytes in saline environments.
N.B. thanks the Australian Government's Endeavour Europe Award and The University of Western Australia Scholarship for International Fees. Support was also received from the School of Plant Biology (UWA), the Western Australian Government's Centre for Ecohydrology and the Future Farm Industries Cooperative Research Centre. We are grateful for the technical assistance of Mr Meir Altman of the Department of Agriculture and Food of WA.