While it is becoming increasingly clear that plant cells sense and respond to salinity stress by activating multiple sensing networks, much of our knowledge on root salt sensing and signalling has utilized uniform conditions, with no such studies attempted for heterogeneous salinities. Such experiments will generate valuable information on how salt sensing at the single cell level is integrated into organ-scale processes, revealing how the signal propagates and its effects on root system architecture, developmental trade-offs and root plasticity.Scaling up to the whole root level, the root system is a highly dynamic physical network that enables a plant to forage for re-sources and rapidly explore favourable soil patches. Under spatially heterogeneous soil salinities, preferential root growth can occur in the least saline compartment, compensating to different degrees for root growth inhibition in the saline patches . A split-root experiment that closely examined root morphology under hetero-geneous salinities revealed that compensatory root growth in the non-saline areas was associated with increased lateral root growth, which doubled compared to plants with both root halves in non-saline conditions . However, root proliferation in the non-saline compartment does not always occur, with several studies showing no differences, or even a decline, in root growth compared to measurements under uniform non-saline conditions . This highlights the complexity of interpreting how heterogeneous conditions alter root growth,blueberry grow bag because responses depend on timescale, salt concentration and species sensitivity to salinity. To understand root foraging it will be necessary to deter-mine whether heterogeneous salinities affect root anatomical features, in addition to any effects on root morph-ology.

Even when heterogeneous salinities do not alter root architecture, it remains possible that traits that reduce the metabolic cost of soil exploration, such as cortical cell enlargement and cortical senescence, could be beneficial. This has yet to be tested. Nevertheless, the anatomical traits that reduce the metabolic cost of root soil exploration are currently considered an advantage in water-, nitrogen- and O2 -limited soils as these improve water and nutrient uptake per unit investment in roots . This topic therefore merits greater research efforts to identify key root traits that maximize soil resource capture under heterogeneous salinity.Irrespective of environmental heterogeneity within the root-zone, plant water uptake is essential to maintain photosyn-thesis. Typically, water uptake from the non-saline side of the root system increases significantly, which is not always accompanied by increased root biomass . Roots can dynamically alter their water transport capacity to acclimate to the ever-changing soil conditions and rapidly explore favourable soil patches. Under heterogeneous salinity, preferential water uptake from the regions with the least negative water potentials are mediated by changes in root hydraulic conductivity that occur within hours of salt exposure . These are achieved through changes in the abundance or activity of water channel proteins named aquaporins that facilitate water diffusion across cell membranes . The activity of aquaporins is regulated at many levels, including altered transcription levels, channel gating between an open/ closed state by various mechanisms including phosphoryl-ation, pH or Ca2+, and changed cellular trafficking . Under heterogeneous conditions, increased water uptake from the non-saline roots has largely been attributed to changes in aquaporin expression levels . After applying 200 mm NaCl to one root half in split-root cotton seedlings , gene expression profiling revealed several aquaporin genes were up-regulated within 3 h in the non-salinized root half, resulting in 16 % higher root hydraulic conductivity when measured against NaCl-free controls . By contrast, both root hydraulic conduct-ivity and most of the differentially expressed aquaporin genes were largely inhibited in the high-salinity side .

The deposition of hydrophobic lignin and suber in in the cell walls of the exo- and endodermis also alters root hydraulic conductivity and restricts the free diffusion of solutes and water, including restricting entry of Na+ and Cl− from the soil into the vascular stream with high root-zone salinity . Accordingly, dynamic regulation of root hydraulic conductivity under heterogeneous salinities was also associated with altered expression of genes associated with cutin, suberin and wax biosynthesis in the salinized root portions . This could potentially explain decreased endo-dermal and exodermal permeabilities, which limit water and solute transport from the highly saline areas. The ability of plants to acquire and transport water from the roots to the leaves also depends on root anatomy and archi-tecture, and the combined hydraulic conductivities among root types and along the root length . Thus, over the longer term , increases in new root growth and altered root architecture and anatomy may have a more significant effect than localized changes in root hydraulic conductivity at the single root level. Nevertheless, our under-standing of the timescale and concentration-dependent drivers of the long- and short-term responses of roots to localized salinity is inadequate. As highlighted in the following sections and above, responses are expected to become increasingly complex when heterogeneous salinity interacts with other environmental factors, such as heterogeneous nutrients as discussed below, and their impacts on plant nutrient and water acquisition.Heterogeneous salinity can induce variable degrees of stomatal closure, with stomatal conductance similar to uniform salinity in some studies . However, most split-root studies indicate greater plant water use under heterogeneous than uniform salinity, at the same average root-zone salinity. This is mostly because plant water uptake from the nonsalinized part of the root system substantially increases, even exceeding water uptake from roots of nonsalinized plants .

Long-distance signalling in planta is implicated in regulating these plant water relationships under heterogeneous salinity by modulating root hydraulic conductivity and stomatal conductance. Although leaf water status is regarded as an important regulator of stomatal responses , it is generally determined by the non-salinized part of the root-zone under heterogeneous salinities . Considerable stomatal closure of these plants suggests non-hydraulic mechanisms of stomatal closure. Homogeneous salinity induced multiple phytohormonal changes in salinized roots, according to the duration of exposure, with phytohormones such as abscisic acid , auxin and cytokinins , and their crosstalk, mediating the balance between growth and sal-inity stress responses . The same applies to heterogeneous salinities. In cotton grown with heterogeneous salinity , 200 mm NaCl induced only tran-sient increases in root ABA concentration. Root ABA levels were similar to controls after 24 h, presumably as sustained up-regulation of ABA catabolism genes influenced root ABA concentrations more than concurrent up-regulation of ABA biosynthesis genes . Paradoxically, root ABA concentrations of the non-salinized roots exceeded those of salinized roots throughout the experiment, despite a limited and transient up-regulation of genes, implying considerable ABA transport into these non-salinized roots. Further studies are needed to elucidate the source of this additional ABA, since the shoot can regulate root ABA concentration , which in turn upregulates root hydraulic conductance . Heterogeneous salinity also altered the concentrations of other phytohormones in the non-salinized portion of split-root cotton plants: with indole acetic acid, isopentenyladenine and zeatin riboside concentrations increasing compared to their concentrations in plants that were not exposed to salinity . In this case,blueberry grow bag size increased root cytokinin concentrations correlated with increased expression of IPT genes, which were maximal 3 h after salinizing the other part of the root system. Measuring root water potential in a transpiring plant may help determine whether this was a transient response to altered root water relationships in the non-salinized roots. Such measurements are required in girdled and non-girdled plants to determine whether local root water relationships and/or a cumulative message from other parts of the plant regulate gene expression. Under heterogeneous salinity, Na+ accumulation in the non-salinized portion of the root system doubled compared to roots from non-salinized controls. Such Na+ accumulation depended on phloem transport from the salinized roots, as girdling prevented Na+ transport to these roots . Whether girdling eliminates changes in root phytohormone concentration in non-salinized roots, when the other part of the root system is exposed to salinity, needs to be addressed. Irrespective of whether changes in root phytohormone concentration occur, it is uncertain whether they actually affect shoot phytohormone concentrations and physiological responses, since root-to-shoot signalling under heterogeneous soil condi-tions depends on relative sap flow from different parts of the root system . Under heterogeneous salinity, changes in root phytohormone concentration in the salinized root system may have little impact on shoot physiology since these roots contribute relatively little to total transpirational flow . Interestingly, changes in root phytohormone concentration in the non-salinized roots may have a greater influence on shoot physiology, since these roots contribute most of the total water flux.

Grafting techniques allow the relative contribution of different parts of the root system to root phytohormone export to be evaluated , but to date this has only been attempted in plants exposed to different soil moisture levels and such experiments should be applied to plants with heterogeneous root-zone salinity.Though probably a common occurrence in drip irrigated crops, very few studies have simultaneously varied both salinity and nutrient distribution. The following discussion first considers experiments with only nutrient heterogeneity, before dis-cussing the integration of nutrients with salinity heterogeneity. In tomato, preferential nitrate uptake was found to occur from areas of the root-zone with higher electrical conductivity generated by locally high nutrient concentrations , suggesting a local response of roots exposed to high concen-trations probably due to their enhanced NO3 − uptake kinetics. Mathematical simulations of nutrient uptake under heterogeneous conditions of NO3 − and phosphate using the Barber–Cushman model found a greater impact of soil heterogeneity and root plasticity, with NO3 − uptake increasing 7–20 times under heterogeneous conditions . Root proliferation and increased uptake kinetics from the enriched root-zones accounted for up to 75 % of NO3 − supply of a plant and over 50 % of PO4 3− acquired from enriched soil patches. Simulations demonstrated that plants lacking plasticity of root growth or uptake always acquired less nutrients under heterogeneous NO3 − and PO4 3− distributions. In a split-root solution culture experiment on Lolium multiflorum, less than 24 h after depriving NO3 − from half the root volume, net NO3 − influx to roots in the nitrate-rich area increased, with root growth increments observed only after 1 week . Brassica napus responded similarly . This rapid variation in NO3 − uptake was strongly associated with altered root hydraulic conductivities, with a sudden increase in NO3 − concentration around the roots almost simultaneously increasing root hydraulic conductivity and preferential water uptake from the nitrate-rich patch . Split-root experiments applying NO3 − to a portion of the root system demonstrated a localized and reversible response, with N starvation on one side of the root system leading to compensatory and enhanced NO3 − uptake in the other root portion . Heterogeneous NO3 − distribution to split-root Acer rubrum and Betula papyrifera plants demonstrated a species-dependent response, with two-fold more fine roots measured for B. papyrifera in the high NO3 − portion than A. rubrum, yet similar total NO3 − uptake rate . Under heterogeneous conditions, A. rubrum had smaller leaves and N deficiency symptoms in the shoot portion directly above the nutrient-deficient root portion, while B. papyrifera had regular leaves with no visible deficiency symptoms. Vascular system architecture may explain this differential response . In species with sectored vascular systems , in which contiguous and largely exclusive vascular traces occur from a specific root to a specific branch, N deficiencies occurring in isolated parts of the canopy reflect the nutritional status of the specific root that feeds that branch. In contrast, other species have an integrated vascular system allowing nutrient transfer from an individual root to the canopy as a whole, avoiding the consequences of patchy nutritional deficiencies.In Betula pendula, dry matter allocation to roots can be modified in three different ways when the availability of mineral nutrients is limited: increased root growth in N-, P- or S-limited soils; decreased root growth when K+, Mg2+ and Mn2+ is limited; and no effect on root growth when Ca2+, Fe2+ and Zn+ are limited . Root growth plasticity in patchy soil enhances the ability of plants to fill the soil volume rich in nutrients and was the most important trait influencing species success . The ability of a plant to ‘find’ the nutrient-rich patch is essential if morphological/physiological root responses are to be expressed. For instance, while nutrient heterogeneity in Lolium perenne did not lead to preferential root growth in the nutrient-rich soil patches , there were overall increases in specific root length and root elongation throughout the entire soil profile compared to the uniform N treatment . This suggests that the overall plant N deficiency induced root elongation and not the patchiness perse.