If one could make a plant transpire at any rate E, the trajectory of E vs leaf water potential (Ψ) should look something like Fig. 1 (solid curve; Sperry et al., 2002). When E is zero, the leaf Ψ would equal the bulk soil Ψ (ignoring gravitational effects). As E is increased the Ψ will drop disproportionally because the hydraulic conductance of the flow path declines with increasingly negative Ψ. There are two well understood reasons for this decline: cavitation in the xylem, and soil drying in the rhizosphere between bulk soil and the root surface. Although there may be additional changes in hydraulic conductance with E, such as variable aquaporin activity in root or leaf membranes (Henzler et al., 1999), or variable KCl concentration in xylem sap (Zwieniecki et al., 2001), the Ψ-dependence of these factors is not well characterized, as opposed to the inevitable physical processes of rhizosphere drying and xylem cavitation.
The E vs Ψ trajectory cannot go to infinity, but has a maximum steady-state value of E: Ecrit (Fig. 1, open symbols) with an associated Ψcrit. Any higher steady-state rate of E is impossible, because the drop in pressure drives the remaining hydraulic conductance in the bulk soil-leaf pathway to zero, breaking apart the hydraulic continuum. The critical values of E and Ψ describe a physical boundary to gas exchange with respect to soil and plant hydraulics. Transpiration and plant Ψ must be regulated to stay within these physical limits or else canopy desiccation will occur. A drier soil will have a lower Ψ intercept and a flatter E vs Ψ trajectory with a lower Ecrit (Figure 1, dashed curve). Maximizing gas exchange while avoiding hydraulic failure means operating on the edge of dysfunction, and requires rapid stomatal control of E to prevent it from exceeding Ecrit.
Under many situations, xylem cavitation is more limiting than rhizosphere drying. In these cases, Ψcrit approximates the Ψ causing 100% loss of hydraulic conductance (Fig. 1, Ψcrit≈Ψ100). Brodribb & Holbrook estimate safety factors from hydraulic failure by measuring the leaf Ψ in detached leaves at full stomatal closure. This Ψ represents the lowest leaf Ψ permitted by stomatal regulation (vertical Ψmin≈Ψclosure line). The difference (Ψmin–Ψcrit) gives the safety margin in terms of leaf Ψ (Fig. 1, safety margin). Seedless vascular plants had broader safety margins than angiosperms from the same habitat, as shown diagrammatically in Fig. 1. The difference was primarily due to a less negative Ψclosure in the seedless vascular plants (Fig. 1P) vs angiosperms (Fig. 1A). The vulnerabilities to cavitation (and hence Ψcrit) did not differ systematically between the two groups. Note that Brodribb & Holbrook report safety margins slightly differently – as (Ψclosure–Ψ50), where Ψ50 is the leaf Ψ at 50% loss of conductivity.
Xylem cavitation has been recognized as a threat to water transport since the cohesion-tension theory was first proposed over 100 years ago. John Milburn was probably the first to postulate a link between control of xylem cavitation and stomatal regulation (Milburn, 1973). The concept is simple: it is mal-adaptive for stomata to permit the transpiration stream to dry itself up by cavitation. Avoiding this fate requires coordination between stomatal regulation and the process of cavitation so that the demand for water by the foliage does not exceed the xylem's supply capacity (Box 1). An efficient match of supply and demand has the stomata maximizing CO2 uptake by pushing the xylem to its carrying capacity (Jones & Sutherland, 1991). Just as a rope will hold the most weight while its fibers are yielding to the verge of mechanical failure, so will the xylem achieve its greatest flow rate when its conduits have cavitated to the verge of complete blockage (Fig. 1). Exploiting this strategy for getting the most out of xylem transport requires rapid stomatal responses to avoid hydraulic failure, and a mechanism for daily reversal of cavitation so that the feat can be repeated. Brodribb & Holbrook (see pp. 663–670 in the issue) are the first to take a comparative and evolutionary perspective on the coordination of stomatal and hydraulic conductances in plants.
‘The hallmark of important results is that they expose new avenues of investigation’
Ancestral stomatal regulation – club mosses and ferns
Brodribb & Holbrook compared seedless vascular plants (species of Lycophyta and Pteridophyta) with selected flowering plants from the same seasonal dry forest in Costa Rica. They found dramatic differences in the coordination of stomatal vs hydraulic conductances. The angiosperms closed their stomata at lower (more negative) values of leaf water potential (Ψ) and much higher values of leaf xylem cavitation – sufficient to drop the hydraulic conductance of the leaf by over 50%. This is consistent with a tight coordination between stomatal regulation and hydraulic failure – a riskier, but presumably more efficient, match of supply with demand (Fig. 1, narrow safety margin, A). By contrast, the seedless vascular plants closed their stomata at relatively high leaf Ψ and at less than 10% drop in hydraulic conductance from cavitation (Fig. 1, broad safety margin, P). This indicates a much more conservative strategy – keeping demand well below levels that would challenge supply. The striking difference between the groups was in their stomatal regulation rather than xylem traits because the comparison groups had similar resistances to cavitation.
The implication is that the ancestral state is a conservative pattern of stomatal regulation which keeps xylem pressures from rarely if ever causing cavitation. As the authors point out, such large safety margins from hydraulic failure would compensate for sluggish stomatal responses to vapor pressure deficit, changing plant conductance, and soil moisture. Walking the tightrope between maximizing gas exchange and avoiding hydraulic failure requires rapid stomatal responses to these factors. Furthermore, large safety margins in seedless vascular plants would be necessary in the absence of efficient cavitation repair mechanisms. If cavitation is forever, then its prevention becomes much more imperative. Many seedless vascular plants do show root pressure and thus have the potential to repair cavitation when soil is wet and transpiration is minimal (Sperry, 1983). However, xylem refilling has not been demonstrated in these basal vascular plants, and it is unknown whether they can reverse cavitation under transpirational conditions as seems to occur in many flowering plants (Bucci et al., 2003).
These conclusions are consistent with observations from many gymnosperms, strengthening the notion that large safety factors from hydraulic failure may be an ancestral condition. To mention a few examples, Pinus taeda, P. ponderosa and P. edulis all close their stomata at approx. −2 MPa, while their stem xylem does not even begin to cavitate until −3 MPa or below (Linton et al., 1998; Hacke et al., 2000; Hubbard et al., 2001). Like seedless vascular plants, under well-watered conditions, these conifers maintain enormous safety margins from cavitation. A difference is that these trees have considerably greater resistance to cavitation than the seedless vascular plants studied by Brodribb & Holbrook, which were completely cavitated by −3 MPa. The quantum leap in cavitation resistance may be related to the evolution of torus-margo pit membranes in gymnosperms, which appears to afford much lower flow resistance for the same protection from cavitation as the homogenous type of pit membrane in other groups (Hacke et al., 2004). It should be noted that while large safety factors accompany well-watered conditions, drought does cause significant root cavitation which can severely limit gas exchange in conifers (Linton et al., 1998; Hacke et al., 2000).
Evolution of xylem and specialized stomatal physiology
Brodribb & Holbrook's study makes one realize that the evolution of more specialized stomatal physiology is just as important as the much better-documented trends in xylem evolution. The evolution of vessels, in particular, is often cited as important for the tremendous success of angiosperms. However, vessels have also evolved in Permian Gigantopterids (Li et al., 1996), ferns (Carlquist & Schneider, 1997), and Gnetophytes (Doyle, 1998), yet these groups never achieved the dominance of angiosperms. Furthermore, some early angiosperm lineages appear to have actually lost vessels (Feild et al., 2002). Finally, vessel-bearing wood is not necessarily more efficient on a conductivity per area basis, as Brodribb & Holbrook show in their comparison of ferns vs angiosperms. Vessels by themselves may be only half the story. It seems possible that the additional evolution of rapid stomatal response mechanisms could provide a crucial competitive advantage by allowing better optimization of the water-for-carbon trade-off. Physiologists tend to neglect the basal vascular plants, but studying the comparative physiology of these organisms is important for a better understanding of macro-evolutionary trends in plants.
Brodribb & Holbrook provide a valuable ‘snapshot’ assay of safety factors from excessive cavitation across diverse growth forms, using the leaf Ψ at stomatal closure in detached leaves as a proxy for minimum leaf Ψ (Box 1). In following up on this work it would be useful to determine the in situ safety factors throughout the active growing period and also to discriminate between the importance of phylogenetic background vs growth form. The seedless vascular plants could be conservative in stomatal regulation because of their basal phylogenetic position, or because their growth form is more compatible with a water-miser strategy where conserving water is more important than maximizing gas exchange. A comparison of palms and tree ferns in similar habitats, for example, would minimize differences in growth form and circumstances that could select for different stomatal regimes independently of phylogenetic background. The hallmark of important results is that they expose new avenues of investigation.