Declining Hydraulic Efficiency As Transpiring Leaves Desiccate

T. J. Brodribb & N. M. Holbrook

Plant, Cell and Environment (2006) 29, 2205–2215 doi: 10.1111/j.1365-3040.2006.01594.x

Declining hydraulic efﬁciency as transpiring leaves desiccate: two types of response*

TIM J. BRODRIBB1,2 & N. MICHELE HOLBROOK2

1Department of Plant Science, University of Tasmania, PO Box 252-55, TAS, Australia 7001, and 2Department of Organismic and Evolutionary Biology, Harvard University, 16 Divinity Avenue, Cambridge, MA, USA

ABSTRACT between sites of water delivery and water loss while not

interfering with light harvesting or the CO2 diffusion path- The conductance of transpiring leaves to liquid water (K ) leaf way. In most plants, these demands are met by a hierarchi- was measured across a range of steady-state leaf water cal branching network of veins that terminate in extremely potentials ( ). Manipulating the transpiration rate in Yleaf small xylem conduits, tens to hundreds of microns from the excised leaves enabled us to vary in the range 0.1 MPa Yleaf - sub-stomatal cavities where the bulk of evaporation takes to less than 1.5 MPa while using a flowmeter to monitor - place (Wylie 1943; Roth-Nebelsick et al. 2001). Current the transpiration stream. Employing this technique to mea- research suggests that this vascular arrangement generates sure how desiccation affects K in 19 species, including leaf a large resistance to water flow through leaves, representing lycophytes, ferns, gymnosperms and angiosperms, we found between 30 and 90% of the hydraulic resistance of the two characteristic responses. Three of the six angiosperm whole plant (Salleo, Nardini & Lo Gullo 1997; Nardini, species sampled maintained a steady maximum K while leaf Tyree & Salleo 2001; Sack et al. 2002). One important con- remained above 1.2 MPa, although desiccation of Yleaf - sequence of this is that even in plants with good access to leaves beyond this point resulted in a rapid decline in K . leaf soil water and high stem water potential, stomatal closure In all other species measured, declining led to a pro- Yleaf could be induced during the day due to water potential portional decrease in K , such that midday of leaf Yleaf gradients generated in the hydraulic passage from petiole unstressed plants in the field was sufficient to depress K leaf to sub-stomatal cavities. It follows therefore, that the effi- by an average of 37%. It was found that maximum K was leaf ciency of leaf water transport plays a governing role in strongly correlated with maximum CO assimilation rate, 2 processes linked to leaf water status, most importantly, sto- while K = 0 occurred at a slightly less negative than leaf Yleaf matal behaviour and photosynthetic gas exchange. at leaf turgor loss. A strong linear correlation across species The efficiency of water transport within leaves, or leaf between Yleaf at turgor loss and Yleaf at Kleaf = 0 raises the hydraulic conductance (Kleaf), has been demonstrated to was related to declining cell possibility that declining Kleaf vary enormously between species (Tyree et al. 1999; Bro- turgor in the leaf prior to the onset of vein cavitation. The dribb et al. 2005), ecological niches (Nardini & Salleo 2005; vulnerability of leaves rehydrating after desiccation was Sack, Tyree & Holbrook 2005) and seasons (Salleo et al. compared with vulnerability of leaves during steady-state 2002; Brodribb & Holbrook 2003a). Among this variation, evaporation, and differences between methods suggest perhaps the best physiological correlate with Kleaf is sto- that in many cases vein cavitation occurs only as Kleaf matal conductance (Brodribb & Holbrook 2004b; Brodribb approaches zero. et al. 2005; Nardini, Salleo & Andri 2005). Considering that these parameters represent liquid and gas phase con- INTRODUCTION ductances of water moving in a serial pathway through the leaf, this correlation, and similar relationships between Most of the water transported in vascular plants is destined stem hydraulic conductance and gs (Nardini & Salleo 2000; to replace leaf water sacrificed during the diffusive uptake Meinzer 2002) indicate that water potential gradients in of atmospheric CO2 for photosynthetic carbon fixation. non-stressed plants are relatively conservative. That is, dur- Irrigation of the leaf mesophyll thus represents the crux ing evolution plants appear to increase the conductance of of a plant’s vascular function. Leaves make complicated the vascular system in order to accommodate increased demands of the vascular system, requiring close proximity transpirational demand rather than operating at increased water potential gradients. Photosynthetic performance has Correspondence: T. J. Brodribb. Fax: 61 362 262698; e-mail: also been linked with Kleaf over a diverse selection of plants, [email protected] although the nature of the relationship is unclear in tropical angiosperms, where values of K appear to be higher than *This work was supported by an Australian Research Council leaf Fellowship (TJB), National Science Foundation (grant no. IBN in other species, but diurnally variable (Brodribb et al. 0212792) and by the National Geographic Society (grant no. 2005). Emerging relationships between Kleaf and anatomical 7475-03). characters that limit leaf gas exchange such as stomatal

2206 T. J. Brodribb & N. M. Holbrook

pore area index and palisade thickness (Aasamaa, Sober & examine the relationship between Kleaf and Ψleaf in plants Rahi 2001; Sack et al. 2003; Sack & Frole 2006) indicate that spanning a large range of morphological and anatomical hydraulic efﬁciency in the leaf vascular system is highly complexity, from lycopod to angiosperm. We investigate adaptive. Kleaf under conditions that closely replicate those experi-

While there is an ever-expanding library of data for Kleaf enced by leaves in situ by measuring Ψleaf under known variation between species, data describing the vulnerability conditions of leaf transpiration (Tyree et al. 1999). In a of the whole-leaf hydraulic pathway to dysfunction under novel application of this technique, we were able to create water stress remains sparse. Extensive work on stems has a large range of transpiration (E) in each species by using shown that the conductivity of the xylem is critically depen- a variable fan to force water loss, thus giving a measure of dent on water potential (Ψ), usually declining rapidly as Ψ Kleaf at a range of Ψleaf. Data from this steady-state tech- inside the xylem apoplast falls below a threshold value nique are compared with Kleaf vulnerability to dehydration (Sperry & Tyree 1988). Leaves are clearly sensitive to water determined by the non-steady-state pressure relaxation stress-induced depression of hydraulic conductance (Lin- technique (Brodribb & Holbrook 2003b), to indicate the ton & Nobel 2001; Cochard 2002; Brodribb & Holbrook processes responsible for impeding water flow at low water 2003b; Lo Gullo et al. 2003; Brodribb & Holbrook 2004a), potential. and due to the disproportionately large contribution leaves make to whole-plant hydraulic resistance, leaf vulnerability has the potential to dictate how plants respond to short- MATERIALS AND METHODS term water stress. This has been borne out by recent studies Plant material demonstrating a good correspondence between turgor loss, stomatal closure and leaf hydraulic dysfunction (Brodribb A list of 19 species, designed to span a large climatic as well & Holbrook 2003b). However, the exact nature of the as morphological and phylogenetic range, was sampled in temperate forest in Hobart, Australia, and Harvard Forest, decline in Kleaf with leaf water potential, remains poorly understood, hampered by a lack of techniques for probing USA and tropical forest at Santa Rosa National Park, Costa Rica; Lake Eacham, Australia; and Mt. Dzumac, Kleaf while leaves are simultaneously exposed to significant negative water potentials. To date, the only methods used New Caledonia. Among this selection were six lycophytes, two ferns, five gymnosperms (including a cycad) and six to examine the impact of water stress on Kleaf have used the angiosperms (see Table 1). All gymnosperm and angio- kinetics of Ψleaf relaxation (Brodribb & Holbrook 2003b), or measured infiltration rates of droughted leaves exposed sperm leaves were collected from small trees (< 4 m) in full to sub-atmospheric pressures (Trifilò et al. 2003). While sun, while only two of the fern species were collected in the both these techniques have successfully shown responses of sun and the other six collected in forest understorey. Only healthy mature leaves of a similar age were used in each Kleaf to drought, the conditions under which leaves are measured in both cases are rather distant from those experi- species sample so as to minimize within species variation to a minimum. enced by leaves in the field. In the case of Ψleaf relaxation, leaves are measured during a rapid collapse of the water potential gradient, while leaves exposed to vacuum infiltra- K determined by evapotranspiration tion are measured with intercellular spaces flooded and leaf Ψ virtually no gradient in leaf. If we are to determine whether This method calculates Kleaf of a leaf transpiring at a known reductions in Kleaf in the field are likely to be rare events steady-state as the ratio of transpiration flux over the associated with significant plant stress, or common events pressure differential between water entering the leaf and that limit photosynthesis on a diurnal basis, it is desirable steady-state Ψleaf (Boyer 1974). Excised, transpiring leaves to measure the response of Kleaf to desiccation in transpiring were connected to a flowmeter that measured the transpi- leaves exposed to a natural range of Ψleaf. Such conditions ration stream as it was sucked into the leaf. A modified provide the greatest chance of capturing the full range of flowmeter similar to those used to measure the hydraulic processes operating in leaves under natural conditions. conductivity of excised twigs (Brodribb & Feild 2000) was

The process responsible for reduced Kleaf at low water used where a filtered, degassed 0.01 M KCl solution passed potential is assumed to be the same as in stems, that is, from a reservoir, through a capillary tube and into the embolism derived from air-seeding of the pit membrane petiole of the sample leaf. A pressure transducer (PX-136; (Zimmermann 1983), and this is supported by observation Omega Engineering Inc. Stamford, CT, USA) measured of embolism in petioles (Bucci et al. 2003) and in veins the water pressure between a calibrated capillary tube and (Canny 2001; Salleo et al. 2001). However, other dynamic the leaf, and this pressure (sub-atmospheric due to the suc- processes are also thought to affect the efficiency of water tion created by the leaf) was logged and converted into a flow through leaves, including aquaporin activity (Nardini flow rate. The length of the capillary tube was tailored to et al. 2005) and conduit deformation under water stress the type of leaf being measured such that the pressure in (Cochard et al. 2004; Brodribb & Holbrook 2005). To com- the flowmeter remained in the range −0.05 to −0.15 bars, prehend the likely contribution of these (and other) factors thus avoiding tube cavitation while generating sufficient in the decline of Kleaf with Ψleaf, it is essential to comprehen- pressure to allow accurate calculation of flow. sively understand the response of Kleaf to Ψleaf. Here we A problem with the evapotranspiration method is that

Vulnerability of leaf hydraulics 2207

Table 1. Details of the 19 species sampled for Kleaf measurements

Ψleaf at

Species Family Kleaf response Habitat Kleaf = 0 Kleaf max. gs max. A max.

Pteridophytes Diphasiastrum digitatum Lycopodicaeae Linear Shade, temperate 1.90 4.71 0.0794 4.09 Lycopodium annotinium Lycopodicaeae Linear Shade, temperate 1.80 2.93 0.0857 2.96 Lycopodium clavatum Lycopodicaeae Linear Shade, temperate 2.16 3.70 0.0617 2.89 Lycopodium obscurum Lycopodicaeae Linear Shade, temperate 1.83 3.42 0.0571 4.56 Lygodium venustum Lycopodicaeae Linear Sun, tropical 2.25 5.23 0.109 5.38 Selaginella longipinnae Selaginellaceae Linear Shade, tropical 1.80 2.03 0.0728 1.70 Selaginella pallescens Selaginellaceae Linear Sun, tropical 2.50 4.83 0.2445 6.15 Tectaria conﬂuens Tectariaceae Linear Shade, tropical 2.45 3.19 0.0719 1.85 Gymnosperms Callitris rhomboidea Cupressaceae Linear Sun, temperate 4.01 11.30 0.1684 12.31 Cycas media Cycadaceeae Linear Sun, tropical 1.85 7.65 0.093 9.65 Pinus strobilus Pinaceae Linear Sun, temperate 2.10 10.20 0.1719 9.80 Retrophyllum comptonii Podocarpaceae Linear Sun, tropical 1.43 4.58 0.0711 5.72 Tsuga canadensis Pinacaea Linear Sun, temperate 1.61 6.60 0.1083 8.89 Angiosperms Eucalyptus globulus Myrtaceae Linear Sun, temperate 2.97 13.20 0.3592 16.85 Byrsonima crassifolia Malpigiaceae Sigmoid Sun, tropical – 17.19 0.4488 15.59 Curatella americana Dillenaceae Linear Sun, tropical 2.30 21.10 0.4043 15.21 Dalbergia refusa Fabaceae Sigmoid Sun, tropical – 19.50 0.6619 19.02 Genipa americana Rubiaceae Linear Sun, tropical l 2.55 17.00 0.4161 14.29 Rehdera trinervis Verbenaceae Sigmoid Sun, tropical – 20.6 0.4118 17.03

Parameters include the type of regression found to best describe the response of Kleaf to decreasing Ψleaf; habitat from which leaves were sampled; X-intercept of the linear function Kleaf = f(Ψleaf) + c showing the water potential at which Kleaf = 0 (only in species with linear −2 −1 −1 −2 −1 response functions); mean maximum Kleaf (mmol m s MPa ); mean maximum stomatal conductance (mmol m s ) and mean maximum −2 −1 instantaneous rate of CO2 uptake (µmol m s ). rapid hydration of leaves often results in the hydropassive by two fine wire thermocouples held in contact with the closure of stomata, thus arresting transpiration. This was adaxial surface of the sample leaf by a coarse nylon mesh. overcome here by minimizing the time between leaf exci- Temperatures of the water entering the leaf as well as the sion and connection to the flowmeter (to typically 60– flowmeter were also measured by thermocouples, and the 180 s), avoiding excessive leaf wetting, and by driving tran- whole system (including the leaf) maintained as isothermal spiration with heated air as soon as the leaf was connected as possible (by directing a regulated proportion of the heat to the flowmeter. These measures avoided a rapid rise in gun flow over the flowmeter) to avoid problems associated

Ψleaf as leaves were connected to the flowmeter, thus elim- with temperature gradients. Because of the complicating inating hydropassive stomatal closure. effect of temperature on Kleaf calculation (due to viscosity Leaves were measured between 1000 and 1530 h and effects and membrane fluidity), the leaf temperature range were maintained in full sun (1700–2000 µmol photons for each species was confined to 5 °C. In order to minimize m−2 s−1) after sampling and during measurement. Leaves the correlation between leaf temperature and E, airflow were always acclimated to sunlight for a minimum of 2 h over the leaf was modulated to ensure that the warmest prior to measurement. The sampling process involved cut- leaves of each species were measured at both low and high ting a small branch underwater, trying to avoid leaf wetting. steady-state evaporation rates. To avoid any Kleaf signal gen- The sample leaf was carried to the flowmeter where it was erated by circadian rhythmicity (Nardini et al. 2005), leaves re-cut under the perfusing solution at the petiole and rap- were measured between 1000 and 1530 h, and the imposed idly connected to the flowmeter. A heat gun (HG 1100; evaporative gradient was varied such that equal numbers Makita, Aichi, Japan) was then used to create the desired of leaves in the morning and afternoon were exposed to transpiration flow by modifying the airflow and/or temper- low and high Ψleaf. Sample size per species ranged between ature of the leaf. Leaves were maintained in full sun at 30 and 80 replicate leaves from a minimum of five plants. temperatures ranging from 25 C to a maximum of between The half-time for water potential equilibration after 38 and 40 °C; the upper temperature limit for each species alteration in E was either measured from the kinetics of was determined by the maximum temperature found to Ψleaf relaxation in excised leaves rehydrating underwater, produce a reversible 20% decrease in CO2 uptake in the or taken from previous work (Brodribb & Holbrook leaves of each species. This was considered a reasonable 2003b). Among the species sampled here, the range of Ψleaf test to ensure that membrane fluidity was minimally relaxation half-times was 8 to 40 s, thus requiring that impacted by temperature. Leaf temperature was measured leaves were maintained at steady state (< 3% change in

flow over 20 s) for at least 3 min before being removed from KFleaf= n leaf(∆Y leaf A, leaf ) (2) the flowmeter and immediately sealed into humidified plas- −2 −1 tic bags that were then placed into a pressure chamber where Kleaf = leaf hydraulic conductance (mmol m s −1 (SoilMoisture, Santa Barbara, CA, USA) for determina- MPa ); νleaf = viscosity of water in the sample leaf relative tion of Ψleaf. The balance pressure was determined with a to 20 °C; ∆Ψleaf = tube to leaf water potential difference 2 pressure gauge (± 0.1 psi), and a dissecting microscope used (MPa); Aleaf = leaf area (m ). to scrutinize the cut end of the petiole for the exact moment of sap exudation. Precise determination of Kleaf requires Vulnerability determined by Yleaf relaxation that Ψleaf measured by the pressure chamber accurately reflects the driving force behind hydraulic flow in the leaf. In a sub-sample of five species (Byrsonima, Callitris,

We were therefore careful to test for changes in the balance Curatella, Genipa and Rehdera), the response of Kleaf to Ψ pressure of leaves after the initial determination of leaf Ψleaf was measured by water potential relaxation to pro- (approximately 20 s after removal from the flowmeter) as vide a comparison with data collected under steady-state this might reflect redistribution of water within the leaf due evaporation. Six branches from three trees of each spe- to possible tissue compartmentalization. Typically leaves cies were collected in the morning and each branch sub- were measured immediately after removal from the flow- divided into three to four samples, all of which were meter, and then re-measured 5 min (and occasionally allowed to desiccate to a range of water potentials from 10 min) later to determine the extent of possible drift in approximately −0.4 to −3.0 MPa before being carefully

Ψleaf due to heterogeneous pressure distribution in the leaf. bagged to arrest water loss. We employed the technique

In about 3% of leaves, drift in the balance pressure was of Brodribb & Holbrook (2003b) to calculate Kleaf from observed to exceed 5%, in which case it was assumed that the kinetics of Ψleaf relaxation in leaves rehydrated Ψ insufﬁcient time had elapsed for steady-state leaf to estab- through the petiole. Initial Ψleaf was determined either by lish, and the reading was discarded. Following determina- measuring leaves neighbouring the sample leaf, or in tion of Ψleaf, leaf area was measured with a digital camera large leaves by sampling a small part of the target leaf (Nikon, Tokyo, Japan) and image analysis software (Image prior to rehydration. Sample leaves were then cut at the J, National Institute of Health, USA). In the case of the petiole, while underwater, and allowed to rehydrate in lycopods and Selaginella, whole shoots were used, and leaf full sun for 30, 60 or 90 s depending on the initial Ψleaf. area was measured with all leaves/microphylls removed. Final Ψleaf was measured with the pressure chamber and

Clogging of the cut end of the petiole proved to be prob- Kleaf calculated from the ratio of initial to final Ψleaf and lematic in some species, particularly those with resins in the the leaf capacitance (Eqn 3). petiole. To minimize the impact of clogging, the cortex was KC= ln[]YY( t n ) , (3) removed where possible, and the petiole cut several times leaf leaf o f leaf in clean perfusing solution to ensure that the initial influx where Ψo = initial water potential (MPa); Ψf = final water of water into the leaf did not carry with it resins or chemi- potential (MPa); t = duration of rehydration (s); Cleaf = leaf cals likely to block the xylem. In the case of species where capacitance (mmol m−2 MPa−1). resin canals were distributed throughout the petiole/stem, a branch was cut and all leaves removed except the target Pressure–volume (PV) relations leaf. This relatively large segment of stem was then attached to the flowmeter applying the principle that providing an Two leaves from each of three replicates of each species excess of xylem pathways to the leaf should overcome were sampled for determination of leaf turgor dynamics localized flow interruption. In general, leaves were and leaf capacitance from PV analysis (Tyree & Hammel removed and re-cut if the flow had not stabilized after 1972). Mean leaf capacitance (Cleaf) for each species was 10 min. measured from six fully expanded leaves using the slope During the flowmeter measurements, leaf temperature, of the leaf pressure–volume relationship (Tyree & Ham- air temperature, capillary tube temperature, relative mel 1972). Branches were cut underwater in the morning humidity and pressure transducer voltage were all logged and rehydrated until Ψleaf was > −0.05 MPa, after which at 5 s intervals and the data stored on a datalogger (CR- leaves were detached for PV determination. Leaf weight 10X; Campbell Scientific Inc., Logan, UT, USA). The flow and water potential were measured periodically during into the leaf was determined by multiplying the pressure slow desiccation of sample leaves in the laboratory. The differential across the capillary tube by its hydraulic con- initial (linear) slopes of the relative water content ductance (Eqn 1), and Kleaf by dividing the steady-state flow (RWC) versus Ψleaf curves yielded the leaf capacitance Ψ by leaf (Eqn 2). function in terms of RWC. Calculation of Kleaf (mmol m−2 s−1 MPa−1) requires that leaf capacitance be calcu- FPK=∆ 1 n , (1) tube tube lated in absolute terms and normalized by leaf area. To where F = flow into petiole (mmol s−1); ∆P = the pressure do this, the capacitance calculated from the PV curve differential across the capillary tube (MPa); Ktube = was multiplied by the saturated mass of water in the leaf capillary tube hydraulic conductance (mmol s−1 MPa−1); and then divided by leaf area (Koide et al. 1991; Bro-

νtube = viscosity of water in capillary tube relative to 20 °C. dribb & Holbrook 2003b). Leaf areas were measured as

) Statistics

–1 40 s

Both linear and sigmoid curves were ﬁtted to Kleaf versus –2 Ψleaf data from each species obtained using both vulnerabil- m ity methods. Responses were classiﬁed as sigmoid when the 30 r2 > 0.80 for the sigmoid function, while an r2 > 0.80 for a

linear regression signiﬁed a linear response of Kleaf to

Ψleaf. An exponential sigmoid function of the form ab()Yleaf − 20 y =+100/e( 1 ), where y = % loss of Kmax ﬁtted (Pam-

menter & Van der Willigen 1998), and Kmax was deﬁned as

the mean Kleaf at Ψleaf > −1 MPa. Curves were ﬁtted using 10 the statistical package JMP (SAS Inst., Cary, NC, USA).

RESULTS Transpiration flux (mmol 0 Application of the heat gun to modify the temperature and 0.0 0.5 1.0 1.5 2.0 2.5 boundary layer of sample leaves was highly effective in Leaf water potential (–MPa) producing a large range of E in all species (Fig. 1). Maxi- mum flow rates were achieved by the combination of Figure 1. Typical data from leaves [Rehdera trinervis (open high airflow over the leaf and high leaf temperature ° circles) and Eucalyptus globulus (closed circles)] attached to a (Tmax = 40 C in tropical tree species). Immediately upon flowmeter providing water to the petiole at approximately directing heated air over the leaf, the water flow into the atmospheric pressure. Each point represents a steady-state petiole was observed to rise rapidly to a maximum value, measurement from one leaf, and the pooled data for each species stabilizing typically after 60–90 s (Fig. 2). Leaves exposed shows the response of leaf water potential to increasing rates of to higher airflow and air temperature reached progressively transpiration. In both species transpiration increased, in response to increased airflow and leaf temperature, to a maximum rate, higher steady-state E. Maximum transpirational fluxes above which E was non-responsive to forcing due to stomatal induced were up to three times larger than maximum E closure. measured in the field (unpublished data). Once a threshold rate of evaporation was exceeded however, stomatal clo- projected areas with a digital camera and image analysis sure was triggered and E reached a steady-state below its software. initial maximum (Fig. 2). Increasing E was associated with a decline in Ψleaf, however, the relationship between E and

Maximum assimilation and ﬁeld Yleaf 40 Mean maximum assimilation rate was measured in each leaf = –1.20 MPa –1

species from a sample of 10 leaves. Measurements were s ) made at 0700–0800 h while plants were maximally –2 30 hydrated. A portable gas exchange system (Li-6400; Li-Cor,

Lincoln, NE, USA) was used to measure CO2 uptake, transpiration and stomatal conductance in a ventilated 20 cuvette which provided a saturating quantum flux of 1800 µmol quanta m−2 s−1 during measurements. Leaf tem- + heat 10 perature was not controlled, but remained in the range 25– leaf = –1.77 MPa 36 °C in all measurements. Small shoots from lycopods, Water flow at petiole (mmol m Water flow at petiole (mmol Selaginella and some conifer species were dissected and photographed following removal from the cuvette to 0 0100200 300 400 500 600 ensure that the total projected leaf area within the cuvette Time (s) was measured. In the same sub-sample of species used for determination Figure 2. Two traces showing water flow into the petiole of two Ψ of vulnerability by leaf relaxation, diurnal trends in field leaves of Byrsonima crassifolia and the leaf water potential Ψleaf were measured during the wet season under conditions immediately after removal from the flowmeter. The first leaf (open of high soil moisture availability (pre-dawn Ψleaf ≅ 0). Three circles) shows no evidence of stomatal closure or depression of Kleaf −2 leaves from three trees of each Costa Rican species were even at the very high transpiration rate of close to 30 mmol m −1 sampled hourly from 0800 to 1500 h on two cloudless days s . By contrast, the second leaf (closed circles) initially exceeds a transpiration rate of 34 mmol m−2 s−1 apparently triggering (2 and 18 August 2005), while 10 leaves of the Australian stomatal closure, after which the steady-state E is markedly fern species (Tectaria confluens) were sampled between reduced. Despite the reduction in steady-state E, the Ψleaf of this

1200 and 1500 h on two dry sunny days (20 and 22 Novem- second leaf remained highly depressed, indicating that Kleaf is much ber 2005). lower than the ﬁrst leaf.

30 22 Rehdera Genipa 20 ) 25 ) 18 –1 –1 16 MPa MPa

20 14 –1 –1 s s 12 –2 15 –2 m m

10 8 10 6 (mmol (mmol

leaf 5 leaf 4 K K 2 0 0 Byrsonima Tectaria

) ) Figure 3. Changes in Kleaf in response to

–1 20 –1 3 decreasing Ψleaf under forced evaporation.

MPa MPa Two species (angiosperms Rehdera trinervis

–1 15 –1 and Byrsonima crassifolia) show sigmoidal s s Ψ –2 –2 2 responses to leaf while the other two (the m m 10 angiosperm Genipa americana and fern Tectaria conﬂuens) show linear responses

to decreasing Ψleaf. Mean minimum diurnal

(mmol (mmol 1 5 Ψleaf for each of these species measured in leaf leaf K K the ﬁeld during the wet season is shown as a vertical line (with dotted SD). According 0 0 to these data, both species with linear K 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 leaf response to Ψleaf should exhibit signiﬁcant

Leaf water potential (–MPa) Leaf water potential (–MPa) diurnal depression of Kleaf on sunny days.

Ψleaf was non-linear in most species due to a disproportion- response of Kleaf to Ψleaf, midday values of Ψleaf (with fully ate decrease in Ψleaf as E increased. All species exhibited a hydrated soil) remained above the threshold inducing sig- maximum limit for E, above which stomatal closure was nificant depression of Kleaf (Fig. 3). apparently triggered by low leaf water potential (Fig. 2). Maintaining forced evaporation from the leaf once sto- Comparison of methods matal closure was initiated resulted in low leaf water potentials accompanied by significantly reduced E (Fig. 2). Drift In five species from each of the vulnerability classes in balance pressure of leaves 10 min after removal from the described earlier, the response of Kleaf to Ψleaf was tested flowmeter was less than 5% indicating a high degree of using the pressure relaxation method to help pinpoint the pressure homogeneity in the leaf while transpiring at steady tissue responsible for inhibiting Kleaf as Ψleaf declines. All state. species including those with linear responses under steady-

The relationship between E and Ψleaf was non-linear in state flow (above) exhibited sigmoid-shaped responses of all species measured, resulting in two characteristic pat- Kleaf to Ψleaf when vulnerability was measured by relaxation terns in the response of Kleaf to Ψleaf. Most species, includ- (Fig. 4). Despite the contrasting shape of vulnerability ing all sampled conifers, cycads, lycophytes ferns and curves in two of the four species, the extrapolated maxi- three of the six angiosperms, exhibited a linear decline in mum Kleaf (at Ψleaf = 0) derived by each method was similar, Ψ Kleaf with leaf as leaves were exposed to increasingly except in Byrsonima crassifolia where maximum Kleaf deter- large evaporative fluxes (Fig. 3). A second relationship mined by relaxation was significantly higher than that mea- was characterized in the remaining three angiosperms sured by evaporative flux (24.4 mmol m−2 s−1 MPa−1 versus −2 −1 −1 sampled, where the response of Kleaf to Ψleaf presented a 17.3 mmol m s MPa , respectively) more ‘typical’ sigmoidal shape as seen in stem xylem vul- Among the 16 out of 19 species that demonstrated a nerability curves (Fig. 3). Linear or sigmoid regressions linear response of Kleaf to Ψleaf, the X-intercept of the Kleaf Ψ fitted to each respective Kleaf versus leaf plot yielded versus Ψleaf function was calculated to determine the water highly significant correlation coefficients in all cases potential at which Kleaf = 0. This analysis produced a range 2 (r > 0.72; P < 0.01). of Ψleaf from −1.4 MPa in the rain forest conifer Retrophyl-

Measurements of Ψleaf in all angiosperms and a single lum comptonii to −4.0 MPa in the dry forest conifer Callitris fern in the ﬁeld indicated that for the species with a linear rhomboidea. A strong correlation between the water Ψ Ψ response of Kleaf to leaf, ﬁeld leaf at midday would result potential at turgor loss and Ψleaf at Kleaf = 0 was observed 2 in a depression of Kleaf by an average of 37 ± 6% (SD; n = 4 (r = 0.77; P < 0.01), however, the slope of this relationship species) of maximum Kleaf expected at Ψleaf = 0 MPa (0.6) indicates that in most species, bulk leaf turgor loss

(Fig. 3). In the angiosperm species exhibiting a sigmoidal occurred before Kleaf reached zero (Fig. 5).

30 Rehdera Genipa ) ) 20 –1 25 –1 MPa MPa

20 15 –1 –1 s s

–2 –2 15 m m 10

(mmol (mmol 5 5 leaf leaf K K

0 0 0123 01234

Leaf water potential (–MPa) Leaf water potential (–MPa)

30 30 Figure 4. Responses of Kleaf, measured Byrsonima Curatella by steady-state evaporation (closed ) )

–1 –1 Ψ 25 25 circles) and leaf relaxation (open circles), to Ψleaf in four angiosperms. Sigmoid MPa MPa 20 20 regressions best described the water –1 –1 s s potential relaxation data in all species, –2 –2 15 15 while Genipa americana and Curatella m m americana (and most other species; see

10 10 Table 1) showed linear responses to Ψleaf

(mmol (mmol when Kleaf was measured during steady- 5 5 leaf leaf state evaporation. Only three species K K (including Rehdera and Byrsonima shown 0 0 here) produced sigmoidal responses to 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Ψleaf when Kleaf was measured under Leaf water potential (–MPa) Leaf water potential (–MPa) conditions of steady-state evaporation.

Ψ Kleaf versus instantaneous CO2 uptake here, the manner in which Kleaf responded to leaf fell into two distinct patterns. From a sample size of 19 species, 16 Maximum values of Kleaf for each species were strongly species demonstrated a linear decline in Kleaf with Ψleaf. The assimilation rate correlated with mean maximum CO2 remaining three species, all angiosperms, presented a (Fig. 6). The best ﬁt for this correlation was a curve describing an exponential rise to a maximum assimilation rate of 19.7 µmol m−2 s−1, that is, 4.5 −× []A =−35.. + 2321( −e 01. Kleaf ) . This indicates a decreasing 4.0 sensitivity of maximum CO2 uptake to increasing Kleaf at −2 −1 −1 values of Kleaf above 10 mmol m s MPa . The 3.5 angiosperms represented here are all sun plants with −2 −1 −1 Kleaf > 10 mmol m s MPa , and hence in this sample of 3.0 angiosperms, CO2 ﬁxation was relatively insensitive to variation in K . By contrast, the non-angiosperm sample, leaf 2.5 which included many understorey species, demonstrated a high sensitivity of CO uptake to K . 2 leaf 2.0

DISCUSSION 1.5 In a potent demonstration of the limitations of the leaf vascular system, we found that excised leaves given unim- Leaf water potential at turgor loss (–MPa) 1.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 peded access to water were easily desiccated, by forced evaporation, to water potentials low enough to induce sto- Water potential at Kleaf = 0 (–MPa) matal closure. Manipulating leaf evaporation enabled the creation of a large range of steady-state Ψleaf from which Figure 5. The relationship between the water potential at which Ψ the response of K to Ψ could be calculated for the first Kleaf = 0 [or the X-intercept of the linear function Kleaf = f( leaf)] leaf leaf Ψ time in transpiring leaves. Declining leaf water potential and the mean (n = 5) leaf at bulk leaf turgor loss. In those species where Kleaf was linearly related to Ψleaf (16 out of 19 species), a compromised the conductance of whole leaves to liquid highly significant linear regression described the relationship water in the range of water potentials likely to be experi- between these variables (r2 = 0.77, P < 0.01). All data fall below the enced by leaves in the field. While the impact of low leaf 1:1 line (dotted) indicating that in all species, the water potential Ψ water potential was detrimental to all 19 species sampled at turgor = 0 was higher than leaf at Kleaf = 0.

25 to exhibit linear vulnerability to Kleaf, diurnal variation in

Ψleaf under conditions of abundant soil moisture would pro-

duce an expected diurnal depression of Kleaf in the order of ) 20 –2 37% (e.g. Genipa americana; Fig. 3). Assuming some s

–2 degree of isohydry in leaves of these species, a 37% reduc- m

2 tion in Kleaf would lead to a similar reduction in sustainable 15 CO

water loss from the leaf, and hence in CO2 assimilation

compared with the expected rate if Kleaf was to remain mol m maximal (Franks 2006). While the impact of a labile Kleaf on 10 gas exchange is relatively easy to predict, it is critical to

understand the physiology of the process leading to Kleaf depression so as to determine how different species mani- 5 Maximum A ( fest differences in the vulnerability of leaves to impaired

hydraulic function, and whether the depression of Kleaf is reversible. 0 051015 20 25 There are several candidates for producing impaired ﬂow under conditions of decreasing Ψleaf; the most standard is Maximum K (mmol m–2 s–1 MPa–1) leaf cavitation of the vein xylem. A large body of evidence indicates that xylem cavitation in the stem is common Figure 6. Shows the relationship between K and mean leaf under water stress, and the role of desiccation-induced cav- maximum instantaneous CO2 uptake (n = 10) for all species. A saturating exponential curve (r2 = 0.95) provided an excellent ﬁt itation in leaves is supported by observation of embolism for the data. However, when our sample of six sun-dwelling in major veins of plants (Canny 2001; Cochard 2002; Bucci angiosperms (closed circles) were considered in isolation, there et al. 2003). Certainly, xylem cavitation is inevitable at some was no dependence of assimilation upon Kleaf. point as leaves desiccate, and studies of water potential relaxation data (including those shown here), indicate a

dramatic loss of Kleaf at water potentials less than −2 MPa. classic sigmoid response of Kleaf to desiccation where Kleaf However, it seems less likely that cavitation is the cause of declined rapidly beyond a threshold Ψleaf. rapid linear decline in water potential observed here in

Among the few techniques for measuring Kleaf vulnera- most species at Ψleaf > −1 MPa. The only suggestion of leaf bility to desiccation, this novel technique of using transpir- vein embolism in this range of Ψleaf comes from a reported ing leaves provides the most realistic measure of leaf decrease in dye infiltration of fine veins; however, this was performance under natural conditions. Other techniques not typically linked to a decrease in Kleaf (Salleo et al. 2001; such as vacuum infiltration and pressure relaxation can pro- Trifilò et al. 2003). Indeed it seems improbable that leaves vide information about loss of Kleaf by embolism, if it is should invest carbon into a fine vein network that becomes assumed that emboli repair slowly, but cannot be used to embolized by water potentials equivalent to the midday determine the influence of flow or pressure-dependent pro- Ψleaf of well-watered plants. Such highly vulnerable fine cesses upon Kleaf. It could be argued that the high maximum veins would only function under conditions of low evapo- rates of evaporation induced here are also likely to create ration such as low light or high humidity; conditions under an unrealistic environment for the leaf; however, in all spe- which the enhanced Kleaf they confer will be of little service cies at least half of Kleaf determinations were made with to the leaf. leaves transpiring within their natural range. Hence, the We contend that the decline of Kleaf at water potentials water potential gradients and vapour fluxes in this half of above the leaf turgor loss point may arise from either tur- the leaves measured were equivalent to those known to gor-related changes in the conductivity of tissue down- occur at midday on plants with good access to water. The stream from the xylem conduits, or a change in the only shortcoming of this technique is that it is not possible proportion of evaporation from different hydraulic com- to lower the delivery pressure of water at the petiole to less partments of the leaf. While both turgor and osmotic signals than approximately −30 kPa, and hence the effect of drying are thought to induce changes in membrane properties soil cannot be directly simulated. (Lew 1996; Heidecker et al. 2003), turgor is a better candi- Setting aside this limitation for a moment and focusing date for signalling changes in tissue conductivity as it on undroughted plants, we can say with confidence that the changes in proportion with Ψleaf, while osmotic potential data presented here indicate that Kleaf in many species is changes in a non-linear fashion. Several features of the likely to change continually with variations in transpiration parenchymatous bundle sheath that surrounds the minor during the course of a day. Such excursions have been veins make it the probable location for such a turgor- observed in direct measurements of Kleaf (Bucci et al. 2003; sensitive governor in the leaf hydraulic pathway (Sack & Brodribb & Holbrook 2004a), although most techniques Holbrook 2006). It is thought that much of the transpira- currently used for measuring Kleaf might mask Kleaf depres- tion stream is shunted through the bundle sheath symplast sion due to the fact that leaves are measured only while by a suberized layer in perpendicular walls of bundle fully hydrated. Among the angiosperm species found here sheath cells (Van Fleet 1950; Canny 1990; Sack, Streeter &

25 under steady-state evaporation conditions (Fig. 4). If it is assumed that a signiﬁcant proportion of the hydraulic resistance of leaves lies downstream of the bundle sheath )

–1 20 (Boyer 1974; Salleo et al. 2003; Sack et al. 2004), then during pressure relaxation measurements the water potential of

MPa the bundle sheath will rise much more rapidly than the –1 15

s average leaf water potential (as measured by the pressure

–2 bomb). This would have the effect of relieving the ﬂow inhibition created by reduced bundle sheath turgor early in 10 the rehydration time course, thus erroneously inﬂating the

measured Kleaf at low initial leaf water potentials. Vulnera- (mmol m bility responses generated by pressure relaxation are there- leaf 5

K fore likely to indicate the onset of vein cavitation while

being relatively insensitive to perturbations in Kleaf caused by changes in bundle sheath turgor. 0 One ﬁnal possible explanation for a linear decline in Kleaf 0.0 0.5 1.0 1.5 2.0 during desiccation is xylem collapse (Cochard et al. 2004). Average leaf turgor pressure (MPa) This phenomenon has only been observed in a few species of conifers (Cochard et al. 2004; Brodribb & Holbrook

Figure 7. The relationship between Kleaf and leaf turgor pressure 2005) while its importance in other plant families is generated from regressions between both leaf turgor and Kleaf unknown. Two pieces of evidence tend to rule against this Ψ against leaf. Two species Curatella americana (open triangles) and as a likely factor however. Firstly, the water potentials Genipa americana (closed circles) are shown, representing the shown to initiate cell collapse in the xylem are substantially range of responses in the species sample. Most species showed more negative than −1 MPa, yet we see strong depression responses similar to that of Curatella with a proportion of of Kleaf in all species in the Ψleaf range 0 to −1 MPa. Secondly, maximum Kleaf remaining at leaf turgor loss. xylem collapse appears to respond sigmoidally to declining

Ψleaf, making it an unlikely candidate for producing the

Holbrook 2004). As such, bundle sheath cells are probably linear sensitivity of Kleaf to Ψleaf observed here. the first living cells traversed by water passing from the Interestingly, we found that three species among our stem to leaf. If this is so, then the hydraulic conductivity of sample of 19 demonstrated sigmoidal responses of Kleaf to this gateway into the mesophyll is susceptible to changes in declining Ψleaf when measured using both steady-state evap- membrane properties and possibly cell turgor. The concept oration and water potential relaxation techniques (Fig. 4). of a turgor-limited passage through the bundle sheath is These three species, all angiosperms, showed good agree- supported here by a good correlation between the leaf tur- ment between techniques in maximum Kleaf as well as the Ψ gor loss point and the Ψleaf at which Kleaf fell to zero (Fig. 5). threshold leaf leading to Kleaf depression (Fig. 4). This may In all species (except the three angiosperms displaying sig- indicate that the conductivity of leaf tissue in these moid vulnerability functions), a linear relationship between angiosperms is insensitive to turgor, or that the turgor of bulk leaf turgor and Kleaf could be defined (Fig. 7). Most the bundle sheath is maintained by osmotic adjustment, species however, conserved a small proportion of leaf con- thereby maintaining maximum Kleaf until the cavitation duction at zero leaf turgor. This is expected considering that threshold is reached. Considering the obvious benefit for around 30% of leaf hydraulic resistance is due to resis- the plant in avoiding Kleaf depression under average field tances between the fine veins and the sites of evaporation Ψleaf, it is probable that this behaviour is adaptive, and (Sack et al. 2003). As such, while the leaf is transpiring, the possibly restricted to angiosperms. The disadvantage of bundle sheath cells will be at a higher water potential than such a threshold in Kleaf sensitivity to Ψleaf is that once a those cells downstream at the evaporating end of the minimum value of Ψleaf is transgressed, Kleaf falls precipi- hydraulic pathway (the cells assumed to contribute the bulk tously, presumably by xylem cavitation. Avoiding cavitation of Ψleaf). Hence, bundle sheath cells may retain a small thus requires closely regulated isohydry (Tardieu & Simon- proportion of turgor while cells at the sites of evaporation neau 1998), a condition that is satisfied by each of the are at turgor loss. angiosperms found here to express sigmoidal vulnerability The discrepancy between pressure relaxation kinetics (T.J. Brodribb, unpublished data). and steady-state evaporation techniques can also be recon- By contrast, a linear decrease in Kleaf during drought may ciled if Kleaf is sensitive to bundle sheath turgor. All mea- attenuate the impact of drying soil upon leaf conduction by surements of leaves, both here and in previous studies expanding the range of leaf water potentials to which sto- (Brodribb & Holbrook 2003b, 2004a) have shown that leaf matal must respond to maintain leaf evaporation within the Ψ vulnerability to Kleaf depression is best described by a sig- limits defined by Kleaf. In doing so, the decline in leaf in moid function whenever it is measured by water potential response to drought should be slowed, and hence Kleaf and relaxation kinetics. Interestingly, this includes those species Ψleaf are less likely to plummet catastrophically as might be found here to express a linear response when measured the case with a sigmoidal vulnerability. This concept is

© 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd, Plant, Cell and Environment, 29, 2205–2215 2214 T. J. Brodribb & N. M. Holbrook supported by observations of stomatal closure in response Foundation. We also thank Jill Britton for field support and to leaf water potentials significantly higher than those the staff of Parque Nacional Santa Rosa. found to initiate leaf vein cavitation in a number of ferns species (Brodribb & Holbrook 2004b). REFERENCES These two types of Kleaf dynamics in angiosperms are strongly reminiscent of the two types of photosynthetic Aasamaa K., Sober A. & Rahi M. (2001) Leaf anatomical charac- responses to water deficit described by Lawlor & Cornic teristics associated with shoot hydraulic conductance, stomatal conductance and stomatal sensitivity to changes of leaf water (2002). Photosynthetic data mostly from crop plants sug- status in temperate deciduous trees. Australian Journal of Plant gest that in one group of species, CO2-saturated assimila- Physiology 28, 765–774. tion declines in proportion with declining RWC, while in Boyer J. (1974) Water transport in plants: mechanism of apparent the other group maximum assimilation is insensitive to changes in resistance during absorbtion. Planta 117, 187–207. changes in RWC (while RWC > 75%). These similarities in Brodribb T.J. & Feild T.S. (2000) Stem hydraulic supply is linked to leaf photosynthetic capacity: evidence from New Caledonian Kleaf and photosynthetic responses to RWC may not be coincidental but rather indicate coordinated limitation of and Tasmanian rainforests. Plant, Cell & Environment 23, 1381– 1388. these two processes. Brodribb T.J. & Holbrook N.M. (2003a) Changes in leaf hydraulic Much interest has been generated recently by correla- conductance during leaf shedding in seasonally dry tropical for- tions between Kleaf and leaf anatomical traits such as vein est. New Phytologist 158, 295–303. density, and palisade thickness (Aasamaa et al. 2001; Sack Brodribb T.J. & Holbrook N.M. (2003b) Stomatal closure during & Frole 2006). It is hoped that these correlations will enable leaf dehydration, correlation with other leaf physiological traits. the prediction of leaf gas exchange properties and hence Plant Physiology 132, 2166–2173. ecological character from leaf anatomy. As such it is impor- Brodribb T.J. & Holbrook N.M. (2004a) Diurnal depression of leaf hydraulic conductance in a tropical tree species. Plant, Cell & tant to consider that in most of the species measured here, Environment 27, 820–827. Ψ Kleaf declines linearly with leaf, and hence the Kleaf of that Brodribb T.J. & Holbrook N.M. (2004b) Stomatal protection Ψ species measured at leaf = 0 MPa will not provide an accu- against hydraulic failure: a comparison of coexisting ferns and rate indication of the Kleaf expressed in transpiring leaves angiosperms. New Phytologist 162, 663–670. even with fully hydrated soil. Even in terms of understand- Brodribb T.J. & Holbrook N.M. (2005) Water stress deforms trac- heids peripheral to the leaf vein of a tropical conifer. Plant ing the relationship between maximum CO2 fixation and Physiology 137, 1139–1146. leaf hydraulics, it must be borne in mind that maximum Brodribb T.J., Holbrook N.M., Zwieniecki M.A. & Palma B. (2005) CO2 fixation in the field can only be realized under condi- Leaf hydraulic capacity in ferns, conifers and angiosperms: tions of maximum irradiance and moderate air movement, impacts on photosynthetic maxima. New Phytologist 165, 839– conditions that must result in Ψleaf falling significantly below 846. zero. For this reason, Kleaf will always be depressed below Bucci S.J., Scholtz F.G., Goldstein G., Meinzer F.C. & Sternberg L. maximum whenever leaves are photosynthesizing maxi- (2003) Dynamic changes in hydraulic conductivity in petioles of mally (except in species with sigmoid vulnerability). This two savanna tree species: factors and mechanisms contributing to the refilling of embolized vessels. Plant, Cell & Environment may be part of the explanation for the weak correlation 26, 1633–1645. between CO2 uptake and Kleaf in angiosperms (Fig. 6), given Canny M. (1990) What becomes of the transpiration stream? New that half the measured angiosperms displayed linear- and Phytologist 114, 341–368. half-sigmoidal vulnerability. Canny M. (2001) Embolism and refilling in the maize leaf lamina In summary, we note that the hydraulic efficiency of and the role of the protoxylem lacuna. American Journal of leaves of each species examined in this study decreased in Botany 88, 47–51. a predictable manner as transpirational flux increased. Cochard H. (2002) Xylem embolism and drought-induced stomatal closure in maize. Planta 21, 466–471. Between species however, the response of Kleaf to transpi- Cochard H., Froux F., Mayr S. & Coutard C. (2004) Xylem wall ration was highly variable. All leaves measured under well- collapse in water-stressed pine needles. Plant Physiology 134, watered conditions in the field operated at water potentials 401–408. capable of inducing some degree of Kleaf depression, indi- Franks P.J. (2006) Higher rates of leaf gas exchange are associated cating that the phenomenon is probably widespread among with higher leaf hydrodynamic pressure gradients. Plant, Cell & plants. Combining this knowledge with the fact that the Environment 29, 584–592. Heidecker M., Wegner L.H., Binder K.A. & Zimmermann U. leaves contribute a large proportion of the whole-plant (2003) Turgor pressure changes trigger characteristic changes in hydraulic resistance, it is clear that the vulnerability of the electrical conductance of the tonoplast and the plasmale- leaves to Kleaf depression represents a defining character of mma of the marine alga Valonia utricularis. Plant, Cell & Envi- a plant’s physiology. ronment 26, 1035–1051. Koide R.T., Robichaux R.H., Morse S.R. & Smith C.M. (1991) Plant water status, hydraulic resistance and capacitance. In Plant ACKNOWLEDGMENTS Physiological Ecology (eds R.W. Pearcy, J. Ehleringer, H.A. Mooney & P.W. Rundel), pp. 161–183. Chapman and Hall, New We thank The Arnold Arboretum of Harvard University, York, NY, USA. the National Science Foundation (IBN 0212792), The Lawlor D.W. & Cornic G. (2002) Photosynthetic carbon assimila- Australian Research Council, The Harvard Forest, The tion and associated metabolism in relation to water deficits in National Geographic Society and the Andrew W. Mellon higher plants. Plant, Cell & Environment 25, 275–294.

Lew R. (1996) Pressure regulation of the electrical properties of of water flow through sugar maple and red oak. Plant Physiology growing Arabidopsis thaliana root hairs. Plant Physiology 112, 134, 1824–1833. 1089–1110. Sack L., Tyree M.T. & Holbrook N.M. (2005) Leaf hydraulic archi- Linton M.J. & Nobel P.S. (2001) Hydraulic conductivity, xylem tecture correlates with regeneration irradiance in tropical cavitation, and water potential for succulent leaves of Agave rainforest trees. New Phytologist 167, 403–413. deserti and Agave tequilana. International Journal of Plant Sci- Salleo S., Nardini A. & Lo Gullo M. (1997) Is sclerophylly of ences 162, 747–754. Mediterranean evergreens an adaptation to drought? New Phy- Lo Gullo M.A., Nardini A., Trifilò P. & Salleo S. (2003) Changes tologist 135, 603–312. in leaf hydraulic and stomatal conductance following drought Salleo S., Lo Gullo M.A., Raimondo F. & Nardini A. (2001) Vul- stress and irrigation in Ceratonia siliqua (Carob tree). Physiolo- nerability to cavitation of leaf minor veins: any impact on leaf gia Plantarum 117, 186–194. gas exchange? Plant, Cell & Environment 24, 851–859. Meinzer F.C. (2002) Co-ordination of vapour and liquid phase Salleo S., Nardini A., Lo Gullo M.A. & Ghirardelli L.A. (2002) water transport properties in plants. Plant, Cell & Environment Changes in stem and leaf hydraulics preceding leaf shedding in 25, 265–274. Castanea sativa L. Biologia Plantarum 45, 227–234. Nardini A. & Salleo S. (2000) Limitation of stomatal conductance Salleo S., Raimondo F., Trifilò P. & Nardini A. (2003) Axial-to- by hydraulic traits: sensing or preventing xylem cavitation? Trees radial water permiability of leaf minor veins: a possible deter- 15, 14–24. minant of the impact of vein embolism on leaf hydraulics? Plant, Nardini A. & Salleo S. (2005) Hydraulic efficiency of the leaf Cell & Environment 26, 1749–1758. venation system in sun- and shade-adapted species. Functional Sperry J.S. & Tyree M.T. (1988) Mechanism of water stress-induced Plant Biology 32, 953–961. xylem embolism. Plant Physiology 88, 581–587. Nardini A., Tyree M.T. & Salleo S. (2001) Xylem cavitation in the Tardieu F. & Simonneau T. (1998) Variability among species of leaf of Prunus laurocerasus L. and its impact on leaf hydraulics. stomatal control under fluctuating soil water status and evapo- Plant Physiology 125, 1700–1709. rative demand: modeling isohydric and anisohydric behaviours. Nardini A., Salleo S. & Andri S. (2005) Circadian regulation of leaf Journal of Experimental Botany 49, 419–432. hydraulic conductance in sunflower (Helianthus annuus). Plant, Trifilò P., Nardini A., Lo Gullo M.A. & Salleo S. (2003) Vein Cell & Environment 28, 750–759. cavitation and stomatal behaviour of sunflower (Helianthus Pammenter N. & Van der Willigen C. (1998) A mathematical and annuus) leaves under water limitation. Physiologia Plantarum statistical analysis of the curves illustrating vulnerability of 119, 409–417. xylem to cavitation. Tree Physiology 18, 589–593. Tyree M.T. & Hammel H.T. (1972) The measurement of the turgor Roth-Nebelsick A., Uhl D., Mosbrugger V. & Kerp H. (2001) pressure and the water relations of plants by the pressure-bomb Evolution and function of leaf architecture: a review. Annals of technique. Journal of Experimental Botany 23, 267–282. Botany 87, 533–566. Tyree M.T., Sobrado M.A., Stratton L.J. & Becker P. (1999) Diver- Sack L. & Frole K. (2006) Leaf structural diversity is related to sity of hydraulic conductance in leaves of temperate and tropical hydraulic capacity in tropical rainforest trees. Ecology 87, 483– species: possible causes and consequences. Journal of Tropical 491. Forest Science 11, 47–60. Sack L. & Holbrook N.M. (2006) Leaf hydraulics. Annual Review Van Fleet D.S. (1950) The cell forms and their common substance of Plant Physiology and Molecular Biology 57, 361–381. reactions, in the parenchyma-vascular boundary. Bulletin of the Sack L., Melcher P.J., Zwieniecki M.A. & Holbrook N.M. (2002) Torrey Botanical Club 77, 340–353. The hydraulic conductance of the angiosperm leaf lamina: a Wylie R.B. (1943) The role of the epidermis in foliar organization comparison of three measurement methods. Journal of Experi- and its relations to the minor venation. American Academy of mental Botany 53, 2177–2184. Botany 30, 273–280. Sack L., Cowan P.D., Jaikumar N. & Holbrook N.M. (2003) The Zimmermann M.H. (1983) Xylem Structure and the Ascent of Sap. ‘hydrology’ of leaves: co-ordination of structure and function in Springer-Verlag, Berlin, Germany. temperate woody species. Plant, Cell & Environment 26, 1343– 1356. Received 22 May 2006; received in revised form 28 July 2006; Sack L., Streeter C. & Holbrook N.M. (2004) Hydraulic analysis accepted for publication 18 August 2006