Ld

HHa

b

c

M

a

ARR1AEA

KMRDPWX

1

lhteMHtpoepi1

h0

Flora 229 (2017) 1–8

Contents lists available at ScienceDirect

Flora

j o ur na l ho me page: www.elsev ier .com/ locate / f lora

ow resistance to cavitation and xylem anatomy partly explain theecrease in the endemic Rhamnus ludovici-salvatoris

anan El Aou-ouad a,b, Rosana López c, Martín Venturas c, Sebastián Martorell a,ipólito Medrano a, Javier Gulías a,∗

Department of Biology, the University of the Balearic Islands, 07122 Palma de Mallorca, SpainDépartement de Biologie, Université Abdelmalek Essaadi, Faculté des sciences, B.P. 2121, 93002 Tetouan, MoroccoGENFOR Grupo de Investigación en Genética y Fisiología Forestal, E.T.S.I. Montes, Universidad Politécnica de Madrid, Ciudad Universitaria S/N, 28040adrid, Spain

r t i c l e i n f o

rticle history:eceived 27 April 2016eceived in revised form9 December 2016ccepted 10 January 2017dited by Hermann Heilmeiervailable online 19 January 2017

eywords:

a b s t r a c t

The low recruitment success of some endemic species under Mediterranean environmental conditions isprimarily the result of a low ability to create an efficient (highly conductive) and safe (able to maintain anintact water column under negative pressure) hydraulic system. Consequently, their lower resistance todrought-induced cavitation may induce mortality in seedlings and, during summer droughts, in matureindividuals. In this study, the hydraulic safety (water potential at 50% loss of conductivity, P50), hydraulicefficiency (specific conductivity, Ks) and xylem anatomy were compared between Rhamnus ludovici-salvatoris, an endemic species of the Balearic Islands whose distribution area is being reduced, and twopopulations of Rhamnus alaternus, which is widely distributed along the Mediterranean basin. R. ludovici-

editerranean shrubshamnaceaerought stresslant hydraulicsood anatomy

ylem cavitation

salvatoris was found to be more susceptible to drought-induced cavitation and less efficient at conductingwater in comparison with R. alaternus. Moreover, R. ludovici-salvatoris demonstrated a lower vessel area,wood density and inter-vessel wall strength than R. alaternus. These results are in accordance with thelower ability of R. ludovici-salvatoris to recruit seedlings under Mediterranean conditions in comparisonwith R. alaternus; this may partially explain the reduction in the distribution area of R. ludovici-salvatoris.

© 2017 Elsevier GmbH. All rights reserved.

. Introduction

Mediterranean ecosystems, which feature an almost completeack of precipitation during the summer period, coinciding withigh temperatures and high light intensities, are considered amonghe most demanding environments for plant survival (Nardinit al., 2014). Endemic plants account for 7–8% of the flora of theediterranean islands (Cardona and Contandriopoulous, 1979).owever, judging from their declining distribution and high extinc-

ion rates (Alomar et al., 1997), some endemic species seem to beoorly adapted to their present environmental conditions. More-ver, endemic and relict plant species are especially vulnerable tonvironmental changes (natural or human-induced) owing to smallopulation sizes, which are exposed to increased inbreeding, lower

ndividual fitness and lower genetic variability (Ellstrand and Elam,993; Aizen et al., 2002; Angeloni et al., 2011).

∗ Corresponding author.E-mail address: javier.gulias@uib.es (J. Gulías).

ttp://dx.doi.org/10.1016/j.flora.2017.01.005367-2530/© 2017 Elsevier GmbH. All rights reserved.

Rhamnus ludovici-salvatoris R. Chodat is an endemic speciesfound only in the Gymnesic Islands (Mallorca, Menorca andCabrera) of the Balearic Archipelago, located in the westernMediterranean basin (Alomar et al., 1997; Fig. 1). By contrast, theclosely related species Rhamnus alaternus L. is distributed widelyacross the Mediterranean basin, from Portugal to Crimea (Tutinet al., 1968). These two sclerophyllous shrubs represent a typicalcase of contrasting distributions between phylogenetically relatedspecies that possess similar morphological characteristics. As nogenetic incompatibility exists between them, both species mayshare a common ancestor and isolation by time could be themain cause of the sympatric speciation (Ferriol et al., 2009). How-ever, the distribution area of R. ludovici-salvatoris has decreasedduring recent decades, with some populations becoming extinctrecently (Alomar et al., 1997). The low recruitment success of thisspecies due to high seedling mortality during the first summerdrought (Traveset et al., 2003, 2012) limits its regeneration after

disturbances to the ecosystem. In addition, R. ludovici-salvatorisdemonstrates low biomass accumulation regardless of the waterregime (El Aou-ouad et al., 2015)

2 H. El Aou-ouad et al. / Flora 229 (2017) 1–8

Fig. 1. Distribution of the Balearic endemic Rhamnus ludovici-salvatoris, including the location of the populations from which samples of both R. alaternus and R. ludovici-salvatoris were collected, with their Walter-Lieth climodiagrams (above the diagram area, from left to right: elevation above sea level of the population, mean annualtemperature and mean annual rainfall. To the left of the temperature axis, from top to bottom: the mean of the years absolute maximum temperature, the mean of the dailym minimm atic dm

mqla1t1t(2mc

ttspustt2xdv

aximum temperatures of the warmest month of the year, the mean of the daily

inimum temperature. Black bars below the months indicate freezing periods. Climaterial 1).

Drought leads to a water deficit in the leaf tissue; this affectsany physiological processes and can have significant conse-

uences on plant growth and survival. Among these processes, theoss of hydraulic conductivity in the xylem has been recognizeds playing an important role in plant survival (Tyree and Sperry,989). This phenomenon is due to xylem cavitation, a breakage ofhe water column under negative xylem pressure (Zimmermann,983), and the formation of embolisms. Resistance to xylem cavita-ion is positively correlated with drought resistance in plant speciesSperry and Tyree, 1988; Cochard et al., 2008; Blackman et al.,010). A reduction in water transport can result in declines in sto-atal conductance (Pratt et al., 2005) and lowered photosynthetic

apacity (Brodribb and Field, 2000).Survival in a given environment largely depends on plant ability

o create an efficient (highly conductive) and safe (able to main-ain an intact water column under negative pressure) hydraulicystem. Much research has been devoted to understanding howlant hydraulic systems have evolved to accommodate survivalnder different environmental conditions. Generally, wider ves-els allow large hydraulic conductance (Sperry et al., 2006), andherefore, greater rates of stomatal conductance and photosyn-hesis (Santiago et al., 2004; Zhang and Cao, 2009; Meinzer et al.,010). Conversely, the relationship between xylem anatomy and

ylem safety is more complex (Tyree et al., 1994). It has beenemonstrated that drought-induced embolism is related to theessel wall structure, that is, inter-vessel wall thickness (Hacke

um temperatures of the coldest month of the year, the mean of the year absoluteata were obtained from the records collected by MAGRAMA (2013) (Supplementary

et al., 2001; Cochard et al., 2008; Fichot et al., 2009), pit-pore andpit-membrane area (Wheeler et al., 2005; Christman et al., 2009),and pit-membrane porosity or physical-chemical properties (Chenet al., 2009; Cochard et al., 2009).

Losses in conductivity due to vessel implosion may explain thecorrelations between cavitation resistance and wood density (WD),xylem mechanical strength, and vessel and fiber anatomical prop-erties (Hacke et al., 2001; Baas et al., 2004; Jacobsen et al., 2005).WD is often taken as starting point when comparing species; this isbecause it has strong ecological implications (Falster, 2006). HighWD is associated with high survival (Sterck et al., 2006; Sperryet al., 2008) and resistance to physical damage by herbivores andpathogens (McCarthy-Neumann and Kobe, 2008).

Both Rhamnus species included in this study have beendescribed as anisohydric and desiccation-tolerant plants(Martínez-Ferri et al., 2000; Gulías et al., 2002). Consequently,because they operate with narrow hydraulic safety margins duringdrought conditions, they are often exposed to hydraulic failure(Gucci et al., 1997; Martínez-Ferri et al., 2000; Nardini et al., 2014).The observed differences between R. ludovici-salvatoris and R.alaternus in terms of germination and seedling emergence andsurvival (El Aou-ouad et al., 2014; Gulías et al., 2004; Travesetet al., 2003), biomass accumulation, water use efficiency (WUE)

and photosynthetic capacity (Gulías et al., 2002), could supportthe following predictions for these two species: i) R. alaternushas a higher resistance to drought stress-induced cavitation than

l. / Flo

RtdDmaia

2

2

ovosMmvOaes6usif

2

vpoca1RsmoTcmtuen

sew1fil

2v

wa

H. El Aou-ouad et a

. ludovici-salvatoris. This trait would explain to some extentheir contrasting fitness under Mediterranean conditions and theecline in distribution of the endemic R. ludovici-salvatoris. ii)ifferences in water availability in their main distribution areasay result in different xylem anatomy. iii) There is a relationship

mong hydraulic safety (water potential at 50% loss of conductiv-ty; P50), hydraulic efficiency (specific conductivity; KS) and xylemnatomy.

. Material and methods

.1. Plant material and study site

The study was performed in two populations of R. alaternus andne population of R. ludovici-salvatoris. In all cases, considered indi-iduals were adult plants (at least 15 years old). The first populationf R. alaternus is located at Esporles (39◦ 40′ N, 2◦ 34′ E), 200 m aboveea level (a.s.l.), on the North Slope of a hill on the southwesternallorca Island (Fig. 1). This population is in the understory of aixed pine-oak forest with clay-calcareous soil, where surrounding

egetation includes: Pistacia lentiscus, Ampelodesmos mauritanica,lea europaea and Cistus albidus. The second population of R.laternus is located at the University of the Balearic Islands (UIB)xperimental field (80 m a.s.l.; 39◦ 38′ N, 2◦ 38′ E; Fig. 1). Theampled R. ludovici-salvatoris population is located at Binifaldó, at00 m a.s.l., near the Lluc Monastery (39◦ 49′ N, 2◦ 53′ E; Fig. 1) in thenderstory of a large oak (Quercus ilex) forest with clay-calcareousoil. The mean annual temperatures and the mean annual rainfalln the habitats of these populations range from 12.7 to 18.1 ◦C, androm 468 to 1388 mm, respectively (Fig. 1).

.2. Xylem vulnerability to cavitation

The degree of vulnerability to embolism was inferred fromulnerability curves (VCs), created by plotting the xylem waterotential (�xyl, MPa) against the corresponding percentage lossf hydraulic conductivity (PLC, %). Two techniques were used toonstruct these curves: the bench-top dehydration method (Sperrynd Tyree, 1988) and the standard centrifuge method (Alder et al.,997), using a rotor that fits 14-cm stem segments. In the case of. ludovici-salvatoris, we used both methods to construct VCs withamples from the same population to test the agreement betweenethods. In fact, several studies have suggested that the results

btained by both methods are similar (Cochard et al., 2010, 2013).he R. alaternus samples used in the centrifuge-force method wereollected at UIB and the samples for the dehydration bench-topethod were collected in Esporles. Samples used for the dehydra-

ion bench-top method were collected three days earlier than thosesed for the centrifuge method. All samples were collected at thend of winter (mid-February 2012) when soil water availability isot limiting.

Prior to branch collection, the maximum length of xylem ves-els was determined using the air infiltration method along thentire length of large branches (Zimmermann and Jeje, 1981). Airas injected at the distal end of branches (2–3 mm diameter) at

00 kPa. The longest vessels found in six samples of R. alaternus andve samples of R. ludovici-salvatoris branches were 14 and 10 cm in

ength, respectively.

.3. Bench-top dehydration and standard centrifugeulnerability curves

In the bench-top dehydration method, two-year-old branchesere collected (cut in air) in order to prevent previously embolized

nd non-functional conduits of older xylem rings. Branches were

ra 229 (2017) 1–8 3

collected early in the morning and placed into plastic bags imme-diately in order to minimize dehydration. In order to preventcavitation by air-entry, only branches longer than 50 cm were sam-pled and segments were cut under water.

The hydraulic conductivity (K, kg s−1 m MPa−1) of shoot seg-ments of both species was measured following the methodof Sperry and Tyree (1988). K was measured at low pressure(2–3 × 10−3 MPa) in order to minimize the displacement of airbubbles in open vessels. All sample segments were perfused withultra-pure, deionized, degassed and filtered (0.2 �m) water with10 mM KCl. The flow rate was determined by repeated measure-ments of the water mass on an electronic balance and K wascalculated as the ratio between the flow through each segment andthe corresponding hydrostatic pressure gradient.

Branches were gradually dehydrated to cover a wide range of�xyl values (measured by using a pressure chamber, Soil Mois-ture Equipment. Corp., Santa Barbara, CA, USA). For each �xyl point(measured in five previously covered leaves), K was measured.For R. alaternus, which possesses high conductivity, 10-cm seg-ments from near the base of second-year shoots were used, andthe water flow through the segments over 2 min was measured. ForR. ludovici-salvatoris, which has low conductivity, 8-cm segmentsfrom near the bases of second-year shoots were used, and the waterflow through the segments over 4 min was measured. On aver-age, 20 s-year shoots were sampled per species. After measuring�xyl, shoots with a diameter of 2–3 mm were excised underwater;2–3 cm of shoot was discarded in order to avoid embolisms beinginduced by cutting. Following this, the initial hydraulic conductivity(Kin) was measured. Segments were repeatedly perfused (pressureof 175 kPa) to dissolve and expel air bubbles, until the maximumconductivity (Kmax) was obtained.

In the standard centrifuge method, shoots 14 cm in length weresimilarly prepared as explained in the bench-top method. Seg-ments were perfused and, after determining Kmax, branches werecentrifuged for 6 min at successively higher velocities, achievingtensions of −1, −2, −3, −4, −5, −6 and -7 MPa. K was measuredusing a XYL’EM system (Xylem Embolism Meter, Bronkhorst, Mon-tigny les Cormeilles, France) after imposing and relaxing eachpressure level.

The PLC for both methods was estimated as:PLC = 100 × (Kmax − K)/Kmax. The experiment was halted whenthe PLC reached approximately 90%. VCs were fitted usingSigmaPlot 10 software (Systat Software Inc., San Jose, CA, USA) toa logistic function (Pammenter and Vander-Willigen, 1998):

PLC(%) = 100/(1 + e(s/(25∗(P − P50))))

Where P50 is the pressure inducing a 50% loss of xylem conductivityand s is the slope of the VC at this point. In addition, xylem pressureat 12% loss of conductance (P12), an estimate of the xylem waterpotential at which embolism begins, and xylem pressure at 88%loss of conductance (P88), a proxy of the xylem water potential atfull embolism (Domec and Gartner, 2001), were calculated as:

P12 = P50 + 50/s

P88 = P50 − 50/s

Maximum xylem-specific conductivity (KS-max,kg m−1 s−1 MPa−1) was calculated by dividing Kmax by its xylemcross-sectional area (m2; e.g. Hultine et al., 2006).

2.4. Xylem anatomy

All the anatomical characteristics were examined in the sam-ples that were collected for the centrifuge VCs, that is, R. alaternuscollected at UIB and R. ludovici-salvatoris collected at Binifaldó. In

4 H. El Aou-ouad et al. / Flora 229 (2017) 1–8

Table 1Hydraulic xylem characteristics in two Mediterranean shrubs, Rhamnus alaternus and the Balearic endemic Rhamnus ludovici-salvatoris. Mean values ± standard errors (n = 5)are presented. KS-max (kg m−1 s−1 MPa−1), maximum xylem-specific conductivity; Slope (MPa−1), slope parameter (s) of the fitted vulnerability curves; P12 (MPa), pressure atwhich a 12% loss of conductivity is produced; P50 (MPa), pressure at which a 50% loss of conductivity is produced; P88 (MPa), pressure at which an 88% loss of conductivity isreached. P-values from a nested ANOVA (species and method nested under species as factors) are presented.

Variable R. alaternus R. ludovici-salvatoris P-value

UIBa Esporlesb Binifaldóa Binifaldób Species Method

KS-max 2.36 ± 0.38 1.03 ± 0.13 1.23 ± 0.14 0.43 ± 0.03 0.007 0.002Slope 8.2 ± 3.7 18.5 ± 3.2 16.6 ± 2.6 17.4 ± 3.7 0.300 0.162P12 −0.4 ± 0.6 −2.2 ± 0.5 −0.4 ± 0.4 −0.9 ± 0.6 0.241 0.136P50 −6.6 ± 0.3 −5.7 ± 0.3 −3.7 ± 0.2 −3.9 ± 0.3 <0.001 0.273P −12.8 ± 0.8 −9.2 ± 0.7 −6.5 ± 0.8 −6.9 ± 0.6 <0.001 0.030

elivss1oiHamtvsvvgtsqcatt(

2

Soswwvpfi7Wu

2

SCsdca

Ψxyl (MPa)

-7 -6 -5 -4 -3 -2 -1 0

PLC

(%)

0

20

40

60

80

100

Fig. 2. Vulnerability curves constructed for two-year-old xylem samples of Rhamnusalaternus (circles) and Rhamnus ludovici-salvatoris (triangles). PLC: percentage lossof hydraulic conductivity; �xyl: xylem water potential. Open circle: UIB population;Closed circle: Esporles population. Triangles: Binifaldó population. Full symbols cor-

88

a Vulnerability curves constructed with the dehydration bench method.b Vulnerability curves constructed with the centrifugal-force method.

ach selected branches, 4–5 cm-long segments were cut from theeft basal pieces of the branches selected; they were then preservedn FAA (formaldehyde, acetic acid and ethanol 70%, 10:5:85). Trans-erse sections (approximately 40 �m thick) were prepared using aliding microtome (Leica SM2400, Leica Microsystems GmbH, Nus-loch, Germany). Sections were stained using 0.5% safranine for–3 min and were mounted using a xylene-based medium. Imagesf the cross sections were taken with a light microscope system

nterfaced with a digital camera (Leica DM LA, Leica Microsystems,eerbrugg, Switzerland). Four randomly-selected 45◦ sectors werenalysed per sample. Calibration of each picture was made with aicrometer before measuring. The following variables were used

o describe the xylem anatomy: mean vessel diameter (Dmean, �m),essel frequency (VF, vessels mm−2), grouped lumen area of ves-els in relation to total xylem area (VGA, %), number of groupedessels (NGV), and ray frequency (RF). The percentage of groupedessels (PGV, %) was estimated by counting all the grouped and non-rouped vessels. The tangential lumen span (b) and the double-wallhickness (t) of all adjacent vessels within five randomly-selectedquare sectors (0.5 mm2) were measured in each sample. Subse-uently, the intervessel wall strength within these sectors wasalculated as (t/b)2 following the method of Hacke et al. (2001). Inddition, hydraulic vessel diameter (Dh) was calculated in each sec-or by dividing the sum of the vessel diameters to the fifth power byhe sum of vessel diameters to the forth power (Dh =

∑VD5/

∑VD4)

Kolb and Sperry, 1999).

.5. Wood density

Xylem density of all the samples was determined as follows.ix to seven stem segments per species were cut from one endf each sample and were trimmed to a length of 2–3 cm. Thesetem segments were split longitudinally, and the bark and pithere removed using a razor blade. The resultant xylem segmentsere soaked overnight in degassed water. Following this, the fresh

olume of each sample was determined according to Archimedes’rinciple (Hacke et al., 2000), by immersing the sample in a water-lled test-tube placed on a balance. The samples were stored at5 ◦C for 48 h, and their dry weight was subsequently measured.D (g/cm3) was calculated as the ratio of dry weight to fresh vol-

me.

.6. Data analysis

Regression coefficients and correlations were obtained usingigma Plot 10.0 software package (Systat Software Inc., San Jose,A, USA). A nested ANOVA with two factors: species and mea-

urement method nested under species was performed to revealifferences in hydraulic conductivity and xylem vulnerability toavitation parameters. Xylem anatomy and wood density data werenalyzed by performing a one-way ANOVA to reveal differences

respond to centrifugal-force methodology curves and open symbols to dehydrationbench-top methodology curves.

between species. The SPSS 16 software package (SPSS, Chicago, IL)was used to perform both ANOVA.

3. Results

The maximum xylem-specific conductivity (KS-max) of the sam-ples used to construct the VCs was found to differ between thetwo methodologies used (p = 0.002, Table 1). The observed dif-ferences could be attributed to the different age of the samplesand the variation of sample length and diameter required in thetwo methods; larger samples in the centrifuge method resulted inincreased resistance to water flow between vessel elements in com-parison with the shorter samples used in the dehydration method.Despite these differences in KS-max, the VCs generated by the dif-ferent methods did not differ in most of the parameters (p > 0.05;Table 1, Fig. 2), only in P88 differences between methods were sig-nificant (P = 0.03; Table 1), supporting the comparison of resultsobtained for these species using the centrifuge and the dehydra-tion method. The differences between the species were consistentand KS-max was significantly (p = 0.007) higher in R. alaternus thanin R. ludovici-salvatoris. In addition, both P50 and P88 were signifi-cantly (p < 0.001) lower in R. alaternus than in R. ludovici-salvatoris(Table 1).

Both species differed in several anatomical traits. R. alaternus

showed significantly higher wood density (WD), hydraulic vesseldiameter (Dh), vessel frequency (VF), and inter-vessel wall strength(t/b2) in comparison with R. ludovici-salvatoris (Table 2). By con-trast, ray frequency (RF) was significantly lower in R. alaternus

H. El Aou-ouad et al. / Flora 229 (2017) 1–8 5

Table 2Xylem anatomic features of Rhamnus alaternus UIB population and of Rhamnusludovici-salvatoris Binifaldó population. Mean values ± standard errors (n = 4) andthe P-value from a one-way ANOVA are presented. VF, vessel frequency; Dh,hydraulic vessel diameter; PGV, percentage of grouped vessels; PLA, percentage oflumen area; (t/b)2, intervessel wall strength; RF, ray frequency; WD, wood density;VL, vessel length.

Variable Units R. alaternus R. ludovici-salvatoris P-value

VF Vessels mm−2 518.0 ± 83.0 271.7 ± 44.4 0.040Dh �m 25.3 ± 1.8 19.9 ± 1.0 0.040PGV % 51.2 ± 12.4 36.9 ± 9.2 0.081PLA % 47.8 ± 8.4 53.9 ± 8.32 0.623VL cm 13.3 ± 2.7 9.3 ± 4.5 0.170(t/b)2 �m �m−2 0.06 ± 0.003 0.04 ± 0.006 0.009

tbv

o(iDsr

4

pMec

Rceso(vRfraeerhcP(Igc2saimRtVmi

Fig. 3. Correlation of (a) maximum xylem-specific conductivity (KS-max) and xylemwater potential at 50% loss of hydraulic conductivity (P50), (b) xylem water potentialat 50% loss of hydraulic conductivity (P50) and hydraulic vessel diameter (Dh), and (c)

2

RF Rays mm−2 27.6 ± 5.7 79.1 ± 17.9 0.033WD g cm−3 0.74 ± 0.02 0.56 ± 0.04 0.005

han in R. ludovici-salvatoris. No significant differences were foundetween the species in both the average vessel length (VL) and theessel grouped lumen area (PLA) (Table 2).

No relationship was found between the WD and the abilityf vessels to resist implosion, as measured with the parametert/b)2 (R2 = 0.4; P = 0.103; Fig. 3c). The pressure at which a 50% lossn conductivity is induced (P50) was somewhat correlated with

h (R2 = 0.48; P = 0.056; Fig. 3b). However, the maximum xylem-pecific conductivity (KS-max) showed a significant and negativeelationship with P50 (R2 = 0.66; P = 0.015; Fig. 3a).

. Discussion

Differences in vulnerability to cavitation among species canlay a key role in the recruitment success of young individuals inediterranean areas (Vilagrosa et al., 2012) and can explain to some

xtent the lower recruitment capacity of some endemic species inomparison with their more widely distributed counterparts.

The data reported in this study support the hypothesis that. ludovici-salvatoris is more susceptible to water stress-inducedavitation than R. alaternus. This greater susceptibility couldxplain the observed low seedling recruitment rates of R. ludovici-alvatoris (Traveset et al., 2003) and the dieback of large branchesf many individuals during periods of drought in the summerpersonal observation). Cavitation resistance (estimated by P50)aried considerably between both species, ranging from −3.7 for. ludovici-salvatoris to −6.6 MPa for R. alaternus. These values

all within the range described for other chaparral and Mediter-anean basin species, which exhibit a mean P50 between −3.9nd −5.1 MPa, and a range from −1 MPa to −11 MPa (Jacobsent al., 2007; Nardini et al., 2014 and references therein). How-ver, in spite of P50 of R. ludovici-salvatoris being similar to thoseeported for Quercus ilex (Limousin et al., 2010), it appears to beigher than that shown for drought adapted species like Pista-ia lentiscus (Vilagrosa et al., 2003) and it is among the highest50 of the typical Mediterranean and Chaparral shrub speciesJacobsen et al., 2007; Nardini et al., 2014 and references therein).n fact, those species displayed different drought-avoidance strate-ies, by water spending in P. lentiscus (Vilagrosa, 2002), and bylosing the stomata and deep-rooting in Q. ilex (Rambal et al.,003). Rhamnus alaternus and R. ludovici-salvatoris did not demon-trate any trade-off between water transport efficiency (KS-max)nd hydraulic safety (P50) (Fig. 3a). Despite both species exhibit-ng diffuse-porous wood characteristics, and solitary vessel or short

ultiple vessels forming a dendritic pattern characteristics (Fig. 4),. alaternus constructed a safer and more efficient xylem system

han R. ludovici-salvatoris (Tables 1 and 2). The larger Dh and higherF provided this species with higher hydraulic conductance, butay have also affected the hydraulic safety owing to redundancies

n the water pathways. High VF allows plants to transport water

vessel resistance to implosion or (t/b) and wood density (WD). Rhamnus alaternus(circles) and Rhamnus ludovici-salvatoris (triangles). P-value and r2 correspond tothe plotted regression line.

despite deactivation of part of the xylem (Zimmermann, 1983). Inan anatomical study on Mediterranean shrubs and trees, De Miccoet al. (2008) observed that VF increased from mesic to xeric envi-ronments, and that R. alaternus, together with Cistus monspeliensis,had the most redundant xylem. Dmean and length of the two speciesexamined in this study were in accordance with values of otherRhamnus species found in Israel and Turkey (diameter 15–40 �m;length 130–370 mm), which are adapted to dry Mediterranean or

desert conditions (Carlquist, 1977; Fahn et al., 1986; Akkemik et al.,2008).

Previous studies have highlighted that higher investment inconduit wall per conduit volume relates to cavitation resistance

6 H. El Aou-ouad et al. / Flora 229 (2017) 1–8

F (b, d)

asteRhts2rRans2

pwpsa

esv1

ig. 4. Cross-sectional xylem images: Rhamnus alaternus (a, c), R. ludovici-salvatoris

nd higher WD. Consequently, the latter could be used as a valuablecreening tool for identifying species that may be less vulnerableo variation in water availability in a specific community (Hacket al., 2001; Jacobsen et al., 2007). This seems to be the case with. alaternus; in comparison with R. ludovici-salvatoris, R. alaternusas denser wood, a greater degree of mechanical strength at theissue level ((t/b)2), which would help xylem cells to resist thetresses created by negative hydrostatic pressures (Hacke et al.,001; Jacobsen et al., 2005), and greater cavitation resistance. Theseesults follow the same trend as that observed in a study of ninehamnaceae species in the California chaparral (Pratt et al., 2008)nd are in accordance with the reported greater ability of R. alater-us to accumulate biomass and to survive under moderate droughttress (El Aou-ouad et al., 2015; Gulías et al., 2004; Traveset et al.,003).

Finally, higher RF in R. ludovici-salvatoris could entail higher res-iration rates, affecting its competitiveness under dry conditionshen photosynthesis is limited. In fact, higher stem and whole

lant respiration rates have been observed in R. ludovici-salvatoriseedlings in comparison with R. alaternus under both well-waterednd water-limited conditions (El Aou-ouad et al., 2015).

Endemic and relict plant species are especially vulnerable to

nvironmental changes, mainly owing to their small populationizes, which are exposed to increased inbreeding, lower indi-idual fitness and lower genetic variation (Ellstrand and Elam,993; Aizen et al., 2002; Angeloni et al., 2011). Furthermore,

. Scale bars are 500 �m for top images (a, b), and 200 �m for lower images (c, d).

vulnerability to cavitation has important implications for the sur-vival of adult individuals, and this vulnerability could be even morecritical in the recruitment of young individuals (Davis et al., 1999;Pratt et al., 2008; Kursar et al., 2009). Rhamnus alaternus seedlingsshow higher growth rates and biomass accumulation in compar-ison to R. ludovici-salvatoris (El Aou-ouad et al., 2015). Moreover,R. ludovici-salvatoris seedling recruiment is known to be impairedby summer drought (Traveset et al., 2003). Lázaro-Nogal et al.(2013) have previously demonstrated that R. ludovici-salvatoris dis-plays a narrow safety margin in the face of drought. The highervulnerability to cavitation, the lower vessel area and the lowerWD and inter-vessel wall strength (t/b)2 observed in R. ludovici-salvatoris in comparison with R. alaternus are in accordance withits lower competitive ability, i.e. its smaller distribution area, andlower ability to recruit seedlings under Mediterranean conditionsthan its congeneric species R. alaternus. Consequently, the resultsof this study may partially explain the differences observed inthe ability of both species to colonize environments, in particu-lar the low rates of establishment and recovery after degradationof R. ludovici-salvatoris and the reduction in its distribution area.Moreover, taking into consideration that climate change modelspredict more intense and longer drought periods for the Mediter-

ranean Basin (IPCC, 2013), the distribution area of the endemic R.ludovici-salvatoris could decrease even more in the coming decades,compromising the survival of this species.

l. / Flo

A

M

A

i0

R

A

A

A

A

A

B

B

B

C

C

C

C

C

C

C

C

D

D

D

E

E

E

F

F

H. El Aou-ouad et a

cknowledgment

This research was supported by an AVERROES grant (ErasmusUNDUS, European Union Commission).

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.flora.2017.01.05.

eferences

izen, M.A., Ashworth, L., Galetto, L., 2002. Reproductive success in fragmentedhabitats: do compatibility systems and pollination specialization matter? J.Veg. Sci. 13, 885–892.

kkemik, Ü., D’Arrigo, R., Cherubini, P., Köse, N., Jacoby, G.C., 2008. Tree-ringreconstructions of precipitation and stream flow for northwestern Turkey. Int.J. Climatol. 28, 173–183.

lder, N.N., Pockman, W.T., Sperry, J.S., Nuismer, S., 1997. Use of centrifugal forcein the study of xylem cavitation. J. Exp. Bot. 48, 665–674.

lomar, G., Mus, M., Rosselló, J.A., 1997. Flora Endèmica De Les Balears. ConsellInsular de Mallorca, Palma de Mallorca, Spain.

ngeloni, F., Ouborg, N.J., Leimu, R., 2011. Meta-analysis on the association ofpopulation size and life history with inbreeding depression in plants. Biol.Conserv. 144, 35–43.

aas, P., Ewers, F.W., Davis, S.D., Wheeler, E.A., 2004. Evolution of xylemphysiology. In: Hemsley, A.R., Poole, I. (Eds.), The Evolution of Plant Physiology.Elsevier, London, pp. 273–295.

lackman, C.J., Brodribb, T.J., Jordan, G.J., 2010. Leaf hydraulic vulnerability isrelated to conduit dimensions and drought resistance across a diverse range ofwoody angiosperms. New Phytol. 188, 1113–1123.

rodribb, T.J., Field, T.S., 2000. Stem hydraulic supply is linked to leafphotosynthetic capacity: evidence from New Caledonian and Tasmanian rainforests. Plant Cell Environ. 23, 1381–1388.

ardona, M.A., Contandriopoulous, J., 1979. Endemism and evolution in the islandsof the western Mediterranean. In: Bramwell, D. (Ed.), Plants and Islands.Academic Press, London, pp. 113–169.

arlquist, S., 1977. Ecological factors in wood evolution: a floristic approach. Am. J.Bot. 64, 887–896.

hen, J.W., Zhang, Q., Cao, K.F., 2009. Inter-species variation of photosynthetic andxylem hydraulic traits in the deciduous and evergreen Euphorbiaceae treespecies from a seasonally tropical forest in south western China. Ecol. Res. 24,65–73.

hristman, M.A., Sperry, J.S., Adler, F.R., 2009. Testing the ‘rare pit’ hypothesis forxylem cavitation resistance in three species of Acer. New Phytol. 182, 664–674.

ochard, H., Barigah, S.T., Kleinhentz, M., Eshel, A., 2008. Is xylem cavitationresistance a relevant criterion for screening drought resistance amongstPrunus species? J. Plant Physiol. 165, 976–982.

ochard, H., Holtta, T., Herbette, S., Delzon, S., Mencuccini, M., 2009. New insightsinto the mechanisms of water-stress induced cavitation in conifers. PlantPhysiol. 151, 949–954.

ochard, H., Herbette, S., Barigah, T., Badel, E., Ennajeh, M., Vilagrosa, A., 2010. Doessample length influence the shape of xylem embolism vulnerability curves? Atest with the Cavitron spinning technique. Plant Cell Environ. 33, 1543–1552.

ochard, H., Badel, E., Herbette, S., Delzon, S., Choat, B., Jansen, S., 2013. Methodsfor measuring plant vulnerability to cavitation: a critical review. J. Exp. Bot. 64,4779–4791.

avis, S.D., Sperry, J.S., Hacke, U.G., 1999. The relationship between xylem conduitdiameter and cavitation caused by freeze-thaw events. Am. J. Bot. 86,1367–1372.

e Micco, V., Aronne, G., Baas, P., 2008. Wood anatomy and hydraulic architectureof stems and twigs of some Mediterranean trees and shrubs along amesic-xeric gradient. Trees 22, 643–655.

omec, J.C., Gartner, B.L., 2001. Cavitation and water storage capacity in bolexylem segments of mature and young Douglas-fir trees. Trees 15, 204–214.

l Aou-ouad, H., Medrano, H., Lamarti, A., Gulías, J., 2014. Seed germination atdifferent temperatures and seedling emergence at different depths of Rhamnusspp. CEJB 9, 569–578.

l Aou-ouad, H., Florez-sarasa, I., Ribas-Carbó, M., Flexas, J., Medrano, H., Gulías, J.,2015. Trade-offs between seedling growth, plant respiration and water-useefficiency in two Mediterranean shrubs Rhamnus alaternus and Rhamnusludovici-salvatoris. Photosynthetica 53, 537–546.

llstrand, N.C., Elam, D.R., 1993. Population genetic consequences of smallpopulation-size implications for plant conservation. Annu. Rev. Ecol. Syst. 24,217–242.

ahn, A., Werker, E., Baas, P., 1986. Wood Anatomy and Identification of Trees andShrubs from Israel and Adjacent Regions. The Israel Academy of Sciences andHumanities, Jerusalem, Israel.

alster, D.S., 2006. Sapling strength and safety: the importance of wood density intropical forests. New Phytol. 171, 237–239.

ra 229 (2017) 1–8 7

Ferriol, M., Llorens, L., Gil, L., Boira, H., 2009. Influence of phenological barriers andhabitat differentiation on the population genetic structure of the Balearicendemic Rhamnus ludovici-salvatoris Chodat and R. alaternus L. Plant Syst. Evol.277, 105–116.

Fichot, R., Laurans, F., Monclus, R., Moreau, A., Pilate, G., Brignolas, F., 2009. Xylemanatomy correlates with gas exchange, water-use efficiency and growthperformance under contrasting water regimes: evidence from Populusdeltoides × Populus nigra hybrids. Tree Physiol. 29, 1537–1549.

Gucci, R., Massai, R., Casano, S., Gravano, E., Lucchesini, M., 1997. The effect ofdrought on gas exchange and water potential in leaves of seven Mediterraneanwoody species. In: Mohren, G.M.J., Kramer, K., Sabaté, S. (Eds.), Impacts ofGlobal Change on Tree Physiology and Forest Ecosystems. Kluwer, Dordrecht,pp. 225–231.

Gulías, J., Flexas, J., Abadia, A., Medrano, H., 2002. Photosynthetic responses towater deficit in six Mediterranean sclerophyll species: possible factorsexplaining the declining distribution of Rhamnus ludovici-salvatoris, anendemic Balearic species. Tree Physiol. 22, 687–697.

Gulías, J., Traveset, A., Riera, N., Mus, M., 2004. Critical stages in the recruitmentprocess of Rhamnus alaternus L. Ann. Bot. 93, 723–731.

Hacke, U.G., Sperry, J.S., Pittermann, J., 2000. Drought experience and cavitationresistance in six shrubs from the Great Basin, Utah. Basic Appl. Ecol. 1, 31–41.

Hacke, U.G., Sperry, J.S., Pockman, W.T., Davis, S.D., McCulloh, K.A., 2001. Trends inwood density and structure are linked to prevention of xylem implosion bynegative pressure. Oecologia 126, 457–461.

Hultine, K.R., Koepke, D.F., Pockman, W.T., Fravolini, A., Sperry, J.S., Williams, D.G.,2006. Influence of soil texture on hydraulic properties and water relations of adominant warm-desert phreatophyte. Tree Physiol. 26, 313–323.

IPCC, 2013. Climate Change The Physical Science Basis. Cambridge UniversityPress, Cambridge.

Jacobsen, A.L., Ewers, F.W., Pratt, R.B., Paddock, W.A., Davis, S.D., 2005. Do xylemfibers affect vessel cavitation resistance? Plant Physiol. 139, 546–556.

Jacobsen, A.L., Pratt, R.B., Ewers, F.W., Davis, S.D., 2007. Cavitation resistance amongtwenty-six chaparral species of southern California. Ecol. Monogr. 77, 99–115.

Kolb, K.J., Sperry, J.S., 1999. Differences in drought adaptation between subspeciesof sagebrush (Artemisia tridentata). Ecology 80, 2373–2384.

Kursar, T.A., Engelbrecht, B.M.J., Burke, A., Tyree, M.T., El Omari, B., Giraldo, J.P.,2009. Tolerance to low leaf water status of tropical tree seedlings is related todrought performance and distribution. Funct. Ecol. 23, 93–102.

Lázaro-Nogal, A., Forner, A., Traveset, A., Valladares, F., 2013. Contrasting waterstrategies of two Mediterranean shrubs of limited distribution: uncertainfuture under a drier climate. Tree Physiol. 33, 1284–1295.

Limousin, J.M., Longepierre, D., Huc, R., Rambal, S., 2010. Change in hydraulic traitsof Mediterranean Quercus ilex subjected to long-term through fall exclusion.Tree Physiol. 30, 1026–1036.

MAGRAMA, 2013. Database of the Spanish Ministry of Agriculture, Food andEnvironment, http://sig.magrama.es/siga (Accessed July 2015).

Martínez-Ferri, E., Balaguer, L., Valladares, F., Chico, J.M., Manrique, E., 2000.Energy dissipation in drought-avoiding and drought-tolerant tree species atmidday during the Mediterranean summer. Tree Physiol. 20, 131–138.

McCarthy-Neumann, S., Kobe, R.K., 2008. Tolerance of neighborhood soilpathogens co-varies with shade tolerance across species of tropical treeseedlings. Ecology 89, 1883–1892.

Meinzer, F.C., McCulloh, K.A., Lachenbruch, B., Woodruff, D.R., Johnson, D.M., 2010.The blind men and the elephant: the impact of context and scale in evaluatingconflicts between plant hydraulic safety and efficiency. Oecologia 164,287–296.

Nardini, A., Lo Gullo, M.A., Trifilò, P., Salleo, S., 2014. The challenge of theMediterranean climate to plant hydraulics: responses and adaptations.Environ. Exp. Bot. 103, 68–79.

Pammenter, N.W., Vander-Willigen, C., 1998. A mathematical and statisticalanalysis of the curves illustrating vulnerability of xylem to cavitation. TreePhysiol. 18, 589–593.

Pratt, R.B., Ewers, F.W., Lawson, M.C., Jacobsen, A.L., Brediger, M., Davis, S.D., 2005.Mechanisms for tolerating freeze–thaw stress of two evergreen chaparralspecies: Rhus ovata and Malosma laurina (Anacardiaceae). Am. J. Bot. 92,1102–1113.

Pratt, R.B., Jacobsen, A.L., Mohla, R., Ewers, F.W., Davis, S.D., 2008. Linkage betweenwater stress tolerance and life history type in seedlings of nine chaparralspecies (Rhamnaceae). J. Ecol. 96, 1252–1265.

Rambal, S., Ourcival, J.M., Joffre, R., Mouillot, F., Nouvellon, Y., Reichstein, M.,Rocheteau, A., 2003. Drought controls over conductance and assimilation of aMediterranean evergreen ecosystem: scaling from leaf to canopy. Glob. ChangeBiol. 9, 1813–1824.

Santiago, L.S., Goldstein, G., Meinzer, F.C., Fisher, J.B., Machado, K., Woodruff, D.,Jones, T., 2004. Leaf photosynthetic traits scale with hydraulic conductivity andwood density in Panamanian forest canopy trees. Oecologia 140, 543–550.

Sperry, J.S., Tyree, M.T., 1988. Mechanism of water stress-induced xylemembolism. Plant Physiol. 88, 581–587.

Sperry, J.S., Hacke, U.G., Pittermann, J., 2006. Size and function in conifer tracheidsand angiosperm vessels. Am. J. Bot. 93, 1490–1500.

Sperry, J.S., Meinzer, F.C., McCulloh, K.A., 2008. Safety and efficiency conflicts in

hydraulic architecture: scaling from tissues to trees. Plant Cell Environ. 31,632–645.

Sterck, F.J., Poorter, L., Schieving, F., 2006. Leaf traits determine the growth survivaltrade-off across rain forest tree species. Am. Nat. 167, 758–765.

8 l. / Flo

T

T

T

T

T

V

H. El Aou-ouad et a

raveset, A., Gulías, J., Riera, N., Mus, M., 2003. Transition probabilities frompollination to establishment in a rare dioecious shrub species (Rhamnusludovici-salvatoris) in two habitats. J. Ecol. 91, 427–437.

raveset, A., Gonzalez-Varo, J.P., Valido, A., 2012. Long-term demographicconsequences of a seed dispersal disruption. Proc. R. Soc. B Biol. Sci. 279,3298–3303.

utin, T.G., Heywood, V.H., Burges, N.A., et al., 1968. Flora europaea. In: Vol. II.Rosaceae to Umbelliferae. Cambridge University Press, Cambridge.

yree, M.T., Sperry, J.S., 1989. Vulnerability of xylem to cavitation and embolism.Annu. Rev. Plant Physiol. Plant Mol. Biol. 140, 19–36.

yree, M.T., Davis, S.D., Cochard, H., 1994. Biophysical perspectives of xylem

evolution: is there a trade-off of hydraulic efficiency for vulnerability todysfunction? IAWA J. 15, 335–360.

ilagrosa, A., Bellot, J., Vallejo, V.R., Gil-Pelegrín, E., 2003. Cavitation, stomatalconductance, and leaf dieback in seedlings of two co-occurring Mediterraneanshrubs during an intense drought. J. Exp. Bot. 54, 2015–2024.

ra 229 (2017) 1–8

Vilagrosa, A., Chirino, E., Peguero-Pina, J.J., Barigah, T.S., Cochard, H., Gil-Pelegrín,E., 2012. Xylem cavitation and embolism in plants living in water-limitedecosystems. In: Aroca, R. (Ed.), Plant Responses to Drought Stress. Springer,Berlin, pp. 83–108.

Vilagrosa, A., 2002. Estrategias de Resistencia al Déficit Hídrico en Pistacia lentiscusL.y Quercus coccifera L. Implicaciones en la Repoblacion Forestal. PhD thesis.Universidad de Alicante, Alicante.

Wheeler, J.K., Sperry, J.S., Hacke, U.G., Hoang, N., 2005. Inter-vessel pitting andcavitation in woody Rosaceae and other vesselled plants: a basis for a safetyversus efficiency trade-off in xylem transport. Plant Cell Environ. 28, 800–812.

Zhang, J.L., Cao, K.F., 2009. Stem hydraulics mediates leaf water status, carbon gain,nutrient use efficiencies and plant growth rates across dipterocarp species.

Funct. Ecol. 23, 658–667.

Zimmermann, M.H., Jeje, A.A., 1981. Vessel-length distribution in stems of someAmerican woody plants. Can. J. Bot. 59, 1882–1892.

Zimmermann, M.H., 1983. Xylem Structure and the Ascent of Sap. Springer-Verlag,Berlin.