Journal of Experimental Botany, Vol. 50, No. 331, pp. 201–209, February 1999
Root system hydraulic conductivity in species with
contrasting root anatomy
Mark Rieger1 and Paula Litvin
Department of Horticulture, University of Georgia, Athens, GA 30602, USA
Received 11 June 1998; Accepted 27 August 1998
Abstract
Previous research suggested that the hydraulic properties of root systems of intact plants could be
described by two parameters: the hydraulic conductivity (Lp ) or the slope of the flow-density/water potenr
tial gradient relationship, and the offset or minimum
water potential gradient required to induce flow. In
this study, Lp and offset were correlated with anatomr
ical features of the root radial path in plants with
contrasting root anatomy. Two woody and three herbaceous species were examined which exhibit a range
of root anatomical features: Asparagus densiflorus
(Kunth) Jessop (asparagus), Dendrobium superbum
Rchb. f. (dendrobium), Glycine max (L.) Merr. (soybean),
Prunus persica (L.) Batsch. (peach), Citrus aurantium
L. (sour orange). Lp varied about 8-fold, and the offset
r
varied about 6-fold among the five species. Lp was
r
inversely related to root diameter (r2=0.39) and cortex
width (r2=0.55), suggesting that species with thinner
roots or roots with a thin cortex had the highest Lp .
r
Further observations suggested that the cortex width
was a stronger determinant of Lp than root diameter.
r
However, the offset was not correlated with root diameter, stele diameter or cortex width, but was >2-fold
higher in species having an exodermis in the root radial
path (sour orange, asparagus, and dendrobium) compared to those lacking an exodermis (peach and
soybean). The data of root Lp obtained were similar
r
to those given in the literature for both intact plants
and excised roots which have been measured with
different techniques. It is concluded that Lp and
r
offset, which describe the flow-water potential relationship for intact root systems, are related to differences in the root cortex; specifically, its thickness and
the presence/absence of a suberized exodermis.
Hence, these anatomical differences may, in part,
cause the variability in root hydraulic properties that
exists among plant species.
Key words: Endodermis, exodermis, hypodermis, cortex,
drought, water transport, root hydraulics.
Introduction
Roots impose the greatest resistance to liquid water flow
in the soil–plant–atmosphere continuum (SPAC ). Thus,
their hydraulic conductivity (per unit root surface area
conductance, Lp ) has been the subject of numerous
r
studies. From this research, paradigms have arisen concerning the radial pathway of water flow through roots
and the nature and position of the resistance elements in
roots. Three pathways are generally considered: the apoplastic pathway, the symplastic pathway and the transcellular pathway; the ‘cell-to-cell’ pathway is a combination
of the latter two (see reviews by Steudle, 1994; Steudle
and Peterson, 1998). Traditionally, the apoplastic path
was favoured, where water follows a low resistance apoplastic route around cortical cell protoplasts until it
reaches the Casparian bands of the endodermis (or in
some plants the exodermis), where it must traverse the
inner and outer tangential membranes of these cells, or
travel symplastically before reaching the xylem. This
interpretation was based partially on experiments
employing dyes which showed apoplastic barriers to solution movement in the root (Peterson et al., 1981; Hanson
et al., 1985). Dye was deposited in the intercellular spaces
and cell walls of the cortex, but blocked at the apoplast
of the endodermis, although a completely apoplastic route
to the xylem has been demonstrated along the margins
of lateral roots (Peterson et al., 1981). Other lines of
evidence include positive correlations between the number
of passage cells in the endodermis and root hydraulic
conductivity (Huang et al., 1995, cited in Peterson and
1 To whom correspondence should be addressed. Fax: +1 706 542 0624. E-mail: [email protected]
© Oxford University Press 1999
202
Rieger and Litvin
Enstone, 1996), and the observation that suberization of
the endodermis is frequently associated with a drastic
reduction in water uptake in more mature regions of the
root (Clarkson and Robards, 1975; Melchior and
Steudle, 1993).
The advent of the root pressure probe (Steudle and
Jeschke, 1983) has, in some plants, provided evidence
favouring the symplastic or transcellular pathways, or a
combination of the two (cell-to-cell ), where water enters
the epidermis and flows symplastically or across membranes of each cell toward the stele. In other plants, the
apoplastic path dominated or the relative importance of
components varied depending on the conditions (Steudle
et al., 1987). In barley, the summation of membrane
resistances of cortical cells in a radial file was found to
be similar in magnitude to the whole-root hydraulic
resistance, suggesting that the transcellular route may be
operating (Steudle and Jeschke, 1983). Puncturing the
endodermis of maize roots did not change their hydraulic
conductivity materially. This indicated that the endodermis may not be the major resistance element in roots
(Peterson et al., 1993, Steudle et al., 1993; Steudle and
Peterson, 1998). However, excised roots may be
artefactual systems ( Zimmermann et al., 1992), and since
most of the evidence against the traditional apoplastic
pathway has come from excised roots, the importance of
the symplastic or transcellular pathways remains unclear.
Recently, a composite transport model of water flow
across roots has been proposed which rectifies many of
the experimental inconsistencies associated with older
paradigms of root water uptake (Steudle, 1994; Steudle
and Peterson, 1998).
In a study of root water uptake of intact, transpiring
plants, Rieger and Motisi (1990) reported that water
uptake of whole root systems was a function of two
parameters: the hydraulic conductivity (Lp ), given by the
r
slope of the flow/water potential gradient relationship,
and the offset, given by the intercept of this relationship
to the water potential gradient axis. The offset has been
interpreted as the minimum water potential gradient
across roots required to initiate flow (Passioura and
Munns, 1984; Steudle, 1994), and can be eliminated by
severing the root system of an intact plant under water
( Rieger and Motisi, 1990). Both Lp and offset were
r
found to be lower in sour orange than in peach root
systems (Rieger and Motisi, 1990). Since sour orange
roots are greater in diameter than peach, and possess a
suberized exodermis (poorly developed or absent in
peach), this raised the possibility that the greater diameter
or the presence of an extra suberized layer in sour orange
was responsible for the lower Lp and offset. Here, the
r
hypothesis was tested that Lp and offset were related to
r
anatomical features of the root radial path using whole,
intact root systems of five species with widely different
root anatomy.
Materials and methods
Plant material
Five species were selected based on descriptions given in
Perumalla et al. (1990) and Peterson and Perumalla (1990) to
give a wide range of root anatomical features: soybean, peach,
sour orange, asparagus, and dendrobium. In particular, these
species were reported to differ in the structure of the exodermis,
ranging from none in soybean to a multiseriate exodermis with
suberized Casparian bands in asparagus. Peach and sour orange
are both reported to have a uniseriate exodermis with Casparian
bands (Perumalla et al., 1990), but sour orange roots are
generally much thicker than those of peach, and the exodermis
in sour orange is more anatomically distinct. Dendrobium has
a typical orchid root anatomy with a multiseriate epidermis
(velamen) which is water-permeable, and a uniseriate exodermis
with Casparian bands.
Soybean, peach, and sour orange were grown from seeds,
and were approximately 2, 4, and 6 months old at the time of
the experiment (respectively), and of comparable size. Asparagus
was propagated by division of older plants and grown for about
3 months to regenerate new root systems. Dendrobium orchids
were 2–3-years-old, obtained from the Georgia State Botanical
Garden. All plants except dendrobium were grown in coarse
sand media in a greenhouse for 1–3 months prior to measurement. Dendrobium was grown in coarse pine bark mulch
typical for orchid culture.
Some of the peach and soybean plants were subjected to
drought stress for about 3 weeks prior to measurement to
induce a reduction in Lp (Rieger, 1995) and attempt to
r
correlate this with drought-induced changes in anatomical
features. Periodically, drought-stressed plants were given c.
100 ml of dilute nutrient solution to prevent severe stress during
the 3-week period. Pots were sub-irrigated to keep the majority
of the roots constantly dry. Drought stress imposed in this
manner reduced plant dry mass by 35% and 54% for soybean
and peach, respectively (P<0.05), and induced some abscission
of lower leaves.
Root hydraulic conductivity and offset
The Lp and offset were estimated using whole, intact plants
r
following the procedure of Rieger and Motisi (1990). Basically,
plants were enclosed in a 42 l gas exchange system, with the
root systems and several basal leaves (or shoots) left outside
the chamber. Plants were watered to leach excess fertilizer salts
and pots were placed in basins of water to raise soil water
potentials above −10 kPa. Transpiration of basal leaves outside
the chamber was prevented by wrapping them in parafilm and
foil, and they were used to estimate the xylem pressure potential
at the stem base (Hellkvist et al., 1974). Various rates of
transpiration were induced by changing chamber conditions.
Steady-state water flux was verified by sequential measurements
of wrapped leaf water potential (pressure chamber) and whole
plant transpiration rate.
One or two replicates of each species were initially subjected
to three or more different conditions to verify linearity of the
water uptake/water potential relationship. Subsequent replicates
were subjected to only two conditions yielding a maximum and
minimum rate of transpiration. Minimum rates of transpiration
were obtained with high humidity and low light (<10
mmol m−2 s−1) conditions which induced a nearly complete
stomatal closure. Maximum rates were obtained under low
humidity and saturating light (>1500 mmol m−2 s−1); these
rates differed from 2–20-fold. The steady-state values of whole
plant transpiration ( y-axis) were plotted against the water
Root hydraulic conductivity versus anatomy 203
potential gradient across the root system (x-axis). The latter is
given by the magnitude of the wrapped leaf water potential
assuming that the water potential at the root surface was zero,
and the contribution of osmotic potential to the water potential
gradient was insignificant. The slope of this relationship yields
Lp , and the x-intercept the offset. It was assumed that Lp
r
r
reflected the conductivity of the root radial path since the axial
resistance to water flow in roots is one to several orders of
magnitude lower than the radial resistance (Steudle and
Peterson, 1998). However, cavitation of root xylem elements in
stressed plants could have reduced axial conductance, and
hence Lp .
r
Root systems were removed and total length was measured
using the line intersect method (Tennant, 1975) directly after
measurements were conducted. Root hydraulic conductivity
was expressed on a root surface area basis (Lp , m s−1 MPa−1).
r
Surface areas were calculated from diameter and length
measurements assuming roots were cylindrical. Root hydraulic
conductivity on a root length basis (Lp , m2 s−1 MPa−1) has
l
been included in Table 2 for comparative purposes with other
literature. Root and shoot dry mass were recorded after drying
to constant mass at 70 °C.
Root anatomy
Two to four root cross-sections were prepared from root
systems of each plant in the study. Root samples were preserved
in FAA (formaldehyde, acetic acid, and ethanol, 1055585, by
vol.) and dehydrated in ethanol prior to embedding in resin
(Historesin). Cross-sections 5 mm thick and 10–30 mm behind
the root tip were used for measurements of root diameter,
endodermis diameter, and exodermis diameter. This section of
the root was sampled since it represented the characteristic
anatomical features of each species, and water uptake is known
to be maximal in this area (Clarkson and Robards, 1975). The
width of the cortex was estimated as the difference between
radii of the root epidermis (innermost layer in the case of
dendrobium and asparagus) and the endodermis. Roots were
stained with either aniline blue or safranin, and photographed
with a calibrated length reference using an Olympus Vanox
AH-3 microscope at 40–100×. Diameters were measured with
a ruler from photographs, and units converted using the
calibrated reference. Comparable cross-sections were stained
with the suberin-specific dye Sudan III or Sudan Black ( Vaughn,
1987) to detect the presence of suberin in the endodermis and
exodermis. Perumalla et al. (1990) reported that the presence
of suberin was indicative of water-impermeable Casparian
bands in over 150 species. Thus, in the present study, Sudan
staining in the exodermis or endodermis was taken as an
indication of an apoplastic barrier to water flow.
Fig. 1. The relationship was developed using four different
chamber conditions, and shows high linearity and a
significant offset. Relationships for other species were
similar, except for dendrobium which had lower r2 values
(as low as 0.72) probably due to reduced accuracy in
measuring whole plant transpiration, which was very low
for this species (maximal rates of about 0.53E–10 m2 s−1).
Means of root hydraulic conductivity varied 8.4-fold
among species ( Table 1). Root hydraulic conductivity
was lowest in asparagus, intermediate for peach and
dendrobium, with sour orange and soybean among the
highest.
In peach and soybean, drought stress reduced mean
Lp about 4.5-fold ( Table 1). The magnitude of reduction
r
in Lp is consistent with previous findings for droughtr
stressed plants which have been obtained for intact plants
and excised roots (Nobel and Huang, 1992; North and
Nobel, 1992; Rieger, 1995).
Offsets varied several-fold among species, and drought
stress increased the offset in peach, but not in soybean
( Table 1). Offsets were lowest in soybean and highest in
sour orange.
Values of Lp derived from intact plants in this study
r
compare favourably with values reported for a variety of
plant species and measurement techniques (excised roots
and intact plants; Table 2). Values for peach and sour
orange were similar to those previously reported (Rieger
and Duemmel, 1992; Rieger, 1992, 1995). Peach, however,
had an Lp value which was lower by a factor of 3–4
r
than that reported earlier by Rieger and Motisi (1990).
Values for soybean were somewhat lower than those given
by Fiscus (1977). These values for sour orange were
somewhat higher than those reported by others (Graham
Statistics
There were six to eight replicates for each species, and one
Lp /offset estimate and 2–4 root cross-sections per replicate.
r
Data were analysed by analysis of variance for differences
among species, and by correlation or regression for associations
between Lp and offset and root anatomical features (Proc
r
CORR and GLM, SAS, SAS Institute, Cary NC ).
Results
Root hydraulic conductivity and offset
An example of water uptake/water potential gradient
relationship for intact soybean plants is presented in
Fig. 1. A typical water uptake versus water potential gradient curve
developed on intact soybean plants. Different rates of water uptake
were induced by exposing the plant canopy to various light intensity
and humidity conditions.
204
Rieger and Litvin
Table 1. Root anatomical characteristics and root system hydraulic conductivity for the five species chosen for study
The drought-stressed peach and soybean plants were the same age and cultivar as the well-watered group, but were exposed to moderate water
deficits for about 3 weeks prior to measurement.
Species
Exodermis
Asparagus
Yes
Dendrobium
Yes
Peach
non-stressed
Peach
drought-stressed
Sour orange
No
Soybean
non-stressed
Soybean
drought-stressed
No
Yes
No
No
Root
diameter
(mm)
Stele
diameter
(mm)
Cortex
width
(mm)
Specific root
length
(m g−1)
Lp ×108
r
(m s−1 MPa−1)
Offset
(MPa)
1500 aa
(700)
1500 a
(200)
640 b
(150)
580 b
(200)
810 b
(240)
570 b
(210)
670 b
(140)
440 a
(220)
330 ab
(60)
160 d
(40)
160 d
(40)
240 bcd
(90)
200 cd
(80)
290 bc
(100)
460 a
(220)
280 b
(40)
240 b
(60)
210 b
(80)
250 b
(70)
180 b
(80)
190 b
(90)
16.3 c
(2.5)
4.7 d
(0.5)
45.1 b
(18.1)
21.9 c
(5.0)
5.8 d
(0.9)
45.9 b
(4.6)
63.1 a
(4.7)
0.9 d
(0.9)
4.2 b
(1.6)
3.6 bc
(1.8)
0.8 d
(0.4)
7.9 a
(2.4)
7.6 a
(3.8)
1.7 cd
(1.1)
0.20 b
(0.07)
0.12 ab
(0.02)
0.18 b
(0.06)
0.49 c
(0.14)
0.47 c
(0.02)
0.08 a
(0.05)
0.08 a
(0.09)
aMeans are of 6–8 replicates per species. Means followed by the same lower case letters are not significantly different, Duncan’s Multiple Range
Test, P=0.05.
et al., 1987; Zekri and Parsons, 1989). Overall, agreement
is reasonably good considering that plant age, environmental conditions, plant species (e.g. woody versus herbaceous), and measurement technique all influence the
value of Lp (Steudle and Heydt, 1997). Within the
r
scatter, there were no differences when comparing data
obtained from excised roots with those of roots of
intact plants.
Root anatomy
Among the species chosen, anatomical features of roots
were highly variable and generally consistent with descriptions given by Perumalla et al. (1990) and Peterson and
Perumalla (1990) (Fig. 2). Within species, root diameters
varied little: only a small fraction of the root system (not
more than 5%) had secondary growth, and root browning
was not observed. Root and stele diameters, and the width
of the cortex varied about 2–3-fold among species, but were
not significantly altered by drought stress in peach or
soybean (Table 1). Specific root length (SRL, m of root
length g−1 dry mass) varied more than 10-fold among
species and was inconsistently affected by drought stress:
SRL decreased in peach but increased in soybean. Since
root and stele diameters were similar between droughtstressed and non-stressed plants, changes in SRL were not
due to root thickening or epidermal/cortex sloughing.
Sudan staining, indicative of suberin, was observed in
the endodermis of all species. According to Sudan staining
a distinct exodermis was observed in asparagus, dendrobium, and sour orange, but not in peach and soybean
roots. This agrees with results obtained using more sophisticated microscopic tests (Perumalla et al., 1990) with
two exceptions: (1) a suberized exodermis was not
observed in peach, and (2) asparagus roots had only a
uniseriate, not a multiseriate exodermis. Reasons for these
discrepancies are unclear, but could be due to a number
of influences including differences in root age, sampling
distance from root tip, growing conditions, or genotype.
Lp , offset—anatomy correlations
r
Since drought affected Lp and offset without affecting
r
measured root anatomical features, data for droughtstressed plants were excluded from correlation analyses
to obtain a clearer picture of the relationships between
root anatomy and Lp and offset.
r
Lp was negatively correlated with root diameter, indicatr
ing that thinner roots had greater hydraulic conductivity
(Fig. 3). However, Lp was not significantly correlated with
r
SRL which is sometimes used as an index of root thickness.
This suggested that the radial path length, not the mass per
unit of root length governed Lp more strongly.
r
The width of the cortex was negatively correlated with
Lp ( Fig. 4). Regression of Lp on cortex width explained
r
r
more of the variation in Lp than any other single
r
parameter in this study (55%).
Lp was negatively correlated with specific endodermal
r
surface area, calculated as the m2 of tangential endodermal cell area m−1 of root (data not shown). This
suggested that roots presenting more suberized endodermal wall area per unit of root length had lower
conductivities, although the strength of the relationship
was quite low (r2=0.23). This would likely occur in roots
having the endodermis positioned closer to the epidermis
than to the root centre.
The offset was negatively correlated with SRL (r=
−0.51), suggesting that roots less massive per unit length
had a lower minimum water potential gradient required
to initiate flow. Since the offset was not correlated with
Root hydraulic conductivity versus anatomy 205
Table 2. Comparison of root hydraulic conductivity values derived from this study and the literature
Root hydraulic conductivities are presented on a root length basis (Lp ), and a root surface area basis (Lp ).
l
r
Species
Technique
Soybean (Glycine max),
peach (Prunus persica), sour
orange (Citrus aurantium),
asparagus (Asparagus
densiflorus), dendrobium
(Dendrobium superbum)
Woody plants
Citrus rootstocks
(Citrus)
Douglas Fir (Pseudotsuga
menziesii)
Loblolly pine (Pinus taeda)
Root systems of intact plants
Norway spruce (Picea abies)
Oak (Quercus)
Peach, other Prunus species
Peach, sour orange
Peach, sour orange, citrange
(Poncirus trifoliata×Citrus
sinensis)
Peach, olive (Olea
europaea), citrumelo
(Poncirus trifoliata×Citrus
paradisi), Pistacia integerrima
Sour orange
Herbaceous plants
Barley (Hordeum vulgare)
Bean (Phaseolus vulgaris)
Bean
Broad bean (Vicia faba),
bean, maize (Zea mays),
sunflower (Helianthus
annuus), tomato
(Lycopersicon esculentum)
Cotton (Gossypium hirsutum)
Cotton
Cotton
Ferocactus acanthoides,
Opuntia ficus-indica
Maize
Maize
Maize
Sorghum (Sorghum bicolor)
Soybean
Sunflower, barley, maize
Wheat (Triticum aestivum),
maize
Wheat
Lp ×108
r
m s−1 MPa−1
0.9–7.9
Lp ×1011
l
m2 s−1 MPa−1
Reference
2.6–19.0
This study
Pressure chamber, excised
root systems
Pressure chamber, excised
root systems
Pressure chamber, excised
root systems
Root pressure probe,
hydrostatic force, excised
root systems
Root pressure probe,
hydrostatic force, excised
root systems
Root systems of intact plants
Root systems of intact plants
Root systems of intact plants
—
1.0–3.1
Zekri and Parsons (1989)
—
2.0–3.0
Coleman et al. (1990)
—
—
—
6.6–10.7
7.5–26.7
5.8–12.0
Rieger and Duemmel (1992)
Rieger and Motisi (1990)
Rieger (1992)
Root systems of intact plants
—
0.9–6.8
Rieger (1995)
Pressure chamber, excised
root systems
—
0.2–0.7
Graham et al. (1987)
Root pressure probe,
hydrostatic force, excised
individual roots
Pressure chamber, excised
root systems
Pressure chamber, excised
root systems
Osmotically induced water
exudation from whole root
systems
Root systems of intact plants
Pressure chamber, excised
root systems
Root systems of intact plants
Suction applied to excised,
individual roots
Microcapillary apparatus,
osmotic force imposed
across cortical sleeves of
excised, individual roots
Root pressure probe,
hydrostatic force, excised
individual roots
Pressure or suction applied
to excised, individual roots
Suction applied to excised,
individual roots
Pressure chamber, excised
root systems
Root systems of intact plants
Osmotically induced back
flow, excised individual roots
Osmotically induced back
flow, excised root
systems
14.0
—
Sands et al. (1982)
4.9–7.8
—
Rudinger et al. (1994)
0.5–3.1
—
Steudle and
Meshchereryakov (1996)
0.4–1.3
—
Steudle and Jeschke (1983)
1.0–10.2
—
O’Leary (1971)
—
Fiscus and Markhart (1979)
—
Newman (1973)
10–23
—
—
4.3–55.5
Radin and Eidenbock (1984)
Brar et al. (1990)
—
10–50
4–8
—
Radin (1990)
Lopez and Nobel (1991)
10.5–21.8
—
Schambil and
Woermann (1989)
10–20
—
Miller (1985)
10.3
—
Shone and Clarkson (1988)
6–22
—
Cruz et al. (1992)
22–30
—
Fiscus (1977)
0–16
1–6
—
—
Aston and Lawlor (1979)
Jones et al. (1988)
0.5
—
Jones et al. (1983)
10–60
0.5–6.1
206
Rieger and Litvin
Fig. 2. Root cross-sections 10–30 mm behind the root tip for (A) peach, (B) asparagus, (C ) soybean, (D) sour orange, (E ) dendrobium.
Abbreviations: ex=exodermis, ep=epidermis, en=endodermis, c=cortex. Bar in lower right hand corner of each photograph equals 100 mm. Peach
(A) has a collapsed epidermis, no exodermis, an endodermis with suberized interior tangential walls and Casparian bands in the radial wall.
Asparagus (B) shows a multiseriate epidermis and an exodermis with suberized radial and outer tangential walls; the endodermis shows passage
cells opposite xylem vessels, with suberized radial and tangential walls of single endodermal cells interspersed. Soybean (C ) has a collapsed
epidermis, no exodermis, and an endodermis with weak suberin staining in the interior tangential wall. Sour orange (D) has an intact epidermis,
an exodermis with suberized outer tangential walls, and an endodermis with large passage cell areas at the protoxylem poles and suberized interior
tangential walls of cells between protoxylem poles. Dendrobium (E ) shows a multiseriate epidermis or velamen, an exodermis with suberized radial
and tangential walls with occasional passage cells, and an endodermis also with groups of cells having suberized radial and tangential walls
interspersed with passage cells opposite large xylem elements.
root, stele or exodermal diameters, or cortex width, this
was not due to root thickness or the position of suberized
layers in roots, but the dry matter content of the tissue.
However, when species were pooled into two groups—
those with exodermis and endodermis versus those with
endodermis only—the offset was 0.28 MPa for species
with two suberized layers versus 0.12 MPa for those with
only one suberized layer, significantly different at P<0.05.
Discussion
In this study, a similar approach has been taken to that
of Newman (1973), who reported that Lp differed among
r
five herbaceous species by an order of magnitude, yet was
not correlated with root anatomical features. He concluded that the lack of correlation of Lp with root
r
diameter suggested that most of the resistance to water
flow lay within the endodermis. In this study, however,
significant correlations were found between Lp and root
r
diameter ( Fig. 3) and Lp and cortex width ( Fig. 4),
r
unlike Newman’s study. This suggests that the main
resistance does not reside in only one cell layer, because
the longer the radial path to the xylem, the lower the Lp .
r
Since cell walls can impose an appreciable resistance to
water flow (Peterson and Steudle, 1993), either symplastic, transcellular or apoplastic flow across the cortex
would be consistent with the inverse relationship between
Lp and root diameter or cortex width. Thus, relationships
r
in Figs 3 and 4 do not necessarily rule out an apoplastic
pathway across the root, but suggest that the hydraulic
resistance of the cortex is significant compared to that of
suberized layer(s).
Root hydraulic conductivity versus anatomy 207
Fig. 3. Root system hydraulic conductivity (Lp ) as a function of root
r
diameter in five species. Data for drought-stressed plants was excluded
(n=36).
Fig. 4. Root system hydraulic conductivity (Lp ) as a function of width
r
of the root cortex in five species. Data for drought-stressed plants was
excluded (n=36).
One possible confounding effect in the root diameter
correlation with Lp is the positive correlation between
r
the degree of exodermal suberization and root diameter
as reported for Citrus ( Eissenstat, 1992). If the degree of
suberization of the exodermis (or endodermis) increases
with root diameter for plants in general, then correlation
between root diameter and Lp found here could be a
r
casual association. It should be noted that the staining
technique used only allowed an assessment of the presence
or absence of suberin. It did not allow the extent of
suberization to be judged.
Other observations from this study are consistent with
the idea of appreciable resistance to water flow through
the cortex. For example, soybean (non-stressed) and sour
orange had similar Lp and root diameter, yet soybean
r
lacked an exodermis whereas sour orange had both a
suberized exodermis and endodermis. If suberized cell
layers dominated Lp , one would expect sour orange with
r
two suberized layers to have lower Lp than soybean with
r
only one suberized layer. Furthermore, when the five
species were pooled into two classes—those with exodermis and endodermis versus those with endodermis
only—Lp values were 4.5×10−8 and 5.7×10−8
r
m s−1 MPa−1, respectively, which were not statistically
different (P>0.05).
Further support for the importance of the cortex in
Lp is found when comparing peach and dendrobium.
r
Due to the very large stele and presence of the velamen
in dendrobium, these species had different root diameters,
yet similar cortex widths. They also had similar Lp ,
r
supporting a strong role for the cortex in determining
Lp . Also, peach had only a suberized endodermis,
r
whereas dendrobium had both a suberized endodermis
and exodermis, so one would expect peach to have higher
Lp if suberized cell layers largely determined resistance
r
to water movement. This argues against the traditional
view of an apoplastic pathway where the resistance in the
cortex is thought to be much lower than that of the
endodermis (or exodermis).
A similar observation is made when comparing asparagus and dendrobium. These species had similar root
diameter, and both had two apoplastic barriers to water
flow, yet different cortex width. Despite similar root
diameter and an equal number of suberized layers, the
species with the thinner cortex (dendrobium) had the
higher Lp .
r
Drought stress reduced Lp by about 80% in peach and
r
soybean ( Table 1). This was comparable to differences
between the highest and lowest Lp values among nonr
stressed plants. Thus, drought, and perhaps other environmental stresses (O’Leary, 1971; Zekri and Parsons, 1989)
can cause differences in Lp of similar magnitude to
r
natural differences that occur among species having
widely different root radial anatomy. The fact that
drought reduced Lp without inducing gross anatomical
r
changes in roots indicates that factors not readily
observed in root cross-sectional features are also important in determining Lp . Roots of drought-stressed plants
r
are often suberized closer to tips (Drew, 1987), and/or
the extent of suberization is increased by drought (Nobel
and Huang, 1992; North and Nobel, 1992), which could
explain the drop in Lp due to drought in this study.
r
However, air lacunae formation in the cortex also occurs
following drought, which may contribute to reduced
radial conductivity (North and Nobel, 1992; Nobel and
Huang, 1992). Although not quantified, air lacunae were
observed more often in drought-stressed than non-stressed
roots in our study, and thus the drought-induced drop in
Lp may have been partially attributed to disruption of
r
the cortex. Alternatively, aquaporins (membrane water
channels) may be important in determining root hydraulic
conductivity (Steudle, 1997), and it is possible that
208
Rieger and Litvin
drought stress negatively influenced the development and
condition of aquaporins in our plants, causing the reduction in Lp .
r
Data from this study support previous reports (Rieger
and Motisi, 1990; Rieger, 1992) of a greater offset in
plants with two suberized layers compared to plants with
only one such layer. Furthermore, the association between
the offset, but not Lp , and number of suberized layers
r
in roots suggests that apoplastic flow barriers in roots are
more important in determining the standing gradient of
water potential at zero flow (offset), rather than the
steepness of the flux/gradient relationship (conductivity).
The offset was also correlated with SRL, but not root
diameter, suggesting that more dense tissue resists water
movement under small water potential gradients to a
greater extent than less dense tissue.
Drought induced an increase in the offset for peach,
but not soybean ( Table 1). Given the negative correlation
between SRL and the offset for non-stressed plants, this
could be related to an associated decline in SRL for
peach due to drought. However, a decrease in the offset
for soybean would have been expected if SRL strongly
governed the offset. Since soybean offsets were low in the
non-stressed state, there was little potential for a decrease
in offset in response to drought, as zero is the lowest
value possible for the offset.
These data, derived from intact, anatomically diverse
root systems, suggest that the resistance through the
cortex is of equal or greater importance in determining
Lp than the resistance through suberized layers. To some
r
extent, this is consistent with experimental results derived
from excised individual roots (Nobel and Huang, 1992;
Steudle and Jeschke, 1983) which favour a strong role
for the cortex in determining root hydraulic conductivity,
but inconsistent with more recent results from maize roots
where the development of the exodermis reduced Lp
r
( Zimmermann and Steudle, 1998). Although the physiological meaning of the offset is not completely clear, the
present study supports previous findings of the offset
being positively related to the number of suberized cell
layers in the root radial path. The offset was originally
proposed to result from an interaction between osmotic
and hydrostatic forces acting on water at low flow rates
(Dalton et al., 1975; Fiscus, 1975), giving rise to nonlinear force/flux curves. Extrapolated values of driving
force at zero flow (offsets) were thought to be equivalent
to the osmotic potential of the rooting medium. However,
as discussed by Passioura (1988), this explanation does
not hold for our data nor for numerous similar results
reported by others. The offset was not correlated with,
and was therefore independent of, Lp , and was affected
r
by the presence/absence of an exodermis, unlike Lp . It
r
is concluded that the offset is a real component of root
system hydraulics, not a measurement or calculation
artefact, and is related to apoplastic barriers in the root.
Acknowledgements
The authors thank Dr Ned Friedman, Dr William L Arthur,
Dr Hazel Y Wetzstein, and Dr Bingru Huang for the use of
their equipment and expertise for the root microscopy.
References
Aston MJ, Lawlor DW. 1979. The relationship between
transpiration, root water uptake, and leaf water potential.
Journal of Experimental Botany 30, 169–181.
Brar GS, McMichael BL, Taylor HM. 1990. Hydraulic conductivity of cotton roots as influenced by plant age and rooting
medium. Agronomy Journal 82, 264–266.
Clarkson DT, Robards AW. 1975. The endodermis, its structural
development and physiological role. In: Torrey JG, Clarkson
DT, eds. The development and function of roots, New York:
Academic Press, 415–436.
Coleman MD, Bledsoe CS, Smit BA. 1990. Root hydraulic
conductivity and xylem sap levels of zeatin riboside and
abscisic acid in ectomycorrhizal Douglas fir seedlings. New
Phytologist 115, 275–284.
Cruz RT, Jordan WR, Drew MC. 1992. Structural changes and
associated reduction of hydraulic conductance in roots of
Sorghum bicolor L. following exposure to water deficit. Plant
Physiology 99, 203–212.
Dalton FN, Raats PAC, Gardner WR. 1975. Simultaneous
uptake of water and solutes by plant roots. Agronomy Journal
67, 334–339.
Drew MC. 1987. Function of root tissues in nutrient and water
transport, In: Gregory PJ, Lake JV, Rose DA, eds. Root
development and function. Cambridge University Press,
71–101.
Eissenstat DM. 1992. Costs and benefits of constructing roots
of small diameter. Journal of Plant Nutrition 15, 763–782.
Fiscus EL. 1975. The interaction between osmotic- and pressureinduced water flow in plant roots. Plant Physiology 55,
917–922.
Fiscus EL. 1977. Determination of hydraulic and osmotic
properties of soybean root systems. Plant Physiology 59,
1013–1020.
Fiscus EL, Markhart AH. 1979. Relationship between root
system water transport properties and plant size in Phaseolus.
Plant Physiology 64, 770–773.
Graham JH, Syvertsen JP, Smith ML. 1987. Water relations of
mycorrhizal and phosphorus-fertilized non-mycorrhizal Citrus
under drought stress. New Phytologist 105, 411–419.
Hanson PJ, Sucoff EI, Markhart III AH. 1985. Quantifying
apoplastic flux through red pine root systems using trisodium,
3-hydroxy-5,8,10-pyrenetrisulfonate. Plant Physiology 77,
21–24.
Hellkvist J, Richards GP, Jarvis PG. 1974. Vertical gradients
of water potential and tissue water relations in Sitka spruce
trees measured with a pressure chamber.Journal of Applied
Ecology 11, 637–668.
Jones H, Leigh RA, Wyn Jones RG, Tomos AD. 1988. The
integration of whole-root and cellular hydraulic conductivities
in cereal roots. Planta 174, 1–7.
Jones H, Tomos AD, Leigh RA, Wyn Jones RG. 1983. Waterrelation parameters of epidermal and cortical cells in the
primary root of Triticum aestivum L. Planta 158, 230–236.
Lopez FB, Nobel PS. 1991. Root hydraulic conductivity of two
cactus species in relation to root age, temperature, and soil
water status. Journal of Experimental Botany 42, 143–149.
Melchior W, Steudle E. 1993. Water transport in onion Allium
Root hydraulic conductivity versus anatomy 209
cepa L. roots: changes of axial and radial hydraulic
conductivities during development. Plant Physiology 101,
1305–1315.
Miller DM. 1985. Studies of root function in Zea mays. III.
Xylem sap composition at maximum root pressure provides
evidence of active transport into the xylem and a measurement
of the reflection coefficient of the root. Plant Physiology
77, 162–167.
Newman EI. 1973. Permeability to water of the roots of five
herbaceous species. New Phytologist 72, 547–555.
Nobel PS, Huang B. 1992. Hydraulic and structural changes
for lateral roots of two desert succulents in response to soil
drying and rewetting. International Journal of Plant Science
153, S163–S170.
North GB, Nobel PS. 1992. Drought-induced changes in
hydraulic conductivity and structure in roots of Ferocactus
acanthoides and Opuntia ficus-indica. New Phytologist 120,
9–19.
O’Leary JW. 1971. Salinity induced changes in hydraulic
conductivity of roots. Proceedings of a Symposium on the
Structure and Function of Primary Root Tissues, Tatranska
Lomnica, Czechoslovakia.
Passioura JB. 1988. Water transport in and to roots. Annual
Review of Plant Physiology and Plant Molecular Biology
39, 245–265.
Passioura JB, Munns R. 1984. Hydraulic resistance in plants.
II. Effects of rooting medium and time of day in barley and
lupin. Australian Journal of Plant Physiology 11, 341–350.
Perumalla CJ, Peterson CA, Enstone DE. 1990. A survey of
angiosperm species to detect hypodermal casparian bands. I.
Roots with a uniseriate hypodermis and epidermis. Botanical
Journal of the Linnean Society 103, 93–112.
Peterson CA, Emanuel ME, Humphreys GB. 1981. Pathway of
movement of apoplastic fluorescent dye tracers through the
endodermis at the site of secondary root formation in corn
(Zea mays) and broad bean (Vicia faba). Canadian Journal of
Botany 59, 618–625.
Peterson CA, Enstone DE. 1996. Functions of passage cells in
the endodermis and exodermis of roots. Physiologia Plantarum
97, 592–598.
Peterson CA, Murrmann M, Steudle E. 1993. Location of the
major barriers to water and ion movement in young roots of
Zea mays L. Planta 190, 127–136.
Peterson CA, Perumalla CJ. 1990. A survey of angiosperm
species to detect hypodermal casparian bands. II. Roots with
a multiseriate hypodermis and epidermis. Botanical Journal
of the Linnean Society 103, 113–125.
Peterson CA, Steudle E. 1993. Lateral hydraulic conductivity of
early metaxylem vessels in Zea mays L. roots. Planta
189, 288–297.
Radin JW. 1990. Responses of transpiration and hydraulic
conductance to root temperature in nitrogen- and phosphorus-deficient cotton seedlings. Plant Physiology 92,
855–857.
Radin JW, Eidenbock MP. 1984. Hydraulic conductance as a
factor limiting leaf expansion of phosphorus-deficient cotton
plants. Plant Physiology 373–377.
Rieger M. 1992. Growth, gas exchange, water uptake, and
drought response of seedling- and cutting-propagated peach
and citrus rootstocks. Journal of the American Society for
Horticultural Science 117, 834–840.
Rieger M. 1995. Offsetting effects of reduced root hydraulic
conductivity and osmotic adjustment following drought. Tree
Physiology 15, 379–385.
Rieger M, Duemmel MJ. 1992. Comparison of drought
resistance among Prunus species from divergent habitats.
Tree Physiology 11, 369–380.
Rieger M, Motisi A. 1990. Estimation of root hydraulic
conductivity on intact peach and citrus rootstocks.
HortScience 25, 1631–1634.
Rüdinger M, Hallgren SW, Steudle E, Schulze E-D. 1994.
Hydraulic and osmotic properties of spruce roots. Journal of
Experimental Botany 45, 1413–1425.
Sands R, Fiscus EL, Reid CPP. 1982. Hydraulic properties of
pine and bean roots with varying degrees of suberization,
vascular differentiation and mycorrhizal infection. Australian
Journal of Plant Physiology 9, 559–569.
Schambil F, Woermann D. 1989. Radial transport of water
across cortical sleeves of excised roots of Zea mays L. Planta
178, 488–494.
Shone MGT, Clarkson DT. 1988. Rectification of radial water
flow in the hypodermis of nodal roots of Zea mays. Plant
and Soil 111, 223–229.
Steudle E. 1994. Water transport across roots. Plant and Soil
167, 79–90.
Steudle E. 1997. Water transport across plant tissue: role of
water channels. Biology of the Cell 89, 259–273.
Steudle E, Heydt H. 1997. Water transport across tree roots,
In: Rennenberg H, Eschrich W, Ziegler H, eds. Trees—
contributions to modern tree physiology. Leiden, The
Netherlands: Backhuys Publishers, 239–255.
Steudle E, Jeschke WD. 1983. Water transport in barley roots.
Planta 158, 237–248.
Steudle E, Meshcheryakov AB. 1996. Hydraulic and osmotic
properties of oak roots. Journal of Experimental Botany
47, 387–401.
Steudle E, Murrmann M, Peterson CA. 1993. Transport of
water and solutes across maize roots modified by puncturing
the endodermis. Plant Physiology 103, 335–349.
Steudle E, R Oren, Schulze ED. 1987. Water transport in maize
roots. Plant Physiology 84, 1220–1232.
Steudle E, Peterson CA. 1998. How does water get across roots?
Journal of Experimental Botany 49, 775–788.
Tennant D. 1975. A test of a modified line intersect method of
estimating root length. Journal of Ecology 63, 995–1001.
Vaughn KC (ed). 1987. CRC handbook of plant cytochemistry.
II. Other cytochemical staining procedures. Boca Raton,
Florida: CRC Press, 8.
Zekri M, Parsons LR. 1989. Growth and root hydraulic
conductivity of several citrus rootstocks under salt and
polyethylene glycol stresses. Physiologia Plantarum 77,
99–106.
Zimmermann HM, Steudle E. 1998. Apoplastic transport across
young maize roots: the effect of the exodermis. Planta 203,
(in press).
Zimmermann U, Rygol J, Balling A, Klock G, Metzler A, Haase
A. 1992. Radial turgor and osmotic pressure profiles in intact
and excised roots of Aster tripolium. Plant Physiology
99, 186–196.