Plant Physiol. (1993) 101: 1305-1315
Water Transport in Onion (Allium cepa 1.) Roots'
Changes of Axial and Radial Hydraulic Conductivities during Root Development
Walter Melchior and Ernst Steudle*
Lehrstuhl für Pflanzenokologie, Universitat Bayreuth, Postfach 1O1 251, D-8580 Bayreuth, Germany
1982; Passioura, 1988; Steudle, 1992). Variable hydraulic root
resistance is closely associated with changes of root structure
and the pathways used for water movement in the root, both
for axial transport in vessels and for the radial movement
across the root cylinder.
During root development, L, increases due to the breakdown of cross walls between vessel members. The main radial
hydraulic resistance of the root often has been attributed to
the endodermis or exodermis, where hydrophobic deposits
block the transport of water and ions in the cell walls (Passioura, 1988; Peterson, 1989). Therefore, a dependency on
the state of development of endodermis or exodermis could
be expected, and water uptake should occur mainly in the
region near the root tip (Boyer, 1985). However, a substantial
water uptake in zones with suberized endodermis (state 11)
was reported for barley roots (Sanderson, 1983), which coqtradicted the experiments of Robards et al. (1973). In the
former study, water flow declined as the number of endodermal cells in state 111, i.e. cells exhibiting thick layers of
cellulose, increased.
Recently, Peterson et al. (1993) demonstrated that the state
I endodermis (with a Casparian band) was hardly a main
barrier to radial water flow in maize roots. The authors
dissected parts of the root cortex or punctured the endodermis
with narrow tips of microcapillaries to characterize different
resistances on the radial path of water. It was found that,
different from solute permeability, the water permeability
(hydraulic conductivity) was evenly distributed within the
living tissue.
The recently developed root pressure probe has been employed to measure both Lp, and L, of maize roots grown
hydroponically (Frensch and Steudle, 1989). It allows the
resolution of both parameters with higher precision along the
root than other techniques available. Frensch and Steudle
(1989) showed that the contribution of axial hydraulic resistance to the overall resistance was not important except for
the apical 15 mm where early metaxylem was not mature. In
more basal parts, Lp, was constant at distances between 20
The hydraulic architecture of developing onion (Allium cepa L.
cv Calypso) roots grown hydroponically was determined by measuring axial and radial hydraulic conductivities (equal to inverse of
specific hydraulic resistances). I n the roots, Casparian bands and
suberin lamellae develop in the endodermis and exodermis (equal
to hypodermis). Using the root pressure probe, changes of hydraulic conductivities along the developing roots were analyzed
with high resolution. Axial hydraulic conductivity (L,) was also
calculated from stained cross-sections according to Poiseuille's law.
Near the base and the tip of the roots, measured and calculated 1,
values were similar. However, at distances between 200 and 300
mm from the apex, measured values of L, were smaller by more
than 1 order of magnitude than those calculated, probably because
of remaining cross walls between xylem vessel members. During
development of root xylem, L. increased by 3 orders of magnitude.
I n the apical 30 mm (tip region), axial resistance limited water
transport, whereas in basal parts radial resistances (low radial
hydraulic conductivity, Lp,) controlled the uptake. Because of the
high axial hydraulic resistance in the tip region, this zone appeared
to be "hydraulically isolated from the rest of the root. Changes of
the Lp. of the roots were determined by measuring the hydraulic
conductance of roots of different length and referring these data
to unit surface area. At distances between 30 and 150 mm from
the root tip, Lp, was fairly constant (1.4 X lO-' m s-' MPa-'). In
more basal root zones, Lp, was considerably smaller and varied
between roots. The low contribution of basal zones t o the overall
water uptake indicated an influence of the exodermal Casparian
bands and/or suberin lamellae in the endodermis or exodermis,
which develop at distances larger than 50 to 60 mm from the root
tip.
Water transport through plants is governed by water potential differences and hydraulic conductances. Both the conductances and the driving forces may vary greatly in short or
long time scales. For roots, it has been reported that different
root zones contribute to different extents to the overall uptake
of water (Rosene, 1937; Brouwer, 1954; Sanderson, 1983;
Haussling et al., 1988; Frensch and Steudle, 1989; North and
Nobel, 1991). On the other hand, the L, may strongly depend
on environmental factors such as high salinity and drought
of the soil or anoxia (Azaizeh and Steudle, 1991; Azaizeh et
al., 1992; Cruz et al., 1992; Birner and Steudle, 1993). Furthermore, L, may be a function of the flow rate (Weatherley,
Supported by a grant from the Deutsche Forschungsgemeinschaft, Sonderforschungsbereich 137.
* Corresponding author; fax 49-921-552564.
1305
Abbreviations: A,, root surface area (mm'); I , length of root end
segment (mm); L,, axial hydraulic conductance (m3 s-' MPa-'); Lp,,
radial hydraulic conductivity (m s-' MPa-I); L,, hydraulic conductance (m3s-' MPa-'); L,, axial hydraulic conductivity (m' s-' MPa-');
n, number of measurements; N, number of roots measured; P,,root
pressure; P,,, stationary root pressure; z, distance from the root tip
("1.
Melchior and Steudle
1306
and 120 mm from the root tip. Because of the development
of laterals, the authors could not measure Lp, over larger
distances where Casparian bands and suberin lamellae in the
exodermis occurred (Perumalla and Peterson, 1986).
In this study, the pressure probe technique has been applied to onion (Allium cepa L.) roots, which show an anatomy
somewhat different from maize (St. Aubin et al., 1986;
Frensch and Steudle, 1989). Onion roots are unbranched
over long distances (up to 400 mm from the apex). When
grown in hydroponics, this species develops Casparian bands
in the exodermis and suberin lamellae in endodermis and
exodermis at distances from the root tip of z > 50 to 60 mm,
i.e. earlier than in roots of maize. Compared with maize, the
large vessels of the late metaxylem mature closer to the root
tip in a zone where no laterals have crossed the outer root
tissue. Thus, onion roots should be better suited to follow
changes of hydraulic properties of roots during development.
The dependency of both the Lp, and L, on the position
along the root was studied. The measured L, was compared
with that calculated from vessel dimensions obtained from
cross-sections according to Poiseuille’s law. These data on
changes of root hydraulics, i.e. the functions Lp,(z) and L,(z),
are needed to create a detailed modeling of water transport
across roots.
MATERIALS A N D METHODS
Plant Material
Bulbs of onion (Allium cepa L. cv Calypso) obtained from
Vanderhave Seeds (NL-4410AA Rilland, The Netherlands),
were kept in polyethylene bags for 9 d at 6OC. They were
then grown for 3 d in vermiculite moistened with one-third
Miller solution (Miller, 1980. Composition in mo1 m-3: K*
1.0, Ca2+0.5, Mg2+0.67, NH4+1.33, N03- 2.67, P043- 0.5,
S042- 0.5. Micronutrients in mmol m-3: C1- 50, Fe3+36 and
Na+ 36, added as FeNa-EDTA, B033- 25, Mn2+ 2, Zn2+ 2,
Cu2+0.5, Mo0d2- 0.5. Osmotic concentration 7 mOsmol kg-’;
pH 6.0, adjusted with sulfuric acid). After this time, each bulb
was transferred to hydroculture in a nontransparent 15-L
polyvinyl chloride container with aerated one-third Miller
solution. Nutrient solution was exchanged every 7 d. A11
plants were grown in a climatic chamber at a lightldark
rhythm of 14/10 h and a PAR of 500 to 650 pmol m-2 s-‘.
The temperature was kept at 20 to 21OC and the RH was 70
to 80% (water vapor pressure deficit = 4-7 Pa kPa-I).
Root Pressure Probe
After 6 to 22 d in hydroculture, roots were 140 to 510 mm
in length. They were excised close to the bulb, transferred to
a 1-L container with aerated nutrient solution, and fixed to a
root pressure probe (Fig. 1).The roots were left on the probe
for at least 24 h. Only those roots were taken that built up
root pressures of larger than 0.2 MPa (maximum, 0.62 MPa).
Apical parts of the roots that did not have visible laterals were
used to avoid difficulties in determining root surface area.
Experiments were carried out at a teinperature of 21.0 f
0.5OC. Only a few roots were used from each bulb (5-10 of
about 50-100).
The root pressure probe connected the xylem of the cut
Plant Physiol. Vol. 101, 1993
micrometer
Figure 1. Root pressure probe for measuring root pressure and
hydraulic conductivity of roots. An excised root ora root segment
open at both sides was connected tightly to the probe by a silicone
seal to determine Lp, and L, of root segments. Root pressures were
monitored continuously by a pressure transducer in the probe.
Water flows were induced with the aid of a metal rod. The meniscus
between water and silicone oil served as a reference to measure
volume changes in the capillary (for further explanations, see text).
root to a pressure transducer (Fig. 1).Root pressure could be
recorded continuously. The probe was filled with silicone oil
and water so that a meniscus formed in the measuring
capillary that was used as a point of reference. With the aid
of a metal rod connected with a micrometer screw, volume
changes were induced in the measuring system and responses
of root pressure were monitored on a chart recorder. Volume
changes were calculated from the i.d. of the measuring capillary (200 or 250 pm) and from the shift of the meniscus in
the capillary. Severa1step changes of pressure (volume) were
produced in each direction to determine the elastic coefficient
of the system (APr/AVs), i.e. the relation between pressure
(APJ and volume changes (AVs). AP,/AVs ranged between
0.7 and 3.6 MPa mm-3 (7-36 bar/pL). Thus, changes of
pressure with time measured by the root pressure probe (dP,/
dt) could be converted into water flows (dVJdt).
Measurement of 1,
To measure the L, of the root, a stationary root pressure
was first established. Then the meniscus was shifted instantaneously and held in the new position. This induced a
volume (water) flow from the root xylem into the medium
when P, was increased (“exosmotic”)or in the opposite direction when P, was lowered (“endosmotic“).After the change,
root pressure tended to reach a new stationary value following a response curve that was close to exponential with time
(“pressure relaxation”). From the rate constant of the waterflow equilibration (kWr) and from the measured elasticity of
the system (AP,/AVs), the L, in m3 s-’ MPa-’ was evaluated
(Steudle and Jeschke, 1983; Steudle et al., 1987):
T,, = half time of water exchange of root. For details of the
Hydraulic Conductivity of Onion Roots
calculation and for the background of the theory, see the
references given. Although pressure relaxations were virtually exponential (f > 0.91, mean -+ SD = 0.986 & 0.011, n
= 187, N = 28), the P,(t) curves appeared to incorporate
other components. They could be, in fact, described as a sum
of three exponential functions (Peterson and Steudle, 1993).
The first phase, which has a very short time constant (4s),
has been interpreted to be due to an extensibility of the
measuring system and/or to difficulties in rapidly adjusting
the position of the meniscus. The last (third) phase was very
slow. Because it did not appear in “osmotic experiments”
when the water flow was much smaller, it was attributed to
concentration polarization effects in the xylem (Peterson and
Steudle, 1993). The very short and the very slow phases
could be distinguished visually from the main body of the
curves when digitizing the recorder strips with a digitizing
tablet (DIGIKON, Kontron Registriertechnik, Eching, Germany) and transforming the P( t ) curves to semilog plots of
In(P - P,) as a function of time, where P,is the end pressure
reached after the relaxation.
In a preliminary experiment using a different onion cultivar, it could be shown that half-times of the second and the
third phase were not correlated (f = 0.008, P = 0.58, n =
40, N = 10; half-times 3-16 s and 26-690 s, respectively).
Fitting the sum of two exponential functions to the time
courses, it was shown that the influence of the slower (third)
phase on the determination of the half-time of the second
phase could be neglected, because corrected and uncorrected
values differed by only 1%.Thus, the evaluation of TrKafter
visual exclusion of the first and third phases was justified.
The absolute value of P, largely influenced the half-times
calculated. Therefore, a fit program was used and the optimum P,for the second phase was calculated. Compared with
the visual determination of P,, the fit resulted in changes of
the rate constants (and L,) by factors of 0.72 to 2.42 (mean f
SD = 1.43 f 0.26, n = 187, N = 28) and 0.35 to 2.00 (mean
2 SD = 1.15 f 0.19, n = 266, N = 31) in intact and open root
segments, respectively.
Measurement of 1, Using the Root Pressure Probe
First
Method
After excising the apical part (length > 45 mm) of a root
fixed to the probe, root pressure fel1 to zero because of the
opening of mature xylem. The L. was much larger than the
radial so that the water flow induced across the remaining
root segment still attached to the probe was dominated by
the flow along the xylem and out of the cut vessels. Therefore,
under these circumstances, the axial hydraulic conductivity
per unit of root length ( L , = L, X e; where e= length of the
remaining root segment) could be directly determined from
pressure relaxations, neglecting the radial component of
water flow. L, in m4 s-’ MPa-’ was calculated from:
1307
“Second Method”). Hence, the L, of the part of the root
outside the seal could be also evaluated.
To eliminate effects of a reduced conductance at the seal
during regular relaxations with intact roots (apical ends
closed), it was ensured that, after cutting off the root at the
seal, the half-time decreased by at least a factor of 10. The
hydraulic resistances of the root segments remaining in the
seal were subtracted from the total hydraulic resistances
measured. It should be noted that L, represented the axial
hydraulic conductivity per unit length of root in m4s-’ MPa-’
(m3m s-’ MPa-’), whereas L, was the hydraulic conductance
in m3 s-’ MPa-I. Hence, the values of L, and L, cannot be
compared directly.
Second Method
Using a different technique for determining L,, 18-mm root
segments that were left in the seal of the equipment after
cutting off the root right at the seal were measured (Frensch
and Steudle, 1989). This technique avoided any radial component of water flow. It was used to determine L, as a
function of the position along the root (L,[z]) since small
segments of roots could be measured.
Evaluation of
f, from the Anatomy of Root Xylem
Two staining procedures were used to identify mature,
conducting xylem vessels on free-hand cross-sections. (a) The
fluorescent staining technique of Brundrett et al. (1988) with
berberine sulfate (1 g/L, exposure, 1 h) and aniline blue (5
g/L, 1 h) resulted in a bright staining of the lignified xylem
vessels. With this method, Casparian bands in the endodermis and exodermis could be detected. Photographs were
taken using a fluorescence microscope ‘(Zeiss, Oberkochen,
Germany), exciting wavelengths 390 to 420 nm, dichroitic
mirror FT 425, barrier filter LP 450 nm. (b) In a quicker
procedure, toluidine blue was used for staining (0.3 g/L, 1
min). With the latter technique, cross-sections were photographed under white light.
To determine the dimensions of mature vessels, the largest
diameter and the diameter perpendicular to it of the crosssections of stained xylem elements were digitized using a
digitizing tablet. From the data, the L, per unit of length was
calculated according to Poiseuille’s law:
(3)
for cylindrical vessels (ri = radius of the vessel; 1 = viscosity
of water [9.7 X 10-” MPa s at 2l0C]). Because cross-sections
of vessels were not circular, a mean radius (ri =
a=
largest radius and b = smallest radius) was used in Equation
3. Cross-sections could be better described as ellipses. Therefore, a different equation was used (Nonweiler, 1975):
m;
1 Ai x m:
L,=-ZD
In Equation 2, L, included the hydraulic resistance of the
root segment in the seal that was determined separately (see
ki
(4)
(A, cross-sectional area of vessel; mi,”mean hydraulic radius”
Melchior and Steudle
1308
Plant Physiol. Vol. 101, 1993
r
Figure 2. Freehand cross-sections of onion (A. cepa L.) roots grown hydroponically. Mature xylem was stained with
toluidine blue O (A) or with berberine sulfate and aniline blue (B-D). The latter technique also stained Casparian bands
in endodermis (B, arrows) and exodermis (C). A, Section taken at a z = 300 mm from the apex. Xylem vessels showed a
radial arrangement with the late metaxylem (LMX) in the center. Except for the "passage cells" (PC, arrow), the endodermis
had thickened cell walls in this state (Co, cortex). B, Mature xylem vessels and Casparian bands (CB) in the endodermis
(En; z = 230 mm). C, Rhizodermis (Rh) and exodermis (Ex) 300 mm from the apex. In part, two parallel Casparian bands
are visible in the exodermis. D, Typical example of a cross-section in which some of the vessels (marked by "?") could
not be classified as either mature or immature (z = 155 mm).
of vessel, j, which is given by Ai divided by the circumference;
k{, coefficient depending on the geometry of the vessel.) For
an elliptical cross-section of the eccentricity t'i (= Va2 — b2/a),
ki is given by (Nonweiler, 1975):
1+
(5)
In some cases, the staining did not clearly indicate whether
a vessel was conducting or not. These vessels were digitized
separately. This resulted in a certain error of the calculated
conductivity. Other errors (e.g. digitizing errors) could be
neglected compared with this uncertainty.
RESULTS
Anatomy of Onion Roots
To evaluate Lx of the roots, stained cross-sections were
taken (see below). Staining procedures were used to identify
lignified xylem vessels as well as thickenings of the endodermal cell walls (toluidine blue, Fig. 2A) and Casparian
bands in the endodermis and exodermis (berberine sulfate
and aniline blue, Fig. 2, B and C). It is known from the
literature that Casparian bands develop very early in the
endodermis of onion roots even when plants are grown
hydroponically (Perumalla and Peterson, 1986). In this study,
Hydraulic Conductivity of Onion Roots
1309
Casparian bands were already found in the endodermis 10
mm from the apex. Casparian bands in the exodennis (equal
to the hypodermis) appeared at larger distances from the root
tip (about 50 mm). The development of the endodermis and
exodermis was not investigated systematically. However, it
was found that Casparian bands did not appear simultaneously in a11 exodermal cells. Rather, they first formed
"patches." Therefore, on cross-sections, interruptions in the
exodermal ring of Casparian bands were observed. With the
two staining methods, it was not always possible to identify
all xylem vessels as mature (i.e. as lignified) or not. An
example of such a transition in the staining is shown in Figure
2D. This uncertainty resulted in the most important error in
the determination of the L, from the cross-sections (see "Materials and Methods" and below).
vessels. Thus, the system should exhibit properties comparable to those of a leaky electric cable. It has been treated
theoretically by Landsberg and Fowkes (1978) and extended
by Frensch and Steudle (1989). At least for the very tip of
the root where the xylem is yet immature, the axial resistance
should contribute substantially to the overall hydraulic resistance or even should be the limiting component. On the
other hand, the axial conductance for water flow across short
root segments open at both ends is quite large if the xylem is
mature so that it will dominate the total conductance. That
is why the L, could be determined separately from the radial
and why it is treated here first. From L, and from the total
conductance of roots, the Lp, was then evaluated.
L,
L, per unit of root length (in m4 s-l MPa-') as calculated
from the xylem vessel dimensions according to Poiseuille's
law increased with increasing z in the apical 100 to 200 mm.
L, ranged over more than 3 orders of magnitude (Fig. 3, A
and B). At distances of about 200 mm from the apex, all
xylem vessels including the central late metaxylem element
The hydraulic conductance measured on intact root segments (apical ends closed) was composed of axial and radial
conductances in series and in parallel, respectively. Water
taken up radially will be continuously moved away in the
I
I
I
I
L,
I
I
I
t
1o-0
T
10-'
E
10-10
4
i
3
1 o-"
10-"
U
2
cross sections /
berberine and aniline
íO-"
n
IO-~
10-
toluidine
D
C
3
I
10-'O
B
o
1
10-'
(d
a
2
10-'O
10-'O
.-I
I
o
*E
1o-"
10-"
a
10-12
10-12
U
root segments in seal
0
I
I
I
100
200
300
z
I"[
400 0
I
I
I
I
100
200
300
400
10-13
I"[
z
Figure 3. L, per unit of root length of onion roots as a function of z.Values were calculated from cross-sections according
to Poiseuille's law (A and 6) or measured using the root pressure probe (C and D). L, values are presented in a logarithmic
scale because they ranged over more than 3 orders of magnitude. The axial hydraulic resistances (logarithmicdata) were
fitted using the function R, = 1/L, = a X exp (-b X z) + c; a, b, and c > O (see Table I). A, Mature xylem stained with
berberine sulfate/aniline blue (n = 101, N = 11). B, Sections stained with toluidine blue O (n = 89, N = 14). C, Short
open root segments outside the seal were measured excluding the effect of the part of the root inside the seal (see D).
D, Only the parts of roots inside the seal were measured, avoiding radial water flow, after cutting the root right at the
seal. As in C, roots had been connected to the pressure probe for at least 24 h and root pressures were >0.2 MPa. Error
bars in C and D represent the SD (n = 3-5 and 2-11, respectively). In A and B, the main error resulted from the
uncertainty of identifying mature and immature vessels (see "Materials and Methods").
Melchior and Steudle
1310
Table 1. Parameters of the fit function R, = I / L , = a x exp(-b X z)
Determination
Cross-sections/fluorescent staining
Cross-sections/toluidine blue
Pressure probe/open segments ( L < 300 mm)
Pressure probelroot segments in the seal
+ c, shown in Figure 3
log (4
log (a)
I
m4 5-l MPa-’
mm
m4 5-’ MPa-’
0.050 rt 0.004
8.71 f 0.06
8.70 k 0.08
10.15 k 0.1 1
9.72 2 0.1 5
11.87-+_0.14
11.06f0.19
12.7 2 1.2
12.07 -+_ 0.47
were mature so that L, did not vary in the basal part of long
roots ( z > 200 mm). It can be seen from Figure 3, A and B,
that the two staining procedures used (berberine sulfate/
aniline blue and toluidine blue O) resulted in similar values
for L, at the base of the root (z = I ) and for the distance at
which the lignification of xylem was complete. However, the
fluorescent staining resulted in somewhat lower values of L,
near the root tip (Fig. 3, A and B; Table I).
L, were calculated using Poiseuille’s law (Eq. 3) or the
equations for elliptical cross-sections (Eqs. 4 and 5). Equation
3 always resulted in somewhat higher values of L, than
Equations 4 and 5. It is known that the differences are small
for ellipses of low eccentricities (Preston, 1938). This is the
reason why L, calculated from Equations 4 and 5 differed
from those calculated from Equation 3 only by a factor of
1.30 at maximum (mean f SD: 1.050 f 0.044; n = 190
cross-sections; N = 21 roots). Differences were larger near
the root tip, i.e. at low L,, where the small vessels of the
protoxylem and of the early metaxylem largely contributed
to the overall L,. These vessels were more eccentric than the
Figure 4. L, (A and B) and Lp, (C and D) as a
function of the geometric A, or I of the intact
root segments. L, represented radial conductance to a good approximation because of high
axial conductance, i.e. low resistance, of xylem
water flow (see Fig. 5). Each data point represents one root. A, For A, < 550 mm’, there was
a linear relationship between L, and A,, indicating a constant hydraulic conductivity (Lp, = dL,/
dA,) in this root zone (? = 0.69, N = 14). The
intercept with the x axis was 90 mmz and represented the apical area that was ”hydraulically
isolated“ from the rest of t h e root. For A, > 550
mm2,no dependency on A, could be measured
for L, (? = 0.019, significance P = 0.64, N =
14). This means that in this root zone, Lp, was
small as compared with the apical 550 mmz. B,
Due to the linear relationship between A, and
I, the dependency of L, on I was similar to that
on A,. C, The apparent hydraulic conductivity
(LppPP = L,/A,) increased with increasing surface
area in the apical 400 mm2. D, When A, was
corrected by the area of the”’hydraulically isolated” tip zone (Ao),the Lp, did not depend on
A, in the apical 500 mm2, With further increasing surface area, Lp, decreased because of the
small contribution of the basal root zone to the
overall uptake. Error bars indicate the SD ( n =
4-1 1 experiments per root).
Plant Physiol. Vol. 101, 1993
0.048 2 0.007
0.073 k 0.046
0.029 2 0.01 O
n
N
1o1
11
89
18
31
14
18
31
large elements of late metaxylem. A11 data shown in Figure
3, A and B, and TabIe I were calculated using Equations 4
and 5.
L, measured with the root pressure probe were lower than
those calculated from the radii of mature xylem vessels (Fig.
3, C and D). The values determined on short, open root
segments outside the seal were lower by 1 to 2 orders of
magnitude than those evaluated from the stained crosssections at z = 200 to 300 mm. However, they were similar
at distances of 50 mm < z < 100 mm and z approximately
equal to 400 mm (Fig. 3C). Thus, after xylem vessels were
lignified, they were not fully conducting, probably because
of partly remaining cross walls (see “Discussion”). Before
cutting a root for the determination of L,, a P, of more than
0.2 MPa (2 bar) was built up over at least 24 h, indicating a
proper sealing of the root. In addition, this increase and
subsequent stability of P, indicated that active processes
(active ion pumping) were maintained even 1 d after roots
were excised from the bulb and attached to the pressure
probe (see also Birner and Steudle, 1993). Possible effects of
I:i
I
l
i
7
0.4
I
0.2
0.0 1
O?
500
*o
A,
1000
O
1500
#
II
I
I
,
100
200
300
[“‘I
.+
I
sl
a
z
d
I
E
u
rO
-4
h
a
I
8
Y0
500
1000
[“‘I
1500
0
O
500
1000
1500
Hydraulic Conductivity of Onion Roots
the hydraulic resistance of those parts of the roots that were
in the seal could be excluded because the data were corrected
for this, i.e. the data obtained using the second method (Fig.
3D) were used to correct the data of Figure 3C.
The lowest values of L, were obtained when roots were
cut right at the seal after being fixed to the probe for at least
24 h (Fig. 3D; before the cut P, > 0.2 MPa). This indicated a
compression of some of the root xylem vessels due to sealing.
It should be noted that significant contributions of the seal
(leading to extremely low L, values) could be easily excluded.
As in the determination of the overall hydraulic conductance,
only those experiments were taken into account in which
half-times of water exchange of the intact (closed) root segment were by at least 1 order of magnitude larger compared
with the half-times measured after the cut right at the seal.
Because the data of Figure 3C were corrected for effects of
compressing root xylem in the seals, they refer to the same
state as in the intact root.
1311
However, the apical 90 mm2 (Ao, corresponding to a root
length of about 27 mm) were considered as "hydraulically
isolated," i.e. they did not contribute much to water uptake
of the root because of immature xylem. The effect is shown
as a positive intercept of the regression line with the A, axis
(Ao, Fig. 4A). Ao was subtracted from the geometric surface
area of the root to give the "effective surface area." The
effective surface area of the roots was then used to recalculate
Lp, (Fig. 4D). The correction had a large effect for short roots
as compared with the "apparent hydraulic conductivity,"
which refers to the total (geometric) surface area (Fig. 4C).
Because of the linear relationship between A, and I (A, = 4.21
mm X I - 66 mm2, ? = 0.966; N = 28; for I > 63 mm), the
critica1 value of A, of 550 mm2 up to which L, was linearly
related to A, corresponded to z = 146 mm. The regression
line of A,(I) did not pass through the origin because of the
decrease of root diameter toward the root tip.
L, versus Radial Hydraulic Conductance
LPr
The contribution of the L, a t a given position from the apex
The overall L, increased with increasing A, and 1 of the root
segment (Fig. 4, A and B). There was a linear relationship of
L, as a function of A , for values of A, < 550 mm2, as one
would expect (Fig. 4A, ? = 0.686, significance P < 0.001, N
= 14). The slope of this regression line represented the radial
hydraulic conductivity Lp, = 1.37 X 10-7 m s-' MPa-I.
However, for values of A, > 550 mm2,the L,(A,) curve reached
a saturation value, and there was no significant correlation
between L, and A, (r' = 0.019, P = 0.64, N = 14). This means
that due to root development, apical parts of the root contributed to a higher extent to the overall water uptake than
basal. Lp, of basal root zones was reduced substantially.
10-'
(z) was evaluated by comparing the total hydraulic conduct-
ance of the part of the root distal from that position (LJ with
the axial conductance between that position and the base of
the root, L, (Fig. 5). That value was evaluated by integrating
the specific axial hydraulic resistance (R, = l/L,) over the
distance between the given position and the base of the root
(1 = 300 mm was taken) and calculating the inverse of this
integrated value. For z > 60 mm, the L, calculated from the
cross-sections following Equations 4 and 5 (or Poiseuille's
law) was larger than the total conductance of the apical part
by at least 2 orders of magnitude. If the L, values measured
with the root pressure probe were taken (Fig. 3, C and D), L,
was larger than L, by a factor of at least 5 for a total root
length of 300 mm. However, no measurements of L, were
made for short roots ( I < 63 mm), and determination of L,
using the pressure probe was restricted to z > 51 mm.
Therefore, no comparison between measured L, and L, could
be made for the root zone close to the apex.
DlSCUSSlON
O
200
100
300
I"[
Figure 5. Comparison between the axial conductance of a root
segment from the root base to a given position z from the root apex
( L , lines) and the total hydraulic conductance of a closed root
segment distal to that position (Lr, symbols).The values for L, were
calculated assuming a root length of 300 m m and using the fit
functions of the L, (Fig. 3 and Table I). In the range where L , could
be measured ( z > 63 mm), the L, between t h e position z and the
root base was larger than the hydraulic conductance of the apical
root part by a factor of at least 5, if the measured L, values were
taken (measured on open root segments or on the part of the roots
inside the seal). I f L, was calculated o n t h e base of the L, values
from the cross-sections(stained with berberine/aniline or toluidine),
it was even larger (2 or more orders of magnitude difference to L,).
The results show that in onion roots, both the L, and Lp,
change strongly from the tip toward the base. L, increased
over more than 3 orders of magnitude with increasing z,
whereas the Lp, was lower by a factor of at least 5 in basal
relative to apical root zones, although the small contribution
to radial water uptake in the basal parts was quite variable
and, therefore, difficult to quantify. Using the root pressure
probe technique, L,(z) and Lp,(z) could be determined at a
rather high resolution that, in principle, could be increased
further. This is hardly possible with other available
techniques.
L X
The results of the staining procedures involve some uncertainty because the decision whether a vessel was conducting
or not was difficult. The technique could be improved by
making longitudinal sections in addition to cross-sections to
1312
'
I
Melchior and Steudle
examine at which position in the root and to what extent
cross walls in the vessels are dissolved. Another possibility
to improve the method would be to suck dye across root
segments that would stain only conducting vessels, the dimensions of which are then used in the calculations (Peterson
and Steudle, 1993).
At I = 200 to 300 mm, L, values calculated according to
Poiseuille's law were larger than those measured with the
root pressure probe by more than 1 order of magnitude. This
difference is in accordance with Zimmermann and Brown
(1971); Greacen et al. (1976); Giordano et al. (1978); and
Frensch and Steudle (1989). The results support the view that
the application of Poiseuille's law to xylem transport should
be done with caution. Deviations from the theory seem to be
due mainly to remaining cross walls. On the other hand, the
elliptical shape of vessel cross-sections and the roughness of
walls play a minor role if any (Nonweiler, 1975).
In pressure probe experiments, a proper sealing was indicated by root pressures of more than 0.2 MPa that were built
up over time periods of at least 24 h. Preliminary experiments
showed that extremely high apparent Lp, (Lp:PP
= LJA,) of
up to 5 x 10-6 m s-' MPa-' indicated a bypass of water at
the seal. They were measured only at P,, below 0.15 MPa.
For P, > 0.2 MPa, there was no dependency of Lp$PP on P,.
The method of Frensch and Steudle (1989), which involved
a cutting of root segments attached to the pressure probe
right at the seal and measuring the remaining parts of the
roots inside the seal, resulted, in part, in extremely low values
of L, (not shown in Fig. 3D). Even when only those experiments were chosen where the half-times of water flow were
lower by a factor of at least 10 after the cut at the seal than
in intact root segments, low L, values were calculated for z >
300 mm. These cases indicated a substantial compression of
the xylem by the seal, although this did not interfere with
the determination of the overall hydraulic conductance of
the root. Without considering these rare cases, a good correspondence was found between this method and the determination of L, on short, open root segments outside the seal,
where the values were corrected for the effect of the seal
(compare Fig. 3, C and D). Thus, in the pressure probe
experiments presented, "sealing effects" were not important.
This is readily shown by comparing the two methods to
determine L,.
Clarkson et al. (1984) perfused onion root segments of a
length of 65 mm. In agreement with the data found in this
study, these authors measured an increase of L, with increasing z even for z > 250 mm. They explained the increase by
the existence of cross walls between members of lignified
xylem vessels. Similarly, in barley roots, the maximal axial
conductivity was reached at distances of 150 to 250 mm from
the root tip. This was correlated with the complete breakdown of cross walls between vessel members (Sanderson et
al., 1982).
In the present study, remains of cross walls between vessels
were found in basal root parts and should be studied systematically in the future using seria1 cross- and longitudinal
sections. A narrowing by remains of cross walls may result
in turbulent flow of water and in a reduction of L,. This may
also occur as a consequence of circular stiffenings of vessel
walls that cannot be seen on cross-sections. Giordano et al.
Plant Physiol. Vol. 101, 1993
(1978) proved that measured axial conductivities can differ
greatly from calculated ones according to Poiseuille's law if
the vessels were periodically narrower, leading to turbulences. Thus, small Reynolds numbers by themselves do not
indicate that water flow would be laminar. The absolute
values of L, calculated from the data given by Clarkson et al.
(1984) range from 2.3 X 10-" m4 s-' MPa-' ( z = 70-135
mm) to 7.2 X 10-l' m4 s-' MPa-' (z = 300-365 mm) and
correspond with the values measured in this study (10-l2to
10-9 m4 s-' MPa-', Fig. 3, C and D). However, in the present
study, the changes during root development were recorded
at a resolution that was higher by a factor of 3 to 4.
lnfluence of L, on Determination of Lp,
Over the whole range of length of the root segments
measured in this study (60-320 mm), L, to a hypothetical
root base at z = 300 mm was larger by a factor of at least 5
than the total L, of the segment. The linear relationship
between L, and the A, indicated a low influence of L, on L,
(for z > 63 mm). This result shows that for onion roots as
well as for roots of other species, a two-compartment model
(which neglects the longitudinal component of LI) may be
used to evaluate Lp, to a good approximation (Steudle and
Jeschke, 1983; Steudle et al., 1987). However, it has to be
taken into account that in the tip region, immature xylem
would cause a high axial resistance that would hydraulically
isolate this apical zone from the basal root parts. The effect
is shown as a positive intercept with the A, axis (AO,Fig. 4A).
Due to the large variation between roots, the intercept had a
large error because the slope of the regression line and
intercept with the L, axis had errors of 20 and 80%, respectively. The surface area of the hydraulically isolated tip zone
could be higher than Ao = 90 mm2 (corresponding to z = 27
mm) because the inverse relationship (A, plotted as a function
of L,) had an intercept with the A, axis of 171 f 39 mm2
(corresponding to z = 51 f 12 mm). In the experiments of
Clarkson et al. (1984), the apical 60 mm had high axial
hydraulic resistances that contributed substantially to the
overall resistance.
LPr
Using segments of primary roots of maize of a length of
18 to 140 mm, which were shorter than those used in this
paper, Frensch and Steudle (1989) reported an increase of
the "apparent Lp," (measured with the root pressure probe)
with increasing root length. However, when it was taken into
account that the apical 15 mm of the segments were "hydraulically isolated" because of immature metaxylem, the Lp,
Treferred to the effective surface area) was constant over the
entire root length. The authors state that a decrease of Lp, in
more basal root zones (z > 140 mm, as measured in this
study) could, however, not be excluded. In maize roots,
Casparian bands and suberin lamellae in the exodermis were
detected at distances from the root tip of z = 120 to 130 mm,
whereas in onion roots they develop at z = 50 to 60 mm
(Perumalla and Peterson, 1986). In addition, the low contribution of basal zones of onion roots could be explained by a
minor contribution of apbplasmic bypasses in those areas
Hydraulic Conductivity of Onion Roots
where secondary root initials penetrate the endodermis. In
part, Frensch and Steudle (1989) interpreted the high contribution of the apoplasmic pathway to radial water transport
by a flow across the latter structures.
An increased water uptake at the site of secondary root
formation was also indicated by experiments with apoplasmic
tracers (Peterson et al., 1981) and was discussed for Norway
spruce (Haussling et al., 1988). If the exodermis would efficiently block off the apoplasmic water movement in onion
roots, as indicated by the work of Peterson and coworkers
(see the refs. given by Peterson, 1989), the penetration of
root primordia across the cortex (consisting of 10-15 cell
layers) would only slightly affect the water taken up through
the cell walls unless the laterals wound the exodermis. Although it was found in this study that laterals develop much
earlier than they emerge from the rhizodermis (similar to
maize), the position where the exodermis was penetrated and
wounded was not examined. However, the differences in the
formation of laterals as well as the differences in the formation of Casparian bands and suberin lamellae in endodermis
and exodermis in both species seem to be important for the
difference of the Lp,(z) characteristics observed for the two
species.
North and Nobel (1991) measured the apparent Lp, of
Agave deserti by sucking solution across root segments about
80 mm in length. They found that the apparent Lp, of apical
parts of nodal roots ("young roots") was larger by a factor of
5 compared with that of basal parts ("older roots"). Using the
model of Landsberg and Fowkes (1978), North and Nobel
evaluated Lp, to be 1.8 X 10-6 m s-' MPa-' and 3.8 X 10-8
m s-' MPa-' for apical and basal parts of the roots, respectively, i.e. there was a decrease in Lp, by a factor of about 50
during root development. The result was mainly explained
by the suberization of endodermal and exodermal cell walls.
In a study on roots of water-stressed Sorghum bicolor, Cruz
et al. (1992) applied the same technique to determine L, and
L,. However, these authors did not correct for the apical
zone with low axial conductance. Rather, they gave an apparent Lp, of 2.2 X 10-7 m s-' MPa-' and 0.6 X 10-7 m s-'
MPa-' for controls and water-stressed plants, respectively. L,
declined by 2 orders of magnitude due to the drought treatment (1.8 X 10-l' m4s-' MPa-' and 1.7 X 10-l2 m4s-' MPa-'
in control and stressed plants, respectively). Compared with
hydroponically grown onion roots, the low L, of waterstressed Sorghum roots was similar to L, close to the root tip,
where only the small protoxylem vessels were mature. The
large decrease of L, of Sorghum roots during drought was
correlated with smaller vessel diameters and a lower proportion of lignified xylem elements (Cruz et al., 1992). In stressed
roots, exodermis and endodermis cell walls were lignified
and suberized earlier. It should be noted that, different from
the root pressure probe technique, in the experiments of
North and Nobel(l991) and Cruz et al. (1992), there was no
criterion that roots were well sealed and not wounded before
measuring the conductances. A bypass of water at the seal
could possibly be detected by an overestimation of axial and
radial conductivities. However, the results of North and
Nobel(l991) and Cruz et al. (1992) show that even with the
more simplifying techniques, there were indications of large
changes of L, and Lp, caused by environmental factors such
1313
as drought stress. With these techniques, however, hydraulic
parameters could not be resolved in finer detail.
Rosene (1937) carefully examined the contribution of different root zones to the overall water absorption of intact
onion roots (still attached to the bulbs). She used micropotometers and found that a11 parts of the root from the tip to
the base (the root cap itself was not measured) absorbed
water but at different rates. In longer (older) roots ( I > 80
mm), the water uptake rate was maximal in the tip zone
between z = O and z = 40 mm and declined strongly at z =
60 to 70 mm. Water absorption ranged from 3 X 10-9 to 8 X
10-* m s-', with a large variation between individual plants.
There was good evidence that intact and excised roots of
corresponding age and length showed similar behavior with
respect to absolute rates and axial gradients of water uptake
(Rosene, 1941a). Rosene (1941b) determined the driving force
for water uptake by applying osmotic solutions to different
root zones at concentrations just sufficient to stop the water
flow. Taking into account a reflection coefficient of about
0.65 for the solute used (KNO,; our unpublished result), the
driving force in the absorption experiment could be calculated
to be about 0.4 MPa. Thus, Lp, would have ranged from 7 X
10-9 to 2 X 10-7 m s-' MPa-' for intact roots measured in an
atmosphere saturated with water.
These values are smaller than the Lp, measured in the
present study (pressure relaxations driven by gradients of
hydrostatic pressure) and larger than Lp, measured in osmotically driven relaxations (our unpublished results). However,
it is doubtful whether RH really was 100% as stated by
Rosene (1941b), so that even at near water saturation, transpiration (pressure)-driven water uptake could have substantially contributed to total water absorption. Therefore, the
driving force in the experiments of Rosene (1937) may have
been a mixed hydrostatic and osmotic gradient and the
absolute value of Lp, would be, thus, understandable.
Recently, Zimmermann et al. (1992) criticized the root
pressure probe technique as being inadequate for the determination of water and solute relations parameters for severa1
reasons. (a) The assumption of a high rigidity of the xylem
would not hold because (from their own measurements) the
xylem would have an elastic modulus of a few tenths of 1
MPa. This would mean that this structure would double its
volume when subjected to pressure increases of a few bar.
(b) Zimmermann et al. (1992) are concerned about effects of
unstirred layers during the measurement of water flow. However, they overlooked that a rigorous analysis has proven
that results of pressure probe experiments (eg. low reflection
coefficients of roots) could not be completely explained by
unstirred layers (Steudle and Frensch, 1989). (c) Zimmermann et al. (1992) are also concerned by the application of
the theory of irreversible thermodynamics to complex composite structures such as roots. However, this is possible as
shown by basic theoretical work ( e g Kedem and Katchalsky,
1963a, 1963b) as well as by papers where it has been applied
to roots (e.g. Fiscus, 1975; Pitman et al., 1981; Steudle and
Jeschke, 1983; Dainty, 1985; Miller, 1985; Steudle et al., 1987;
Schambil and Woermann, 1989). Application of irreversible
thermodynamics to roots has led to the "composite transport
model of the root" (Steudle, 1992).
The problems of Zimmermann et al. (1992) arise from their
1314
Melchior and Steudle
Plant Physiol. Vol. 101, 1993
own cell pressure probe measurements on root cortex cells of
LITERATURE ClTED
Aster tripolium (measurements at the root level could not be
performed). The results indicated that in young Aster roots
Azaizeh H, Gunse B, Steudle E (1992) Effects of NaCl and CaCI2
on water transport across root cells of maize (Zea mays L.) seedlings.
Plant Physiol99 886-894
Azaizeh H, Steudle E (1991) Effects of salinity on water transport
of excised maize (Zea mays L.) roots. Plant Physiol97: 1136-1145
Birner T, Steudle E (1993) Effects of anaerobic conditions on water
and solute relations and active transport in root of maize (Zea mays
L).Planta (in press)
Boyer JS (1985) Water transport. Annu Rev Plant Physiol 3 6
473-516
Brouwer R (1954) The regulating influence of transpiration and
suction tension on the water and salt uptake by the roots of intact
Vicia faba plants. Acta Bot Neerl 3 264-312
Brundrett MC, Enstone DE, Peterson CA (1988)A berberine-aniline
blue fluorescent staining procedure for suberin, lignin, and callose
in plant tissue. Protoplasma 1 4 6 133-142
Clarkson DT, Williams L, Hanson JB (1984) Perfusion of onion
root xylem vessels: a method and some evidence of control of the
pH of the xylem sap. Planta 162 361-369
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 Physiol 9 9
203-212
Dainty J (1985) Water transport through the root. Acta Hortic 171:
21-31
Fiscus EL (1975) The interaction between osmotic- and pressureinduced water flow in plant roots. Plant Physiol55 917-922
Frensch J, Steudle E (1989) Axial and radial hydraulic resistance to
roots of maize (Zea mays L.). Plant Physiol91: 719-726
Giordano R, Salleo A, Salleo S, Wanderlingh F (1978) Flow in
xylem vessels and Poiseuille’s law. Can J Bot 5 6 333-338
Greacen EL, Ponsana P, Barley KP (1976) Resistance to water flow
in the roots of cereals. In OL Lange, L Kappen, E-D Schulze, eds,
Ecological Studies, Vol 19: Water and Plant Life. Springer Verlag,
New York, pp 86-100
Haussling M, Jorns CA, Lehmbecker G, Hecht-Buchholz C, Marschner H (1988) Ion and water uptake in relation to root development in Norway spruce (Picea abies (L.) Karst.). J Plant Physiol
133: 486-491
Kedem O, Katchalsky A (1963a) Permeability of composite membranes. Part 2: Parallel elements. Transactions of the Faraday
Society 5 9 1931-1940
Kedem O, Katchalsky A (1963b) Permeability of composite membranes. Part 3: Series array of elements. Transactions of the Faraday Society 5 9 1941-1953
Landsberg JJ, Fowkes ND (1978) Water movement through plant
roots. Ann Bot 42: 493-508
Miller DM (1980) Studies of root function in Zea mays. I. Apparatus
and methods. Can J Bot 5 8 351-360
Miller DM (1985) Studies of root function in Zea mays. 111. 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 Physiol77: 162-167
Nonweiler TRF (1975) Flow of biological fluids through non-ideal
capillaries. In MH Zimmermann, JA Milbum, eds, Encyclopedia of
Plant Physiology, New Series, Vol 1:Transport in Plants, I: Phloem
Transport, Appendix I. Springer Verlag, New York, pp 474-477
North GB, Nobel PS (1991) Changes in hydraulic conductivity and
anatomy caused by drylng and rewetting roots of Agave deserti
(Agavaceae).Am J Bot 78: 906-915
Passioura JB (1988) Water transport in and to roots. Annu Rev Plant
Physiol Plant Mo1 Biol39 245-265
Perumalla CJ, Peterson CA (1986) Deposition of Casparian bands
and suberin lamellae in the exodermis and endodermis of young
com and onion roots. Can J Bot 6 4 1873-1878
Peterson CA (1989) Significance of the exodermis in root function.
In BC Loughman, O Gasparikova, J Kolek, eds, Structural and
Functional Aspects of Transport in Roots. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 35-40
Peterson CA, Emanuel ME, Humphreys GB (1981) Pathway of
movement of apoplastic fluorescent dye tracers through the en-
there was a steep inwardly directed gradient of turgor and
osmotic pressure. Astonishingly, this resulted in a constant
water potential of close to zero across the entire cortex, even
when the roots were subjected to 300 m NaCl (osmotic
pressure approximates 1.4 MPa). Zimmermann et al. (1992)
claim that this was due to complete exclusion of NaCl from
the apoplast of the young roots. Even more surprising, a zero
water potential in the cortex was found despite a high hydraulic conductivity of cortical cells that equilibrated with
half-times of about 14 s with their surroundings. When roots
were excised from the plants, gradients of turgor vanished
within 15 to 30 min. The authors claim that this indicates
that the use of excised roots for measuring water relations
should be treated with caution. However, they do not resolve
the inconsistencies in their own data that refer (a) to the
exclusion of salts from the apoplast of young roots; (b) to the
lack of any tendency of an equilibration of water potential of
the roots with the medium of high salinity; and (c)to the fact
that if excision would alter turgor at a given water potential;
the ionic composition of cortical cells would also change. This
means that the ionic concentration of root cells should rapidly
change upon excision. This has not yet been observed. Cell
pressure probe work indicated that there were no gradients
of cell turgor in the cortex of intact roots of wheat and maize
(Pritchard et al., 1989; Spollen and Sharp, 1991). Thus, the
results of Zimmermann et al. (1992) would need some verification. Contradictions need to be resolved.
In conclusion, the data presented in this paper show that
L, and Lp, may both vary substantially along a developing
root. The deposition of Casparian bands in the exodermis
and the formation of suberin lamellae in the endodermis and
exodermis appear to be very important. For a detailed modeling of the water uptake of roots, i.e. of their efficiency, the
pattern of changes of hydraulic conductivities along the roots
is crucial. Root pressure probes can be used to evaluate the
governing parameters at the resolution necessary to work out
the hydraulic architecture of root systems. The technique
would also allow measurement of. interactions between water
and solute (nutrient) flows. Using the cell pressure probe, the
analysis could be extended further from the tissue or organ
level to that of individual cells (Steudle et al., 1987; Zhu and
Steudle, 1991). It has been demonstrated that this is important since the hydraulic conductivity of cell membranes may
change in response to environmental changes (Azaizeh et al.
1992; Bimer and Steudle, 1993).
ACKNOWLEDCMENTS
The authors are indebted to Dr. C.A. Peterson and Ms. D.E.
Enstone for the introduction to the fluorescent staining technique.
We wish to thank A. Elend and M. Scherer, who were involved in
part of the anatomical work, and M. Lauerer and Dr. M. Paul for
carefully reading the manuscript. The expert technical assistance of
B. Stumpf is gratefully acknowledged.
Received September 8, 1992; accepted December 14, 1992.
Copyright Clearance Center: 0032-0889/93/101/1305/11.
-
Hydraulic Conductivity of Onion Roots
dodennis at the site of secondary root formation in com (Zea mays)
and broad bean (Vicia faba). Can J Bot 5 9 618-625
Peterson CA, Murrmann M, Steudle E (1993) Location of the major
bamers to water and ion movement in young roots of Zea mays L.
Planta (in press)
Peterson CA, Steudle E (1993) Lateral hydraulic conductivity of
early metaxylem vessels in Zea mays L. roots. Planta 1 8 9 288-297
Pitman MG, Wellfare D, Carter C (1981) Reduction of hydraulic
conductivity during inhibition of exudation from excised maize
and barley roots. Plant Physiol 67: 802-808
Preston RD (1938) The contents of the vessels of Fraxinus americana
L., with respect to the ascent of sap. Ann Bot 2: 1-21
Pritchard J, Williams G, Wyn Jones RG, Tomos AD (1989) Radial
turgor pressure profiles in growing and mature zones of wheat
roots-a
modification of the pressure probe. J Exp Bot 4 0
567-571
Robards AW, Jackson SM, Clarkson DT, Sanderson J (1973) The
structure of barley roots in relation to the transport of ions into
the stele. Protoplasma 77: 291-311
Rosene HF (1937) Distribution of the velocities of absorption of
water in the onion root. Plant Physiol12: 1-19
Rosene HF (1941a)Comparison of rates of water intake in contiguous
regions of intact and isolated roots. Plant Physioll6 19-38
Rosene HF (1941b) Control of water transport in local root regions
of attached and isolated roots by means of the osmotic pressure of
the extemal solution. Am J Bot 28: 402-410
Sanderson J (1983) Water uptake by different regions of the barley
root. Pathways of radial flow in relation to development of the
endodermis. J Exp Bot 3 4 240-253
Sanderson J, Whitbread FC, Fielding WJ (1982) Axial conductivity
in the xylem vessels of the seminal root of barley. In Agricultura1
Research Council Letcombe Lab Annual Report 1981, pp 70-72
Schambil F, Woermann D (1989) Radial transport of water across
cortical sleeves of excised roots of Zea mays L. Planta 178:
488-494
Spollen WG, Sharp RE (1991) Spatial distribution of turgor and root
growth at low water potentials. Plant Physiol96 438-443
1315
St Aubin G, Canny MJ, McCully ME (1986) Living vessel elements
in the late metaxylem of sheathed maize roots. Ann Bot 5 8
577-588
Steudle E (1992) The biophysics of plant water: compartmentation,
coupling with metabolic processes, and flow of water in plant
roots. In GN Somero, CB Osmond, CL Bolis, eds, Water and
Life: Comparative Analysis of Water Relationships at the Organismic, Cellular, and Molecular Levels. Springer Verlag, Berlin,
pp 173-204
Steudle E, Frensch J (1989) Osmotic responses of maize roots. Water
and solute relations. Planta 177: 281-295
Steudle E, Jeschke WD (1983) Water transport in barley roots.
Measurements of root pressure and hydraulic conductivity of roots
in parallel with turgor and hydraulic conductivity of root cells.
Planta 158: 237-248
Steudle E, Oren R, Schulze E-D (1987) Water transport in maize
roots. Measurement of hydraulic conductivity, solute permeability,
and of reflection coefficients of excised roots using the root pressure probe. Plant Physiol84 1220-1232
Weatherley PE (1982) Water uptake and flow in roots. In OL Lange,
IS
' Nobel, CB Osmond, H Ziegler, eds, Encyclopedia of Plant
Physiology, New Series, Vol 128: Physiological Plant Ecology, 11:
Water Relations and Carbon Assimilation. Springer Verlag,. New
York, pp 79-109
Zhu GL, Steudle E (1991) Water transport across maize roots.
Simultaneous measurement of flows at the cell and root leve1 by
double pressure probe technique. Plant Physiol95 305-315
Zimmermann MH, Brown CL (1971) Transport in the xylem. I n
Trees. Structure and Function. Springer Verlag, New York, pp
169-220
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. Pressure probe measurements and
nuclear magnetic resonance-imaging analysis. Plant Physiol 99:
186-196