Annals of Botany 64, 325-336, 1989
325
Different Aspects of Cavitation Resistance in Ceratonia siliqua, a Drought-Avoiding Mediterranean Tree S. SALLEO and M. A. LO GULLO Islituto di Botanica, Universita di Messina, via P. Castelli 2, 98100 Messina, Italy Accepted: 3 April 1989
ABSTRACT Potted plants of Ceratonia siliqua L., growing in a greenhouse, were used to detect xylem cavitation (in terms of ultrasound acoustic emissions AE) in internodes and node-to-petiole (N-P) junctions, after different periods of drought (9, 16 and 23 d). Diurnal AE were only 100 in internodes of watered (W) plants but 320, 1250 and 2460 in 9-, 16- and 23-d stressed ones. In N - P junctions, AE were only 15 to 20% with respect to internodes. Stem perfusion with dye allowed measurement of the percentage of xylem conduit transverse area blocked by cavitation. This was 2 % in internodes of W-plants and 5-2, 13-8 and 40-4% in those of 9-, 16and 23-d stressed ones. In N - P junctions, 18-5% of the xylem conduit transverse area was blocked in the 23-d stressed plants only. The major resistance to cavitation exhibited by the N - P junctions is interpreted in terms of their greater number of narrow xylem conduits. The percentage of blocked xylem conduits within a range of diameters showed that the narrower a xylem conduit, the less likely it was that cavitation would occur. After rewatering, the release of the xylem blockage caused by cavitation occurred within 2 h. Our data suggest that C. siliqua can be considered to be a cavitation avoider, especially in its stem-toleaf transition zones. Key words: Ceratonia siliqua L., Carob tree, cavitation avoidance, xylem architecture, ultrasonic acoustic
emissions. below 94% even in summer (Lo Gullo and Salleo, Ceratonia siliqua L. is one of the most typical components of the vegetation growing in the Mediterranean Basin region (Pignatti, 1982). In Sicily, the species can be found in very xeric and thermophilic zones at altitudes between 0 and 500 m. During the dry summer period (from May to September) rainfall is less than 80 mm and temperatures vary between 22-25 °C at night and 28 °C to more than 40 °C during the day. Previous studies (Lo Gullo and Salleo, 1988; Lo Gullo, Salleo and Rosso, 1986) have shown that C. siliqua withstands drought by avoiding water stress. This is made possible by extracting sufficient water from the soil rapidly enough to compensate for water loss. Drought-avoiding 'water spending' species (Levitt, 1980) are characterized by relatively high leaf conductances to water vapour (g : ) (about 4-5 mm s"1 in adult C. siliqua trees in the field) and very small changes in their leaf water content (RWC). C. siliqua diurnal RWC never dropped 0305-7364/89/090325+12 $03.00/0
Since stomata of C. siliqua are maintained open for many hours during the day, the species is likely to be exposed to the risk of water cavitation in its xylem conduits because of the high tensions which would develop in them. On the basis of preliminary cavitation measurements made in the field (unpublished), this seemed to be the case. On the other hand, it is very rare to observe damage to C. siliqua foliage as a consequence of summer drought. Therefore, some hypotheses-can be advanced about the eventual strategy adopted by C. siliqua to resist xylem cavitation. Hypothesis I. The species could have developed a xylem structure preventing cavitation (cavitation avoidance), e.g. numerous narrow and short xylem conduits which are less vulnerable to cavitation than wider and longer ones. A range of xylem conduit sizes could satisfy the species' demand of © 1989 Annals of Botany Company
326
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
Even when stems are cut off very carefully, however, manipulations can cause small air bubbles to be produced in the solution which enter the xylem and affect subsequent measurements. Moreover, Newbanks, Bosch and Zimmermann (1983) and Zimmermann (1983), using microcinematographic analysis, showed that dyes ascended through embolized xylem conduits by capillarity when they were put in contact with dyes at atmospheric pressure. Therefore, they proposed to perfuse twigs or intact plants with dyes at reduced pressure (10 kPa). The authors used a modified Ursprung (1913) apparatus. To overcome both dye ascent by capillarity and Hypothesis II. The species could tolerate the xylem embolism caused by manipulation, we used effects of the blockage of a given percentage of its a simple apparatus, similar to that described by xylem conduits (cavitation tolerance), e.g. because Zimmermann, but modified to allow the stem to be its xylem is overefficient or the cambium is very cut off with minimal manipulation (see Materials active in producing new water-conducting ele- and Methods). ments. Previous data (Salleo and Lo Gullo, 1989) The aim of the present study was to investigate: have shown that up to 30% of the water- (a) the time course of water cavitation in interconducting cross-sectional area of Vitis vinifera nodes (I) and node-to-petiole (N-P) junctions of could be lost due to cavitation without any visible C. siliqua plants subjected to a progressively severe damage to leaves. water stress; (b) the percentage of water-conducting cross-sectional area permanently lost at Hypothesis III. Xylem conduits could undergo different levels of water stress intensity; (c) the cavitation only transiently, i.e. water conduction relative vulnerability to cavitation of xylem concould be restored in them as soon as transpiration duits with different diameters; and (d) the time declines (in the presence of an adequate water taken by water stressed plants to stop cavitation supply). This is theoretically possible only if xylem after rewatering and the extent of the eventual conduits are water vapour-filled (Tyree and Dixon, recovery of their water-conducting cross-sectional 1986; Tyree et al., 1986). To the best of our area. knowledge, it is not known how much time is needed for air to enter a cavitated (i.e. water MATERIALS AND METHODS vapour-filled) xylem conduit and convert it to the embolized state (i.e. air-filled). Tyree and Dixon Studies were conducted on 40 three-year-old (1986) suggested that the transition from cavitated C. siliqua plants, about 120 cm tall, with about 25 to embolized state should occur over a period of leaves each. Plants were grown in pots in a minutes to hours. Crombie, Hipkins and Milburn greenhouse (Institute of Botany, University of (1985) reported, on the contrary, that cavitation Messina, Sicily). Nictemeral changes of temperaand embolism are simultaneous phenomena in ture and air relative humidity were adjusted to Rhododendron xylem conduits. Unfortunately, it is vary between 22 and 32 °C and between 30 and difficult to visualize both water vapour-filled and 65%, respectively. All the measurements were air-filled xylem conduits. Visual detection of the made between mid-April and June 1988. Ten xylem blockage induced by cavitation is usually plants were irrigated every 48 h with 200 ml water made using dye solutions. Other techniques (e.g. while the other three groups of 10 plants each were hydraulic conductivity measurements, Sperry, deprived of irrigation for 9, 16 and 23 d. Donnelly and Tyree, 1988), do not give information about the relative vulnerability to cavitation of single xylem conduits. This would be Water relations parameters useful information for crop plant selection purTo estimate the changes in the plant water poses. economy caused by the water stress applied, some To infiltrate xylem with dyes, stems are usually water relations parameters were measured hourly cut off under water and put in contact with dyes between 0800 and 2000 h: (a) leaf conductance to which are then pulled into them using a vacuum water vapour (gx); (b) leaf water potential ( ^ ) ; pump (Sperry, 1986) or by transpiration of leafy and (c) leaf relative water content (RWC). twigs (Salleo and Lo Gullo, 1989; Zimmermann, gj was measured using a steady-state porometer 1983). (LI-COR model 1600) on 20 leaflets offivedifferent safety and efficiency of water transport (Salleo and Lo Gullo, 1986, 1989; Tyree and Dixon, 1986; Zimmermann, 1978, 1982, 1983). In fact, narrower xylem conduits have been found in localized plant regions such as junctions where branches meet (Ikeda and Suzaki, 1984; Zimmermann, 1978, 1983; Zimmermann and Jeje, 1981), nodal regions (Salleo, Rosso and Lo Gullo, 1982) and leaf traces (Isebrands and Larson, 1977; Larson and Isebrands, 1978). Such 'constricted zones' have been interpreted as 'safety zones' preventing xylem conduit embolism (Salleo, Lo Gullo and Siracusano, 1984).
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree mature leaves each time (C. siliqua has compoundpinnate leaves). \lrl was measured at 20 °C on three mature leaves each time using a pressure bomb (Scholander et al., 1965) with a sheet of wet filter paper inside to minimize water loss during measurements. Immediately after i/r1 measurements, leaves were weighed within about 20 s and their f. wts (FW) recorded. Leaf weights at full turgor (TW) were measured after leaves had remained for 3 h in the dark, covered with plastic bags and their cut ends immersed in distilled water. Then \jrx was measured again to check that leaves were fully turgid (i.e. ijrl was near zero), but that no over-saturation with water had occurred. In fact, }fr1 was about —001 MPa. Leaves were then dried at 70 °C for 3 d, cooled in Petri dishes for 30 mins and reweighed to obtain their d. wt (DW). RWC was calculated by: RWC = F W - D W / T W - D W x 100. Measurements of leaf water potential isotherms at 20 °C and at different symplasmic water losses (Tyree and Hammel, 1972) were made for three mature leaves at the beginning of the experiments to measure the \jrl value at the turgor loss point in order to estimate the impact of the water stress applied. The procedure employed to measure the \jri isotherms is described in detail elsewhere (Lo Gullo et al., 1986; Salleo, 1983). Cavitation measurements Water cavitation in the xylem of C. siliqua plants was detected between 0800 and 2000 h in terms of ultrasound acoustic emissions (AE) (Dixon, Grace and Tyree, 1984; Salleo and Lo Gullo, 1986, 1989; Sandford and Grace, 1985; Tyree and Dixon, 1983, 1986; Tyree et al., 1984, 1986). AE were detected using two AE transducers (Bruel and Kjaer, model 8312) connected to a wideband conditioning amplifier (Bruel and Kjaer, model 2368) with a high pass filter set at 01 to 2 MHz. The number of AE was counted using a programmable AE counter (UltraSci, model AElc). The two transducers were clamped to the exposed wood of an internode and of the nearest N - P junction. The exposed wood was covered with a thin layer of silicon grease to prevent wood dehydration and secure a good contact between transducers and wood. AE were either recorded every 60 s for 5 mins alternatively from internodes and N - P junctions or continuously from one stem region (Salleo and Lo Gullo, 1986). Dye infiltration and anatomical measurements Figure 1 illustrates the experimental set-up designed to infiltrate stems with dye under reduced pressure. The apparatus consisted of a 2 cm thick-
327
walled plexiglas chamber with holes at the opposite sides through which the intact stem could be fitted tightly. The chamber was filled partly with 01 % Safranin solution, the stems were immersed at a depth of 3 cm and supported by a block. The leaves inserted on the stem at distances corresponding to the chamber length were cut off a week before the experiments so that new cork could be produced on the cut surfaces. Through the upper cover of the chamber, a vertical sliding axis was mounted, bearing a disposable razor blade. This allowed stems to be cut off under the dye solution without manipulating the stems directly and so disturbing the solution. The chamber was connected to a manometer and to a vacuum pump allowing the pressure inside the box to be set at desired values. Running the pump periodically, it was possible to maintain the pressure between 8 and 12 kPa. This procedure should prevent dye ascent through embolized xylem conduits by capillarity and reduce substantially the pressure gradient between the dye solution and the cavitated (water vapour-filled) xylem conduits. The vacuum pump was run immediately after the stem was cut and the desired pressure was reached within 2 mins. The cut stems remained in contact with the dye for 2 h during which a small brush, fixed on one side of the razor blade, was moved continually on the stem cut surface to remove air bubbles caused by the depressurization. Stems were then disconnected from the box and internodes and N - P junctions were razor crosssectioned between 8 and 10 cm from the cut surface. The number of red stained (i.e. efficient) xylem conduits was counted by light microscopy and their inside diameters were calculated as described previously (Lo Gullo and Salleo, 1988; Salleo et al., 1982; Salleo and Lo Gullo, 1986). Knowing the xylem conduit diameters, the efficient and non-efficient cross-sectional areas were calculated as £nr2 where r is the xylem conduit radius. The distance from the stem cut surface at which internodes and N - P junctions were studied was decided on the basis of preliminary data (Fig. 2). Irrigated and pre-stressed plants were perfused with Safranin as described above. Cut stems were then razor cross-sectioned, serially and each section observed by light microscopy to determine the percentage of unstained xylem conduits. In irrigated plants this was typically low at the cut surface (due to contact staining). At longer distances it increased until a plateau was reached at about 9 % which was maintained up to 18 cm stem length. The subsequent increase in the percentage of unstained xylem conduits indicated that the dye was no longer transported uniformly.
328
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
FIG. 1. Schematic view (not to scale) of the experimental set-up used to mark the blocked xylem conduit cross-area and to detect the ultrasound acoustic emissions (AE). The intact plant was fitted through the chamber. Insert 1 illustrates the chamber, partly filled with Safranin. The stem was cut off using the blade D fixed on the axis C. The infiltration occurred at reduced pressure, maintained at the desired value (displayed by the manometer B) by the vacuum pump A. The dye was drained through the valve E. Insert 2 shows the stem regions to which the AE transducers were clamped, i.e. internodes (I) and node-to-petiole (N-P) junctions (J).
A similar pattern of dye ascent was recorded in 23d stressed plants where the plateau was at about 40 % between 3-5 and 12-5 cm from the cut surface. I and (N-P) J were studied within this plateau. Rewatering experiments To measure the decline of cavitation and the eventual recovery of water conduction in cavitated xylem conduits, two plants out of each group of pre-stressed plants were rewatered the day after the experiments described above. They were also
mounted through the apparatus for dye infiltration. At different daytimes (between 1400 and 1030 h depending on the daytime at which AE began) plants were rewatered (through a hole made on the upper surface of the plastic pot) with sufficient water volumes to reach a soil water content of 0-37 g H 2 O g"1 soil. During plant rewatering experiments, gl and ijrl were measured as described above and AE were counted continuously. 2 h after rewatering, stems were cut off and put in contact with the dye.
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
329
100
8 £
6
8 10 12 14 16 18 20 22 Distance from the stem cut surface (cm)
24
26
28
30
FIG. 2. Percentages of xylem conduits appearing unstained (i.e. non-efficient) after perfusion with dye of cut stems of C. siliqua L. watered (W), 23-d stressed plants (23-d S). The plateaux indicate the stem lengths along which the dye was transported uniformly; all anatomical studies were conducted within these plateaux.
In N - P junctions, the number of cAE was significantly less than in internodes, being only The amount of the water stress applied can be about 20 in watered and 9-d stressed plants, 230 estimated from the diurnal changes in g, and ^ t of and 370 in 16- and 23-d stressed ones. Such irrigated plants with respect to the pre-stressed differences in the number of cAE cannot be ones (Fig. 3 A), g, decreased as water stress attributed to the greater number of xylem conduits increased, from about 1-4 mm s'1 in watered plants of internodes with respect to that of N - P junctions, to about 0-1 mm s"1 in 23-d stressed ones which is i.e. 784+ 16-82 vs. 178 + 3-25 (mean±s.e.m.) belikely to represent cuticular water loss only. \jrl not cause (Table 1 column 3) the ratio of cAE to the only decreased with the water stress applied but it number of xylem conduits per section, as calculated failed to recover in the afternoon in the two groups in the four different water stress conditions, was in of severely stressed plants (16 and 23 d). The all stressed plants significantly higher in internodes turgor loss point of C. siliqua leaves (at \jrl = than in N-P junctions (0-42 to 0-11 in 9-d stressed — 2-45 MPa, as indicated in Fig. 3B), was nearly plants, 1-58 to 1-29 in 16-d stressed plants and 315 reached in 9-d stressed plants already (at 1400 h) to 208 in the most severely stressed ones). while in 16- and 23-d stressed ones, }/r1 dropped Moreover (Table 1 column 4), the percentage of non-efficient xylem cross-sectional area (in terms below this point (Fig. 3 A). 2 In spite of these very severe water stress of Znr of all unstained xylem conduits with conditions, C. siliqua leaves appeared green and respect to the total water-conducting transverse healthy in all plant groups except for 23-d stressed area), was 2 % in internodes of irrigated plants, 5-2, 13-8 and 40-4% in internodes of 9-, 16- and 23plants where they began to fold. The number of cumulative acoustic emissions d stressed plants vs. 0, 0, 0, and 18-5 % in N-P (cAE) (Fig. 4) as recorded between 0800 and 2000 junctions of plants at the same water stress h was about 100 in internodes of watered plants; it intensities. increased to 320 in 9-d stressed plants and further The distribution of xylem conduit diameters in to 1250 and 2460 in 16- and 23-d stressed ones, the two plant regions studied was very different respectively. As expected, AE began to be recorded (Fig. 5). In internodes, xylem conduits with earlier and earlier in the morning as water stress diameters between 40 and 60 /im were about 14% intensity increased (at 1100 h in 9-d stressed plants, of the total number while narrow xylem conduits at 0930 and 0830 h in 16- and 23-d stressed ones, (between 10 and 20 /im in diameter) were 33 % and respectively). the narrowest ones (8 to 10/tm in diameter) were RESULTS
A
1-5
E
0-5
w
J 1
0
8.
^ - .
1
1
1 x
2 1
9d-S
1
1
i
i
i
X fr—••
I
1
i
i
i
i
i
i
i
i
|
1
I
1
-
•-,_• t
i
I6d-S
t
0 1 2
i
23d-S §'
V
0-5
§ a.
-
5o
I I
l
l
i
i
8
i
-
•
-5,
-
*np
tup
3 4 i
8
i
i
10
i
12
14
i
i
i
i
i
i
i
i
i
i
16
18
20
8
10
12
14
16
18
20
Hour of day (h)
10 W o - W/Wo(%)
FIG. 3. A, Diurnal time course of leaf conductance to water vapour (g,) and water potential (^,) in watered (W), 9d, 16-d and 23-d stressed plants (9-d S, 16-d S and 23-d S). Horizontal dashed lines, labelled as ^ t l p , yield the corresponding leaf water potential at the turgor loss point. Means are given plus or minus the standard deviation unless eclipsed by the symbol. B, Leaf water potential (ijr), osmotic potential (tjrn) and turgor pressure (Pt) of a typical C. siliqua L. leaf vs. the relative symplasmic water loss. Wo and W represent the leaf symplasmic water content at full turgor and after pressurization. respectively.
15
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree 30 Internodes
25-
o A • A
W 9d-S I6d-S 23d-S
A
A
A
A
A
•
•
•
•
A
20-
A A
15A j t •••
•
10-
A
« .
'
5-
A • A A * ,
• A
10
A
ft
12
A O
A
A
A
A
A
A
A
0
O
O
O
O
0
O
14
i
16
18
20
Hour of day (h)
FIG. 4. Diurnal time course of cumulative acoustic emissions (AE) recorded in internodes and node-topetiole (N-P) junctions of watered (W), 9-d, 16-d and 23d stressed plants (9-d S, 16-d S and 23-d S). only 0-5%. In N - P junctions, on the contrary, 20 % of the xylem conduits were as narrow as 8 to 10 fim and 6 3 % were between 10 and 20 fim in diameter. No xylem conduit wider than 30 fim in diameter was found in N - P junctions. In other words, over 80 % of xylem conduits in the N - P junctions of C. siliqua was less than 20 /im in diameter while in the adjacent internodes about 70 % of the xylem conduits were wider. The relative vulnerability to cavitation of xylem conduits with different diameters was estimated on the basis of the percentages of unstained (i.e. nonefficient) xylem conduits within the different diameter intervals (Table 2 columns 4 and 7). For this purpose, only internodes of 16- and 23-d stressed plants have been considered and data from all the experiments have been pooled. Narrow xylem conduits (10 to 20 fim in diameter) were very resistant to cavitation (only 3-3
331
and 13-7% of them were blocked in 16- and 23-d stressed plants). These percentages increased with the xylem conduit diameter up to 31 and 50% of the xylem conduits between 50 and 60 fim in diameter and up to 44 and 100% of the widest xylem conduits ( > 60 fim in diameter) in 16- and 23-d stressed plants, respectively. When pre-stressed plants were rewatered (Fig. 6), gj and \jrl appeared to increase very quickly (within a matter of minutes). In particular, ijrl as recorded in 16- and 23-d stressed plants increased over the turgor loss point (Fig. 6 H and K.) and at 1900 h it was not much lower than that recorded in irrigated plants. After rewatering, AE stopped almost instantaneously both in 9- and in 16-d stressed plants while in 23-d stressed ones AE decreased significantly in number but they really stopped only after 3 h 15 mins. In other words, C. siliqua plants appeared to be very responsive to rewatering: although gj remained lower than in irrigated plants, xjr^ recovered quickly and also cavitation stopped. In terms of recovery of the efficient xylem conduit transverse area of internodes (Table 3 column 4), our data show that it was complete only in 9-d stressed plants while it was partial in 16-d stressed plants. Here, the percentage of nonconducting xylem transverse area decreased from 13-8 to 8-7%, i.e. of about 1/3, 2 h after the plants had been rewatered. No recovery was detected in 23-d stressed plants either in internodes or N-P junctions. DISCUSSION A significant number of AE was counted in C. siliqua internodes at the daytime when \jrl fell to the turgor loss point (i.e. to ^ , = —2-45 MPa) (Fig. 6E, F). Water deprivations of 16 and 23 d caused \jrl to drop below this point (down to — 3-5 and —3-95 MPa, respectively). In spite of the very severe water stress to which plants were subjected, the number of cAE as recorded between 0800 and 2000 h was only 2460 in the 23-d stressed plants. This was much less than that recorded in more. sensitive species like V. vinifera (over 4000) (Salleo and Lo Gullo, 1989) or Chorisia insignis (about 9000) (Salleo and Lo Gullo, 1986), under similar physiological conditions. Therefore, we can take C. siliqua as a cavitation-resistant species. Significantly lower numbers of cAE were counted in N - P junctions (about 15 to 20 %) with respect to those recorded in the nearest internodes. Accordingly, the percentage of the blocked xylem transverse area was much less in N - P junctions
332
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
T A B L E 1. Mean ratio of the cumulative acoustic emissions (AE) recorded between 0800 and 1900 h to the number of conduits per section (column 3) in internodes (I) and node-to-peptiole (N-P) junctions of Ceratonia siliqua L. watered (W), 9-d, 16-d and 23-d stressed plants (9-d S, 16-d S and 23-d S). Crosssectional blocked xylem area (column 4) as the sum of square radii of the blocked (i.e. unstained) xylem conduits (r|) as a percentage of the total xylem cross-sectional area. Cumulative AE Water stress
Stem region
Conduit No
W
I N-P.J I N-P.J I N-P.J I N-P,J
013 ±002 011+002 10-42+ 002 011+002 * 1-58+003 1-29 + 004 f 315 + 009 208+009
9-d S 16-d S 23-d S
1Ewr*, % 202 000 5-20 000 13-79 000 40-41 ±2-28 18-46+1-05
Means are given plus or minus the s.e. of the mean. Lines are drawn connecting values significantly different from each other (n = 4).
M-P Junctions
i
8 10 20 30 40 50 60 70
8
f
i
i
10 20 30 40 50 60 70
Xylem conduit diameters (^itn) FIG. 5. Distribution of xylem conduit diameters for internodes and node-to-petiole (N-P) junctions.
than in internodes (0 vs. 13.8% in 16-d stressed plants and 18-5% vs. 40-4% in 23-d stressed ones). Such data further support the view of the transition zones between stems and leaves as 'protective regions' against water cavitation. In our opinion, this is due to the narrower diameters
of xylem conduits in these plant regions. Here, in fact, the analysis of the distribution of the xylem conduit diametres showed that over 80 % of xylem conduits were less than 20 fim in diameter while in internodes, over 40 % of them were more than 30 /im in diameter. The percentages of xylem conduits
333
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
T A B L E 2. Number of xylem conduits {columns 2 and 5) belonging to different diameter intervals {column 1). Number of blocked {i.e. unstained) xylem conduits {columns 3 and 6) and percentages of blocked xylem conduits vs. their total number {columns 4 and 7) in 16-d and 23-d stressed {16-d S and 23-d S) plants of Ceratonia siliqua L. 16-d S Conduit diameter (jim)
< 10 10-0-20 201-30 30-1-40 401-50 501-60 >60
Conduit no.
Blocked conduit no.
10 1037 818 863 387 48 9
0 34 110 104 70 15 4
23-d S
%
Conduit no.
Blocked conduit no.
%
000 3-28 13-45 1205 1808 31-25 44-44
12 1012 797 840 377 46 6
1 138 137 346 190 23 6
8-33 13-66 21-71 4119 50-40 5000 10000
appearing blocked within the different diameter intervals (Table 3) indicated that the narrower a xylem conduit, the safer (only 3 % of xylem conduits 10 to 20 fim in diameter were blocked in 16-d stressed plants vs. 30 % in conduits between 50 and 60 fim and 40 % in those wider than 60 /tm in diameter). This functional interpretation was first advanced by Zimmermann (1978, 1982, 1983), Zimmerman and Brown (1974), Isebrands and Larson, (1977) and more recently by Zimmerman and Jeje (1981), Ikeda & Suzaki (1984), Tyree and Dixon (1986) (at least within one species) and by us (Salleo and Lo Gullo, 1986, 1989; Salleo et al., 1982, 1984). Such xylem architecture, based on a more efficient (and vulnerable) water-conducting system in the stem and zones with opposite features localized at petiole junctions, is likely to protect leaves against the blockage of water transport into them, caused by the very negative pressure potentials occurring in the petiole xylem conduits under water stress conditions. It seems, however, that once cavitation is initiated in petioles, xylem conduits are irreversibly lost. In fact, 23-d stressed plants lost their leaves after about 35 d from the beginning of the experiments although they were rewatered at the 24th day. Our data would support hypothesis I (see above), i.e. a kind of overall cavitation avoidance strategy adopted by C. siliqua involving xylem conduits with reduced diameters which would allow the species to resist cavitation much better than others with wider xylem conduits (e.g. V. vinifera or Chorisia insignis). Localized plant regions like the transition zones between stem and petioles would be well protected against cavitation by narrower (and probably shorter) xylem conduits which are
capable of resisting greater negative pressure potentials without cavitating. On the other hand, C. siliqua 1-year-old internodes lost about 14% of their water-conducting cross-sectional area without any visible damage to leaves (Table 1). This would mean that a loss of water flow of jlhis order of magnitude could be tolerated by the species. At least within these limits, therefore, C. siliqua can be regarded as a cavitation tolerating species (hypothesis II, see above). The prompt recovery of \j/l in 9- and 16-d stressed plants after rewatering, suggests that water was transported very rapidly to leaves through the intact xylem conduits of stems and petioles (the majority of which were still efficient). Also AE stopped within a few minutes of rewatering. This indicates that the species is capable of recovering from water stress very rapidly and this is probably what happens in the field during the night when air humidity (which is often over 6 5 % in Sicily) condenses on the soil (Salleo, 1983). Two hours after rewatering, 100% of the xylem conduit blockage was released in 9-d stressed plants while about 3 3 % was released in 16-d stressed ones. It is possible that xylem conduits refill with sap so rapidly because they are still in the cavitated state (i.e. water vapour-filled) and not yet in the embolized state (i.e. air-filled). In 23-d stressed plants, \jfy also recovered, but AE continued to be produced (although at a lower rate) for over 3 h. In terms of water-conducting transverse area, no recovery was recorded at 2 h after rewatering in this plant group. In our opinion, the ^ , recovery (which occurred only partly until AE were produced and was completed later, Fig. 6K), was due to the water transport through the
9d-S
W
% £
1-0
^
0-5
I6d-S
23d-S
o 0.
I A A A
25 20 18 16 14 12 10 8 6 4 2 0
§'
AA A A A A A A
f 8
-A-A-
10
A A A A A A A A
12
14
16
18
A A A
20
8
10
A
A A A A I
A A A S A A A A A
12
14
16
18 Hour
20
A AJ
8
10
A A A A A A A .
12
14
16
18
20
8
.
»
•
10
•
•
12
of day ( h )
FIG. 6. Diurnal time course of leaf conductance to water vapour (g,), water potential (0-,) and cumulative acoustic emissions (AE) of watered (W), 9-d, 16-d and 23-d stressed plants (9-d S, 16-d S and 23-d S). Arrows indicate the daytime at which pre-stressed plants were rewatered; open symbols indicate the diurnal time course of g,, tjrl and AE before and after rewatering. Horizontal dashed lines as in Fig. 3 A.
•
•
14
16
18
20
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree
335
T A B L E 3. Percentages of blocked xylem crossLITERATURE CITED sectional area (Zm2B) in internodes (I) and node-toCROMBIE, D. S., HIPKINS, M. F. and MILBURN, J. A., petiole (N-P) junctions of watered (W), 9-d, 16-d 1985. Gas penetration of pit membranes in the and 23-d stressed plants (9-d S, 16-d S and 23-d S) xylem of Rhododendron as the cause of acoustically (column 3) and after rewatering (Re W) (column 4) detectable sap cavitation. Australian Journal of Plant Physiology 12, 445-53. Water Stem DIXON, M. A., GRACE, J. and TYREE, M. T., 1984. stress region ReW £m\(%) Concurrent measurements of stem density, leaf and r«S(%) stem water potential, stomatal conductance and cavitation on a sapling of Thuja occidentals L. I W 202 000 Plant, Cell and Environment 7, 615-8. N-P 000 000 IKEDA, T. and SUAZKI, T., 1984. Distribution of xylem I 5-20 9-d S 000 resistance to water flow in stems and branches of N-P 000 000 hardwood species. Journal of the Japanese Forestry I 13-79 16-dS 8-72 Society 66, 229-36. N-P 000 000 1 ISEBRANDS, J. G. and LARSON, P. R., 1977. Vascular 23-d S t40-41±2-28 f43-20 + 2-42 anatomy of the nodal regions in Populus deltoides N-P 18-46+105 1900+112 Bartr. American Journal of Botany 64, 1066-77. LARSON, P. R. and ISEBRANDS, J. G., 1978. Functional
Means are given plus or minus the s.e. of the mean. Lines are drawn connecting values significantly different from each other, (n = 4).
significance of the nodal constricted zone in Populus deltoides Bartr. Canadian Journal of Botany 56, 801-4. LEVITT, J., 1980. Responses of Plants lo Environmental Stresses. Academic Press, New York.
60% xylem transverse area still functioning. The seemingly permanent loss of 40% xylem crossarea suggests that at very high water stress intensities, the transition from the cavitated to the embolized state in xylem conduits is very rapid (or xylem cavitation and embolism are really simultaneous phenomena as suggested by Crombie et al., 1985). In other words, we think that under mild water stress conditions, xylem cavitation in C. siliqua can be regarded as a transient phenomenon which probably occurs with nictemeral rhythm during the summer period: beyond given limits of water stress intensity, it would cause plastic and no longer elastic water strain. C. siliqua was shown to be a drought-avoiding 'water spending' species (Lo Gullo and Salleo, 1988): to adopt such a water stress resistance strategy, the species should assure an efficient water supply to its leaves. The high resistance to cavitation of its overall water-conducting system, and especially of that from stem to leaves, would contribute to the efficiency of water transport, at least in terms of its continuity. Nonetheless, leaves are typical 'disposable' organs: when water stress intensity exceeds given limits, some of them would fall and transpiration decline proportionally. It would be of great interest to investigate further the limits of water stress at which plants suffer permanent damage and must, therefore, spend energy to produce new xylem, as well as the xylem conduit diameter interval that allows the best equilibrium between efficiency and safety of water transport in Mediterranean plants.
Lo GULLO, M. A. and SALLEO, S., 1988. Different
strategies of drought resistance in three Mediterranean sclerophyllous trees growing in the same environmental conditions. New Phytologist 108, 267-76. and Rosso, R., 1986. Drought avoidance strategy in Ceralonia siliqua L., a mesomorphicleaved tree in the xeric Mediterranean area. Annals of Botany 58, 745-56. NEWBANKS, D., BOSCH, A. and ZIMMERMANN, M. H.,
1983. Evidence for xylem dysfunction by embolization in Dutch Elm disease. Phytopathology 73, 1060-3. PIGNATTI, S., 1982. Flora d'ltalia, p. 625. Edagricole, Bologna. SALLEO, S., 1983. Water relations parameters of two Sicilian species of Senecio (Groundsel) measured by the pressure bomb technique. New Phytologist 95, 179-88. and Lo GULLO, M. A., 1986. Xylem cavitation in nodes and internodes of whole Chorisia insignis H. B. et K. plants subjected to water stress: relations between xylem conduit size and cavitation. Annals of Botany 58,431-41. 1989. Xylem cavitation in nodes and internodes of Vitis vinifera L. plants subjected to water stress. Limits of restoration of water conduction in cavitated xylem conduits. In Structural and functional responses to environmental stresses: water shortage (Proceedings of the XlVth International Botanical Congress, Berlin (West), Germany, 1987), eds. K. H. Kreeb, H. Richter and T. M. Hinckley, (in press). and SIRACUSANO, L., 1984. Distribution of
vessel ends in stems of some diffuse- and ringporous trees: the nodal regions as 'safety zones' of the water conducting system. Annals of Botany 54, 543-52.
336
Salleo and Lo Gullo—Cavitation Resistance in Carob Tree Rosso, R. and Lo GULLO, M. A., 1982. Hydraulic architecture of Vitis vinifera L. and Populus deltoides Bartr. 1-year-old twigs. II. The nodal regions as 'constricted zones' of the xylem system. Giornale Botanico Italiano 116, 15—27.
SANDFORD, A. P. and GRACE, J., 1985. The measurement
and interpretation of ultrasounds from woody stems. Journal of Experimental Botany 36, 298-311. SCHOLANDER, P. F., HAMMEL, H. T., BRADSTREET, E. D . and HEMMINGSEN, E. A., 1965. Sap pressures in
vascular plants. Science 148, 339-46. SPERRY, J. S., 1986. Relationship of xylem embolism to xylem pressure potential, stomatal closure and shoot morphology in the palm Rhapis excelsa. Plant Physiology 80, 110-15. DONNELLY, J. R. and TYREE, M. T., 1988. A method
for measuring hydraulic conductivity and embolism in xylem. Plant, Cell and Environment 11, 35-40. TYREE, M. T. and DIXON, M. A., 1983. Cavitation events
in Thuja occidentalis L. 2. Ultrasonic acoustic emissions from sapwood can be measured. Plant Physiology 72, 1094-9. 1986. Water stress induced cavitation and embolism in some woody plants. Physiologia Plantarum 66, 397-405. and HAMMEL, H. T., 1972. The measurement of the
turgor pressure and water relations of plants by the
pressure bomb technique. Journal of Experimental Botany 23, 267-82. DIXON, M. A., TYREE, E. L. and JOHNSON, R., 1984.
Ultrasonic acoustic emissions from the sapwood of Cedar and Hemlock. An examination of three hypotheses regarding cavitation. Plant Physiology 75, 988-92. Fiscus, E. L., WULLSCHLEGER, S. D. and DIXON,
M. A., 1986. Detection of xylem cavitation in corn under field conditions. Plant Physiology 82, 597-9. URSPRUNG, A., 1913. Ueber die Bedeutung der Kohaesion fuer das Saftsteigen. Berichte der Deutschen Botanische Gesellschafl 31, 401-12. ZIMMERMANN, M. H., 1978. Hydraulic architecture of some diffuse-porous trees. Canadian Journal of Botany 56, 2286-95. 1982. Functional xylem anatomy of angiosperm trees, pp. 59-70. In New Perspectives in Wood Anatomy, ed. Pieter Baas. The Hague, The Netherlands. 1983. Xylem Structure and the Ascent of Sap, ed. T. E. Timell, Springer Verlag, Berlin. and BROWN, C. L., 1974. Trees: Structure and Function, pp. 169-182. Springer Verlag, Berlin. and JEJE, A. A., 1981. Vessel-length distribution in stems of some American woody plants. Canadian Journal of Botany 59, 1882-92.
