Trees (2001) 15:472–482 DOI 10.1007/s00468-001-0131-9

O R I G I N A L A RT I C L E

Sabine Rosner · Peter Baier · Silvia B. Kikuta

Osmotic potential of Norway spruce [Picea abies (L.) Karst.] secondary phloem in relation to anatomy

Received: 8 August 2001 / Accepted: 19 September 2001 / Published online: 10 November 2001 © Springer-Verlag 2001

Abstract Variations in water status of secondary phloem of Picea abies (L.) Karst., caused by (1) radial and (2) vertical differences within the tree, (3) seasonal influences, and (4) tree class, and their relation to bark anatomy were investigated. The water status parameters measured were the osmotic potential at full saturation [Ψo (sat)], the in situ osmotic potential [Ψo (in situ)], the in situ water content (Cin situ), and the in situ relative water content (Rin situ). Ψo (sat) reached most negative values in the conducting part of the secondary phloem, whereas Ψo (in situ) was similar in conducting (PC) and non-conducting secondary phloem (PN). The remarkable discontinuity in the radial course of Cin situ and Ψo (sat) at the transition from PC to PN can be attributed to the degeneration of sieve cells and Strasburger cells. In PC, the vertical decrease of Ψo (sat) towards the crown was compensated by an increase in Rin situ, so that Ψo (in situ) did not change along the stem. With stem height, Ψo (sat) decreased and PC width increased. The determining factor for vertical gradients in Ψo (sat) was the distance to the sources; similar gradients were also measured in PN. Seasonal differences in Ψo (sat) could only be detected in PC, where they corresponded exactly to changes of Ψo (sat) in needles. Suppressed trees showed less negative Ψo (sat) values in PC, smaller annual secondary phloem increments and smaller radial lumen diameter of living sieve cells than predominant or dominant trees. Keywords Osmotic potential · Picea abies (L.) Karst. · Potential gradients · Relative water content · Secondary phloem S. Rosner (✉) · S.B. Kikuta Institute of Botany, University of Agricultural Sciences, Gregor Mendel-Strasse 33, 1180 Vienna, Austria e-mail: [email protected] Fax: +43-1-476543180 P. Baier Institute of Forest Entomology, Forest Pathology and Forest Protection, University of Agricultural Sciences, Hasenauerstrasse 38, 1190 Vienna, Austria

Introduction Growth conditions, including water, mineral and light supply, influence the width both of secondary phloem and wood annual rings of trees (Holdheide 1951). Suppressed trees show not only reduced tree height and diameter, but also narrower secondary phloem increments compared to dominant trees (Bäucker et al. 1998; Rosner 1998). The main driving force for cell enlargement is the turgor pressure, which depends on the osmotic potential of the cells (Hsiao 1973; Turner and Jones 1980). Cell growth starts when loosening processes cause cell walls to relax. This reduces cell turgor and induces a water potential gradient between the cell interior and the apoplast (Cosgrove 1993). Cell volume is increased by water movement caused by the potential gradient. For sustained growth, solutes must continuously be imported into growing cells so that osmotic potential does not increase due to inflowing water. Solute import assures maintenance of the water potential gradient that will sustain inward water movement (Kozlowsky and Pallardy 1997). Measurement of osmotic potential of leaf tissues is a standard procedure for characterising the water status of a tree. It is therefore surprising that data for osmotic potential in secondary phloem are scarce in the literature, probably due to methodological problems. The term “secondary phloem” comprises both conducting and non-conducting tissue. In Picea abies (L.) Karst., conducting secondary phloem is only a narrow zone (0.2–0.5 mm) adjacent to the vascular cambium (Huber 1939; Rosner 1998), where sieve cells are alive and thus able to transport assimilates. Osmotic potentials of secondary phloem of deciduous tree species were mostly measured to verify the mass flow theory of Münch (1930; Pfeiffer 1937; Kaufmann and Kramer 1967; Hammel 1968; Rogers and Peel 1975). The mass flow or pressure-flow hypothesis states that assimilates are transported passively via the sieve elements of the phloem along a gradient of turgor pressure. This gradient is established osmotically, mainly by

473 Table 1 Characterisation of study sites and harvested trees Study site

Height a.s.l. (m)

E

N

Tree age (years)

Average tree height (m)

Time of sampling

Number of trees

Murau Manhartsberg Rosalia

1200 500 600

14°10′ 15°46′ 16°17′

47°05′ 48°33′ 47°42′

85–90 95–100 75–80

22.0 28.2 24.0

300

14°50′

48°05′

20

15.0

8 August 1995 18 August 1995 30 May 1996 22 July 1996 28 September 1999

6 6 3 3 1

Amstetten

different concentrations of sucrose. Sugar concentration is high (and osmotic potential low) in the conduits where synthesis of assimilates takes place or where stored carbohydrates are mobilised (source); it is low (and osmotic potential high) where assimilates are utilised in respiration, growth, or storage (sink) (Esau 1969). A turgor gradient can only develop where the gradient of osmotic potential is steeper than the gradient of water potential (Kaufmann and Kramer 1967). Loading and unloading of sieve cells is an active, metabolically controlled process (Spanner 1975; Cronshaw 1981; Gamalei 1991; Van Bel and Gamalei 1992; Evert et al. 1996). In Norway spruce phloem (un)loading and radial transport processes seem to proceed mainly by the symplastic pathway (Langenfeld-Heyser 1987; BlechschmidtSchneider 1990; Blechschmidt-Schneider et al. 1997). Preconditions for the accurate measurement of osmotic potentials of secondary phloem are (1) the exact definition of the sample position and (2) an adequate method for determining osmotic potential at full saturation [Ψo (sat)]. Vertical gradients of osmotic potential in bark tissues of P. abies were investigated by Kraemer (1953) and Merker (1956). Since the authors did not consider phloem anatomy, interpretation of their data is difficult. Agedependent changes in secondary phloem, such as degeneration of sieve cells and Strasburger cells, dilatation (plus increase of secondary resin canal diameter), differentiation of sclerenchyma cells and formation of calcium oxalate crystals may cause drastic radial differences in water status. Osmotic potential gradients in radial direction have been investigated in deciduous trees, but with different methods (e.g. press saps, incision, aphid stylets; Pfeiffer 1937; Rogers and Peel 1975). Up to now, only in situ osmotic potential [Ψo (in situ)], influenced by fluctuating water content, has been measured in secondary phloem. Determination of osmotic potential at a standardised relative water content, e.g. at full saturation [Ψo (sat)], is absolutely essential for investigating vertical or radial gradients, seasonal changes, and variations within one tree species or between different species. Furthermore, important information on the development of potentials in symplastic solutions can be obtained by measurement of corresponding water content and relative water content. The relative water content is the actual water content of a tissue as a fraction of the water in the fully hydrated tissue and therefore an indirect estimation of the turgor pressure.

The purpose of this study was a detailed investigation of various water status parameters [in situ osmotic potential, osmotic potential at full saturation, in situ water content (Cin situ), water content at full saturation (Csat), in situ relative water content (Rin situ)] in secondary phloem of P. abies and their relation to bark anatomy. First, it was necessary to define representative sites for sampling with regard to morphological changes in radial direction. Second, we examined vertical gradients in defined tissue regions. Third, variations in water status caused by the tree’s social class were investigated. Correlation between the different water status parameters should improve the analysis of their development. The results should provide basic knowledge for further investigations, for instance on water stress, and on seasonal and site-specific differences in water status. Investigations on the relation between water status and anatomy may help to explain the different “productivity” within and between trees.

Materials and methods Plant material Samples of secondary phloem were taken from Norway spruce trees [P. abies (L.) Karst.] growing at various sites in Austria: Murau, where occurrence is autochthonous; and Manhartsberg, Rosalia and Amstetten, where occurrence is allochthonous. For further characterisation of study sites and trees, see Table 1. Trees were classified after Kraft (1884) as predominant, dominant, codominant, dominated and suppressed. All trees were harvested at noon. Samples for histological and physiological investigations were taken at three heights: at 3/10 length of the tree (3/10 h), at the base of the crown (B) and from the middle part of the crown (M), or at breast height (1.3 m). Microscopy Bark plus one to two of the latest annual increments of wood were punched out of the stem with a chisel. The samples had a tangential surface area of about 0.7–1.0 cm2. Samples were fixed in FAA (formalin–acetic acid–alcohol, Brown 1964) for at least 24 h. The fixed material was run to 50% alcohol and cut into 30-µm transverse sections with a sliding microtome (Reichert, Vienna). Sections were stained with astrablue/safranin, dehydrated, and mounted in Euparal (Gerlach 1984). The width of conducting secondary phloem and the latest annual phloem increment were measured under a binocular microscope (Fig. 1a; Wild Leitz, Heerbrugg, Switzerland). If the latest increment was not completely developed, the increment of the year before was measured, as long as the sieve cells were still open and not collapsed. Additionally, the number of sieve cell rows of the latest completely developed increment was determined under a light microscope (Fig. 1b, c). The

474

475 mean value was computed from 10 sieve cell row counts, excluding the cells of the parenchyma cell band. The radial diameter of living sieve cells was measured with a light microscope connected to a Macintosh computer via a CCD camera using the NIH-Image analysis system (Wayne Rasband, National Institute of Health, USA). For micrographs of secondary phloem, the fixed material was dehydrated in ethanol and embedded in Technovit 7100 (Heraeus Kulzer, Wehrheim, Germany). Semi-thin sections (1–2 µm) were made on a Reichert Jung Rotocut microtome (model 1165, Nußloch, Germany) using Kulzer Histoknive H (Heraeus Kulzer). Sections were stained with toluidine blue (Riedel-de Haen, Seelze, Germany) (Gerlach 1984) and mounted in Entellan (Merck, Darmstadt, Germany). Tree-vigour indices were measured on wood discs sampled at breast height. The tree-vigour index was defined as the incremental area of 1 year ring divided by the sapwood area multiplied by 100 (Waring et al. 1980; Führer et al. 1997). The percent increment of the previous 10 years was obtained. Water status parameters of secondary phloem The following water status parameters of secondary phloem were measured: Cin situ, Csat, Rin situ, Ψo (in situ), and Ψo (sat). In situ denotes values of parameters at the moment of sample collection. Secondary phloem samples were either taken immediately after harvesting, or spruce logs (60 cm length) were stored in a cold room (4°C, 80% r.h.) until sample collection. Preliminary measurements showed that storing spruce logs in a cold room for at least 4 days would not result in significant changes between tree harvesting and sampling in secondary phloem Ψo (sat), the water status parameter very often used for comparisons within and between trees. Water status parameters were determined on tissue discs. With this method, different water status parameters, such as (relative) water content and Ψo at different water contents, can be obtained on the same sample. The in situ osmotic potential is measured on saturated, frozen discs that were dehydrated to initial in situ fresh weight on the micro-balance prior to measurements (Rosner 1998). Preparation of discs

▲

A core of bark plus one or two annual increments of wood was punched out with a cork-borer of 6 mm diameter. To avoid water Fig. 1a–d Transverse sections of secondary phloem. a Bark plus one almost differentiated increment of wood of a suppressed spruce tree (M3/2). Sample was taken in the middle of the crown on 8 August 1995 in Murau; bar 500 µm. Conducting secondary phloem consists of two annual increments with open sieve cell rows. The arrow marks the vascular cambium. b, c Transition zone between conducting and non-conducting secondary phloem of a suppressed (b 7 June 1997) and a dominant (c 4 August 1997) tree. The arrow marks the border between latest and penultimate annual phloem increment. In the dominant tree, one increment consisted of about 10 sieve cell rows; two parenchyma cell bands (dark filling caused by polyphenols) were produced each year. In the suppressed tree, one increment consisted of about 4 sieve cell rows. Only one parenchyma cell band (dark filling caused by polyphenols) was produced each year. d Secondary resin canal in the non-conducting secondary phloem of Picea abies (18 May 1996). Sieve cells are collapsed and parenchyma cells (dark filling caused by polyphenols) are inflated. The light spots in the parenchyma cells are starch grains. Samples for b–d were taken at breast height of 50-year-old trees in Kindberg (Styria, Austria); bar 150 µm. (CO cortex, EW earlywood, LW latewood, PC conducting secondary phloem, PN non-conducting secondary phloem, PRC primary resin canal, VC vascular cambium)

loss during transport to the laboratory the cores were tightly wrapped in Parafilm (American National Can, Greenwich) and stored in a styrofoam box kept cool with ice packs. Starting from the vascular cambium, discs were cut tangentially from the core with a scalpel. On the disc cut next to the cambium the water status of the conducting secondary phloem, consisting of uncollapsed sieve cells and parenchyma cells, was determined. The thickness of the discs had to be chosen according to the specimen and varied from 0.2 to 0.5 mm, since in P. abies the width of the conducting secondary phloem depends on genetics, growth conditions, and the physiological age of the bark (Holdheide 1951; Bäucker et al. 1998; Franceschi et al. 1998; Rosner 1998). By cutting the discs under a binocular microscope it was possible to distinguish between vascular cambium, conducting and non-conducting secondary phloem (Fig. 1a). About 0.5-mm-thick discs were also cut from the central part of the non-conducting secondary phloem 1.5–2.5 mm distant from the vascular cambium (Fig. 1d).

Determination of Rin situ, and Ψo (sat) and Ψo (in situ) After determining the in situ fresh weight (F) on a micro-balance (ME/BE22, resolution 10–3 mg, Mettler, Zürich, Switzerland), the discs were saturated with distilled water. A fully saturated polyurethane foam sheet with holes 6 mm in diameter was put into a small plastic box (7×3 cm) with a tightly fitting lid. The discs were placed into the holes, thus providing direct contact between the foam sheet and the edges of the discs. After saturation (5 min for conducting secondary phloem, 10 min for non-conducting secondary phloem) the discs were removed from the box and weighed again to obtain the saturated weight (S). Non-conducting secondary phloem required longer saturation because of changes occurring in the tissue at the end of its functional period, such as collapse of sieve cells, dilatation growth, sclerification of parenchyma cells, and accumulation of calcium oxalate crystals (Esau 1969). Water loss from the discs was kept at a minimum by performing all manipulation steps very rapidly. The short saturation times used were sufficient to reach full saturation (Rosner 1998). Saturated discs were put in 0.7 ml micro-tubes (Roth, Karlsruhe, Germany), killed in a repeated freeze-thaw cycle (Kikuta and Richter 1992) to eliminate turgor potential, and stored at –20°C. After weighing, discs were transferred immediately into the sample holder of the Wescor 5500 Vapor Pressure Osmometer (Logan, Utah, USA), and the Ψo (sat) was measured. Time for weighing and transfer to the osmometer did not exceed 5 s. For the determination of Ψo (in situ) discs were transferred, after Ψo (sat) measurement, from the osmometer to the micro-balance and allowed to dehydrate to the in situ fresh weight (F) plus a few tenths of a milligram. The potential was measured, and the sample was weighed again. The mean between the weights before and after the Ψo measurement should correspond to the initial in situ fresh weight. A number of samples were weighed before and after the measurement of the potential. Thus, sufficient information on the weight losses during Ψo measurements was provided. The weight to which the samples had to be dehydrated was calculated as: (Weight of disc before Ψo measurement +Weight of discs after Ψo measurement)/2=In situ weight

(1)

The potential measured is a reasonable approximation to Ψo (in situ), since it is not possible to obtain reliable values for in situ osmotic potential with the conventional disc method where measurements are done on discs frozen without prior saturation. Water loss during manipulation steps, such as weighing and killing or transfer to vials and osmometer, lowers osmotic potentials considerably (Rosner 1998). The osmolality values (mosmol/kg) obtained were converted into osmotic potentials at 25°C by means of factors derived from tables published by Lang (1967).

476 Discs were dried to constant weight for 12 h at 85°C (Litvay and McKimmy 1975). The F and S values, and the dry weight (D) were used to calculate the following parameters: (2) (3) (4) Percentage values were obtained by multiplying Cin situ, Csat, and Rin situ by 100. Osmotic potential of fully saturated needles Shoots with needles of the last two age classes were saturated for 24 h in distilled water. After carefully drying the shoots with filter paper, needles were severed with a razor blade and put in plastic vials. The vials were immersed in liquid nitrogen (–196°C) and stored at –20°C. Needle press saps were obtained with a hydraulic laboratory press (Kikuta and Richter 1992). Ψo (sat) values were calculated as the mean value of two measurements. Statistical analyses Statistical data analysis was carried out with SPSS 7.5.1 for Windows software. Mean values of normally distributed data were tested for significance of differences either with the t-test for independent samples or analysis of variance and subsequent Duncan’s test. Associations between two variables were examined using linear regression analyses. Differences or correlations were accepted as significant if P was ≤0.05, * P≤0.05, ** P≤0.01 and *** P≤0.001. In tables, significant differences are marked with different letters. In the text, mean values that showed no significant differences are underlined (for instance: A B C means that group A and B are not significantly different, whereas A and C, and B and C are significantly different).

Results Radial gradients in the secondary phloem For correct measurements of water relations of secondary phloem it is absolutely necessary to define the position for sample collection exactly. Cin situ reached the highest and Ψo (sat) the most negative values in the narrow zone adjacent to the vascular cambium. This area corresponds to the conducting secondary phloem. The transition zone to the non-conducting secondary phloem is marked by a clear discontinuity in the radial courses of Cin situ (Fig. 2a) as well as Ψo (sat) (Fig. 2b). For Ψo (sat), no significant differences could be detected in the non-conducting zone farther than 1.25 mm from the vascular cambium. Cin situ was not significantly different between 1.75 and 3.25 mm from the vascular cambium. Thus, for measurements of water relations parameters of secondary phloem we recommend two positions for sample collection: first, the conducting secondary phloem (PC) and second, the middle part of the non-conducting secondary phloem (PN) at 1.5–2.5 mm distance from the vascular cambium. Both sites are suitable for comparisons. Trees from the Rosalia site served for a more detailed investigation into the differences between PC and PN. Cin situ was significantly higher in all samples of PC (Table 2). Rin situ reached lowest

Fig. 2 Radial gradients of the in situ water content (a) and the osmotic potential at full saturation (b) of secondary phloem at 0.5 m from the ground (n=4). Distance from vascular cambium was measured to 1/2 of the disc’s width. Secondary phloem had a width of about 4 mm. Samples were taken from a 15-m-high tree harvested on 28 September 1999 in Amstetten

values in PN. For the samples collected at the end of July, the radial difference in Rin situ was pronounced. In both investigation periods (May and July), Ψo (sat) values were more negative in PC, whereas the water contents corresponding to the measurements of the potentials (Csat) were significantly higher than in the non-conducting zone. The most striking result was that Ψo (in situ) did not differ significantly between PC and PN. In particular, trees harvested at the end of May had very similar Ψo (in situ) values at all heights investigated. Seasonal changes in osmotic potential were detected only in PC; Ψo (sat) was significantly less negative at all heights at the end of May (3/10 h, B and M: P≤0.05, n=3, Table 2) than at the end of July. This result corresponds well to the seasonal changes of Ψo (sat) in spruce needles. Ψo (sat) of the last needle age class (1996) decreased significantly from –1.19±0.09 in May to –2.26±0.21 MPa in July (P≤0.001, n=3), Ψo (sat) of the penultimate class (1995) from –1.71±0,09 MPa to –2.11±0.16 MPa (P≤0.05, n=3). Vertical gradients in the secondary phloem Osmotic potential at full saturation of PC from trees harvested at Manhartsberg clearly changed along a vertical gradient. Numbers Mb11 and Mb12 showed significant changes in Ψo (sat) along the stem (P≤0.05, Fig. 3). The

477 Table 2 Mean values±SD of the width of conducting secondary phloem (PC), annual phloem increment, number of annually produced sieve cell rows, in situ water content (Cin situ), in situ relative water content (Rin situ), in situ osmotic potential [Ψo (in situ)], water content at full saturation (Csat) and osmotic potential at full saturation [Ψo (sat)]. Samples were taken at 3/10 tree height (3/10 h), at the base of the crown (B) and in the middle part of the Date

Relative Sample Width height PC (µm)

May 30 3/10 h

July 22

PC PN

B

PC PN

M

PC PN

3/10 h

PC PN

B

PC PN

M

PC PN

Phloem increment (µm)

crown (M) from predominant, dominant and codominant trees harvested at Rosalia (R1–R3, 30 May 1996 and R7–R9, 22 July 1996). Average age of trees was 75 years, average height 23 m. Significant differences between conducting secondary phloem (PC) and non-conducting secondary phloem (PN) are marked with different letters (t-test, n=3)

Sieve Cin situ (%) cell rows (no.)

272.5±66.0 170.7±11.9 8.8±1.3

Rin situ (%)

Ψo in situ (MPa)

Csat (%)

Ψo sat (MPa)

302.9±22.8 b 167.1±14.7 a P≤0.001 363.7±23.9 203.7±18.7 10.0±1.7 304.9±23.5 b 172.3±7.0 a P≤0.001 439.1±32.3 251.4±24.3 13.0±1.0 363.0±69.3 b 174.3±16.0 a P≤0.05

67.1±5.2 a –0.84±0.04 a 457.8±39.6 b 60.7±3.9 a –0.81±0.06 a 275.4±13.2 a P≤0.01 69.7±4.3 a –0.90±0.08 a 441.7±49.3 b 68.1±5.6 a –0.83±0.04 a 254.4±12.7 a P≤0.01 83.2±2.3 b –0.90±0.05 a 431.5±63.1 b 66.6±9.2 a –0.91±0.00 a 263.2±25.3 a P≤0.05 P≤0.01

–0.77±0.05 a –0.61±0.08 b P≤0.05

292.4±28.7 168.0±16.0 7.0±0.0

80.0±4.5 b 55.7±7.7 a P≤0.01 87.2±4.3 b 58.6±5.7 a P≤0.01 88.9±4.0 b 63.9±10.1 a P≤0.01

–0.85±0.10 a –0.52±0.02 b P≤0.05 –0.93±0.14 a –0.50±0.04 b P≤0.05 –0.97±0.07 a –0.60±0.05 b P≤0.01

286.5±42.6 b 139.8±17.7 a P≤0.01 357.2±41.2 201.0±39.1 9.0±2.6 358.9±59.3 b 139.9±12.4 a P≤0.05 474.9±4.7 255.3±29.8 12.7±1.1 353.2±22.2 b 155.2±19.8 a P≤0.001

–1.10±0.10 a 356.0±39.9 b –0.91±0.10 a 250.1±10.4 a P≤0.01 –1.07±0.18 a 421.2±35.4 b –0.95±0.11 a 257.8±15.6 a P≤0.01 –1.08±0.08 a 398.0±18.9 b –0.97±0.05 a 255.2±11.0 a P≤0.001

–0.59±0.02 a –0.49±0.03 b P≤0.01 –0.63±0.09 a –0.57±0.07 a

Fig. 3 Vertical gradients of Ψo (sat) in conducting secondary phloem of two predominant (Mb11, Mb12) and two dominant (Mb21, Mb22) trees harvested at Manhartsberg. Samples were taken at 3/10 tree height (3/10 h), at the base of the crown (B) and in the middle part of the crown (M); n=2–4

Fig. 4 Vertical gradients of Ψo (sat) (open symbols) and Ψo (in situ) (black symbols) in non-conducting secondary phloem of six predominant, dominant, and codominant trees harvested on 30 May (R1–R3) and 22 July 1996 (R7–R9) at the Rosalia site, n=3. Samples were taken at 3/10 tree height (●), at the base of the crown (■) and in the middle part of the crown (▲)

mean value of the four trees in 3/10 h, –0.67±0.02 MPa, decreased to –0.77±0.05 MPa in B, and to –0.86± 0.19 MPa in M. Similar results were obtained on trees from the Rosalia site. At the end of May, the mean values of Ψo (sat) of conducting secondary phloem at M were significantly more negative than at B and 3/10 h (P≤0.05, Table 2). In contrast, values of Ψo (in situ) did not change. This was probably caused by the significant decrease of Rin situ down the stem (P≤0.01, n=3), which led to a passive decrease of osmotic potential and compensated for the increase in Ψo (sat) (Table 2).

Vertical gradients of Ψo (sat) were also detected in the PN. At the end of July, mean values of Ψo (sat) were significantly less negative at 3/10 h and B than at M (P≤0.05, n=3), whereas in May only slight vertical differences were observed (Table 2). Ψo (in situ) was significantly more negative at M than at B or 3/10 h in trees harvested at the end of May (P≤0.05, n=3). In contrast to PC, vertical changes of Rin situ were not significant in PN. Analyses of single trees showed that in only one tree (R8) Ψo (sat) did not change significantly along the stem (Fig. 4). With the exception of trees R2 and R9, no significant vertical changes in Ψo (in situ) were observed

478 Table 3 Mean values of the radial lumen diameter of living sieve cells in conducting secondary phloem at 3/10 tree height (3/10 h), at the base of the crown (B) and in the middle part of the crown (M). Trees M11–32 were harvested in Murau, Mb11–32 at Manhartsberg and R1–9 at Rosalia. Significant differences between different relative heights are marked with different letters (Duncan’s-test, n=20)

Tree sample M11 M12 M21 M22 M31 M32 Mb11 Mb12 Mb21 Mb22 Mb31 Mb32 R1 R2 R3 R7 R8 R9

Lumen 3/10 h (µm)

Lumen B (µm)

Lumen M (µm)

P

14.30±2.50 a 16.74±3.27 a 21.43±5.67 a 17.57±4.65 a 16.30±3.63 a 16.89±1.48 a 17.74±3.18 a 20.24±3.56 a 21.12±3.33 a 22.32±5.02 a 20.20±3.23 a 20.88±4.08 a

26.92±5.32 28.47±2.97 22.08±2.89 20.34±3.49 19.01±3.83 18.45±3.41 20.11±3.19 b 24.16±4.08 b 22.52±4.66 a 23.28±5.04 b 18.44±2.00 b 20.34±4.35 b 21.51±2.83 b 20.52±3.05 a 24.37±4.32 b 24.49±3.11 a 23.07±5.19 ab 22.93±4.91 ab

20.92±2.58 b 26.99±3.58 c 20.75±2.97 a 24.51±4.61 b 18.59±2.88 b 21.93±3.13 b 19.39±3.76 a 23.98±4.78 b 23.02±3.14 ab 24.64±4.72 a 24.13±5.01 b 24.85±4.07 b

<0.001 <0.001

(Fig. 4), which was probably caused by the small increase of Rin situ along the stem. Vertical gradients of Ψo (sat) corresponded to age-dependent changes in the width of the most recent secondary phloem increments, and therefore to the width of PC. Trees from the Rosalia site produced 7.9±1.3 sieve cell rows per year at 3/10 h, 9.5±2.1 at B, and 12.8±1.0 at M (3/10 h B M, P≤0.001, n=6). This led to a significant increase of the most recent secondary phloem increments towards the crown, from 169.3±12.7 µm at 3/10 h to 202.4±27.5 µm at B and to 253.3±24.4 µm at M (3/10 h B M, P≤0.001, n=6, Table 2). The correlation between annual phloem increment and number of sieve cell rows was very close (r=0.93***, n=18). The radial lumen diameter of sieve cells in the conducting secondary phloem increased in almost all investigated trees significantly towards the crown (Table 3). At the end of May and in July, the conducting zone consisted of two almost complete secondary phloem increments. In May, the last increment was not fully completed, in July some sieve cell rows of the penultimate increment had collapsed. The width of conducting secondary phloem was very similar for both investigation periods (Table 2). Thickness increased from 282.4± 46.8 µm at 3/10 h to 360.4±30.4 µm at B, and 457.0± 28.5 µm at M (3/10 h B M, P≤0.001, n=6), whereas values of Ψo (sat) became more negative with increasing tree height. For trees harvested at the Rosalia site in May, a negative linear regression between Ψo (sat) and the width of PC with r=–0.80** and a quadratic regression with r=–0.87** (n=9) was calculated. In predominant and dominant trees from Manhartsberg, Ψo (sat) was also very closely correlated to the thickness of PC (linear: r=–0.96***, quadratic: r=–0.98***, n=12, Fig. 5): Ψo (sat) became less negative with decreasing width of PC.

<0.001 <0.05 <0.001 <0.01 <0.01 <0.05 <0.05 <0.05

Fig. 5 Correlation between the osmotic potential at full saturation [Ψo (sat)] and the width (width PC) of conducting secondary phloem of two predominant [Mb11 (● ● ), Mb12 (▲ ▲ )] and two dominant [Mb21 (●), Mb22 (▲)] trees harvested at Manhartsberg. The quadratic equation was: Ψo (sat) [MPa] = –522.6–1674.0 × width PC–646.5 × (width PC)2 (µm), r=–0.98***, n=12

Osmotic potential and tree class Tree class had a conspicuous influence on the osmotic potentials at full saturation and on secondary phloem increments. In suppressed trees from the Murau and Manhartsberg sites, the annual secondary phloem increment consisted of only 5–7 sieve cell rows, which were less than 150 µm in width at the base of the crown (Fig. 6). Predominant and dominant trees formed 8–13 sieve cell rows per year; the annual increment at the base of crown was therefore more than 180 µm. The radial lumen diameter of the living sieve cells was significantly smaller in suppressed trees M31 and M32 than in dominant trees M11 and M12 harvested in Murau (Table 3, Duncan’s test, P≤0.001). Ψo (sat) of PC correlated negatively with the latest phloem increment, which was at that time (August) fully completed. Despite the small number of trees investigat-

479 Table 4 Correlation between the tree-vigour indices (TVI) and the latest annual secondary phloem increment from trees harvested at Manhartsberg and Rosalia. TVI10 is the relative sapwood increment of the last 10 years, TVI5 of the last 5 years, TVIB of the previous year and TVIA of the year of harvesting (n=6)

Fig. 6 Correlation between Ψo (sat) of conducting secondary phloem and the width of the latest phloem increment at the base of the crown of predominant (p), dominant (d), and suppressed (s) trees harvested in Murau (filled symbols, linear equation: r=–0.90**, n=6) and at Manhartsberg (open symbols, linear equation: r=–0.92**, n=6)

Site

Relative height

TVI10

TVI5

TVIB

TVIA

Manhartsberg

3/10 h B M

0.90** 0.90* 0.81*

0.75 0.66 0.52

0.77 0.69 0.54

0.76 0.69 0.54

Rosalia

3/10 h B M

0.86* 0.94** 0.82*

0.79 0.90* 0.81*

0.70 0.75 0.69

indices of the last 5 years or of the previous year showed weaker correlation (Table 4, Fig. 7).

Discussion

Fig. 7 Tree-vigour indices of the last 10 years before harvesting (1995) of trees from the Manhartsberg (a) and Rosalia (b) sites

ed, correlation coefficients for quadratic regressions were very high: r=–0.98** for Murau (n=6) and –0.95* for Manhartsberg (n=6). PC of suppressed trees also showed lower relative water contents. (Pre)dominant spruce trees from Murau had an average Rin situ of 85.8±2.4% (n=4) at the base of the crown; suppressed trees had only 69.4±4.9%. Correlations between annual phloem increments and the mean tree-vigour indices of the last 10 years were very strong for trees harvested at the Manhartsberg and Rosalia sites. Annual phloem increments and tree-vigour

The radial differences in water relation parameters observed in P. abies can be explained by age-dependent structural changes of secondary phloem. In spruce, secondary phloem consists of sieve cells, which form the axial transport system for the assimilates, and of parenchyma cells. Radial transport of assimilates is carried out by parenchyma cells arranged into rays. Sieve cells and ray parenchyma cells are connected by Strasburger cells (Esau 1969). Some rays may contain secondary resin canals, schizogenous ducts lined by epithelial cells. Depending on the physiological age of the bark, the individual constitution of the tree, and general growth conditions, 5–13 new rows of sieve cells are differentiated each year (Huber 1939; Holdheide 1951; Bäucker et al. 1998; Rosner 1998). The latest annual increment of phloem in the trees studied was in the range described by Huber (1939) and Holdheide (1951) (0.1–0.3 mm). Usually, only one single-rowed tangential band of axial parenchyma cells is formed each year (Holdheide 1951; Huber 1961). However, in conditions favouring growth or depending on the tree’s genetics, two bands may be formed, one early, the other later in the season (Figs. 2, 3; Huber 1939; Franceschi et al. 1998). Sieve cells usually stay alive for two seasons [conducting secondary phloem (PC)], before they stop assimilate transport and collapse (Fig. 1a–d). Apart from the collapse of sieve cells, Esau (1969) described another three phenomena bringing about structural differences in secondary phloem. These are (1) dilatation growth resulting from enlargement and division of parenchyma cells, (2) sclerification, that is, development of lignified secondary walls in parenchyma cells, and (3) accumulation of calcium oxalate crystals. In conifers, the enlargement and proliferation of axial parenchyma cells represent the main form of dilatation and may continue for many years (Holdheide 1951; Esau 1969). In older parts of the non-conducting secondary phloem (PN) secondary resin canals enlarge to a great extent, so that they become rather conspicuous.

480

The remarkable discontinuity in the radial course of Cin situ and Ψo (sat) near the vascular cambium (Fig. 2) can be attributed to a drastic structural change in the secondary phloem: the breakdown (degeneration) of sieve cells and Strasburger cells. PC turns into PN (Fig. 1b, c). The continuous decrease of Cin situ (Fig. 2a) can be explained by a gradual increase in dry weight, which is caused by the presence of sclerenchyma cells and formation of calcium oxalate crystals. In addition, decrease in Cin situ could be influenced by the enlargement of secondary resin canals due to dilatation growth. Although Ψo (sat) reached its most negative values in the conducting part of the secondary phloem, Ψo (in situ) was similar in PC and PN. Whereas comparable values of Ψo (in situ) in PN were mainly caused by passive concentration of osmotically active substances, in cells of PC they were also reached by active processes. Pfeiffer (1937) observed a positive gradient of Ψo (in situ) in secondary phloem of Castanea vesca in August/September. Starting from the living sieve tubes, Ψo (in situ) became less negative outwards. Ψo (in situ) values of sieve tubes were measured on sap obtained after incision, those of PN on press saps. According to Rogers and Peel (1975), Ψo (in situ) of sieve tube sap of 2- to 3-year-old Salix viminalis trees obtained with severed aphid stylets was more negative than that of press sap of whole secondary phloem. Since data were obtained with different methods, these papers do not prove that symplastic solutions of the sieve elements have more negative Ψo (in situ) values than of parenchyma cells in PC and PN. It may have been the application of different methods that caused these results. The methods used have various sources of error, and the measured values may be either too low or too high. First, evaporative loss, which lowers Ψo passively, may occur in the small quantities of sieve tube sap obtained by aphid stylets (Rogers and Peel 1975) or incision. Second, measurements of osmotic potential on press sap may lead to less negative values; measurements on tissue samples turned out to deliver the more correct results (Kikuta and Richter 1992). The increase of Ψo (sat) from PC to PN (Fig. 2b, Table 2) indicates a radial decrease in the absolute mass of osmotically active substances. Whereas in living (intact) sieve cells sucrose is the most important osmotically active carbohydrate, older parts of P. abies secondary phloem also contain large amounts of mono- and trisaccharides (Ziegler 1975; Dünisch 1993). In P. abies, assimilates may also be stored as fats (Ziegler 1964) and starch. The polysaccharide starch is osmotically inactive. Stored in parenchyma and ray cells, the amount of starch is higher in PN than in PC. Mineral elements also function as osmolytes, with potassium playing an important role in loading and unloading of sieve elements (Spanner 1975). Dünisch et al. (1996) found that the content of potassium and phosphorus in the vascular cambium zone and the neighbouring PC zone of P. abies is higher than in the following older parts of the secondary phloem. Vertical gradients of osmotic potential in whole secondary phloem or sieve tubes of deciduous trees have been reported by many authors (Pfeiffer 1937;

Kaufmann and Kramer 1967; Hammel 1968; Rogers and Peel 1975). Similar gradients were also detected in conifers. According to Kraemer (1953) and Merker (1956), in situ osmotic potential of P. abies bark press saps becomes more negative towards the crown. During the summer months they measured about –0.7 MPa at the stem base, –1.0 MPa in the middle of the stem, and –1.2 MPa in the stem beneath the crown. The authors gave no exact definition for the site of sample collection. They obtained press saps from a mixture of tissues which they termed alternately “living bark” or “secondary phloem”, consisting probably of conducting secondary phloem, non-conducting secondary phloem and cortex (for terminology see Esau 1969; Trockenbrodt 1990). Our investigations showed that for in situ osmotic potential vertical gradients do not exist in all parts of the secondary phloem, whereas for osmotic potential at full saturation gradients do exist (Table 2, Figs. 3, 4). In PC, a vertical decrease of Ψo (sat) was superseded by an increase in Rin situ, so that Ψo (in situ) did not change along the stem. Ψo (sat) of PC became gradually more negative from 3/10 tree height to the middle part of the crown. The most negative values were measured in the needles from the upper crown. This indicates that the concentration of osmolytes, mainly sucrose, increases towards the source (the needles). According to Rogers and Peel (1975), in sieve cells and parenchyma cells concentrations of sucrose and potassium in sieve tubes of young S. viminalis were higher at the top of the stem than at the base. Ψo (in situ) of PN became more negative with tree height, because Rin situ showed smaller changes, compared to conducting secondary phloem. Rin situ and Cin situ of both conducting and non-conducting secondary phloem changed in a similar way along the stem (Table 2). An increase in water content of whole P. abies bark towards the crown was reported by Bednar (1975). Vertical differences of water status correspond to agedependent changes in the thickness of conducting secondary phloem (Fig. 5). The width of PC increases towards the crown, because the number of annually produced sieve cell rows is higher (Huber 1939; Bäucker et al. 1998), and the radial lumen diameter is larger in younger bark (Table 3). The production of more sieve cell rows in higher stem regions is favoured by a shorter distance to the sources. A greater width of PC in regions with a small stem diameter guarantees assimilate supply for lower regions with higher diameter and roots. Shorter distance to the sources enables more negative Ψo (sat), and thus higher turgor pressure. The increase of Rin situ in PC with stem height indicated an increase of turgor pressure along the stem. It is very likely that similar vertical gradients exist in the cambial region when new sieve cells are produced, causing a larger radial lumen diameter of the sieve cells in higher stem regions. Growth conditions of the tree influence the width of secondary phloem increments (Huber 1939; Holdheide 1951; Bäucker et al. 1998). The annual phloem increments from trees of different social classes correlated

481

well with Ψo (sat) of PC (Fig. 6). Less negative Ψo (sat) values indicate that suppressed trees had a lower concentration of osmotically active substances in the PC than predominant or dominant ones. Their smaller, shaded crowns produce fewer assimilates. Less negative Ψo (sat) only enables less turgor maintenance of the developing tissues (Turner and Jones 1980; Kozlowsky and Pallardy 1997) compared to dominant trees. The low turgor maintenance capacity in suppressed trees may result in a smaller radial lumen diameter of the sieve cells produced. In midsummer, less turgor pressure in PC of suppressed trees was indicated by their lower Rin situ. Strong correlation between the mean tree-vigour index of the last 10 years gave evidence that the recent annual secondary phloem increment is a reliable parameter for describing the vitality or individual growth capacity of the tree (Table 4, Fig. 7). Weaker correlation between tree-vigour index of the last 5 years, the previous or a recent year and the phloem increment showed that phloem increment is not as strongly influenced by climatic conditions as wood increment. Our study shows that care has to be taken to obtain comparable data on water status [Rin situ, Ψo (sat)] of secondary phloem. First, the position of sampling has to be defined exactly, because of radial and vertical gradients in the secondary phloem. Second, the trees studied should be of the same class, because this parameter has an enormous influence on water status. Third, secondary phloem shows seasonal differences in osmotic potential at full saturation which correspond to changes in saturated needles. Similar seasonal changes in needle Ψo (sat) have been reported for many conifer species (Tyree et al. 1978; Gross et al. 1980; Ritchie and Shula 1984; Colombo 1987; Colombo and Teng 1992; Baier 1993; Zine el Abidine et al. 1994). Acknowledgements This work was supported by Project P 10485Bio and the Special Research Program Forest Ecosystem Restoration (SF008), funded by the Austrian Science Foundation and the Austrian Ministry of Agriculture and Forestry. We thank Hanno Richter for helpful discussions and critical reading of the manuscript. Erwin Führer (Institute of Forest Entomology, Forest Pathology, and Forest Protection, BOKU, Vienna) is thanked for project coordination. We thank Birgit Kartusch for expert advice and for passing on her great knowledge of plant anatomy.

References Baier P (1993) Untersuchungen zur Indikation der Befallsdisposition der Fichte für rindenbrütende Stammschädlingsarten. Doctoral thesis, University of Agricultural Sciences, Vienna Bäucker E, Bues CT, Vogel M (1998) Radial growth dynamics of spruce (Picea abies) measured by micro-cores. IAWA J 19:301–309 Bednar H (1975) Vertikale Feuchteverteilung der Rinde bei Fichte und Tanne. Centralbl Ges Forstwes 92:129–138 Blechschmidt-Schneider S (1990) Phloem transport in Picea abies (L.) Karst. in mid-winter. I. Microautoradiographic studies on 14C-assimilate translocation in shoots. Trees 4:179–186 Blechschmidt-Schneider S, Eschrich W, Jahnke S (1997) Phloem loading, translocation and unloading processes. In: Rennenberg H, Eschrich W, Ziegler H (eds) Trees – contributions to modern tree physiology. Backhuys, Leiden, pp 139–163

Brown CL (1964) The influence of external pressure on the differentiation of cells and tissues in vitro. In: Zimmermann MH (ed) The formation of wood in forest trees. Academic Press, New York, pp 389–404 Colombo SJ (1987) Changes in osmotic potential, cell elasticity, and turgor relationships of 2nd-year black spruce container seedlings. Can J For Res 17:365–369 Colombo SJ, Teng Y (1992) Seasonal variation in the tissue water relations of Picea glauca. Oecologia 92:410–415 Cosgrove DJ (1993) How do plant cell walls extend? Plant Physiol 102:1–6 Cronshaw J (1981) Phloem structure and function. Annu Rev Plant Physiol 32:465–84 Dünisch O (1993) Über die Bedeutung der Elementversorgung von Fichten [Picea abies (L.) Karst.] für die Holzbildung. Doctoral thesis, University of Hamburg Dünisch O, Bauch J, Puls J, Müller M (1996) Biological and chemical wood properties of long-term polluted spruce [Picea abies (L.) Karst.] at high-altitude stands of the Erzgebirge. Holzforschung 50:497–506 Esau K (1969) The phloem. In: Zimmermann W, Ozenda P, Wulff HD (eds) Handbuch der Pflanzenanatomie. Gebrüder Borntraeger, Berlin Evert RF, Russin WA, Botha CEJ (1996) Distribution and frequency of plasmodesmata in relation to photoassimilate pathways and phloem loading in the barley leaf. Planta 198:572–9 Franceschi VR, Krekling T, Berryman AA, Christiansen E (1998) Specialized phloem parenchyma cells in Norway spruce (Pinaceae) bark are an important site of defense reactions. Am J Bot 85:601–615 Führer E, Lindenthal J, Baier P (1997) Baummortalität bei Fichte: Zusammenhänge zwischen prämortaler Vitalitätsdynamik und dem Befall durch rindenbrütende Insekten. Mitt Dtsch Ges Allg Angew Entomol 11:645–648 Gamalei Y (1991) Phloem loading and its development related to plant evolution from trees to herbs. Trees 5:50–64 Gerlach D (1984) Botanische Mikrotechnik. Thieme, Stuttgart Gross K, Pham-Nguyen T, Unger H (1980) Tägliche und saisonale Änderungen des Wasserpotentials und seiner Komponenten in den Kronen von Fichten unterschiedlichen Alters. Allg Forst Jagdztg 151:69–79 Hammel HT (1968) Measurement of turgor pressure and its gradient in the phloem of oak. Plant Physiol 43:1042–1048 Holdheide W (1951) Anatomie mitteleuropäischer Gehölzrinden (mit mikrophotographischem Atlas). In: Handbuch der Mikroskopie in der Technik, vol 5, part 1. Umschau, Frankfurt, pp 193–314 Hsiao TC (1973) Plant responses to water stress. Annu Rev Plant Physiol 24:519–570 Huber B (1939) Das Siebröhrensystem unserer Bäume und seine jahreszeitlichen Veränderungen. Jahrb Wiss Bot 88: 176–242 Huber B (1961) Grundzüge der Pflanzenanatomie. Springer, Berlin Göttingen Heidelberg Kaufmann MR, Kramer PJ (1967) Phloem water relations and translocation. Plant Physiol 42:191–194 Kikuta SB, Richter H (1992) Leaf discs or press saps? A comparison of techniques for the determination of osmotic potentials in freeze-thawed leaf material. J Exp Bot 43:1039–1044 Kozlowsky TT, Pallardy SG (1997) Growth control in woody plants. In: Mooney HA (ed) Academic Press, San Diego Kraemer GD (1953) Die kritischen Grenzen der Brutbaumdisposition für Borkenkäfer an Fichte (Picea excelsa L.). Z Ang Entomol 34:463–512 Kraft G (1884) Beiträge zur Lehre von den Durchforstungen, Schlagstellungen und Lichtungshieben. Verein dt forstl Versuchsanstalt, Hannover Lang ARG (1967) Osmotic coefficients and water potentials of sodium chloride solutions from 0 to 40°C. Aust J Chem 20:2017–2023 Langenfeld-Heyser R (1987) Distribution of leaf assimilates in the stem of Picea abies L. Trees 1:102–109

482 Litvay JD, McKimmy MD (1975) Determining moisture content and moisture sorption in douglas-fir bark. Wood Sci 7:284– 291 Merker E (1956). Der Widerstand von Fichten gegen Borkenkäferfraß. Allg Forst Jagdztg 127:129–45; 165–187. Münch E (1930) Die Stoffbewegungen in der Pflanze. Fischer, Jena Pfeiffer M (1937) Die Verteilung der osmotischen Werte im Baum im Hinblick auf die Münchsche Druckstromtheorie. Flora 132:1–47 Ritchie GA, Shula RG (1984) Seasonal changes of tissue-water relations in shoots and root systems of Douglas-fir seedlings. For Sci 30:538–548 Rogers S, Peel AJ (1975) Some evidence for the existence of turgor pressure gradients in sieve tubes of willow. Planta 126:259–267 Rosner S (1998) Vergleichende Untersuchungen der Anatomie und des Wasserhaushaltes der Rinde von Fichte [Picea abies (L.) Karst.]. Doctoral thesis, University of Agricultural Sciences, Vienna Spanner DC (1975) Electroosmotic flow. In: Zimmermann MH, Milburn JA (eds) Encyclopedia of plant physiology, vol 1. Springer, Berlin Heidelberg New York, pp 301–327

Trockenbrodt M (1990) Survey and discussion of the terminology used in bark anatomy. IAWA Bull n.s. 11:141–166 Turner NC, Jones MM (1980) Turgor maintenance by osmotic adjustment: a review and evaluation. In: Turner NC, Kramer PJ (eds) Adaptation of plants to water and high temperature stress. Wiley, New York, pp 87–103 Tyree MT, Cheung YNS, MacGregor ME, Talbot AJB (1978) The characteristics of seasonal and ontogenetic changes in the tissue–water relations of Acer, Populus, Tsuga, and Picea. Can J Bot 56:635–647 Van Bel AJE, Gamalei YV (1992) Ecophysiology of phloem loading in source leaves. Plant Cell Environ 15:265–270 Waring RH, Thies WG, Muscato D (1980) Stem growth per unit of leaf area: a measure of tree vigour. For Sci 26:112–117 Ziegler H (1964) Storage, mobilization and distribution of reserve material in trees. In: Zimmermann MH (ed) The formation of wood in forest trees. Academic Press, New York, pp 303–320 Ziegler H (1975) Nature of transported substances. In: Zimmermann MH, Milburn JA (eds) Encyclopedia of plant physiology, vol 1. Springer, Berlin Heidelberg New York, pp 59–100 Zine el Abidine A, Bernier PY, Plamondon, AP (1994) Water relations parameters of lowland and upland black spruce: seasonal variations and ecotypic differences. Can J For Res 24:587–593