The distribution of xylem hydraulic resistance in the fruiting truss of tomato

Transcription

1 Plant, Cell and Environment (2001) 24, The distribution of xylem hydraulic resistance in the fruiting truss of tomato M. MALONE & J. ANDREWS Horticulture Research International, Wellesbourne, Warwicks. CV35 9EF, UK ABSTRACT The distribution of hydraulic resistances in xylem throughout the pathway leading to the tomato fruit was investigated. Previous work had indicated that there were large resistances within the supporting sections of this pathway (the peduncle and pedicel), perhaps associated with interruptions in the xylem. These high resistances are believed to impede calcium flux into the fruit and thus impair fruit development. It is shown here that fruit on intact plants do not shrink detectably during drought, even when the drought is sufficient to cause marked shrinkage of leaves and visible wilting of the shoot. In explants, it is possible to induce back-flow from the fruit into the stem (probably via the xylem) but this flow is small and very slow. These observations support the view that there is a large hydraulic resistance in the pathway between fruit and stem. When pulses of water were made available within explants, by scorching of one leaflet, there was a rapid swelling of leaves and sepals. Such rapid fluxes indicate the presence of strong hydraulic (xylem) connections throughout the pathway between leaf and calyx. This shows that there are no significant hydraulic constrictions in the xylem proximal to the calyx. This finding is contrary to some previous conclusions but it is supported by experiments with dyes which showed continuous, functional xylem throughout the peduncle and pedicel. Calculations show that over 90% of the hydraulic resistance between stem and fruit must reside within the fruit pericarp. Implications for calcium nutrition are discussed. Key-words: Lycopersicon esculentum L. (Mill); hydraulic architecture; pedicel anatomy; xylem. INTRODUCTION The tomato fruit offers a good system for the study of plant transport: it is large, accessible, and it forms a strong sink at the end of a narrow transport pathway (Lee 1989). Various tissues within the fruit are capable of some photosynthesis (Hetherington, Smillie & Davies 1998) but the vast majority of the growing fruit s requirements, including all its water and minerals, must be supplied from the shoot via the Correspondence: M. Malone. Fax: ; mike.malone@hri.ac.uk fruit peduncle (truss stalk) and pedicel (fruit stalk; Ho, Grange & Picken 1987). The high sugar and low calcium content of the mature fruit indicate that most influx occurs via the phloem rather than the xylem. Ho et al. (1987) estimate that over 90% of the total volume of the fruit arrives via the phloem. The small residual contribution from xylem is consistent with the low rate of transpiration found in the major part of the fruit (the berry), which has a thick cuticle (Andrews et al. 2000) and no stomata (Johnson, Dixon & Lee 1992). There can be significant transpiration from the calyx (Ehret & Ho 1986; Araki, Kitano & Eguchi 1997). Xylem-borne flux to the fruit, although minor in volumetric terms, is important for fruit quality because it delivers virtually all the fruit s calcium (Adams & Ho 1993). Phloem sap contains minimal levels of free calcium (Raven 1977; Brauer et al. 1998). Calcium deficiency is a common problem in fruit crops throughout the world. In tomato, it causes a sporadic necrosis of the distal part of the fruit ( blossom-end rot, Adams & Ho 1993). Xylem water flux is also important in cracking in some types of tomato (Ohta et al. 1997) and it contributes to fruit growth rate. To assess these various roles of xylem flux, several studies have considered xylem anatomy in the tissues which support and supply the growing fruit: the peduncle and pedicel. Such studies have often reported restrictions or interruptions (Lang & Ryan 1994; Greenspan, Shackel & Matthews 1994). Lee (1989) summarized physiological evidence for a xylem restriction in the tomato pedicel, and identified a valve site at the knuckle, at which the xylem was visibly reduced. The knuckle is a prominent swelling found midway along the pedicel in most tomato varieties; it contains the future abscission zone of the fruit. Similarly, at certain stages of development in a range of soft fruit, gaps have been reported in the xylem leading into the fruit (Wolswinkel, Ammerlaan & Koerselman-Kooij 1999). Soft fruit of several species can show diurnal shrinkage. This is believed to reflect xylem-borne back-flow, as well as transpiration, from the fruit at times when shoot water potential, and xylem pressure, falls below that of the fruit (Tromp 1984; Lang & Thorpe 1989; Greenspan et al. 1994). Tomato fruit, by contrast, show little or no diurnal shrinkage (Ehret & Ho 1986; Johnson et al. 1992; Pearce, Grange & Hardwick 1993; Kitano et al. 1996). This is further evidence of an unusually high xylem resistance at some point between fruit and shoot Blackwell Science Ltd 565

2 566 M. Malone and J. Andrews Displacement transducers can provide non-destructive analysis of hydraulic architecture in plant tissues (Turquois & Malone 1996). These are here employed, together with dye tracing and histological approaches, to examine hydraulic restrictions in the xylem leading into the tomato fruit. MATERIALS AND METHODS Plant material Tomato plants (Lycopersicon esculentum L. [Mill] cv Counter) were grown in the glasshouse in 15 dm 3 pots of Levingtons no. 2 compost (Fisons plc, Ipswich, UK). When they reached about 2 m in height they were brought into the laboratory, placed under fluorescent lighting of about 100 mmol m -2 s -1 photosynthetic photon flux density (PPFD) measured at the height of the first fully expanded leaf and allowed to acclimatize for 2 d prior to experiment. A supplementary mercury vapour lamp, 250 W, was sometimes switched on at 1 m above the plants ( SL in Fig. 1). In some experiments, leaf and sepal thickness were measured on explants. The explants comprised one leaf (trimmed to 1 3 leaflets) and one truss (4 7 green fruit) connected by some 20 cm of stem. They were cut from mature glasshouse plants (cv Counter) on the morning of the experiment, and brought to the laboratory in polythene bags. The explant system is inherently more sensitive to small flows than the intact plant system. This is because explants have a limited hydraulic capacity, so that a given small flow causes a greater effect on tissue thickness than it would in the intact plant. Functional xylem connections within explants were probed by scorching one leaflet with a flame for about 3 s. This kills the cells of the leaflet and releases their water to the apoplast. From there the water can be sucked into the xylem, which is not damaged by scorching. The released water will then flow rapidly to any region that has good xylem connections with the scorched tissue, causing swelling there (Malone 1993, 1996). Swelling kinetics can thus be used to map functional xylem connections between tissues (Turquois & Malone 1996). Transducer measurements The thickness of leaves and sepals (calyx) was measured with displacement transducers (Turquois & Malone 1996). transducers were always included to help identify artefacts. These were exactly the same as the experimental transducers except that they contained no leaves. For fruit growth, displacement sensors based on lightemitting diodes (LDS; model Z4W-V25R; Omron Corp., Kyoto, Japan) were preferred because they can be applied at any angle and because they impose no mechanical contact which might damage the surface of the fruit (Kitano et al. 1996).The output from each LDS was applied to a precision resistor (RS Components, Corby, UK) and logged, usually at 15 s intervals, on an eight-channel analogue-todigital converter (ADC-16; Pico Technology, Cambridge, UK). The LDS was mounted on a plastic tray. The fruit was retained gently in the corner of the tray by a wide elastic band. The measuring beam from the LDS fell approximately normal to the equator of the fruit so as to measure fruit diameter. Tracer experiments LYCH (lucifer yellow CH; L-0259; Sigma-Aldrich, Poole, UK) was used with caution (see Results) to assess functional xylem connections through the pedicel and into the fruit. The apical 2 m was excised from mature glasshousegrown tomato plants and placed in the laboratory under supplementary fluorescent lighting (approximately 100 mmol m 2 s -1 PPFD). The cut end was placed in water. After several hours, freshly prepared LYCH solution (100 mg dm -3 ) was added to the cut end. At various times thereafter, the overhead fluorescent tubes were switched off briefly, and the intact, attached, fruit were photographed under UV light (GEC type MBW/U; 125 W, 365 nm; Marconi, Coventry, UK) using a camera positioned about 20 cm from the fruit. Exposure time was set at 8 s, and a high sensitivity film (Kodak GPZ; Kodak Ltd, Hemel Hempstead, UK) was used. In some of these experiments, the plants were steam girdled on the day prior to the experiment. This was done by fitting a cup-shaped collar around the stem and filling it with hot wax (80 C). By the following day, the girdled region was severely shrivelled. To improve image contrast and facilitate reproduction, the original fluorescent images were converted to negativeimage grey scales; dyed regions now appear dark against a light background. The original colour of some of these images can be seen from the cover of Plant Cell & Environment (2000) 23 (4). Xylem histology Sections of fruit and pedicel were cut using a cryostat microtome (Bright Ltd, Huntingdon, UK) at about 10 mm thickness. The specimen was first supported with an embedding compound (Tissue-Tek; Agar Scientific, Stansted, UK) then frozen at 30 C. Some material was stained for lignin after sectioning, using 0 05% auramine O-SO 2 (O. Mattsson, University of Copenhagen, personal communication). All experiments were repeated at least three times. RESULTS Xylem flow Fruit diameter and leaf thickness was measured on large intact plants in the laboratory. An example is shown in which two cycles of drought were imposed over a period of several days (Fig. 1). The shoot became visibly wilted during the drought periods, and leaf thickness declined markedly. Leaf thickness provides a sensitive measure of

3 Tomato hydraulics SL on SL off Fruit 1 Fruit 2 Scorch Adjacent Leaflet Calyx (berry removed) Calyx (berry present) 2 h Scale: 100 mm (fruit) 25 mm (leaf) Leaf Figure 1. Fruit growth and leaf thickness in a tomato plant subjected to drought. Leaf thickness (thicker lines) was monitored over a period of 3 d, on one leaf of a 2 5-m tall tomato plant. Simultaneously, diameter was measured on fruit from two different trusses. Fruit 1 and 2 were 25 and 18 mm in diameter, respectively. Water was added to the compost at the times indicated by the dashed lines, in the amounts (cm 3 ) indicated on the figure. The light and dark periods are indicated (lower). At the times indicated by the solid lines, a supplementary lamp above the plants was switched on ( SL on ), then off. At the times indicated by the two asterisks, the shoot became visibly wilted. shoot water status (Malone 1992). Fruit growth was inhibited by drought but there was negligible fruit shrinkage even at times of maximal drought and severe leaf shrinkage (Fig. 1). When drought was relieved by adding aliquots of water to the compost, as shown in Fig. 1, leaf thickness recovered rapidly. Sometimes this was matched by strong increases in fruit diameter but at other times it was not. We did not observe the sharp decreases in fruit diameter reported by Kitano et al. (1996) on re-watering of tomato. The thickness of leaflets and sepals was monitored on explants from mature tomato plants (Fig. 2). Brief scorching of one leaflet with a flame was used to mobilize water within the explant. The water thus released will enter the xylem and flow rapidly to any region that has good xylem connections with the scorched tissue (Turquois & Malone 1996). The ensuing pattern of tissue swelling can thus reveal functional xylem connections within the explant. Figure 2 shows that within seconds of scorching one leaflet, thickness increased rapidly in a neighbouring leaflet. This was expected because leaflets, as the main organs of transpiration, should have ample xylem connections with the petiole and stem and thus with each other. Sepals on the explant also swelled rapidly after the leaflet was scorched (Fig. 2). This indicates that the sepals had good xylem connections with the leaves, and that there was no significant hydraulic constriction at any point between leaf and calyx. This included the entire pathway through the peduncle and pedicel. The calyx responded similarly whether its subtending fruit (berry) was present or not. No such rapid increase in leaf or sepal thickness occurred 2 h 5 mm 10 min Leaflet Figure 2. Effect of leaf scorching on thickness of leaves and sepals on a tomato explant. The explant consisted of three leaflets connected by 12 cm of stem to a truss of five green fruit. Leaf and sepal thickness were monitored with linear-variable differential transformers (LVDTs). On one of the two calyces monitored, the fleshy part of the fruit (the berry) was intact (thicker line); on the other, most of the berry was cut away several hours prior to the experiment. At the time indicated by the vertical line, one of the leaflets (not the one under the transducer) was scorched with a flame for 3 s. when the fruit berry was scorched, either with a flame (Fig. 3) or by partial immersion in water at 80 C (not shown). Some small variations in thickness are apparent in Fig. 3, but they are not associated with scorching of the fruit. They occurred in both treated material and in the untreated blank and are probably temperature artefacts. When explants were exposed on the laboratory bench their leaf thickness declined slowly, as water was lost by transpiration. Figure 4 shows that removal of green fruit from an explant will greatly accelerate this decline. This indicates that, prior to fruit removal, leaf thickness was maintained by a slow (reversed) flow from the fruit. This reversed flow was not affected by steam girdling of the pedicel (not shown) and it must therefore occur in the xylem rather than the phloem. Similarly, in explants that remained on the bench for 2 3 d, the smaller fruit (1 3 cm diameter) developed a distinct crinkling on the cheek of the berry. This reflects shrivelling of the pericarp as water flows out gradually to the leaves. Thus, although back-flow from 10 min Scorch 1 Scorch 2 5 mm Calyx of 3 cm fruit Calyx of 4 cm fruit Leaflet Figure 3. Leaf and calyx thickness after scorching a fruit berry. Leaf and calyx thickness was measured on an explant as in Fig. 2. At the time indicated by the first vertical line (scorch 1) the berry of one fruit on the explant was heated for 10 s with a flame. The calyx of this fruit was not under a transducer. At the time indicated by the second vertical line (scorch 2) a berry subtending one of the measured sepals (that shown by the thicker line) was heated for 10 s with a flame.

4 568 M. Malone and J. Andrews 1 h 10 mm Remove all fruit from explant A the fruit was not apparent with intact plants, it was detectable in the more sensitive explant system. Xylem structure and continuity Leaflet explant B Leaflet explant A Figure 4. Effect of fruit removal on leaf thickness in explants. Leaf thickness was monitored in explants as in Fig. 2. At the time indicated by the vertical line, all fruit were excised from explant A (thicker line). No fruit were removed from explant B. A blank transducer and time and scale markers are also shown. Note much longer time scale than in some previous figures. Substantial xylem was visible in transverse sections from all positions along the peduncle and pedicel in fruit that was more than a few days old. Even the smallest, distal regions of pedicel contained large amounts of heavily lignified xylem (Fig. 6a). The transverse area of this xylem increased rapidly during fruit development (Andrews unpublished). LYCH dye applied to the basal cut end of excised tomato shoots was found to move throughout the shoot within a few hours. The yellowish dye began to appear in the fruit calyx and pericarp within about 1 h of its addition to the cut end, and it accumulated progressively during subsequent hours (Fig. 5; dye appears as dark patches). To reach the fruit from the basal cut end of the shoot, the dye would have to travel through at least 1 m. Closer inspection of the stained pericarp tissue in Fig. 5 revealed that the dye was located in xylem vessels at about 1 mm below the surface (Fig. 6b). This dye movement was not affected by steam girdling of the stem or pedicel. Dye movements must be interpreted with caution because they may not follow the normal pathway of water through a tissue (Canny 1990). In the present experiments, the dye was drawn up passively by the tissue rather than being forced in at positive pressure. The latter can flood air spaces and may recruit artificial flow pathways. Furthermore, the absence of staining was not used here as evidence of absence of connections. This avoids problems associated with limited dye uptake, rather than limited hydraulic conductivity. LYCH does not cross membranes readily (Oparka & Read 1994) so it should enter the xylem, but not the phloem, at the cut surface. Steam girdling was regularly used as a further check against phloem movement of dye. In view of these precautions, the LYCH movements observed here should be a reliable indicator of uninterrupted xylem pathways that are functional in vivo. They show that continuous xylem connections are present from the stem throughout the peduncle and pedicel, and into the fruit pericarp. Dye movement in the reversed direction was also observed. In one experiment the distal third of the berry was excised from a fruit on an explant. The cut surface of the attached part of the fruit was immersed to a depth of 2 mm in dye solution. After a range of incubation times, the pedicel was frozen and sectioned. The dye was found to have moved from the distal cut surface of the fruit into the pedicel within as little as 4 h. Xylem connections through the calyx-berry junction were studied in longitudinal sections. These were taken at about 16 h after applying LYCH dye to the cut end of excised shoot apices (as above). Substantial and continuous conducting xylem was observed to traverse this junction region (not shown). DISCUSSION Transient shrinkage is common in many plant organs. In low-transpiring organs, such as many fruit, it indicates that back-flow from the organ, via the xylem, exceeds influx via the phloem. Tomato fruit have previously been observed to show little or no diurnal shrinkage (Ehret & Ho 1986; Pearce et al. 1993; Kitano et al. 1996). Figure 1 demonstrates negligible fruit shrinkage even when the intact tomato plant was Figure 5. Progressive movement of xylem borne dye into the calyx and berry of an intact, attached tomato fruit. The same fruit was photographed sequentially after applying LYCH dye solution to the cut basal end of the shoot, some 1 m distant from the fruit. The time since dye application is indicated in each panel (top). Areas of dye appear dark. This fruit was approximately 25 mm in diameter, and approximately 20 d postanthesis.

5 Tomato hydraulics 569 Figure 6. Xylem in pedicel and pericarp. The original greenish fluorescent images have been converted to negative gray scale, and the stained areas thus appears dark. (A) A young pedicel (25 d postanthesis) is shown in transverse section. Auramine O-SO 2 was applied after sectioning, to stain lignin. A substantial and complete annulus of xylem is apparent, with a small ring of phloem fibres outside it, and some protoxylem scattered within the pith. The cortex is barely visible, but the position of the outer epidermis is marked by a dashed line. (B) Pericarp from a tomato berry was sectioned after xylem-loading of the shoot with LYCH for 5 h (as in Fig. 5). The dye is found within xylem vessels, seen here in longitudinal section. subjected to severe drought. Such drought would greatly decrease xylem pressures in the shoot, and should thus have overturned xylem pressure gradients (and water potential gradients, Johnson et al. 1992) into the fruit. The absence of fruit shrinkage, despite this treatment, suggests strongly that back-flow from the fruit is restricted because of high hydraulic resistance in the pathway between fruit and stem. The failure of tomato fruit to shrink readily on plants subjected to drought is unfortunate in horticultural terms because it precludes the use of diurnal cycles of modest water stress to ebb-and-flow additional calcium-rich xylem water into the growing fruit (Wiebe, Schatzler & Kuhn 1977). This might otherwise have offered a useful strategy against blossom-end rot. The evidence presented above indicates high hydraulic resistance between stem and fruit. However, no major hydraulic restrictions were found at any point proximal to the calyx. This is clear from Fig. 2, which shows that sepals, like leaves, swell rapidly when water is made available to the xylem of the stem. Such rapid swelling is diagnostic of good xylem connections (Turquois & Malone 1996). Consistent with these findings, extensive xylem was visible throughout the pedicel from an early stage of fruit development (Fig. 6a). Tests with dyes also demonstrated that continuous functional xylem connections extend through the entire peduncle and pedicel, and into the fruit (Fig. 5). Xylem restriction zones such as that reported by Lee (1989) in the knuckle of the pedicel must therefore affect few, if any, of the main conducting vessels. It is possible that only the mechanical components of the xylem (fibres and xylem parenchyma) are interrupted at such zones. This would give a visible reduction in the total xylem tissue but without greatly reducing hydraulic conductivity. This would also be consistent with the knuckle s future role as an abscission zone. No evidence was found to suggest that the knuckle region functions as a valve in any hydraulic sense. From data of the type shown in Figs 2 and 3, a quantitative estimate can be made of the distribution of xylem hydraulic resistance about the fruit calyx. Electrical analogue models show that the half-time of tissue swelling, t 1/2, will be roughly proportional to the product of the resistance and capacitance of the system, at least for a thin tissue such as a leaf or sepal (Malone 1993). Thus, t 1/2 ª krc, where R is the resistance of the flow path; C is the capacitance of the tissue; and k is a proportionality constant relating t 1/2 to the true time constant. When comparing flow into the calyx from proximal tissues with that from distal tissues (Figs 2 and 3, respectively) the capacitance is, for both fluxes, that of the sepal. Clearly then, the ratio of these half-times will equal the ratio of the pathway resistances: t t 12 proximal 12 distal krproximalcsepal R = = kr C R distal sepal proximal distal The half-time in Fig. 3 is too slow to measure accurately. However, it is evident that the swelling in Fig. 2 is at least 10-fold faster than that in Fig. 3, and thus that the pathway from calyx to fruit must have at least 10 times greater hydraulic resistance than that from calyx to stem. Thus it has been demonstrated that the major hydraulic resistance between stem and fruit lies within fruit itself, at some position distal to the calyx. The nature of this resistance remains unknown but it may reflect the general properties of the fruit pericarp, rather than some particular site of xylem constriction. The experiments with dyes indicated that functional xylem continues all the way from the stem into the pericarp (Fig. 6b). However, xylem vessels within the tomato pericarp may be unevenly distributed, too widely spaced, or partly non-functional (Belda & Ho 1993) thus contributing to low conductivity. A similar situation may occur within fruit of several other species, including grape and apple. These can also show calcium-deficiency disorders. Here, too, the xylem may be non-functional or may lose functionality at certain stages of fruit development (Greenspan et al. 1994; Lang & Ryan 1994; Wolswinkel et al. 1999). The restricted xylem flux into

6 570 M. Malone and J. Andrews these fruits may be required to promote phloem unloading and fruit softening (Wolswinkel et al. 1999). High xylem resistance may also serve to protect such organs from excessive back-flow at times when hydraulic tension is high in the xylem of the shoot. This might otherwise drain off the extracellular fluids, with their high solute content, which are commonly found in ripening fruit (Lang & Thorpe 1989). The restricted xylem connections may have supplied adequate calcium to the fruit of ancestral tomato lines, but are not adequate for the fast-growing, large-fruited crop lines of today. A similar situation seems to obtain in a range of other tissues, such as developing seeds, which also show high levels of apoplastic solutes coupled with a high degree of hydraulic isolation (Welbaum et al. 1992). ACKNOWLEDGMENTS This work was funded by MAFF. The authors are grateful to Richard Sampson of HRI Wellesbourne for help with photography. REFERENCES Adams P. & Ho L.C. (1993) Effects of environment on the uptake and distribution of calcium in tomato and on the incidence of blossom-end rot. Plant & Soil 154, Andrews J., Malone M., Thompson D.S., Ho L.C. & Burton K.S. (2000) Peroxidase isozyme patterns in the skin of maturing tomato fruit. Plant Cell & Environment 23, Araki T., Kitano M. & Eguchi H. (1997) Respiration, sap flux, water balance and expansive growth in tomato fruit. Biotronics 26, Belda R.M. & Ho L.C. (1993) Salinity effects on the network of vascular bundles during tomato fruit development. Journal of Horticultural Science 68, Brauer M., Zhong W.-J., Jellito T., Schobert C., Sanders D. & Komor E. (1998) Free calcium ion concentration in the sieve tube sap of Ricinus communis L. seedlings. Planta 206, Canny M.J. (1990) What becomes of the transpiration stream? New Phytology 114, Ehret D.L. & Ho L.C. (1986) Effects of osmotic potential in nutrient solution on diurnal growth of tomato fruit. Journal of Experimental Botany 37, Greenspan M.D., Shackel K.A. & Matthews M.A. (1994) Developmental changes in the diurnal water budget of the grape berry exposed to water deficits. Plant, Cell & Environment 17, Hetherington S.E., Smillie R.M. & Davies W.J. (1998) Photosynthetic activities of vegetative and fruiting tissues of tomato. Journal of Experimental Botany 49, Ho L.C., Grange R.I. & Picken A.J. (1987) An analysis of the accumulation of water and dry matter in tomato fruit. Plant, Cell & Environment 10, Johnson R.W., Dixon M.A. & Lee D.R. (1992) Water relations of the tomato during fruit growth. Plant, Cell & Environment 10, Kitano M., Hamakoga M., Yokomakura F. & Eguchi H. (1996) Interactive dynamics of fruit and stem growth in tomato plants as affected by root water condition. 1. Expansion and contraction of fruit and stem. Biotronics 25, Lang A. & Ryan K.G. (1994) Vascular development and sap flow in apple pedicels. Annals of Botany 74, Lang A. & Thorpe M.R. (1989) Xylem, phloem, and transpiration flows in a grape: application of a technique for measuring the volume of attached fruits to high resolution using Archimedes principle. Journal of Experimental Botany 40, Lee D.R. (1989) Vasculature of the abscission zone of tomato fruit: implications for transport. Canadian Journal of Botany 67, Malone M. (1992) Kinetics of wound-induced hydraulic signals and variation potentials in wheat seedlings. Planta 187, Malone M. (1993) Hydraulic signals. Philosophical Transactions of the Royal Society B341, Malone M. (1996) Rapid, long-distance signal transmission in higher plants. Advances in Botanical Research 22, Ohta K., Hosoki T., Matsumoto K., Ohya M., Ito N. & Inaba K. (1997) Relationships between fruit cracking and changes of fruit diameter associated with solute flow to fruit in cherry tomatoes. Journal of the Japanese Society for Horticultural Science. 65, Oparka K.J. & Read N.D. (1994) The use of fluorescent probes for studies of living plant cells. In Plant Cell Biology: a Practical Approach (eds N. Harris & K.J. Oparka), pp Oxford University Press, Oxford. Pearce B.D., Grange R.I. & Hardwick K. (1993) The growth of young tomato fruit I. Effects of temperature and irradiance on fruit grown in controlled environments. Journal of Horticultural Science 68, Raven J.A. (1977) H + & Ca 2+ in phloem and symplast: relation of relative immobility of the ions to the cytoplasmic nature of the transport paths. New Phytology 79, Tromp J. (1984) Diurnal fruit shrinkage in apple as affected by leaf water potential and vapour pressure deficit of the air. Scientia Horticulturae 22, Turquois N. & Malone M. (1996) Non-destructive assessment of developing hydraulic connections in the graft union of tomato. Journal of Experimental Botany 47, Welbaum G.E., Meinzer G.E., Grayson R.L. & Thornham K.T. (1992) Evidence for and consequences of a barrier to solute diffusion between the apoplast and vascular bundles in sugarcane stalk tissue. Australian Journal of Plant Physiology 19, Wolswinkel P., Ammerlaan A. & Koerselman-Kooij J. (1999) Changes in the function of the xylem pathway during the development of fleshy fruits: implications for phloem transport, water movement and fruit growth. In Proceedings of International Conference on Assimilate Transport and Partitioning, p Wiebe H.J., Schatzler H.P. & Kuhn W. (1977) On the movement and distribution of calcium in white cabbage in dependence of the water status. Plant & Soil 48, Received 20 June 2000; received in revised form 30 October 2000; accepted for publication 24 November 2000