- Email: [email protected]
Response of tomato plants to a step-change in root-zone salinity under two different transpiration regimes
Scientia Horticulturae 93 (2002) 267±279
Response of tomato plants to a step-change in root-zone salinity under two different transpiration regimes Ya Ling Lia,1, Cecilia Stanghellinia,*, Hugo Challab a
DLO Institute of Agricultural and Environmental Engineering (IMAG-DLO), P.O. Box 43, NL-6700 AA Wageningen, The Netherlands b Department of Agrotechnology and Food Sciences, Wageningen University, Bomenweg 4, 6703 HD Wageningen, The Netherlands Accepted 19 July 2001
Abstract The response of a tomato crop to a step-change in salinity was investigated under different potential transpiration conditions. A crop growing for 5 months under saline irrigation water (EC 9 dS m 1) was given thereafter a standard nutrient solution with an EC of 2 dS m 1. The previous effects of salinity were largely reversed, especially for fruits and leaves that had not yet reached the rapid growth phase. After a period of 8 weeks, the ®nal weight of fruits reached that of ``normal'' (EC 2 dS m 1) fruits. There was a high incidence of fruit cracking, even greater in the low transpiration treatment than the high one. The peak incidence of cracking was in fruits that were harvested some 25 days after lowering the EC. The chance of cracking was positively affected by the increase in skin expansion rate due to a change in EC and further enhanced by reduced potential transpiration (high ambient humidity). New leaves formed after the EC was lowered were comparable with those grown in low EC, but leaves that were fully expanded at that moment did not respond to the change in EC. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Tomato (Lycopersicon esculentum L.); Salinity; Transpiration; Fruit cracking; Leaf size; EC-change
1. Introduction Most plants respond to salinity with reduced growth, whenever salt concentration in the root environment exceeds a threshold value, according to a model originally proposed by * Corresponding author. Tel.: 31-317-476322; fax: 31-317-425670. E-mail address: [email protected] (C. Stanghellini). 1 Permanent address: Department of Horticulture, Shanxi Agricultural University, 030801 Taigu, Shanxi, PR China.
0304-4238/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 2 3 8 ( 0 1 ) 0 0 3 2 9 - 6
268
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Maas and Hoffman (1977). In most subsequent publications on the subject (e.g. Dalton and Poss, 1990; Shannon and Grieve, 1999; Sonneveld, 2000) threshold EC and yield decrease were determined for various crops, under conditions of constant salinity in the root environment. In tomato, the salinity-induced yield reduction is mainly caused by a decreased in¯ow of water into fruits (Ehret and Ho, 1986a), combined with a shortening of the fruit growth period (Mizrahi, 1982). There is general agreement that other factors (most notably shoot environment), may modify yield response to salinity, as already pointed out by Maas and Hoffman (1977). We have shown that the ``environment'' effect can be quanti®ed by potential evaporation; in particular, that reducing potential evaporation limits the damage caused by salinity (Li et al., 2001). However, natural conditions are seldom constant: for instance, growers may have to use irrigation water of changing quality; or rainfall may wash out salts accumulated in the root environment. Not much is known about plant response to changing root-zone salinity. AlarcoÂn et al. (1994) showed that a decrease in relative growth rate and leaf area ratio in tomato plants, in response to increasing osmotic pressure of the nutrient solution, could be detected within an experimental period of 17 days. Van de Sanden and Uittien (1995) showed that the fruit growth rate of tomato decreased and that the decrease of fruit size was related to relative exposure time at high salinity. In the present study, the response of a tomato crop to a step change of root-zone salinity was observed in order to establish to what extent plants recover after a prolonged exposure to high salinity. Therefore the response to a step-change from a 5-month-exposure to 9 dS m 1 root-zone salinity (EC) to 2 dS m 1, rather than the other way round, was determined in this experiment. In addition, the response under two levels of potential transpiration was determined to get some insight into the response dynamics and underlying mechanisms. 2. Materials and methods The general set-up of the series of experiments has been described in detail before (Li et al., 2001) and only the points that are relevant to this paper will be outlined here. Tomato, cv Chaser, was grown in rockwool slabs in two compartments (300 m3 each) of a Venlo-type glasshouse. The two compartments were controlled so as to have two potentialtranspiration climates, as explained below. Two (constant) salinity treatments (expressed by electrical conductivity, EC, dS m 1) were given each to one-half of the rows in each compartment. Oversupply (drain fraction at least 70%) ensured that the EC of drain water (that was monitored) was practically the same as the EC of the irrigation water. In addition, the concentration of the nutrient solution in the root-zone was manually controlled twice a week in blended random samples drawn from the slabs. This particular experiment started after the crop had been subjected to two constant-salinity treatments, both 9 dS m 1, between February and June 1997. In one treatment sodium chloride was added to a standard nutrient solution of 2 dS m 1 (NaCl treatment) whereas the other one was a more concentrated version of the standard nutrient solution (high EC, HEC). At the start of the experiment (4 July), the solution containing NaCl was stepwise ¯ushed out (during 10 days), and refreshed with a standard nutrient solution (EC 2 dS m 1 : lowered EC
Treatment
EC (dS m 1)
pH
NH4 (mmol l 1)
NaCl EC# HEC (before) HEC (after)
8.9 2.2 9.1 8.0 CNS (MPa)
6.1 6.1 5.6 5.2
0.1 5.6 0.1 4.4 0.2 21.4 0.2 23.3 Fe (mmol l 1) Mn (mmol l 1)
11.2 5.1 6.9 2.4 23.8 11.8 19.4 7.8 Zn (mmol l 1) B (mmol l 1)
22.7 13.0 54.3 37.0
8.6 3.5 13.3 7.7
NaCl EC# HEC (before) HEC (after)
0.37 0.09 0.35 0.30
K (mmol l 1)
7.1 2.0 29.4 25.5
Ca (mmol l 1)
Mg (mmol l 1)
64.6 28.5 166.6 117.0
NO3 (mmol l 1)
SO4 (mmol l 1)
19.7 7.3 14.2 4.6 56.4 13.6 51.0 11.7 Cu (mmol l 1) Mo (mmol l 1) 2.7 1.4 4.8 4.1
H2PO4 Na Cl (mmol l 1) (mmol l 1) (mmol l 1) 1.5 0.6 4.4 4.9
49.6 3.2b 6.3 5.1
52.4 0.8b 4.7 2.0
1.6 0.9 1.8 0.9
a Also given is the calculated osmotic potential of the nutrient solution (Slatyer, 1967). Values are means of 7 (before) and 2 (after) measurements unless otherwise indicated. b One measurement: the other sample was taken during the phasing out of NaCl.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Table 1 Mean composition of the solution drawn from the rockwool before and after lowering the ECa
269
270
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
treatment, EC#). Nothing was changed in the other treatment. The mean composition of the solution extracted from the rockwool slabs in both treatments is listed in Table 1, for this experiment as well as for the preceding period. Climate in one compartment was controlled according to Dutch standard practice, i.e.: pre-®xed set-points of day- and night-temperature (20 and 18 8C), respectively, with an allowance for solar radiation and crop stage; ambient humidity controlled by setting minimum values (day- and night-time) for vapour pressure de®cit (VPD). This compartment was our reference (high potential transpiration treatment, HET0). Climate control in the other compartment (low transpiration treatment, LET0) aimed at reducing potential transpiration by a third, by manipulating as far as possible only ambient humidity, by a combination of venting and misting. This caused, however, the temperature to be about 1 8C lower (average over the growing period, Fig. 1). The model used for calculating the transpiration rate has been described by Stanghellini (1987) and the climate control algorithm by Stanghellini and Van Meurs (1992). The effectiveness of the transpiration control was checked by determining water uptake (Fig. 2) through the balance of supply and drain data from each treatment. Six groups of four plants were marked at random in the central rows of each treatment. Twice a month, length and width of every leaf (longer than 15 cm) present on one preselected plant in each group were measured non-destructively. Since leaves were routinely picked, as is done in commercial tomato production, there were about 25 leaves each time, whose position from the root was recorded as well. Each 4-plant group was harvested twice a week, and marketable and unmarketable yields (weight and number) were determined. Fruit development period (FDP) was de®ned as the difference between a ``¯owering'' and a ``harvesting'' time, estimated as follows. The position of all fruit trusses with respect to the leaves was recorded on the six plants per treatment that were monitored. Very few exceptions were observed (Li and Stanghellini, 1999) to the characteristic of tomato that there is a ¯ower truss for each three leaves above the ®rst in¯orescence (Shishido and Hori, 1977). Then, the number of the truss ¯owering at any moment (NTf, counting from the root) was estimated by relating it to the leaf observations as follows: NTf 13 NL
N0
Fig. 1. Mean temperature in the high (solid line) and low (dashed one) transpiration greenhouses (28-day moving averages). DOY is day of year.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
271
Fig. 2. Daytime water uptake (litres per plant per day) in low potential transpiration (LET0) vs. the one in high potential transpiration (HET0), between 7 July and 20 August 1997. Symbols: (*) HEC and (*) EC#. Line is the intended treatment: Y 23 X.
where NL is the number of the youngest measured leaf (longer than 7 cm), counting from the root; and N0 the number of leaves below the first source±sink unitÐa truss together with the three leaves immediately below it (Tanaka and Fujita, 1974). The date of ¯owering of each truss was estimated by linear interpolation between such two-a-month observations, one mean date for each treatment. The date of the ®rst harvest from each truss was recorded. A mean (of 24 plants per treatment) harvest date of each truss was determined, and assigned to all fruits of a truss. FDP was de®ned as the time elapsed between the estimated mean dates of ¯owering and harvesting. 3. Results 3.1. Fruit growth Fruits in the NaCl treatment were consistently smaller than the fruits in the corresponding HEC (Fig. 3). This may be caused by the slight difference in osmotic potential (Table 1)
Fig. 3. Difference (g) between mean weight of fruits from the two nutrition treatments before (circles, NaCl± HEC) and after (triangles, EC#±HEC) lowering the EC. Closed and open symbols refer to high and low transpiration, respectively. The arrow indicates the time of lowering the EC. DOY is day of year.
Table 2 Parameters of plant development during the 2 weeks before lowering the EC (before) and two final weeks of the present experiment (after)a High transpiration
1
EC root-zone (dS m ) Leaf length (cm) Fruit trusses (per plant) Fruit weight (g) Fruit DM (%) FDP (days) a
Low transpiration
1996 EC 9
Before
9.3 32.4 0.6 ± 55.0 1.9 6.58 0.03 ±
9.1 37.1 7.0 56.8 6.52 52
After
HEC
!
NaCl
0.6 0.3 0.7 0.05
9.0 37.0 6.9 58.9 6.72 53
0.2 0.2 0.8 0.02
EC#
HEC
1998 EC 2
1996 EC 9
2.2 8.0 2.3 9.0 34.5 0.4 32.3 1.0 36.9 0.9 35.0 0.4 6.3 0.1 6.1 0.3 5.8 0.4 ± 74.2 1.8 45.6 2.3 71.9 0.8 59.2 1.2 ± ± ± 6.78 0.03 46 39 48 ±
Before
After
HEC 9.1 37.0 7.5 59.4 6.31 59
!
NaCl
0.3 0.2 1.1 0.03
8.8 36.3 7.7 58.3 6.42 60
0.5 0.3 1.4 0.02
EC#
HEC
1998 EC 2
2.2 7.9 2.4 35.1 1.1 33.0 0.6 38.9 1.3 6.9 0.3 6.3 0.3 6.5 0.2 76.6 1.6 51.4 1.2 75.9 2.6 ± ± ± 49 40 50
Leaf length is the average of the 15 lowermost leaves on six plants for each treatment. Fruit weight and fruit dry matter content (fruit DM) are the average of marketable fruits. ``Fruit trusses'' is the mean number of trusses on the plants in that time. ``FDP'' is the fruit development period. The codes 1996 and 1998 indicate corresponding values (in the same period as before and after) from a similar crop grown at constant EC 9:5 dS m 1 (high concentrated nutrient solution) in 1996 and EC 2 dS m 1 in 1998 (Li et al., 2001). Figures are mean standard error (n 6, except for fruit DM (%), where n 40).
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
273
Fig. 4. Mean fruit weight (g), of each harvest, from LET0 vs. corresponding weight in HET0. Symbols: (*) HEC, (*) NaCl and (~) EC#. Solid line shows the best-fit relationships of HEC (P < 0:0001: slope 1.074) and dotted line shows of EC# points (P < 0:0001: slope 1.009).
between the two treatments (Sonneveld, 2000). However, the difference between mean weights at the time of starting this experiment (if one does not account for previous history) was not signi®cant (Table 2). After the EC was lowered, the mean weight of fruits harvested in EC# soon became larger than that of the HEC fruits in the same greenhouse, and the difference in weight kept increasing with time (Fig. 3). At the end of the experiment the average fruit weight of EC# was 63 and 49% higher than that of HEC in the high and the low transpiration houses, respectively (Table 2). HEC fruits were larger in the low transpiration house than in the high transpiration one (Fig. 4), as had been observed also the year before (Li et al., 2001). 3.2. Fruit cracking Lowering the EC caused fruit cracking, particularly in the low transpiration treatment (Fig. 5). The ®rst cracked fruits were harvested soon after lowering the EC, whereas the highest proportion of cracked fruits was harvested 17 and 24 days later in LET0 and HET0, respectively, when it exceeded 90% in LET0 and 30% in HET0 (Fig. 5). That cracking was almost exclusively caused by lowering the EC is shown by the lower panel of Fig. 5, where the (absence of) incidence of cracking in the HEC treatment is shown. Over the whole period of the experiment, 44% of all fruits harvested from EC# in LET0 were cracked, compared with 12% in HET0. 3.3. Fruit development period The declining trend of FDP with truss number (Fig. 6) is fully explained by the opposite trend of temperature (refer to Fig. 1), according to the model of de Koning (1994). That model can explain as well the difference in FDP between HET0 and LET0, as caused by the difference in temperature between the two houses. After starting this experiment, FDP in EC# gradually extended, ®nally to become 20% longer than in HEC, in both greenhouses (Fig. 6). If this effect were related to osmotic potential, this could
274
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Fig. 5. Percentage of cracked fruits in each harvest, for the treatments NaCl (squares) and EC# (triangles) in (A) and HEC (circles) in (B). Closed and open symbols indicate HET0 and LET0, respectively. The arrow shows the time of lowering the EC. Vertical bars are plus and minus standard error n 6. The standard error is not shown in (B) for clarity.
explain as well the difference of about 3 days between FDP in NaCl and in HEC in HET0, observed in the ®rst 15 trusses that were harvested before this experiment. The slightly lower root-zone EC in NaCl (Table 2) may have masked this effect in the other house. 3.4. Vegetative growth Fig. 7 shows the mean pro®les of leaf length measured at the start of the experiment, and the boundary of the mean pro®les (the maximum length) measured subsequently (last
Fig. 6. FDP plotted against truss number (referring to the position of the truss counting from the root) in HEC (*, *), NaCl (& , &) and EC# (~, ~). Closed and open symbols represent HET0 and LET0, respectively. The arrows together with lines indicate the trusses that were being harvested at the time of lowering the EC. Points show averages of each truss for 24 plants per treatment.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
275
Fig. 7. Profiles of leaf length in NaCl ( ) and HEC (Ð) when starting this experiment (shown by the two pointers), and the boundary of the mean profiles (the maximum length) measured subsequently in EC# ( ) and HEC (Ð) (last measured profile was 28 August). Leaf number refers to the position of the leaves along the stem, counting from the root. Lines are three-value running averages of mean profiles of 12 plants per treatment.
measured pro®le was 28 August). The abscissa refers to the position of each leaf, starting from the root. Since there was no leaf response to the transpiration treatment, data are pooled for each EC treatment. There was a small difference between HEC and NaCl pro®les at the beginning of the experiment. However, what is important to notice is that leaves that were fully expanded (about 37 cm) when the EC was lowered did not expand much further, whereas smaller leaves grew to become longer than the HEC ones. After 2 months, most leaves that were still present had been formed after the treatment started and the maximum length at full expansion (average of 15 leaves from bottom) in EC# was 6% more than that in HEC (Table 2). This agrees with our previous ®nding (Li and Stanghellini, 2001) that prolonged exposure to high EC decreases leaf length by 3.7% per dS m 1, in excess of 6.5 dS m 1. 4. Discussion 4.1. Fruit growth In a previous paper (Li et al., 2001) it was shown that the main effect of high EC was a decrease of water in¯ow into the fruits. Consequently, lowering the EC at the beginning of this experiment must have increased water in¯ow into fruits, i.e., fresh growth rate. Indeed, the weight of ripe fruits gradually increased with time (Fig. 3) until fruits were harvested that had developed fully under the new EC. The experiment probably ®nished around the time that this stage was reached, and so no plateau could be distinguished in Fig. 3. That the weight of ripe fruits at the end of the experiment did not ``bear memory'' of the previous exposure of the crop to high EC is partly con®rmed by Table 2. The EC 2 values in 1998, put there as reference, refer to fruits that were harvested in the same compartment and period the year thereafter, from a crop subjected to a constant low-EC treatment, during an experiment that was similar in design to this one (Li et al., 2001).
276
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
4.2. Fruit cracking The occurrence of fruit cracking after lowering the EC of the solution was to be expected. Indeed, it is well known that fruits may crack when there is a sudden increase in water availability, as a result of irrigation, or rain after prolonged drought (e.g. Peet, 1992; Opara et al., 1997), or a lowering of the conductivity of the fertiliser solution (Peet, 1992). Cracking is associated with decreased epidermis elasticity at the mature-green or breaker stage (Kamimura et al., 1972; Bakker, 1988) that causes rupture under the stretching caused by increased water in¯ow. In order to pinpoint the relationship between growth stage and susceptibility to cracking, it is useful to describe fruit growth rate in the two cases, since ®nal fruit size and fruit development period before and after lowering the EC are known. The growth curves in Fig. 8 (left panel) are based on the logistic function with the parameters given in Table 2. In both cases a ripening period equal to 20% of FDP has been taken into account (Ho and Hewitt, 1986; Bakker, 1991). For the sake of this analysis, fruit weight has been converted into surface area, assuming a spherical shape and unit density. The right panel of Fig. 8 shows the derivative (growth rate) of the curves ``before'' lowering the EC. It also shows the difference between the growth rates after and before lowering the EC. The abscissa in the right panel of Fig. 8 is the time remaining to harvest, to facilitate comparison with Fig. 5. This may seem confusing, unless one realises that, since Fig. 5 refers to ripe fruits, points to the right of any given day in that ®gure represent fruits that were still unripe at that moment. The right-hand panel of Fig. 8 implies that fruits whose growth rate was maximal at the time EC was lowered would be harvested some 30 days thereafter, whereas fruits subjected to the largest ``strain'' (difference between ``old'' and ``new'' growth rate) would be some 25 days from harvest. Comparison of this with Fig. 5 implies that fruits were most susceptible to cracking when the difference between the growth rate in the new and old situation was maximal, rather than when the absolute growth rate was maximal. This analysis supports the hypothesis that the epidermis tends to crack when the newly required
Fig. 8. Left panel: growth curve of the surface area (cm2) of tomato fruits before and after lowering the EC, in HET0 (Ð) and LET0 (- - -). Right panel: the original growth rate before lowering the EC (i.e., the derivative of the ``before'' curves in the left panel) and the difference between the new (after) and the original (before) growth rate vs. the time remaining to harvest.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
277
expansion is much larger than the previous expansion rate (Kamimura et al., 1972). If the hypothesis of Peet (1992), that rapidly growing fruit might be especially predisposed to cracking were true, the maximum occurrence of cracking would have taken place later. According to Fig. 8, low transpiration would stretch the time-to-harvest in both cases by a few days, which is obviously not consistent with Fig. 5. The one factor taken into account here, however, is ®nal size, which was larger in LET0. For instance, the increased incidence of cracking of tomato fruits under low potential evaporation (high relative humidity, small vapour pressure de®cit) is well documented (e.g. Peet, 1992; Maroto et al., 1995), most recently by Leonardi et al. (2000). Peet (1992) inferred that high humidity effects on fruit cracking were related to gas and water pressure increases, due to an increase in water supply or a temperature increase. The excess water supply (or water pressure increase) within the fruit, when fruit transpiration is small such as in a humid environment (Ehret and Ho, 1986b; Leonardi et al., 1999) can only be relieved by expansion of the epidermis. Indeed, our results seem compatible with the hypothesis that susceptibility to fruit cracking is proportional both to the difference in expansion rate between the new and the old situation and to the difference in water in¯ow and out¯ow to and from the fruits, which is determined by the potential evaporation. 4.3. Fruit development period The difference of FDP between the two greenhouses before this experiment started, could be fully explained by the difference in temperature. Therefore there is no need to postulate a humidity effect, which, indeed Bakker (1991) did not observe. Shortened FDP with NaCl in HET0 might be explained by a sodium-induced reduction of potassium absorption (Hecht-Buchholz et al., 1979), since a low concentration of potassium reduces FDP (Besford and Maw, 1975). Since depressed transpiration is likely to reduce uptake of Na (Tsuchiya et al., 1992), the absence of an FDP response to NaCl in LET0 does not necessarily go against this hypothesis. This may also explain why the FDP extension after lowering EC is apparently greater in HET0 than in LET0. The extension of FDP after lowering the EC is probably comparable with the observations of Mizrahi (1982) that high salinity (6.6 dS m 1) could shorten the duration of fruit development (by 4±15% in various tomato cultivars). A shortening of FDP caused by high EC has been observed in our series of experiments (Li, 2000). This may be quite similar to the shortening of development period under water stress, observed by many researchers (e.g. Salter, 1958; Wolf and Rudich, 1988; Hsiao, 1993). Hsiao (1993) stated that mild to moderate water stress during the generative phase can be bene®cial for tomato plants by promoting early partition of assimilates to fruits and, consequently, early maturity. 4.4. Vegetative growth As shown before (Li and Stanghellini, 2001), the transpiration treatment had no effect on leaf area, which is con®rmed here. Although an enhanced elongation of single, still growing leaves, could be observed soon after lowering the EC, it took the 8-week duration of the experiment before the leaf pro®les signi®cantly diverged. The response of leaves and of fruits to a step change in osmotic pressure in the root-zone is similar in one aspect.
278
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Organs formed and developed under the new situation are fully adapted to it, despite the prolonged exposure of the plant to high EC. Organs that are already formed can adapt only to a limited extent, as shown in Figs. 3 and 7. Munns et al. (1982) observed the same in a short-term experiment with barley plants. A rapid reversal of the inhibition of leaf growth, after removal NaCl from the root medium, was also observed by Neumann (1993). They all concluded that leaf growth was limited in high salinity by water de®cit in the elongating leaf tissue rather than by ion excess. 5. Conclusion The negative effect of high salinity on growth and yield is mainly related to the water balance of the plant. This is indirectly con®rmed by the higher occurrence of cracked fruit in LET0, a treatment that was likely to make more water available for uptake into the fruits. This paper also shows that the water balance can be restored even after quite a long exposure to high salinity. However, the time course for recovery is comparable to the duration of the organs' development. Acknowledgements The project has been ®nanced by the Dutch Ministry of Agriculture and Environment (research program 256) and by the European Union (ICA3-CT-1999-0009 HORTIMED). Financial support from the World Laboratory, Geneva, Switzerland for the stay in Holland of Ya Ling Li is gratefully acknowledged. We are indebted to the staff of the horticultural farm of IMAG and to Ferry Corver in particular, for the success of the experiments. We owe thanks to Johan van Gaalen for helping with the yield measurements and to Wim Van Meurs and Rein Bijkerk for solving all our problems with the climate control routine and data collection. References AlarcoÂn, J.J., SaÂnchez-Blanco, M.J., BloarõÂn, M.C., Torrecillas, A., 1994. Growth and osmotic adjustment of two tomato cultivars during and after saline stress. Plant Soil 166, 75±82. Bakker, J.C., 1988. Russeting (cuticle cracking) in glasshouse tomatoes in relation to fruit growth. J. Hort. Sci. 63 (3), 459±463. Bakker, J.C., 1991. Analysis of humidity effects on growth and production of glasshouse fruit vegetables. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 155 pp. Besford, R.T., Maw, G.A., 1975. Effect of potassium nutrition on tomato plant growth and fruit development. Plant Soil 42, 395±412. Dalton, F.N., Poss, J.A., 1990. Water transport and salt loading: a unified concept of plant response to salinity. Acta Hort. 278, 187±293. de Koning, A.N.M., 1994. Development and dry matter distribution in glasshouse tomato: a quantitative approach. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 240 pp. Ehret, D.L., Ho, L.C., 1986a. The effect of salinity on dry matter partitioning and fruit growth in tomatoes grown in nutrient film culture. J. Hort. Sci. 61 (3), 361±367. Ehret, D.L., Ho, L.C., 1986b. Effect of osmotic potential in nutrient solution on diurnal growth of tomato fruit. J. Exp. Bot. 37 (182), 1294±1302.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
279
Hecht-Buchholz, Ch., Mix, G., Marschner, H., 1979. Effect of NaCl on mineral content and fine structure of cell in plant species with different salt tolerance. In: Proceedings of the Seventh Intern. Coll. Plant Anal. Fert. Problems, Vol. 1, pp. 147±156. Ho, L.C., Hewitt, J.D., 1986. Fruit development. In: Thereon, J.G., Rudich, J. (Eds.), The Tomato Crop: A Scientific Basis for Improvement. Chapman & Hall, London, pp. 201±239. Hsiao, T.C., 1993. Growth and productivity of crops in relation to water status. Acta Hort. 335, 137±148. Kamimura, S., Yoshikawa, H., Ito, K., 1972. Studies on fruit cracking in tomatoes. Bull. Hort. Res. Station Morioka, Iwate, Japan 7, 73±138. Leonardi, C., Baille, A., Guichard, S., 1999. Effects of fruit characteristics and climatic conditions on tomato transpiration in a greenhouse. J. Hort. Sci. Biotechnol. 74 (6), 748±756. Leonardi, C., Guichard, S., Bertin, N., 2000. High vapour pressure deficit influence growth, transpiration and quality of tomato fruits. Sci. Hort. 84, 258±296. Li, Y.L., 2000. Analysis of greenhouse tomato production in relation to salinity and shoot environment. Ph.D. Dissertation. Wageningen University, Wageningen, 96 pp. Li, Y.L., Stanghellini, C., 1999. The leaf characteristics of a tomato plant in rockwool during long season growing. J. Shanxi Agric. Univ. 19 (3), 214±217. Li, Y.L., Stanghellini, C., 2001. Analysis of the effect of EC and potential transpiration on vegetative growth of tomato. Sci. Hort. 89, 9±27. Li, Y.L., Stanghellini, C., Challa, H., 2001. Effect of EC and transpiration on production of greenhouse tomato (Lycopersicon esculentum L.). Sci. Hort., 88, 11±29 Maas, E.V., Hoffman, G.J., 1977. Crop salt tolerance-current assessment. J. Irrig. Drain. Div., Proc. Am. Soc. Civil Eng. 103 (IR2), 115±134. Maroto, J.V., Bardisi, A., LoÂpez, S., Pascual, B., Alagarda, J., 1995. Influence of relative humidity in the appearance of cracking in tomato (Lycopersicon esculentum Mill.) fruit. Acta Hort. 412, 306±312. Mizrahi, Y., 1982. Effect of salinity on tomato fruit ripening. Plant Physiol. 69, 966±970. Munns, R., Greenway, H., Delane, R., Gibbs, J., 1982. Ion concentration and carbohydrate states of the elongating leaf tissue of Hordeum vulgare growing in high external NaCl. II. Cause of the growth reduction. J. Exp. Bot. 33, 574±583. Neumann, P.M., 1993. Rapid and reversible modifications of extension capacity of cell walls in elongation maize leaf tissues responding to root addition and removal of NaCl. Plant Cell Environ. 16, 1107±1114. Opara, L.U., Studman, C.J., Banks, N.H., 1997. Fruit skin splitting and cracking. Hort. Rev. 19, 217±262. Peet, M.M., 1992. Fruit cracking in tomato. HortTechnology 2 (2), 216±223. Salter, P.J., 1958. The effect of difference water-regimes on the growth of plants under glass. IV. Vegetative growth and fruit development in the tomato. J. Hort. Sci. 33, 1±12. Shannon, M.C., Grieve, C.M., 1999. Tolerance of vegetable crops to salinity. Sci. Hort. 78, 5±38. Shishido, Y., Hori, Y., 1977. Studies on translocation and distribution of photosynthetic assimilate in tomato plants. II. Distribution pattern as affected by phyllotaxis. Tohoku J. Agric. Res. 28 (2), 82±95. Slatyer, R.O., 1967. Plant±water Relationships. Academic Press, London, 366 pp. Sonneveld, C., 2000. Effects of salinity on substrate grown vegetables and ornamentals in greenhouse horticulture. Ph.D. Dissertation. Wageningen University, Wageningen, 151 pp. Stanghellini, C., 1987. Transpiration of greenhouse crops: an aid to climate management. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 150 pp. Stanghellini, C., Van Meurs, W.Th.M., 1992. Environmental control of greenhouse crop transpiration. J. Agric. Eng. Res. 51, 297±311. Tanaka, A., Fujita, K., 1974. Nutrio-physiological studies on the tomato plant. IV. Source±sink relationship and structure of the source±sink unit. Soil Sci. Plant Nutr. 20, 57±68. Tsuchiya, M., Naito, H., Ehara, H., Ogo, T., 1992. The physiological response of rice to salinity. I. The relationship between Na uptake and transpiration under different humidity and salinity conditions. Jpn. J. Crop Sci. 61 (1), 16±21. Van de Sanden, P.A.C.M., Uittien, J.J., 1995. Root environment water potential and tomato fruit growth. Acta Hort. 401, 531±536. Wolf, S., Rudich, J., 1988. The growth rates of fruits on different parts of the tomato plant and the effect of water stress on dry weight accumulation. Sci. Hort. 34, 1±11.
Response of tomato plants to a step-change in root-zone salinity under two different transpiration regimes Ya Ling Lia,1, Cecilia Stanghellinia,*, Hugo Challab a
DLO Institute of Agricultural and Environmental Engineering (IMAG-DLO), P.O. Box 43, NL-6700 AA Wageningen, The Netherlands b Department of Agrotechnology and Food Sciences, Wageningen University, Bomenweg 4, 6703 HD Wageningen, The Netherlands Accepted 19 July 2001
Abstract The response of a tomato crop to a step-change in salinity was investigated under different potential transpiration conditions. A crop growing for 5 months under saline irrigation water (EC 9 dS m 1) was given thereafter a standard nutrient solution with an EC of 2 dS m 1. The previous effects of salinity were largely reversed, especially for fruits and leaves that had not yet reached the rapid growth phase. After a period of 8 weeks, the ®nal weight of fruits reached that of ``normal'' (EC 2 dS m 1) fruits. There was a high incidence of fruit cracking, even greater in the low transpiration treatment than the high one. The peak incidence of cracking was in fruits that were harvested some 25 days after lowering the EC. The chance of cracking was positively affected by the increase in skin expansion rate due to a change in EC and further enhanced by reduced potential transpiration (high ambient humidity). New leaves formed after the EC was lowered were comparable with those grown in low EC, but leaves that were fully expanded at that moment did not respond to the change in EC. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Tomato (Lycopersicon esculentum L.); Salinity; Transpiration; Fruit cracking; Leaf size; EC-change
1. Introduction Most plants respond to salinity with reduced growth, whenever salt concentration in the root environment exceeds a threshold value, according to a model originally proposed by * Corresponding author. Tel.: 31-317-476322; fax: 31-317-425670. E-mail address: [email protected] (C. Stanghellini). 1 Permanent address: Department of Horticulture, Shanxi Agricultural University, 030801 Taigu, Shanxi, PR China.
0304-4238/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 2 3 8 ( 0 1 ) 0 0 3 2 9 - 6
268
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Maas and Hoffman (1977). In most subsequent publications on the subject (e.g. Dalton and Poss, 1990; Shannon and Grieve, 1999; Sonneveld, 2000) threshold EC and yield decrease were determined for various crops, under conditions of constant salinity in the root environment. In tomato, the salinity-induced yield reduction is mainly caused by a decreased in¯ow of water into fruits (Ehret and Ho, 1986a), combined with a shortening of the fruit growth period (Mizrahi, 1982). There is general agreement that other factors (most notably shoot environment), may modify yield response to salinity, as already pointed out by Maas and Hoffman (1977). We have shown that the ``environment'' effect can be quanti®ed by potential evaporation; in particular, that reducing potential evaporation limits the damage caused by salinity (Li et al., 2001). However, natural conditions are seldom constant: for instance, growers may have to use irrigation water of changing quality; or rainfall may wash out salts accumulated in the root environment. Not much is known about plant response to changing root-zone salinity. AlarcoÂn et al. (1994) showed that a decrease in relative growth rate and leaf area ratio in tomato plants, in response to increasing osmotic pressure of the nutrient solution, could be detected within an experimental period of 17 days. Van de Sanden and Uittien (1995) showed that the fruit growth rate of tomato decreased and that the decrease of fruit size was related to relative exposure time at high salinity. In the present study, the response of a tomato crop to a step change of root-zone salinity was observed in order to establish to what extent plants recover after a prolonged exposure to high salinity. Therefore the response to a step-change from a 5-month-exposure to 9 dS m 1 root-zone salinity (EC) to 2 dS m 1, rather than the other way round, was determined in this experiment. In addition, the response under two levels of potential transpiration was determined to get some insight into the response dynamics and underlying mechanisms. 2. Materials and methods The general set-up of the series of experiments has been described in detail before (Li et al., 2001) and only the points that are relevant to this paper will be outlined here. Tomato, cv Chaser, was grown in rockwool slabs in two compartments (300 m3 each) of a Venlo-type glasshouse. The two compartments were controlled so as to have two potentialtranspiration climates, as explained below. Two (constant) salinity treatments (expressed by electrical conductivity, EC, dS m 1) were given each to one-half of the rows in each compartment. Oversupply (drain fraction at least 70%) ensured that the EC of drain water (that was monitored) was practically the same as the EC of the irrigation water. In addition, the concentration of the nutrient solution in the root-zone was manually controlled twice a week in blended random samples drawn from the slabs. This particular experiment started after the crop had been subjected to two constant-salinity treatments, both 9 dS m 1, between February and June 1997. In one treatment sodium chloride was added to a standard nutrient solution of 2 dS m 1 (NaCl treatment) whereas the other one was a more concentrated version of the standard nutrient solution (high EC, HEC). At the start of the experiment (4 July), the solution containing NaCl was stepwise ¯ushed out (during 10 days), and refreshed with a standard nutrient solution (EC 2 dS m 1 : lowered EC
Treatment
EC (dS m 1)
pH
NH4 (mmol l 1)
NaCl EC# HEC (before) HEC (after)
8.9 2.2 9.1 8.0 CNS (MPa)
6.1 6.1 5.6 5.2
0.1 5.6 0.1 4.4 0.2 21.4 0.2 23.3 Fe (mmol l 1) Mn (mmol l 1)
11.2 5.1 6.9 2.4 23.8 11.8 19.4 7.8 Zn (mmol l 1) B (mmol l 1)
22.7 13.0 54.3 37.0
8.6 3.5 13.3 7.7
NaCl EC# HEC (before) HEC (after)
0.37 0.09 0.35 0.30
K (mmol l 1)
7.1 2.0 29.4 25.5
Ca (mmol l 1)
Mg (mmol l 1)
64.6 28.5 166.6 117.0
NO3 (mmol l 1)
SO4 (mmol l 1)
19.7 7.3 14.2 4.6 56.4 13.6 51.0 11.7 Cu (mmol l 1) Mo (mmol l 1) 2.7 1.4 4.8 4.1
H2PO4 Na Cl (mmol l 1) (mmol l 1) (mmol l 1) 1.5 0.6 4.4 4.9
49.6 3.2b 6.3 5.1
52.4 0.8b 4.7 2.0
1.6 0.9 1.8 0.9
a Also given is the calculated osmotic potential of the nutrient solution (Slatyer, 1967). Values are means of 7 (before) and 2 (after) measurements unless otherwise indicated. b One measurement: the other sample was taken during the phasing out of NaCl.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Table 1 Mean composition of the solution drawn from the rockwool before and after lowering the ECa
269
270
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
treatment, EC#). Nothing was changed in the other treatment. The mean composition of the solution extracted from the rockwool slabs in both treatments is listed in Table 1, for this experiment as well as for the preceding period. Climate in one compartment was controlled according to Dutch standard practice, i.e.: pre-®xed set-points of day- and night-temperature (20 and 18 8C), respectively, with an allowance for solar radiation and crop stage; ambient humidity controlled by setting minimum values (day- and night-time) for vapour pressure de®cit (VPD). This compartment was our reference (high potential transpiration treatment, HET0). Climate control in the other compartment (low transpiration treatment, LET0) aimed at reducing potential transpiration by a third, by manipulating as far as possible only ambient humidity, by a combination of venting and misting. This caused, however, the temperature to be about 1 8C lower (average over the growing period, Fig. 1). The model used for calculating the transpiration rate has been described by Stanghellini (1987) and the climate control algorithm by Stanghellini and Van Meurs (1992). The effectiveness of the transpiration control was checked by determining water uptake (Fig. 2) through the balance of supply and drain data from each treatment. Six groups of four plants were marked at random in the central rows of each treatment. Twice a month, length and width of every leaf (longer than 15 cm) present on one preselected plant in each group were measured non-destructively. Since leaves were routinely picked, as is done in commercial tomato production, there were about 25 leaves each time, whose position from the root was recorded as well. Each 4-plant group was harvested twice a week, and marketable and unmarketable yields (weight and number) were determined. Fruit development period (FDP) was de®ned as the difference between a ``¯owering'' and a ``harvesting'' time, estimated as follows. The position of all fruit trusses with respect to the leaves was recorded on the six plants per treatment that were monitored. Very few exceptions were observed (Li and Stanghellini, 1999) to the characteristic of tomato that there is a ¯ower truss for each three leaves above the ®rst in¯orescence (Shishido and Hori, 1977). Then, the number of the truss ¯owering at any moment (NTf, counting from the root) was estimated by relating it to the leaf observations as follows: NTf 13 NL
N0
Fig. 1. Mean temperature in the high (solid line) and low (dashed one) transpiration greenhouses (28-day moving averages). DOY is day of year.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
271
Fig. 2. Daytime water uptake (litres per plant per day) in low potential transpiration (LET0) vs. the one in high potential transpiration (HET0), between 7 July and 20 August 1997. Symbols: (*) HEC and (*) EC#. Line is the intended treatment: Y 23 X.
where NL is the number of the youngest measured leaf (longer than 7 cm), counting from the root; and N0 the number of leaves below the first source±sink unitÐa truss together with the three leaves immediately below it (Tanaka and Fujita, 1974). The date of ¯owering of each truss was estimated by linear interpolation between such two-a-month observations, one mean date for each treatment. The date of the ®rst harvest from each truss was recorded. A mean (of 24 plants per treatment) harvest date of each truss was determined, and assigned to all fruits of a truss. FDP was de®ned as the time elapsed between the estimated mean dates of ¯owering and harvesting. 3. Results 3.1. Fruit growth Fruits in the NaCl treatment were consistently smaller than the fruits in the corresponding HEC (Fig. 3). This may be caused by the slight difference in osmotic potential (Table 1)
Fig. 3. Difference (g) between mean weight of fruits from the two nutrition treatments before (circles, NaCl± HEC) and after (triangles, EC#±HEC) lowering the EC. Closed and open symbols refer to high and low transpiration, respectively. The arrow indicates the time of lowering the EC. DOY is day of year.
Table 2 Parameters of plant development during the 2 weeks before lowering the EC (before) and two final weeks of the present experiment (after)a High transpiration
1
EC root-zone (dS m ) Leaf length (cm) Fruit trusses (per plant) Fruit weight (g) Fruit DM (%) FDP (days) a
Low transpiration
1996 EC 9
Before
9.3 32.4 0.6 ± 55.0 1.9 6.58 0.03 ±
9.1 37.1 7.0 56.8 6.52 52
After
HEC
!
NaCl
0.6 0.3 0.7 0.05
9.0 37.0 6.9 58.9 6.72 53
0.2 0.2 0.8 0.02
EC#
HEC
1998 EC 2
1996 EC 9
2.2 8.0 2.3 9.0 34.5 0.4 32.3 1.0 36.9 0.9 35.0 0.4 6.3 0.1 6.1 0.3 5.8 0.4 ± 74.2 1.8 45.6 2.3 71.9 0.8 59.2 1.2 ± ± ± 6.78 0.03 46 39 48 ±
Before
After
HEC 9.1 37.0 7.5 59.4 6.31 59
!
NaCl
0.3 0.2 1.1 0.03
8.8 36.3 7.7 58.3 6.42 60
0.5 0.3 1.4 0.02
EC#
HEC
1998 EC 2
2.2 7.9 2.4 35.1 1.1 33.0 0.6 38.9 1.3 6.9 0.3 6.3 0.3 6.5 0.2 76.6 1.6 51.4 1.2 75.9 2.6 ± ± ± 49 40 50
Leaf length is the average of the 15 lowermost leaves on six plants for each treatment. Fruit weight and fruit dry matter content (fruit DM) are the average of marketable fruits. ``Fruit trusses'' is the mean number of trusses on the plants in that time. ``FDP'' is the fruit development period. The codes 1996 and 1998 indicate corresponding values (in the same period as before and after) from a similar crop grown at constant EC 9:5 dS m 1 (high concentrated nutrient solution) in 1996 and EC 2 dS m 1 in 1998 (Li et al., 2001). Figures are mean standard error (n 6, except for fruit DM (%), where n 40).
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
273
Fig. 4. Mean fruit weight (g), of each harvest, from LET0 vs. corresponding weight in HET0. Symbols: (*) HEC, (*) NaCl and (~) EC#. Solid line shows the best-fit relationships of HEC (P < 0:0001: slope 1.074) and dotted line shows of EC# points (P < 0:0001: slope 1.009).
between the two treatments (Sonneveld, 2000). However, the difference between mean weights at the time of starting this experiment (if one does not account for previous history) was not signi®cant (Table 2). After the EC was lowered, the mean weight of fruits harvested in EC# soon became larger than that of the HEC fruits in the same greenhouse, and the difference in weight kept increasing with time (Fig. 3). At the end of the experiment the average fruit weight of EC# was 63 and 49% higher than that of HEC in the high and the low transpiration houses, respectively (Table 2). HEC fruits were larger in the low transpiration house than in the high transpiration one (Fig. 4), as had been observed also the year before (Li et al., 2001). 3.2. Fruit cracking Lowering the EC caused fruit cracking, particularly in the low transpiration treatment (Fig. 5). The ®rst cracked fruits were harvested soon after lowering the EC, whereas the highest proportion of cracked fruits was harvested 17 and 24 days later in LET0 and HET0, respectively, when it exceeded 90% in LET0 and 30% in HET0 (Fig. 5). That cracking was almost exclusively caused by lowering the EC is shown by the lower panel of Fig. 5, where the (absence of) incidence of cracking in the HEC treatment is shown. Over the whole period of the experiment, 44% of all fruits harvested from EC# in LET0 were cracked, compared with 12% in HET0. 3.3. Fruit development period The declining trend of FDP with truss number (Fig. 6) is fully explained by the opposite trend of temperature (refer to Fig. 1), according to the model of de Koning (1994). That model can explain as well the difference in FDP between HET0 and LET0, as caused by the difference in temperature between the two houses. After starting this experiment, FDP in EC# gradually extended, ®nally to become 20% longer than in HEC, in both greenhouses (Fig. 6). If this effect were related to osmotic potential, this could
274
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Fig. 5. Percentage of cracked fruits in each harvest, for the treatments NaCl (squares) and EC# (triangles) in (A) and HEC (circles) in (B). Closed and open symbols indicate HET0 and LET0, respectively. The arrow shows the time of lowering the EC. Vertical bars are plus and minus standard error n 6. The standard error is not shown in (B) for clarity.
explain as well the difference of about 3 days between FDP in NaCl and in HEC in HET0, observed in the ®rst 15 trusses that were harvested before this experiment. The slightly lower root-zone EC in NaCl (Table 2) may have masked this effect in the other house. 3.4. Vegetative growth Fig. 7 shows the mean pro®les of leaf length measured at the start of the experiment, and the boundary of the mean pro®les (the maximum length) measured subsequently (last
Fig. 6. FDP plotted against truss number (referring to the position of the truss counting from the root) in HEC (*, *), NaCl (& , &) and EC# (~, ~). Closed and open symbols represent HET0 and LET0, respectively. The arrows together with lines indicate the trusses that were being harvested at the time of lowering the EC. Points show averages of each truss for 24 plants per treatment.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
275
Fig. 7. Profiles of leaf length in NaCl ( ) and HEC (Ð) when starting this experiment (shown by the two pointers), and the boundary of the mean profiles (the maximum length) measured subsequently in EC# ( ) and HEC (Ð) (last measured profile was 28 August). Leaf number refers to the position of the leaves along the stem, counting from the root. Lines are three-value running averages of mean profiles of 12 plants per treatment.
measured pro®le was 28 August). The abscissa refers to the position of each leaf, starting from the root. Since there was no leaf response to the transpiration treatment, data are pooled for each EC treatment. There was a small difference between HEC and NaCl pro®les at the beginning of the experiment. However, what is important to notice is that leaves that were fully expanded (about 37 cm) when the EC was lowered did not expand much further, whereas smaller leaves grew to become longer than the HEC ones. After 2 months, most leaves that were still present had been formed after the treatment started and the maximum length at full expansion (average of 15 leaves from bottom) in EC# was 6% more than that in HEC (Table 2). This agrees with our previous ®nding (Li and Stanghellini, 2001) that prolonged exposure to high EC decreases leaf length by 3.7% per dS m 1, in excess of 6.5 dS m 1. 4. Discussion 4.1. Fruit growth In a previous paper (Li et al., 2001) it was shown that the main effect of high EC was a decrease of water in¯ow into the fruits. Consequently, lowering the EC at the beginning of this experiment must have increased water in¯ow into fruits, i.e., fresh growth rate. Indeed, the weight of ripe fruits gradually increased with time (Fig. 3) until fruits were harvested that had developed fully under the new EC. The experiment probably ®nished around the time that this stage was reached, and so no plateau could be distinguished in Fig. 3. That the weight of ripe fruits at the end of the experiment did not ``bear memory'' of the previous exposure of the crop to high EC is partly con®rmed by Table 2. The EC 2 values in 1998, put there as reference, refer to fruits that were harvested in the same compartment and period the year thereafter, from a crop subjected to a constant low-EC treatment, during an experiment that was similar in design to this one (Li et al., 2001).
276
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
4.2. Fruit cracking The occurrence of fruit cracking after lowering the EC of the solution was to be expected. Indeed, it is well known that fruits may crack when there is a sudden increase in water availability, as a result of irrigation, or rain after prolonged drought (e.g. Peet, 1992; Opara et al., 1997), or a lowering of the conductivity of the fertiliser solution (Peet, 1992). Cracking is associated with decreased epidermis elasticity at the mature-green or breaker stage (Kamimura et al., 1972; Bakker, 1988) that causes rupture under the stretching caused by increased water in¯ow. In order to pinpoint the relationship between growth stage and susceptibility to cracking, it is useful to describe fruit growth rate in the two cases, since ®nal fruit size and fruit development period before and after lowering the EC are known. The growth curves in Fig. 8 (left panel) are based on the logistic function with the parameters given in Table 2. In both cases a ripening period equal to 20% of FDP has been taken into account (Ho and Hewitt, 1986; Bakker, 1991). For the sake of this analysis, fruit weight has been converted into surface area, assuming a spherical shape and unit density. The right panel of Fig. 8 shows the derivative (growth rate) of the curves ``before'' lowering the EC. It also shows the difference between the growth rates after and before lowering the EC. The abscissa in the right panel of Fig. 8 is the time remaining to harvest, to facilitate comparison with Fig. 5. This may seem confusing, unless one realises that, since Fig. 5 refers to ripe fruits, points to the right of any given day in that ®gure represent fruits that were still unripe at that moment. The right-hand panel of Fig. 8 implies that fruits whose growth rate was maximal at the time EC was lowered would be harvested some 30 days thereafter, whereas fruits subjected to the largest ``strain'' (difference between ``old'' and ``new'' growth rate) would be some 25 days from harvest. Comparison of this with Fig. 5 implies that fruits were most susceptible to cracking when the difference between the growth rate in the new and old situation was maximal, rather than when the absolute growth rate was maximal. This analysis supports the hypothesis that the epidermis tends to crack when the newly required
Fig. 8. Left panel: growth curve of the surface area (cm2) of tomato fruits before and after lowering the EC, in HET0 (Ð) and LET0 (- - -). Right panel: the original growth rate before lowering the EC (i.e., the derivative of the ``before'' curves in the left panel) and the difference between the new (after) and the original (before) growth rate vs. the time remaining to harvest.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
277
expansion is much larger than the previous expansion rate (Kamimura et al., 1972). If the hypothesis of Peet (1992), that rapidly growing fruit might be especially predisposed to cracking were true, the maximum occurrence of cracking would have taken place later. According to Fig. 8, low transpiration would stretch the time-to-harvest in both cases by a few days, which is obviously not consistent with Fig. 5. The one factor taken into account here, however, is ®nal size, which was larger in LET0. For instance, the increased incidence of cracking of tomato fruits under low potential evaporation (high relative humidity, small vapour pressure de®cit) is well documented (e.g. Peet, 1992; Maroto et al., 1995), most recently by Leonardi et al. (2000). Peet (1992) inferred that high humidity effects on fruit cracking were related to gas and water pressure increases, due to an increase in water supply or a temperature increase. The excess water supply (or water pressure increase) within the fruit, when fruit transpiration is small such as in a humid environment (Ehret and Ho, 1986b; Leonardi et al., 1999) can only be relieved by expansion of the epidermis. Indeed, our results seem compatible with the hypothesis that susceptibility to fruit cracking is proportional both to the difference in expansion rate between the new and the old situation and to the difference in water in¯ow and out¯ow to and from the fruits, which is determined by the potential evaporation. 4.3. Fruit development period The difference of FDP between the two greenhouses before this experiment started, could be fully explained by the difference in temperature. Therefore there is no need to postulate a humidity effect, which, indeed Bakker (1991) did not observe. Shortened FDP with NaCl in HET0 might be explained by a sodium-induced reduction of potassium absorption (Hecht-Buchholz et al., 1979), since a low concentration of potassium reduces FDP (Besford and Maw, 1975). Since depressed transpiration is likely to reduce uptake of Na (Tsuchiya et al., 1992), the absence of an FDP response to NaCl in LET0 does not necessarily go against this hypothesis. This may also explain why the FDP extension after lowering EC is apparently greater in HET0 than in LET0. The extension of FDP after lowering the EC is probably comparable with the observations of Mizrahi (1982) that high salinity (6.6 dS m 1) could shorten the duration of fruit development (by 4±15% in various tomato cultivars). A shortening of FDP caused by high EC has been observed in our series of experiments (Li, 2000). This may be quite similar to the shortening of development period under water stress, observed by many researchers (e.g. Salter, 1958; Wolf and Rudich, 1988; Hsiao, 1993). Hsiao (1993) stated that mild to moderate water stress during the generative phase can be bene®cial for tomato plants by promoting early partition of assimilates to fruits and, consequently, early maturity. 4.4. Vegetative growth As shown before (Li and Stanghellini, 2001), the transpiration treatment had no effect on leaf area, which is con®rmed here. Although an enhanced elongation of single, still growing leaves, could be observed soon after lowering the EC, it took the 8-week duration of the experiment before the leaf pro®les signi®cantly diverged. The response of leaves and of fruits to a step change in osmotic pressure in the root-zone is similar in one aspect.
278
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
Organs formed and developed under the new situation are fully adapted to it, despite the prolonged exposure of the plant to high EC. Organs that are already formed can adapt only to a limited extent, as shown in Figs. 3 and 7. Munns et al. (1982) observed the same in a short-term experiment with barley plants. A rapid reversal of the inhibition of leaf growth, after removal NaCl from the root medium, was also observed by Neumann (1993). They all concluded that leaf growth was limited in high salinity by water de®cit in the elongating leaf tissue rather than by ion excess. 5. Conclusion The negative effect of high salinity on growth and yield is mainly related to the water balance of the plant. This is indirectly con®rmed by the higher occurrence of cracked fruit in LET0, a treatment that was likely to make more water available for uptake into the fruits. This paper also shows that the water balance can be restored even after quite a long exposure to high salinity. However, the time course for recovery is comparable to the duration of the organs' development. Acknowledgements The project has been ®nanced by the Dutch Ministry of Agriculture and Environment (research program 256) and by the European Union (ICA3-CT-1999-0009 HORTIMED). Financial support from the World Laboratory, Geneva, Switzerland for the stay in Holland of Ya Ling Li is gratefully acknowledged. We are indebted to the staff of the horticultural farm of IMAG and to Ferry Corver in particular, for the success of the experiments. We owe thanks to Johan van Gaalen for helping with the yield measurements and to Wim Van Meurs and Rein Bijkerk for solving all our problems with the climate control routine and data collection. References AlarcoÂn, J.J., SaÂnchez-Blanco, M.J., BloarõÂn, M.C., Torrecillas, A., 1994. Growth and osmotic adjustment of two tomato cultivars during and after saline stress. Plant Soil 166, 75±82. Bakker, J.C., 1988. Russeting (cuticle cracking) in glasshouse tomatoes in relation to fruit growth. J. Hort. Sci. 63 (3), 459±463. Bakker, J.C., 1991. Analysis of humidity effects on growth and production of glasshouse fruit vegetables. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 155 pp. Besford, R.T., Maw, G.A., 1975. Effect of potassium nutrition on tomato plant growth and fruit development. Plant Soil 42, 395±412. Dalton, F.N., Poss, J.A., 1990. Water transport and salt loading: a unified concept of plant response to salinity. Acta Hort. 278, 187±293. de Koning, A.N.M., 1994. Development and dry matter distribution in glasshouse tomato: a quantitative approach. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 240 pp. Ehret, D.L., Ho, L.C., 1986a. The effect of salinity on dry matter partitioning and fruit growth in tomatoes grown in nutrient film culture. J. Hort. Sci. 61 (3), 361±367. Ehret, D.L., Ho, L.C., 1986b. Effect of osmotic potential in nutrient solution on diurnal growth of tomato fruit. J. Exp. Bot. 37 (182), 1294±1302.
Y.L. Li et al. / Scientia Horticulturae 93 (2002) 267±279
279
Hecht-Buchholz, Ch., Mix, G., Marschner, H., 1979. Effect of NaCl on mineral content and fine structure of cell in plant species with different salt tolerance. In: Proceedings of the Seventh Intern. Coll. Plant Anal. Fert. Problems, Vol. 1, pp. 147±156. Ho, L.C., Hewitt, J.D., 1986. Fruit development. In: Thereon, J.G., Rudich, J. (Eds.), The Tomato Crop: A Scientific Basis for Improvement. Chapman & Hall, London, pp. 201±239. Hsiao, T.C., 1993. Growth and productivity of crops in relation to water status. Acta Hort. 335, 137±148. Kamimura, S., Yoshikawa, H., Ito, K., 1972. Studies on fruit cracking in tomatoes. Bull. Hort. Res. Station Morioka, Iwate, Japan 7, 73±138. Leonardi, C., Baille, A., Guichard, S., 1999. Effects of fruit characteristics and climatic conditions on tomato transpiration in a greenhouse. J. Hort. Sci. Biotechnol. 74 (6), 748±756. Leonardi, C., Guichard, S., Bertin, N., 2000. High vapour pressure deficit influence growth, transpiration and quality of tomato fruits. Sci. Hort. 84, 258±296. Li, Y.L., 2000. Analysis of greenhouse tomato production in relation to salinity and shoot environment. Ph.D. Dissertation. Wageningen University, Wageningen, 96 pp. Li, Y.L., Stanghellini, C., 1999. The leaf characteristics of a tomato plant in rockwool during long season growing. J. Shanxi Agric. Univ. 19 (3), 214±217. Li, Y.L., Stanghellini, C., 2001. Analysis of the effect of EC and potential transpiration on vegetative growth of tomato. Sci. Hort. 89, 9±27. Li, Y.L., Stanghellini, C., Challa, H., 2001. Effect of EC and transpiration on production of greenhouse tomato (Lycopersicon esculentum L.). Sci. Hort., 88, 11±29 Maas, E.V., Hoffman, G.J., 1977. Crop salt tolerance-current assessment. J. Irrig. Drain. Div., Proc. Am. Soc. Civil Eng. 103 (IR2), 115±134. Maroto, J.V., Bardisi, A., LoÂpez, S., Pascual, B., Alagarda, J., 1995. Influence of relative humidity in the appearance of cracking in tomato (Lycopersicon esculentum Mill.) fruit. Acta Hort. 412, 306±312. Mizrahi, Y., 1982. Effect of salinity on tomato fruit ripening. Plant Physiol. 69, 966±970. Munns, R., Greenway, H., Delane, R., Gibbs, J., 1982. Ion concentration and carbohydrate states of the elongating leaf tissue of Hordeum vulgare growing in high external NaCl. II. Cause of the growth reduction. J. Exp. Bot. 33, 574±583. Neumann, P.M., 1993. Rapid and reversible modifications of extension capacity of cell walls in elongation maize leaf tissues responding to root addition and removal of NaCl. Plant Cell Environ. 16, 1107±1114. Opara, L.U., Studman, C.J., Banks, N.H., 1997. Fruit skin splitting and cracking. Hort. Rev. 19, 217±262. Peet, M.M., 1992. Fruit cracking in tomato. HortTechnology 2 (2), 216±223. Salter, P.J., 1958. The effect of difference water-regimes on the growth of plants under glass. IV. Vegetative growth and fruit development in the tomato. J. Hort. Sci. 33, 1±12. Shannon, M.C., Grieve, C.M., 1999. Tolerance of vegetable crops to salinity. Sci. Hort. 78, 5±38. Shishido, Y., Hori, Y., 1977. Studies on translocation and distribution of photosynthetic assimilate in tomato plants. II. Distribution pattern as affected by phyllotaxis. Tohoku J. Agric. Res. 28 (2), 82±95. Slatyer, R.O., 1967. Plant±water Relationships. Academic Press, London, 366 pp. Sonneveld, C., 2000. Effects of salinity on substrate grown vegetables and ornamentals in greenhouse horticulture. Ph.D. Dissertation. Wageningen University, Wageningen, 151 pp. Stanghellini, C., 1987. Transpiration of greenhouse crops: an aid to climate management. Ph.D. Dissertation. Wageningen Agricultural University, Wageningen, 150 pp. Stanghellini, C., Van Meurs, W.Th.M., 1992. Environmental control of greenhouse crop transpiration. J. Agric. Eng. Res. 51, 297±311. Tanaka, A., Fujita, K., 1974. Nutrio-physiological studies on the tomato plant. IV. Source±sink relationship and structure of the source±sink unit. Soil Sci. Plant Nutr. 20, 57±68. Tsuchiya, M., Naito, H., Ehara, H., Ogo, T., 1992. The physiological response of rice to salinity. I. The relationship between Na uptake and transpiration under different humidity and salinity conditions. Jpn. J. Crop Sci. 61 (1), 16±21. Van de Sanden, P.A.C.M., Uittien, J.J., 1995. Root environment water potential and tomato fruit growth. Acta Hort. 401, 531±536. Wolf, S., Rudich, J., 1988. The growth rates of fruits on different parts of the tomato plant and the effect of water stress on dry weight accumulation. Sci. Hort. 34, 1±11.