Fitness consequences of natural variation in ... - Wiley Online Library

New Phytologist

Research

Fitness consequences of natural variation in floodinginduced shoot elongation in Rumex palustris Xin Chen1,2, Eric J. W. Visser1, Hans de Kroon1, Ronald Pierik2, Laurentius A. C. J. Voesenek2 and Heidrun Huber1 1

Department of Experimental Plant Ecology, Institute for Water and Wetland Research, Radboud University Nijmegen, Heyendaalseweg 135, 6525 AJ

Nijmegen, the Netherlands; 2Department of Plant Ecophysiology, Institute of Environmental Biology, Utrecht University, Padualaan 8, 3584 CH Utrecht, the Netherlands

Summary Author for correspondence: Eric J. W. Visser Tel: +31 243653382 Email: [email protected] Received: 21 October 2010 Accepted: 21 December 2010

New Phytologist (2011) 190: 409–420 doi: 10.1111/j.1469-8137.2010.03639.x

Key words: cost–benefit of plasticity, fitness, flooding-induced elongation, natural variation, petioles, Rumex palustris, selection regimes, submergence.

• Plants can respond to their environment by morphological plasticity. Generally, the potential benefits of adaptive plastic responses are beyond doubt under predictable environmental changes. However, the net benefits may be less straightforward when plants encounter temporal stresses, such as flooding in river flood plains. • Here, we tested whether the balance of costs and benefits associated with flooding-induced shoot elongation depends on the flooding regime, by subjecting Rumex palustris plants with different elongation capacity to submergence of different frequency and duration. • Our results showed that reaching the surface by shoot elongation is associated with fitness benefits, as under less frequent, but longer, flooding episodes plants emerging above the floodwater had greater biomass production than plants that were kept below the surface. As we predicted, slow-elongating plants had clear advantages over fast-elongating ones if submergence was frequent but of short duration, indicating that elongation also incurs costs. • Our data suggest that high costs select for weak plasticity under frequent environmental change. In contrast to our predictions, however, fast-elongating plants did not have an overall advantage over slow-elongating plants when floods lasted longer. This indicates that the delicate balance between benefits and costs of flooding-induced elongation depends on the specific characteristics of the flooding regime.

Introduction Plastic changes of the morphology are ubiquitous responses of plants to their environments as these changes can buffer unfavourable environmental conditions and ultimately enhance plant performance (Schmitt & Wulff, 1993; Sultan, 1995; Schmitt et al., 1999; Schlichting & Smith, 2002). Whether plastic responses are selected for depends on the relation between the costs and benefits associated with such responses. Under conditions where environmental conditions change frequently, the selection for plasticity can be expected to be weak because expression of the plastic phenotype may be costly under the alternate environment (DeWitt et al., 1998; Givnish, 2002). However, alternating environmental states are difficult to mimic and therefore

2011 The Authors New Phytologist 2011 New Phytologist Trust

this hypothesis, though frequently conveyed, has rarely been tested. One of the environmental stresses which induces plastic responses and may occur at unpredictable times and with varying duration is flooding. Therefore, flooding is an ideal scenario to test the effect of frequency and duration of a stress invoking plastic morphological changes on selection for plasticity. In river forelands, flooding events are relatively unpredictable and may vary greatly in timing and duration (Nabben et al., 1999; Vervuren et al., 2003). Flooding in these riverine areas typically comprises complete submergence of the shoot and, therefore, imposes strong adverse effects on the growth of the majority of terrestrial plants. Aerobic respiration is strongly restrained underwater as a result of slow oxygen diffusion (reviewed in Visser et al.,

New Phytologist (2011) 190: 409–420 409 www.newphytologist.com

New Phytologist

410 Research

2003; Visser & Voesenek, 2004; Bailey-Serres & Voesenek, 2008), and underwater photosynthesis is reduced as a result of low light intensities and carbon dioxide concentration (Vervuren et al., 2003; Mommer et al., 2006). Consequently, submerged terrestrial plants have only limited carbohydrates available to continue respiration and growth (Das et al., 2005). Plants growing in flood-prone areas have evolved specific morphological adaptations that help ameliorate or avoid the adverse effects of flooding, such as flooding-induced formation of aerenchyma and adventitious roots, which facilitate internal diffusion of gases (Colmer, 2003), and floodinginduced elongation of shoot organs such as internodes and petioles (Ridge, 1987; Kende et al., 1998; Sauter, 2000; Voesenek et al., 2006; Bailey-Serres & Voesenek, 2008; Jackson, 2008; Chen et al., 2010). With the latter response, plants bring leaves closer to the surface into better illuminated water layers (Mommer et al., 2006), and eventually above the water surface. Although it is clear that such elongation will deliver benefits by restoring contact with the air above the floodwater, thus improving internal aeration for aerobic respiration and allowing for partly aerial photosynthesis, it has also been suggested that shoot elongation may be associated with costs (Voesenek et al., 2004; Pierik et al., 2009), as energy and carbohydrates are needed for cell division and elongation. This may ultimately even cause plant death when energy reserves are depleted before reaching the water surface (Das et al., 2005). In rice culture, therefore, genotypes that slow down growth and respiration when flooded are chosen for cultivation in areas prone to sudden deep flooding of short duration, whereas genotypes that strongly elongate during flooding are used in flood plains where deep flooding persists for at least a month, with some of the genotypes being able to elongate by > 25 cm d)1 (reviewed in Bailey-Serres & Voesenek, 2008). Next to depletion of resources, an additional problem may be that leaves that have elongated during submergence are often biomechanically weaker and have a thinner cuticle (Mommer et al., 2005, 2007). These leaves may die soon after flooding subsides, and therefore constitute a cost to plants when flooding lasts only for a short period of time, as, because of the short duration of flooding, plants had not been able to benefit from the investment into leaf elongation. We therefore conclude that the net benefit of flooding-induced elongation may strongly depend on the specific characteristics of the flooding regime. Voesenek et al. (2004) therefore hypothesized that flooding-induced shoot elongation is only beneficial when floods are shallow enough for plants to reach the water surface, and when floods are relatively long-lasting, as under these conditions costs of elongation are balanced by the benefits of improved gas exchange and enhanced photosynthesis. In the present study we aimed to elucidate the fitness consequences of variation in flooding-induced shoot elong-

New Phytologist (2011) 190: 409–420 www.newphytologist.com

ation in Rumex palustris under different flooding regimes. We selected the two most responsive and the two least responsive populations from a previous study on floodinginduced shoot elongation (Chen et al., 2009) and subjected them to a range of treatments mimicking floods of different frequency and duration. The natural variation in floodinginduced shoot elongation provided us with the opportunity to experimentally test fitness consequences of different elongation patterns under a range of flooding regimes. We used survival and biomass as fitness indicators (cf. van der Sman et al., 1993b) to test whether the balance of costs and benefits associated with a given magnitude and elongation rate depends on the specific flooding regime. In general, we expected to find a net benefit of elongation in conditions where leaves are able to reach the water surface, thereby increasing assimilation, that is, under less frequent and longer flooding. When floods are infrequent and last for an extended period of time, the costs of elongation are compensated for and we predicted the fast-elongating genotypes would have a relative advantage over the slowelongating genotypes, as they will be able to reach the water surface and benefit from increased gas exchange earlier. We further predicted that when flooding is frequent and of short duration, plants showing fast elongation would be at a disadvantage compared with slow-elongating plants because, among other costs, the fast-elongating populations lose more leaf biomass after de-submergence. Under these conditions, we expected slow elongation to be selected for. To properly test these predictions, we varied duration and frequency of submergence while keeping the total duration of submergence over the course of the experiment the same. In this way, effects of flooding pattern remain unconfounded with the effects of the intensity of the stress imposed on the plants. For every flood duration we designed two submergence treatments, one in which the plants were allowed to grow leaves above the water surface and express benefits as well as costs, and another in which the leaves were forced to stay under water and in which only the costs of elongation would be expressed.

Materials and Methods Plant growth Rumex palustris Sm. is a terrestrial plant that generally occurs in riverine flood plains, but which can also be found in areas with more stagnant water tables. Most seeds of this species germinate in nonflooded conditions in the spring. In favourable conditions, plants can flower in the first growing season, but they usually postpone flowering until the second year. Flooding in the growing periods often has negative effects on growth and performance, while winter flooding hardly affects the performance of the dormant plants (van Eck et al., 2005). However, the species can


New Phytologist rapidly elongate its leaf petioles upon flooding in the growing season and thereby reach the water surface if the flood water is not too deep (Voesenek et al., 1993). Four populations of R. palustris were selected from a previous experiment based on the leaf elongation during the initial 7 d of complete submergence (Chen et al., 2009). Two of the four populations were characterized by relatively fast elongation (12 ± 0 mm (± SE) per 7 d for the third oldest petiole and 88 ± 4 mm per 7 d for the fifth oldest petiole), while the other two populations had relatively slow elongation (7 ± 0 mm per 7 d for the third oldest petiole and 64 ± 4 mm per 7 d for the fifth oldest petiole). For each of these four populations, seeds from the same eight mother plants were used in the current experiment as in the previous experiment (Chen et al., 2009). Plants growing from seeds of different mother plants are referred to as seed families in the following text. Seeds from each seed family were geminated on filter paper moistened with tap water in separate Petri dishes in a germination cabinet (12 h light, 10–30 lmol m)2 s)1 photosynthetic photon flux density (PPFD), at 25C, and 12 h dark, at 10C) for 10 d. Thereafter seedlings were individually transplanted to plastic pots containing 300 ml mixture of commercially available sieved potting compost (Lentse potgrond, nr 4; Hortimena groep, Elst, the Netherlands) and sand (1 : 1, v : v) and grown in a heated glasshouse. During this pregrowth period, the average temperature in the heated glasshouse was 22C during the day and 15.5C at night. Light was supplemented by high-pressure sodium lamps (Hortilux-Schreder, 600W, Monster, the Netherlands) whenever irradiance in the glasshouse dropped below 400 lmol m)2 s)1 between 06:00 and 22:00 h. After transplanting, pots were covered with transparent plastic film for 3 d to prevent dehydration, after which the foil was removed. Twenty days after transplanting, each pot received 20 ml nutrient solution containing 4 mM N, 3 mM K, 0.5 mM P, 2 mM Ca, 0.5 mM Mg, 1.75 mM S, 90 lM Fe, 65 lM Na, 50 lM Cl, 25 lM B, 2 lM Mn, 2 lM Zn, 0.5 lM Cu and 0.5 lM Mo. Thirty-one days after growing in the heated glasshouse, plants were transplanted to 4 l plastic pots (18 cm in diameter and 17.5 cm in depth). Care was taken not to disturb the roots by transferring the plants with the original soil containing the roots to the larger pots. The pots were filled with the same mixture of sieved potting compost and sand (1 : 1, v : v) and 18 g of a 1 : 1 (w : w) mixture of two types of the slow-release fertilizer Osmocote to make sure that the concentrations of both N and K were sufficient (the first type containing 9% N, 11% P, 18% K, 2% MgO, micronutrients, and the second type containing 15% N, 9% P, 9% K, 3% MgO, micronutrients; Scotts International B.V., Waardenburg, the Netherlands). The slow-release fertilizer used was manufactured to guarantee nutrient release at ambient temperature in moist soil for 8–9 months. After this second transplant,


Research

plants were grown in a nonheated glasshouse to adapt to the outdoor conditions in the experimental garden of Radboud University Nijmegen (the Netherlands) for 15 d. All plants were regularly watered with tap water throughout the experiment. Treatments Plants were subjected to eight treatments which combined different flooding frequencies (including different duration of individual submergence events; Fig. 1) and either allowing plants to elongate or preventing them from elongating above the water surface during submergence. The experiment was performed in outdoor basins of 1.80 m in diameter and 0.90 m in depth (total volume c. 2.3 m3) in which plants could be submerged in tap water. The resulting maximum water level was thus 0.70 m above the pot surface. The basins were filled with water at c. 3 m3 h)1; immediately after filling, oxygen concentrations were assumed to be in equilibrium with atmospheric pressure (21 kPa), and carbon dioxide concentrations were c. 18.5 Pa. Because of the relatively stagnant conditions in the basins, oxygen and carbon dioxide concentrations were likely to vary in time and local position in the basin depending on depth, temperature and wind speed throughout the flooding episodes. At drainage, all water was removed from the basins at a speed of 5 m3 h)1. Plants were watered twice a week during the drained period. The experiment started in April 2007 and lasted for 24 wk. Plants in all treatments were subjected to a total of 12 wk of submerged and 12 wk of drained conditions. We applied four different submergence treatments mimicking different flooding frequencies (Fig. 1). Plants were subjected to 2, 4 or 12 wk of submergence, periods that were alternated with

1 2W

2 4W

3 Early 12 W

4 Late 12 W

Fig. 1 Schematic representation of the flooding frequency treatments. Filled space indicates submerged, and open space indicates drained conditions. 2W (high flooding frequency), 2 wk of submergence alternated with 2 wk of drained condition; 4W (intermediate flooding frequency), 4 wk of submergence alternated with 4 wk of drained condition; early12 W (early prolonged flooding), 12 wk of submergence in the early developmental stages followed by 12 wk of drained conditions; late 12W (late prolonged flooding), 12 wk of submergence started after 10 wk of drained conditions. For all treatments the total duration of submergence was 12 wk, and the total duration of the drained condition was also 12 wk.


411

New Phytologist

412 Research

drained conditions of the same duration. As the plant size at the onset of submergence is likely to affect the ability of plants to respond to submergence, plants were subjected to 12 wk of submergence either at the beginning of the experiment or after 10 wk. In all flooding treatments, half of the plants were prevented from outgrowing the flood water, whereas the other half of the plants could potentially emerge from the water through flooding-induced leaf elongation. The aim of the former treatment was to prevent plants from benefiting from shoot elongation by regaining contact with the atmosphere, enabling us to test for the combined metabolic and structural costs of elongation. This was realized by covering half of each basin with wired netting at 5 cm below the water surface, which prevented leaves from emerging above the water surface. We used standard-brand plasticcoated chicken wire with a mesh width of 1 cm. This size of wiring has proven not to reduce light conditions during submergence, thereby preventing a potential interaction of submergence responses with responses to reduced light availability. In each half of the basins, there was one replicate of every seed family of every population. Four replicates per seed family per population per treatment were used. To keep algal growth in the water to a minimum, water fleas (Daphnia sp.) were added once a week to the basins filled with water. As plant survival can only be assessed after plants had time to regrow after de-submergence (van Eck et al., 2006), plants were harvested after a recovery period under drained conditions of at least 2 wk (see Fig. 1 for the exact duration of the recovery period for the different treatments). Measurements All plants were harvested in October. Survival and root biomass were determined at harvest. As harvesting took place at the end of the growing season, the biomass of roots, which were mainly storage roots, is a good indicator of performance in the coming year and of the resources available potentially to be invested into inflorescences. Leaf biomass was measured as an indicator of present photosynthetic ability and inflorescence weight as an indicator of reproductive allocation. At harvest, roots were carefully washed free of soil substrate and plants were separated into roots, leaves and flowering stalks. The weights of these organs were measured after drying them to constant weights at 70C. For a separate set of plants, petiole porosity was measured using the microbalance method described by Visser & Bo¨gemann (2003). Plant leaf length was measured right before the start of the treatments and after c. 3 months (i.e. halfway through the experiment). All measurements of leaf length were done after the submergence period, immediately upon de-submergence, that is, for plants subjected to the 2 wk submergence treatments after the third flooding episode, and for plants subjected to 4 wk submergence after the second flooding episode. Plants subjected to the late 12 wk treatments were


measured 10 wk after the onset of the experiment to obtain an indication of leaf length before the onset of flooding. Owing to the large size of the basins, which hampered access to the most central pots and to the considerable depth of the flood water, it was impossible to determine when each individual plant reached the water surface, without causing damage to the fragile leaves of the submerged plants (i.e. lifting pots to identify the origin of emerging leaves). We therefore decided not to consistently record the elongation of individual plants during submergence periods. However, we did measure the maximum length of leaves immediately after de-submergence. These data give a good indication of whether plants were able to reach the water surface in a given treatment, and the leaf length over three time points. These data were in line with the more incidental observations on whether leaves had reached the surface in the course of a given flooding episode (data not shown). Statistical analyses Biomass data were analysed using three-way mixed model nested ANOVA, with flooding frequency, preventing plants from reaching the water surface (i.e. with vs without netting), and elongation type (slow vs fast) being the main effects, and population nested within elongation type and seed family nested within populations and elongation type. Flooding frequency, netting and elongation type were treated as fixed factors, and population and seed family as random factors. When higher-order interactions complicate the interpretation of multifactorial designs, the subgroups can be analysed separately to facilitate the interpretation of the results (Montgomery, 1996). As the three-way interaction flooding frequency · netting · elongation type was significant, we analysed each flooding frequency treatment separately with two-way mixed model nested ANOVA. In this analysis, netting and elongation type were the fixed factors, and population was nested within elongation type, and seed family was nested within populations and elongation type. All biomass data were log-transformed before analyses to ascertain homogeneity of the variances. Survival data and the likelihood of flowering were analysed with logistic regression. The model used in these analyses was constructed similarly to the model used in the analyses of the biomass data. All data were analysed using SAS (version 9.1, SAS Institute Inc., Cary, NC, USA).

Results Effects of flooding frequency Flooding frequency had significant effects on the survival of plants (Table 1, Fig. 2), even though the total duration of submergence in the different flooding frequency treatments was the same. Plants continuously submerged for 12 wk


New Phytologist

Research

Table 1 Results of mixed model nested ANOVA for the effects of flooding frequency, prevention from reaching the water surface (by the use of netting), and elongation type (fast vs slow) on survival rate and flower rate of Rumex palustris plants Chi-square

Frequency Preventing to reach the water surface (netting) Elongation type Frequency · netting Frequency · elongation type Netting · elongation type Frequency · netting · elongation type Population (elongation type) Frequency · population (elongation type) Netting · population (elongation type) Frequency · netting · population (elongation type)

df

Survival rate

Flower rate

3 1 1 3 3 1 3 2 6 2 6

46.1512*** 0.0006ns 0.0015ns 0.0684ns 7.9884* 0.0013ns 0.7957ns 0.0273ns 16.7723* 1.0670ns 1.5149ns

31.0332*** 0.0029ns 0.0005ns 0.0130ns 0.5430ns 0.0004ns 0.9959ns 0.0025ns 0.0039ns 0.0012ns 0.0038ns

Chi-squares and their significance are given. Significance levels are as follows: ns, P > 0.1; *, 0.01 < P < 0.05; ***, P < 0.001.

had higher mortality rates than when shorter periods of submergence were alternated with drained conditions. However, the response to extended submergence depended on the size of plants at the onset of submergence. Smaller plants (early prolonged flooding treatments) had lower survival rates and lower biomass production than larger plants (late prolonged flooding treatments). In contrast to the high mortality of plants submerged continuously for 12 wk, surviving plants subjected to continuous late submergence produced more biomass than plants that were subjected to shorter but more frequent periods of submergence. Flooding frequency also significantly affected biomass production and biomass allocation to the different compartments (Fig. 2, Table 2). The high biomass of plants subjected to late prolonged submergence at the end of the experiment was mainly the result of larger allocation to roots. In plants subjected to short and frequent periods of submergence, total biomass as well as root and leaf weights were at their lowest. The frequency and timing of submergence also affected investment into reproduction. Continuous long-term submergence in the early developmental stages resulted in the lowest flowering rates and flower stalk weight of all flooding treatments (Fig. 2, Tables 1, 2). Plants under short, frequent submergence (2 wk flooding treatments) had a flowering probability of nearly 100%, but the weight of flower stalks was much less than in plants subjected to other flooding frequency treatments. Intermediate flooding frequency and continuous late long-term submergence did not appear to affect flowering frequency and flower stalk weight differently (Tables 1, 2). The frequency of submergence significantly affected the length of leaves produced during submergence (Tables 4, 5). At the onset of the experiment, the average longest leaf length (± SE) of both slow- and fast-elongating populations


was 26.6 ± 0.4 cm. Interestingly, in plants subjected to 12 wk of early submergence, the longest leaf lengths at the end of submergence were shorter, rather than longer, than the starting lengths and the lengths at 2 and 4 wk after the onset of submergence. None of the plants had leaves extending above the water surface at the end of the flooding episode (Tables 4, 5). On average, plants subjected to intermediate flooding frequency were able to produce leaves longer than 70 cm, which enabled plants to extend their leaves above the water surface. By contrast, only a small part of the plants subjected to short but frequent periods of flooding were able to reach a leaf length of 70 cm during a given submergence episode. For plants subjected to submergence in the late developmental stages, initial leaf length (± SE) was 60.8 ± 1.4 cm as a result of normal growth in nonflooded conditions before submergence. As a consequence, all plants submerged at late developmental stages were able to reach the water surface within a few days after the onset of submergence (X. Chen et al., pers obs). It should be noted that none of the full-grown leaves that formed underwater survived de-submergence. Only the youngest, just developing and nonelongated leaves did not wilt within a few hours after de-submergence and finally developed into the terrestrial leaf phenotype. Further regrowth of the shoot depended on carbohydrate storage in the root system, given the small size of the remaining leaves. Effects of potentially restoring aerial contact Plants were either allowed or prevented to emerge from the flood water by shoot elongation by covering half of the number of plants by plastic-coated netting just below the water surface. Generally, the ability to extend above the water surface did not affect survival, biomass and flowering significantly (Tables 1, 2). However, the effect of the possibility to restore aerial contact differed among flooding frequency treatments


413

New Phytologist

414 Research 1.2 (a)

25 (b)

Slow, netting Slow, no netting Fast, netting Fast, no netting

20

Total weight (g)

Survival rate

1.0 0.8 0.6 0.4

15 10 5

0.2

(a)

0.0

0

18 (c) 16

3.5 (d) 3.0 2.5

Leaf weight (g)

Root weight (g)

14 12 10 8 6

2.0 1.5 1.0

2

0.5

0

0.0

1.2 (e)

3.5 (f)

1.0

3.0

Flower stalk weight (g)

Flower rate

4

0.8 0.6 0.4 0.2 0.0

2.5 2.0 1.5 1.0 0.5

2W

4 W Early 12 W Late 12 W Frequency of flooding

0.0

2W

4 W Early 12 W Late 12 W Frequency of flooding

Fig. 2 Survival rate (1 = 100% survival) (a), total DW (b), root DW (c), leaf DW (d), flower rate (1, 100% of the plants had an inflorescence) (e), and flower stalk DW (f) of Rumex palustris plants. The first two bars (black and light grey) represent the mean (+ SE) values of the two slow-elongating populations (n = 2) and the second two bars (dark grey and white) represent the mean (+ SE) values for the two fastelongating populations (n = 2). The black and dark grey bars (netting) represent submergence treatments where plants were prevented from extending above the water surface, and the light grey and white bars (no netting) represent plants which were allowed to extend above the water surface. Frequency of flooding abbreviations: 2W, high flooding frequency; 4W, intermediate flooding frequency; early 12W, early prolonged flooding; late 12W, late prolonged flooding. For an explanation of the flooding frequency treatments, see Fig. 1. Statistical analyses of the data are presented in Tables 1–3.

(Table 3). Plants subjected to intermediate flooding frequency which were able to extend above the water surface produced a significantly higher root biomass and tended to invest more into leaf biomass than plants of the same flooding frequency treatments which were prevented from reaching the water surface. An opposite response in root and leaf allocation was observed in plants subjected to prolonged submergence at late developmental stages (Fig. 2, Table 3). Having the potential opportunity to extend above the water surface had no effect on survival or biomass production for


plants being subjected to short and frequent flooding episodes or for plants subjected to prolonged submergence in the early developmental stages (Tables 2, 3), probably because most plants did not reach the water surface in these treatments (Table 4). Plants that were allowed to reach the water surface tended to produce longer, but not shorter, leaves (Tables 4, 5). This effect was strongest for plants subjected to intermediate flooding frequency. Plants subjected to an intermediate (4 wk) frequency of submergence produced significantly


New Phytologist

Research

Table 2 Results of mixed model nested ANOVA for the effects of flooding frequency, prevention from reaching the water surface (by the use of netting), elongation type, population, and seed family of Rumex palustris plants on biomass parameters F-value

Frequency Netting Elongation type Frequency · netting Frequency · elongation type Netting · elongation type Frequency · netting · elongation type Population (elongation type) Frequency · population (elongation type) Netting · population (elongation type) Frequency · netting · population (elongation type) Seed family (elongation type population) Frequency · seed family (elongation type population) Netting · seed family (elongation type population) Frequency · netting · seed family (elongation type population)

df

Root weight

Leaf weight

Stem weight

Total weight

3 1 1 3 3 1 2-3 2 6 2 4-6 28 66-83 28 55-74

50.62*** 2.70 ns 14.52$ 11.82** 1.70ns 22.91* 15.72** 4.94* 3.95** 0.14 ns 0.26 ns 0.90 ns 0.93 ns 1.63* 1.43*

24.64*** 15.80$ 4.69 ns 3.84$ 3.45$ 54.67* 4.30$ 5.39* 2.31* 0.06 ns 0.46 ns 1.08 ns 0.91 ns 1.03 ns 1.05 ns

21.73** 0.01 ns 1.72 ns 2.47 ns 1.18 ns 0.41 ns 0.28 ns 6.96** 2.83* 4.90* 1.29 ns 3.36*** 1.39* 1.18 ns 1.20 ns

49.60*** 1.95 ns 11.68$ 7.39* 2.02 ns 6.95 ns 7.35* 7.30** 4.14** 0.54 ns 0.53 ns 1.03 ns 0.96 ns 1.61* 1.45*

Biomasses were log-transformed before analyses. F-values and their significance are given. Significance levels are as follows: ns, P > 0.1; $ , 0.05 < P < 0.1; *, 0.01 < P < 0.05; **, 0.001 < P < 0.01; ***, P < 0.001.

longer leaves when they were allowed to extend above the water surface, resulting in a significant flooding frequency · netting interaction (Table 5). Plants from the same treatment also produced significantly more root and marginally significantly more leaf and total biomass (Fig. 2, Table 3). Together, these results indicate that plants benefited from elongating above the water surface in terms of increased biomass production. Generally, populations within elongation type also differed significantly in the response of leaf length and investment into flowering. In the 4 wk treatment, populations within elongation type responded significantly differently to having the possibility of outgrowing the floodwater in terms of biomass production. Together, these results indicate that there is variation within elongation type in the ability to actually reach the water surface and subsequent performance (Tables 2, 5). Effects of elongation types Populations were chosen based on their different floodinginduced leaf elongation rate in artificial climate-chamber conditions. These populations kept their potential for differences in elongation rate in the more natural conditions in the large outdoor basins and, despite the similarity of their longest leaf lengths (± SE) at the beginning of the experiment (fast-elongating, 27.0 ± 0.6 cm; slow-elongating, 26.3 ± 0.3 cm), the difference in elongation rate translated into different effects of submergence treatments on plant performance. The effect of elongation type on survival and biomass production depended on the duration and frequency of flooding (Fig. 2, Tables 1, 2). When subjected to frequent and intermediate flooding, plants from fast-elongating


populations produced significantly lower biomass than plants from slow-elongating populations, whereas slow- and fast-elongating plants produced similar biomass in response to prolonged submergence. These results suggest a potential cost of fast elongation in conditions where submergence is short and thus alternating with frequent de-submergence. A significant interaction between the effects of flooding frequency and elongation type on survival occurred (Fig. 2, Table 1). In both the frequent 2 wk submergence and the late, prolonged 12 wk submergence treatments, plants from slow-elongating populations had higher survival probability than plants from fast-elongating ones, whereas plants subjected to intermediate flooding or plants submerged for a prolonged period of time in the early developmental stages showed no clear difference in the survival between fast- and slow-elongating populations. As the capacity of gas diffusion from emerged tissue to the submerged parts of plants depends on the porosity of the tissues, we measured the porosity of the young elongating petioles of submerged plants from slow- and fast-elongating populations. Both groups had similar petiole porosity (26.0 ± 0.4%, and 27.5 ± 0.3%, respectively; n = 16).

Discussion Selection for plastic responses depends on the relation between costs and benefits associated with the expression of these responses, which in turn depends on the frequency and duration of alternate environments (Ernande & Diekmann, 2004; Huber et al., 2004, 2009; van Kleunen & Fischer, 2005; Dechaine et al., 2007). Flooding is an ideal scenario to test the effect of frequency and duration of


415

New Phytologist

416 Research

Table 3 Results of mixed model nested ANOVA for the effects of prevention from reaching the water surface (by the use of netting), elongation type, population, and seed family of Rumex palustris plants on biomass parameters for each flooding frequency separately F-value df

Root weight

2 wk Netting Elongation type Netting · elongation type Population (elongation type) Netting · population (elongation type) Seed family (elongation type population) Netting · seed family (elongation type population)

1 1 1 2 2 28 28

4.33 ns 140.77** 4.19 ns 0.47 ns 0.49 ns 1.20 ns 0.89 ns

2.04 ns 34.77* 0.01 ns 1.56 ns 0.17 ns 0.88 ns 0.70 ns

2.37 ns 8.79$ 5.33 ns 9.12*** 0.44 ns 1.40 ns 0.73 ns

2.44 ns 87.62* 1.53 ns 1.14 ns 0.51 ns 1.10 ns 0.93 ns

4 wk Netting Elongation type Netting · elongation type Population (elongation type) Netting · population (elongation type) Seed family (elongation type population) Netting · seed family (elongation type population)

1 1 1 2 2 28 28

38.16* 28.02* 5.10 ns 3.05$ 1.00 ns 0.65 ns 0.40 ns

9.76$ 10.86$ 0.50 ns 2.97$ 3.53* 0.87 ns 0.69 ns

6.61 ns 6.14 ns 0.02 ns 15.98*** 6.31** 1.54* 0.66 ns

11.70$ 12.56$ 0.48 ns 9.49*** 4.85* 0.64 ns 0.42 ns

Early 12 wk Netting Elongation type Netting · elongation type Population (elongation type) Netting · population (elongation type) Seed family (elongation type population) Netting · seed family (elongation type population)

1 1 1 2 2 10-27 2-20

1.15 ns 0.89 ns 16.67$ 5.16* 0.20 ns 0.48 ns 1.34 ns

1.30 ns 0.26 ns 9.79$ 2.13 ns 0.50 ns 1.37 ns 2.72**

–

1.42 ns 0.81 ns

0.81 ns 0.79 ns 12.97$ 4.81* 0.28 ns 0.47 ns 1.28 ns

Late 12 wk Netting Elongation type Netting · elongation type Population (elongation type) Netting · population (elongation type) Seed family (elongation type population) Netting · seed family (elongation type population)

1 1 1 2 2 28 25-26

0.10 ns 1.15 ns 0.34 ns 11.89*** 2.07 ns 1.31 ns 1.01 ns

0.03 ns 0.47 ns 3.30 ns 11.97*** 0.68 ns 1.15 ns 0.76 ns

270.22** 0.32 ns 1183.96*** 9.11*** 0.00 ns 0.55 ns 0.67 ns

Leaf weight

9.09$ 0.25 ns 0.58 ns 6.30** 0.08 ns 0.90 ns 0.84 ns

Stem weight

3.01 ns – 0.20 ns –

Total weight

Biomasses were log-transformed before analyses. F-values and their significance are given. Significance levels are as follows: ns, P > 0.1; $ , 0.05 < P < 0.1; *, 0.01 < P < 0.05; **, 0.001 < P < 0.01; ***, P < 0.001.

a stress on selection for plasticity, as under natural conditions flooded and nonflooded regimes can alternate with various frequencies. Submergence has been shown to select for flooding-induced shoot elongation, which enables submerged plants to restore contact with the air above the water surface and thus to restore aerobic respiration and aerial photosynthesis (Mommer et al., 2006). Whilst the potential benefits associated with this elongation are well documented (Voesenek et al., 2004; Bailey-Serres & Voesenek, 2008; Pierik et al., 2009), the net benefit of the response may strongly depend on the specific characteristics of the flooding regime. In the present study we aimed to elucidate the fitness consequences of variation in flooding-induced shoot elongation in R. palustris under different flooding regimes, using four populations from our previous study (Chen et al., 2009) that had the most contrasting elongation rates in response


to submergence, yet similar final lengths. We mimicked selection regimes characterized by different duration and frequency of relatively deep flooding. We found that if flooding lasted for an intermediate period of time, plants which were allowed to grow above the water surface had benefits in terms of higher biomass production compared with plants which were kept underwater throughout submergence. This shows that being able to elongate above the water surface is actually associated with fitness benefits. Under short and frequent floods, however, slow-elongating plants that were not able to reach the water surface during submergence had clear advantages over fast-elongating ones, indicating that there are also costs associated with elongation. Our results thus support the general but rarely tested prediction that high costs select for weak plasticity if environments change again before plants can benefit from plastic changes induced by a previous change of environmental conditions. In contrast to


New Phytologist

Research

submergence episodes, as the fast-elongating plants had a disadvantage in terms of reduced survival and biomass accumulation under conditions characterized by frequent submergence and de-submergence. Costs of floodinginduced leaf elongation include both direct costs of investment into elongation and indirect costs of losing leaf biomass when submergence subsides. Submergence-acclimated, elongated leaves are biomechanically weak, and thus tip over and partly break after de-submergence. Owing to thinner cuticles developed under water for better gas exchange (Mommer et al., 2005, 2007), leaves adapted to submergence desiccate quickly upon de-submergence. For future photosynthesis, previously submerged plants depend on leaves that are newly produced in the post-submergence period. Consequently, this leaf turnover forms a considerable loss of investment in flooding-acclimated structures. Although both slow- and fastelongating populations had such losses, the fast ones suffered more because they had initially invested a higher amount of energy and structural carbohydrates into the longer leaves, even though differences in carbohydrate depletion between fast- and slow-elongating plants were not obvious in earlier experiments (Chen et al., 2010). Faster leaf elongation may also require relatively more energy for, for example, rapid cue perception and translation into plastic elongation responses (DeWitt et al., 1998). This disadvantage disappeared if submergence lasted sufficiently long for the fast-elongating plants to actually benefit from their elongation rate. These results also confirm the hypothesis of Voesenek et al. (2004), and elaborated by Bailey-Serres & Voesenek (2008), who predicted that when flooding is short in

Table 4 Effects of flooding frequency on the longest leaf length (cm) present at the end of a given flooding episode (measured halfway through the experiment immediately after the end of a flooding episode; population means ± SE) of Rumex palustris plants

2W 4W Early 12W

Slow, netting

Slow, no netting

Fast, netting

Fast, no netting

58.5 ± 3.0 75.6 ± 3.6 14.9 ± 0.6

59.0 ± 3.5 84.5 ± 6.1 17.6 ± 0.1

63.5 ± 1.3 76.7 ± 0.3 13.4 ± 0.8

67.2 ± 2.3 84.8 ± 1.7 15.2 ± 4.5

Data of plants from the late prolonged flooding treatment were excluded as these plants had just started their first submergence treatment at this stage of the experiment. 2W, high flooding frequency; 4W, intermediate flooding frequency; early 12W, early prolonged flooding; slow, slow-elongating populations; fast, fast-elongating populations; netting, plants were kept under water when submerged; no netting, plants were allowed to extend above water. For explanation of the frequency treatments, see Fig. 1. The depth of the water column during submergence was kept at 70 cm above the pot surface. Statistical analyses are presented in Table 5.

our predictions, however, fast-elongating plants did not have an overall advantage over slow-elongating plants when floods were of longer duration. Here we discuss these partly surprising results in the context of the benefits and costs of the elongation response. Costs of flooding-induced shoot elongation Our results confirm the hypothesis that investment into fast elongation will be selected against under frequent and short

Table 5 Results of mixed model nested ANOVA for the effects of flooding frequency, prevention from reaching the water surface (by the use of netting), elongation type, population and seed family on longest leaf length at the end of a given flooding episode (measured halfway through the experiment immediately after the end of a flooding episode) of Rumex palustris plants

Frequency Netting Elongation type Frequency · netting Frequency · elongation type Netting · elongation type Frequency · netting · elongation type Population (elongation type) Frequency · population (elongation type) Netting · population (elongation type) Frequency · netting · population (elongation type) Seed family (elongation type population) Frequency · seed family (elongation type population) Netting · seed family (elongation type population) Frequency · netting · seed family (elongation type population)

df

All treatments

2 wk

4 wk

Early 12 wk

2 1 1 2 2 1 2 2 4 2 4 28 56 28 56

715.35*** 13.00$ 0.30 ns 13.39* 3.60 ns 0.05 ns 1.29 ns 15.00*** 2.82* 4.41* 0.76 ns 0.70 ns 0.65 ns 0.32 ns 0.49 ns

– 14.52$ 3.24 ns – – 8.92$ – 6.81** – 0.23 ns – 0.91 ns – 0.56 ns –

– 28.16* 0.02 ns – – 0.05 ns – 11.25*** – 2.26 ns – 0.66 ns – 0.35 ns –

– 1.37 ns 0.51 ns – – 0.05 ns – 3.55* – 2.63$ – 0.72 ns – 0.50 ns –

Only length > 0 was used. F-values and their significance are given. Significance levels are as follows: ns, P > 0.1; $, 0.05 < P < 0.1; *, 0.01 < P < 0.05; **, 0.001 < P < 0.01; ***, P < 0.001. [Correction added after online publication 25 January 2011: the significance level of **(0.001 < P < 0.01) has been added.] Data of plants from the late prolonged flooding treatment were excluded as these plants had just started their submergence treatment at this stage of the experiment. The third column shows the results of the analyses including all treatments; in the fourth to sixth column the analyses were done for each flooding treatment separately. These latter three columns hence only report the effects of netting and elongation type.



417

418 Research

duration, costs involved in fast elongation will select against fast elongation rate. Costs of elongation are also found in cultivars of rice, a species from a very different phylogenetic clade and growth form. Setter & Laureles (1996) showed a negative correlation between flooding-induced shoot elongation and survival under complete submergence of rice plants. Although the 1.5-fold difference in elongation in our natural slow- and fast-elongating populations is relatively small compared with the extreme nonelongating vs elongating response (about a sixfold difference) in the experiments with rice, even this relatively small difference in elongation potential found in natural populations of R. palustris resulted in fitness differences. This shows that even a relatively small natural variation in elongation rate is likely to result in different selection depending on the prevalent environmental conditions. It is relatively small variation like this that commonly occurs in natural populations and forms the basis for evolutionary processes taking place in natural conditions. Our results show that this variation is sufficient to affect the ratio between costs and benefits of submergence-induced elongation responses, thereby shaping the evolutionary outcome in response to different selection regimes. Long-lasting flooding in the early developmental stages of plants affected fitness negatively, as it led to a low survival and flower rate. This would be comparable to spring flooding in the field, which can occur soon after germination and last for extended periods of time (Vervuren et al., 2003). Early in the growing period, plants are still in their juvenile stage, and may have to delay their reproduction for one more year. Such an extension of the life cycle may have negative effects on population growth rates, as a result of both higher risk of mortality in case flowering is delayed and an increased generation time (Childs et al., 2004; Shea et al., 2010). It has been suggested that in years with early spring floods, only very few plants survive and that populations have to regrow from the seed bank (Voesenek & Blom, 1992). Moreover, if early spring floods occur in successive years, this may impose a strain on population viability and the populations may eventually get extinct. Benefits of flooding-induced shoot elongation In the current study, plants did benefit from reaching the water surface if flooding lasted for 4 wk, but not if flooding lasted for shorter periods of time, which was consistent with the fact that most plants had reached the surface in the 4 wk treatment but not in the 2 wk treatment (Table 4). This benefit in terms of relatively higher biomass production, compared with plants kept under water throughout submergence, was found for both slow- and fast-elongating populations, indicating that plants can benefit from elongation responses when elongation results in restored contact with the air. Despite comparable costs of elongation, plants of the same seed families that elongated their leaves to the same


New Phytologist extent but were kept below the water surface had a lower biomass than plants that could grow above the water surface and thus benefit from enhanced gas exchange. These results clearly show that there are costs associated with the expression of elongation. These costs may select against elongation if plants cannot reach the water surface, and increased light and CO2 interception thus cannot compensate for the previously incurred costs. This interpretation is supported by the fact that, in rice, artificially reduced elongation by applying plant growth hormone inhibitors resulted in increased survival under complete submergence (Setter & Laureles, 1996). We also expected a positive effect of elongation when flooding was continuous for an extended period of time, as under these conditions plants can benefit from reaching the water surface without frequently losing leaf biomass after de-submergence, and particularly so for fast-elongating plant types. Interestingly, we did not find any benefits of fast elongation in plants subjected to long flooding, either the 4 wk or for the 12 wk submergence treatment. Voesenek et al. (2004) predicted that a fast elongation rate is beneficial when flooding is prolonged and plants can reach the water surface, because the benefits of improved gas exchange and thus aerobic respiration outweigh the costs of leaf elongation. Our prolonged flooding treatment in the early developmental stages meets these criteria, given the fact that most plants had reached the water surface at c. 8 wk before de-submergence (Table 4). However, most leaves that emerged decayed soon after, and new leaves from the same plants were too small to reach the water surface, resulting in much shorter leaf length displayed after 12 wk of submergence than after 4 wk (Table 4). We have no leaf turnover estimates from submerged plants in the field to compare these observations with, but leaves of R. palustris plants in a long-term submergence experiment in climate room conditions showed similar fast turnover rates (Mommer & Visser, 2005). It may therefore be anticipated that, similar to these aquatic leaves, the leaves that reach the water surface only persist for a limited time, and are rapidly replaced by newly formed leaves if the size of carbohydrate reserves, which are at least partially drawn from the tap root (Nabben, 2001), is sufficient to continue production of long leaves. The flooding conditions that we provided were relatively deep, and plants subjected to prolonged flooding at an early developmental stage may have had insufficient resources (i.e. a small tap root size) for such continued development of new large leaves. Different benefits as a result of a 1–2 wk difference in timing of emergence between the slow- and fast-elongating populations may have faded away in the long post-emergence period in this treatment. Therefore, subjecting plants to shallower flooding than applied in the current experiments, and measuring biomass sooner after de-submergence, may be helpful to reveal benefits of fast elongation. Another explanation for the fact that we did not find benefits of fast elongation could have been a potential difference


New Phytologist among populations in the porosity of their shoots and roots. The porosity of submerged tissues determines to a large extent the internal oxygen diffusion towards the cells (Colmer, 2003) and within and between species variation in porosity can have direct growth and fitness consequences (Huber et al., 2009; Pierik et al., 2009). However, the fast- and slowelongating populations in our experiment had similar porosity in the submerged, elongating petioles, implying that these populations were equally capable of transporting gases down to the roots. Therefore, petiole porosity is not likely to be responsible for the lack of benefits of reaching out of the water. When plants lose all shoot biomass during or immediately after submergence, the roots, and particularly the tap roots, may still remain alive, and shoots can regenerate from these roots after de-submergence (Armstrong et al., 1994; Nabben, 2001). When energy and resources are limited, such as during flooding, plants can reallocate biomass from vegetative parts to inflorescences to support reproduction (Heilmeier et al., 1986). Accordingly, we found that in the prolonged flooding treatment at later developmental stages, plants of fast-elongating populations had lower root biomass when they were allowed to grow above the water surface, but a much higher flower stalk biomass than the plants that were kept under water continuously. As R. palustris is a monocarpic species, the plants die after producing seeds. Species with this type of life cycle usually mobilize all stored resources and invest them into reproduction, which may in our case explain the reduced allocation to roots. These results also indicate that plants may respond to submergence with plasticity of their life history (van der Sman et al., 1993a). However, we do not know whether the increased investment into flowering was caused by the fact that these initially larger plants were able to reach the reproductive stage and the other smaller plants had to delay reproduction as a result of the limited resource acquisition, or that investment into the flowering stalks may be an actual adaptation to flooded conditions. The latter could be the case if the flowering stalks, which typically are longer than the leaves, also provide a low-resistance gas diffusion pathway, leading to a better oxygen supply to submerged plant parts (such as shown for Myriophyllum plants by Schuette & Klug, 1995). Conclusion To the best of our knowledge, this is one of the first studies relating the frequency and duration of environmental stresses to the consequences of plasticity, showing that there are actually costs associated with flooding-induced elongation when conditions change rapidly. We have also shown to what extent there is within and among the population natural variation in the consequences of different elongation rates and that slow- and fast-elongating plant types of R. palustris confer different costs and benefits, depending on the specific flooding regime. Our results confirm the


Research

general prediction that high costs select for weak plasticity if environmental changes are more rapid than will allow plants to take advantage of their responses. In contrast to our predictions, fast-elongating plants did not have an overall advantage over slow-elongating plants when floods were of longer duration. As the depth and duration of flooding can be most unpredictable from year to year in natural flood plains, this spatiotemporal variation in flooding regimes is likely to be responsible for the maintenance of genetic variation in flooding responses within populations.

Acknowledgements We are grateful to Harry van de Steeg for collecting the seeds, Katarzyna Banach, Gerard Bo¨gemann, Hannie de Caluwe, Yvette Evers, Sara Go´mez, Marloes Hendriks, Ronald Jansen, Aleid de Jong, Danny Keijsers, Leon Klaassen, Erik Lu¨ckers, Tamara van Mo¨lken, Maartje Reijnders, Gemma Rutten, Marieke Sanders, Annemiek Smit-Tiekstra, Loes Vervoort, Jinfeng Wang and Rob Welschen for assistance in the glasshouse. This study is part of a project financed by the Centre for Wetland Ecology, a partnership of The Netherlands Institute of Ecology, Radboud University Nijmegen, Utrecht University and the University of Amsterdam.

References Armstrong W, Brandle R, Jackson MB. 1994. Mechanisms of flood tolerance in plants. Acta Botanica Neerlandica 43: 307–358. Bailey-Serres J, Voesenek LACJ. 2008. Flooding stress: acclimations and genetic diversity. Annual Reviews in Plant Biology 59: 313–339. Chen X, Huber H, de Kroon H, Peeters AJM, Poorter H, Voesenek LACJ, Visser EJW. 2009. Intraspecific variation in the magnitude and pattern of flooding-induced shoot elongation in Rumex palustris. Annals of Botany 104: 1057–1067. Chen X, Pierik R, Peeters AJM, Poorter H, Visser EJW, Huber H, de Kroon H, Voesenek LACJ. 2010. Endogenous ABA as a key switch for natural variation in flooding-induced shoot elongation. Plant Physiology 154: 969–977. Childs DZ, Rees M, Rose KE, Grubb PJ, Ellner SP. 2004. Evolution of size-dependent flowering in a variable environment: construction and analysis of a stochastic integral projection model. Proceedings of the Royal Society B: Biological Sciences 271: 425–434. Colmer TD. 2003. Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant, Cell & Environment 26: 17–36. Das KK, Sarkar RK, Ismail AM. 2005. Elongation ability and nonstructural carbohydrate levels in relation to submergence tolerance in rice. Plant Science 168: 131–136. Dechaine JM, Johnston JA, Brock MT, Weinig C. 2007. Constraints on the evolution of adaptive plasticity: costs of plasticity to density are expressed in segregating progenies. New Phytologist 176: 874–882. DeWitt TJ, Sih A, Wilson DS. 1998. Costs and limits of phenotypic plasticity. Trends in Ecology and Evolution 13: 77–81. van Eck WHJM, Lenssen JPM, Rengelink RHJ, Blom CWPM, de Kroon H. 2005. Water temperature instead of acclimation stage and oxygen concentration determines responses to winter floods. Aquatic Botany 81: 253–264.


419

420 Research van Eck WHJM, Lenssen JPM, van de Steeg HM, Blom CWPM, de Kroon H. 2006. Seasonal dependent effects of flooding on species survival and zonation. A comparative study of 10 terrestrial plant species. Hydrobiologia 565: 59–69. Ernande B, Diekmann U. 2004. The evolution of phenotypic plasticity in spatially structured environments: implications of intraspecific competition, plasticity costs and environmental characteristics. Journal of Evolutionary Biology 17: 613–628. Givnish TJ. 2002. Ecological constraints on the evolution of plasticity in plants. Evolutionary Ecology 16: 213–242. Heilmeier H, Schulze ED, Whale DM. 1986. Carbon and nitrogen partitioning in the biennial monocarp Arctium tomentosum Mill. Oecologia 70: 466–474. Huber H, Jacobs E, Visser EJW. 2009. Variation in flooding-induced morphological traits in natural populations of white clover (Trifolium repens L.) and their effects on plant performance during soil flooding. Annals of Botany 103: 377–386. Huber H, Kane NC, Heschel MS, Von Wettberg EJ, Banta J, Leuck A-M, Schmitt J. 2004. Frequency and microenvironmental pattern of selection on plastic shade-avoidance traits in a natural population of Impatiens capensis. American Naturalist 163: 548–563. Jackson MB. 2008. Ethylene-promoted elongation: an adaptation to submergence stress. Annals of Botany 101: 229–248. Kende H, van der Knaap E, Cho HT. 1998. Deepwater rice: a model plant to study stem elongation. Plant Physiology 118: 1105–1110. van Kleunen M, Fischer M. 2005. Constraints on the evolution of adaptive phenotypic plasticity in plants. New Phytologist 166: 49–60. Mommer L, Pons TL, Visser EJW. 2006. Photosynthetic consequences of phenotypic plasticity in response to submergence: Rumex palustris as a case study. Journal of Experimental Botany 57: 283–290. Mommer L, Pons TL, Wolters-Arts M, Venema JH, Visser EJW. 2005. Submergence-induced morphological, anatomical and biochemical responses in a terrestrial species affect gas diffusion resistance and photosynthetic performance. Plant Physiology 139: 497–508. Mommer L, Visser EJW. 2005. Underwater photosynthesis in flooded terrestrial plants: a matter of leaf plasticity. Annals of Botany 96: 581–589. Mommer L, Wolters-Arts M, Andersen C, Visser EJW, Pedersen O. 2007. Submergence-induced leaf acclimation in terrestrial species varying in flooding tolerance. New Phytologist 176: 337–345. Montgomery DC. 1996. Design and analysis of experiments, 4th edn. New York, NY, USA: John Wiley and Sons. Nabben RHM. 2001. Metabolic adaptations to flooding-induced oxygen deficiency and post-anoxia stress in Rumex species. PhD thesis, University of Nijmegen, Nijmegen, the Netherlands. Nabben RHM, Blom CWPM, Voesenek LACJ. 1999. Resistance to complete submergence in Rumex species with different life histories: the influence of plant size and light. New Phytologist 144: 313–321. Pierik R, van Aken JM, Voesenek LACJ. 2009. Is elongation-induced leaf emergence beneficial for submerged Rumex species? Annals of Botany 103: 353–357. Ridge I. 1987. Ethylene and growth control in amphibious plants. In: Crawford RMM, ed. Plant life in aquatic and amphibious habitats. Oxford, UK: Blackwell Publishing, 53–76.


New Phytologist Sauter M. 2000. Rice in deep water: how to take heed against a sea of troubles. Naturwissenschaften 87: 289–303. Schlichting CD, Smith H. 2002. Phenotypic plasticity: linking molecular mechanisms with evolutionary outcomes. Evolutionary Ecology 16: 189–211. Schmitt J, Dudley SA, Pigliucci M. 1999. Manipulative approaches to testing adaptive plasticity: phytochrome-mediated shade-avoidance responses in plants. American Naturalist 154: S43–S54. Schmitt J, Wulff RD. 1993. Light spectral quality, phytochrome and plant competition. Trends in Ecology and Evolution 8: 47–51. Schuette JL, Klug MJ. 1995. Evidence for mass flow in flowering individuals of the submersed vascular plant Myriophyllum heterophyllum. Plant Physiology 108: 1251–1258. Setter TL, Laureles EV. 1996. The beneficial effect of reduced elongation growth on submergence tolerance of rice. Journal of Experimental Botany 47: 1551–1559. Shea K, Jongejans E, Skarpaas O, Kelly D, Shepard YW. 2010. Optimal management strategies to control local population growth or population spread may not be the same. Ecological Applications 20: 1148–1161. van der Sman AJM, Blom CWPM, Barendse GWM. 1993a. Flooding resistance and shoot elongation in relation to developmental stage and environmental conditions in Rumex maritimus L. and Rumex palustris Sm. New Phytologist 125: 73–84. van der Sman AJM, Joosten NN, Blom CWPM. 1993b. Flooding regimes and life-history characteristics of short-lived species in river forelands. Journal of Ecology 81: 121–130. Sultan SE. 1995. Phenotypic plasticity and plant adaptation. Acta Botanica Neerlandica 44: 363–383. Vervuren PJA, Blom CWPM, de Kroon H. 2003. Extreme flooding events on the Rhine and the survival and distribution of riparian plant species. Journal of Ecology 91: 135–146. Visser EJW, Bo¨gemann GM. 2003. Measurement of porosity in very small samples of plant tissue. Plant and Soil 253: 81–90. Visser EJW, Voesenek LACJ. 2004. Acclimation to soil flooding – sensing and signal-transduction. Plant and Soil 274: 197–214. Visser EJW, Voesenek LACJ, Vartapetian BB, Jackson MB. 2003. Flooding and plant growth. Annals of Botany 91: 107–109. Voesenek LACJ, Banga M, Thier RH, Mudde CM, Harren FJM, Barendse GWM, Blom CWPM. 1993. Submergence-induced ethylene synthesis, entrapment, and growth in two plant species with contrasting flooding resistances. Plant Physiology 103: 783–791. Voesenek LACJ, Blom CWPM. 1992. Germination and emergence of Rumex in river floodplains. I. Timing of germination and seedbank characteristics. Acta Botanica Neerlandica 41: 319–329. Voesenek LACJ, Colmer TD, Pierik R, Millenaar FF, Peeters AJM. 2006. How plants cope with complete submergence. New Phytologist 170: 213–226. Voesenek LACJ, Rijnders JHGM, Peeters AJM, van de Steeg HM, de Kroon H. 2004. Plant hormones regulate fast shoot elongation under water: from genes to communities. Ecology 85: 16–27.
