When Oskar meets Alice: Does a lack of trade-off in r/K-strategies make Prunus serotina a successful invader of European forests?

Forest Ecology and Management 247 (2007) 120–130 www.elsevier.com/locate/foreco When Oskar meets Alice: Does a lack of trade-off in r/K-strategies make Prunus serotina a successful invader of European forests? Déborah Closset-Kopp a, Olivier Chabrerie a, Bertille Valentin b, Hermine Delachapelle b, Guillaume Decocq a,* a University of Picardie Jules Verne, Department of Botany, 1 rue des Louvels, F-80037 Amiens, France b Conservatoire Botanique National de Bailleul, Hameau de Haendries, F-59270 Bailleul, France Received 8 June 2006; received in revised form 12 April 2007; accepted 12 April 2007 Abstract Alien plant invasions result from a complex interaction between the species life traits (i.e. ‘invasiveness’) and the recipient ecosystem attributes (i.e. ‘invasibility’). However, little is known about the demographical strategy of invaders and its plasticity among similar ecosystems. To assess the role of demographical attributes and their interaction with soil and light conditions on the durable integration of an exotic invasive tree species into a recipient forest, we analyzed population structure, sexual and clonal reproduction, and growth characteristics of the American black cherry (Prunus serotina Ehrh.) in a European forest. As seeds, P. serotina is able to enter closed-canopy forests and form a long-living sapling bank, according to the ‘Oskar syndrome’ (no height growth, diameter increment < 0.06 mm year 1). Suppressed saplings typically develop a ‘sit-and-wait’ strategy so that the invader had a head start on native species when a disturbance-induced gap occurs. Once released, suppressed saplings grow rapidly (height growth > 56 cm year 1) to reach the canopy, fill in the gap and produce numerous seeds (6011 per tree on average). During the self-thinning process characterizing the aggrading phase, overtopped saplings die back but subsequently resprout from roots and stumps, going back to ‘Oskar’ stage. This ‘Alice behaviour’ would enable individuals to decrease in size, delay mortality and locally self-maintain in the understories. These results suggest that P. serotina may successfully invade European forests thanks to a combination of traits which fits well the disturbance regime of the recipient ecosystems. It would behave as a shade-tolerant K-strategist in juvenile stages by giving priority to persistence, but as a light-demanding r-strategist once released, by allocating high energy in growth and reproduction. Initial stages of colonisation are weakly affected by soil but strongly by light conditions. # 2007 Elsevier B.V. All rights reserved. Keywords: Biological invasion; r/K demographical strategies; Forest ecosystem; Gap dynamics; Prunus serotina; Resprouting ability; Sexual reproduction; Fluctuating resources 1. Introduction Invasive alien plants are an increasingly important environmental problem, altering natural ecosystems worldwide by displacing native species and modifying key ecological processes (Vitousek et al., 1996; Higgins et al., 1999; Mooney, 1999; Williamson, 1999). It is thus of outstanding importance to understand the processes influencing the invasion of natural ecosystems. Invasion dynamics largely depends upon the complex interaction between the characteristics of non-native * Corresponding author. E-mail address: guillaume.decocq@u-picardie.fr (G. Decocq). 0378-1127/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2007.04.023 species that enable them to efficiently invade an ecosystem (i.e. ‘invasiveness’) and the properties of the new recipient ecosystems that make them susceptible to invasion (i.e. ‘invasibility’) (Williamson, 1996; Lonsdale, 1999; Alpert et al., 2000; Kolar and Lodge, 2001; Lake and Leishman, 2004). The first stages of invasion, i.e. dispersal and recruitment, are thought to be relatively habitat-independent and mainly determined by a species-specific set of traits (Reichard and Hamilton, 1997), but this hypothesis remains poorly tested. Despite extensive research, identifying particular traits that enable a species to invade a new habitat and compete with its new neighbours has been difficult (Williamson and Fitter, 1996; Reichard and Hamilton, 1997; Dukes and Mooney, 1999; Williamson, 1999; Alpert et al., 2000). Traits like short D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 generation time, long fruiting period, large seed number, small long-living seed, anemochory or zoochory, phenotypic plasticity, clonal growth, rapid growth rate, are likely to confer greater invasiveness (Baker, 1965; Bazzaz, 1986; Lodge, 1993; Rejmanek and Richardson, 1996; Williamson and Fitter, 1996; Dukes and Mooney, 1999; Schweitzer and Larson, 1999; Alpert et al., 2000). It is noteworthy that those life history traits are usually considered as characteristic of r-strategists according to the classical MacArthur and Wilson (1967) scheme, i.e. species adapted for colonisation and reproduction in expanding populations. According to the r/K-model, r-strategists characterize unstable environments that provide available energy and/or nutrients, and host few competing species (Pianka, 1970; Silvertown, 1982). It is not surprising therefore that the most invasible habitats are also the most disturbed (Crawley, 1987; Hobbs and Huenneke, 1992; Williamson, 1999; Levine and D’Antonio, 1999; Alpert et al., 2000) or, more generally, those providing the largest resource fluctuations (Davis et al., 2000). On the contrary, stable environments, i.e. habitats with no or low resource fluctuations, are usually dominated by Kstrategists and expected to be more resistant to invasions. Because of the expected trade-off between the r-strategy (i.e. maximisation of population growth and dispersal) and Kstrategy (i.e. maximisation of biomass accumulation and persistence), no plant can theoretically be successfully adapted to both dispersal and persistence (Pianka, 1970; Silvertown, 1982; Krebs, 2001), even if populations are often polymorphic and a plant phenotype may vary according to environmental conditions (Bradshaw, 1965; Schlichting, 1986; Miner et al., 2005). This probably explains that many plant invaders dominate the early-successional stages following disturbance, but progressively decrease or even become extinct over time in the absence of new disturbance (Rejmanek, 1989; Gibson et al., 2000). Indeed, to become permanently established in a relatively undisturbed recipient ecosystem, the invader must be a better competitor than native species. In other terms, it would be able to shift from a r-strategy to a K-strategy that is apparently a paradox. However, recent studies suggest that some invasive tree species (e.g. the European Acer platanoides L. in north America) would be less constrained than natives by this ecological trade-off (Sanford et al., 2003; Webster et al., 2005). Here, we designed a demographic analysis to explore the possibility for an exotic plant species to behave in contravention of the r/K-trade-off rule to successfully invade a forest ecosystem. The use of demographic studies to understand biological invasions is widely seen as a suitable way for developing and testing invasion theory (Mack, 1985; Crawley, 1986; Parker et al., 1999; Sheppard et al., 2002), modelling invasion dynamics (Higgins and Richardson, 1996; Cannas et al., 2003), and elaborating effective management strategies (Navas, 1991; Sheppard, 2000; Sheppard et al., 2002). Exploring population dynamics of invasive plants may also show contrasts between native and exotic ranges and thus help to explain why these species have become more problematic outside their native habitats (Weiss and Milton, 1984; Lonsdale 121 and Segura, 1987; Noble, 1989; Grigulis et al., 2001). Several mechanisms responsible for the stronger competitive ability of a species in its exotic range compared to its native range have been recently suggested, like allelopathy (Callaway and Aschehoug, 2000) or release from pathogens (Mitchell and Power, 2003) for example, which all directly affect plant demography and may vary according to the biogeographical region. Within this framework, our research focuses on the demography and population dynamics of black cherry (Prunus serotina Ehrh.), a North American forest tree which is successfully spreading in European forests. In its native range, it is a bird- and mammal-dispersed, gap-dependent species which is able to colonize openings in mixed temperate forests and transiently dominate forest stands during secondary succession (Auclair and Cottam, 1971; Auclair, 1975). It has been introduced in Europe for ornamental, timber production and soil amelioration purposes since the 17th century. It was considered as naturalized in the first half of the 19th century, but described as an invasive species only in the second half of the 20th century (Starfinger, 1997). It has become a major problem in forestry, since it is able to perpetuate in the canopy and form monospecific stands, particularly on well-drained, acidic, nutrient-poor soils, impeding native tree regeneration. Although the basic ecology of P. serotina is well known in its native range (e.g. Auclair and Cottam, 1971; Auclair, 1975; Marquis, 1990), few demographic works are available in its exotic range, with the notable exception of Starfinger (1991) who has analysed population structure both in native and exotic (Berlin area) ranges. According to his results, the most important trait enabling P. serotina to be a successful invader is its ‘Oskar syndrome’, that is the ability of tree seedlings to survive as ageing juveniles under the dense shade of the closed canopy (Silvertown, 1982). If a canopy gap occurs (e.g. after a windfall), these Oskars grow rapidly to fill it, flower and reproduce (Hibbs and Fischer, 1979). Unfortunately, to date, no information is available about the population processes involved in the P. serotina invasion in the European forest landscapes. Recently, Deckers et al. (2005) have looked at the population structure of P. serotina in Belgium, but their study was conducted within an agricultural landscape and did not concern forest ecosystems. In order to fill in this gap of knowledge, we designed a study on the population dynamics of P. serotina, in a heavily invaded forest. More specifically, we aimed at testing the following hypotheses: (i) Like in its native range, it is a gap-dependent species because light availability controls its growth and/or seed production. (ii) It is more prone to invade acidic, nutrient-poor soils because soil fertility controls its growth and/or reproduction. (iii) It is able to durably incorporate the silvigenetic cycle because it exhibits a set of demographical traits which is lacking in native tree species. 122 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 2. Material and methods Table 1 Sampling design along the soil pH gradient 2.1. Study area Soil type The study was carried out in Compiègne forest, which is located in northern France (498220 N, 28540 E, 32–148 m altitude). This is a temperate deciduous forest covering 14,417 ha. The climate is of oceanic type with moderate continental influences (mean annual temperature of 10.3 8C and annual rainfall of 677 mm). Soils are developed on sedimentary substrates (sands, chalks and limestones) largely covered by quaternary loess. They are distributed among four main soil types (FAO-UNESCO, 1989): calcareous soils (leptosols, 17% of the total forest area), poorly colonized by P. serotina, podzolized soils (podzols, 45%) highly invaded, and more (luvisols) or less (cambisols) leached brown soils (38%) with an intermediate level of colonisation. The major part of the forest is managed as even-aged plantations of Common Beech (Fagus sylvatica, 44% of the forest area), Pedunculate Oak (Quercus robur, 25%) and Scots Pine (Pinus sylvestris, 7%). Management operations consists of thinnings every 5–15 years and clearcuttings every 150–220 years. Some stands at the forest periphery are still managed as hornbeam coppice-with-oak standards. The whole coppice and part of the standards are harvested every 30 years (ONF, 1995). More than half of the forest is currently invaded by P. serotina. The species was probably introduced in the second half of the 19th century for unclear purposes. It has been described as a ‘forest pest’ for the first time in 1970. Its largescale spread has been related to several storm events, quite important in the past 20 years, namely in 1983 (1040 m3 of fallen trees over 43 ha), 1984 (9004 m3 over 333 ha), 1987 (1037 m3 over 317 ha), 1988 (2577 m3 over 200 ha), 1990 (21,000 m3 over 434 ha), 1993 (3524 m3 over 157 ha), 1999 (1942 m3 over 421 ha). 2.2. Methods We sampled P. serotina populations among four main soil types that were arranged along an increasing soil pH gradient from podzols to leptosols through luvisols and cambisols (Table 1). This gradient positively correlates with nutrient availability (Duchaufour, 1995) and thus represents an increasing soil fertility. All the sampled soils were welldrained and exhibit a sandy-silty texture. We used available soil maps to delineate soil types in the field, and checked the profile by examining a soil core. 2.2.1. Sexual reproduction All racemes were counted in June 2003 from 35 randomly selected trees growing on the different soil types (Table 1, FAOUNESCO, 1989). These 35 trees were distributed among 3 types of light conditions: along stand edges (n = 23), within a forest gap (n = 10) and under closed-canopy conditions (n = 2). It was particularly difficult to include trees in the third category since P. serotina usually remained sterile in the shade. Moreover we had to cope with the scarcity of fertile individuals pH Tree sampling for raceme couting Tree sampling for seed germation test Plot settlement for demographical characteristics Podzols Humic Orthic 4.0–4.5 4.5–5.0 – 5 16 15 – 7 Luvisols Acid 4.5–5.5 17 15 7 Cambisols Dystric Luvic Humic Eutric 5.5–6.5 6.0–7.0 6.0–7.5 6.5–7.5 – 2 7 3 16 15 15 15 – – 7 – Leptosols Rendzic 7.0–8.0 1 15 7 35 122 28 Total on certain soil types, which led us to follow an unbalanced sample design. Six racemes from the crown periphery were then randomly chosen from 122 randomly selected trees that were growing in openings along the soil fertility gradient (Table 1). Racemes were bagged in June to avoid bird predation and then harvested after fruit maturation, in early September 2003, to count fruits. Twenty seeds per tree (10 trees per soil subtype, see Table 1) were randomly selected to assess their germination capacity. The endocarp and the seed coat were removed in order to break external dormancy and facilitate water imbibition before incubation. This treatment ensures rapid and high germination rates and avoids a long cold stratification (Suszka, 1967). Seeds without embryo were counted and deleted. Other seeds were kept for treatment by the gibberellin growth hormone (400 mg L 1) and sown in 5.5 cm Petri dishes on filter paper, which was kept moist with distilled water. All Petri dishes were placed on a table in a laboratory, with a thermoperiod at 20 8C day and 16 8C night and a 12 h photoperiod. Seeds were considered to have germinated at radicle emergence. The seeds were incubated for 40 weeks. 2.2.2. Demographical attributes For each of the 4 main soil types, seven 25 m  25 m plots were randomly settled in the invaded forest stands during summer 2004. Within each plot, all stems of P. serotina were counted, distinguishing resprouts from standards (i.e., single stem originating from a seed), and assigned to one of the following cohorts—stage I: seedlings with their cotyledons; stage II: <20-cm saplings; stage III: >20-cm sterile saplings; stage IV: fertile (usually > 7-m) shrubs and trees. Growth stages were preferred to age stages for defining cohorts because of the plastic growth of the plant (Silvertown, 1982). One hundred and ninety-one individuals from stages II and 173 from stage III were then randomly uprooted for percentage estimation of root suckers in two plots. To quantify the light 123 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 arrival at the forest floor, we used the Daubenmire method (1959) which defined six separated cover classes. The two first classes (class 1 and 2 corresponding respectively to 0–5% and 5–25% canopy cover) were assimilated to forest edges. The two next classes (3 and 4 covering 25–50% and 50–75%) corresponded to the forest gap notion, and the last classes (5 and 6: 75–95% and 95–100% coverage) were considered as closed-canopy conditions. In order to define growth variations among cohorts, 40 individuals (16 under closed canopy and 24 in a canopy-gap) were randomly sampled among standards from stage II to IV in 5 other plots. We used an unbalanced sampling design to account for the larger size variability in light conditions. Diameter and height were measured and age was determined by counting tree rings at the stem basis, which is the best method for determining the age of actively growing seedlings (Collet et al., 2002). Finally, 20 light-growing and 4 shade-growing <1.5 m individuals were randomly selected for a more accurate tree ring analysis. They were cut at the stem basis and thin sections were obtained with a microtome and then coloured with methyl green solution. Rings were observed using a binocular microscope, and widths were measured with a micrometer (1/100 mm accuracy). Ring width data were used to analyse plant growth during the first years of life. dynamics. Within the gap, we sampled 30 saplings every 10 cm along a transect randomly disposed from the gap centre to the gap edge. We measured height and diameter on the 30 individuals and counted tree rings at the stem basis of 10 of them. 2.2.4. Statistical analysis Due to the low number of samples and/or to avoid assumptions of normality, all statistical analyses used nonparametric tests. Effects of soil type and canopy cover on raceme and fruit production, seed germination, and stump resprouting were tested using the Kruskal–Wallis’ H-test ( p < 0.05). Population structure (i.e. distribution of individuals among cohorts) and growth attributes (i.e. age, height, stem diameter) were compared between the different light conditions using the Mann and Whitney’s U-test ( p < 0.05). Relations between growth attributes themselves and between growth attributes and distance from the fallen P. serotina in the special case of the gap structure, were analysed by calculating pairwise Spearman rank correlation coefficients (r, p < 0.05). All statistical analyses were implemented using STATISTICA1 software version 5.1 (Statsoft, 1998). Results were expressed as means  standard errors. 3. Results 2.2.3. Gap dynamics To analyse more precisely P. serotina population dynamics into forest gaps, we selected a 12 m-diameter gap, corresponding to a 2 years old P. serotina windfall. As preliminary field surveys have shown that the patterns of P. serotina populations among gaps were highly repeatable, we preferred to intensively survey a single gap than sampling parts of several ones. Individual density and species composition was determined in the whole gap in order to assess species colonisation and gap 3.1. Sexual reproduction A total of 24,468 racemes were counted from the 35 selected trees. A total of 6345 fruits were collected from the 732 racemes bearing by 122 trees (6 racemes per tree on average). The distribution of racemes and fruits among soil types is shown in Table 2. Neither raceme nor seed production were significantly influenced by the soil type. Seed production Table 2 Seed and raceme production of Prunus serotina Ehrh. among different soil types and light or shade conditions Soil type Racemes Seeds n Total MV  S.E. Humic podzols Orthic podzols Acid luvisols Dystric cambisols Luvic cambisols Humic cambisols Eutric cambisols Rendzic leptosols – 5 17 3 2 7 – 1 – 1805 14335 2240 1310 4413 – 165 – 361.0  121.1 843.2  161.1 813.3  203.3 655.0  325.0 630.4  144.5 – – Total 35 24468 Kruskal–Wallis test Light exposition Forest edges Forest gaps Forest understories Kruskal–Wallis test – Seeds/racemes ratio Total MV  S.E. 16 15 15 16 15 15 15 15 797 685 810 814 730 805 750 954 49.8  7.9 45.7  4.3 54.0  6.1 50.9  5.7 48.7  3.1 53.7  4.3 50.0  4.4 63.6  8.23 122 6345 n 4.63 ns 23 10 2 18535 5478 455 805.9  122.1 547.8  125.9 227.5  182.5 4.44 ns 9.3 7.8 9.1 8.4 8.2 8.9 8.6 8.8 – – 6.62 ns – – – – – – – – – – – – – – – – – n: number of samples; MV: mean value; S.E.: standard error. Significance of Kruskal–Wallis tests—ns: not significant, *p < 0.05. 124 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 Table 3 Percentages of empty seeds and germinated seeds among those having an embryo Soil type n % empty seeds % germinated seeds Humic podzols Orthic podzols Acid luvisols Dystric cambisols Luvic cambisols Humic cambisols Eutric cambisols Rendzic leptosols 200 200 200 200 200 200 200 200 25.5  5.2 (b) 24.5  7.7 (b) 11.5  2.5 (ab) 30.0  6.5 (b) 16.0  5.7(ab) 12.0  5.1(ab) 21.0  8.6 (ab) 4.5  1.7 (a) 48.9  7.8 50.6  10.1 55.6  7.0 43.1  9.2 55.4  9.9 62.1  6.0 28.8  7.8 46.9  8.6 Kruskal–Wallis test – 16.8** 8.37 ns n: total number of collected seeds. Significance of Kruskal–Wallis test—ns: not significant, *p < 0.05, **p < 0.01. clearly increased with the light availability even if the difference between the three light environments assessed was not significant due to the low number of fertile trees found under shade conditions. On average, each tree produced 699 racemes, corresponding to 6011 seeds (8.6 seeds per raceme on average). However the variability was high as indicated by large standard error values. Of the 1600 seeds tested for germination capacity, 290 (18%) were without embryo. In the Petri dishes, 669 seeds germinated, representing 51.1% of the 1310 seeds with embryo and 41.8% of the whole tested seeds (Table 3). The proportion of seeds without embryo was significantly higher for seeds from parents growing on dystric cambisols and podzols, compared to rendzic leptosols. Conversely, the soil type did not significantly influence the germination percentage of viable seeds (Table 3). 3.2. Demographical attributes 3.2.1. Population structure The distribution of individuals among cohorts showed important differences between gaps and understories (Fig. 1). It showed a typical reversed J-shape in the shade, and rather a bell shape in full light conditions. Stage I-seedlings were much more numerous under closed canopy (mean value: 61  17%) than in full light conditions (24  7%). Conversely, stage IIIshrubs were more abundant in full light conditions (43  7%) than under dense shade (17  4%). There was no significant difference for stage II and IV. Stage I, II and III-individuals Fig. 1. Distribution of Prunus serotina saplings among cohorts under light (gap) and shade (understorey) conditions and standard error (N = 28. Stage I: seedlings with their cotyledons; stage II: <20 cm saplings; stage III: >20-cm sterile saplings; stage IV: fertile (usually >7 m) shrubs and trees; significance of Mann–Whitney U-tests—ns: not significant, *p < 0.05; **p < 0.01). were significantly much more abundant on acidic soils (i.e. podzols and luvisols) than on cambisols and rendzic leptosols (Table 4). Among the 4994 individuals recorded in the 28 plots, 3603 (72.1%) were stage I-seedlings and 1391 corresponded to saplings (stages II–IV), of which 1148 (82.5%) were standards and 243 (17.5%) were resprouts. The latter consisted of 4.8%, 26.2% and 51.7% resprouted individuals distributed among stages II, III and IV, respectively. This distribution was neither influenced by the soil type (H = 2.39 ns, H = 1.34 ns and H = 2.77 ns for resprouts of stages II, III and IV, respectively), nor by light availability (U = 74.5 ns, U = 58.5 ns and U = 213 ns for stages II, III and IV, respectively). In the 2 plots where saplings were uprooted, 50 (13.7%) of the 364 individuals were root suckers. Under similar light conditions, seedlings (stage I) and saplings (stages II–IV) showed the same growth characteristics. Height was highly correlated with diameter (r = 0.904*** and r = 0.990*** for light and shade conditions, respectively), as well as height with age (r = 0.787*** and r = 0.853***, respectively). In the shade, stage II- and stage III-individuals with a same diameter and height were significantly older than those under full light conditions (Table 5). After the age of 5, individuals growing in the light were significantly taller and bigger than those growing in the shade. The extreme performances were a height of 1.59 m for a 3 years old P. serotina growing in a canopy-gap and of 40 cm for a 10 years old sapling under closed canopy. Table 4 Distribution of individuals among cohorts and soil types Soil type n Stage I, MV  S.E. Stage II, MV  S.E. Stage III, MV  S.E. Stage IV, MV  S.E. Orthic podzols Luvisols Humic cambisols Rendzic leptosols Kruskal–Wallis test 7 7 7 7 7 397.3  160.3 (a) 109.0  60.8 (a) 5.7  2.6 (b) 2.7  2.3 (b) 16.4*** 35.3  12.5 (a) 37.6  14.2 (a) 8.4  1.7 (b) 5.7  4.4 (b) 8.6 * 34.4  18.7 (a) 26.1  6.0 (a) 9.1  4.5 (b) 3.1  3.5 (b) 11.7** 2.1  2.4 1.0  1.2 1.0  1.4 0 5.8 ns n: number of samples; MV: mean value; S.E.: standard error. Significance of Kruskal–Wallis tests—ns: not significant, *p < 0.05; **p < 0.01; ***p < 0.001. 125 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 Table 5 Sapling growth characteristics under light and shade conditions n All stages Diameter (cm) Height (cm) Age 24 Stage II Diameter (cm) Age 3 Stage III Diameter (cm) Age 16 Stage IV Diameter (cm) Age 5 0–5 years Diameter (cm) Height (cm) 11 6–10 years Diameter (cm) Height (cm) 7 +10 years Diameter (cm) Height (cm) 6 Gap, MV  S.E. n Understorey, MV  S.E. U 1.4  0.5 259.3  95.7 12.3  1.8 ns ns ns 0.1  0.0 5.6  0.6 ns ** 0.9  0.4 13.8  2.3 ns *** 6.0  0.5 14.0  0.0 ns ns 0.2  0.0 17.7  3.8 ns ns 0.4  0.3 47.8  20.5 ** ** 2.6  0.9 482.0  157.8 ** ns 16 2.6  0.7 326.5  84.5 8.1  1.2 3 0.2  0.0 3.0  1.0 11 1.2  0.5 6.2  0.9 2 8.6  0.5 17.2  1.2 3 0.3  0.0 38.0  12.7 5 2.2  2.5 268.2  80.0 8 7.5  1.3 923.3  139.0 N: total number of samples; MV: mean value; S.E.: standard error; significance of Mann–Whitney U-tests—ns: not significant; *p < 0.05; **p < 0.01; ***p < 0.001. 3.2.2. Growth characteristics The analysed saplings of the random plots were aged from 2 to 11 years (Fig. 2). All presented a high growth increment during the 2–3 first years of life. At this stage, height ranged from 10 to 25 cm. After that short period, ring width strongly decreased irrespective of the light conditions. Difference of ring width between light and shade conditions became significant from the fourth year. Ring width progressively increased when the sampled saplings were growing in full light conditions. Conversely, ring width remained constant, never exceeding 0.06 mm, for saplings growing in the understories. The lack of significance after the seventh year was due to the low number of samples. Radial increment was significantly higher under light conditions than in the shade whatever the sapling age (Table 6). The same trend was observed for height increment although the difference was rarely significant because of the high variability of values. Fig. 2. Mean values and standard error of ring width of 24 P. serotina individuals growing under light or shade conditions (significance of Mann– Whitney U-tests—ns: not significant, *p < 0.05). 3.3. Single tree fall-induced gap dynamics We recorded 1335 saplings in the 2 years old studied tree-fall, corresponding to a density of 11.8 stems m 2. Those stems corresponded to P. serotina (89%), Q. robur (4%), Sorbus aucuparia (3%), F. sylvatica (3%) and Picea abies (1%) individuals. The P. serotina population structure showed a typical bell shape. Individuals were 7  4 years old, 56.2  19.2 cm tall and 3.5  0.7 mm diameter wide on average. The further from the gap centre, the smaller the saplings (r = 0.963***). We observed the same trend for diameter and age, which progressively decreased from the centre to the margins (r = 0.927*** and r = 0.963*** for the diameter/distance and age/distance correlations, respectively). 4. Discussion 4.1. Sexual reproduction We found a clear trend toward a raceme and seed production increasing with light availability. As very few reproducing adults could be found under dense shade, the statistics have a low power. Regarding our first hypothesis, this confirms that P. serotina is a light-demanding species in its adult stage, and suggests that persistent shade is theoretically able to sterilize adults. This may occur if neighbouring trees developed their crown sufficiently to overtop P. serotina, for example in small gaps. Our results clearly show that both raceme and seed production did not depend on soil conditions. Conversely, the proportion of viable seeds increased from podzols to leptosols through cambisols following the soil nutrient Table 6 Height and diameter increments under light and shade conditions Height increment (cm year 1) All saplings 0–5 years 6–10 years >10 years Radial increment (mm year 1) n Gap, MV  S.E. n Understorey, MV  S.E. U-test Gap, MV  S.E. Understorey, MV  S.E. U-test 24 7 11 6 30.3  5.6 15.1  6.6 32.0  8.9 56.3  9.3 16 5 3 8 17.2  6.5 3.5  0.8 5.6  1.9 29.6  11.6 ns ns * ns 1.2  0.2 0.5  0.1 1.0  0.3 2.2  0.4 0.5  0.2 0.2  0.0 0.2  0.0 0.8  0.3 ** * * * n: number of samples; MV: mean value; S.E.: standard error; significance of Mann–Whitney U-tests—ns: not significant; *p < 0.05; **p < 0.01. 126 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 gradient, but this gradient did not influence germination rate of viable seeds. Overall, we found a moderate proportion of seeds able to germinate in experimental conditions (42% on average). This value is weaker than the one reported in the native range of P. serotina (e.g. 68% germinations in Pennsylvania; Hough, 1965). The high number of seeds produced by tree (i.e. 6011 on average), particularly in full light conditions, suggests that seed viability may not be a limiting factor for sexual reproduction. However, we did not take into consideration seed predation and post-dispersal mortality, though we presumed they were quite high given the numerous perforated pits we observed on the forest floor during field investigations. Surprisingly, the higher the soil fertility, the higher the number of viable seeds that were produced and the lower the invasibility of the stands (see Table 4). Regarding our second hypothesis, this suggests that soil conditions poorly control sexual reproduction, but rather sapling growth and/or adult tree establishment. This may also explains the paradox that, in its native range, P. serotina performs better on nutrient-rich soils (Auclair and Cottam, 1971; Auclair, 1975) while in Europe, it mainly establishes on nutrient-poor soils (Starfinger, 1997). We conclude that sexual reproduction is very efficient and would allow P. serotina to massively disperse throughout the surrounding landscape, and thus colonize new suitable forest habitats. As a seed, P. serotina is able to enter closed-canopy forests, since in the study area it is mainly a mammal-dispersed species. Moreover endozoochory offers several advantages to the invader, as a possible long-distance dispersal, a germination facilitation thanks to seed coat abrasion in the gut, and a fertilizing effect of the faeces at the time of seedling establishment (Shiferaw et al., 2004). 4.2. Demographical attributes 4.2.1. Growth and population structure According to our tree ring analysis, light became a significant control on sapling growth only after the third year following seedling emergence. Growth of P. serotina in its earliest life stages (stages I and II) was independent from light availability. This period would correspond to the utilisation of seed reserve (Grime and Jeffery, 1965). Between the second and third year, we found a decreasing growth rate, irrespective of light conditions. This may corresponds to the acclimatising time for new resource (i.e. soil water and nutrients, light), following exhaustion of seed reserve. In the understorey, the lack of light induced a sharp ring width reduction (<0.06 mm after the fifth year of growth) associated with a strong decrease of height growth (see Tables 4 and 5), indicating growth suppression. For instance, suppressed saplings can reach the age of 6 years despite a 20 cm-height and 0.2 cm-diameter. This growth characteristic corresponds to the ‘Oskar syndrome’ described by Silvertown (1982) after the character in the novel from Günter Grass, ‘Die Blechtrommel’ (1959), who stopped growing at the age of 3. In Compiègne forest, Oskars’ height varied between 10 and 25 cm. This is consistent with Marquis (1990) for northern America, but slightly differs from the values reported by Starfinger (1991) in Berlin area, where Oskars were taller than 1 m. However, in our study area, forest structure and understorey environment were similar to those of the forests studied by Marquis. Moreover, it is well-known that the size of suppressed saplings primarily depends on the light availability in the understorey (Brown et al., 1999; Messier et al., 1999). In our study, saplings sharing similar height and diameter were significantly older under dense shade than in a canopy gap (e.g. difference of 16 years for trees with a diameter and height of 0.7 and 159 cm, respectively). The distribution of individuals among cohorts followed a typical reversed J-shape, with seedlings (stage I) and ‘Oskars’ (stage II) dominating the population (see Fig. 1), that is typical of shade-tolerant species, but may also indicate a strongly expanding population (Silvertown, 1982). As shade is known to improve germination of P. serotina seeds by helping to maintain stable moisture (Marquis, 1990; Mulligan and Munro, 1981), this may partly explain the accumulation of several generations of seedlings and Oskars in the understorey. Hence, P. serotina is able to constitute a long-lived seedling bank, which may be activated as soon as light would be available. This is typically a ‘sit-and-wait’ strategy. Under favourable light conditions, we found an average diameter increment of 0.12 cm year 1,that is far from the one reported in its native range: 0.635 cm year 1 in Pennsylvania (Fowells, 1965), as well as from the one reported in its exotic range: 0.629 cm year 1 in a Belgian agricultural landscape (Deckers et al., 2005). Stage III-saplings strongly dominated the population, indicating an active aggrading stage. This result contrasts with those recently reported by Deckers et al. (2005), since they found that the population’s age structure followed a negative exponential function. However the authors studied a P. serotina metapopulation structure at a landscape scale, within a dense hedgerow network and in full light conditions, though we focused on populations scattered into small forest gaps. Here, seedlings (stage I) may have rapidly reached stage III through stage II without any growth suppression. The scarcity of seedlings may be explained either by the lack of reproducing adults able to replenish the seedling bank, or by a lower efficacy of germination associated with a higher seedling mortality due to unfavourable moisture conditions (Marquis, 1990; Mulligan and Munro, 1981). Stage II-saplings may have been released from suppression and reached stage III. Stage IV-trees remain scarce, either because of a self-thinning process which would prevent most of the stage III-saplings from reaching the canopy or because all stage III-saplings did not have time to reach stage IV yet. P. serotina presented higher height increment than native trees: up to 56.3  9.3 cm year 1 against 40 cm year 1 for F. sylvatica (Schober, 1975) and 20–30 cm year 1 for Q. robur (Lanier, 1994), for example. Rapid height growth is a wellknown trait of other successful gap invaders, such as Ailanthus altissima (Knapp and Canham, 2000) or Robinia pseudoacacia (Lee et al., 2004), for example. We conclude that P. serotina fails to satisfy the welldescribed trade-off between survivorship in the understorey and fast-growth in openings (Canham, 1989; Pacala et al., 1996; Kobe, 1997; Sanford et al., 2003), since it is both a superior D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 survivor in the shade and a superior competitor in gaps by growing faster than shade-intolerant species. This strongly contrasts with Sanford et al. (2003) who concluded that alien woody species persist in intact forest understorey through high survivorship, or successfully occupy canopy gaps through rapid growth rates, but are not capable of both strategies. It is noteworthy that such a lack of trade-off was also suggested for the invasive wetland herb species Lythrum salicaria in northern America, but in experimental conditions (control of nutrient concentration and shade) (Keddy et al., 1994). 4.2.2. Vegetative regeneration Our results show that P. serotina is able to reproduce vegetatively by producing both stump and root suckers. We found a high proportion of stump resprouts among all but stage I cohorts (from 4.8 to 51.7%), irrespective of light conditions, suggesting a high probability for resprouting before maturity. However, this value is lower than the one reported in the native range: from 35.4% to 79.7% among all age classes (Auclair and Cottam, 1971; Auclair, 1975). Basal sprouting in juvenile stages has been reported to promote survival under a variety of stressful conditions, including suppression by canopy trees, and prolong the life span of individuals in case of disturbance (Del Tredici, 2001). We found that 13.7% of stages II and III-saplings were root suckers. This value was probably underestimated since it was particularly difficult to recognize root suckers in the field. For example, they may grow independently after removal from the parent plant (e.g. following natural death of the main stem or disturbance) and thus, may have been counted as genets instead of ramets. Root suckering probably plays an important role in vegetative propagation and clonal growth at a local scale, but may also enhance vegetative expansion when a canopy gap develops (Del Tredici, 2001). Stump and root sprouting are important strategies. When a portion of the main stem has broken, suffered crown dieback, or been cut down, the organism remains alive, retaining live root and stem material, and then resprouts (Putz and Brokaw, 1989). Resprouting compensates for growth reduction and permits to maintain regeneration with low energy costs, rapid and vigorous stem growth. This may also be important for individuals spending their entire lives in the understorey (Greig, 1993), that is potentially the case of P. serotina’s Oskars. We observed that when P. serotina was overtopped during the aggradation phase, it suffered crown dieback and then was able to resprout from stump and roots. This is an important demographic process through which individuals decrease in size and delay mortality (Paciorek et al., 2000; Del Tredici, 2001). This may greatly contribute to P. serotina invasiveness. This kind of alternation of shoot increase (i.e. resprout growth) and shoot reduction (i.e. aboveground biomass die back) may be viewed as a way to optimize resource conservation (Givnish, 1988). This would enable individuals to maximise their fitness for the light resource fluctuation regime and thus, to maintain themselves for a long time in a stochastic environment. To emphasize this important trait and to draw a parallel with the Oskar’s metaphor, we can 127 term it ‘Alice behaviour’ after the character in the Lewis Carroll’s novel ‘Alice’s adventures in wonderland’, who changes her size to better adapt to a changing environment, by eating mushrooms. We conclude that ‘Alice behaviour’ would allow P. serotina to perpetuate individuals under a closed canopy and to persist after a disturbance (e.g. a natural treefall or an artificial clearcut), thus to self-sustain local populations once established. Such long-term survival of resprouted individuals has been documented in temperate forests (Peterson and Pickett, 1991; Harcombe and Marks, 1993; Del Tredici, 2001). Here, this theoretically prevents P. serotina from local extinction, that is of course a major competitive advantage. 4.3. Single tree fall-induced gap dynamics Our results were only based on the analysis of a single gap, that is probably not sufficient to capture the entire variability of the gap dynamics. However, our field observations revealed similar patterns in other forest gaps, irrespective of their size. This study should be pursued further with large number analyses of gap colonisation in order to definitively and statistically valid these observations. In our case study, the zero-event corresponded to the death and fall of an adult black cherry, that had suddenly increased the amount of available resources, particularly solar radiation at the forest floor. According to Oldeman’s model of forest dynamics, this started the innovation phase, which is the first step in the development of a new eco-unit (Oldeman, 1990). The P. serotina population growing in the studied gap had a typical bell-shaped structure, with the highest stem density, biomass and height in the gap centre. Both height and diameter might decrease along the decreasing gap-created light gradient, suggesting a sapling growth rate approximately proportional to light availability. This contrasts with demographical studies conducted in the native area of the species, where population structure into gaps have a reversed bell-shape (i.e. a U-shape), with tallest saplings at the periphery. It has been hypothesized that soil pathogens, particularly Pythium spp., were responsible for this pattern, in accordance with the Janzen-Connell hypothesis (Packer and Clay, 2000, 2003). Conversely, in its exotic range the soil community would not impaired P. serotina recruitment, and thus, allow the invader to reproduce below itself (Reinhart et al., 2003). We found that 2 years after the zero-event (tree fall) P. serotina could represent 89% of the stems. It seemed able to colonize the whole gap area, inaugurating the aggradation phase sensu Oldeman (1990), i.e. the phase beginning with rapid height growth. The high density of its stems may induce a heavily shaded understorey where nearly nothing else can develop, especially juveniles of other tree species. There could also be a negative feedback on conspecific seedlings and saplings, inducing a self-thinning process as indicated by the ring width reduction and the number of resprouts. Seedlings may become suppressed and go back to an Oskar stage. Saplings rather die back and subsequently resprout following the ‘Alice behaviour’. 128 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 The aggradation phase should normally lead to the biostatic phase according to Oldeman’s model, which is the phase of stand maturation. Although we did not accurately study this phase, our field observations on the whole forest allow us some comments. For small gaps (e.g. single treefalls), the aggrading phase may be transient. A higher structural ensemble may progressively be built by neighbouring (or even surviving) higher trees (e.g. beech, oak, maple, hornbeam). Then P. serotina crowns usually die back and follow ‘Alice behaviour’. As a shade-intolerant species, P. serotina would be expected to have few regeneration opportunities in small gaps where it should be overcompeted by shade-tolerant species (Poulson and Platt, 1989; Brokaw and Busing, 2000). Our results clearly show the contrary since it is able to integrate the natural sylvigenetic cycle without invading the whole forest community. Conversely, for large gaps (e.g. large windthrow or clearcut), the aggradation phase may be brought into stasis as long as P. serotina remains alive (at least 30 years according to a longterm invasion monitoring in this forest), so that the biostasic phase is never reached. P. serotina exhibits a shrub or small tree habit (5 m-height) and forms a low, closed carpet (Starfinger, 1991). Hence, it is able to durably build a selfsustaining population thanks to both Oskar syndrome and Alice behaviour. Such expanding populations impair the development of other tree species since few native species are able to develop in the understories, due to the heavy shade and the thick litter layer (Facelli and Pickett, 1991; Bosy and Reader, 1995). This typically characterizes the invasion stage with deleterious effects on ecosystem functioning. We conclude that P. serotina holds two major competitive advantages that could greatly influence the gap capture process. Firstly, the ‘Oskar syndrome’ confers it a head start on species which must disperse into the gap from the surroundings (Peterson and Pickett, 1995). Similarly, the ‘Alice behaviour’ enables its advanced regenerations to persist in the understorey, survive damage during gap creation and exploit the resulting gap (Putz and Brokaw, 1989; Paciorek et al., 2000). It has been shown that residual plants that reestablish vegetatively following disturbance usually achieve large sizes more quickly than those that start from seed (Brokaw and Busing, 2000). It is well documented that gaps are filled mostly by chance occupants (i.e. species that simply happen to be present in the understorey at the time of gap creation) rather than by best adapted species, and thus, pre-gap patterns in the understorey, not postgap partioning, largely determine gap composition (Fraver et al., 1998; Brokaw and Busing, 2000). This ‘sit-andwait’ strategy is quite common among co-occuring tree species in P. serotina’s native range (e.g. Fagus grandifolia, Quercus alba, Acer pensylvanicum, Tsuga canadensis; Silvertown, 1982) but is almost unknown for tree species of European temperate forests. Secondly, as it may grow faster than many common native species once released, it is able to quickly capture gaps and competitively exclude regeneration of native species and thus, form pure stands and ultimately dominate native forests. 4.4. Synthesis of demographical traits: does P. serotina lack the usual trade-off between r- and K-strategies? The r/K-strategy scheme developed by MacArthur and Wilson (1967) distinguishes two contrasting types of life history strategies based upon those which are adapted for dispersal and those adapted for persistence, respectively named r-strategists and K-strategists. The two strategies are often envisaged as opposite ends of a continuum (Gadgil and Solbrig, 1972; Pianka, 1974; Hallé et al., 1978; Bazzaz, 1996). The rselection characterizes unstable environments where interspecific competition is the major driver, while K-selection rather characterizes stable ones, where habitat carrying capacity and both inter- and intra-specific competition determine the outcome. It is postulated that natural selection should maximize r or K but not both, and that there is a trade-off between the two extremes of the continuum (Tilman, 1990; Grime, 2001). Recently, invasive trees were even reported as developing either a stronger K-strategy than native Kstrategists, or a stronger r-strategy than native r-strategists (Petit et al., 2004). However some studies comparing the ecophysiology of native and exotic congeners have also found that exotic species seems to experience less of a trade-off in adaptation to variable light environments relative to native species (Lei and Lechowicz, 1990; Pattison et al., 1998). We consider that P. serotina behaves as a K-strategist when juvenile, since it is able to durably maintain in the recipient community understorey, by providing (1) a very low growth rate, (2) a long life span as an ‘Oskar’ and (3) a strong resprouting capacity following the ‘Alice behaviour’. It is thus able to infiltrate and establish in a closed-canopy forest with many competing species and few free niches. Conversely, P. serotina rather behaves as a r-strategist in its adult life-stage since it tends to maximise population growth and dispersal, thanks to (1) a rapid growth rate, (2) a quite short life-span (usually a few decades) and (3) an early sexual maturity, (4) an abundant seed production and (5) a sexual reproduction weakly influenced by soil properties. It is thus able to rapidly colonize a canopy-gap by monopolising the available resources, particularly light, and build expanding populations. Subsequently it may reproduce, both sexually to disperse seeds and establish new seedling banks throughout the surrounding forest landscape (‘Oskar syndrome’), and vegetatively to locally selfsustain (‘Alice behaviour’). Consistently with our third hypothesis, we conclude that this seeming lack of trade-off between r and K strategy may confer to P. serotina a competitive advantage over native tree species, which may contribute to its invasiveness. A similar strategy may explain the success of other invasive tree species like, e.g. Ailanthus altissima (Knapp and Canham, 2000), Acer platanoides (Sanford et al., 2003; Webster et al., 2005) or Robinia pseudoacacia (Lee et al., 2004). Acknowledgements We thanks Brice Normand, Christine Beugin for their help during field investigations, and Jérôme Jaminon (Office D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 National des Forêts) for technical support and informations about Compiègne forest. We also thank the two anonymous referees for their helpful comments on the initial draft. This work has been financially supported by the French ‘Ministère de l’Ecologie et du Développement Durable’ (INVABIO II program, CR no. 09-D/2003). References Alpert, P., Bone, E., Holzapfel, C., 2000. Invasiveness, invasibility and the role of environmental stress in preventing the spread of non-native plants. Persp. Plant Ecol. Evol. Syst. 3, 52–66. Auclair, A.N., 1975. Sprouting response in Prunus serotina Erhr.: Multi-variate analysis of site, forest structure and growth rate relationships. Am. Midl. Nat. 94, 72–87. Auclair, A.N., Cottam, G., 1971. Dynamics of black cherry (Prunus serotina Erhr.) in Southern Wisconsin Oak Forests. Ecol. Monogr. 41 (2), 153–177. Baker, H.G., 1965. Characteristics and modes of origin of weeds. In: Baker, H.G., Stebbins, C.L. (Eds.), The Genetics of Colonizing Species. Academic Press, New York, USA, pp. 147–169. Bazzaz, F.A., 1986. Life history of colonizing plants: some demographic, genetic, and physiological features. In: Baker, H.G., Mooney, H.A., Drake, J.A. (Eds.), Ecology of Biological Invasions of North America and Hawaii. Ecological Studies. Springer-Verlag, New York, pp. 96–110. Bazzaz, F.A., 1996. Plants in Changing Environments. Linking Physiological, Population and Community Ecology. Cambridge University Press, Cambridge. Bosy, J.L., Reader, R.J., 1995. Mechanisms underlying the suppression of forb seedling emergence by grass (Poa pratensis) litter. Funct. Ecol. 9, 635–639. Bradshaw, A.D., 1965. Evolutionary significance of phenotypic plasticity in plants. Adv. Genet. 13, 115–155. Brokaw, N., Busing, R.T., 2000. Niche versus chance and tree diversity in forest gaps. Trends Ecol. Evol. 15, 183–187. Brown, S.L., Schroeder, P., Kern, J.S., 1999. Spatial distribution of biomass in forests of the eastern USA. For. Ecol. Manage. 123, 81–90. Callaway, R.M., Aschehoug, E.T., 2000. Invasive plants versus their new and old neighbours: a mechanism for exotic invasion. Science 290, 521– 523. Canham, C.D., 1989. Different responses to gaps among shade-tolerant tree species. Ecology 70, 548–550. Cannas, S.A., Marko, D.E., Paez, S.A., 2003. Modelling biological invasions: species traits, species interactions, and habitat heterogeneity. Math. Biosci. 183, 93–110. Collet, C., Lanter, O., Pardos, M., 2002. Effects of canopy opening on the morphology and anatomy of naturally regenerated beech seedlings. Trees 16, 291–298. Crawley, M.J., 1987. What makes a community invasible? In: Gray, A.J., Crawley, M.J., Edwards, P.J. (Eds.), Colonization, Succession and Stability. Blackwell Scientific Publications, Oxford, pp. 429–453. Crawley, M.J., 1986. Life History and Environment. Chapter 8. In: Crawley, M.J. (Ed.), Plant Ecology. Blackwell Scientific Publication, Oxford. Daubenmire, R.F., 1959. A canopy-coverage method of vegetation analysis. Northwest Sci. 33, 43–64. Davis, M.A., Grime, J.P., Thompson, K., 2000. Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol. 88, 528–534. Deckers, B., Verheyen, K., Hermy, M., Muys, B., 2005. Effects of landscape structure on the invasive spread of black cherry Prunus serotina in an agricultural landscape in Flanders, Belgium. Ecography 28, 99–109. Del Tredici, P., 2001. Sprouting in temperate trees: a morphological and ecological review. Bot. Rev. 67, 121–140. Duchaufour, P., 1995. Pédologie, 4ème éd. Masson, p. 324. Dukes, J.S., Mooney, H.A., 1999. Does global change increase the success of biological invaders? Trends Ecol. Evol. 14 (4), 135–139. Facelli, J.M., Pickett, S.T.A., 1991. Plant litter: its dynamics and effects on plant community structure. Bot. Rev. 57, 1–32. FAO-UNESCO, 1989. Carte mondiale des sols, légende révisée , p. 125. 129 Fowells, H.A., 1965. Silvics of forest trees of the United States. U.S. Department of Agriculture, Forest Service Agriculture, Handbook no. 271. , pp. 642–648. Fraver, S., Brokaw, N.V.L., Smith, A.P., 1998. Delimiting the gap phase in the growth cycle of Panamanian forest. J. Trop. Ecol. 14, 674–681. Gadgil, M., Solbrig, O.T., 1972. The concept of r- and K-selection: evidence from wild flowers and some theoretical considerations. Am. Nat. 106, 14–31. Gibson, N., Keighery, G., Keighery, B., 2000. Threatened plant communities of the western Australia 1, the ironstone communities of the Swan and Scott Coastal Plains. J. R. Soc. Western Aust. 83, 1–11. Givnish, T.J., 1988. Adaptation to sun and shade: a whole-plant perspective. Aust. J. Plant Physiol. 15, 63–92. Greig, N., 1993. Regeneration mode in neotropical Piper: habitat and species comparisons. Ecology 74, 2125–2135. Grigulis, K., Shepard, A.W., Ash, J.E., Groves, R.H., 2001. The comparative demography of the pasture weed Echium plantagineum between its native and invaded ranges. J. Appl. Ecol. 38, 281–291. Grime, J.P., 2001. Plant Strategies, Vegetation Processes, and Ecosystem Properties, 2nd ed. John Wiley, Sons Ltd., Chichester. Grime, J.P., Jeffery, D.W., 1965. Seedling establishment in vertical gradients of sunlight. J. Ecol. 53, 621–642. Hallé, F., Oldeman, R.A.A., Tomlinson, P.B., 1978. Tropical Trees and Forests: An Architectural Analysis. Springer, Berlin, Germany. Harcombe, P.A., Marks, P.L., 1993. Five years of tree death in a FagusMagnolia forest, southeast Texas (USA). Oecologia 57, 49–54. Hibbs, D.E., Fischer, B.C., 1979. Sexual and vegetative reproduction of striped mapple (Acer pennsylvanicum L.). Bull. Torrey Bot. Club 106, 222–227. Higgins, S.I., Richardson, D.M., 1996. A review of models of alien plant spread. Ecol. Model. 87, 249–265. Higgins, S.I., Richardson, D.M., Crowling, R.M., Trinder-Smith, T.H., 1999. Predicting the landscape-scale distribution of alien plants and their threat to plant diversity. Conserv. Biol. 13, 303–313. Hobbs, R.J., Huenneke, L.F., 1992. Disturbance, diversity and invasion: implications for conservation. Conserv. Biol. 6, 324–337. Hough, A.F., 1965. Black cherry (Prunus serotina Ehrh.) in silvics of forest trees in the United States. In: Fowells, A. (Ed.), Agriculture Handbook 271. Department of Agriculture, Washington, pp. 539–545. Keddy, P.A., Twolan-Strutt, L., Wisheu, I.C., 1994. Competitive effect and response rankings in 20 wetland plants: are they consistent across three environments? J. Ecol. 82, 635–643. Knapp, L.B., Canham, C.D., 2000. Invasion of an old-growth forest in New York by Ailanthus altissima: sapling growth and recruitment in canopy gaps. Bull. Torrey Bot. Club 127, 307–315. Kobe, R.K., 1997. Carbohydrate allocation to storage as a basis of interspecific variation in sapling survivorship and growth. Oikos 80, 226–233. Kolar, C.S., Lodge, D.M., 2001. Progress in invasion biology: predicting invaders. Trends Ecol. Evol. 16, 199–204. Krebs, C.J., 2001. Ecology: the experimental analysis of distribution and abundance. Benjamin Cummings, New York. Lake, J.C., Leishman, M.R., 2004. Invasion success of exotic plants in natural ecosystems: the role of disturbance, plant attributes and freedom from herbivores. Biol. Conserv. 117, 215–226. Lanier, L., 1994. Précis de sylviculture. Ecole National du Génie Rural des Eaux et Forêts. Nancy, France. Lee, C.S., Cho, H.J., Yi, H., 2004. Stand dynamics of introduced black locust (Robinia pseudoacacia L.) plantation under different disturbance regimes in Korea. For. Ecol. Manage. 189, 281–293. Lei, T.T., Lechowicz, M.J., 1990. Shade adaptation and shade tolerance in saplings of three Acer species from eastern North America. Oecologia 84 (2), 224–228. Levine, J., D’Antonio, C.M., 1999. Elton revisited: a review of evidence linking diversity and invasability. Oikos 87, 15–26. Lodge, D.M., 1993. Biological invasions: lessons for ecology. Trends Ecol. Evol. 8, 133–137. Lonsdale, W.M., Segura, R., 1987. A demographic study of native and introduced populations of Mimosa pigra. In: Lemerle, D., Leys, A.R. (Eds.), Proceeding of the Eighth Australian Weed Conference. Council of Australian Weed Science Societies, Sydney, Australia, pp. 163–166. 130 D. Closset-Kopp et al. / Forest Ecology and Management 247 (2007) 120–130 Lonsdale, W.M., 1999. Global patterns of plant invasions and the concept of invasibility. Ecology 80, 1522–1536. MacArthur, R.H., Wilson, E.O., 1967. The Theory of Island Biogeography. Princeton University Press, Princeton. Mack, R.N., 1985. Invading plants: their potential contribution to population biology. In: White, J.A. (Ed.), Studies on Plant Demography. Academic Press, London, pp. 127–142. Marquis, D.A., 1990. Prunus serotina. In: Burns, R.M., Honkala, B.H. (Eds.), Silvics of North America, vol. 2. Hardwoods Agriculture Handbook, Washington, DC, pp. 594–604. Messier, C., Doucet, R., Ruel, J.C., Claveau, Y., Kelly, C., Lechowicz, M.J., 1999. Functional ecology of advance regeneration in relation to light in boreal forests. Can. J. For. Res. 29, 811–823. Miner, B.G., Sultan, S.E., Morgan, S.G., Padilla, D.K., Relya, R.A., 2005. Ecological consequences of phenotypic plasticity. Trends Ecol. Evol. 20, 685–692. Mitchell, C.E., Power, A.G., 2003. Release of invasive plants from fungal and viral pathogens. Nature 421, 625–627. Mooney, H.A., 1999. Species without frontiers. Nature 397, 665–666. Mulligan, G.A., Munro, D.B., 1981. The biology of Canadian weeds. Prunus virginiana L and Prunus serotina Ehrh. Can. J. Plant Sci. 61, 977–992. Navas, M.L., 1991. Using plant population biology in weed research: a strategy to improve weed management. Weed Res. 31, 171–179. Noble, I.R., 1989. Attributes of invaders and the invading process: terrestrial and vascular plants. In: Drake, J.A., Mooney, H.A., di Castri, F., Groves, R.H., Kruger, F.J., Rejmanek, M., Williamson, M. (Eds.), Biological Invasions: A Global Perspective. John Wiley & Sons, Chichester, pp. 301–313. Office National des Forêts, 1995. Aménagement forestier. Forêt domaniale de Compiègne. Révision d’aménagement 1996–2010. Rapport Interne, France. Oldeman, R.A.A., 1990. Forests: Elements of Silvology. Springer-Verlag, Berlin, Heidelberg. Pacala, S.W., Canham, C.D., Saponara, J.A., Silander, J., Kobe, R.K., Ribbens, E., 1996. Forest models defined by field measurements: estimation, error analysis and dynamics. Ecol. Monogr. 66, 1–44. Paciorek, C.J., Condit, R., Hubbell, S.P., Foster, R.B., 2000. The demographics of resprouting in tree and shrub species of a moist tropical forest. J. Ecol. 88, 765–777. Packer, A., Clay, K., 2000. Soil pathogens and spatial patterns of seedling mortality in a temperate tree. Nature 404, 278–281. Packer, A., Clay, K., 2003. Soil pathogens and Prunus serotina seedling and sapling growth near conspecific trees. Ecology 84, 108–119. Parker, I.M., Simberloff, D., Lonsdale, W.M., Goodell, K., Wonham, M., Williamson, M.H., von Holle, B., Moyle, P.B., Byers, J.E., Goldwasser, L., 1999. Impact: toward a framework for understanding the ecological effects of invaders. Biol. Invasions 1, 3–19. Pattison, R.R., Goldstein, G., Ares, A., 1998. Growth, biomass allocation and photosynthesis of invasive and native Hawaiian rainforest species. Oecologia 117, 449–459. Peterson, C.J., Pickett, S.T.A., 1991. Treefall and resprouting following catastrophic windthrow in an old-growth hemlock-hardwoods forest. For. Ecol. Manage. 42, 205–217. Peterson, C.J., Pickett, S.T.A., 1995. Forest reorganization: a case study in an old-growth forest catastrophic blowdown. Ecology 76, 763–774. Petit, R.J., Bialozyt, R., Garnier-Géré, P., Hampe, A., 2004. Ecology and genetics of tree invasions: from recent introductions to Quaternary migrations. For. Ecol. Manage. 197, 117–137. Pianka, E.R., 1970. On r- and k-selection. Am. Nat. 104, 592–597. Pianka, E.R., 1974. Niche overlap and diffuse competition. Proc. Natl Acad. Sci. 71, 2141–2145. Poulson, T.L., Platt, W.J., 1989. Gap light regimes influence canopy tree diversity. Ecology 70, 553–555. Putz, F.E., Brokaw, V.L., 1989. Sprouting of broken trees on Barro Colorado Island, Panama. Ecology 70, 508–512. Reinhart, K.O., Packer, A., van der Putten, W., Clay, K., 2003. Plant-soil biota interactions and spatial distribution of black cherry in its native and invasive ranges. Ecol. Lett. 6, 1046–1050. Reichard, S.H., Hamilton, C.W., 1997. Predicting invasions of woody plants introduced into North America. Conserv. Biol. 11, 193–203. Rejmanek, M., 1989. Invasibility of plant communities. In: Drake, J.A., Mooney, H.A., di Castri, F., Groves, R.H., Kruger, F.J., Rejmanek, M., Williamson, M. (Eds.), Biological Invasions: A Global Perspective. John Wiley, New York, pp. 369–383. Rejmanek, M., Richardson, D.M., 1996. What attributes make some plant species more invasive? Ecology 77, 1655–1661. Sanford, N.K., Harrington, R.A., Fownes, J.H., 2003. Survival and growth of native and alien woody seedlings in open and understorey environments. For. Ecol. Manage. 183, 377–385. Schlichting, C.D., 1986. The evolution of phenotypic plasticity in plants. Annu. Rev. Ecol. Systemat. 17, 667–693. Schober, R., 1975. Ertragstafeln wichtiger Baumarten. Dritte Auflage Sauerländer Verlag Frankfurt a. M. Schweitzer, J.A., Larson, K.C., 1999. Greater morphological plasticity of exotic honeysuckle species may make them better invaders than native species. J. Torrey Bot. Soc. 1256, 15–23. Sheppard, A.W., 2000. In: B. Sindel, R.G. (Ed.), Weed ecology and population dynamics. Australian Weed Management Systems. F.J. Richardson Publishers, Melbourne, Australia, pp. 39–60. Sheppard, A.W., Hodge, P., Payner, Q., Rees, M., 2002. Factors affecting invasion and persistence of broom Cytisus scoparius in Australia. J. Appl. Ecol. 39, 721–734. Shiferaw, H., Teketay, D., Nemomissa, S., Assefa, F., 2004. Some biological characteristics that foster the invasion of Prosopsis juliflora (Sw.) DC. at Middle Awash Rift Valley Area, north-eastern Ethiopia. J. Arid Environ. 58, 135–154. Silvertown, J.W., 1982. Introduction to Plant Population Ecology. Longman, London. Starfinger, U., 1991. Population biology of an invading tree species Prunus serotina. In: Seitz, A., Loeschke, V. (Eds.), Species Conservation: A Population Biology Approach. Birkhäuser Verlag, Basel, pp. 171–184. Starfinger, U., 1997. Introduction and naturalization of Prunus serotina in central Europe. In: Brock, J.H., Wade, M., Pysek, P., Green, D. (Eds.), Plant invasions: studies from North America and Europe. Backhuys, Leiden, pp. 161–171. Statsoft, I., 1998. STATISTICA for Windows (Computer Program Manual). Statsoft Inc., Tulsa. Suszka, B., 1967. Studia nad spoczynkiem i kielkowaniem nasion roznych gatunkow z rodzaju Prunus L. (Studies on the dormancy and germination of seeds from various species of the genus Prunus L.). Arboretum Kornickie 12, 221–282. Tilman, D., 1990. Constraints and trade-offs: toward a predictive theory of competition and succession. Oikos 58, 3–15. Vitousek, P.M., D’Antonio, C.M., Loope, L.L., Westbrooks, R., 1996. Biological invasions as global environmental change. Am. Sci. 84, 218–228. Webster, C.R., Nelson, K., Wangen, S.R., 2005. Stand dynamics of an invasive tree, Acer platanoides. For. Ecol. Manage. 208, 85–99. Weiss, P.W., Milton, S., 1984. Chrysanthemoides monilifera and Acacia longifolia in Australia and South Africa. In: Dell, B. (Ed.), Proceedings of the Fourth international Conference of Mediterranean Ecosystems. Botany Department University of Western Australia, pp. 159–160. Williamson, M., 1996. Biological Invasions. Chapman & Hall, London, UK. Williamson, M., 1999. Invasions. Ecography 22, 5–12. Williamson, M., Fitter, A., 1996. The characteristics of successful invaders. Biol. Conserv. 78, 163–170.