Pollen dispersal inferred from paternity analysis in a mixed oak stand of Quercus robur L. and Q. petraea (Matt.) Liebl.

Paternity analysis was used to determine the spatial distribution of male parents of 984 offspring collected from 13 identified mother trees in a natural stand of 5.76 ha and comprising 296 adult trees of Quercus petraea and Q . robur . For seven of the 13 maternal progeny arrays sampled, we found an excess of nearby matings and a preferential direction of pollination. For the remaining progeny arrays, no departure from random distribution of male parents was detected. A common trend among all families was a high percentage (averaging 65% for Q. robur and 69% for Q. petraea ) of offspring that were pollinated by male parents from outside the study site. By pooling the data over all families, the average pollen dispersal curve within the stand was inferred and fitted to a negative exponential distribution. This model extrapolated for distances over the spatial scale of the study stand was insufficient to explain the high level of gene flow detected by the paternity analysis, suggesting a substantial level of long-distance pollination events. The genetic composition of the pollen pools received by each maternal tree was compared and showed significant differentiation that could be attributed to differences in male reproductive success. By contrast, no significant differentiation between the pollen clouds from outside and inside the study stand was detected, suggesting genetic homogeneity between the surrounding forest and the study stand.     

Pollen and seed dispersal inferred from seedling genotypes: the Bayesian revolution has passed here too

Understanding precisely how plants disperse their seeds and pollen in their neighbourhood is a central question for both ecologists and evolutionary biologists because seed and pollen dispersal governs both the rate of spread of an expanding population and gene flow within and among populations. The concept of a 'dispersal kernel' has become extremely popular in dispersal ecology as a tool that summarizes how dispersal distributes individuals and genes in space and at a given scale. In this issue of Molecular Ecology, the study by Moran & Clark (2011) (M&C in the following) shows how genotypic and spatial data of established seedlings can be analysed in a Bayesian framework to estimate jointly the pollen and seed dispersal kernels and finally derive a parentage analysis from a full-probability approach. This approach applied to red oak shows important dispersal of seeds (138 m on average) and pollen (178 m on average). For seeds, this estimate contrasts with previous results from inverse modelling on seed trap data (9.3 m). This research gathers several methodological advances made in recent years in two research communities and could become a cornerstone for dispersal ecology.     

A comment on García et al. (2005, 2007) and related papers on mating patterns and gene dispersal in Prunus mahaleb

In two studies on mating patterns and spatial components of pollen and seed dispersal of Prunus mahaleb based on parentage analysis, García et al. (2005, 2007) depicted their 196 focal trees as a spatially isolated population where all reproductive trees had been genotyped. Additional distributional data for P. mahaleb trees in their study area, however, revealed that García and colleagues’ depiction of their study system bears little resemblance to reality. The trees these authors studied did not form a discrete, geographically isolated population. Around 300 ungenotyped reproductive trees occurred within the 1.5‐km distributional gap to the nearest population proclaimed by García and colleagues. Since exhaustive sampling of potential parental genotypes is essential in parentage analyses, the occurrence of a large number of ungenotyped trees in the immediate neighbourhood of focal trees can severely affect the main conclusions of García et al. (2005, 2007) as well as of several related publications on gene dispersal and mating patterns of P. mahaleb conducted on the same trees and relying on the same false premises of spatial isolation and exhaustive sampling.     

Patterns and levels of gene flow in Rhododendron metternichii var. hondoense revealed by microsatellite analysis

Parentage analysis was conducted to elucidate the patterns and levels of gene flow in Rhododendron metternichii Sieb. et Zucc. var. hondoense Nakai in a 150 x 70 m quadrant in Hiroshima Prefecture, western Japan. The population of R. metternichii occurred as three subpopulations at the study site. Seventy seedlings were randomly collected from each of three 10 x 10 m plots (S1, S2, and S3) on the forest floor of each subpopulation (A1, A2, and A3). Almost all parents (93.8%) of the 70 seedlings were unambiguously identified by using 12 pairs of microsatellite markers. Within the quadrant, adult trees less than 5 m from the centre of the seedling bank (plots S1, S2, and S3) produced large numbers of seedlings. The effects of tree height and distance from the seedling bank on the relative fertilities of adult trees were highly variable among subpopulations because of the differences in population structure near the seedling bank: neither distance nor tree height had any significant effect in subpopulation A1; distance from the seedling bank had a significant effect in subpopulation A2; and tree height had a significant effect in subpopulation A3. Although gene flow within each subpopulation was highly restricted to less than 25 m and gene flow among the three subpopulations was extremely small (0-2%), long-distance gene flow from outside the quadrant reached 50%. This long-distance gene flow may be caused by a combination of topographical and vegetational heterogeneity, differences in flowering phenology, and genetic substructuring within subpopulations.     

Tree rooting patterns and soil water relations of healthy and damaged stands of mature oak (Quercus robur L. and Quercus petraea [Matt.] Liebl.)

At three sites in northwestern Germany, which represent the centres of the present oak damage, root distribution and biomass beneath healthy and damaged trees of mature pedunculate oak (Quercus robur L.; Neuenburg site) and sessile oak (Q. petraea [Matt.] Liebl.; Lappwald and Sprakensehl sites) were investigated, and soil texture, bulk density, duration of waterlogging periods and the water available in the mineral soil were determined. For Neuenburg and Sprakensehl, the available soil water was related to leaf water parameters determined in a separate investigation. At the clayey and hydromorphic sites of Neuenburg and Lappwald, the measurements were performed in each one healthy and one damaged part of the site, which differed in the number of oaks with crown damage. In the damaged stand of Neuenburg, the clay content of the subsoil was higher than in the healthy stand, and the soil water availability was reduced especially in dry periods. Compared to healthy oaks of the healthy stand, the density of finest plus fine roots as well as the biomasses of finest roots were lower beneath damaged oaks of the damaged stand. With decreasing relative available soil water (actually available water in relation to water available at the saturation state), the relative leaf water content decreased in damaged, but not in healthy oaks. At Lappwald, similar differences in soil water availability between the healthy and the damaged stand were found, but had no effect on the distribution or biomass of the roots. At the sandy site (Sprakensehl), the available soil water decreased drastically during a dry period, and predawn leaf water potentials of both healthy and damaged oaks declined with decreasing relative available soil water. However, the damaged oaks were not inferior to the healthy ones with respect to root density and biomass. It is concluded that, in the damaged stand of Neuenburg, the high clay content of the subsoil, which results in prolonged periods of waterlogging, in sharp changes from waterlogging to drought and decreased water availability in dry periods, is the reason for the reduced biomass and density of roots of the pedunculate oak. Thus, in northwestern Germany, unfavourable soil water relations are considered as a factor contributing to crown damage of pedunculate oak at hydromorphic sites, but not to damage of sessile oak.     

Natural hybridisation between Quercus petraea (Matt.) Liebl. and Quercus pubescens Willd. within an Italian stand as revealed by microsatellite fingerprinting

Interspecific gene flow is frequently reported in the genus Quercus . However, interfertile oak species often seem to remain distinct, even within areas of sympatry. This study employed molecular markers to verify, at a fine scale, the presence of interspecific gene flow in a natural population of Quercus petraea and Quercus pubescens . Within a delimited area of 6 ha, all adult trees belonging to the studied oak complex and seeds from a subsample of such trees were collected and analysed using molecular microsatellite markers. A low interspecific genetic differentiation and a high level of interspecific genetic admixture suggested past hybridisation. Paternity inference of seeds allowed the estimation of pollination frequencies from the three groups of pollen donors ( Q. petraea, Q. pubescens , intermediate). We also assayed pollen viability and germinability of each species group. We observed natural hybridisation between Q. petraea and Q. pubescens, with a predominant component in the direction Q. petraea → Q. pubescens : Q. pubescens displayed a higher level of heterospecific pollination by Q. petraea (25.8%) and intermediate morphotypes (14.7%), compared to Q. petraea acting as pollen receptor (with less than 5% heterospecific pollinations). Intermediate 'mother trees' were pollinated in similar proportions by Q. petraea (23.1%), Q. pubescens (37.8%) and intermediate morphotypes (39.1%). The asymmetrical introgression observed for the studied generation may be caused, among other factors, by the relative abundance of trees from each species group in the studied area.     

The Description of Clones of Quercus robur L. and Q. petraea (Matt.) Liebl. with Microsatellites and AFLP in an Ancient Woodland

Abstract: In densely populated areas autochthonous Quercus robur L. (pedunculate oak) and Q. petraea (Matt.) Liebl. (sessile oak) (Fagaceae) populations have been maintained as ancient devastated woodlands. The continuous cutting, grazing and resprouting of such woodlands has enabled the development of clonal structures. For conservation purposes, an analysis of the actual number, size and spatial distribution of clones is necessary, especially when there is an interest in genetic variation of the population. This study describes for the first time - based on microsatellite and AFLP™ analysis - clones in an autochthonous mixed Q. robur and Q. petraea population that has been coppiced and grazed for several centuries. Based on six microsatellite loci and 69 polymorphic AFLP markers, only 14 unique genotypes were detected in a plot that consisted of 80 trees. Clones were observed for both Q. robur and Q. petraea. The largest clone diameters were observed for Q. robur, with distances up to 5.8 m. The observed clone sizes may indicate the age of the trees.     

Morphological and physiological reactions of young deciduous trees (Quercus robur L., Q. petraea [Matt.] Liebl., Fagus sylvatica L.) to waterlogging

One-year-old seedlings of Quercus robur L., Q. petraea (Matt.) Liebl. and Fagus sylvatica L. were cultivated in lysimeters and subjected to waterlogging for 17 weeks, interrupted by a five-week drainage period during summer. The growth of Q. robur was less affected by waterlogging than that of Q. petraea and Fagus. Waterlogging resulted in the formation of adventitious roots in Q. robur and Q. petraea, but not in Fagus. In contrast to Fagus, Q. robur and, to a lesser extent, Q. petraea were able to generate roots even below the water table. The hydraulic conductance of the excised root systems, the stomatal conductance and, in Fagus, the leaf water potential and the leaf-mass related hydraulic conductance were decreased by waterlogging. The decrease in the hydraulic conductance was largest in Fagus, and smallest in Q. robur. The roots of Fagus responded to anaerobic conditions with an increase in ethanol concentration. The measurements of nitrate reductase activities in roots and leaves provided no indications of a persistent contribution of NO₃ ⁻ metabolism to the alleviation of waterlogging-induced stress. It is concluded that Q. robur and, to a lesser extent, Q. petraea can tolerate waterlogging periods better than Fagus due to a different pattern of root formation, and to a better adjustment of leaf biomass production to the hydraulic conductivity of the root system. This revised version was published online in June 2006 with corrections to the Cover Date.     

Seeing the forest through the trees: comprehensive inference on individual mating patterns in a mixed stand of Quercus robur and Q. petraea

Background and Aims

Sexual reproduction is one of the most important moments in a life cycle, determining the genetic composition of individual offspring. Controlled pollination experiments often show high variation in the mating system at the individual level, suggesting a persistence of individual variation in natural populations. Individual variation in mating patterns may have significant adaptive implications for a population and for the entire species. Nevertheless, field data rarely address individual differences in mating patterns, focusing rather on averages. This study aimed to quantify individual variation in the different components of mating patterns.

Methods

Microsatellite data were used from 421 adult trees and 1911 seeds, structured in 72 half-sib families collected in a single mixed stand of Quercus robur and Q. petraea in northern Poland. Using a Bayesian approach, mating patterns were investigated, taking into account pollen dispersal, male fecundity, possible hybridization and heterogeneity in immigrant pollen pools.

Key Results

Pollen dispersal followed a heavy-tailed distribution (283 m on average). In spite of high pollen mobility, immigrant pollen pools showed strong genetic structuring among mothers. At the individual level, immigrant pollen pools showed highly variable divergence rates, revealing that sources of immigrant pollen can vary greatly among particular trees. Within the stand, the distribution of male fecundity appeared highly skewed, with a small number of dominant males, resulting in a ratio of census to effective density of pollen donors of 5·3. Male fecundity was not correlated with tree diameter but showed strong cline-like spatial variation. This pattern can be attributed to environmental variation. Quercus petraea revealed a greater preference (74 %) towards intraspecific mating than Q. robur (36 %), although mating preferences varied among trees.

Conclusions

Mating patterns can reveal great variation among individuals, even within a single even-age stand. The results show that trees can mate assortatively, with little respect for spatial proximity. Such selective mating may be a result of variable combining compatibility among trees due to genetic and/or environmental factors.     

Complex patterns of mating revealed in a Eucalyptus regnans seed orchard using allozyme markers and the neighbourhood model

The neighbourhood model apportions offspring of individual mother plants to self-fertilization, outcrossing to males within a circumscribed area around the mother plant (the neighbourhood), and outcrossing to males outside the neighbourhood. Formerly the model was applied only to haploid pollen gametes in the offspring of conifers, but is extended so that it can be used with genotypic data from diploid offspring of both angiosperms and gymnosperms. In addition, it is shown that the mating parameters can be estimated without independent estimates of allele frequencies in the pollen pools outside the neighbourhood; thus the model might be applied effectively to natural populations exposed to unknown external pollen sources. Parameters of the neighbourhood mating model were estimated for a 10-year-old seed orchard population of the insect-pollinated tree, Eucalyptus regnans, in southeast Australia, which contained a mixture of two geographical provenances (Victoria and Tasmania). The mating patterns revealed were complex. Crosses between trees of the same provenance occurred three times more often than crosses between trees of different provenances. Levels of self-fertilization and patterns of mating within neighbourhoods were influenced by provenance origin, crop fecundity and orchard position (central vs. edge) of mother trees. Gene dispersal, however, was extensive, with approximately 50% of effective pollen gametes coming from males more than 40 m away from mother trees (average distance between neighbouring trees was 7.4 m). Thus, insect pollinators are efficient promoters of cross-fertilization in this orchard, with the result that the effective number of males mating with each female is large.     

Fat-tailed gene flow in the dioecious canopy tree species Fraxinus mandshurica var. japonica revealed by microsatellites

Pollen flow, seed dispersal and individual reproductive success can be simultaneously estimated from the genotypes of adults and offspring using stochastic models. Using four polymorphic microsatellite loci, gene flow of the wind-pollinated and wind-seed-dispersed dioecious tree species, Fraxinus mandshurica var. japonica, was quantified in a riparian forest, in northern Japan. In a 10.5-ha plot, 74 female adults, 76 male adults and 292 current-year seedlings were mapped and genotyped, together with 200 seeds. To estimate dispersal kernels of pollen and seeds, we applied normal, exponential power, Weibull, bivariate t-distribution kernels, and two-component models consisting of two normal distribution functions, one with a small and one with a large variance. A two-component pollen flow model with a small contribution (26.1%) from short-distance dispersal (sigma = 7.2 m), and the rest from long-distance flow (sigma = 209.9 m), was chosen for the best-fitting model. The average distance that integrated pollen flows inside and outside the study plot was estimated to be 196.8 m. Tree size and flowering intensity affected reproduction, and there appeared to be critical values that distinguished reproductively successful and unsuccessful adults. In contrast, the gene flow model that estimated both pollen and seed dispersal from established seedlings resulted in extensive seed dispersal, and the expected spatial genetic structures did not satisfactorily fit with the observations, even for the selected model. Our results advanced small-scale individual-based parentage analysis for quantifying fat-tailed gene flow in wind-mediated species, but also clarified its limitations and suggested future possibilities for gene flow studies.     

Long distance pollen-mediated gene flow at a landscape level: the weed beet as a case study

Gene flow is a crucial parameter that can affect the organization of genetic diversity in plant species. It has important implications in terms of conservation of genetic resources and of gene exchanges between crop to wild relatives and within crop species complex. In the Beta vulgaris complex, hybridization between crop and wild beets in seed production areas is well documented and the role of the ensuing hybrids, weed beets, as bridges towards wild forms in sugar beet production areas have been shown. Indeed, in contrast to cultivated beets that are bi-annual, weed beets can bolt, flower and reproduce in the same crop season. Nonetheless, the extent of pollen gene dispersal through weedy lineages remains unknown. In this study, the focus is directed towards weed-to-weed gene flow, and we report the results of a pollen-dispersal analysis within an agricultural landscape composed of five sugar beet fields with different levels of infestation by weed beets. Our results, based on paternity analysis of 3240 progenies from 135 maternal plants using 10 microsatellite loci, clearly demonstrate that even if weedy plants are mostly pollinated by individuals from the same field, some mating events occur between weed beets situated several kilometres apart (up to 9.6 km), with rates of interfield-detected paternities ranging from 11.3% to 17.5%. Moreover, we show that pollen flow appears to be more restricted when individuals are aggregated as most mating events occurred only for short-distance classes. The best-fit dispersal curves were fat-tailed geometric functions for populations exhibiting low densities of weed beets and thin-tailed Weibull function for fields with weed beet high densities. Thus, weed beet populations characterized by low density with geographically isolated individuals may be difficult to detect but are likely to act as pollen traps for pollen emitted by close and remote fields. Hence, it appears evident that interfield pollen-mediated gene flow between weed beets is almost unavoidable and could contribute to the diffusion of (trans)genes in the agricultural landscape.     

Strength, diversity and plasticity of postmating reproductive barriers between two hybridizing oak species (Quercus robur L. and Quercus petraea (Matt) Liebl.)

Very little is known about the nature and strength of reproductive isolation (RI) in Quercus species, despite extensive research on the estimation and evolutionary significance of hybridization rates. We characterized postmating pre- and postzygotic RI between two hybridizing oak species, Quercus robur and Quercus petraea, using a large set of controlled crosses between different genotypes. Various traits potentially associated with reproductive barriers were quantified at several life history stages, from pollen-pistil interactions to seed set and progeny fitness-related traits. Results indicate strong intrinsic postmating prezygotic barriers, with significant barriers also at the postzygotic level, but relatively weaker extrinsic barriers on early hybrid fitness measures assessed in controlled conditions. Using general linear modelling of common garden data with clonal replicates, we showed that most traits exhibited important genotypic differences, as well as different levels of sensitivity to micro-environmental heterogeneity. These new findings suggest a large potential genetic diversity and plasticity of reproductive barriers and are confronted with hybridization evidence in these oak species.     

Pollen and seed flow patterns of Carapa guianensis Aublet. (Meliaceae) in two types of Amazonian forest

Various factors affect spatial genetic structure in plant populations, including adult density and primary and secondary seed dispersal mechanisms. We evaluated pollen and seed dispersal distances and spatial genetic structure of Carapa guianensis Aublet. (Meliaceae) in occasionally inundated and terra firme forest environments that differed in tree densities and secondary seed dispersal agents. We used parentage analysis to obtain contemporary gene flow estimates and assessed the spatial genetic structure of adults and juveniles. Despite the higher density of adults (diameter at breast height ≥ 25 cm) and spatial aggregation in occasionally inundated forest, the average pollen dispersal distance was similar in both types of forest (195 ± 106 m in terra firme and 175 ± 87 m in occasionally inundated plots). Higher seed flow rates (36.7% of juveniles were from outside the plot) and distances (155 ± 84 m) were found in terra firme compared to the occasionally inundated plot (25.4% and 114 ± 69 m). There was a weak spatial genetic structure in juveniles and in terra firme adults. These results indicate that inundation may not have had a significant role in seed dispersal in the occasionally inundated plot, probably because of the higher levels of seedling mortality.     

Comparison of earlywood vessel variables in the wood of Quercus robur L. and Quercus petraea (Mattuschka) Liebl. growing at the same site

The differences in the microscopic structure of wood based on the variables of earlywood vessel area and tree ring width were analysed in 6 trees of Pedunculate oak (Quercus robur L.) and 6 trees of Sessile oak (Quercus petraea (Mattuschka) Liebl.) in the same forest stand at a site in the Vizovice Highland (Czech Republic). The aim of this paper was to assess any differences between the two oak species when grown in the same location. Also the effect of tree-ring width and the effect of age on earlywood vessel variables were analysed. Mean values of tree-ring widths were not different between species. Earlywood vessel area chronologies were synchronized well between species. The mean values of earlywood vessel area (average vessel area, average vessel area of the first row of vessels, area of the largest vessel) showed a significant difference. Finally, we can conclude that there were differences in vessel features between these oak species.     

Measuring gene flow in the cultivation of transgenic cotton (Gossypium hirsutum L.)

Transgenic Bt cotton NewCott 33B and transgenic tfd A cotton TFD were chosen to evaluate pollen dispersal frequency and distance of transgenic cotton (Gossypium hirsutum L.) in the Huanghe Valley Cotton-producing Zone, China. The objective was to evaluate the efficacy of biosafety procedures used to reduce pollen movement. A field test plot of transgenic cotton (6×6 m) was planted in the middle of a nontransgenic field measuring 210×210 m. The results indicated that the pollen of Bt cotton or tfd A cotton could be dispersed into the environment. Out-crossing was highest within the central test plot where progeny from nontransgenic plants, immediately adjacent to transgenic plants, had resistant plant progeny at frequencies up to 10.48%. Dispersal frequency decreased significantly and exponentially as dispersal distance increased. The flow frequency and distance of tfd A and Bt genes were similar, but the pollen-mediated gene flow of tfd A cotton was higher and further to the transgenic block than that of Bt cotton (χ² = 11.712, 1 degree of freedom, p<0.001). For the tfd A gene, out-crossing ranged from 10.13% at 1 m to 0.04% at 50 m from the transgenic plants. For the Bt gene, out-crossing ranged from 8.16% at 1 m to 0.08% at 20 m from the transgenic plants. These data were fit to a power curve model: y=10.1321x ^−1.4133 with a correlation coefficient of 0.999, and y=8.0031x ^−1.483 with a correlation coefficient of 0.998, respectively. In this experiment, the farthest distance of pollen dispersal from transgenic cotton was 50 m. These results indicate that a 60-m buffer zone would serve to limit dispersal of transgenic pollen from small-scale field tests.     

Characterisation, prediction and relationships between different wavebands of solar radiation transmitted in the understorey of even-aged oak (Quercus petraea, Q. robur) stands

Solar radiation transmission in forest stands affects many processes, including biomass and diversity of understorey vegetation and tree seedling regeneration (growth and morphogenesis). However, understorey light availability is not easy for forest managers or scientists to measure. Therefore, different models have been developed to predict light transmission in forest stands according to tree or stand structure. However, these models are generally too complex to be used operationally. This paper reports the assessment of light transmission according to stand parameters commonly measured by foresters in inventories. We measured transmittance in 29 even-aged oak stands in France for various wavebands, total solar radiation (TSR, 300–3000 nm), PAR (400–700 nm), red (R, 660 nm) and far-red (FR, 730 nm), and demonstrated that transmittance in a given waveband can be predicted from the measurement of another waveband. The R:FR ratio can be predicted according to TSR or PAR transmittance, but the opposite is also true; PAR or TSR transmittance can be predicted from the R:FR ratio. Transmittance variability was characterised, and the variation coefficient ranged from 5 to 45% with a trend to increase with tree density. By analogy to Beer–Lambert's law, we established that mean daily transmittance for the different wavebands can be assessed according to stand basal area and stand age with good accuracy (R ²>0.74). Results are discussed in comparison with other models based on the principle of parsimony.     

Immigration history and gene dispersal: allozyme variation in Nordic populations of the red campion, Silene dioica (Caryophyllaceae)

Most of the Nordic region was ice-covered during the last (Weichselian) glaciation. During the postglacial period, plant and animal species recolonized the region from several directions and the geographic structuring of genetic variation within Nordic species may still contain a historic component that reflects patterns of postglacial immigration. The present investigation of 69 populations of Silene dioica represents the first large-scale allozyme study of a widespread herbaceous plant in the Nordic region. Although the frequencies of individual alleles showed a range of different geographic patterns, mapping of the axis scores from an ordination of variation at eight polymorphic loci revealed a division into two main geographic groups of populations. The broadly south-western and north-eastern distributions of these two groups of populations suggest that immigration into the region may have involved both eastern and southern geographic sources. However, the geographic boundaries between the two groups of populations are diffuse, and the relatively low between-population component of genetic diversity (G_ST = 16.4%) suggests a history of extensive gene dispersal by pollen.  © 2002 The Linnean Society of London, Biological Journal of the Linnean Society , 2002, 77 , 23–34.     

The effects of seed size, cotyledon reserves, and herbivory on seedling survival and growth in Quercus rugosa and Q. laurina (Fagaceae)

In a greenhouse experiment, seedling survival of two oak species (Quercus rugosa and Q. laurina) was greatly affected by the excision of cotyledons 1 mo after germination, with a greater impact on Q. laurina. The effect of seed size was also significant for both species, with a positive correlation between seed mass and survival and growth. The effect of cotyledon excision on seedling growth persisted throughout the first growing season in Q. rugosa and was not analyzed for Q. laurina due to the low number of seedlings that survived cotyledon excision. Seed size significantly affected seedling height, diameter, leaf area, and biomass at 6 mo. Seed size and cotyledon retention affected the ability of Q. rugosa to recover from herbivory, as both factors had a significant effect on relative growth rates after aerial biomass removal. The results show that seedlings originating from large seeds can better endure loss of cotyledons and aerial biomass and thus are better equipped to confront stress early in their lives.