Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

High Correlated Paternity Leads to Negative Effects on Progeny Performance in Two Mediterranean Shrub Species

High Correlated Paternity Leads to Negative Effects on Progeny Performance in Two Mediterranean Shrub Species

  • Sofia Nora, 
  • Abelardo Aparicio, 
  • Rafael G. Albaladejo


Anthropogenic habitat deterioration can promote changes in plant mating systems that subsequently may affect progeny performance, thereby conditioning plant recruitment for the next generation. However, very few studies yet tested mating system parameters other than outcrossing rates; and the direct effects of the genetic diversity of the pollen received by maternal plants (i.e. correlated paternity) has often been overlooked. In this study, we investigated the relation between correlated paternity and progeny performance in two common Mediterranean shrubs, Myrtus communis and Pistacia lentiscus. To do so, we collected open-pollinated progeny from selected maternal plants, calculated mating system parameters using microsatellite genotyping and conducted sowing experiments under greenhouse and field conditions. Our results showed that some progeny fitness components were negatively affected by the high correlated paternity of maternal plants. In Myrtus communis, high correlated paternity had a negative effect on the proportion and timing of seedling emergence in the natural field conditions and in the greenhouse sowing experiment, respectively. In Pistacia lentiscus, seedling emergence time under field conditions was also negatively influenced by high correlated paternity and a progeny survival analysis in the field experiment showed greater mortality of seedlings from maternal plants with high correlated paternity. Overall, we found effects of correlated paternity on the progeny performance of Myrtus communis, a self-compatible species. Further, we also detected effects of correlated paternity on the progeny emergence time and survival in Pistacia lentiscus, an obligate outcrossed species. This study represents one of the few existing empirical examples which highlight the influence that correlated paternity may exert on progeny performance in multiple stages during early seedling growth.


Mating patterns are fundamental for determining the amount and distribution of genetic variation within and between populations of plant species [1, 2]. Anthropogenic habitat deterioration may promote changes in mating patterns [3, 4], which can directly influence the genetic variation in progeny and, even more importantly, may be reflected in fitness costs that will compromise plant recruitment in the forthcoming generation [4, 5]. Mating system can be described by parameters such as selfing, outcrossing rates and correlated paternity. Numerous studies have found high variability in mating parameters such as outcrossing rates and correlated paternity levels among species [4], among populations within species [6, 7] and even among maternal plants (e.g. [8, 9, 10, 11, 12]). This variation is highly context-dependent and reflects the influence of diverse ecological factors such as landscape heterogeneity, vegetation structure and the local neighbourhood of conspecific plants [9, 10, 13]. Nevertheless, mating system parameters are mainly controlled by plant reproductive traits that permit or prevent self-fertilization (i.e. the breeding system; see [14]).

However, although numerous studies aimed at evaluating variation in mating systems within species [5, 15], only a few have ever focused on their influence on progeny performance [16, 17]. Furthermore, most current studies consider no other mating system parameter than selfing rates and overlook the direct effects of the genetic diversity of the pollen received by the maternal plants (but see [18, 19, 20]). Correlated paternity (the proportion of full-sibs within maternal progeny arrays) is an important parameter that can provide deep insights into the pollination biology of plant species [21, 22]. Pollen diversity received by the maternal plants may be advantageous for progeny fitness because it can promote pollen competition, thereby allowing post-pollination selection for those males with the fastest-growing pollen tubes [23, 24], and/or may increase female choice by post-pollination mechanisms occurring both before ovule fertilization and during seed development (as cryptic female choice) and enable them to select between progeny that differ in quality and/or compatibility [23, 24].

In this study, we analysed how correlated paternity affects progeny performance in two common shrub species, Pistacia lentiscus L. (Anacardiaceae) and Myrtus communis L. (Myrtaceae) (Pistacia and Myrtus, hereafter). Even though their life-history traits are similar in a number of ways (e.g. both are long-lived and their seeds are dispersed mainly by birds), they differ in their breeding and pollen dispersal systems. Myrtus has a mixed-mating system with hermaphroditic flowers pollinated by insects, while Pistacia is a wind-pollinated dioecious species (and thus an obligate outcrosser). The mating systems of both species have been extensively studied. For instance, Albaladejo et al. [25, 26] found a wide range for correlated paternity values in Pistacia linked to significant spatiotemporal variation (from 0.03 to 0.23), which, to some extent, could be attributed to inherent individual factors (such as phenological synchronisation) or the local neighbourhood of conspecific plants. Additionally, Pistacia is capable of extensive pollen movement and high rates of pollen flow [25]. Regarding Myrtus pollen dispersal, less information is known. Myrtus populations often show very high levels of correlated paternity (in the range 0.40–0.61; [11, S. Nora, unpublished results]). Moreover, variation in outcrossing rates in Myrtus populations (from 0.13 to 0.62) has been shown to significantly affect progeny performance, with higher outcrossing rates being associated with higher seedling emergence and survival under greenhouse conditions [27]. However, to date no attempts have been made to test the impact of correlated paternity variation on progeny performance in these two species.

Here, we combined mating system analysis with data on progeny performance measurements based on both greenhouse and natural field sowing experiments. Specifically, we aimed (1) to assess the influence of correlated paternity on early progeny performance (here evaluated as seedling emergence, seedling emergence time, seedling growth, biomass, lifetime and survival) in Myrtus communis and Pistacia lentiscus, and (2) to verify the consistency of the observed relationships both in a greenhouse environment and under natural (field) conditions, where they could be potentially masked by other factors such as the response to environmental heterogeneity.

Material and Methods

Ethics statement

Permission for conduct this study was obtained from the Delegación Provincial de la Consejería de Medio Ambiente de la Junta de Andalucía, Spain. This study did not involve endangered or protected species.

Study species and system

Myrtus communis and Pistacia lentiscus are widespread and abundant shrubs in sclerophyllous Mediterranean woodlands. In the study area, Myrtus grows up to 4 m in height and blooms massively in early summer (from mid-June to early July) and its white flowers are hermaphroditic and pollinated mostly by hymenopterans and dipterans [28]. Myrtus fruit are berries (mean ± SD = 5.2 ± 2.7 seeds; [27]) that turn dark blue when mature. Fruits mature from mid-October to late November and seeds are dispersed by passerine birds, mostly Sylviidae and Turdidae [29]. Pistacia is a dioecious shrub up to 4 m in height. It is wind-pollinated and blooms massively between mid-March and late April [30]. Its fruits are small black one-seeded drupes and mature in September–December and are consumed by a similar guild of dispersers as those of Myrtus [31].

This study was conducted in the Guadalquivir river valley (southwestern Spain), a large (21,000 km2), fertile and intensively cultivated Mediterranean lowland. This region has a long history of human intervention and management [32] and remnant woodland patches cover less than 1% of their potential area [33]. The study area, Dehesa de las Yeguas (36°33'15''N, 6°08'08''W), is ca. 100 hectares of semi-natural stone pine (Pinus pinea L.) forest (Fig 1). In the study area, shrub cover represents 63.82% of the total study area and is mainly composed of small (<1 m) dry-fruited shrubs (Cistaceae, Lamiaceae and Leguminosae) and tall (>1 m) fleshly fruited shrubs, amongst which Pistacia and Myrtus are the dominant species. The climate is typically Mediterranean, with a mean annual precipitation of ca. 650 mm and an intense summer drought (approximately 10 mm rainfall in July and August), and a mean monthly temperature of 19°C (ranging from 9°C in February to 25°C in August) (data from subsistema CLIMA available at

Fig 1. Map of the Guadalquivir River Valley (southwestern Spain), the location of the stone pine forest selected as a study area (Dehesa de las Yeguas).

The unshaded area represents lowland areas (< 200 m above sea level), the remaining woodland patches are shaded in black and higher elevations and legally protected areas are in light and dark grey, respectively. The arrow indicates the location of the study area.

In November 2010, we collected naturally produced mature fruits from throughout the canopy of 18 Myrtus and 18 Pistacia maternal plants. Study plants were selected to cover the range of natural heterogeneity in the local environment (mainly plant density, from clumped to isolated plants). Therefore, local variation was chosen a priori to try to capture the highest variability in mating system parameters at individual plant level [9, 10, 13, 24].

Microsatellite genotyping and individual mating system analyses

For genetic analysis we randomly selected 16 seeds per maternal plant. Pistacia embryos were easily excised from the maternal endocarp and endosperm, so they did not require germination. In Myrtus only one seed per fruit was used to avoid the effects of correlated pollination in a single pollination event. Seeds from each maternal plant were germinated in separate Petri dishes containing filter paper saturated with distilled water. Previous studies indicate that germination rates in Myrtus are very high [11] and in this study almost all seeds germinated within a few days (mean germination rates ± SD = 95.0% ± 6.9%). For DNA isolation, Myrtus seedlings were harvest once both cotyledons were fully developed.

We extracted DNA from Myrtus seedlings and Pistacia embryos with the Invisorb DNA Plant HTS 96 Kit (Invitek, Germany) following the manufacturer’s protocol. We amplified seven and eight polymorphic nuclear microsatellites in Pistacia and Myrtus, respectively. Further details regarding the amplification reactions and PCR cycle conditions can be found in Albaladejo et al. [34, 35] and Nora et al. [36]. We analysed fluorescently labelled (with 6-FAM, NED, VIC and PET dyes) PCR products with an ABI 3730 DNA Analyzer (Applied Biosystems, Foster City, US) at the Unidad de Genómica (UCM, Madrid, Spain). Fragment sizes were automatically scored with GeneMapper 3.7 (Applied Biosystems, Foster City, US) and corrected manually when necessary. We only retained multilocus genotypes that successfully scored with at least four loci. Maternal genotypes had already been analysed for a companion study [36]. We checked for genotyping errors by identifying mother-offspring mismatches and obtained an overall mismatching estimate of 1.90% for Myrtus and 2.14% for Pistacia.

Independence among pairs of loci of maternal genotypes was checked through linkage disequilibrium (LD) tests using the Genepop web server ( and tested for null alleles with Micro-Checker [37]. We calculated maternal genetic diversity levels via the homozygosity by loci (HL) index, a calculation that improves raw heterozygosity estimates by weighting the contribution of each locus to the individual homozygosity value in terms of their allelic variability [38]. We calculated HL with the R script Genhet [39]. We estimated the multilocus (tm) and single locus (ts) outcrossing rates and the correlated paternity (rp) for maternal plants with the software Mltr 3.4 [40]. Pollen and ovules were constrained to have the same gene frequencies and standard errors, and confidence intervals were assessed by bootstrap (1000 replicates) with individuals within families as the resampling unit. For Pistacia (as this species is dioecious), we used only the parameters tm and ts to estimate the biparental inbreeding (tm-ts, i.e. mating between genetically related individuals).

Progeny performance under greenhouse conditions

In April 2011, we individually sowed a total of 720 seeds (20 seeds from each maternal plant of both species) in trays of 60 pots (5 x 5 cm and 17 cm in depth). We filled each pot with horticultural blend (5:1 peat and perlite) and sowed the seeds at a depth of c. 0.5 cm. We watered the trays twice a week and randomly changed their position every 2–3 weeks to ensure homogeneous conditions inside the greenhouse throughout the 424 days of the experiment. We regularly monitored ‘seedling emergence’ (every 2–3 days for two months after sowing) and ‘seedling mortality’ until the end of the experiment. A seedling was regarded to have emerged when it extended cotyledons. For each emerged seedling we also recorded the following data: (1) ‘seedling emergence time’ as the number of days between sowing and seedling emergence (monitored every 2–3 days for two months after sowing), (2) ‘seedling growth’, measured as seedling height at four dates distributed regularly throughout the experiment (days 36, 113, 190 and 267) and including a final measurement at the end (day 424), and (3) ‘seedling biomass’ measured as the dry weight at the end of the experiment. To measure the biomass, after harvesting we dried the surviving plants at 60° C for 72 h and, in order to obtain accurate measurements of the biomass allocation, we separately measured the shoot and root dry weight of each plant.

Progeny performance under field conditions

We performed a sowing experiment using 30 seeds from each maternal plant (totalling 1080 seeds) to assess seedling performance under natural conditions in Dehesa de las Yeguas (Spain). To avoid bias between seedling emergency of sowed seeds of our experiment and seedlings from natural dispersed seeds, we started this experiment in early March 2011, the end of the natural dispersion period (cf. González-Varo et al. [41]). We sowed seeds in three distinct microhabitats: (1) under conspecific shrubs (Myrtus seeds under Myrtus shrubs and Pistacia seeds under Pistacia shrubs), (2) under heterospecific shrubs (Myrtus seeds under Pistacia shrubs and Pistacia seeds under Myrtus shrubs), and (3) in open ground under tree cover. We focused our experiment on these microhabitats because they are traditionally considered suitable places for birds to drop seeds and are consequently favourable places for seedling recruitment (e.g. [41, 42, 43, 44]). We established 10 sowing plots per microhabitat (30 plots in total). In each plot, we removed any naturally dispersed seedlings and sowed uniformly one seed from each of the selected maternal plants (i.e. 18 seeds of each species) at a depth of 0.5–1 cm. To match natural conditions as close as possible, we then added a thin layer of litter on top of the seed after sowing. Each plot was protected by a wire mesh cage (with an upper surface grid area of 15 × 10 cm and 10 cm in height) to prevent predation by rodents.

We conducted this experiment and monitored each emerged seedling until September 2011 (i.e. after the first summer). We monitored ‘seedling emergence’ and ‘seedling mortality’ weekly for three months and fortnightly thereafter until the end of the experiment. Like in the greenhouse, in this field experiment, seedlings were regarded to have emerged when they fully extended cotyledons. For each seedling we also recorded the following variables: (1) ‘seedling emergence time’, as the number of days between sowing and seedling emergence and (2) ‘seedling lifetime’, as the total number of days a seedling lived from emergence until death or until the end of the experiment (for those seedlings that survived). To control soil moisture, we used a time domain reflectometer (TDR, Campbell Scientific Inc., Logan, UT, USA) with 12-cm depth rods to measure soil volumetric water content (VWC; %). This measure is fundamental because the water deficit, especially during the summer, is a critical factor for seedling survival in many Mediterranean plant species (e.g. [42, 44]). Measurements were taken at an adjacent point (~20 cm) to each sowing plot every two weeks in the first three months and monthly thereafter.

Statistical analyses

For each species, we compared seedling performance between experiments (greenhouse vs. field) using the non-parametric Wilcoxon paired test. Correlations between seedling height and seedling total dry biomass, as well as between shoot and root dry biomass were made using Spearman’s correlations. We reported data as means ± SE.

We evaluated the relationships between the correlated paternity (rp) and progeny performance measurements using generalized linear models (GLMs) in R 3.0.2 (R Developmental Core Team 2016). We conducted all analyses at maternal plant level and used a gamma distribution with a log link function for modelling the response variables. To check whether rp was correlated with other mating system variables we performed Spearman’s correlation tests and observed no significant correlations between correlated paternity and any other genetic parameter (HL, tm-ts or tm) neither in Myrtus nor in Pistacia (S1 Table). We accounted for the effects of seed mass on the response variables ‘proportion of seedling emergence’, ‘seedling growth’ and ‘seedling lifetime’, and the effects of ‘seedling emergence time’ on the variables ‘seedling growth’, ‘seedling biomass’ and ‘seedling lifetime’ by including these variables as covariates in the models. To obtain the mean seed mass for each maternal plant we randomly sampled 20 seeds per maternal plant (one seed per Myrtus fruit) and weighed it to the nearest 0.1 mg.

We evaluated seedling survival rates of Pistacia in the field sowing experiment with survival analyses [45] and we tested the effects of correlated paternity using Cox’s proportional hazard regression [46]. This analysis was performed with the software Statistica v.6 (StatSoft 2001).We considered seedlings that at the end of the study were still alive as censored data. No Myrtus seedlings survived the first summer in any of the microhabitats (see Results) and so this analysis could not be performed for this species.


Genetic diversity and mating parameters

We genotyped open-pollinated progeny from 18 maternal plants of Myrtus (n = 266) and Pistacia (n = 283). Microsatellite markers were highly polymorphic in both species with a total of 57 and 40 different alleles identified across all maternal plants for Myrtus and Pistacia, respectively. Linkage disequilibrium tests showed that all pairs of loci in the two species were independent and null allele presence was not significant at any loci within any species.

Overall, Myrtus maternal plants had high levels of homozygosity (mean HL = 0.386 ± 0.035) and correlated paternity (mean rp = 0.395 ± 0.045) (S2 Table). On the other hand, Pistacia maternal plants had relatively low levels of both homozygosity (mean HL = 0.271 ± 0.034) and correlated paternity (mean rp = 0.060 ± 0.006). Both parameters presented a large variation among Myrtus maternal plants: HL ranged from 0.18 to 0.69 and rp from 0.10 to 0.87. In Pistacia maternal plants, we also found high variability in HL levels, ranging from 0.00 to 0.49 but rp values were more stable among maternal plants, ranging from 0.03 to 0.11. As expected in a dioecious species, estimated outcrossing rate (tm) in Pistacia was very close to one (S2 Table). The mean value of tm in Myrtus maternal plants was 0.704 ± 0.044, ranging from 0.25 to 0.98 (S2 Table). Myrtus displayed relatively higher rates of biparental inbreeding than Pistacia (tm-ts = 0.264 ± 0.027 vs. 0.065 ± 0.005).

Progeny performance under greenhouse conditions

The mean percentage of seedling emergence was 53.33% (± 2.26) in Myrtus and 49.17% (± 1.73) in Pistacia. Myrtus seedlings took on average 20.04 (± 0.40) days to emerge (emergence time was 14–36 days after sowing), while Pistacia seedlings took 18.90 (± 0.40) days for emergence (range of 14–44 days) (S3 Table). Seedling mortality was low throughout the experiment in the greenhouse (15.03% in Myrtus and 16.38% in Pistacia). Approximately one month after the start of the experiment, Pistacia seedlings had a mean height of 26.9 (± 0.8) mm and Myrtus seedlings a mean of 16.7 (± 0.8) mm (S3 Table). At the end of the experiment, the mean heights were 391.6 (± 9.8) mm for Pistacia seedlings and 257.5 (± 9.3) mm for Myrtus seedlings. Pistacia seedlings had a final mean dry biomass of 2.87 (± 0.04) g and Myrtus seedlings 2.39 (± 0.04) g (S3 Table). In both species, seedling height at the end of the experiment was highly correlated with seedling total dry biomass (Spearman’s rMyrtus = 0.87, P < 0.001, and rPistacia = 0.77, P < 0.001). Shoot and root dry biomass were also highly correlated in both species (Spearman’s rMyrtus = 0.92, P < 0.001 and rPistacia = 0.71, P < 0.001).

The influence of correlated paternity on seedling performance in the greenhouse environment is summarized in Table 1. Correlated paternity had no effect on Myrtus ‘proportion of seedling emergence’, ‘seedling growth’ or ‘seedling biomass’ but had a significant positive effect on ‘seedling emergence time’ (Fig 2). Correlated paternity had no significant effect on Pistacia seedling performance measurements (‘proportion of seedling emergence’, ‘seedling emergence time’, ‘seedling growth’ and ‘seedling biomass’) (Table 1).

Fig 2. Relationships between correlated paternity and seedling performance variables for Myrtus communis derived from GLMs in the greenhouse (left) and field (left) experiment.

Table 1. Results from generalized linear models (GLMs) examining the influence of correlated paternity on fitness measurements in seedlings of Myrtus communis and Pistacia lentiscus grown under greenhouse conditions.

The covariate ‘seed mass’ had a significant positive effect on ‘proportion of seedling emergence’ in Myrtus and in the first measurement of seedling height (‘seedling height after 36 days’) in both species (Table 1). ‘Seedling emergence time’ only had a significant effect on seedling height in Pistacia.

Progeny performance under field conditions

Pistacia had similar seedling emergence rates in field and greenhouse environments (Wilcoxon’s test; Z = 0.15, P = 0.88). However, field conditions had a strongly significant negative effect on Myrtus seedling emergence (Wilcoxon’s test; Z = 3.59, P < 0.001). Both Myrtus and Pistacia seedlings took longer to emerge in the field than in the greenhouse (Wilcoxon’s test; Z = 3.72, P < 0.001 and Z = 3.72, P < 0.001, respectively) (S3 Table). Myrtus seedlings in the field took on average 39.19 (± 0.95) days to emerge, whereas Pistacia seedlings emerged slightly earlier, 34.24 (± 0.60) days.

We did not conduct GLMs on seedling survival due to low seedling survival in Pistacia (only 40 seedlings survived to the end of the experiment) and due to the lack of surviving Myrtus seedlings. Myrtus seedling mortality rate rose dramatically from 12.38% at day 29 to approximately 100% at day 78 and was clearly associated with the measured volumetric water content in the ground (see Fig 3). For this reason, the potential of seedling survival for this species was assessed indirectly by using seedlings’ lifetimes as a response variable (see Table 2).

Fig 3. Cumulative percentage of Myrtus communis seedling mortality in field environment (dots) and temporal dynamics in mean soil volumetric water content (%) in the plot where sowing experiments were carried out (thin line).

The influence of correlated paternity on seedling performance in the field experiment is summarized in Table 2. Contrary to the effects detected in the greenhouse experiment, we found no significant trend towards the influence of correlated paternity on ‘seedling emergence time’ and ‘seedling lifetime’ in Myrtus under field conditions. For this species we found a significant negative relationship of correlated paternity on ‘proportion of seedling emergence’ (Fig 2). Correlated paternity had no association with Pistacia seedling performance measurements ‘seedling emergence’ and ‘seedling lifetime’. However, we found a significant relationship between correlated paternity and ‘seedling emergence time’ in Pistacia under field conditions. The covariate ‘emergence time’ had a significant effect on ‘seedling lifetime’ in Myrtus.

Table 2. Results of the generalized linear models (GLMs) examining the influence of correlated paternity on fitness measurements in seedlings of Myrtus communis and Pistacia lentiscus grown under natural field conditions.

Survival analysis regression test detected the influence of correlated paternity in Pistacia seedling survival. When running the analyses we found significant negative effect of correlated paternity (Chi2 = 4.09, df = 1, P = 0.043; Fig 4), indicating that progeny from maternal plants with high correlated paternity had lower survival rates than progeny from maternal plants with low correlated paternity.

Fig 4. Pistacia lentiscus seedlings survival as a function of correlated paternity in the field environment.


Plant recruitment is an essential stage of plant population dynamics and the long-term persistence of plant populations. In order to understand plant recruitment, it is fundamental to determinate which factors are affecting the demographic processes involved in early seedling performance (e.g. seedling emergence, growth and first-year survival). Our study is one of the few to date that has investigated whether the pollen diversity naturally received by maternal plants–here assessed by the correlated paternity–has an influence on the subsequent progeny performance. To accomplish our goal, we integrated mating system analysis with data on progeny performance both in greenhouse and field-sowing experiments.

The effects of correlated paternity on progeny performance were more evident in Myrtus. Under greenhouse conditions, we observed that high correlated paternity was associated to a delay in seedling emergence. In Mediterranean environments early emergence is of critical importance in determining subsequent seedling growth [47], as was also observed in our study, in which emergence time influenced seedling fitness. Furthermore, the negative influence of high correlated paternity on the percentage of seedling emergence was also evident in the field environment, i.e. under more stressful conditions. These results fit the hypothesis that maternal plants that receive high pollen diversity (i.e. have low correlated paternity) generate progeny with increased fitness. As hypothesized in the literature, this relation could be due to female selection for more compatible pollen [24, 48], a process that could occur both before fertilization (e.g. by pollen germination and pollen-tube growth) or after [23, 49]. In addition, our results concur with the few other empirical studies in which the effects of correlated paternity on fitness progeny have been tested. For instance, Llorens et al. [20] reported both direct and indirect (through the increase of seed size) relationships of paternal diversity on seed germination, seedling weight and survival in Banksia sphaerocarpa. Also, Breed et al. [18, 19] reported negative effects of high correlated paternity on sapling growth in Swietenia macrophylla and Eucalyptus socialis.

Contrary to the results obtained under greenhouse conditions, seedling emergence time in the field was not significantly affected by correlated paternity, even though progeny from maternal plants with high correlated paternity had greater emergence times that progeny from maternal plants with low correlated paternity. We probably only observed this relationship in the greenhouse because under field conditions the influence of correlated paternity was masked by other factors that may condition this fitness variable more critically. Several studies documented the particularly importance of abiotic factors (such as water, light, temperature, microhabitats and disturbances) on seedling establishment under the harsh summer conditions of the Mediterranean climate (e.g. [41, 44]).Still, even under this type of natural conditions, we were able to detect the influence of correlated paternity in the percentage of seedling emergence in Myrtus.

It has been suggested that higher pollen diversity moderates the effects of inbreeding depression on progeny performance by lessening the likelihood that pollen with expressed deleterious alleles is involved in reproduction [50]. In our study, although it is difficult to completely disentangle the effects of correlated paternity from the effects of inbreeding, in some cases it was possible to detect a trend indicating that correlated paternity could have an effect on progeny fitness independently of inbreeding (whenever an effect of rp independently of tm or HL results is observed), which agrees with findings in other studies (e.g. [19]). Nevertheless, this subject still needs further research.

In Pistacia, no significant impact of correlated paternity on progeny performance was evident in the greenhouse conditions. Pistacia is an obligate outcrossed species so it is not surprising that it has low levels of correlated paternity. Thus, the inherent pollination biology of Pistacia, (leading to inherent low levels of correlated paternity) is diluting the potential trends that we found in Myrtus.

However, under field conditions we were able to detect the influence of correlated paternity on Pistacia seedling performance. For instance, and even though abiotic factors (especially drought) may be a major cause of the low seedling establishment in the Mediterranean [42, 44, 51], even more than genetic factors [41], our survival analyses did find some evidence of the positive influence of low correlated paternity in a very important stage of the progeny fitness: survival after the first summer. Surviving the first summer in the harsh Mediterranean conditions is critical and this period is the most likely to act as a bottleneck for plant recruitment [41, 44, 52, 53].

Under field conditions, we also observed the influence of correlated paternity in delaying Pistacia emergence time, a fundamental characteristic in the Mediterranean, as mentioned before. In fact, several Pistacia characteristics such as early seedling emergence make this species competent and able to survive efficiently in the Mediterranean climate [54]. Early emergence leads to benefits for seedling growth and fecundity, especially important in a Mediterranean environment in which species have to grow sufficiently during rainy spring to survive the summer drought [47]. The timing of emergence will in fact determine the seedling’s fate as a plant. In our study, it is likely that the early emergence of Pistacia seedlings (as compared to Myrtus) in the field coincided with more benign conditions that resembled those in the greenhouse, which may have led to the similar percentage of emergence observed in the two sowing experiments for this plant. After the first summer, only 8% of Pistacia seedlings survived in the field experiment, a low survival rate but higher than that of Myrtus (in which no seedlings survived to the end of the experiment). Summer water deficit is the main factor of stress and mortality for seedlings in Mediterranean woody plants [41, 42, 44, 51]. Nearly all dead seedlings in our field experiment were found to be desiccated (S. Nora, personal observation) and so seedling mortality is probably mostly attributable to drought. In fact, Myrtus seedling mortality was already 48.57% after just two months of the experiment (approximately at the end of April), by which time the measured volumetric water content in the ground had fallen dramatically. Plant species that do not form persistent soil seed banks (such as Myrtus or Pistacia) depend on a short temporal window after dispersal or on rare rainfall events to be able to recruit. Bearing in mind these conditions, the probability of natural seedling establishment in these two species–and especially in Myrtus–is extremely limited [41].

In this study, seedlings assessed for progeny fitness were not genotyped, which have been ideally the best approach in order to obtain direct fitness–mating system associations. However, we rather used family mating system estimates to detect associations between correlated paternity and progeny fitness. We considered this approach in order to avoid bias in the estimation of the mating system parameters, because genotyping only emerged seedlings could have biased upwardly the mating system estimates (considering that the percentage of emerged seeds in the greenhouse was ~ 50% for both species and in the field experiment the percentage was even lower). Bearing this in mind, it is fundamental to reckon that in this study the detected effects between mating system parameters and the fitness variables are indirect.

Our results on progeny fitness-correlated paternity associations found multiple non-significant effects and thus, we recognize that our results might lack significance once a multiple correction is applied. However, we considered that our study present novel results which could promote the forthcoming of new researches on the effects correlated paternity in other plant species and future meta-analysis studies.


We found that the correlated paternity has an impact on different aspects of progeny performance. In this study, we report a strong influence of correlated paternity on the progeny performance of Myrtus, a self-compatible species. Even so, in Pistacia (outcrossed species) the negative consequences of high correlated paternity on progeny emergence time and survival under field conditions were also detected. We conclude that our findings, in combination with a handful of recent empirical studies, suggest that mating pattern parameters–other than outcrossing rates–influence progeny performance in multiple stages during early seedling growth.

Supporting Information

S1 Table. Correlations among maternal plant-level mating system estimates (Maternal homozygosity by loci, HL; biparental inbreeding, tm-ts; outcrossing rate, tm and correlated paternity rp).

Pearson’s correlation coefficients shown.


S2 Table. Individual and overall (SE) genetic parameters (maternal homozygosity by loci and mating system parameters) computed for the selected maternal plants.


S3 Table. Progeny performance of Myrtus communis and Pistacia lentiscus under greenhouse and field conditions.

Data are given as means ± SE.



The authors thank Encarni Rubio for essential assistance and the staff at the greenhouse facility Servicio de Invernadero of the University of Seville.

Author Contributions

  1. Conceptualization: RGA SN AA.
  2. Formal analysis: SN RGA.
  3. Funding acquisition: RGA AA.
  4. Methodology: SN RGA.
  5. Project administration: RGA AA.
  6. Writing – original draft: SN.
  7. Writing – review & editing: SN RGA AA.


  1. 1. Duminil J, Fineschi S, Hampe A, Jordano P, Salvini D, Vendramin G, et al. Can population genetic structure be predicted from life-history traits? Amer Nat. 2007; 169: 662–672.
  2. 2. Hamrick JL, Godt MJW. Effects of life history traits on genetic diversity in plant species. Philos T Roy Soc B. 1996; 351: 1291–1298.
  3. 3. Coates DJ, Sampson JF, Yates CJ. Plant mating systems and assessing population persistence in fragmented landscapes. Aust J Bot. 2007; 55: 239–249.
  4. 4. Eckert CG, Kalisz S, Geber MA, Sargent R, Elle E, Cheptou PO, et al. Plant mating in a changing world. Trends Ecol Evol. 2009; 25: 35–43. pmid:19683360
  5. 5. Lowe AJ, Boshier D, Ward M, Bacles CFE, Navarro C. Genetic resource impacts of habitat loss and degradation; reconciling empirical evidence and predicted theory for neotropical trees. Heredity. 2005; 95: 255–273. pmid:16094300
  6. 6. Mimura M, Barbour RC, Potts BM, Vaillancourt RE, Watanabe KN. Comparison of contemporary mating patterns in continuous and fragmented Eucalyptus globulus native forests. Mol Ecol. 2009; 18: 4180–4192. pmid:19769693
  7. 7. Rocha OJ, Aguilar G. Variation in the breeding behaviour of the dry forest tree Enterolobium cyclocarpum (Guanacaste) in Costa Rica. Am J Bot. 2001; 88: 1600–1606. pmid:21669693
  8. 8. Byrne M, Elliot CP, Yates C, Coates DJ. Extensive pollen dispersal in a bird-pollinated shrub, Calothamus quadrifidus, in a fragmented landscape. Mol Ecol. 2007; 16: 1303–1314. pmid:17391415
  9. 9. de Lucas A, Robledo-Arnuncio JJ, Hidalgo E, González-Martínez SC. Mating system and pollen gene flow in Mediterranean maritime pine. Heredity. 2008; 100: 390–399. pmid:18212806
  10. 10. García C, Arroyo JM, Godoy JA, Jordano P. Mating patterns, pollen dispersal, and the ecological maternal neighbourhood in a Prunus mahaleb L. population. Mol Ecol. 2005; 14: 1821–1830. pmid:15836653
  11. 11. González-Varo JP, Albaladejo RG, Aparicio A. Mating patterns and spatial distribution of conspecific neighbours in the Mediterranean shrub Myrtus communis (Myrtaceae). Plant Ecol. 2009a; 203: 207–215.
  12. 12. Oddou-Muratorio S, Klein E, Demesure-Musch B, Austerlitz F. Real-time patterns of pollen flow in the wild-service tree, Sorbus torminalis (Rosaceae). III. Mating patterns and the ecological maternal neighbourhood. Am J Bot. 2006; 93: 1650–1659. pmid:21642110
  13. 13. Franceschinelli EV, Bawa K. The effect of ecological factors on the mating system of a South American shrub species (Helicteres brevispira). Heredity. 2000; 84: 116–123. pmid:10692018
  14. 14. Neal PR, Anderson GJ. Are “mating systems” and “breeding systems” f inconsistent and confusing terminology in plant reproductive biology? Or is it the other way around? Plant Syst Evol. 2005; 250: 173–185.
  15. 15. Ghazoul J. Pollen and seed dispersal among dispersed plants. Biol Rev. 2005; 80: 413–443. pmid:16094807
  16. 16. Breed MF, Ottewell KM, Gardner MG, Marklund MHK, Stead MG, Harris JBC, et al. Mating system and early viability resistance to habitat fragmentation in a bird-pollinated eucalypt. Heredity. 2015; 115: 100–107. pmid:23188172
  17. 17. Breed MF, Christmas MJ, Lowe AJ. Higher levels of multiple paternities increase seedling survival in the long-lived tree Eucalyptus gracilis. PLOS ONE, 2014: 9, e90478. pmid:24587373
  18. 18. Breed MF, Gardner MG, Ottewell KM, Navarro CM, Lowe AJ. Shifts in reproductive assurance strategies and inbreeding costs associated with habitat fragmentation in Central American mahogany. Ecol Lett. 2012a; 15: 444–452.
  19. 19. Breed MF, Marklund MHK, Ottewell KM, Gardner MG, Harris BC, Lowe AJ. Pollen diversity matters: revealing the neglected effect of pollen diversity on fitness in fragmented landscapes. Mol Ecol. 2012b; 21: 5955–5968.
  20. 20. Llorens TM, Yates CJ, Byrne M, Nistelberger HM, Williams MR, Coates DJ. Complex interactions between remnant shape and the mating system strongly influence reproductive output and progeny performance in fragmented populations of a bird-pollinated shrub. Biol Cons. 2013; 164: 129–139.
  21. 21. Hardy OJ, González-Martínez SC, Colas B, Fréville H, Mignot A, Olivieri I. Fine-scale genetic structure and gene dispersal in Centaurea corymbosa (Asteraceae). II. Correlated paternity within and among sibships. Genetics. 2004; 148: 1601–1614.
  22. 22. Verdú M, González-Martínez SC, Montilla AI, Mateu I, Pannell JR. Ovule discounting in an outcrossing, cryptically dioecious tree. Evolution. 2006; 60: 2056–2063. pmid:17133862
  23. 23. Pannel JR, Labouche AM. The incidence and selection of multiple mating in plants. Philo T Roy Soc B. 2013; 368: 20120051.
  24. 24. Yasui Y. The ‘genetic benefits’ of female multiple mating reconsidered. Trends Ecol Evol. 1998; 13: 246–250. pmid:21238286
  25. 25. Albaladejo RG, Guzmán B, González-Martínez SC, Aparicio A. Extensive pollen flow but few pollen donors and high reproductive variance in an extremely fragmented landscape. PLOS ONE. 2012; 7: e49012. pmid:23152842
  26. 26. Albaladejo RG, González-Martínez SC, Heuertz M, Vendramin GG, Aparicio A. Spatiotemporal mating pattern variation in a wind-pollinated Mediterranean shrub. Mol Ecol. 2009; 18: 5195–5206. pmid:19889041
  27. 27. González-Varo JP, Albaladejo RG, Aparicio A, Arroyo J. Linking genetic diversity, mating patterns and progeny performance in fragmented populations of a Mediterranean shrub. J Appl Ecol. 2010; 47: 1242–1252.
  28. 28. González-Varo JP, Arroyo J, Aparicio A. Effects of fragmentation on pollinator assemblage, pollen limitation and seed production of Mediterranean myrtle (Myrtus communis). Biol Cons. 2009b; 142: 1058–1065.
  29. 29. González-Varo JP. Fragmentation, habitat composition and the dispersal/predation balance in interactions between the Mediterranean myrtle and avian frugivores. Ecography. 2010; 33: 185–197.
  30. 30. Jordano P. Polinización y variabilidad de la producción de semillas en Pistacia lentiscus L. (Anacardiaceae). Anales Jard Bot Madrid. 1988; 45: 213–231.
  31. 31. Jordano P. Pre-dispersal biology of Pistacia lentiscus (Anacardiaceae): cumulative effects on seed removal by birds. Oikos. 1989; 55: 375–386.
  32. 32. Valbuena-Carabaña M, Heredia UL, Fuentes-Utrilla P, González-Doncel I, Gil L. Historical and recent changes in the Spanish forests: A socio-economic process. Rev Palaeobot Palyno. 2010; 162: 492–506.
  33. 33. Aparicio A. Descriptive analysis of the ‘relictual’ Mediterranean landscape in the Guadalquivir River valley (southern Spain): a baseline for scientific research and the development of conservation action plans. Biodivers Conserv. 2008; 17: 2219–2232.
  34. 34. Albaladejo RG, González-Martínez SC, González-Varo JP, Vendramin GG, Aparicio A. Isolation of microsatellite markers for the common Mediterranean shrub Myrtus communis (Myrtaceae). Am J Bot. 2010; 97: e23–e25. pmid:21622432
  35. 35. Albaladejo RG, Sebastiani F, Aparicio A, Buonamici A, González-Martínez SC, Vendramin GG. Development and characterization of eight polymorphic microsatellite loci from Pistacia lentiscus L. (Anacardiaceae). Mol Ecol Resour. 2008; 9: 904–906.
  36. 36. Nora S, Albaladejo RG, Aparicio A. Genetic variation and structure in the Mediterranean shrubs Myrtus communis and Pistacia lentiscus under different landscape contexts. Plant Biol. 2015; 17: 311–319. pmid:25262762
  37. 37. Oosterhout CV, Hutchinson WF, Wills DPM, Shipley P. MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Notes.2004; 4: 535–538.
  38. 38. Aparicio A, Ortego J, Cordero J. What should we weigh to estimate heterozygosity, alleles or loci? Mol Ecol. 2006; 15: 4659–4665. pmid:17107491
  39. 39. Coulon A. GenHet: an easy-to-use R function to estimate individual heterozygosity. Mol Ecol Resour. 2010; 10: 167–169. pmid:21565003
  40. 40. Ritland K. Extensions of models for the estimation of mating systems using n independent loci. Heredity. 2002; 88: 221–228. pmid:11920127
  41. 41. González-Varo JP, Nora S, Aparicio A. Bottlenecks for plant recruitment in woodland remnants: An ornithochorous shrub in a Mediterranean ‘relictual’ landscape. Perspect Plant Ecol Evol Sys. 2011; 14: 111–122.
  42. 42. Herrera CM, Jordano P, Lopez-Soria L, Amat JM. Recruitment of a mast-fruiting, bird-dispersed tree: Bridging frugivore activity and seedling establishment. Ecol Monogr. 1994; 64: 315–344.
  43. 43. Izhaki I, Walton PB, Safriel UN. Seed shadows generated by frugivorous birds in an Eastern Mediterranean. J Ecol. 1991; 79: 575–590.
  44. 44. Rey PJ, Alcántara JM. Recruitment dynamics of a fleshy-fruited plant (Olea europaea): connecting patterns of seed dispersal to seedling establishment. J Ecol. 2000; 82: 622–633.
  45. 45. Kleinbaum DG. Survival Analysis: A Self-Learning Text. New York: Springer-Verslag Inc; 1996.
  46. 46. Cox DR. Regression models and Life-Tables. J Roy Statist Soc. 1972; 34: 187–220.
  47. 47. Verdú M, Traveset A. Early emergence enhances plant fitness: a phylogenetically controlled meta-analysis. Ecology. 2005; 86: 1385–1394.
  48. 48. Skogsmyr IO, Lankinen Å. Sexual selection: an evolutionary force in plants? Biol Rev. 2002; 77: 537–562. pmid:12475053
  49. 49. González-Varo JP, Traveset A. Among-individual variation in pollen limitation and inbreeding depression in a mixed-mating shrub. Ann Botany. 2010; 106: 999–1008.
  50. 50. Armbruster WS, Rogers DG. Does pollen competition reduce the cost of inbreeding? Am J Bot. 2004; 91: 1939–1943. pmid:21652341
  51. 51. García-Fayos P, Verdú M. Soil seed bank, factors controlling germination and establishment of a Mediterranean shrub: Pistacia Lentiscus L. Acta Oecol. 1998; 19: 357–366.
  52. 52. Gómez-Aparicio L. Spatial patterns of recruitment in Mediterranean plant species: linking the fate of seeds, seedlings and saplings in heterogeneous landscapes at different scales. J Ecol. 2008; 96: 1128–1140.
  53. 53. Traveset A, Gulias J, Riera N, Mus M. Transition probabilities from pollination to establishment in a rare dioecious shrub species (Rhamnus ludovici-salvatoris) in two habitats. J Ecol. 2003; 91: 427–437.
  54. 54. Verdú M, García-Fayos P. Ecología reproductiva de Pistacia lentiscus L. (Anacardiaceae): un anacronismo evolutivo en el matorral mediterráneo. Rev Chil Hist Nat. 2002; 75: 57–65.