Molecular Ecology (2004) 13, 2567–2577

doi: 10.1111/j.1365-294X.2004.02251.x

Increased selfing and correlated paternity in a small population of a predominantly outcrossing conifer, Pinus sylvestris

Blackwell Publishing, Ltd.

J . J . R O B L E D O - A R N U N C I O ,*,† R . A L Í A † and L . G I L * *Unidad de Anatomía, Fisiología y Genética, ETSI de Montes, Ciudad Universitaria s/n, 28040 Madrid, Spain; †Centro de Investigación Forestal, INIA, Madrid, Spain

Abstract Outcrossing rate, the rates of ovule and seed abortion, and levels of correlated paternity were estimated in a small population of Pinus sylvestris, a predominantly outcrossing conifer, and were compared with estimates from two widely dispersed woodlands of the same species, showing a range of densities. On average, seed trees of the small population showed an eight-fold higher selfing rate (25 vs. 3%) and a 100-fold greater incidence of correlated paternity (19.6 vs. 0.2%) than did trees from the large populations. No evidence was found of pollen limitation within the remnant stand, as suggested by ovule abortion rates. Investigation of the mating patterns in the small population, based on the unambiguous genealogy of 778 open-pollinated seeds, showed a large departure from random mating. Only 8% of the possible mating pairs within the stand were observed. Correlated paternity rate within a maternal sibship was negatively associated (rs = − 0.398, P < 0.050) with the distance to the nearest neighbour, and shared paternity among maternal sibships was negatively correlated (rs = − 0.704, P < 0.001) with the distance between seed trees. Numerical simulations, based on the estimated individual pollen dispersal kernel, suggest that restricted dispersal might have been the key factor affecting mating patterns in the small population and, together with low population density, may account for the observed mating system variation between the small and the large populations. The results of this study show that a severe size reduction may substantially affect the mating system of a wind-pollinated, typically outcrossed plant species. Keywords: mating system, Pinus sylvestris, pollen dispersal, population size Received 24 February 2004; revision received 7 May 2004; accepted 7 May 2004

Introduction The genetic consequences of small population size are a central concern in the evolutionary and conservation biology of plant species. Besides a direct risk of extinction through demographic and environmental forces (Lande 1988), a reduced number of individuals may involve increased levels of inbreeding and genetic drift, which may in turn influence patterns of genetic diversity and fitness (Ellstrand & Ellam 1993). Inbreeding can greatly reduce individual fitness by increasing homozygosity in species carrying recessive deleterious alleles (Charlesworth & Charlesworth 1987). Increased levels of genetic drift may lead to population genetic Correspondence: L. Gil. Tel. 34 913-365-039; Fax: 34 915 - 439-557; E-mail: [email protected] © 2004 Blackwell Publishing Ltd

diversity erosion and divergence among populations (Hartl & Clark 1997). Empirical results suggest that genetic diversity levels in small plant populations may be sensitive to the continuing action of genetic drift, although the short-term effects of a severe loss of individuals and subsequent inbreeding may often be of greater importance (Young et al. 1996). The first step to improving our understanding of the genetic consequences of small population size in plants is to assess the immediate effects of size reduction on the mating system (Sork et al. 1999). The mating system determines the transmission of genes from one generation to the next through sexual reproduction. It is typically characterized in plants by the outcrossing rate and by the level of correlated paternity, defined as the portion of full-sibpairs among outcrossed maternal progenies (Schoen & Clegg 1984; Ritland 1989). Increased selfing may reduce the

2568 J . J . R O B L E D O - A R N U N C I O , R . A L Í A and L . G I L individual fitness of parents through inbreeding depression in the offspring, especially in typically outcrossed species. More frequent selfing will also result in declining effective population size, which in turn will favour local inbreeding and genetic drift (Loveless & Hamrick 1984). The level of correlated paternity defines the probability that a seed tree draws two male gametes from the same pollen donor, which can be regarded as the inverse of an effective pollination neighbourhood size, analogous to Wright’s neighbourhood size when considering only the dispersal variance of male gametes (Austerlitz & Smouse 2001). The level of correlated paternity, together with selfing rate, will thus condition the departure from random mating and the significance of genetic drift under the isolation-by-distance model (Ritland 1989). Correlated mating may also influence patterns of selection and competition among siblings (Karron & Marshall 1990). Among the major factors that may limit outcrossed paternal multiplicity (defined as the inverse of correlated paternity) in wind-pollinated plants are pollen limitation (Surles et al. 1990), spatially restricted pollen dispersal (Smouse et al. 1999), asynchronous floral phenology and unequal male fecundity (Erickson & Adams 1989; Burczyk & Prat 1997), and low conspecific density (Smouse & Sork 2004). First, population size reduction may directly affect the mating system by simply reducing the number of local compatible mates, which, even under random mating, will increase the likelihood of correlated mating and selffertilization (Surles et al. 1990). Second, total pollen availability will decrease in small plant populations, which may result in reduced seed set and/or increased selfing (Larson & Barrett 2000). The impact of pollen limitation in windpollinated species, however, remains unclear (Koenig & Ashley 2003). Third, as a consequence of the typically leptokurtic shape of the pollen dispersal kernel in plants (Levin & Kerster 1974), pollen pool diversity around individual seed trees (and thus the probability of selfing and correlated paternity) may be conditioned by the existence of a broad spectrum of long-distance pollen donors, counteracting the proximity advantage of neighbouring individuals (Adams 1992; Ellstrand 1992). This rich background pollination could be severely reduced in small isolated populations. Evidence from early experiments on conifer species suggests that both the quantity and diversity of available pollen in small stands may be significantly lower than in large populations (Sarvas 1962; Koski 1970, 1973). Little is known, however, about the precise consequences of this potential pollen pool impoverishment for the mating system of particular species. Most studies dealing with the consequences of small population size on plant mating systems have focused on the outcrossing rate and reproductive output of insectpollinated species, in which the interaction between population spatial context and pollen–vector foraging behaviour

pose an additional challenge (Levin & Kerster 1974; Van Treuren et al. 1993; Hauser & Loeschcke 1994; Kennington & James 1997; Routley et al. 1999; Paschke et al. 2002). Although no significant effects of small population size were detected in some of these studies, a general trend towards increased selfing and reduced seed set has been observed as population sizes decrease. None of these studies provided information regarding the levels of correlated paternity. Analogous studies on wind-pollinated woody perennials are timely, but have drawn little attention in the literature (see Lian et al. 2001). Scots pine (Pinus sylvestris L.) is a long-lived, monoecious, wind-pollinated, predominantly outcrossing conifer (Muona & Harju 1989), widely distributed across Eurasia. Like many conifers, it lacks a prezygotic self-incompatibility system, but severe inbreeding depression in the seed enforces typically low levels (< 10%) of effective self-fertilization at the seed stage (Sarvas 1962; Kärkkäinen & Savolainen 1993). In this study, we investigated the Scots pine mating system in one small remnant population and two widespread woodlands of the species by assessing the proportion of unpollinated ovules and aborted seeds, as well as the levels of self-fertilization and correlated paternity at the seed stage. We addressed the following questions: 1 What are the levels of selfing and correlated paternity in the small vs. the large populations? 2 Is there evidence of pollen limitation in the small vs. the large populations? We had previously extracted detailed information on the individual pollen dispersal kernel in the same small population, by means of a total-exclusion paternity analysis (Robledo-Arnuncio & Gil 2004). Using numerical simulations based on this dispersal kernel, we additionally explored the following questions: 3 How does intermate distance influence mating patterns in the small stand? 4 To what extent might typical values of male fecundity variation and asynchronous phenology of Scots pine account for the mating system estimates in the small stand? 5 Might differences in population size and density account for the observed mating system variation among the small and the large populations?

Materials and methods Study populations We assessed mating system variation in one small and two large native Scots pine populations, located in the Northern Meseta region, northwest Iberian Peninsula (Fig. 1). The small population (Coca) consists of 36 isolated adult trees, scattered over an approximate area of 15 ha (~2.4 trees/ha) and included within a large maritime © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

M A T I N G S Y S T E M V A R I A T I O N I N P I N U S S Y L V E S T R I S 2569 Table 1 Size, density and exposure of the five Scots pine populations examined

Fig. 1 Map of the native distribution of Scots pine (Pinus sylvestris) in the northwestern Iberian Peninsula, indicating the location of the study populations. Valsaín (Val) and Navafría (Nav) are widely spread monospecific woodlands, while Coca consists of 36 isolated trees, distributed as shown in the enlarged area, growing within a continuous maritime pine (P. pinaster) forest.

pine (Pinus pinaster) forest. It is located in the inner Meseta plains, 30 km from the nearest P. sylvestris population, and represents a remnant stand of the more widespread distribution of this species that occupied this sedimentary plateau until the Late Holocene period (~1500 years bp; Franco-Múgica et al. 2001). We failed to find any other native populations of the species in the region which consist of fewer than several thousand trees. Historical documents indicate that the size of the Coca population of P. sylvestris was still economically significant in the 16th century (Álvarez & Allué 1997). The severe size reduction leading to the current numbers must have occurred in recent times, probably due to the human-mediated lowering of the water table in the area (Franco-Múgica et al. 2001). As a consequence of its geographical isolation, contemporary long-distance immigration of effective pollen into the Coca remnant is below 5% (Robledo-Arnuncio & Gil 2004). The two large populations (Valsaín and Navafría) are monospecific woodlands, each covering thousands of hectares in the Guadarrama Chain, at the southern boundary of the plateau, 60 km to the southeast of Coca (Fig. 1). Two different sites were selected, with different densities, within each of the two large populations (ranging from 80 to 315 trees/ha; Table 1).

Mating system estimation We compared mating system differences of the five study sites by estimating, for each, the proportions of unpollinated ovules (UO), aborted seeds among pollinated ovules (AS ) and selfs among nonaborted seeds (s), and the correlation among paternal gametes within maternal © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

Population

No. of trees

Density (trees/ha)

Exposure

Coca Val1 Val2 Nav1 Nav2

36 > 105 > 105 > 105 > 105

2.4 80.4 183.2 183.7 315.3

All North North North South

outcrossed sibships (rp). In Scots pine, unpollinated ovules shrink, developing wings but not seeds, whereas an aborted (empty) seed will result if an ovule is pollinated but all the embryos subsequently die during seed development (Sarvas 1962). Sound (filled) seeds can develop from either self- or cross-fertilization. The proportion of unpollinated ovules in the seed crop is negatively correlated with total pollen availability in the stand (Sarvas 1962), whereas the relative amount of self-pollen affects the proportion of aborted seeds and the effective self-fertilization rate at the seed stage (Kärkkäinen & Savolainen 1993). We collected 16 randomly selected, open-pollinated cones from each cone-bearing tree (total 35 trees, 560 cones) in the small population, and from each of 15 seed trees within each of the four sampling sites (total 240 cones per sampling site) in the large populations. We opened cones from each tree, extracted the seeds, and counted the number of aborted ovules (wings without seed), empty seeds and filled seeds, following standard ISTA procedures (MacKay et al. 1976). For the filled-seed samples, we obtained estimates of selffertilization and correlated paternity based on two previous molecular assays. In the Coca isolated population, we had performed an unambiguous paternity assignment for 22–24 filled seeds from 34 of the 35 trees, using a totalexclusion set of four chloroplast and one nuclear microsatellites, as described in Robledo-Arnuncio & Gil (2004). We thus had direct estimates of individual and population selfing rates. For this study, and using these categorical paternal designations, we calculated rp for each seed tree in Coca by directly computing the proportion of full-sibs among the total number of outcrossed offspring pairs within its progeny (Ritland 1989). For each of the four sampling sites in Valsaín and Navafría populations, we had previously assessed the paternal contribution to a 12-seed sibship from each of the 15 sampled seed trees by means of three highly polymorphic nuclear microsatellite markers (described in Robledo-Arnuncio et al. 2004). For this study and for each of the sites, we used this molecular information to estimate jointly the multilocus self-fertilization rate at the seed stage (s) and the rate of correlated paternity (rp), using the correlated-mating model of Ritland (2002), as implemented in mltr 2.4 software. We

2570 J . J . R O B L E D O - A R N U N C I O , R . A L Í A and L . G I L obtained variances for the estimates by bootstrapping over families 500 times. We evaluated the variation in the different mating system parameters among the study sites using a Kruskal–Wallis nonparametric test, because population variances were not equal and the variables were not normally distributed, even after being arcsine square-root transformed. We conducted pairwise population comparisons of means using the T3 multiple comparison method for unbalanced designs with unequal variances, included in spss 11.5 statistical package (SPSS 1999).

Mating patterns in the small population We performed a more detailed investigation of the mating patterns within the Coca population, based on the unambiguous genealogies of 778 seeds collected from all conebearing trees in the stand (Robledo-Arnuncio & Gil 2004). To provide a measure of the departure from panmixia, we used the effective number of mating pairs (Nmp), defined as the inverse of the probability that two randomly drawn offspring from among the population total seed crop are full-sibs. We estimated Nmp by computing: N mp

n n   2 = ∑ I ij   n(n − 1) ∑ i j >i  

−1

(2)

where n = 778 is the number of seeds, and Iij is the indicator function of the event ‘the i-th and j-th seeds are full-sibs’, equaling 1 when true and zero when false. Two possible mating pairs were considered for each adult pair, depending on which individual acted as male or female. Because the same number of seeds was collected from each tree, unequal female fecundity was not reflected in Nmp. The expected value of Nmp under panmixia (random mating with self-fertility) in a 36-tree population is Nmp = 362 = 1296. Confidence intervals for Nmp were obtained by bootstrapping (10 000 trials) over offspring in the total seed crop. A previous study showed that the effective self-fertilization rate at the seed stage was negatively correlated (rs = −0.608, P < 0.001) with distance from the seed tree to its nearest conspecific neighbour in the Coca population (RobledoArnuncio & Gil 2004). To explore further the relationship between the spatial location of the trees and the mating patterns, we here investigated the association between correlated paternity (rp ) and the distance to the nearest conspecific pollen donor, testing whether the isolation of seed trees had a significant effect on the diversity of outcrossed male gametes they recovered. We also calculated the correlation among the paternal gametes recovered by different seed tree pairs (rp , as ), by counting the proportion of paternal half-sibs among maternal sibship pairs (OJ Hardy, SC González-Martínez, B Colas, H Fréville, A Mignot & I Olivieri, unpublished), and investigated its association

with the spatial distance between those seed trees. In doing so, we explored how the relative spatial location of a given pair of seed trees in the stand affects the probability that they share the same set of effective pollen donors.

Simulations We performed a series of numerical simulations to model pollen dispersal in the Coca population and further investigated how spatial and flowering phenological separation, as well as male fecundity variation, might have affected mating patterns in the small population. First, we estimated mating system parameters under the null hypothesis of panmixia. To do so, random mating with self-fertility was simulated by generating a 24-embryo sibship for each seed tree of the stand, with the male parent of each offspring randomly chosen from among all the 36 potential pollen donors. Under this null model, spatial, fecundity and phenological factors are completely ignored. Second, we simulated restricted pollen dispersal with a bivariate exponential-power probability density function of dispersal distances from individual trees (dispersal kernel; Clark 1998): b   2 2  b   x +y   p (x, y ) = exp  − ,   a 2πa2 Γ(2/b)     

(1)

where p(x, y) denotes the probability per unit area of pollen dispersal to a point at a distance x2 + y 2 from the individual source, Γ is the gamma function, a is the scale parameter for distance, and b is the shape parameter (Clark 1998). We used parameter estimates of â = 24.08 and b = 0.67, previously obtained from the Coca data set (RobledoArnuncio & Gil 2004). To model mating patterns under the kernel law, we generated a 24-seed sibship for the j-th seed tree in Coca, the paternity of each offspring being assigned to the i-th pollen donor with likelihood πij, given by (Erickson & Adams 1989; Devlin et al. 1992; S OddouMuratorio, EK Klein & F Austerlitz, unpublished) π ij =

λ i ϕ ij pij

∑ λ kϕ kj pkj

,

(2)

k∈N

where pij denotes the probability, given by the dispersal kernel (eqn 1), of a pollen grain travelling the distance that separates trees i and j, N is the total number of individuals in the stand, λi is the male fecundity of the i-th individual, and ϕij is the proportion of the receptivity period of female j coincident with the pollen shedding period of pollen donor i. We considered four different scenarios of restricted dispersal. For the first, we isolated the effect of limited dispersal by assuming equal male fecundity and globally synchronous phenology (λi = ϕij = 1 for all i and j ). For the second scenario, we evaluated the combined impact of © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

M A T I N G S Y S T E M V A R I A T I O N I N P I N U S S Y L V E S T R I S 2571 restricted dispersal and nonrandom spatial distribution of individuals, by considering the same dispersal and flowering hypothesis as above, but with randomized spatial location of the 36 trees, accomplished by generating fictitious spatial coordinates for each from a Poisson distribution, limited within a circular area that held the expected mean pairwise distance among simulated locations equal to the mean pairwise distance among trees in Coca (180 m). The radius of such an area was estimated numerically (r = 198.8 m). For the third scenario, we allowed restricted dispersal, along with unequal male fecundity. Although we lack measurements of flowering characteristics in Coca, we tested how typical values of male fertility variation for Scots pine would affect population means of the mating system parameters. Reported estimates of the coefficient of variation (CV) in male fertility across nine Scots pine populations range from 53 to 136% (mean 86%; Kang et al. 2003). We considered two different values for CV in our simulations, 50 and 100%, individual λi values being randomly drawn from a normal distribution with mean unity and standard deviation 0.5 and 1.0, respectively. For a given variance, we confirmed that the use of a skewed (Weibull) distribution for generating λi-values did not yield significantly different estimates of the mating system parameters considered. The fourth scenario included the combined effects of restricted dispersal, unequal male fecundity and asynchronous phenology. Because phenological information was not available for Coca, we used detailed measurements of individual phenological phases performed on 24 Scots pine clones located in a nearby seed orchard, with material from the Valsaín population, 60 km from Coca (Rodriguez 2000). Individual values for male and female mean date of flowering, and for the length (days) of pollen shedding and female receptivity periods, were generated for each Coca tree by randomly sampling from the observed distributions of these variables in the Valsaín seed orchard. The ϕij

phenological overlapping coefficients in eqn 2 were then calculated as the proportion of the receptivity period of the j-th female coincident with the pollen-shedding period of the i-th male (Erickson & Adams 1989), assuming uniform pollen anthesis and female-cone receptivity. The expected values of self-fertilization, effective number of mating pairs and correlated mating within maternal sibships under each scenario were calculated based on the simulated progenies, and confidence intervals for the estimates were obtained from 1000 random repetitions of the process. We performed a last simulation to explore how variation in the number of Scots pine reproductive individuals might affect population outcrossing and correlated paternity rates. To do so, we simulated a series of populations with increasing numbers of randomly distributed trees, from 36 to 3000, using iteration steps of 75 trees. In a first scenario, population area was held constant (circular 15-ha surface), and we adjusted the density of trees from a low of 2.4 trees/ha, as at Coca, to a high of 200 trees/ha, close to the average value in the large Scots pine populations in the mountains. In a second scenario, we held population density constant at 2.4 trees/ha, as at Coca, but we varied the total population from 15 to 1250 ha. We simulated a 24seed sibship for each seed tree, as described above, under the hypothesis of restricted dispersal, unequal male fecundity (CV = 100%), and asynchronous phenology, from which we estimated the levels of outcrossing and correlated paternity. We computed average values of the target parameters for each population size and density from 10 independent runs.

Results Mating system variation There was significant heterogeneity in all four mating system parameters among the sampled populations (Kruskal– Wallis H = 20.6 – 41.0, P < 0.001; Table 2). The effective

Population

N

n

UO

AS

s

rp

Coca Val1 Val2 Nav1 Nav2 H P (df = 4)

35 15 15 15 15

778 175 175 179 179

39.0 (23.1) A 33.3 (24.0) A 34.1 (22.0) A 29.4 (14.2) A 65.9 (14.9) B 20.6 < 0.001

56.3 (22.0) A 31.8 (16.9) B 34.7 (13.7) B 23.3 (13.5) B 45.3 (19.6) AB 29.6 < 0.001

25.2 (30.3) A 3.3 (6.1) B 6.3 (8.5) B 1.7 (3.4) B 1.1 (2.9) B 23.3 < 0.001

19.6 (21.0) A 0.1 (0.0) B 0.0 (0.0) B 0.2 (0.1) B 0.5 (0.3) B 41.0 < 0.001

SD-values are indicated between brackets. N trees were sampled per population, 16 cones per tree, and n seeds were analysed to obtain s and rp estimates. Arcsine-transformed percentages were compared with a Kruskal–Wallis test (H). Pairwise comparisons were made with the T3 test. Means with at least one common letter are not significantly different at the P = 0.05 level. © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

Table 2 Mating system variation among Pinus sylvestris populations. Variables are the proportion of aborted (unpollinated) ovules (UO), proportion of aborted seeds (AS), self-fertilization at the seed stage (s), and correlated paternity within maternal sibships (rp)

2572 J . J . R O B L E D O - A R N U N C I O , R . A L Í A and L . G I L self-fertilization rate at the seed stage (s) was very high in the small population (mean 25.2%, SD = 30.3), and significantly larger (T3 tests, P < 0.01) than any of the low values observed in the four higher density sites within the widespread woodlands (mean 1.1– 6.3%, SD = 2.9 – 8.5), among which multiple comparisons showed no significant differences ( T3 tests, P > 0.05). Correlated paternity within outcrossed sibships (rp ) was substantially (mean 19.6%, SD = 21.0) and significantly higher ( T3 tests, P < 0.01) in the small population than in any of the sites within the large populations (mean 0.0 – 0.5%, SD = 0 – 0.3), among which estimates were very similar and not significantly different from each other or zero (T3 tests, P > 0.05). A different pattern of variation emerged for the proportions of unpollinated ovules (UO, Table 2). The small population, together with three of the locations within the large populations, formed a homogenous group ( T3 tests, P > 0.05), with values of UO ranging from 29.4 to 39.0% (SD = 14.2–23.1), as opposed to Nav2, the only site with a southern exposure and the one with the highest tree density among the larger populations, which showed a significantly higher value, UO = 65.9% (SD = 14.9; Tables 1 and 2), suggesting pollen limitation. Among pollinated ovules, the proportion of aborted seeds (AS) in Coca was 56.3% (SD = 22.0), an estimate that was significantly higher (T3 tests, P < 0.01) than any observed at Val1 (31.8%, SD = 16.9), Val2 (34.7%, SD = 13.7) and Nav1 (23.3%, SD = 13.5) sites within the large populations, and higher, but not significantly so (T3 test, P > 0.05), than the 45.3% (SD = 19.6) value recorded for the Nav2 site. The AS estimates were not significantly different (T3 test, P > 0.05) among the four sites within the widespread woodlands.

Mating patterns in the small population The effective number of mating pairs (Nmp) contributing to the 778-seed sample collected in the Coca population, as revealed by unambiguous paternity assignment, was 104.4 (SD = 6.2), only 8% of the 362 = 1296 possible under random mating. Indeed, only 264 different mating pairs were detected among the 778-seed sample. The effective selfing rate within maternal sibships ranged from 0 to 92% (mean 25.2, SD = 30.3), and the full-sib rate ranged from 0 to 91% (mean 19.6, SD = 21.0) among outcrossed seed. The rate of correlated paternity within individual outcrossed sibships (rp) showed a weak, but significant, inverse association with the distance to the closest tree (rs = −0.398, N = 34, P = 0.020, Fig. 2a). Values of rp > 0.30 were observed only for seed trees within 9 m of the nearest neighbour, although other seed trees in analogous spatial situations showed lower paternity correlation (0.10 < rp < 0.20, Fig. 2a). Among sibship pairs, correlated paternity was in the range 0 < rp, as < 0.55 (mean 0.03, SD = 0.05), showing a moderate inverse relation with distance among

Fig. 2 Relationship (a) between correlated paternity within maternal sibships, rp, and distance from the index seed tree to the closest neighbouring conspecific tree, and (b) between correlated paternity among sibship pairs, rp, as, and pairwise distance, observed in the Coca Pinus sylvestris population, based on the unambiguous paternity assignment of 22–24 seeds from each seed tree. The scale of the horizontal axis is logarithmic. Spearman rank correlation coefficients between the variables are displayed.

tree pairs (rs = −0.704, N = 561, P < 0.001, Fig. 2b), indicating that the spatial location of seed trees determines the sources of the pollen they sample.

Effective pollen dispersal simulations The simulation analysis of pollen dispersal in Coca highlighted the substantial departure from random mating in the population, with all observed mating system parameters showing large differences from the simulated values under panmixia (Table 3). Restricted dispersal alone, as determined by the estimated individual pollen dispersal kernel, was the most influential factor in elevating the rates of self-fertilization and correlated paternity within maternal sibships, as well as in deflating the number of effective mating pairs, relative to random mating (Table 3). Randomization of the spatial location of trees, holding global density constant, did not result in significant changes in © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

M A T I N G S Y S T E M V A R I A T I O N I N P I N U S S Y L V E S T R I S 2573 Table 3 Observed and simulated mating system parameters for the small Coca population of Pinus sylvestris under different assumptions Scenario

s

Nmp

rp

Observed Panmixia Restricted Restricted-RD Restricted-UMF(50) Restricted-UMF(100) Restricted-UMF(100)-AP

0.252 (0.150 – 0.353) 0.028 (0.027 – 0.028) 0.340 (0.315 – 0.366) 0.392 (0.346 – 0.435) 0.326 (0.298 – 0.352) 0.296 (0.257 – 0.330) 0.262 (0.203 – 0.321)

104.4 (91.8–117.4) 1296.3 (1141.1–1496.7) 187.1 (170.2–205.1) 187.8 (159.7–223.0) 180.0 (159.5–203.7) 163.7 (136.0–196.6) 144.7 (112.3–181.2)

0.196 (0.125–0.266) 0.028 (0.027– 0.028) 0.093 (0.075– 0.117) 0.095 (0.073–0.119) 0.106 (0.080– 0.139) 0.135 (0.097– 0.182) 0.181 (0.125–0.253)

s is self-fertilization rate, Nmp is the effective number of mating pairs, and rp is the correlation of paternity within maternal sibships. The 95% confidence intervals are indicated between brackets. Restricted denotes leptokurtic pollen dispersal, RD random distribution of trees, UMF(50) unequal male fecundity (coefficient of variation), AP asynchronous phenology (see methods for details).

mating system parameters under restricted dispersal, although there was a slight increase in the rate of selfing, relative to restricted dispersal in the observed spatial context. Unequal male fecundity and asynchronous phenology had a similar (additive) effect on mating system parameters, when consecutively added to the model, i.e. increasing correlated paternity, reducing Nmp and decreasing selffertilization rate, relative to restricted dispersal alone (Table 3). It is noteworthy that, under the assumed restricted pollen dispersal model, the expected decrease in selffertilization (relative to the hypothesis of equal fecundity) of the set of individuals with below-average male fecundity is comparatively larger than the expected increase in those with above-average male fecundity, resulting in a lower population rate of self-fertilization when variance in male fecundity increases. The best approximation to the observed estimates was obtained when restricted dispersal, male fertility variation and asynchronous phenology were combined, yielding values of selfing (s = 0.262) and correlated paternity (rp = 0.181) whose 95% confidence intervals included the observed values (Í = 0.252, rp = 0.196; Table 3). The observed value of Nmp (Nmp = 104.4) lay slightly below the 95% confidence interval of the simulated value (Nmp = 144.7) in the final model including all factors. The simulation analysis showed that increasing the number of reproductive individuals through increased density (with the area held constant) had a greater effect on selfing and correlated paternity than increasing the population area (with density held constant; Fig. 3). An increment of density from the initial value of 2.4 trees/ha (N = 36 trees) to 20 trees/ha (N = 300 trees) was enough to decrease both selfing and correlated paternity below 5% in the 15-ha simulated population. The rate of decrease was asymptotic with increasing density; beyond 133 trees/ha (N = 2000 trees), both mating system parameters decreased slightly below 1% ( Fig. 3). Increasing the area, however, with density held constant at 2.4 trees/ha, resulted in a comparatively weak reduction of selfing and correlated © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

Fig. 3 Self-fertilization (s) and correlated paternity within maternal sibships (rp) as functions of the number of reproductive individuals in simulated populations under different demographic scenarios: constant 2.5 trees/ha density (s , rp ); constant 15-ha area (s , rp ). Random distribution of trees, spatially restricted pollen dispersal, unequal male fecundity (CV = 100%), and asynchronous phenology were assumed.

paternity, which reached asymptotic values near 22 and 10%, respectively, beyond 830 ha ( N = 2000 trees; Fig. 3).

Discussion Levels of correlated paternity and self-fertilization at the seed stage were markedly higher in a small Scots pine population than in widespread woodlands of the species. Seed trees of large populations showed (on average) eightfold lower selfing rates and 100-fold lower correlation among paternal gametes than did seed trees of the small stand, among which a quarter of the offspring were selfs and a fifth of outcrossed sib-pairs were full-sibs (Table 2). This unusual departure from the typically highly outcrossed mating system of Scots pine (Muona & Harju 1989) was observed in a population whose severe size reduction has taken place very recently, due to anthropogenic disturbance, reflecting the immediate effect that this process may

2574 J . J . R O B L E D O - A R N U N C I O , R . A L Í A and L . G I L have on the mating patterns of a predominantly outcrossing, wind-pollinated conifer. No evidence of pollen limitation was found at lower plant densities or following population size reduction in this study, however, as estimated by the proportion of aborted (nonpollinated) ovules. In fact, the highest ovule abortion rate was observed within one of the large Scots pine populations, in the site with the highest tree density (Nav2, Table 2). This site is located on a rocky and steep southfacing slope, and it can be hypothesized that local ecological factors may be the reason for the observed difference. By contrast, it must be noted that, although a similar (or even higher) percentage of fertile cone scales was successfully pollinated in trees in the small Coca population than in the widespread woodlands, this was at the cost of increased selfing, suggesting outcrossed pollen limitation (Table 2). Scots pine lacks a prezygotic self-incompatibility system, and the rate of self-fertilization at the seed stage is mainly determined by the amount of self-pollination and the degree of in-cone inbreeding depression (Sarvas 1962; Kärkkäinen & Savolainen 1993). Selfing rate differences among populations may thus arise from environmental factors influencing self-pollen success and from variation in the numbers of embryonic lethal genes (Kärkkäinen et al. 1996). Although the assessment of potential differences in the level of early inbreeding depression would require experimental self-pollinations and cross-pollinations in a common environment, these differences are unlikely to exist between the small and large populations in this study. All these populations belonged to the same widespread woodlands at least until 1500 years ago (Franco-Múgica et al. 2001), and the severe size reduction of Coca, leading to the current dimensions, took place over the last few decades. Because Scots pine generation time is expected to be long (~80 years; Kärkkäinen et al. 1996), a potential increase in selection against recessive lethals through inbreeding (purging) has probably not yet begun. If variation in the level of early inbreeding depression among populations can be disregarded, observed differences in the effective self-fertilization rate at the seed stage and in the proportion of aborted seeds must be consistent with a reasonable range of values for the amount of selfpollination. Pollen-catch experiments with Pinus sylvestris suggest that the mean percentage of self-pollination is between 10 and 25% in large, normal density populations, whereas it reaches 40 – 80% when the size of the stands decreases below 20 – 30 ha (Sarvas 1962; Koski 1970, 1973). According to a theoretical model that accounts for the polyzygotic polyembryony system of P. sylvestris (Savolainen et al. 1992), and assuming the estimated values for the number of embryonic lethal equivalents (5–6) and embryo environmental mortality (0.3) in Scots pine (Sarvas 1962; Kärkkäinen et al. 1996), these amounts of self pollen would result in aborted seed rates ranging from 20 to

35% and effective self-fertilization rates at the seed stage ranging from 0 to 7% in the case of large populations. The corresponding estimates in small stands would be 40–65% for the seed abortion rate and 10–45% for the selffertilization rate. Observed values in the large Valsaín and Navafría populations and in the small Coca remnant match these theoretical expectations quite closely under the formulated assumptions (Table 2). Observed population variation in self-fertilization and seed abortion is thus compatible with reported differences in the amount of self-pollination for different sized populations of this species, under the initial assumption of equal early inbreeding depression. The degree of self-pollination can vary among populations due to the number and spatial distribution of reproductive individuals. In wind-pollinated species, self-pollen fertilization success is expected to be a function of its relative concentration around the tree crown (mass-action hypothesis, Holsinger 1991). Effective pollen contributions from the same tree (selfing) or from neighbouring individuals may be favoured when the total number of potential pollen donors is small or plant density is low (Levin & Kerster 1974; Loveless & Hamrick 1984). Negative relationships between selfing rate and plant density have been found in different wind-pollinated conifer species (Farris & Mitton 1984; Knowles et al. 1987; Shea 1987). Several other studies have shown, however, that high levels of outcrossing may remain stable in a fairly wide range of densities (~20–300 trees/ha) within large conifer forests (Neale & Adams 1985; Morgante et al. 1991; El-Kassaby & Jaquish 1996). Supporting the later line of evidence are the undifferentiated high outcrossing rates observed in different density sites (80–315 trees/ha) within the large Scots pine populations in this study (Table 2). Substantially increased levels of selfing and correlated paternity were found in the 15-ha Coca population, however, where density was 2.4 trees/ha, suggesting that a severe size or density reduction may in fact greatly affect the mating system. Further experimental replication would be needed, however, to generalize this result, because it is based on observations from a single small population. The simulation analysis, based on the individual pollen dispersal kernel estimated for Scots pine, supported the experimental evidence that the typically high (t = 1 − s > 0.9) outcrossing rates of the species may be reached when a fairly low plant density threshold is exceeded (20 trees/ha), even when population size is as small as 15 ha (Fig. 3). The model also suggests that the probability of correlated paternity may be greatly reduced above a similar density level. These predictions strongly rely on the assumption that the pollen dispersal kernel is independent of population size and density. Although this hypothesis could be acceptable for the case of invariable canopy structure (e.g. different number or density of conspecific trees within a normal density, continuous mixed forest), it could © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

M A T I N G S Y S T E M V A R I A T I O N I N P I N U S S Y L V E S T R I S 2575 become invalid when conspecific density variation is linked to total plant density (e.g. a monospecific forest). Pollen movement may be enhanced by a reduction in total vegetative structure (Di-Giovanni & Kevan 1991), which could increase the pollen pool diversity, favouring outcrossing and multiple mating, and thus counteracting the effect of pollen donor density reduction (White et al. 2002). A significant effect of the relative spatial location of individual seed trees on the identity and multiplicity of their mating partners was found in the Coca population. There was a trend towards higher correlated paternity within outcrossed maternal sibships as the distance to the nearest conspecific neighbour decreased (i.e. outcrossed pollen sampled by more isolated trees tended to be more diverse; Fig. 2a). Increased isolation of seed trees in Coca, however, has been reported to yield elevated proportions of selfs among the offspring (Robledo-Arnuncio & Gil 2004). Furthermore, the correlation among paternal gametes sampled by different seed trees was inversely related to the distance between them (Fig. 2b). All these facts suggest that restricted pollen dispersal resulted in a substantial proximity advantage for self-pollen and pollen from the nearby individuals. Interestingly, besides increased selfing, trees that were less closely surrounded by neighbours had a higher number of outcrossed effective pollen donors (lower correlated paternity), as has also been observed in low-density populations of insect-pollinated tropical trees (Stacy et al. 1996). The simulation analysis of pollen dispersal within Coca provided further support to the key role of restricted dispersal in determining mating patterns in the small population, especially in increasing selfing and decreasing the effective number of mating pairs (i.e. limiting mate availability; Table 3). The dispersal simulations also suggested that male fertility variation and asynchronous phenology might contribute as much as limited dispersal to favouring correlated paternity, although this is a hypothesis conditioned by the model assumptions, and would need further experimental support. It is noteworthy that, although intermate distance is widely accepted as a crucial factor affecting mating probability in wind-pollinated tree species (Erickson & Adams 1989; Burczyk & Prat 1997), the observation of high probabilities of correlated paternity is not invariably the case (Adams 1992). As Adams (1992) points out, the proximity advantage of close individuals may be lost in favour of the numeric advantage of distant pollen donors, resulting in the typically large pollination neighbourhood sizes of wind-pollinated species. Consistent with this hypothesis, the large monospecific Scots pine populations in this study showed correlated paternity estimates not significantly differentiated from zero. The effectiveness of air-borne pollen flow in enhancing multiple mating, however, may be compromised by low conspecific density and/or thick canopies (Smouse & Sork 2004). That would seem to be © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

the case for Coca (with 2.4 Scots pine trees per hectare, scattered within a continuous maritime pine forest), with an effective number of outcrossed pollen donors of Nep = (1/rp ) = (1/0.196) = 5.1. An additional explanation for the low mating multiplicity in the small Coca remnant might be that biparental inbreeding is causing severe early inbreeding depression at the seed stage, reducing the number of compatible tree pairs. The absence of inbreeding and genetic relatedness among the adults in the Coca population, however, makes this hypothesis unlikely (our unpublished results). Comparably low values of Nep have been observed in savanna populations of Quercus lobata (Nep = 3–4), forest fragments of Q. humboldtii (Nep = 1–2), and P. echinata mixed conifer–deciduous forests (Nep = 3–6; all cited in Smouse & Sork 2004). In conclusion, the results of this study support the expectation, rarely tested for wind-pollinated trees, that severe population size reduction may cause increased self-fertilization and correlated paternity in predominantly outcrossing species. Across generations, increased selfing and reduced mating multiplicity may lead to substantially reduced viability in smaller populations, relative to that found in large populations. Long-term sustainability of small remnant stands may be compromised, from among a number of demographic and evolutionary processes, by limited outcrossed mate availability and reduced pollination neighbourhood size.

Acknowledgements We are indebted to S. Oddou-Muratorio, E.K. Klein and F. Austerlitz for providing unpublished information on the kernel-based pollen dispersal model. We thank P.E. Smouse for helpful comments on the manuscript and for reviewing the language. We also thank S. Martín for providing helpful information on Coca population. JJR-A was supported by a PhD scholarship from the Universidad Politécnica de Madrid. This work was financed by DGCN-ETSIM project ‘Conservación y mejora de recursos genéticos de coníferas (2000– 03)’ and by CICYT AGL2000-1545 project.

References Adams WT (1992) Gene dispersal within forest tree populations. New Forests, 6, 217–240. Álvarez JC, Allué M (1997) Aspectos forestales en las Ordenanzas de la Comunidad de Villa y Tierra de Coca de 1583. Proceedings of the 2nd Spanish Forest Conference, Gráficas Pamplona, Pamplona, Spain, pp. 383–388. Austerlitz F, Smouse PE (2001) Two-generation analysis of pollen flow across a landscape. II. Relation between Φft, pollen dispersal and interfemale distance. Genetics, 157, 851–857. Burczyk J, Prat D (1997) Male reproductive success in Pseudotsuga menziesii (Mirb.) Franco: the effects of spatial structure and flowering characteristics. Heredity, 79, 638–647. Charlesworth D, Charlesworth B (1987) Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics, 118, 237–268.

2576 J . J . R O B L E D O - A R N U N C I O , R . A L Í A and L . G I L Clark JS (1998) Why trees migrate so fast: confronting theory with dispersal biology and the paleorecord. American Naturalist, 152, 204–224. Devlin B, Clegg J, Ellstrand NC (1992) The effect of flower production on male reproductive success in wild radish populations. Evolution, 46, 1030 –1042. Di-Giovanni F, Kevan PG (1991) Factors affecting pollen dynamics and its importance to pollen contamination: a review. Canadian Journal of Forest Research, 21, 1155 –1170. El-Kassaby YA, Jaquish B (1996) Population density and mating pattern in western larch. Journal of Heredity, 87, 438 – 443. Ellstrand NC (1992) Gene flow by pollen: implications for plant conservation genetics. Oikos, 63, 77– 86. Ellstrand NC, Ellam DR (1993) Population genetic consequences of small population size: implications for plant conservation. Annual Review of Ecology and Systematics, 24, 217–242. Erickson VJ, Adams WT (1989) Mating success in a coastal Douglasfir seed orchard as affected by distance and floral phenology. Canadian Journal of Forest Research, 19, 1248 –1255. Farris MA, Mitton JB (1984) Population density, outcrossing rate, and heterozygote superiority in ponderosa pine. Evolution, 38, 1151–1154. Franco-Múgica F, García-Antón M, Maldonado-Ruiz J, MorlaJuriasti C, Sainz-Ollero H (2001) The Holocene history of Pinus forests in the Spanish Northern Meseta. Holocene, 11, 343–358. Hartl DL, Clark AG (1997) Principles of Population Genetics, 3rd edn. Sinauer Associates, Sunderland, MA. Hauser TP, Loeschcke V (1994) Inbreeding depression and mating distance-dependent offspring fitness in large and small populations of Lychnis flos-cuculi (Caryophyllaceae). Journal of Evolutionary Biology, 7, 609–622. Holsinger KE (1991) Mass-action models of plant mating systems: the genetic basis of mating system evolution. Evolution, 42, 1235 – 1244. Kang KS, Bila AD, Harju AM, Lindgren D (2003) Estimation of fertility variation in forest tree populations. Forestry, 76, 329– 344. Kärkkäinen K, Koski V, Savolainen O (1996) Geographical variation in the inbreeding depression of Scots pine. Evolution, 50, 111– 119. Kärkkäinen K, Savolainen O (1993) The degree of early inbreeding depression determines the selfing rate at the seed stage: model and results from Pinus sylvestris (Scots pine). Heredity, 71, 160– 166. Karron JD, Marshall DL (1990) Fitness consequences of multiple paternity in wild radish, Raphanus sativus. Evolution, 44, 260 –268. Kennington WJ, James SH (1997) The effect of small population size on the mating system of a rare clonal mallee, Eucalyptus argutifolia (Myrtaceae). Heredity, 78, 252–260. Knowles P, Furnier GR, Aleksiuk MA, Perry DJ (1987) Significant levels of self-fertilization in natural populations of tamarack. Canadian Journal of Botany, 65, 1087–1091. Koenig WD, Ashley MV (2003) Is pollen limited? The answer is blowin′ in the wind. Trends in Ecology and Evolution, 18, 157–159. Koski V (1970) A study of pollen dispersal as a mechanism of gene flow in conifers. Communicationes Instituti Forestalis Fenniae, 70, 1–78. Koski V (1973) On self-pollination, genetic load, and subsequent inbreeding in some conifers. Communicationes Instituti Forestalis Fenniae, 78, 1–42. Lande R (1988) Genetics and demography in biological conservation. Science, 241, 1455 –1460.

Larson BMH, Barrett SCH (2000) A comparative analysis of pollen limitation in flowering plants. Biology Journal of the Linnean Society, 69, 503–520. Levin DA, Kerster HW (1974) Gene flow in seed plants. Evolutionary Biology, 7, 139–220. Lian C, Miwa M, Hogetsu T (2001) Outcrossing and paternity analysis of Pinus densiflora ( Japanese red pine) by microsatellite polymorphism. Heredity, 87, 88–98. Loveless MD, Hamrick JL (1984) Ecological determinants of genetic structure in plant populations. Annual Review of Ecology and Systematics, 15, 65–95. MacKay DB, Arder F, Gordon AG, Hutin C (1976) ISTA international rules for seed testing. Seed Science and Technology, 4, 357–409. Morgante M, Vendramin GG, Rossi P (1991) Effects of stand density on outcrossing rate in two Norway spruce (Picea abies) populations. Canadian Journal of Botany, 69, 2704–2708. Muona O, Harju A (1989) Effective population sizes, genetic variability, and mating system in natural stands and seed orchards of Pinus sylvestris. Silvae Genetica, 38, 221–229. Neale DB, Adams WT (1985) The mating system in natural shelterwood stands of Douglas fir. Theoretical and Applied Genetics, 71, 201–207. Paschke M, Abstract C, Schmid B (2002) Effects of population size and pollen diversity on reproductive success and offspring size in the narrow endemic Cochlearia bavarica (Brassicaceae). American Journal of Botany, 89, 1250–1259. Ritland K (1989) Correlated-matings in the partial selfer Mimulus guttatus. Evolution, 43, 848–859. Ritland K (2002) Extensions of models for the estimation of mating systems using n independent loci. Heredity, 88, 221–228. Robledo-Arnuncio JJ, Gil L (2004) Patterns of pollen dispersal in a small population of Pinus sylvestris L. revealed by total-exclusion paternity analysis. Heredity, in press. Robledo-Arnuncio JJ, Smouse PE, Gil L, Alía R (2004) Pollen movement under alternative silvicultural practices in native populations of Scots pine (Pinus sylvestris L.) in central Spain. Forest Ecology and Management, in press. Rodriguez JA (2000) Variabilidad reproductiva y sus implicaciones en el incremento de la producción de semilla en un huerto semillero de Pinus sylvestris. PhD thesis, Polytechnical University of Madrid, Spain. Routley MB, Mavraganis K, Eckert CG (1999) Effect of population size on the mating system in a self-compatible, autogamous plant, Aquilegia canadensis (Ranunculaceae). Heredity, 82, 518–528. Sarvas R (1962) Investigations on the flowering and seed crop of Pinus sylvestris. Communicationes Instituti Forestalis Fenniae, 53, 1–198. Savolainen O, Kärkkäinen K, Kuittinen H (1992) Estimating numbers of embryonic lethals in conifers. Heredity, 69, 308 –314. Schoen DJ, Clegg MT (1984) Estimation of mating system parameters when outcrossing events are correlated. Proceedings of the National Academy of Sciences of the USA, 11, 5258–5262. Shea KL (1987) Effects of population structure and cone production on outcrossing rates in Engelmann spruce and subalpine fir. Evolution, 41, 124–136. Smouse PE, Meagher TR, Kobak CJ (1999) Parentage analysis in Chamaelirium luteum (L.) Gray (Liliaceae): why do some males have higher reproductive contributions? Journal of Evolutionary Biology, 12, 1069–1077. Smouse PE, Sork VL (2004) Measuring pollen flow in forest trees: an exposition of alternative approaches. Forest Ecology and Management, in press. © 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

M A T I N G S Y S T E M V A R I A T I O N I N P I N U S S Y L V E S T R I S 2577 Sork VL, Nason J, Campbell DR, Fernandez JF (1999) Landscape approaches to historical and contemporary gene flow in plants. Trends in Ecology and Evolution, 14, 219 –224. SPSS Inc (1999) SPSS Base 10.0 for Windows. User’s Guide. SPSS, Chicago. Stacy EA, Hamrick JL, Nason JD et al. (1996) Pollen dispersal in low-density populations of three neotropical tree species. American Naturalist, 148, 275 –298. Surles SE, Arnold J, Schnabel A, Hamrick JL, Bongarten BC (1990) Genetic relatedness in open-pollinated families of two leguminous tree species, Robinia pseudoacacia L. and Gleditsia triacanthos L. Theoretical and Applied Genetics, 80, 49 – 56. Van Treuren R, Bijlsma R, Ouborg NJ, Van Delden W (1993) The effects of population size and plant density on outcrossing rates in locally endangered Salvia pratensis. Evolution, 47, 1094– 1104. White GM, Boshier DH, Powell W (2002) Increased pollen flow counteracts fragmentation in a tropical dry forest: an example

© 2004 Blackwell Publishing Ltd, Molecular Ecology, 13, 2567–2577

from Swietenia humilis Zuccarini. Proceedings of the National Academy of Sciences of the USA, 99, 2038–2042. Young A, Boyle T, Brown T (1996) The population genetic consequences of habitat fragmentation for plants. Trends in Ecology and Evolution, 11, 413–418.

This work was conducted as a part of Juan J. Robledo-Arnuncio’s doctoral research. Robledo-Arnuncio is currently investigating dispersal processes of tree species. Ricardo Alía, researcher at the Forest Research Center of the National Institute of Agricultural Research at Madrid, is interested in the management and conservation of adaptive genetic variation of Iberian tree species. Luis Gil, Professor at the Polytechnical University of Madrid, is interested in conservation genetics of Mediterranean forest ecosystems.

Increased selfing and correlated paternity in a small ...

b = 0.67, previously obtained from the Coca data set (Robledo-. Arnuncio & Gil 2004) ..... JJR-A was supported by a PhD scholarship from the Universidad. Politécnica de .... National Academy of Sciences of the USA, 11, 5258 – 5262. Shea KL ...

465KB Sizes 3 Downloads 224 Views

Recommend Documents

Determinants of Paternity Success in a Group of ...
Dec 18, 2010 - Paternity analysis . Vervets ... the species' degree of reproductive seasonality. ... For this, we present genetic data from wild-caught vervets.

Determinants of Paternity Success in a Group of ...
Dec 18, 2010 - M. Krützen. Anthropological Institute and Museum, University of Zurich, 8057 Zurich, Switzerland e-mail: .... different location than the females on St. Kitts and Nevis. Post hoc genetic ..... Princeton: Princeton. University Press.

Paternity Leave.pdf
Page 2 of 3. Historia del pensamiento político en la Edad Media. Walter Ullmann. Historia del pensamiento político en la Edad Media Walter Ullmann. Descargar Historia del pensamiento político en la Edad Medi ...pdf. Leer en línea Historia del pen

Female multiple mating and paternity in free-ranging ...
zDepartment of Animal and Plant Sciences, University of Sheffield .... treat age as a minimum estimate (i.e. age ¼ 1 in year of ..... 2002). This relatedness coefficient estimates the degree ..... C. S., Strobeck, C., Slate, J. & Coltman, D. W. 2005

Bayesian Estimator of Selfing (BES) - GitHub
Next, download some additional modules for particular mating systems. These files are .... This variant is run by specifying -m AndroID.hs on the command line.

Conn: A Functional Connectivity Toolbox for Correlated and ...
hr(x А y) Б r(x, y) ... x and y represent the spatial locations of two arbitrary voxels, hr represents a ...... [Full Text HTML] [Full Text PDF] [Full Text PDF with Links]. 3.

Male phenotype predicts extra-pair paternity in pied ...
mantle was visually estimated, high values indicating a black plumage (Potti .... Data from secondary broods of polygamous males were not considered in ..... tive of their age and breeding date, extra-pair males were larger and displayed.

Correlated Equilibrium and Concave Games
May 1, 2007 - I acknowledge financial support by The Japan Economic Research ... payoff function with respect to the player's own strategy and call the.

paternity test pdf
Loading… Whoops! There was a problem loading more pages. Retrying... Whoops! There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. paternity test pdf. paternity

Correlated topologies in citation networks and the Web
Received 5 November 2003 / Received in final form 26 February 2004. Published online 14 May ... networks since the 1960's yielding local similarity metrics such as co-citation and ...... tributed, topic specific search services. 5 Growth models.

A national sample of US paternity tests: do ...
From the Johns Hopkins Bloomberg School of Public Health,. Baltimore, Maryland ... be greatly improved if administrative data (from birth cer- tificates) on the ...

Extrapair paternity in Mediterranean blue tits
among nest-boxes were estimated from UTM coordinates. The size of each study plot was calculated using a Geographic Information. System (ArcView GIS 3.2) and a layer from a land cover database. (Spanish forest inventory) including information on the

Male phenotype predicts extra-pair paternity in pied ...
episodes leading to EPP. Therefore, as our analyses accounting for breeding ..... related to degree of extra-pair paternity in the reed bunting. — Nature 371: 698-700. ... Johnson, L.S., Hicks, B.G. & Masters, B.S. (2002). Increased cuckoldry as ..

A prescription for error-proofing and increased ...
Advancements in medicine are helping people live longer, resulting in a ... the world's healthcare systems, a worldwide nursing shortage has reached crisis proportions. due to an ... by eliminating the time spent constantly traveling to either the ..

Increased Automaticity and Altered Temporal Preparation Following ...
5.5 ± 0.7 39.2 ± 2.4. Mean RT (ms). 262 ± 6. 279 ± 6. 278 ± 4. 480 ± 24. Std Deviation of RT (ms). 70 ± 7. 81 ± 7. 95 ± 11. 569 ± 79. Number of false alarms.

Conn: A Functional Connectivity Toolbox for Correlated and ...
We describe the methods implemented in the Conn toolbox for the analysis of fcMRI data ...... Kim JH, Lee JM, Jo HJ, Kim SH, Lee JH, Kim ST, Seo SW, Cox RW,.

Correlated Equilibria, Incomplete Information and ... - Semantic Scholar
Sep 23, 2008 - France, tel:+33 1 69 33 30 45, [email protected]. ..... no coalition T ⊂ S, and self-enforcing blocking plan ηT for T generating a.

Placement Performance Report Increased Transparency and ...
developed the Placement Performance report, which offers more insight into where ... Google Analytics, the Placement Performance report becomes a powerful ...

Serambi sees increased traffic and on...
channel, a Facebook page, an RSS feed, and email and text message mailing lists to keep ... Users may also see contextually targeted Google AdWords ads.