Reproductive success relates to population density in self-incompatible populations of the wind-pollinated plant, Plantago maritima
Vanessa Coronel
Master thesis in Biology Examensarbete i biologi, 20p, 2005 Department of Ecology and Evolution, Plant Ecology, Uppsala University Supervisor: Jon Ågren
1 SUMMARY
In animal-pollinated plants pollination success tends to decrease with population density, density-dependent pollination success can be expected also among self-incompatible wind-pollinated plants, but has been little studied. To investigate the occurrence of such density-dependence, I related fruit and seed set to population density of the self- incompatible, wind-pollinated herb Plantago maritima, in the Gryt archipelago, south- eastern Sweden. As predicted, I found strong positive density-dependence reproductive success in the studied populations. Plants in high-density populations produced more fruits and seeds than plants in low-density populations. A lower pollination intensity is a probable explanation for the reduced reproductive success in low-density populations. The results suggest that density-dependent reproductive success may influence patterns of seed production and population dynamics also in wind pollinated plants.
INTRODUCTION
Pollen limitation can reduce plant reproductive success in natural populations, and influence population dynamics and evolution of breeding systems (Culley et al. 2002). In several self-incompatible species, pollen limitation increases with decreasing population density (Burd 1994; Wilcock and Neiland 2002; Ashman et al. 2004). However, most of the studies to date have concerned animal-pollinated plants, and there is little evidence on the importance of density-dependent pollination success in wind-pollinated plants. In self-incompatible animal-pollinated plants, pollen limitation has been attributed to the limitation, behavior and effectiveness of pollinators (Ågren 1996; Kunin 1997; Waites and Ågren 2004). Contrarily, it has been suggested that plants pollinated by wind are less likely to be pollen-limited, and the evolution of wind-pollination has been considered to be driven by pollinator limitation (Proctor et al. 1996; Culley et al. 2002; Wilcock and Neiland 2002). However, recent studies have reported that pollination success in wind-pollinated plants can be strongly affected by the spatial distribution of plants within populations (Knapp et al. 2001; Koening and Ashley 2003, Davis et al. 2004). The efficiency of pollen transfer of wind-pollinated plants decrease rapidly with increasing distance between plants (Culley et al. 2002; Proctor et al. 1996), and wind tunnel experiments with Plantago lanceolata L, have shown that pollen is dispersed over short distances; less than 1 m (Bos et al. 1986; Tonsor 1985; Young and Schmitt 1995). Few experimental studies have investigated the effect of pollen limitation on the reproductive success in wind-pollinated plants. However, some significant exceptions exist. Allison (1990) found that the proportion of fertilized ovules of the dioecious gymnosperm Taxus canadensis increased with increasing pollen production of the nearest male. Knapp et al. (2001) indicated that a low availability of pollen-producing plants reduced seed production in hermaphroditic Quercus douglassii. Similarly, a reduced reproductive success due to pollen limitation has been reported in a low-density
2 population of the hermaphroditic grass Spartina alterniflora, where supplemental hand pollination increased seed production among isolated plants (Davis et al. 2004). A low reproductive output caused by pollen limitation can create a positive density- dependence (Allee effects, Allee et al. 1949), and therefore, reduce the persistence and growth of small plant populations of low density (e.g., Groom 1998; Hackney et al. 2001; Forsyth 2003; Brys et al. 2004). Consequently, to determine the effect of density on plant reproductive success is important for the conservation of declining populations and for management and control of invasive species (Davis et al. 2004; Taylor et al. 2004). In addition, as reproductive assurance has been identified as a major factor promoting the evolution of self-fertilization (Barrett 2002; Culley et al. 2002), studies of pollen limitation and their causes are essential to understand the evolution of breeding systems. In this study, I document the relationship between population density and reproductive success, in the gynodioecious and wind-pollinated herb Plantago maritima based in data collected in 21 populations of the Gryt archipelago, south-eastern Sweden. To test the hypothesis that plant reproductive success is positively related to population density measured as mean distance to nearest pollen donor, plant reproductive components were related to population density between and within populations. Variation in components of reproductive success and population characteristics were evaluated between populations. Fruit and seed-set were related to sex ratio to examine whether there was any cause in pollen limitation with increasing proportion in females in the populations.
MATERIALS AND METHODS
Study species Plantago maritima L (Plantaginaceae) is a wind-pollinated, protogynous and gynodioecious, perennial rosette herb, which grows in salt marshes and seashore meadows. It is a widespread plant in Europe (Hultén and Fries, 1986). Hermaphrodite flowers have four stamens with long filaments and yellow anthers. Flowers are arranged in a dense inflorescence, and the number of inflorescences per plant varies greatly from one to many (> 20). Protogyny occurs both within flowers and inflorescences, and varies considerably (Dinnétz 1997). A strong but incomplete self-incompatibility mechanism has been found in plants of P. maritima from Gryt archipelago, however self- compatibility has been detected in population from northern Sweden (Nilsson 2005).
Study sites This study was conducted in the Gryt archipelago (58°47' N, 16°50' E), south-eastern of Sweden (Fig. 1). A total of 21 populations were studied (Table 1). A population was defined as a group of plants separated by ≥ 50 m from the nearest conspecific plant (Table 1). The size of each population was estimated as the total number of flowering plants (Table 1).
3 Fig. 1. Location of study populations of Plantago maritima in the Gryt archipelago, south-eastern of Sweden. See Table 1 for names of the islands.
Table 1. Characteristics of the studied Plantago maritima populations in the Gryt archipelago, Sweden. Mean distance to Proportion Number of nearest pollen Populations of females flowering donor (%) plants (cm) 1 Tryskär 5.3 12 400 2 St. Myrholmen 9.8 8.6 1000 3 Kättilö norra 64.5 12 17 4 Hamnholmen 9.2 6.1 3000 5 Flöjsholmen 9.9 7.1 1000 6 St. Bockholmen södra 21.6 6.4 30 7 St. Bockholmen norra 78.2 7.8 95 8 Armnö syd 21.9 7.6 26 9 Lilla Fågelö västra 20 14.2 12 10 Stora Fågelholmen 22.9 22.8 277 11 Låga Svedsholmen 25.7 8.7 200 12 Ämtö Bastun 13.6 9.6 320 13 Långholmen (Ämtö) östra 895 0 4 14 Långholmen (Ämtö) västra 155.8 20 25 15 Getholmen södra 97.7 33.3 6 16 Trissholmen östra 31.6 4 177 17 Stora Gråholmen 41.7 12.5 16 18 Långholmen (Armnö) 34.4 12.7 1200 19 St. Skällö östra 6.6 15.2 5000 20 St. Skällö norra 29.9 6.7 2100 21 Lilla Fågelö östra 40.9 11.1 25
Populations characteristics, density and reproductive success To test if plant reproductive success is positively correlated with the density of pollen- producing plants, I documented the reproductive success of 20 plants in each 21 P. maritima populations, plants were sampled along a transect placed across each population. For each plant, I recorded the distance to the nearest pollen donor and the
4 number of inflorescences. Three inflorescences were collected from each plant and were brought to the laboratory. If the plant had less than 3 inflorescences, all inflorescences were collected. In the lab, the number of flowers and mature fruits produced by each of the sampled inflorescences were counted under a stereomicroscope. The number of flowers per plant was estimated as number of flowers per inflorescence × number of inflorescences per plant. Fruit set was quantified as the proportion of flowers that developed into a fruit. To estimate seed set and seed mass, ten fruits were randomly selected from each sampled plant and were opened, and the number of seeds was recorded. The mean number of seeds per fruit was calculated by dividing the total number of seeds by the number of mature fruits. Seeds per plant was estimated as mean number of seeds per fruit × mean number of fruits produced per inflorescence × number of inflorescences per plant. To calculate seed mass (mg), the mass of the seeds from the ten fruits was divided by the number of seeds formed in those fruits. For populations 1 and 2 (Table 1), data on number of flowers and fruits were lost from the 20 plants sampled above. Therefore flowers and fruits per inflorescence were counted on 200 inflorescences collected to estimate the sex expression (see below). This data set was used at the population level, to estimate fruits per flower, flowers per plants and seeds per plant. Flowers per plant was estimated as mean number of flowers per inflorescence (mean from 200 inflorescences) × mean number of inflorescences per plant (mean from 20 plants), and seeds per plant was estimated as mean number of seeds per fruit (mean from 20 plants) × mean fruits per flower (mean from 200 inflorescences) × mean number of inflorescences per plant (mean from 20 plants). I estimated population density as the mean distance to the nearest pollen donor (Table 1). The frequency of female plants was estimated by collecting one inflorescence from 200 plants along a transect placed across each population. In populations of less than 200 individuals, one inflorescence was collected from all individuals. The inflorescences were brought to laboratory and the sex expression was determined under a stereomicroscope. Intermediate plants (with partially reduced male function) were included among the hermaphrodite plants to calculate the frequency of females (Table 1).
ANALYSIS
Populations characteristics To examine among-population variation in density (mean distance to nearest pollen donor), reproductive success (number of flowers per plant, fruit set, number of seeds per fruit, number of seeds per plant), and seed mass, I used one-way analysis of variance (ANOVA). The relationships between population density (mean distance to nearest pollen donor), plant size (mean number of flowers per plant), population size (number of flowering plants), and female frequency (proportion of females), were examined using Pearson correlations.
Population density and reproductive success The occurrence of density-dependent reproductive success was tested with a linear regression. Mean number of flowers per plant, fruit set, number of seeds per fruit, number of seeds per plant and seed mass, were regressed on population density (mean distance to
5 nearest pollen donor). A multiple regression analysis was used to test if among- populations differences in seed output can be explained by variation in plant size (number of flowers per plant) and population density (mean distance to nearest pollen donor). An analysis of covariance was performed to examine how seed output of individual plants varies with number of flowers per plant and distance to the nearest pollen donor. Number of seeds per plant was used as a dependent variable with population as a categorical variable and flowers per plant and distance to nearest pollen donor as covariates. I investigated the interaction terms, and in the case of non-significance they were removed from the model. For this analysis I selected the nine populations (3, 7, 11, 14, 16, 17, 18, 20, 21; Table 1), which included one or more plants with a distance of more than 50 cm to the nearest pollen donor, and consisted of more than 15 individuals. Linear regression analyses were used to determine the relationships between fruit, seed-set, and female frequency. To obtain normality and homogeneity of variances, fruit set was arcsine square-root transformed, and number of flowers per plant and number of seeds per plant were log10 transformed. All statistical tests were performed with SPSS version 11.0 (SPSS, Inc., Chicago IL, USA).
RESULTS
Populations characteristics There were substantial differences among populations in measures of reproductive success (number of flowers per plant, fruit set, number of seeds per fruit and number of seeds per plant), seed mass and density (mean distance to nearest pollen donor; Table 2). Mean distance to nearest pollen donor was negatively correlated with population size (number of flowering plants; Table 3). The female frequency varied from 0 to 33% (Table 1), but was not correlated with population density, size or mean number of flowers produced per plant (Table 3).
Table 2. Variation in distance to nearest pollen donor and components of reproductive success among 21 populations of Plantago maritima in the Gryt archipelago, Sweden, analysed with ANOVA. Mean ± SE calculated based on population mean. Source Mean ± SE df MS F P Distance to nearest pollen donor1 1.21 ± 0.47 20 1.97 14.93 < 0.001 Flowers per plant1 1.92 ± 0.47 18 1.06 5.89 < 0.001 Fruits per flower2 0.85 ± 0.44 18 1.24 8.92 < 0.001 Seeds per fruit 1.26 ± 0.26 20 0.20 3.44 < 0.001 Seeds per plant1 1.60 ± 0.16 18 1.23 3.66 < 0.001 Seed mass 0.48 ± 0.16 20 0.10 4.35 < 0.001 1 Log10 transformed, 2Arcsine square-root transformed.
6 Table 3. Matrix of Pearson correlation coefficients for characteristics of Plantago maritima populations in the Gryt archipelago, Sweden. Variable Distance to Plant size Number of Proportion of Variable nearest (mean number flowering females pollen donor1 of flowers per plants1 plant)1 Distance to nearest pollen donor1 --- 0.65** -0.67** 0.01 Plant size1 0.65** --- -0.58** -0.39 Number of flowering plants1 -0.67** -0.58** --- -0.19 Proportion of females 0.01 -0.39 0.19 --- 1 Log10 transformed. ** (P < 0.01)
Population density and reproductive success Plants from low-density populations, produced more flowers, but had a lower fruit set and produced fewer and lighter seeds than plants from high-density populations. The mean number of flowers per plant was positively related to mean distance to nearest pollen donor (linear regression coefficient, b = 0.25, P < 0.001; Fig. 2a). In contrast, fruit- set (b = -0.22, P < 0.001; Fig. 2b), mean number of seeds per plant (b = -0.10, P = 0.04; Fig.2d) and seed mass (b =- 0.07, P = 0.05; Fig. 3) were negatively related to mean distance to nearest pollen donor. The number of seeds per fruit was not related to mean distance to nearest pollen donor (P > 0.5; Fig. 2c).
Fig. 2. Relationships between components of reproductive success and population density (log mean distance to nearest pollen donor) in Plantago maritima populations in the Gryt archipelago, Sweden.
7 Fig. 3. Relationship between seed mass and population density (log mean distance to nearest pollen donor) in Plantago maritima populations in the Gryt Archipelago, Sweden.
Multiple regression analysis indicated a considerable effect of mean plant size (log10 mean flowers per plant) and population density (log10 mean distance to nearest pollen 2 donor) on mean seed output (F2,18 = 13.53, R = 60.1%, p < 0.001). Mean seed output was positively related to mean plant size (partial regression coefficient, b = 0.91, t = 4.3, p < 0.001, Fig. 4a), and negatively related to population density (b = 0.64, t = -5.02, p < 0.001, Fig. 4b). At the individual plant level, seed output varied among populations (covariance analysis, F8,148 = 155.41 p < 0.001), and was positively related to the number of flowers per plant (F1,148 = 155.41 p < 0.001), and negatively related to distance to the nearest pollen donor (F1,148 = 37.12, p < 0.01).
Fig. 4. Partial regression plots for the relationships (a) seed output vs. mean flowers per plant, and (b) seed output vs. mean distance to nearest pollen donor in Plantago maritima populations in the Gryt archipelago, Sweden.
Fruit and seed-set were not related to frequency of females (Fig. 5).
8 Fig. 5. Relationships between (a) proportion of fruits per flower and frequency of females, and (b) seed output per plant and frequency of females in Plantago maritima populations in the Gryt archipelago, Sweden.
DISCUSSION
This study has documented a strong positive density-dependence of reproductive success in the self-incompatible herb Plantago maritima. Plants in high-density populations produced more fruits and seeds than plants in low-density populations. Limited pollen deposition in isolated plants probably reduced the reproductive success in low-density populations. The density-dependent reproductive success in P. maritima may be explained if pollination success decreases with increasing distance to nearest pollen donor (Nilsson et al. 1987; Allison 1990; Knapp et al. 2001; Davis et al. 2004). According with this suggestion, supplemental hand pollination experiments have been found to increase seed set per flower in self-incompatible populations of P. maritima (Dinnétz 1997). The decrease in fruit-set with increasing distance to the nearest pollen donor, suggest that the pollen of P. maritima is dispersed over short distances. Consistent with these results, Dinnétz (1996) showed experimentally that the seed set per flower in populations of P. maritima, decreased rapidly when the distance to pollen donor increase from one to four meters. Wind tunnel experiments with the congeneric P. lanceolata indicated similar results (Bos et al. 1986; Tonsor 1985; Sharma et al. 1992; Young and Schmitt 1995). The low production of fruits and seeds in populations of low density could be also attributed to low resource availability (Knapp et al. 2001; Davis et al. 2004), but this possibility is less likely, because the plants in the low-density populations tended to be larger and produce more flowers, than plants at high-density populations. The high production of flower could reflect a less intense intraspecific competition and possibly also reallocation of resources from fruit to flower production in plants with low production of fruits. A reduction in population size and density can not only affect the efficiency of pollen deposited on plants, but also can influence the quality of pollen through loss compatibles mates, increased relatedness among mates, or an increase in the proportion of females (Ågren 1966; Asikainen and Mutikainen 2003; Ashman et al. 2004). If so, reproduction
9 is expected to decrease in low-density and small populations, in spite of a successful transport and deposition of pollen between plants (Ågren 1996). However, in the natural populations of P. maritima, the seed output of plants tended to increase when spacing between plants decreased, and no relation was found between seed output and the proportion of females. Thus, the lower reproductive output in low-density populations is most likely to be caused by a limited amount of pollen deposited on plants, rather than pollen incompatibility through the loss of compatibles mates or inbreeding depression. A density-dependent reproductive success caused by pollen limitation could have negative effects on the population viability in self-incompatible species, so called Allee effects (e.g., Kunin 2003, Groom 1998; Hackney et al. 2001; Davis et al. 2004). To examine how reduced reproduction affects the dynamics and survival of low-density and small populations, it is necessary to combine estimates of pollen limitation with demographic information. Reproductive assurance is believed to play an important role in the evolution of self-fertilization (Culley et al. 2002), and self-compatibility has recently been detected in plants of P. maritima from northern of Sweden (Nilsson 2005). It would be interesting to set up experimental populations of self-compatible and self-incompatible plants in populations of low and high density, and compare density-dependence in plants of different breeding system. In conclusion, the reproductive success in natural self-incompatible populations of P. maritima shows positive density-dependence. Plants that grow in high-density populations produced more fruit and seeds than plants in populations of low density. Limited pollen deposition in isolated plants probably reduced the reproductive success in sparse populations. The results suggest that a though wind-pollinated plants are independent of animal pollinators, their pollination success is subject to strong density dependence.
ACKNOWLEDGMENTS
I would like to thank my supervisor Jon Ågren, for the opportunity to carry out this study and for the support and valuable commentaries. I thank to Emil Nilsson for the help and support and for all the constructive commentaries and suggestions during the writing. Thanks to Mats Vedin, my husband for all the ideas and support.
LITERATURE CITED
Ågren, J. 1996. Population size, pollinator limitation and seed set in the self-incompatible herb Lythrum salicaria. Ecology 77:1779-1790.
Allee, W.C., A.E. Emerson, O. Park, T. Park, and K.P. Schmidt. 1949. Principles of animal ecology. Saunders, Philadelphia, Pennsylvania, USA.
Allison, T.D. 1990. Pollen production and plant density effect pollination and seed production in Taxus Canadensis. Ecology 71:516-522.
10 Ashman, T.M., T.M. Knight, J.A. Steets, P. Amarasekare, M. Burd, D. R. Campbell, M. R. Dudash, M. O. Johnston, S. J. Mazer, R. J. Mitchell, M. T. Morgan, and W.G. Wilson. 2004. Pollen limitation of plant reproduction: ecological and evolutionary causes and consequences. Ecology 85:2408–2421.
Asikainen E., P. Mutikainen. 2003. Female frequency and relative fitness of females and hermaphrodites in gynodioecious Geranium sylvaticum (Geraniaceae). American Journal of Botany 90:226-234.
Barrett, S.C.H. 2002. The evolution of plant sexual diversity. Nature reviews 3:274-284. Bos, M., H. Harmens, and K. Vrieling. 1986. Gene flow in Plantago I. Gene flow and neighborhood size in P. lanceolata. Heredity 56: 43-54.
Brys, R., J. Jacquemyn, H. Endels, P. Van Rossum, F. Hermy, M. Triest, L. De Bruyn, and G.D.E. Blust. 2004. Reduced reproductive success in small populations of the self-incompatible Primula vulgaris. Journal of Ecology 92:5-14.
Burd, M. 1994. Bateman's principle and plant reproduction: The role of pollen limitation in fruit and seed set: Botanical Review 60: 83–139.
Culley, T.M., S.G. Weller, and A.K. Sakai. 2002. The evolution of wind pollination in angiosperms. Trends in Ecology and Evolution 17:361-369.
Davis, H.G., C.M. Taylor, J.G. Lambrinos, and D.R. Strong. 2004. Pollen limitation cause an Allee effect in a wind-pollinated invasive grass (Spartina alterniflora). Proceedings of the National Academy of Science USA 101:13804-13807.
Dinnétz, P. 1996. Nuclear-cytoplasmatic male sterility and population dynamics in Plantago maritima. Doctoral thesis. Stockholm University.
Dinnétz, P. 1997. Male sterility, protogyny, and pollen-pistil interference in Plantago maritima (Plantaginaceae), a wind-pollinated, self-incompatible perennial. American Journal of Botany 84:1588-1594.
Forsyth, S.A. 2003. Density-dependent seed set in the Haleakala silversword: evidence for an Allee effect. Oecologia 136:551–557.
Groom, M.J. 1998. Allee effects limit population viability of an annual plant. American Naturalist 151:487-496.
Hackney, E.E. and J.B. McGraw. 2001. Experimental Demonstration of an Allee Effect in American Ginseng. Conservation Biology 15:129-136.
Hultén, E. and M. Fries. 1986. Atlas of North European vascular plants: north of the Tropic of Cancer I-III. Koeltz Scientific Books, Königstein.
11 Knapp, E.E., M.A. Goedde, and K.J. Rice. 2001. Pollen-limited reproduction in blue oak: implications for wind pollination in fragmented populations. Oecologia 128:48-55.
Koening, W.D. and Ashley, M.V. 2003. Is pollen limited? The answer is blowin' in the wind. Trends in Ecology and Evolution 18: 157-159.
Kunin, W.E. 1997. Population size and density effects in pollination: pollinator foraging and plant reproductive success in experimental arrays of Brassica kaber. Journal of Ecology 85:225-234.
Nilsson, E. 2005. Breeding system evolution and pollination success in the wind- pollinated herb Plantago maritima. Doctoral thesis. Uppsala University, Sweden.
Nilsson, S.G., and U. Wästljung. 1987. Seed predation and cross-pollination in mast- seeding beech (Fagus sylvatica) patches. Ecology 68: 260–265.
Proctor, M., P. Yeo, and A. Lack. 1996. The natural history of pollination. Timber press, USA.
Sharma, N., P. Koul, and A.K. Koul, 1992. Reproductive biology of Plantago: Shift from cross- to self-pollination. Annals of Botany 69:7-11.
Taylor, C.M. and A. Hastings. 2005. Allee effects in biological invasions. Ecology Letters 8:895-908.
Tonsor, S.J. 1985. Leptokurtic pollen-flow, non-leptokurtic gene-flow in a wind pollinated herb. Plantago lancoelata L. Oecologia 67: 442-446.
Waites, A.R., and J. Ågren. 2004. Pollinator visitation, stigmatic pollen loads and among- population variation in seed set in Lythrum salicaria. Journal of Ecology 92:512-526.
Wilcock, C., and R. Neiland. 2002. Pollination failure in plants: why it happens and when it matters. Trends in Plant Science 7:270-277.
Young, K.A., J.S. Schmitt. 1995. Genetic variation and phenotypic plasticity of pollen release and capture height in Plantago Lanceolata. Functional Ecology 9:725-733.
12 13