The reproductive ecology of diploid and tetraploid galax urceolata.
Polyploids are organisms with more than two sets of chromosomes (Levin, 2002), and though polyploidy is relatively rare in most groups of animals (Otto and Whitton, 2000; Mable, 2004; but see Legatt and Iwama, 2003), it is extremely common in flowering plants. Indeed, studies suggest 70% or more of all angiosperms have experienced one or more rounds of genome duplication at some point during their evolutionary history (Grant, 1963; Stebbins, 1971; Goldblatt, 1980; Masterson, 1994; Otto and Whitton, 2000; Cui et al, 2006). Polyploids can differ from their diploid relatives in a variety of ways, including but not limited to cytology, gene activity, physiology, development, tolerance to pests and/or pathogens, stress tolerance, competitive ability, and reproductive strategy (Levin, 1983, 2002; Barker et al., 2016). Despite these differences, relatively few studies have compared ecological and evolutionary patterns and processes in natural populations of closely related polyploids and diploids. In particular relatively little is known about the consequences of genome duplication in awfopolyploids, despite growing evidence that they are common in nature (Soltis et al, 2007). Such studies are important because they help us understand the ecological and evolutionary consequences of polyploidy without confounding the effects of genome duplication with interspecific hybridization.
Polyploidy is often thought to lead to "instant speciation" due to reductions in the viability and fertility of intercytotype hybrids (Husband and Sahara, 2003; Bolnick and Fitzpatrick, 2007; Rieseberg and Willis, 2007; Wood et al., 2009; Ramsey, 2011). This reduction in intercytotype hybrid fitness, also reflected in the concept of minority cytotype disadvantage (Levin, 1975), is expected to select for increased prezygodc isolation. Such isolation could occur as a consequence of changes in flowering time (Lumaret et al., 1987; Thompson and Lumaret, 1992; Ramsey, 2011), increased levels of self-fertilization (Stebbins, 1950; Barringer, 2007) and/or changes in the activity and composition of pollinator communities (Thompson and Merg, 2008; Segraves and Anneberg, 2016). Alternatively, the ploidy change itself might alter any of these traits; e.g., flowering time (Ramsey, 2011) or self-incompatibility systems (Miller and Venable, 2000). Recent work has found prezygodc isolation is often larger than post-zygotic isolation, supporting these ideas (Husband et al., 2016). However, geographic separation is a substantial contributor to this isolation. We have less information on the reproductive ecology and its impact on the potential for gene exchange in natural geographically proximate populations of different cytotypes.
Galax is a natural autopolyploid series that includes diploid, triploid, and tetraploid individuals occurring in uniform- as well as mixed-cytotype populations (Baldwin, 1941; Nesom, 1983; Burton and Husband, 1999; Servick et al., 2015). Genome duplication has occurred numerous times in the species, and a recent study suggests 46 independent origins of tetraploid Galax, more than for any other polyploid (Servick et al., 2015). These characteristics bolster the species' reputation as "the classic example" of autopolyploidy (Baldwin, 1941; Stebbins, 1950; Grant, 1971; Soltis et al., 2007). However, despite this reputation, surprisingly little is known about the basic ecology of the species (but see Johnson et al., 2003) and virtually nothing is known about its reproductive biology. Indeed, the mating system of Galax has not been investigated, though levels of heterozygosity suggest that it is at least partially outcrossing (Servick et al., 2015). Nearly two-thirds of populations located in the most dense region of cytotype overlap contain multiple cytotypes (Burton and Husband, 1999). Therefore, an understanding of the species' reproductive ecology would be of value, as it would help us better understand why triploid and tetraploid lineages of Galax have arisen so frequendy, the overall lack of genetic differences among cytotypes (Servick et al., 2015) and whether we should expect levels of genetic differentiation among cytotypes to increase in the future.
We studied six populations (two diploid, two tetraploid, and two of mixed ploidy; i.e., containing both diploids and tetraploids) of Galax located in an area where both cytotypes are common. Our goals were to contribute to understanding of the basic reproductive biology and ecology of the species and to compare diploids and tetraploids with an eye toward assessing the potential for intercytotype gene movement. More specifically, we addressed four questions: (1) Do the flowering phenologies of diploids and tetraploids differ? (2) Do diploids and tetraploids differ in terms of the taxonomic compositions or abundances of their floral visitors? (3) Do levels of self-compatibility differ between diploids and tetraploids and does the species produce seed autonomously? (4) Do diploids and tetraploids differ in terms of pollen limitation and seed production?
MATERIALS AND METHODS
Study system.--Galax urceolata (Poiret) Brummitt (common names: beetleweed, wandflower; Diapensiaceae) is an herbaceous perennial endemic to the southern Appalachian Mountains (Baldwin, 1941; Brummitt, 1972; Nesom, 1983; Burton and Husband, 1999; Johnson et al., 2003). Isolated populations are also reported in Massachusetts, New York, Ohio, Kentucky, Illinois, Wisconsin, and British Columbia, Canada, though the species is generally considered "exodc" in those areas (USDA NRCS, 2008; D.M. Waller, pers. comm.). Galax is monotypic within its genus (Palser, 1963; Brummitt, 1972; Ronblom and Anderberg, 2002) and represents an autotetraploid series, with diploids (2x = 12), triploids (3x= 18) and tetraploids (4x = 24) occurring in uniform- as well as mixed-cytotype populations (Baldwin, 1941; Nesom, 1983; Burton and Husband, 1999).
Cytotype frequencies are spatially structured throughout the species' range, with diploids being most common in the north and tetraploids and rare triploids being most frequent in the south; however, a substantial region of geographic overlap of the cytotypes exists, particularly near the Blue Ridge escarpment (Nesom, 1983; Burton and Husband, 1999; Servick et al., 2015). At finer spatial scales, diploids and tetraploids are found in similar habitats (Servick et al., 2015), although some ecological differences between cytotypes do exist (Nesom, 1983; Johnson et al., 2003). In the region of overlap, almost two-thirds of populations contain multiple cytotypes (Burton and Husband, 1999). Populations occur in shady forested areas, and population sizes range widely, from fewer than 10 to many thousands of individuals (Baldwin, 1941; Nesom, 1983; Burton and Husband, 1999).
Plants are hermaphroditic with individuals reproducing asexually via stolons and sexually via small white flowers attached to long spike-like racemes (Baldwin, 1941; Nesom, 1983; Burton and Husband, 1999). Depending on population location, the species begins to flower in May and continues into July with the peak of flowering during the month of June (B. Barringer, pers. obs.). Whether Galax can reproduce via self-fertilization is not known; however, other members of the family are self-compatible (Ferrer and Good, 2012).
Sampling.--Six populations were used in this study, including two comprised solely of diploid individuals, two comprised solely of tetraploid individuals, and two of mixed-ploidy (i.e., including diploid and tetraploid individuals) (Table 1). For the analyses reported here the diploids and tetraploids from these mixed populations were treated as though they were from separate populations to permit comparing the performance of diploids and tetraploids. Populations with triploid individuals were not included in this study.
Assessing ploidy.--Ploidy was assessed and verified for at least 42 haphazardly-selected plants in each population using chromosome squashes (cf., Baldwin, 1941; Nesom, 1983). Because Galax can reproduce clonally via stolons, it is sometimes difficult to determine where one plant ends and another begins. To reduce the likelihood of sampling twice from the same individual within a study population, we focused on plants spatially separated from each other by >2 m (often >3 m). If the first 42 individuals sampled from a given population were found to be of a single cytotype (diploid or tetraploid), that population was assumed to be comprised solely of either diploid or tetraploid plants. For populations of mixed ploidy at least 21 individuals of each cytotype were identified. In all cases fresh young leaf tissue was harvested from each individual, fixed in a 3:1 ethanol:acetic acid solution, immersed in aceto-carmine stain, and boiled five times. Stained tissue was macerated and squashed on a microscope slide and chromosomes were visualized with a light microscope using oil-immersion (100X). Individuals were assessed as being either diploid (2n = 2x = 12) or tetraploid (2n = 4x = 24) and were used in one of the studies of Galax reproductive ecology detailed below.
Flowering phenology and number.--The flowering phenologies of six plants in each population were studied by counting the number of open flowers on each individual once/week throughout their flowering period (6 wk total; May 25 through July 11, 2009). In mixed-ploidy populations, three diploid and three tetraploid plants were used. Open flowers were defined as those whose petals were at least partially reflexed. Seasonal patterns of flower production for each plant were summarized by the "average flower time," representing the census week that the mean flower was produced. This was calculated by weighting each census week by the proportion of an individual's flowers that were open that week. Open flowers were also summed over the flowering period on each plant to produce an index of total flower production. Average flower time and total flower production were natural log-transformed and ANOVA was used to compare populations. Population was treated as a fixed effect because populations were selected to be diploid, tetraploid or mixed. A priori contrast was then used to compare diploid populations to tetraploid populations. Throughout, all means are reported with standard errors.
Floral visitor observations.--Once flowering began, and continuing through the majority of the flowering period (6 wk), floral visitor observations were conducted weekly in each study population by two observers during 1 h intervals at three different times of the day: morning (0800-0900), afternoon (1300-1400), and evening (1800-1900), for a total of 18 observation periods per population. During each 1 h time interval, each observer visually monitored four to six plants and recorded the identities of floral visitors (taxonomic order) as well as the frequencies and lengths of their visits in seconds. The observation periods were lumped for each population for analysis. Potential differences among populations and pollinator taxa in the number of visitors and duration of visits were evaluated using a two-way ANOVA with a priori contrasts to compare diploid populations to tetraploid populations. The distributions of both variables met assumptions of ANOVA.
Self-compatibility and autonomous self-fertilization.--To determine whether Galax is self-compatible and whether the degree of self-compatibility differs between cytotypes, we self-pollinated flowers on six plants in single cytotype populations and three diploid and three tetraploid plants in mixed cytotype populations. We enclosed a haphazardly-selected inflorescence on each plant before flowers opened using mesh bagging. Bagging was applied loosely to allow flowers to open, mature, and senesce naturally while excluding potential pollination vectors. The relatively small pore size (~150 [micro]m) should have eliminated the possibility of wind pollination via pollen originating on other plants (cf., Berry and Gorchov, 2004). When flowers on the enclosed inflorescence were open, we briefly removed the bagging and self-pollinated two flowers by hand with pollen obtained from other flowers on the same plant. Two to four anthers from each donor flower were used so that the stigmatic surfaces of recipient flowers were saturated with pollen. We then applied a small amount of nontoxic fabric paint to the pedicel of pollinated flowers to allow for the identification of resulting fruit and reapplied the bagging to prevent pollen contamination from other plants. Pollinated flowers were generally near the bottom and among the first to open on a given inflorescence. Hand-pollinated fruits were collected 8-9 wk later and their seeds were counted and recorded.
To estimate seed production as a product of autonomous self-fertilization and to determine whether this differs between cytotypes, we applied loose-fitting mesh bagging to an entire inflorescence on six plants in each population before flowers opened. In mixed-ploidy populations, three diploid and three tetraploid plants were used. At the end of the season (8-9 wk after the first flowers opened) the mesh bagging was removed, fruits were collected, and seed number per fruit was assessed.
Seed production from hand outcrosses and natural pollination.--To estimate the maximum number of seeds that could be produced per flower under outcross pollination conditions, six unopened flowers on each of six plants per population were emasculated while in bud. In mixed-ploidy populations, three diploid and three tetraploid plants were used. On a given plant, two buds were near the top, two near the center, and two near the bottom of a haphazardly-chosen inflorescence. Small mesh bags were then applied to individual flowers to eliminate the possibility of natural cross-pollination via insects or wind. Because Galax flowers are so small with very little space between flowers on an inflorescence, and because we did not know whether the species was self-compatible, we removed flowers directly adjacent to the emasculated flowers by cutting through their pedicels with a scalpel. When stigmas matured (generally within 2-3 d) we briefly removed the mesh bagging around a given flower and saturated the stigmatic surface with equal amounts of pollen obtained from two other individuals from the same population (two anthers per donor). Using mixedpollen loads helped to ensure seed production even if one of the donors shared a self-incompatibility allele with the pollen recipient. Pedicels were labelled with nontoxic fabric paint and the mesh bagging was put back into place. Hand-pollinated fruits were collected 8-9 wk later and their seeds were counted and recorded.
We estimated natural levels of seed production on six plants in each population at the end of the season by counting the number of fruits produced on a haphazardly-chosen inflorescence on each plant and the number of seeds in six fruits, two near the top, two near the center, and two near the bottom of the inflorescence. In mixed-ploidy populations, three diploid and three tetraploid plants were used. Seed production per inflorescence for a plant was estimated as the product of the mean number of seeds per fruit and the number of fruits per inflorescence. Aborted fruits were not counted.
Seed number data were analyzed using a three-way ANOVA with population, pollination type (hand outcrossed vs natural), and flower location (bottom, middle, and top of inflorescence) as fixed effects. Plant number was included as a random effect to account for multiple inflorescence locations sampled on each. Fruit production and total seed number (natural-log transformed) were analyzed using ANOVA with population as the independent variable. In both ANOVAs, a priori contrast was used to compare diploid populations to tetraploid populations.
Flowering phenology and number.--For almost all plants, regardless of population or ploidy, flowering began the first week of June and continued for 6 wk through midjuly (Fig. 1). Although the duration of flowering did not differ between cytotypes, tetraploids generally produced flowers earlier than diploids (peak flower production week for diploids 4.4 [+ or -] 0.07 and tetraploids 4.2 [+ or -] 0.04; [F.sub.1,28] = 5.17; P = 0.03). Differences between the cytotypes were greatest at the peak of flower production and the weeks leading up to it (Fig. 1). Tetraploids had 36% more open flowers than diploids over the season (tetraploids 331.1 [+ or -] 32.5 and diploids 243.2 [+ or -] 26.6; [F.sub.1,28] = 4.32, P < 0.05).
Floral visitor observations.--We observed 171 floral visitors representing two taxonomic orders: Hymenoptera and Diptera. The majority of Hymenoptera visitors were either bumblebees (Bombus spp.), honey bees (Apis mellifera), or sweat bees (Lasioglossum spp.), while the majority of Diptera visitors were either hover flies (Taxorrwrus geminatus) or tachinid flies (Juriniopsis spp.). Hymenoptera were the most common visitor (65% of visits) with an average of 1.67 [+ or -] 0.097/h and spent the most time interacting with flowers x= 11.9 [+ or -] 0.7 s) (Table 2). Diptera comprised 35% of visitors (0.89 [+ or -] 0.069/h), however their visits were relatively brief x= 5.9 [+ or -] 0.5 s). Diploids were visited 16% more often than tetraploids (diploid: 1.36 [+ or -] 0.10; tetraploid 1.19 [+ or -] 0.09), but there was no difference between cytotypes in terms of their visitor fauna or the duration of visits (Table 2).
Self-compatibility and autonomous self-fertilization.--Seventy of the 72 self-pollinated flowers did not produce seed and none of the flowers in bagged inflorescences set seed. Of the two self-pollinated flowers that did make seeds one was located near the bottom of an inflorescence on a diploid plant in a diploid population (six seeds); the other was located near the bottom of an inflorescence on a tetraploid plant in a mixed-cytotype population (10 seeds). These seed numbers are small for fruits located near the bottoms of inflorescences x (22 seeds). Due to low frequency of fruit production and small seed number, we suspect the seeds produced by these two flowers may have been a result of pollen contamination and not self-fertilization.
Seed production from hand outcrosses and natural pollination.--Hand outcrossed flowers produced 16% more seeds than naturally pollinated flowers (Table 3). Seed production also varied among flowers located on different regions of an inflorescence, with flowers located near the bottom producing 25% more seeds (21.6 [+ or -] 0.8) than flowers located near the middle (17.2 [+ or -] 0.7), which in turn produced almost twice as many seeds as flowers located near the top (8.9 [+ or -] 0.6; Table 3). There were no differences in seed number per fruit between diploids and tetraploids and the increase in seed number with cross pollination and toward the base of the inflorescence was consistent across diploids and tetraploids (Table 3).
There were no differences in plant-level measures of performance between diploids and tetraploids. The mean number of fruits per inflorescence (F]i28 = 0.04, P = 0.85) and total seed production per inflorescence ([F.sub.1,28] = 0.07, P = 0.79) did not differ between cytotypes.
Overall there were few differences between diploid and tetraploid Galax urceolata in terms of reproductive ecology. For example, although the peak of flower production for tetraploids did precede that of diploids, the overall distributions of flowering times were similar, resulting in substantial overlap between cytotypes in terms of floral phenology. In addition, although diploids were visited more frequendy by pollinators than tetraploids, we found no evidence for differences between cytotypes in terms of the taxonomic composition of their pollinator communities. Moreover, both diploid and tetraploid Galax appear to be self-incompatible. Finally, fruit production is comparable across cytotypes, with both having increased seed production in hand-outcrossed flowers suggesting pollen limitation. In total, while there are significant differences in some aspects of the reproductive ecology between diploid and tetraploid Galax, the overall pattern is one of similarity.
Our study indicates Galax is self-incompatible and insect pollinated. Hymenoptera were twice as common as Diptera and interacted with flowers for twice as long. Therefore, it is likely that Hymenoptera--most notably a combination of bumble bees, honey bees and sweat bees--are the most important group of pollinators for Galax, facilitating the majority of the species' pollen transfer. Plants set no fruit by autonomous pollination and virtually no fruit by hand self-pollination, indicating that Galax is self-incompatible and relies on insects for between genet pollen movement to produce seeds. Self-incompatibility in both diploid and tetraploid Galax is somewhat unexpected in that polyploidy is often associated with a breakdown in self-incompatibility systems (cf., Miller and Venable, 2000; Robertson et al., 2010; but see Mable, 2004) and polyploid angiosperms often exhibit higher levels of self-fertilization than their diploid relatives (Barringer, 2007).
Our data suggest diploid and tetraploid Galax are both pollen limited. Natural levels of seed production ranged widely, though this variation was not related to cytotype. Hand-pollinated flowers produced on average 16% more seed than open-pollinated flowers, suggesting pollen limitation reduces seed production in natural populations of Galax (whether in terms of total pollen or nonself pollen received). However, the degree of pollen limitation did not depend on cytotype, and there were no differences between cytotypes in terms of the average numbers of seeds per fruit or the numbers of fruits per inflorescence. Interestingly, the occurrence of pollen limitation in Galax suggests pollen competition for access to ovules is limited, which in turn could facilitate the success of between-cytotype seed production even if intercytotype fertilization is infrequent or inefficient.
Intercytotype crosses are expected to produce offspring with low fitness. For example diploid-tetraploid crosses often produce sterile triploids (Levin, 1975; Ramsey and Schemske, 1998; Sutherland and Galloway, 2016). Therefore, in populations in which both diploid and tetraploid cytotypes occur, or in areas in which diploid and tetraploid populations are in relatively close proximity, selection to reduce the frequency of intercytotype crosses may result in character displacement for traits that influence the likelihood such crosses occur, i.e., prezygotic isolation. Our study found few differences between populations of diploid and tetraploid Galax in terms of their reproductive ecology, a finding that can simultaneously help to explain the lack of genetic divergence between cytotypes and the relatively frequent formation of triploid lineages (at least 31 times) and tetraploid lineages (at least 46 times) within the species (Servick et al., 2015). Indeed, the prevalence of triploids and the recurrent formation of both triploid and tetraploid lineages suggests tetraploidy in Galax might commonly arise via a so-called triploid bridge (Harlan and de Wet, 1975; Ramsey and Schemske, 1998), as this process would be facilitated by overlapping flowering phenologies, shared pollinators, and self-incompatibility. The performance of intercytotype crosses has not been investigated and would illuminate this possibility.
Acknowledgments.--The authors thank The Highlands Biological Station and the William Chambers Coker Fellowship in Botanical Research for their support. Thanks also to Brett Jones for assistance in the field.
BALDWIN, J. T. 1941. Galax, the genus and its chromosomes. J. Hered., 32:249-254.
BARKER, M. S., B. C. HUSBAND, AND J. C. PIRES. 2016. Spreading Winge and flying high: the evolutionary importance of polyploidy after a century of study. Am. J. Bot., 103:1139-1145.
BARRINGER, B. C. 2007. Polyploidy and self-fertilization in flowering plants. Am. J. Bot., 94:1527-1533.
BERRY, E.J. AND D. L. GORCHOV. 2004. Reproductive biology of the dioecious understory palm Chamaedorea radicalis in a Mexican cloud forest: pollination vector, flowering phenology and female fecundity .J. Trap. Ecol., 20:369-376.
BOLNICK, D. I. AND B. FITZPATRICK. 2007. Sympatric speciation: models and empirical evidence. Annu. Rev. Ecol. Evol. S" 38:459-487.
BRUMMITT, R. K. 1972. Nomenclatural and historical considerations concerning the genus Galax. Taxon, 21:303-317.
BURTON, T. L. AND B. C. HUSBAND. 1999. Population cytotype structure in the polyploid Galax urceolata (Diapensiaceae). Heredity, 82:381-390.
CUI, L., P. K. WALL, J. H. LEF.BENS-MACK, B. G. LINDSAY, D. E. SOLTIS, J. J. DOYLE, P. S. SOLTIS, J. E. CARLSON, K. ARUMUGANATHAN, A. BARAKAT, V. A. ALBERT, H. MA, AND C. W. DEPAMPHILIS. 2006. Widespread genome duplications throughout the history of flowering plants. Genome Res., 16:738-749.
FERRER, M. M. AND S. V. GOOD. 2012. Self-sterility in flowering plants: preventing self-fertilization increases family diversification rates. Ann. Bot., 110:1079-1081.
GOLDBLATT, P. 1980. Polyploidy in angiosperms. In W. H. Lewis (ed.), Polyploidy--Biological Relevance. Plenum, New York, NY.
GRANT, V. 1963. The Origin of Adaptations. Columbia University Press, New York, NY.
--. 1971. Plant Speciation. Columbia University Press, New York, NY.
HARLAN, J. R. AND J. M.J. DE WET. 1975. On O. Winge and a prayer: the origins of polyploidy. Bot. Rev., 41:361-390.
HUSBAND, B. C., S. J. BALDWIN, AND H. A. SABARA. 2016. Direct vs. indirect effects of whole-genome duplication on prezygotic isolation in Chamerion angustifolium: implications for rapid speciation. Am. J. Bot., 103:1259-1271.
--AND H. A. SABARA. 2003. Reproductive isolation between autotetraploids and their diploid progenitors in fireweed, Chamerion angustifolium (Onagraceae). New Phytol., 161:703-713.
JOHNSON, M. T. J., B. C. HUSBAND, AND T. L. BURTON. 2003. Habitat differentiation between diploid and tetraploid Galax urceolata (Diapensiaceae). Int. J. Plant Sci., 164:703-710.
LEGATT, R. A. AND G. K. IWAMA. 2003. Occurance of polyploidy in the fishes. Rev. Fish Biol. Fisher., 13:237-246.
LEVIN, D. A. 1975. Minority cytotype exclusion in local plant populations. Taxon, 24:35-43.
--. 1983. Polyploidy and novelty in flowering plants. Am. Nat., 122:1-25.
--. 2002. The Role of Chromosomal Change in Plant Evolution. Oxford University Press, New York, NY.
LUMARET, R.,J. L. GUILLERM, J. DELAY, A. AIT LHAJ LOUTFI, J. IZCO, AND M. JAY. 1987. Polyploidy and habitat differentiation in Dactylis glomerata L. from Galicia (Spain). Oecologia, 73:436-446.
MABLE, B. K. 2004. Polyploidy and self-compatibility: is there an association? New Phytol., 162:803-811.
MASTERSON, J. 1994. Stomatal size in fossil plants: evidence for polyploidy in majority of flowering plants. Science, 264:421-424.
MILLER, J. S. AND D. L. VENABLE. 2000. Polyploidy and the evolution of gender dimorphism in plants. Science, 289:2335-2338.
NESOM, G. L. 1983. Galax (Diapensiaceae): geographic variation in chromosome number. Syst. Bot., 8:1-14.
OTTO, S. P. AND J. WHITTON. 2000. Polyploid incidence and evolution. Annu. Rev. Genet., 34:401-437.
PALSER, B. F. 1963. Studies of floral morphology in the Ericales VI. The Diapensiaceae. Bot. Gaz., 124:200-219.
RAMSEY, J. 2011. Polyploidy and ecological adaptation in wild yarrow. P. Natl. Acad. Sci. U.S.A., 108:70967101.
--AND D. W. SCHEMSKE. 1998. Pathways, mechanisms, and rates of polyploid formation in flowering plants. Annu. Ren. Ecol. Syst., 29:467-501.
RIESEBERG, L. H. and J. H. Willis. 2007. Plant speciation. Science, 317:910-914.
ROBERTSON, K, E. E. GOLDBERG, AND B. IGIC. 2010. Comparative evidence for the correlated evolution of polyploidy and self-compatibility in Solanaceae. Evolution, 65:139-155.
RONBLOM, K AND A. A. ANDERBERG. 2002. Phylogeny of Diapensiaceae based on molecular data and morphology. Syst. Bot., 27:383-395.
SEGRAVES, K A. AND T.J. ANNEBERG. 2016. Species interactions and plant polyploidy. Am. J. Bot., 103:13261335.
SERVICE, S., C.J. VISGER, M. A. GITZENDANNER, P. S. SOLTIS, AND D. E. SOLTIS. 2015. Population genetic variation, geographic structure, and multiple origins of autotetraploidy in Galax urceolata. Am. J. Bot., 102:973-982.
SOLTIS, D. E., SOLTIS, P. S., SCHEMSKE, D. W., HANCOCK, J. F, THOMPSON, J. N., HUSBAND, B. C., AND JUDD, W. S. 2007. Autopolyploidy in angiosperms: have we grossly underestimated the number of species? Taxon, 56:13-30.
STEBBINS, G. L. 1950. Variation and Evolution in Plants. Columbia University Press, New York, NY.
--. 1971. Chromosomal Evolution in Higher Plants. Addison-Wesley, London, U.K.
SUTHERLAND, B. L. AND L. F. GALLOWAY. 2016. Postzygotic isolation varies by ploidy level within a polyploid complex. New Phytologist doi:10.1111/nph.l4116.
THOMPSON, J. N. AND K F. MERG. 2008. Evolution of polyploidy and the diversification of plant-pollinator interactions. Ecology, 89:2197-2206.
USDA, NRCS. 2008. The PLANTS Database (http://plants.usda.gov, 22 January 2008). National Plant Data Center, Baton Rouge, LA 70874-4490 U.S.A.
WOOD, T. E., N. TAKEBAYASHI, M. S. BARKER, I. MAYROSE, P. B. GREENSPOON, AND L. H. RIESEBERG. 2009. The frequency of polyploid speciation in vascular plants. P. Natl. Acad. Sci., 106:13875-13879.
SUBMITTED 13 SEPTEMBER 2016
ACCEPTED 4 NOVEMBER 2016
Brian C. Barringer (1)Department of Biology, University of Wisconsin-Stevens Point, 800 Reserve Street, Stevens Point 54481
Laura F. Galloway
Department of Biology, University of Virginia, P.O. Box 400328, Charlottesville 22904
(1) e-mail: email@example.com
Caption: FIG. 1.--Mean ([+ or -] 1 se) number of open flowers per plant for diploids and tetraploids over the 6 wk study period
TABLE 1.--Ploidy (diploid, tetraploid, or mixed ploidy) and locations (all in Macon County, NC) of Galax urceolata populations used in this study Population Ploidy Location Lat./Long. ID (DMM) D1 Diploid Rich Gap Rd., 1.8 miles 35 01.287 east/northeast of Highway 28. -83 10.374 D2 Diploid Big Bear Pin Rd., 0.7 miles 35 03.678 north of Chestnut St. -83 11.272 T1 Tetraploid Clear Creek Rd., 1.4 miles 35 01.452 west/southwest of Highway 28. -83 12.436 T2 Tetraploid Forest Service Rd., 3.0 miles 35 01.700 west of Clear Creek Rd. -83 13.754 Ml Mixed Cliffside Lake Rd., 0.6 miles 35 04.692 east/northeast of Highway 64. -83 14.591 M2 Mixed Glen Falls Rd., 1.1 miles 35 02.108 south of Dillard Rd. -83 14.067 TABLE 2.--Analysis of variance exploring whether the number and duration of floral visits varies among visitor taxa (Hymenoptera and Diptera) or population. Diploid and tetraploid populations were compared with an a priori contrast (ploidy). F-values are given. Source df Number Duration of visits of visits Taxa 1 35.04 *** 30.13 *** Population 7 1.29 1.06 Ploidy 1 4.11 * 1.05 Taxa * Pop 7 0.86 1.73 Error df 116 111 * P < 0.05, *** P < 0.001 TABLE 3.--Analysis of variance evaluating if seed number per fruit depends on whether flowers were hand outcrossed or naturally pollinated (pollination type), their location on the inflorescence (bottom, middle, top), or the population. Diploid and tetraploid populations were compared with an a priori contrast (ploidy). Plant number was included to account for multiple samples per inflorescence and treated as a random factor. F-values are reported for fixed effect and Z scores for random effects Source df F/Z P Pollination type 1 7.39 0.008 Location 2 82.42 <0.001 Population 7 1.45 0.20 Ploidy 1 1.58 0.21 Pollination type * location 2 0.05 0.95 Pop * location 14 0.41 0.97 Pop * pollination type 7 0.65 0.72 Pop*poll type * location 14 0.60 0.86 Plant -- 0.55 0.29 Error 168
|Printer friendly Cite/link Email Feedback|
|Author:||Barringer, Brian C.; Galloway, Laura F.|
|Publication:||The American Midland Naturalist|
|Date:||Apr 1, 2017|
|Previous Article:||Pollination and reproductive biology in a hill prairie population of Nothocalais cuspidata (Asteraceae: cichorieae).|
|Next Article:||Conserving large oaks and recruitment potential while restoring Midwestern Savanna and Woodland.|