| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
School of Biological Sciences, University of Nebraska, Lincoln Nebraska 68588-0118
Received for publication August 5, 1997. Accepted for publication November 24, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Andropogon gerardii • cytotype • enneaploid • hexaploid • Poaceae • polyploidy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Lewis, 1980
; Lumaret 1988a
; Thompson and Lumaret, 1992
; Keeler, 1997
). A number of authors have pointed out the difficulties in establishing a polyploid mutant in a diploid, or lower polyploid, population (Stebbins, 1971
; Grant, 1981
; Levin, 1983
; Fowler and Levin, 1984; Felber, 1991
; Bever and Felber, 1993
; Rodríguez, 1996a,
b
). At the same time, it is obvious that this step must have repeatedly occurred to produce existing patterns of polyploidy and intrapopulation polyploidy.
Autopolyploids, plants with a shared genome, but varying in the number of copies of that genome, are also relatively common (Lewis, 1980
; Jackson, 1982
; Lumaret, 1988a,
b
; Thompson and Lumaret, 1992
; Keeler, 1997
). They are believed to result from nonreduction of the genome during reproduction (Stebbins, 1947
; Jackson, 1976
, 1982
; Bretagnolle and Thompson, 1995
).
Levin (1975,
1983
) and Fowler and Levin (1984)
showed that a new autopolyploid will be at a significant disadvantage as a result of frequency-dependent minority status, and, even if it has the same fitness as the previous cytotype, is likely to be lost. Survival of the polyploid is enhanced by selective (competitive) superiority (Levin, 1983
; Fowler and Levin, 1984
; Felber, 1991
; Rodríguez, 1996a,
b
). New polyploid establishment is also assisted if genetic drift renders the new polyploid locally dominant, reversing the rare cytotype disadvantages (Levin 1983
; Fowler and Levin, 1984
; Felber, 1991
; Rodríguez, 1996b
) or if the species is iteroparous or autogamous (Rodríguez, 1996b
).
Models of establishment of polyploids resulting from selection (Levin, 1983
; Fowler and Levin, 1984
; Rodríguez, 1996a,
b
) appear to have assumed directional selection favoring the higher polyploid. The widespread occurrence of populations containing several polyploid cytotypes (see reviews by Lewis, 1980
; Keeler, 1997
) suggests intraspecific polyploid variation may be stable rather than transitory. Much work is needed on polyploid population genetics generally (Thompson and Lumaret, 1992
; Bever and Felber, 1993
) and on the dynamics of autopolyploidy within a population in particular (e.g., Felber, 1991
). Lacking theoretical analysis of the conditions for stable intraspecific polyploid variation, in this paper we consider the possibility that balanced rather than directional selection may operate in polyploid complexes. We present data that compares the fitness of cytotypes of a grass species with widespread intraspecific polyploidy to test the hypothesis that selection is favoring replacement of the lower polyploids by the higher. The simplest alternative model was that higher polyploids represent unfit mutants and are rapidly eliminated. As the data below indicate, we find the need for alternate models that would address the maintenance of intraspecific autopolyploidy.
Andropogon gerardii Vitman (big bluestem) is one of the dominant grasses of the tallgrass prairie, comprising up to 80% of the biomass on favorable sites (Weaver, 1954
). Populations across its range contain two cytotypes, hexaploid (2n = 6x = 60) and enneaploid (2n = 9x = 90) (Gould, 1965
, 1967
; Norrmann, Quarín, and Keeler, 1997
). In Nebraska, Kansas, and Colorado prairies, these two cytotypes are often in nearly equal proportions (Keeler et al., 1987
; Keeler, 1990,
1992
, and unpublished data). In these populations, other cytotypes are rare (Keeler, 1992
, and unpublished data). Meiosis in the hexaploid is regular and seed set can be high; meiosis in the enneaploid is irregular with a high frequency of gametic loss resulting from unbalanced genomes (Norrmann, Quarín, and Keeler, 1997
). Both cytotypes are obligately outbreeding and when crossed, produce viable offspring of intermediate ploidy level, which vary in fertility and vigor (Norrmann, Quarín, and Keeler, 1997
; Norrmann and Keeler, unpublished data).
The two cytotypes cannot readily be distinguished in the field (Keeler et al., 1987
; Norrmann, Quarín, and Keeler, 1997
), as is often the case for intraspecific polyploids (Lumaret, 1988a
; Keeler, 1997
). This paper reports the results of detailed comparisons looking for differences between the cytotypes that might produce a selective advantage for one or the other of a sort that would result in directional selection. Such a comparison is critical for understanding whether the one polyploid cytotype can be expected to replace the other or whether the intraspecific variation in polyploidy is being maintained within the population.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Norrmann, Quarín, and Keeler, 1997
). Plants compared in this study were collected as rhizomes from sites where both hexaploids and enneaploids were abundant, i.e., western prairies (see Keeler, 1990
). Where native populations have been analyzed in detail, cytotypes other than hexaploid and enneaploid occur with a frequency under 10% (Keeler, 1992
; Norrmann, Quarín, and Keeler, 1997
; Keeler, unpublished data). This paper focuses on the two common cytotypes of mixed populations of A. gerardii (hexaploids and enneaploids) because we had too few observations of other cytotypes to analyze. These intermediate ploidy levels are the subject of ongoing work.
The plants originated at Konza Prairie Research Natural Area, Manhattan, Kansas; Pawnee Prairie, Beatrice. Nebraska; Nine-Mile Prairie, Lincoln, Nebraska (sites described in Keeler, 1990
) and the City of Boulder Mountain Parks and Open Space, Boulder, Colorado (descriptions in Bock, Bennett, and Bock, 1995
; Bock et al., 1995
). All four prairies have a similar pattern: overall a slight majority of hexaploids with local areas varying from 0 to 80% enneaploids (Keeler, 1990,
1992
, unpublished report to City of Boulder Open Space, 1996).
Common gardens were established in 1986–1987 on the University of Nebraska East Campus (designated Lincoln garden, LG) and at the University of Nebraska Agricultural Station in Mead, Nebraska (MG), using, in approximately equal numbers, plants taken as rhizomes from Konza, Pawnee, and Nine-Mile Prairies. Plants had been growing in 10-cm (four-inch) pots in the greenhouse for more than a year. Hexaploids and enneaploids were alternated in a grid. Water was provided to get the plants established, but after that conditions were not managed.
At Konza Prairie Research Natural Area (Konza hereafter) individual plants were marked and cytotyped in 1988–1989, establishing 18 plots of 24–50 marked plants. Naturally occurring shoots were chosen to be at least 1 m apart. These were unlikely to be the same clone (Keeler, 1992
). In Boulder, Colorado (Boulder), nine 10 x10 m plots were established in 1995 and 1996 on Boulder Mountain Parks and Open Space land. All A. gerardii in the plot were mapped and more than half of the 613 clumps (plants) have been cytotyped to date.
Height was measured on the tallest flowering stalks (Table 1) and on the canopy height (Table 2). Tables 1 and 2 contain independently collected data.
|
|
Leaf length was measured on the longest leaf available (some potentially larger leaves were too damaged to measure). Leaf width was taken at the widest part of the widest leaf available. Biomass was determined by clipping the entire canopy at ground level after it had gone dormant in the fall and drying it 2–3 d in a plant drier at 60°C to remove any remaining moisture. Leaves and inflorescences were separately weighed to the nearest 0.1 g.
To understand the influence of the environment between cytotypes plant height was compared under natural conditions in four different sites. The sites were Osage Prairie, Tecumseh, Nebraska; Nine-Mile Prairie, Lincoln, Nebraska; Reller Natural History Study Area, 30 km southwest of Lincoln, Nebraska; and Konza. Konza and Nine-Mile Prairies are native prairies, Reller was replanted to native grass mix about 1970 (R. C. Lommasson, University of Nebraska-Lincoln, personal communication) and Osage includes both native and replanted portions (Nebraska Game and Parks Commission, unpublished report). At each site, areas burned the previous May were compared to areas not burned in the previous year. Hill and low areas are relative: tops of the low hills vs. the base of the hills.
Plant sizes for LG plants were determined by photographing the basal area of the plants at the end of the growing season, projecting the slide onto paper, and mapping and calculating the area with a planimeter. Total area for Boulder plants was determined from the area on to scale maps digitized in ArcView (ESRI, 1997
).
For seed mass, all caryopses on a marked plant were gathered at the end of the growing season, stored at room temperature, and weighed to the nearest 0.01 gm. For the Boulder plants, good seed was determined based on individually removing the seed coat to see whether a filled seed was present in a sample of 100 caryopses (or all the caryopses if under 100). For large reproductive masses, the percentage good seed was calculated by multiplying the frequency of filled seeds in the sample of 100 caryopses by the total number of caryopses actually counted from the plant. Caryopses and seeds discussed in this paper are from sessile spikelets. While seeds from pedicellate spikelets are reported (Boe, Ross, and Wynia, 1983
; Springer, 1991
) seed production by pedicellate spikelets was rare to nonexistent in our data (see Discussion).
Seed viability was determined by planting standard masses or counts of caryopses gathered from the plants of known cytotype at LG and MG, in individual vermiculite-filled 10-cm (four-inch) clay pots in the greenhouse. The number of seedlings emerging were recorded and compared as percentage germinating using ANOVA and a repeated-measures ANOVA. Plant (genotype) was treated as a random variable.
Total annual seed production by a cytotype was estimated by summing the total good seed production for a plot by cytotype.
Statistical significance was determined with Student's t test where a normal distribution was evident and using Mann-Whitney U test where normality could not be assumed, especially in flowering-related characters, which had a high frequency of zeros. The impact of a variety of variables on height was tested using mixed-model analysis of variance. Repeated-measures statistics were used for data sets involving the same plants sampled in different years. Programs used were Statview (1994)
and SAS (1996)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In common gardens, total biomass produced in a season was compared and found to be greater in enneaploids. This was significantly related to time in the Lincoln common garden: enneaploids increased more than that of hexaploids between 1987 and 1988 (the first year after transplanting into the garden) so that subsequent differences are a result of a difference established by 1988 (Fig. 1). At MG, hexaploids transplanted in 1989 averaged 145.9 and 572.2 g dry mass in 1990 and 1992, respectively (N = 66, 64), and enneaploids averaged 171.4 g dry mass and 845.4 (N = 72, 71), respectively. The former is not significantly different, but the latter is (P < 0.001, Student's t test). Total biomass production was thus variable and greater in enneaploids, sometimes significantly.
|
Clone areas were not significantly different in transplants in LG after 3 yr (hexaploid mean area ± 1 SD = 1001.8 ± 633 cm2, enneaploid mean ± 1 SD = 1662.8 ± 1289 cm2, N = 23, t test). In MG after 7 and 8 yr, area was significantly greater, mean ± 1 SD for hexaploids = 208.7 ± 2.4 cm2, mean for enneaploids = 279.0 ± 15.8 cm2, N = 260, P < 0.001, ANOVA). In native prairie in Boulder, the mean area of enneaploids was also larger (hexaploid mean ± 1 SD = 1584.7 ± 3547.2 cm2, enneaploid = 4072.6 ± 5661.3 cm2, N = 108, P < 0.01, Mann Whitney U test).
Total reproductive effort per plant, measured as mass, was significantly greater for enneaploids (Table 3). Good seed, per plant or per square centimetre, was not statistically significantly different. In Boulder in 1995, for hexaploids X ± 1 SD = 18.04 ± 89.3 seeds/plant , enneaploids X ± 1 SD = 38.25 ± 109.14 seeds/plant, df =107, and in 1996 X ± 1 SD for hexaploids = 88.39 ± 427.2 seeds/plant and X ± 1 SD for enneaploids = 144.10 ± 320.7 seeds/plant, df = 86. Recalculated on a per square centimetre basis the results are not statistically significant either: a mean of 0.018 seeds/cm2 and 0.145 seeds/cm2 for hexaploids in 1995 and 1996, respectively, and 0.013 seeds/cm2 and 0.037 seeds/cm2, respectively, for enneaploids. For the plants at LG, repeated-measures analysis indicates a significantly different effect of time on the plants: change in seed production of the two cytotypes was significantly different in 1987–1988 and 1988–1989 but not 1989–1990 (Fig. 2).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), those are uncommon in the field (e.g., Keeler, 1992
, and unpublished data). This study looks at fitnesses of the two cytotypes that make up the bulk of the plants in natural populations of A. gerardii, a necessary but not sufficient step for understanding this polyploid complex.
The hexaploid and enneaploid cytotypes differed in most of the characters measured. Because they produce unequal frequencies of good seed (Norrmann, Quarín, and Keeler, 1997
), the initial expectation was that the enneaploid cytotype was less fit than the hexaploid cytotype. That being the case, enneaploid establishment seemed improbable. However, the widespread existence of enneaploids (Keeler et al., 1987
; Keeler, 1990
, 1992
, and unpublished data) either demonstrates that enneaploid establishment happens or represents a mutation: selection balance with a long time frame. By the time the enneaploid has persisted for more than a decade or reached a metre in diameter, it is reasonable to describe it as "established" and consider the intrapopulation dynamics and consequences resulting from intrapopulational cytotypic variation. Where there are data on plant size in natural populations (Boulder), enneaploids are demonstrated to be big, productive plants.
Enneaploid A. gerardii produce markedly poorer quality seeds than hexaploids (Norrmann, Quarín and Keeler, 1997
; Table 5), but they are bigger plants (Fig. 1, Results), producing more seeds overall (Table 3) and so, on a per plant basis, more good seeds (see Results). The outcome is similar contributions of the two cytotypes to good seed production at a site (Table 5).
Variation is striking (e.g., for flowering stalk height, Table 1). Because there is dramatic between-year variation at a site (Table 1), we attribute this to the plasticity of the species in response to the environment. This is supported by Table 2 and studies of response of A. gerardii to fire (e.g., Collins and Wallace, 1987
).
Boe, Ross, and Wynia (1983)
and Springer (1991)
compare the fertility of seeds from pedicellate and sessile spikelets. They find both to have 70–80% germinable pedicellate and sessile spikelet seeds/caryopsis. This is in strong contrast to a maximum of 28% germinable seeds/caryopsis found here (Table 4). We believe that the difference between their results and ours is that we are working in populations of mixed cytotypes. The common garden planting scheme alternated cytotypes, and intermingling occurs naturally in the native prairies studied. Boe, Ross, and Wynia (1983)
used wild-collected seed, which, given the poor fertility of enneaploids, selects strongly for seed quality and therefore for hexaploidy (Norrmann, Quarín, and Keeler 1997
; see also Table 4). Springer (1991)
used USDA-Soil Conservation Service seed (PI 483446). Since USDA plant breeders generally gather seed from wild populations and propagate plants via seed, those populations too are generally hexaploid, with better seed set. In contrast, mixed populations with high frequencies of aborted pollen and ovules would be expected to have much lower frequencies of good seed. Very low seed production is observed in wild populations (e.g., Masters et al., 1993
; L. Reidel, City of Boulder Open Space, personal communication).
The observations of fertile pedicellate spikelets of Boe, Ross, and Wynia (1983)
and Springer (1991)
also do not match ours. We have only very rarely observed fertile pedicellate spikelets. Filled seeds from pedicellate spikelets were included in total seed counts when observed, but they were rare. At present, we can only ascribe this to the highly significant effect of sites on fertile pedicellate spikelet seed yields found by Boe, Ross, and Wynia (1991)
.
Effects resulting from cytotype are always confounded by genotypic effects, because a particular genotype always has the same cytotype. Where we have looked at genotype, it is a more important determinant of response in A. gerardii than cytotype (Davis, unpublished data) as might be expected. Within a site, significant effects of cytotype may result from significant effects of genotype. In this study, that was controlled for by using samples of both cytotypes combined in a garden from a series of sites or replicate plots within prairies. There is no reason to expect replicate plots or several prairies to differ in a systematic way based on genotype: the emergent effects can thus be attributed to cytotype. For any particular trait, we do not doubt that genotype is the most important variable, but there are consistent effects of cytotype as well.
The absence of the aneuploids that result from crossing hexaploid and enneaploid A. gerardii remains a problematic factor for understanding this system. Since enneaploids would have lower fitness if they crossed with each other rather than with hexaploids (Norrmann, Quarín, and Keeler, 1997
), it seems unlikely that the populations are reproductively isolated. The most likely situation is that strong selection is removing most hybrid seedlings and the realized fitness of enneaploids—number of reproducing progeny—is lower than the data on seed production and germination suggest. Ongoing research is addressing the absence of aneuploids.
Thus, while alternative hypotheses cannot be eliminated at this time, from the data gathered to date, the best explanation is that offsetting selection in A. gerardii appears to be maintaining the two cytotypes, with the hexaploid having more efficient seed set and the enneaploid greater vegetative vigor and so greater total reproduction. Consequently, analysis of selective mechanisms that might be responsible seems appropriate. Because existing theory addresses establishment and appears to assume replacement not polymorphism, we can find little applicable theory.
The distribution of cytotypes in western prairies as compared to eastern prairies (Keeler, 1990
, 1992
) suggests important regional differences, probably in relative cytotype fitness. However, in the populations of mixed cytotypes, maintenance by balanced selection seems the most likely at this time. In other taxa, habitat and flowering time differ among Claytonia virginica cytotypes (Lewis, Suda, and MacBryde, 1967
; Lewis and Suda, 1976
), Dactylis glomerata diploids and tetraploids showed slight habitat preferences (Lumaret et al., 1987
; Lumaret and Barrientos, 1990
), and cytotype is one of the determinants of variation in Panicum virgatum (McMillan and Weiler, 1959
; Porter, 1966
; Hultquist, Vogel, and Kaeppler, 1997
) and Paspalum (Norrmann, Quarín, and Burson, 1987). Thus A. gerardii is one of an array of autopolyploid complexes with significant differences between co-occurring cytotypes. An obvious conclusion is that balanced polymorphism may be adaptive because it can extend the range of tolerance of the species. Intraspecific polyploidy, as a form of genetic variation that may expand environmental tolerance in plant species, requires serious future theoretical and field analysis.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bock, J. H., B. C. Bennett, and C. E. Bock. 1995 A comparison of relictual tallgrass prairie in Colorado with true prairie from the eastern Great Plains. Ninth Annual Meeting of the Society for Conservation Biology, p. 31 (Abstract).
———, T. Seastedt, Y. B. Linhart, and C. E. Bock. 1995 Relictual tallgrass prairie in the western Great Plains of North America. Proceedings of the VI International Congress of Ecology, p. 105 (Abstract).
Boe, A., J. G. Ross, and R. Wynia. 1983 Pedicellate spikelet fertility in big bluestem from eastern South Dakota. Journal of Range Management 36: 131–133.[CrossRef][ISI]
Bretagnolle, F., and J. D. Thompson. 1995 Gametes with the somatic chromosome number: mechanisms for their formation and role in the evolution of autopolyploid plants. New Phytologist 129: 1–22.[CrossRef][ISI]
Collins, S. L., and L. L. Wallace, [eds.]. 1987 Fire in North American tallgrass prairie. University of Oklahoma Press, Norman, OK.
ESRI (Environmental Systems Research Institute). 1997 ArcView 3.0. ESRI Inc., St. Louis, MO.
Felber, F. 1991 Establishment of a tetraploid cytotype in a diploid population: effect of relative fitness of the cytotypes. Journal of Evolutionary Biology. 4: 195–207.
Fowler, N. L., and D. A. Levin. 1984 Ecological constraints on the establishment of a novel polyploid in competition with its diploid progenitor. American Naturalist.124: 703–711.
Gould, F. W. 1965 Variability in big bluestem A. gerardi Vitman in DeKalb and Ogle Counties, Illinois. Southwestern Naturalist 10: 218–220.[CrossRef]
———. 1967 The genus Andropogon in the United States. Brittonia 19: 70–76.
Grant, V. 1981 Plant speciation, 2d ed. Columbia University Press, New York, NY.
Hultquist, S. J., K. P. Vogel, and S. Kaeppler. 1997 DNA content and chloroplast DNA polymorphisms among switchgrasses from remnant midwestern prairies. Crop Science 37: 595–597.
Jackson, R. C. 1976 Evolutionary and systematic significance of polyploidy. Annual Review of Ecology and Systematics 7: 209–234.
———. 1982 Polyploidy and diploidy: new perspectives on chromosome pairing and its evolutionary implications. American Journal of Botany 69: 1512–1523.[CrossRef][ISI]
Jones, K. 1958 Cytotaxonomic studies in Holcus. I. The chromosome complex of Holcus mollis L. New Phytologist 57: 191–210.
Keeler, K. H. 1990 Distribution of polyploid polymorphism in big bluestem, A. gerardii in the tallgrass prairie region. Genome 33: 95–100.
———. 1992 Local polyploid variation in the native prairie grass A. gerardii. American Journal of Botany 79: 1229–1232.
———. 1997 Population biology of polyploidy in species populations. In G. Cheplick [ed.], Population biology of grasses, 183–207. Cambridge University Press, Cambridge.
———, B. Kwankin, P. W. Barnes, and D. W. Galbraith. 1987 Polyploid polymorphism in A. gerardii Vitman (Poaceae). Genome 29: 374–379.
Lewis, W. H. 1980 Polyploidy in species populations. In W.H. Lewis [ed.], Polyploidy: biological relevance, 103–144. Plenum Press, New York, NY.
———, and Y. Suda. 1976 Diploids and polyploids from a single species population: temporal adaptations. Journal of Heredity 67: 391–393.
———, ———, and B. Macbryde. 1967 Chromosome numbers of Claytonia virginica in the St. Louis, Missouri area. Annals of the Missouri Botanical Garden 10: 51–62.
Levin, D. A. 1975 Minority cytotype exclusion in local plant populations. Taxon 24: 35–43.[CrossRef]
———. 1983 Polyploidy and novelty in flowering plants. American Naturalist 124: 1–25.[CrossRef][ISI]
Lumaret, R. 1988a Adaptive strategies and ploidy levels. Acta Oecologia/Oecologia Plantarum 9: 83–93.
———. 1988b Cytology, genetics and evolution in the genus Dactylis. CRC Critical Reviews of Plant Science 7: 55–91.
———, and E. Barrientos. 1990 Phylogenetic relationships and gene flow between sympatric diploid and tetraploid plants of Dactylis glomerata (Gramineae). Plant Systematics and Evolution 169: 81–96.[CrossRef][ISI]
———, J-L. Guillerm, J. Delay, A. L. Loutfi, J. Izco, and J. Jay. 1987 Polyploidy and habitat differentiation in Dactylis glomerata L. from Galicia (Spain) Oecologia (Berlin) 73: 436–446.
Masters, R. A., R. B. Mitchell, K. P. Vogel, and S. S. Waller. 1993 Influence of improvement practices on big bluestem and indiangrass seed production in tallgrass prairies. Journal of Range Management 46: 183–188.[CrossRef][ISI]
McMillan, C., and J. Weiler. 1959 Cytogeography of Panicum virgatum in central North America. American Journal of Botany 46: 590–593.[CrossRef][ISI]
Norrmann, G. A, C. L. Quarín, and B. L. Burson. 1989 Cytogenetics and reproductive behavior of different chromosome races in six Paspalum species. Journal of Heredity 80: 24–28.
———, ———, and K. H. Keeler. 1997 Evolutionary implications of meiotic chromosome behavior, reproductive biology and hybridization in 6x and 9x cytotypes of Andropogon gerardii (Poaceae). American Journal of Botany 84: 201–207.[Abstract]
Porter, C. L., Jr. 1966 An analysis of variance between upland and lowland switchgrass, Panicum virgatum L. Ecology 47: 980–992.[CrossRef][ISI]
Rodríguez, D. J. 1996a A model for establishment of polyploidy in plants. American Naturalist 147: 33–46.[CrossRef][ISI]
———. 1996b A model for the establishment of polyploidy in plants: viable but infertile hybrids, iteroparity, and demographic stochasticity. Journal of Theoretical Biology, 180: 189–196.[CrossRef]
SAS. 1996 SAS statistical program. SAS Institute, Cary, NC.
Springer, T. L. 1991 Caryopsis size and germination of Andropogon gerardii pedicellate and sessile spikelets. Seed Science and Technology. 19: 461–468.[ISI]
Statview. 1994 StatView statistical program. Abacus Concepts, Berkeley, CA.
Stebbins, G. L. 1947 Types of polyploidy I. Their classification and significance. Advances in Genetics 1: 403–429.
———. 1971 Chromosomal evolution in higher plants. Addison-Wesley, Reading, MA.
Thompson, J. D., and R. Lumaret. 1992 The evolutionary dynamics of polyploid plants: origin, establishment and persistence. Trends in Ecology and Evolution 7: 302–307.[CrossRef]
Weaver, J. E. 1954 North American prairie. Johnsen, Lincoln, NE.
This article has been cited by other articles:
|
|
 |
|
  J. Suda, H. Weiss-Schneeweiss, A. Tribsch, G. M. Schneeweiss, P. Travnicek, and P. Schonswetter Complex distribution patterns of di-, tetra-, and hexaploid cytotypes in the European high mountain plant Senecio carniolicus (Asteraceae) Am. J. Botany, August 1, 2007; 94(8): 1391 - 1401. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
