| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biological Sciences, University of Calgary, Calgary, Alberta, Canada T2N 1N4
Received for publication June 2, 2006. Accepted for publication March 28, 2007.
ABSTRACT
Unlike pollen and seed size, the extent and causes of variation in ovule size remain unexplored. Based on 45 angiosperm species, we assessed whether intra- and interspecific variation in ovule size is consistent with cost minimization during ovule production or allows maternal plants to dominate conflict with their seeds concerning resource investment. Despite considerable intraspecific variation in ovule volume (mean CV = 0.356), ovule production by few species was subject to a size–number trade-off. Among the sampled species, ovule volume varied two orders of magnitude, whereas seed volume varied four orders of magnitude. Ovule volume varied positively among species with flower mass and negatively with ovule number. Tenuinucellate ovules were generally larger that crassinucellate ovules, and species with apical placentation (which mostly have uniovulate ovaries) had smaller ovules than those with other placentation types. Seed volume varied positively among species with fruit mass and seed development time, but negatively with seed number. Seeds grew a median 93-fold larger than the ovules from which they originated. Our results provide equivocal evidence that selection minimizes ovule size to allow efficient resource allocation after fertilization, but stronger evidence that ovule size affords maternal plants an advantage in parent–offspring conflict.
Key Words: integument • nucellus • ovule • parent–offspring conflict • placentation • resource allocation • seed • size–number trade-off
Seed production is costly for many plants, so that resource availability commonly limits fecundity (reviewed by Fenner and Thompson, 2005
) and high fecundity during one reproductive season can diminish subsequent growth, survival, and/or reproduction (reviewed by Obeso, 2002
). The direct costs of seed production are paid during two phases, ovule production and seed development, with the latter phase being most expensive because seeds grow many-fold between fertilization and dispersal. Nevertheless, ovule production requires some resource investment and the cost of individual ovules may influence the number of ovules produced per ovary (Burd, 1995
). Curiously, in contrast to the voluminous literature considering pollen size (reviewed by Sarkissian and Harder, 2001
) and seed size (reviewed by Leishman et al., 2000
), ovule size has seldom been measured (although see Igersheim and Endress, 1998
), and the influences on ovule size and its relation to the cost of seed production remain unexplored.
Ovule size may be subject to selection to minimize the overall costs of seed production. A plant's ovule production establishes its potential seed output, which is then modified to determine realized seed production by various processes, including inadequate fertilization, expression of lethal alleles in fertilized embryos, competition among sibling embryos, preferential maternal investment, and consumption by seed predators (reviewed by Fenner and Thompson, 2005
). Given the uncertainty that an ovule will develop into a seed, economical seed production might involve limited investment in individual ovules, with most investment being expended later, after fertilized embryos have demonstrated their potential to become seeds (Lloyd, 1980
; Westoby and Rice, 1982
). Investment in individual ovules should be particularly small if plants produce more ovules than they can mature into seeds to compensate for losses of developing embryos (e.g., Porcher and Lande, 2005
; L. D. Harder, M. B. Routley and S. A. Richards, unpublished manuscript). Plants may commonly produce such compensatory ovules, as an average of only 60% of ovules become seeds, even after abundant pollination (based on 60 species: M. B. Routley, L. D. Harder and S. A. Richards, unpublished manuscript). According to this resource economy hypothesis, ovule size should vary little within and among species, regardless of ultimate seed size, all else being equal, in contrast to the 11 orders of magnitude of variation for seed volume among angiosperm species (see Westoby et al., 1992
).
Another adaptive hypothesis for interspecific variation in ovule size involves the role of ovule tissues in mediating conflict between the maternal plant and future seeds. An ovule is primarily sporophytic tissue, especially the integuments and nucellus, with the female gametophyte comprised of only a few cells (Bouman, 1984
). Westoby and Rice (1982)
proposed that the envelopment of the female gametophyte in sporophytic tissue allows the maternal plant to control resource investment in developing seeds. In seeming contrast to this hypothesis, trends for reduction of both integument number and nucellus size are recurring themes in angiosperm evolution (Philipson, 1974
; Albach et al., 2001
; Soltis et al., 2005
), perhaps reflecting increased economy in ovule production. This evolution should have created some interspecific variation in ovule size, with species with small nucelli (tenuinucellate) and/or one integument (unitegmic) having smaller ovules than those with large nucelli (crassinucellate) and/or two integuments (bitegmic). In addition, the orientation and arrangement of the ovules within the ovary, as determined by the placentation pattern, may create variation in the proximity of individual ovules to maternal resources, thereby influencing both the ability of the maternal plant to govern resource supply to developing seeds and the equality of access to resources by different ovules within an ovary. For example, seed size and number vary among Solanum species (Solanaceae) with placentation (Symon, 1987
). Thus, aspects of interaction between a maternal plant and its offspring may influence the evolution of ovule size.
In contrast to the resource economy and parent–offspring conflict hypotheses, ovule size could vary among species within integument-nucellus classes, or even within species, either in the absence of selection for restricted ovule size or if such selection is constrained by the demands of seed development. For example, species with brief reproductive seasons may not be able to postpone investment in seeds until after fertilization and still have sufficient time to complete seed growth and maturation. If ovule size has not been minimized universally, it may vary positively with seed size. In addition, ovule size could vary positively with flower size if large flowers invest more resources (R) in ovule production. However, for a specific investment in ovule production, ovule size (S) should vary negatively with ovule number (N), so that
|
| (1) |
; Worley et al., 2000
), pollen (Vonhof and Harder, 1995
; Yang and Guo, 2004
), and seeds (Venable, 1992
); however, Golonka et al. (2005)
found no correlation between ovule number and width among 21 Schiedea species.
As an initial exploration of the costs of ovule production relative to seed production and the influences on ovule production, we surveyed ovule and seed characteristics for 45 angiosperm species. Our analysis assesses the occurrence of size–number trade-offs in ovule production within species and considers influences on interspecific variation in ovule size, including flower, ovule, and seed characteristics. We specifically examine whether interspecific variation in ovule production is consistent with resource economy, the resolution of parent–offspring conflict, and constraints associated with seed development. In this analysis, we assume that plant-level patterns of resource allocation to ovules is subsumed in the details of flower production (e.g., Worley et al., 2000
) and so focus on flower-level allocation and do not consider possible effects of flower or inflorescence number and the deployment of flowers within and among inflorescences.
MATERIALS AND METHODS
We measured ovules for 45 angiosperm species. Forty-two species were sampled between May and August 2004 in Alberta, Canada, including Calgary (51°4' N, 114°8' W), near the University of Calgary Barrier Lake Field Station (51°1' N, 115°4' W), and Fortress Mountain (50°49' N, 115°14' W). Of the remaining three species, Narcissus poeticus and N. tazetta were collected near Sommières, France (43°48' N, 4°4' E), during April 2000 and Hosta rectifolia was collected near Bibai, Hokkaido, Japan (43°20' N, 141°52' E) during July and August 2004.
For each species, we collected one fresh flower from 
40 individuals, of which 30 were stored in 70% ethanol until we examined the ovules. From the remaining 10 fresh flowers, we selected five undamaged flowers randomly, dried them overnight in a 40°C oven, and then weighed them as a group. From the flowers preserved in ethanol, we selected 20 randomly, opened their ovaries, counted the ovules, and determined the placentation type. We then selected randomly one ovule per ovary and measured its length (L), width (W), and depth (D) using a dissecting microscope with an ocular micrometer. With these dimensions, we estimated the ovule's volume (V) as though it was a prolate ellipsoid (V = 
LWD/6). We also determined the number of integuments and nucellus state for as many of the sampled species as possible from published accounts of family traits for monotypic families (Johri et al., 1992
) or genus or species traits for variable families (Björnstad, 1970
; Steeves and Steeves, 1991
).
We revisited locations from which we had collected flowers weekly for 25 of the original 45 species to check whether fruits had matured and, if so, to collect fruits. From each of 30 plants, we collected one ripe fruit, avoiding those with visible damage. For 20 of the fresh fruits (see Appendix for sample sizes), we counted the seeds and measured the length, width, and depth of one randomly selected seed per fruit. Seed volume was calculated as described for ovule volume. Five additional unopened fruits were air-dried in individual envelopes for 6 mo and then weighed as a group. Only fruits that were unopened when placed in the envelope to dry were used to determine fruit mass. The duration of seed maturation was recorded as the period between ovule and seed collection in days.
We used general linear models to analyze the influences on ovule and seed volume (Kutner et al., 2005
). All analyses considered the natural logarithms of all continuous variables. For interspecific analysis of variation in ln(ovule volume), the analyses initially considered a species' placentation and nucellus type and its integument number as categorical factors and their interactions with continuous independent variables. Nonsignificant terms were excluded from a model by backward elimination, with the condition that a term could not be excluded if it was involved in a significant (
< 0.05) interaction.
Interspecific analyses of ovule and seed volume accounted for the phylogenetic relationships among species by the generalized least-squares approach (Martins and Hansen, 1997
), as implemented in SAS by Butler et al. (2000)
. This analysis used a model of evolution by Brownian motion to transform both dependent and independent variables prior to analysis by standard general linear models. Given that the 45 species considered by our study represent 24 taxonomic families (based on APG, 1998
, 2003
), we used the program Phylomatic (Webb and Donohue, 2005
; http://www.phylodiversity.net/phylomatic/) to extract a phylogeny for the sampled families from the family-level phylogeny of Davies et al. (2004)
. For families for which we sampled more than two species, we added genus-level phylogenies onto the family phylogeny based on Johansson and Jansen (1993
: Ranunculaceae), Potter et al. (2002
: Rosaceae), Wojciechowski et al. (2004
: Fabaceae), and Funk et al. (2005
: Asteraceae). We also followed Gottschling et al. (2001)
in depicting the Boraginoideae as basal to the Hydrophylloideae (Phacelia), but could not locate a phylogeny for the three genera that we sampled from the Boraginoideae (Hackelia, Lithospermum, and Mertensia). The lack of resolution of relationships within the Boraginoideae had no practical effect, because missing observations for some variables resulted in exclusion of at least one of these genera from all analyses. The resulting phylogeny represents the topology of current understanding of the relationships between the species that we studied; however, we could not incorporate accurate branch lengths, because the composite phylogeny was based on data collected in different manners. Therefore, we assigned branch lengths based on the number of sampled species beyond a node minus one (Grafen, 1989
). This approach probably implies less independence among many observations than is actually the case, because most of the families included in our study have long independent histories (see Davies et al., 2004
).
If ovule or seed number significantly affected variation in ovule or seed volume (i.e., partial regression coefficient, b 
0), respectively, we also tested for size–number trade-offs based on expectations from Eq. 1. Logarithmic transformation of both sides of this equation results in
|
| (2) |
The gynoecia of Anemone parviflora, Dryas dummondii, and Ranunculus pedatifidus are comprised of many separate uniovulate carpels, raising the question of whether ovules and seeds should be counted per carpel or per flower. Based on Akaike's information criterion, interspecific analyses of per-flower counts explained more variation than those based on per-carpel counts, so we present only the former.
RESULTS
Intraspecific variation
Based on the coefficients of variation (CV), ovule volume varied about twice as much as ovule number within species (Table 1). Over all 45 species, the CV for ovule volume ranged from 0.153 (Thermopsis rhombifolia) to 0.663 (Achillea millefolium), with a mean CV of 0.356 (median = 0.322). Twenty-nine species exhibited variation in ovule number among plants. For these species, the average CV for ovule volume was 0.332 (median = 0.319), whereas the CV for ovule number ranged from 0.040 (Sisymbrium altissimum) to 0.378 (Ranunculus pedatifidus), with a mean of 0.184 (median = 0.170). Among these species, the CVs for ovule volume and number correlated positively (r = 0.418, P < 0.025). The CV for ovule volume did not differ significantly between species with fixed or variable ovule numbers (t41 = 1.35, P > 0.1).
|
Interspecific variation
Species differed extensively in ovule volume, ovule number, flower mass, seed volume, seed number, and fruit mass (Appendix).
Ovules
Ovule volume varied by two orders of magnitude for the sampled species, ranging from 0.00010 mm3 (Pinguicula vulgaris and Sisymbrium altissimum) to 0.0132 mm3 (Arnica cordifolia) (Fig. 1, open bars).
|
|
Multiple regression revealed four aspects of resource allocation that affect interspecific variation in seed volume (Table 2). Species with many seeds produce smaller seeds, and the partial regression coefficient for ln(seed number) does not differ significantly from the value of –1 expected for a linear size–number trade-off (t17 = 0.51, P > 0.5). Mean seed volume varies positively among species with dry fruit mass, and the partial regression coefficient does not differ from 1 (t17 = 0.01, P > 0.9), indicating that species invest proportionally in individual seeds, regardless of flower size. Species with long periods of seed development produce larger seeds. Finally, seed size differed among species with different placentation types. In particular, species with axile placentation produce significantly smaller seeds than those with apical, free-central and marginal placentation (F1,17 = 6.61, P < 0.025), which did not differ from one another (F2,17 = 2.01, P > 0.15). After we accounted for these influences, regression analysis did not detect a significant association of seed volume with ovule volume (F1,16 = 0.92, P > 0.25), although variance inflation factors indicated correlations of ovule volume with seed development time and, to a lesser extent, seed number of sufficient magnitude to complicate assessment of the independent effect of ovule volume (Kutner et al., 2005
).
DISCUSSION
Our study of a geographically limited sample of 45 species revealed moderate variation in ovule size both within and among angiosperm species. Based on the observed coefficients of variation, ovule volume varies roughly threefold more among plants within species than does pollen size (Vonhof and Harder, 1995
; Cresswell, 1998
), suggesting that ovule volume either is more phenotypically plastic and/or is subject to weaker stabilizing selection. For example, Ishii and Morinaga (2005)
found that ovule volume, but not pollen volume, varied significantly with petal length both within and among Iris gracilipes plants. Given the extent of intraspecific variation in ovule size revealed by our study, more detailed analysis of the sources of this variation (e.g., flower size and position within the ovary) is warranted. Interspecifically, ovule size varies less than seed size (Fig. 1), suggesting fewer opportunities, or reduced impetus, for diversification of ovule size.
Our results provide contradictory evidence for the resource economy hypothesis that plants minimize resource investment in ovules. The general absence of a negative association of ovule size and number within species (Table 1; also see Solomon, 1988
; Ishii and Morinaga, 2005
) is consistent with this hypothesis, although these intraspecific analyses did not account for variation in flower size (i.e., resource availability). In contrast, several aspects of interspecific variation in ovule volume contradict the resource economy hypothesis. The nucelli of crassinucellate species include more cell layers, but tenuinucellate species tend to have larger ovules. Furthermore, the positive interspecific association of ovule volume with flower mass indicates that ovule size is not minimized, but instead depends upon current investment in individual flowers. The negative interspecific association of ovule size and number also indicates that ovule production depends on resource allocation within flowers, so that plants seem not to produce the smallest possible ovules to minimize pre-zygotic investment.
The preceding conclusion does not completely reject the hypothesis that ovule size is reduced to focus resource allocation on viable offspring (Lloyd, 1980
; Westoby and Rice, 1982
). Indeed, the median 93-fold conversion ratio of ovule volume into seed volume confirms that most investment in seed production occurs post-zygotically. Seeds are probably even more costly than this ratio suggests, if they contain higher proportions of proteins and especially lipids. Nevertheless, the diversity of influences on interspecific variation in ovule volume that are inconsistent with the resource economy hypothesis suggests that additional factors influence the evolution of ovule size.
The observed influences on interspecific variation in ovule volume provide more support for the hypothesis that ovule size mediates parent–offspring conflict over the allocation of maternal resources within flowers (see Lloyd, 1980
; Westoby and Rice, 1982
). Two lines of evidence support this conclusion. First, species with apical placentation have significantly smaller ovules than those with other types of placentation. All seven of the species in our analysis with either uniovulate flowers (five species) or gynoecia comprised of univolate carpels (two species) have apical placentation (the remaining three species in this group have four ovules per ovary). Species with one ovule per ovary have the least opportunity for parent–offspring conflict if resource allocation among ovules occurs primarily within rather than between flowers (which we did not assess) because of similar optimal resource allocations for the maternal plant and developing seed. In contrast, maternal plants of multi-ovulate species may need to exert stronger control over investment in individual seeds, which could favor incorporation of relatively more sporophytic tissue (integuments and nucellus) in ovules. Therefore, the uniovulate ovaries (as opposed to flowers) of most species characterized by apical placentation may obviate the need for them to produce large ovules to allow maternal dominance in parent–offspring conflict. The second line of evidence for a possible role of ovule size in parent–offspring conflict comes from the nature of the interspecific trade-off between ovule volume and number. The partial regression coefficient for this relation was smaller in absolute value than expected from a simple, linear size–number trade-off, even after accounting for variation in resource availability (i.e., flower mass). This result indicates that species with many ovules per ovary invest proportionately more per ovule than those with few ovules, as expected if maternal dominance depends on sporophytic investment in ovule tissue. This intriguing evidence indicates that assessment of a role of parent–offspring conflict in the evolution of ovule size deserves further attention.
In addition to apparent support for the ultimate hypotheses concerning parent–offspring conflict, our results provide clear evidence that ovule size depends proximately on resource allocation constraints. Specifically, species that invest more resources in individual flowers generally also have larger ovules, and those that produce more ovules per flower have smaller ovules. These independent effects involve both levels in the hierarchical allocation of resources to female function within flowers: reproductive investment and the subdivision of investment among individual ovules. In contrast, we found no evidence that ovule size affects eventual seed size. This result suggests that interspecific variation in ovule size primarily reflects direct effects on ovule function, rather than correlated consequences of seed size variation, and argues for a broader analysis of variation in ovule size among angiosperms and the factors that influence its evolution.
APPENDIX. Mean (SD) ovule number and volume, seed number and volume, dry flower mass, dry fruit mass, nucellus type (crassinucellate or tenuinucellate), number of integuments and placentation type for 45 angiosperm species. Mean flower and fruit masses were measured for pooled samples of five flowers, so standard deviations could not be calculated. Ovule characteristics are based on samples of 20 flowers per species, whereas the sample sizes for seed characteristics are indicated in square brackets after a species' name
FOOTNOTES
1 The authors thank B. Smith for taxonomic assistance and M. L. Reid and M. B. Routley for comments on the manuscript. A Discovery Grant (LDH) from the Natural Sciences and Engineering Research Council of Canada provided partial funding for this research. 
2 Present address: Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9
3 Author for correspondence (harder{at}ucalgary.ca
)
LITERATURE CITED
Albach D. C. Soltis P. S. Soltis D. E.. 2001. Patterns of embryological and biochemical evolution in the Asterids. Systematic Botany 26: 242-262.[ISI]
APG [Angiosperm Phylogeny Group].. 1998. An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden 85: 531-553.[CrossRef][ISI]
APG [Angiosperm Phylogeny Group].. 2003. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants. Botanical Journal of the Linnean Society 141: 399-436.[CrossRef][ISI]
Björnstad I. N.. 1970. Comparative embryology of Asparagoideae–Polygonateae, Liliaceae. Nytt Magasin for Botanikk 17: 169-207.
Bouman F.. 1984. The ovule. In B. M. Johri [ed.], Embryology of angiosperms 123-157 Springer-Verlag, Berlin, Germany.
Burd M.. 1995. Ovule packaging in stochastic pollination and fertilization environments. Evolution 49: 100-109.[CrossRef][ISI]
Butler M. A. Schoener T. W. Losos J. B.. 2000. The relationship between sexual size dimorphism and habitat use in Greater Antillean Anolis lizards. Evolution 54: 259-272.[CrossRef][ISI][Medline]
Cresswell J. E.. 1998. Stabilizing selection and the structural variability of flowers within species. Annals of Botany 81: 463-473.
Davies T. J. Barraclough T. G. Chase M. W. Soltis P. S. Soltis D. E. Savolainen V.. 2004. Darwin's abominable mystery: insights from a supertree of the angiosperms. Proceedings of the National Academy of Sciences, USA 101: 1904-1909.
Fenner M. Thompson K.. 2005. The ecology of seeds Cambridge University Press, Cambridge, UK.
Funk V. A. Bayer R. J. Keeley S. Chan R. Watwon L. Gemeinholzer B. Schilling E. Panero J. L. Baldwin B. G. Garcia-Jacas N. Susanna A. Jansen R. K.. 2005. Everywhere but Antarctica: using a supertree to understand the diversity and distribution of the Compositae. Biologiske Skrifter Kongelige Danske Videnskabernes Selskab 55: 343-374.
Golonka A. M. Sakai A. K. Weller S. G.. 2005. Wind pollination, sexual dimorphism, and changes in floral traits of Schiedea (Caryophyllaceae). American Journal of Botany 92: 1492-1502.
Gottschling M. Hilger H. H. Wolf M. Diane N.. 2001. Secondary structure of the ITS1 transcript and its application in a reconstruction of the phylogeny of Boraginales. Plant Biology 3: 629-636.[CrossRef]
Grafen A.. 1989. The phylogenetic regression. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 326: 119-157.[CrossRef]
Igersheim A. Endress P. K.. 1998. Gynoecium diversity and systematics of paleoherbs. Botanical Journal of the Linnean Society 127: 289-370.[CrossRef][ISI]
Ishii H. S. Morinaga S.-I.. 2005. Intra- and inter-plant level correlations among floral traits in Iris gracilipes (Iridaceae). Evolutionary Ecology 19: 435-448.[CrossRef][ISI]
Johansson J. T. Jansen R. K.. 1993. Chloroplast DNA variation and phylogeny of the Ranunculaceae. Plant Systematics and Evolution 187: 29-49.[CrossRef][ISI]
Johri B. M. Ambegaokar K. B. Srivastava P. S.. 1992. Comparative embryology of angiosperms Springer-Verlag, Berlin, Germany.
Kutner M. H. Nachtsheim C. J. Neter J. Li W.. 2005. Applied linear statistical models, 5th ed McGraw-Hill/Irwin, New York, New York, USA.
Leishman M. R. Wright I. J. Moles A. T. Westoby M.. 2000. The evolutionary ecology of seed size. In M. Fenner [ed.], Seeds—the ecology of regeneration in plant communities 31-57 CAB International, Wallingford, UK.
Lloyd D. G.. 1980. Sexual strategies in plants. I. An hypothesis of serial adjustment of maternal investment during one reproductive season. New Phytologist 86: 69-79.[CrossRef][ISI]
Martins E. P. Hansen T. F.. 1997. Phylogenies and the comparative method: a general approach to incorporating phylogenetic information into the analysis of interspecific data. American Naturalist 149: 646-667.[CrossRef][ISI]
Obeso J. R.. 2002. The costs of reproduction in plants. New Phytologist 155: 321-348.[CrossRef][ISI]
Philipson W. R.. 1974. Ovular morphology and the major classification of the dicotyledons. Botanical Journal of the Linnean Society 68: 89-108.[ISI]
Porcher E. Lande R.. 2005. Reproductive compensation in the evolution of plant mating systems. New Phytologist 166: 673-684.[CrossRef][ISI][Medline]
Potter D. Gao F. Esteban Bortiri P. Oh S.-H. Baggett S.. 2002. Phylogenetic relationships in Rosaceae inferred from chloroplast matK and trnL-trnF nucleotide sequence data. Plant Systematics and Evolution 231: 77-89.[CrossRef][ISI]
Sakai S.. 2000. Size–number trade-off and allocation to flower production in animal-pollinated flowers. Plant Species Biology 15: 261-268.[CrossRef]
Sarkissian T. S. Harder L. D.. 2001. Direct and indirect responses to selection on pollen size in Brassica rapa L. Journal of Evolutionary Biology 14: 456-468.[CrossRef][ISI]
Solomon B. P.. 1988. Patterns of pre- and postfertilization resource allocation within an inflorescence: evidence for interovary competition. American Journal of Botany 75: 1074-1079.[CrossRef][ISI]
Soltis D. E. Soltis P. S. Endress P. K. Chase M. W.. 2005. Phylogeny and evolution of angiosperms Sinauer, Sunderland, Massachusetts, USA.
Steeves T. A. Steeves M. W.. 1991. Flower development in Amelanchier alnifolia Maloideae. Canadian Journal of Botany 69: 844-857.
Symon D. E.. 1987. Placentation patterns and seed numbers in Solanum (Solanaceae) fruits. Journal of the Adelaide Botanic Gardens 10: 179-200.
Venable D. L.. 1992. Size–number trade-offs and the variation of seed size with plant resource status. American Naturalist 140: 287-304.[CrossRef][ISI]
Vonhof M. J. Harder L. D.. 1995. Size-number trade-offs and pollen production by papilionaceous legumes. American Journal of Botany 82: 230-238.[CrossRef][ISI]
Webb C. O. Donoghue M. J.. 2005. Phylomatic: tree retrieval for applied phylogenetics. Molecular Ecology Notes 5: 181-183.[CrossRef][ISI]
Westoby M. Jurado E. Leishman M.. 1992. Comparative evolutionary ecology of seed size. Trends in Ecology and Evolution 7: 368-372.[CrossRef]
Westoby M. Rice B.. 1982. Evolution of seed plants and inclusive fitness of plant tissues. Evolution 36: 713-724.[CrossRef][ISI]
Wojciechowski M. F. Lavin M. Sanderson M. J.. 2004. A phylogeny of legumes (Leguminosae) based on analysis of the plastid matK gene resolves many well-supported subclades within the family. American Journal of Botany 91: 1846-1862.
Worley A. C. Baker A. M. Thompson J. D. Barrett S. C. H.. 2000. Floral display in Narcissus: variation in flower size and number at the species, population, and individual levels. International Journal of Plant Sciences 161: 69-79.[CrossRef][ISI][Medline]
Yang C.-F. Guo Y.-H.. 2004. Pollen size–number trade-off and pollen–pistil relationships in Pedicularis (Orobanchaceae). Plant Systematics and Evolution 247: 177-185.[ISI]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
