Printer Friendly

The role of resources and pathogens in mediating the mating system of Kalmia latifolia.

INTRODUCTION

Fruit maturation is a dynamic process and may be influenced by many factors (Casper and Niesenbaum 1993). Among these, the source of pollen can affect which seeds will draw maternal resources required to mature (reviewed in Stephenson and Bertin 1983, Willson and Burley 1983, Lee 1988, Marshall and Folsum 1991). Competition for limited resources, early inbreeding depression, and/or nonrandom distribution of resources can lead to outcrossed fruits being provisioned more richly than self-fertilized fruits (Bushnell 1920, Haber 1928, Urata 1954, Kambal 1969, Stephenson 1981). In such cases, the degree of a plant's resource limitation may influence its "mating system" (i.e., how many selfed seeds it will produce compared to outcrossed seeds) by altering the allocation of resources to outcrossed vs. selfed progeny. Consequently, the dynamic forces that control plant resource levels may greatly affect plant reproductive success and plant mating systems.

Many factors can influence plant resource levels. We chose to examine the effect of pathogens on mating systems, because pathogens have been demonstrated to affect plant reproduction in several ways (Burdon 1987, Dobson and Crawley 1994). These include pathogen-induced castration of a host's chasmogamous flowers (Clay 1991), the alteration of pollinator behavior caused by floral mimicry by pathogens (Roy 1993), patterns of pathogen transmission which can be related to the probabilities of colonization and extinction of host populations (Antonovics 1994), and the direct loss of fruits from a variety of rots and mildews (Rawal 1993, Duffy and Gardner 1994). Disease may also influence plant fitness by mediating the resource provisioning of seeds and/or fruits, since diseases can influence the resource levels available for fruit development and maturation (Burdon 1987).

Pathogens can reduce the pool of resources within individual plants (Burdon 1987). Limited resources have been shown to reduce the quantity of seeds produced (Stephenson 1981, Willson and Burley 1983); but, here we examine how resource limitation may affect the overall quality of progeny.

In the present study we wish to address three questions: (1) Is high infection associated with high outcrossing? (2) Does infection reduce seed provisioning? (3) Can infection influence the proportion of selfed seeds produced? To address these questions, we examined the interaction between a fungal pathogen and the expression of mating system in individuals of Kalmia latifolia. Kalmia latifolia is a long-lived, evergreen shrub, found mainly in the mountains of the east coast of the United States (Jaynes 1988). This hermaphrodite is known to have variable mating system expression across geographic locations at both local and regional spatial scales (Rathcke and Real 1993), and it is commonly attacked by the necrotic leaf spot fungal pathogen, Cercospora kalmiae.

We addressed the questions associated with an interaction between mating system and disease occurrence in two ways. First, within natural populations of K. latifolia, we documented the correlation between two measures of plant reproductive success (percentage fruit set and percentage seedling survival) and disease intensity, and we partitioned the correlation by mating class. For the first study, flowers were self-pollinated, naturally pollinated, or pollinated by hand with outcrossed pollen.

For the second study, in an attempt to mimic the effect of disease, we performed a manipulative field experiment to test for an effect of resource reduction on the fitness consequences of different mating strategies. There were only two pollination treatments: self-pollinated and naturally pollinated. The purpose of this 2 x 2 design (punched vs. unpunched leaves, self-pollinated vs. outcrossed flowers) was to ascertain the interaction between leaf damage and the probability of maturing fruit following selfed vs. outcrossed pollination. In this way, we could test two hypotheses. We predicted that the effect of leaf damage would be localized within a plant, thus damage would only reduce the fruit set of the inflorescence closest to the damaged leaves. We hypothesized that reduced resources may constrain the development of self-fertilized fruits more than outcrossed fruits and, therefore, that resource reduction will increase the proportion of outcrossed progeny. Here we demonstrate that a leaf-damaging pathogen can influence the expression of a plant's mating system.

MATERIALS AND METHODS

The system

Kalmia latifolia produces flowers clustered in corymbs, and one plant can have thousands of flowers open at one time (with an average of 20 flowers distributed on each of [greater than]500 inflorescences; M. A. Levri, unpublished data). Bumblebees (Bombus spp.) are its primary pollinators (Jaynes 1988). The anther sacs are recurved into grooves in the corolla, ready to spring and catapult pollen onto the pollinator's abdomen. The pollinator deposits threaded clumps of pollen tetrads onto the stigma of another flower. If no pollinator visits the flower, the filaments will eventually spring as the corolla senesces, throwing the pollen onto the flower's own pistil. Thus, delayed selfing (Lloyd 1988) may be the predominant mode of selfing.

Kalmia latifolia at the Virginia sites are infected with Cercospora kalmiae, which thrives in moist and shady conditions. Cercospora kalmiae is a nonsystemic fungus that causes circular, necrotic spots visible on one-year-old leaves (Sinclair et al. 1987). Transmission, primarily through water splashing, is thought to progress from old leaves to new leaves on the same plant (K. Rane, personal communication). Infections are initiated on new leaves before and during flowering, and severe infection reduces flower production and stunts plant growth (Sinclair et al. 1987). The necrotic spots result in the loss of photosynthetic area and presumably reduce carbon resources available to reproduction. This reduction in resources may then influence plant mating system expression.

Our study included two populations of K. latifolia near the Mountain Lake Biological Station in the Allegheny Mountains of Virginia (37 [degrees] 22[minutes] N, 80 [degrees] 31[minutes] W). The Bald Knob (BK) site is an exposed, rocky outcrop at 1313 m elevation. The Hedwig Trail (HT) site is 2.9 km northwest of the BK site in an oak-maple understory at 1160 m elevation. In order to assess morphological differences between the two sites, five indicators were quantified: (1) Leaf density per plant was measured by placing a meter tape over the leaf canopy of each plant, counting the number of leaves touching the tape, and dividing that value by the total length of the canopy ("number of leaves per meter"). (2) The size of the arc of the leaf canopy measured by the meter tape was also recorded ("length of leaf transect"). (3) The average number of flowers per inflorescence was noted from 30 haphazardly chosen inflorescences counted on each plant. (4) The number of flowers per plant was extrapolated by multiplying the average number of flowers per inflorescence by (5) the total number of inflorescences on a plant. Variation in female reproductive success and disease incidence were also examined.

Pollinations

Three pollination treatments were used to estimate female reproductive success in 1994 and 1995. Autogamously self-pollinated inflorescences were bagged throughout the flowering season using bridal veiling (selfed treatment). Inflorescences in the "natural" group were left untreated to receive pollen from natural pollinators (primarily bumblebees). For the hand pollination treatment, flowers were hand pollinated each day with pollen collected onto a microscope slide from at least five plants [greater than]5 m away. Every stigma was saturated with pollen. The purpose of the hand pollination treatment was to measure a maximum outcrossed fruit set, unlimited by pollen. Hand-pollinated inflorescences were bagged with bridal veiling in 1994 but not in 1995. We replicated each treatment on five entire inflorescences on each plant. Eighty-one plants were included in the study: 41 plants at HT and 40 plants at BK. Fruit set percentage was calculated for each treatment as the sum of the number of fruits produced divided by the sum of the number of flowers, in all of the inflorescences in that treatment. Because isozyme studies have shown that the Virginia sites predominantly outcross (over 85% of their seeds on average, with as much as 95% outcrossing in some years; M. A. Levri, unpublished data), fruits from natural pollinations and hand pollinations with outcrossed pollen will be referred to as "outcrossed."

After fruits were collected in October 1994, they were stored at room temperature until seeds were planted. The procedure for germination followed the protocol of Jaynes (1988). Single fruits were arbitrarily chosen from randomly chosen infructescences. Five seeds from each fruit were sown in a cell containing a soil mix of perlite, Canadian peat, and sphagnum. Location in the greenhouse (under an intermittent mist) and time of sowing were randomly assigned by site and by treatment. Seedling survival was measured as the percentage of seedlings surviving to 5 wk after sowing. The success of progeny resulting from each pollination treatment (selfed vs. natural vs. hand pollinated) was then compared to disease intensity on the maternal plant. All analyses were performed using the JMP statistical package (SAS Institute 1989). The t tests, ANOVA, and two-way ANOVA that involved comparison of fruit set percentages were performed on the arcsine- and square-root-transformed percentages to meet the assumption of equal variance for the test.

Estimate of disease damage

In 1994, without removing leaves, we traced 25 one-year-old leaves on each of 21 plants at the exposed BK site and 25 plants at the understory HT site. For each plant, leaves were chosen by dividing the leaf canopy into five sections equal in volume and selecting five leaves within each section. Tracings also recorded leaf area damaged by lesions of C. kalmiae. Leaf tracings were then digitized and analyzed using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image/). To estimate the level of infection on each plant in terms of the average proportion of leaf damaged by C. kalmiae, we summed the area damaged on 25 leaves on one plant and divided it by the sum of the total leaf area of the same 25 leaves (i.e., damaged area/total area). Linear regressions between (1) fruit set and level of infection and (2) seedling survival and level of infection were analyzed for each pollination treatment within and between sites. An ANCOVA on fruit set was performed to test for site, site-by-disease, and disease effects. Since there were no significant site (F = 0.22, P = 0.63, df = 1, N = 43) and site-by-disease effects (F = 0.01, P = 0.99, df = 1, N = 43) in this model, the data from both sites were pooled for the above regressions 1 and 2.

Effect of disease-simulating resource limitation

In 1995, we designed an experiment to test the effect of disease-like resource limitation on the fruit set of self-pollinated vs. naturally pollinated (i.e., outcrossed) flowers. Leaf damage artificially applied via hole punching was used as a surrogate for damage sustained through high infection of C. kalmiae. We chose hole punching in order to mimic two obvious effects of the necrotic spots caused by C. kalmiae: the reduction of photosynthetic area and the production of a callus around the damaged part of the leaf. We punched four circular 0.35-[cm.sup.2] holes (two on each side of the midvein) in the leaves immediately subtending 10 inflorescences on 20 of the least infected individuals at the BK site. (No plant was found to be disease free.) The average amount of leaf tissue removed artificially (6.2%) was comparable to the average amount of natural damage (6.6%). On average, five one-year-old leaves subtended each inflorescence. New leaves were untreated. Whole inflorescences received natural or selfed pollinations. It has been shown that there is no difference in fruit set between autogamously selfed flowers and flowers hand pollinated with an abundance of selfed pollen (Rathcke and Real 1993) on the same individuals that were included in our study. Thus the reduction in selfed fruit set is not likely to be attributable to pollen limitation. Ten more inflorescences on the same plant received the same pollination treatments, but the leaves subtending these inflorescences were left unmanipulated.

Fruit set percentages from each of these treatments were calculated. Lost infructescences (i.e., those not present at collection in October due to selective abortion or chance events) were not included in the calculation of fruit set. The proportion of infructescences that were lost was also calculated.

RESULTS

Variation in selfing and pathogenic attack

Plants at the BK site set significantly more self-fertilized fruits in 1994 (9.1% [+ or -] 1.5%) than plants at HT (3.2% [+ or -] 1.0%) (t = 3.67, P [less than] 0.001, df = 1, N = 81). We found the same result in 1995: in self-pollinated flowers, BK set 19.8% [+ or -] 2.6% of its fruits and HT set 9.5% [+ or -] 2.0% (t = 3.51, P [Lambda] 0.001, df = 1, N = 61). At the HT site in 1994, there was no significant difference in fruit set between flowers in the hand and natural and pollination treatments (t = i.28, P = 0.2, df = 1) [ILLUSTRATION FOR FIGURE 1 OMITTED]. At the BK site, there was a significant effect due to pollination treatment: the fruit set for flowers in the hand pollination treatment equaled 75.6% whereas the fruit set for flowers in the natural pollination treatment equaled 69.4% (t = 5.18, P [less than] 0.001, df = 1). The proportion of leaf area infected by C. kalmiae of HT plants ([Mathematical Expression Omitted]) was 36.3% greater than that of BK plants ([Mathematical Expression Omitted]) (t = 3.09, P [less than] 0.004, df = 1).

Evidence for resource limitation

BK plants were exposed to more light than plants in the HT understory (M. A. Levri, unpublished data). BK plants also produced more and smaller leaves, and more flowers per plant distributed on more inflorescences [ILLUSTRATION FOR FIGURE 2 OMITTED]. The average number of flowers per inflorescence did not differ between the sites [ILLUSTRATION FOR FIGURE 2 OMITTED]. Plants at HT showed a correlation between the plant size (measured as the product of the volume of the plant canopy and leaf density) and the number of inflorescences ([R.sup.2] = 0.28, P [less than] 0.0004, N = 41). However, at BK there was no relationship between a plant's size and the number of inflorescences it produces ([R.sup.2] = 0.03, P = 0.34, N = 40).

In addition, the plants at BK that showed greater production of self-fertilized fruits also showed greater production of outcrossed fruits ([R.sup.2] = 0.40, P [less than] 0.002, N = 40). However, there was no relationship between selfed and outcrossed fruit production in the plants at HT ([R.sup.2] = 0.00, P = 0.88, N = 41).

Relationship between disease damage and fruit set

Fruit set was significantly correlated with disease damage for self-pollinated flowers, but no significant correlation was detected in flowers that received the natural and hand pollinations (i.e., outcrossed treatments) (Table 1A). Plants with greater damage by C. kalmiae set a lower proportion of self-fertilized fruits ([R.sup.2] = 0.178, P [less than] 0.005, N = 46) [ILLUSTRATION FOR FIGURE 3 OMITTED].

Relationship between disease damage and seedling survival

In the same way, seedling survival was significantly correlated with disease damage for self-pollinated flowers, but no significant correlation was detected in flowers that received the natural and hand pollinations (i.e., outcrossed treatments) (Table lB). Plants with greater damage by C. kalmiae produced selfed offspring that had lower seedling survival ([R.sup.2] = 0.127, P [less than] 0.038, N = 46) [ILLUSTRATION FOR FIGURE 4 OMITTED].

Experimental manipulation

The interaction between pollination treatment and leaf damage (i.e, hole punching) had a significant effect on fruit set (F = 7.49, P [less than] 0.01, df = 1, 69, N = 20) [ILLUSTRATION FOR FIGURE 5 OMITTED]. Leaf damage reduced the fruit set of outcrossed flowers by 17% compared to undamaged controls (t = 2.28, P [less than] 0.02, df = 1) and reduced the fruit set of self-pollinated flowers by 31% compared to controls (t = 2.57, P [less than] 0.02, df = 1). Artificial leaf damage reduced the fruit set of self-pollinated flowers more than outcrossed flowers. Comparison of the normalized reduction in fruit set between selfed and outcrossed pollinations (correcting for the overall difference in fruit set by comparing the product of outcrossed fruit set and outcrossed punched fruit set to the product of selfed fruit set and selfed punched fruit set) shows that selfed progeny experience an additional 13.6% reduction due to hole punching relative to the reduction experienced by outcrossed progeny (t = 3.87, P [less than] 0.02, df = 1).

The same pattern emerges in the proportion of lost (and presumably aborted) infructescences. There was a significant interaction between pollination treatment and leaf damage (F = 4.92, P [less than] 0.03, df = 1, 76) [ILLUSTRATION FOR FIGURE 6 OMITTED]. For inflorescences receiving outcrossed pollinations, we did not detect a significant effect of damage on infructescence survival (t = 0.26, P = 0.80, df = 1). However, for self-pollinated flowers, punching reduced the survival of infructescences by 44% compared to nondamaged controls (t = 2.76, P [less than] 0.01, df = 1). Thus, artificial leaf damage reduced the survival of selfed infructescences significantly more than outcrossed infructescences (F = 11.19, P [less than] 0.003, df = 1, 76).

DISCUSSION

Our results suggest that resource limitation, caused by infection in this case, can bias a plant's reproductive output, resulting in a greater representation of outcrossed offspring. Three lines of evidence lead us to the conclusion that disease-damaged plants produce proportionally more outcrossed seeds: 11) Fruit set was not significantly correlated with disease damage in outcrossed pollinations (from natural pollinations and pollinations by hand), whereas the performance of selfed flowers was significantly negatively correlated with disease damage. (2) Artificial damage reduced the fruit [TABULAR DATA FOR TABLE 1 OMITTED] set of outcrossed flowers significantly less than selfed flowers. (3) Finally, artificial damage increased infructescence abortion for infructescences produced from the selfed pollination treatment but did not affect infructescences produced from the outcrossed pollination treatment. Thus, the proportion of selfed seeds that the plant will mature seems to be a negative function of the degree of leaf damage due to infection. If inbreeding depression is high and the absolute number of seeds is not greatly reduced at a more infected site, then infected plants may approach the fitness of uninfected plants.

In the hole punch experiment, we observed an interaction between pollination treatment and disease simulation: leaf damage reduced the fruit set of self-pollinated flowers more than the fruit set of outcrossed flowers on the same plant. We would not expect to observe this sort of differential response to damage if these plants were physiologically integrated (in sensu Watson and Casper 1984) due to our experimental design. For each plant, we assigned 20 inflorescences and their subtending leaves to one of the four pollination and damage treatments, and left the rest of the plant's flowers and leaves unmanipulated. If the plants were fully integrated, the flowers close to damaged leaves would be able to sequester resources from undamaged parts of the plant and would exhibit no difference from flowers close to undamaged leaves. Rather, in this case, the leaves immediately subtending inflorescences appear to be important sources of carbon for provisioning maturing fruits. In this way, the resources used for provisioning seeds and fruits as they develop seem to be localized within a plant (e.g., Galen et al. 1993 and references therein). Therefore, disease damage to the leaves subtending each inflorescence affects the provisioning of those fruits and mediates what type of seed (outcrossed vs. selfed) ultimately matures.

In addition, the results from the hole punch manipulation support the prediction that disease inhibits the production of selfed seeds by reducing photosynthetic area, thus lowering resources. This suggests that the natural effects of the fungal spot disease can generate similar pressures modifying selfing rate via resource reduction in the plant population. Thus, some of the natural variation observed in the production of selfed progeny may be attributable to disease incidence and intensity. If a plant is heavily diseased, then it would mature proportionally fewer selfed seeds.

Theory predicts that if outcrossed pollen is limiting, a plant can achieve higher fitness by producing a few poor seeds (i.e., to self) than by not producing any seeds at all (Lloyd 1979, Schoen and Lloyd 1992). But, this prediction considers reproduction within one flowering season. K. latifolia is a long-lived perennial, and some plants that bear a large number of flowers in one year produce few flowers in the next year (M. A. Levri, personal observation). If there is a cost to reproduction (Pyke 1991, Galen 1993, Ashman 1994, Primack et al. 1994), then it may be advantageous to save some resources for next year's production of more outcrossed seeds rather than to spend limited resources on selfed seeds.

The examination of conditional expression of mating system within an individual is not unprecedented. Beccera and Lloyd (1992) proposed a "best of both worlds" hypothesis to explain a resource-based, conditional response in Phormium tenax. They found that when resources were shared within a branch, outcrossed fruits were preferentially matured over selfed fruits. But, when the branches were far enough away that resources were not shared between outcrossed and selfed fruits, selfed seeds did mature. They found that spatial partitioning of resources determined whether or not selfed seeds were matured. Our data support a similar "best of both worlds" hypothesis. In K. latifolia, if the plant is infected, resources are reduced. Outcrossed seeds will be provisioned preferentially; selfed seeds mature only if there are resources to spare. In this way, temporal partitioning of pathogen-mediated resources may be responsible for some variation in the mating system.

The Kalmia-Cercospora interaction suggests a novel way that disease may influence plant population structure via its effects on individual resource availability and maternal provisioning of offspring. Given the ubiquitous occurrence of diseases within natural systems, we suspect that such disease-mediated effects on the mating structure of plant populations is very common and should be considered a potential force shaping mating system expression.

ACKNOWLEDGMENTS

M. Levri wishes to thank Ed Levri, Paul McElhany, Gall LaMoreaux, Dana Dudle, Patrick Hatch, Colleen Hatch, Daniel Rozen, Jonathan Brandt, Jennifer Secki, Leigh Rosenberg, Dr. Mary Crowe, Stasia Skillman, Charlie Werth, Stacie Emery, Dean Kleinhenz, Lori Klukowski, and Henry Wilbur for work in the field. This manuscript was improved by comments from Sandy Davis, Dana Dudle, Ed Levri, Deborah Marc, Paul McElhany, Janet Gehring, Lynda Delph, Keith Clay, Loren Rieseberg, and three anonymous reviewers. M. Levri also thanks the editor, W. Scott Armbruster, for his heroic patience. Grants to M. Levri from Indiana University, Indiana Academy of Sciences, Sigma Xi Scientific Research Society, and University of Virginia's Mountain Lake Biological Station made this research possible.

LITERATURE CITED

Antonovics, J. 1994. Ecological Genetics of metapopulations: the Silene-Ustilago plant-pathogen system. Pages 146-170 in L. A. Real, editor. Ecological genetics. Princeton University Press, Princeton, New Jersey, USA.

Ashman, T. L. 1994. A dynamic perspective on the physiological cost of reproduction in plants. American Naturalist 144:300-316.

Becerra, J. X., and D. G. Lloyd. 1992. Competition dependent abscission of self-pollinated flowers of Phormium tenax (Agavaceae): a second action of self-incompatibility at the whole flower level? Evolution 46:458-469.

Burdon, J. J. 1987. Diseases and plant population biology. Cambridge University Press, Cambridge, UK.

Bushnell, J. W. 1920. The fertility and fruiting habit in Cucurbita. Proceedings of the American Society of Horticultural Science 17:47-51.

Casper, B. B., and R. A. Niesenbaum. 1993. Pollen vs. resource limitation of seed production: a reconsideration. Current Science 65:210-214.

Clay, K. 1991. Parasitic castration of plants by fungi. Trends in Ecology and Evolution 6: 162-166.

Dobson, A., and M. Crawley. 1994. Pathogens and the structure of plant communities. Trends in Ecology and Evolution 9:393-398.

Duffy, B. K., and D. E. Gardner. 1994. Locally established Botrytis fruit rot of Myrica faya, a noxious weed in Hawaii. Plant Disease 78:919-923.

Galen, C. 1993. Cost of reproduction in Polemonium viscosum: phenotypic and genetic approaches. Evolution 47: 1073-1079.

Galen, C., T. E. Dawson, and M. L. Stanton. 1993. Carpels as leaves: meeting the carbon cost of reproduction in an alpine buttercup. Oecologia 95:187-193.

Haber, E. S. 1928. Inbreeding the table queen (Des Moines) squash. Proceedings of the American Society of Horticultural Science 25:111-114.

Jaynes, R. A. 1988. Kalmia, the laurel book. II. Timber Press, Portland, Oregon, USA.

Kambal, A. E. 1969. Flower drop and fruit set in field beans, Vicia faba L. Journal of the Agricultural Society of Cambridge 72:131-138.

Lee, T. D. 1988. Patterns of fruit and seed production. Pages 179-202 in J. Lovett Doust and L. Lovett Doust, editors. Plant reproductive ecology: patterns and strategies. Oxford University Press, New York, New York, USA.

Lloyd, D. G. 1979. Some reproductive factors affecting the selection of self-fertilization in plants. American Naturalist 113:67-79.

-----. 1988. Benefits and costs of biparental and uniparental reproduction in plants. Pages 233-252 in R. E. Michod and B. R. Levin, editors. The evolution of sex. Sinauer, Sunderland, Massachusetts, USA.

Marshall, D. L., and M. W. Folsom. 1991. Mate choice in plants: an anatomical to population perspective. Annual Review of Ecology and Systematics 22:37-63.

Primack, R. B., S. L. Miao, and K. R. Becker. 1994. Costs of reproduction in the pink lady's slipper orchid (Cypripedium acaule): defoliation, increased fruit production, and fire. American Journal of Botany 81: 1083-1090.

Pyke, G. H. 1991. What does it cost a plant to produce floral nectar? Nature 350:58-59.

Rathcke, B., and L. A. Real 1993. Autogamy and inbreeding depression in mountain laurel, Kalmia latifolia (Ericaceae). American Journal of Botany 80:143-146.

Rawal, R. D. 1993. Yield loss in guava by different fruit rots. International Journal of Tropical Plant Diseases 11: 69-72.

Roy, B.A. 1993. Floral mimicry by a plant pathogen. Nature 362:56-58.

SAS Institute. 1989. JMP user's guide. Version 2. SAS Institute, Cary, North Carolina, USA.

Schoen, D., and D. G. Lloyd. 1992. Self and cross-fertilization in plants. II. The selection of self-fertilization. International Journal of Plant Sciences 153:370-380.

Sinclair. W. A., H. H. Lyon, and W. T Johnson. 1987. Diseases of trees and shrubs. Cornell University Press, Ithaca, New York, USA.

Stephenson, A. G. 1981. Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12:253-279.

Stephenson, A. G., and R. I. Bertin. 1983. Male competition, female choice, and sexual selection in plants. Pages 109149 in L. A. Real, editor. Pollination biology. Academic Press, Orlando, Florida, USA.

Urata, U. 1954. Pollination requirements of macadamia. Hawaii Agricultural Experimental Station Technical Bulletin No. 22, Honolulu, Hawaii, USA.

Watson, M. A., and Casper B. B. 1984. Morphogenetic constraints on patterns of carbon distribution in plants. Annual Review of Ecology and Systematics 15:233-258.

Willson, M. F., and N. Burley. 1983. Mate choice in plants: tactics, mechanisms and consequences. Princeton University Press, Princeton, New Jersey, USA.
COPYRIGHT 1998 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Levri, Maureen A.; Real, Leslie A.
Publication:Ecology
Date:Jul 1, 1998
Words:4471
Previous Article:Maintenance of diversity within plant communities: soil pathogens as agents of negative feedback.
Next Article:Floral display, pollinator discrimination, and female reproductive success in two monoecious Begonia species.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |