| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2Department of Biological Science, California State University, Fullerton, California 92834; and 3Department of Ecology and Evolutionary Biology and the Department of Botany, University of Tennessee, Knoxville, Tennessee 37996
Received for publication December 18, 1997. Accepted for publication June 16, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Fabaceae • floral display • Lotus scoparius ssp. scoparius • pollinator visitation • postpollination changes • reproductive success • self-pollination
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
for a review of the subject). Such floral modifications result in the flowers becoming less conspicuous and/or less accessible to pollination vectors. Pollinators typically do not visit older phase flowers, which generally lack nectar and pollen, but the retention of these flowers has been shown to increase the overall floral display and to attract more pollinators (Gori, 1983
; Cruzan, Neal, and Willson, 1986
; Weiss, 1991
; but see Casper and LaPine, 1984
). However, the expected increase in the reproductive success of plants as a result of old-phase flower retention has been more difficult to demonstrate (but see Cruzan, Neal, and Willson, 1986
).
The first concise hypothesis regarding the potential biological significance of postpollination floral changes was formulated by Fritz Müller in a letter to Charles Darwin (Müller, 1877
; see also Weiss, 1991
). In discussing the floral color progression in Lantana camara from yellow to orange to purple that occurs over a 3-d period, he noted that a number of species of butterflies primarily visited the yellow flowers, while visiting the orange flowers less frequently and never visiting the purple flowers. He stated that according to Federico Delpino (1870) such color changes "would serve to show the visiting insects the proper moment for effecting the fertilization of these flowers" (Müller, 1877
, p. 79). Müller then expanded this hypothesis by saying (in reference to the Lantana–butterfly interaction): "If the flowers fell off at the end of the first day the inflorescence would be much less conspicuous; if they did not change colour much time would be lost by the butterflies inserting their proboscis in already fertilized flowers" (Müller, 1877
, p. 79). In more modern terms (Gori, 1983
), Müller's hypothesis would suggest that these floral changes increase the reproductive success of plants without inflicting additional energy costs on pollinators.
The present study was initiated to gather further data to test Müller's hypothesis by investigating the consequences of floral changes that occur in deer weed, Lotus scoparius (Nutt. in T. & G.) Ottley spp. scoparius. A preliminary study by Jones and Cruzan (1982)
concluded that pollinator visitation rates were increased as a result of the retention of old-phase flowers in this species. In this study, we use manipulations of flowers to (1) examine extent of the contribution of old-phase flowers to the overall floral display, and (2) determine the consequences of old-phase flower retention for reproductive success. Using our study, we also discuss the consequences of the presence of old-phase flowers on patterns of pollen dispersal within and among plants and the evolution of this phenomenon.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Deer weed is pollinated by a variety of native solitary and social bees, as well as the introduced honey bee, Apis mellifera. Individual flowers are small and initially yellow. When a bee alights on a flower, the keel is forced downward causing the stigma to contact pollen located on the abdomen of the bee and results in pollen being forced out of the tip of the keel by the movement of the staminal filaments. Pollen is extended in a ribbon-like sticky mass that adheres to the ventral surface of the pollinator, as is common for other species of Lotus (Stephenson et al., 1995
). Following pollination, or eventually as a result of the aging process, the upper petal (banner petal) turns orange and folds down over the keel portion of the flower that includes the stamens and carpel (Jones and Cruzan, 1982
). In addition, we have found that pollinator rewards decrease as the morphological changes in flower color and banner position occurred; nectar declines to nearly zero by the time the flowers have turned orange and the banner petal has completely folded down over the keel. Pollen also disappears as the flowers change, and by the end of the process little or no pollen remains in the anther sacs (Jones and Cruzan, 1982
). Our study population of deer weed was located on the property of the Robert Diemer Water Filtration Plant, in Yorba Linda, Orange County, California. Flowering begins at this site in early April and continues through May or early June. The most common visitors to deer weed in this area were the introduced honey bee, Apis mellifera, and the native bee, Anthidium collectum. Four other insect visitors, Bombus californicus, Bombus sonorus, Osmia bruneri, and Hoplitis sambuci, were observed much less frequently. Field work was conducted from April through August of both 1982 and 1983.
Self-compatibility tests
The numbers of yellow and orange flowers were counted on 20 plants, and the average ratio of yellow to orange flowers was determined. Three entire plants were bagged (using Ward's pollinator exclusion bags, no longer available) before any buds had opened and 105 tagged buds were left untreated to determine whether a pollinator was necessary to effect fruit set. One branch on each of four plants was bagged and, as they opened, five flowers per branch were hand-pollinated with pollen from the same plant and then rebagged to determine whether deer weed was self-compatible.
Pollinator behavior and fruit set experiments
We conducted several experiments to determine bee behavior in relation to the floral changes observed in deer weed. Observations were first made to see whether the pollinators discriminated against the older phase (orange) flowers as has been shown in studies of other species (Gori, 1983
; Casper and LaPine, 1984
; Cruzan, Neal, and Willson, 1986
; Weiss, 1991
). We made individual bee observations periodically throughout the blooming period, May through June, in 1982 and tested whether bee visitation to the flowers of deer weed is independent of flower color. To investigate the role of orange flowers in pollinator attraction, we manipulated the floral display of plants in the following manner. In the first year, four pairs of adjacent plants that were approximately the same size and had equal numbers of flowers were selected. One plant in each of these pairs had all orange banner petals removed daily (while leaving the rest of the flower intact) over the entire blooming period and the other plant in the pair was left unmanipulated. At the beginning and end of the blooming season, one person could remove the petals in 
45 min. During the peak of blooming, teams worked at this for >3 h each day. A total of 105 person-hours were spent in petal removal. On 8 d throughout the flowering season, we observed the number of pollinator visits in an hour to each member of the pairs. This was done to determine whether the frequency of bee visitations is independent of the presence of orange flowers on the plants. These same paired plants were also used to determine whether fruit set is independent of orange color presentation. Each day during the entire blooming period the number of new flowers produced was recorded and summed over the season for each plant. When fruits were mature, all paired plants were harvested, placed in large plastic garbage bags, and returned to the laboratory where we counted the number of fruits produced per plant. We then calculated a percentage fruit set by dividing the total number of fruit harvested by the total number of flowers produced and used a paired t test to test for a difference between the mean fruit sets of the treatments.
The following year, another set of pairwise comparisons was set up to test whether the frequency of bee visits is dependent on the total number of flowers per plant irrespective of the color of those flowers. Three pairs of plants were used, and treated in the following way over the entire blooming period. One plant in each pair had all orange banner petals removed each day so that the floral color presentation was all yellow. From its mate we removed banner petals daily from both yellow and orange flowers and removed entire flowers as needed (leaving developing ovaries intact) to reduce the total floral display to the same size as the plant with only yellow flowers while maintaining the average 36:64 ratio of yellow to orange flowers that was found in unmanipulated plants. This treatment required considerably more time than in the previous year because of the care taken to equalize the numbers of flowers on each plant of the pairs. In total, 220 person-hours were spent on this task. Pollinator visitation observations were increased to 32 h, made periodically throughout the entire blooming period, and the differences in visitation frequency and residency times on flowers and whole plants were tested using a chi-square test. Fruit set for each plant was determined using the techniques described above at the end of the flowering season.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bee behavior experimental results
All six species of bees examined discriminated against the orange flowers of deer weed (Table 1), but varied somewhat in their level of discrimination between the two flower types. For all pollinator species, visits to orange flowers were rare, ranging from as high as 2.1% for Apis to <0.5% for Osmia (Table 1). The null hypothesis that bee visitation to the flowers was independent of flower coloration was therefore rejected.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Therefore, pollination triggers the morphological changes much sooner than the normal aging process. It is assumed that there is a cost to the plant for the maintenance of these older corollas due to water loss and the carbohydrates necessary to sustain the tissues. Our data demonstrate that retaining spent flowers is also beneficial to the plant through increases in total seed production, potential increases in the amount of pollen exported to other plants, and potential reduction in the level of geitonogamy.
Fruit set in deer weed is significantly increased on plants with unmanipulated older orange flowers when compared to experimental plants that had banner petals on older flowers removed. Increases in female reproductive success due to the retention of orange flowers have not been found for other species studied (Gori, 1983
; Casper and LaPine, 1984
); however, Cruzan, Neal, and Willson (1986)
suggested that this effect may be seen in Phyla incisa when pollinators are limiting to fruit production. Since there are only one or two seeds per legume in Lotus scoparius, a single pollinator visit to a flower is probably sufficient to achieve maximum seed production. These results, coupled with the fact that a pollinator is necessary to effect fruit set in deer weed, suggest that fruit set in this population of Lotus scoparius is pollinator limited.
The presence of orange flowers significantly increases the frequency of pollinator visitations to deer weed, which may have resulted in increased fitness through male reproductive success. It is likely that increases in the visitation rate result in an increased amount of pollen being exported to neighboring plants, particularly if there is a pollen-dispensing mechanism that limits the amount of pollen removed at each floral visit (Harder and Thomson, 1989
; Stephenson et al., 1995
). In deer weed, pollen extrusion from the tip of the keel appears to act as such a mechnism since small amounts of pollen are dispensed for several repititions each time the flower is manipulated (personal observation). Although we did not directly measure pollen export, Gori (1983)
suggested that old-phase flower retention may result in increased male reproductive success in Lupinus argenteus, and Cruzan, Neal, and Willson (1986)
demonstrated increased pollen removal for plants with larger numbers of old-phase flowers in Phyla incisa. However, increased visitation rates as a result of the retention of older flowers have not always been found (e.g., Casper and LaPine, 1984
). In this study, we found that deer weed pollinators make long-distance foraging decisions based on the total floral display and not simply on basis of the number of yellow flowers present on the plant.
Our manipulations of floral display composition provide a comparison of the consequences of flower overproduction (Sutherland and Delph, 1984
) and flower-retention strategies for increasing floral display size. While the total number of pollinators visiting plants and the average fruit production were similar when the numbers of flowers in the paired plants was equalized, the residency times of pollinators on individuals of deer weed having only yellow flowers was significantly higher than on those having both yellow and orange flowers. The time spent per flower was equivalent for these two treatments, but pollinators visit more flowers per plant when only yellow flowers are present. The degree of self-pollination would therefore be much greater in the plants with only yellow flowers, since any outcross pollen is probably deposited on the first few flowers visited due to limited pollen carryover (Thomson and Plowright, 1980
; Thomson and Thomson, 1989
). This suggests that the amount of self-pollination would be increased if plants increased their floral display through the overproduction of new flowers rather than the retention of older flowers (Harder and Barrett, 1995
). The retention of orange flowers, when compared to increasing the floral display through overproduction of new flowers, appears to decrease the frequency of self pollination and the potential consequences of inbreeding depression for progeny vigor (Charlesworth and Charlesworth, 1987
). Further studies on selfing frequencies in L. scoparius and on the effects of selfing on progeny vigor will increase our understanding of the factors responsible for old-phase flower retention in this and other species.
Bees also benefit from the presence of orange flowers on deer weed. Once the plant has been identified as in bloom, time and energy are saved by the bee because of its ability to use near-goal orientation cues (von Frisch, 1967
) to distinguish rewarding from nonrewarding flowers. Since overall floral display tends to be correlated with the number of rewarding flowers, inflorescence size is a good predictor of resource availability for pollinators (Cruzan, Neal, and Willson, 1986
). Pollinators making foraging decisions based on the size of the floral display should not experience a decrease in foraging efficiency while foraging on these plants, since bees apparently quickly learn to discriminate between rewarding and nonrewarding flowers.
We have demonstrated that the presence of orange flowers on plants contributes to the attraction of pollinators of deer weed and that these higher visitation frequencies increase the fruit production and probably the pollen dispersal of these plants. We therefore conclude that Müller's hypothesis cannot be rejected and that both male and female reproductive success are significantly increased as a result of old-phase flower retention in Lotus scoparius.
Selective pressures for the evolution of old-phase flower retention
Increased visitation rates to deer weed plants having both yellow and orange flowers represent a parallel situation to that found in Phyla by Cruzan, Neal, and Willson (1986)
. In both studies, bees tended to choose larger floral displays. It has been long noted (Darwin, 1896
) that there is often a low fruit set in flowering hermaphroditic plants. Several explanations have been proposed for the discrepancy between the number of flowers a plant produces and the number of fruits actually set (see, e.g., Stephenson, 1981
; Sutherland and Delph, 1984
; Sutherland, 1986
). One of the proposed hypotheses states that an overproduction of flowers would ensure that the pollinator visitation rate will be sufficient for adequate pollination (Willson and Rathcke, 1974
; Willson and Price, 1977
; Schaffer and Schaffer, 1979
; Stephenson, 1980
). Increasing the size of the floral display increases the visitation rate, but the constant production of more new flowers to maintain a large display is energetically costly, especially in plants that live in stressed (i.e., resource limited) environments (Grime, 1977
, 1979
). Plants living in such habitats (e.g., Lotus scoparius) could achieve the same result, however, by simply maintaining spent flowers on the plant. With the addition of morphological modifications to these flowers, pollinators would learn to differentiate them from rewarding flowers in near-goal orientation situations. These morphological modifications need not be pollination induced to be functional. However, pollination-induced changes may be more advantageous to plants when pollinators are limiting to fruit set or are unpredictable in time (see Gori [1983]
for a review of the possible temporal patterns involved in adaptive floral morphological changes). Our suggestion that spent flower retention is more likely to evolve in stressed environments is supported by a review of the extensive list of such plants in Gori (1983)
and by a review of California xerophytes that exhibit this tendency (Jones and Cruzan, 1988
). Nearly all of these species fit into the categories of stress-tolerant ruderals or stress-tolerant competitors on Grime's (1979)
triangular model of life history variation.
Most flowering plant species are dependent upon the services of pollinators for pollen receipt and dispersal. Since the attraction of pollinators is closely linked to a plant's reproductive success, any mechanism by which plants can increase their pollinator visitation rate should be favored. In addition to spent flower retention, plants are known to increase their floral display through several mechanisms, including colored bracts and sepals (Faegri and van der Pijl, 1979
) and the overproduction of flowers (Stephenson, 1980
). One of the major selective pressures for retention of spent flowers on plants living in resource-stressed environments may be the necessity of having an energetically efficient means to achieve adequate pollination by increasing the floral display and providing a larger signal to attract more pollinators. This would be particularly important for plants that must have an outside vector to effect their pollination as in the case of deer weed. Our suggestion may help explain the relatively high incidence of spent flower retention that occurs in desert and other seasonally dry, resource limited habitats.
Conclusions
Many plant species from a variety of stressed, resource-limited habitats exhibit some form of sequential morphological changes that occur after floral anthesis and prior to corolla abscission. Our data reveal that deer weed increases its reproductive success through increased fruit production and increased pollinator visitation rates by retaining orange, nonrewarding corollas. In addition to increasing the pollinator visitation rate, retention of spent flowers may decrease the relative degree of selfing and consequently increase the genetic diversity of pollen received when compared to flower overproduction as a mechanism of increasing the floral display. Bee pollinators are able to easily and quickly distinguish between yellow (rewarding) and orange (nonrewarding) flowers in deer weed and therefore save both time and energy. We suggest that one reason sequential floral changes are so common in stressed habitats is that it allows plants to increase their floral display, and thus improve their ability to attract pollinators, and acts as a resource-efficient alternative to the overproduction of flowers that is common in other species living in less stressful habitats.
Fritz Müller was right over 100 years ago!
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Charlesworth, D., and B. Charlesworth. 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237–268.[CrossRef][ISI]
Cruzan, M. B., P. R. Neal, and M. F. Willson. 1986 Floral display in Phyla incisa: consequences for male and female reproductive success. Evolution 42: 505–515.[CrossRef]
Darwin, C. 1896 The different forms of flowers on plants of the same species. D. Appleton, New York, NY.
Faegri, K., and L. van der Pijl. 1979 The principles of pollination ecology. Pergamon Press, Oxford.
Gori, D. F. 1983 Post-pollination phenomena and adaptive floral changes. Chapter 2. In C. E. Jones and R. J. Little [eds.], Handbook of experimental pollination biology, 31–45. Scientific and Academic Editions, Division of Van Nostrand Reinhold, New York, NY.
Grime, J. P. 1977 Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111: 1169–1194.[CrossRef][ISI]
———. 1979 Plant strategies and vegetation processes. John Wiley and Sons, New York, NY.
Harder, L. D., and J. D. Thomson. 1989 Evolutionary options for maximizing pollen dispersal of animal-pollinated plants. American Naturalist 133: 323–344.[CrossRef][ISI]
———, and S. C. H. Barrett. 1995 Mating cost of large floral displays in hermaphrodite plants. Nature 373: 512–515.[CrossRef]
Jones, C. E., and M. B. Cruzan. 1982 Floral color changes in deer weed (Lotus scoparius): possible function. Crossosoma 8:1–6.
———, and ———. 1988 Adaptive significance of time dependent, pollination related floral changes in California xerophytes. In R. G. Zahary [ed.], Desert ecology 1986: a research symposium. Southern California Academy of Sciences and Southern California Desert Studies Consortium Special Publication, 124–139. Los Angeles, CA.
Müller, F. 1877 Letters to the editor. Nature 17: 78–79.
Munz, P. A. 1974 A flora of southern California. University of California Press, Berkeley, CA.
Schaffer, W. M., and M. V. Schaffer. 1979 The adaptive significance of variations in reproductive habit in the Agavaceae. II. Pollinator foraging behavior and selection for increased reproductive expenditure. Ecology 60: 1051–1069.[CrossRef][ISI]
Stephenson, A. G. 1980 Fruit set, herbivory, fruit reduction, and the fruiting strategy of Catalpa speciosa (Bignoniaceae). Ecology 61: 57–64.[CrossRef][ISI]
———. 1981 Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12: 253–279.
———, M. R. Quesada, C. D. Schlichting, and J. A. Winsor. 1995 Consequences of variation in pollen load size. Monographs in Systematic Botany 53: 233–244.
Sutherland, S. 1986 Patterns of fruit-set: what controls fruit-flower ratios in plants? Evolution 40: 117–128.[CrossRef][ISI]
———, and L. F. Delph. 1984 On the importance of male-fitness in plants: patterns of fruit set. Ecology 65: 1093–1104.[CrossRef][ISI]
Thomson, J. D., and R. C. Plowright. 1980 Pollen carryover, nectar rewards, and pollinator behavior with special reference to Diervilla lonicera. Oecologia 46: 68–74.[CrossRef][ISI]
———, and B. A. Thomson. 1989 Dispersal of Erythronium grandiflorum pollen by bumblebees: implications for gene flow and reproductive success. Evolution 43: 657–661.[CrossRef][ISI]
von Frisch, K. 1967 The dance language and orientation of bees (translated by L. F. Chadwick). Belknap Press of Harvard University Press, Cambridge, MA.
Weiss, M. R. 1991 Floral colour changes as cues for pollinators. Nature 354: 227–229.[CrossRef][ISI]
Willson, M. F., and P. W. Price. 1977 The evolution of inflorescence size in Asclepias (Asclepiadaceae). Evolution 31: 495–511.[CrossRef][ISI]
———, and B. J. Rathcke. 1974 Adaptive design of the floral display in Asclepias syriaca L. American Midland Naturalist 92: 47–57.[CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  E. R. Valdivia, D. J. Cosgrove, and A. G. Stephenson Role of accelerated style senescence in pathogen defense Am. J. Botany, November 1, 2006; 93(11): 1725 - 1729. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Dohzono, K. Suzuki, and J. Murata Temporal changes in calyx tube length of Clematis stans (Ranunculaceae): a strategy for pollination by two bumble bee species with different proboscis lengths Am. J. Botany, December 1, 2004; 91(12): 2051 - 2059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Montalvo and N. C. Ellstrand Nonlocal transplantation and outbreeding depression in the subshrub Lotus scoparius (Fabaceae) Am. J. Botany, February 1, 2001; 88(2): 258 - 269. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
