| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biological Sciences, University of Texas at Brownsville, 80 Ft. Brown, Brownsville, Texas 78520 USA
Received for publication October 9, 2001. Accepted for publication January 3, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Apoidaea • Penstemon • Scrophulariaceae • size-dependent selection • staminode
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. As a result of visiting flowers, pollinators deposit pollen on stigmas of flowers and also transfer pollen to stigmas of other flowers. Often it is assumed that various floral features, such as corolla diameter, influence the behavior of pollinators and drive the mechanics of pollen deposition and/or transfer (Galen, 1989
). Many features or traits of flowers are considered adaptive and evolving under the force of natural selection (Waser, 1983
; Galen, 1996
; but see Herrera, 1996
for a counter argument). The adaptiveness or "selective advantage" of any floral trait is directly related to the trait's ability to enhance production of offspring or seed. Production of offspring in hermaphroditic plants includes not only seed that it produces (female reproductive success) but also seed sired on other plants through the distribution of pollen to flowers on other plants (male reproductive success; Snow and Lewis, 1993
).
In the Scrophulariaceae, flowers typically possess only four functional stamens, but in approximately 30% of the genera, a fifth sterile stamen, the staminode, is present (Polak, 1900
). Generally, the staminode is a small, simple structure, but in the genus Penstemon the staminode is larger than the fertile stamens and frequently bears bristles on the upper surface of its distal end, therefore the common name, beardtongues. Staminodia are considered to play a role in the pollination biology of many plant species including Penstemon (Straw, 1956
; Torchio, 1974
; Endress, 1984
; Young, 1984
; Atwood, 1985
; Banziger, 1996
; Dieringer and Cabrera R., 2001
; Walker-Larsen and Harder, 2001
). For Penstemon, Straw (1956)
proposed two functions for the staminode. He stated (p. 117), "...they appear to cause pollinating insects to keep their bodies high enough to insure pollination, and they impose a barrier to smaller insects that might otherwise attempt to reach the essential organs...." The phylogeny, evolution, and possible functions of stamondia in general have recently been reviewed by Walker-Larsen and Harder (2000)
. Yet, few experimental studies have been conducted to test how the staminode, as a floral trait, influences components of male and female reproductive success or specifically interacts with visiting pollinators (but see Dieringer and Cabrera R., 2001
; Walker-Larsen and Harder, 2001
).
Penstemon digitalis possesses a staminode with many large bristles on the upper surface of its distal end. The staminode is positioned on the lower side of the corolla throat, and the bristles partially obstruct the entrance of the flower. Its large size, relative to the corolla throat, and its location combine to cause an interaction with visiting bees. Staminode removal experiments by Dieringer and Cabrera R. (2001)
have demonstrated a significant effect on seed production. Open-pollinated flowers without staminodia produced 37% fewer seeds/fruit than unmanipulated flowers.
Flowers of Penstemon digitalis are visited by a diversity of bees varying greatly in size. Visiting bees range in size from the large Bombus spp. to small Ceratina spp. (Clinebell and Bernhardt, 1998
; Dieringer and Cabrera R., 2001
; this study). Bees of differing size are expected to vary in their interaction with the staminode and resultant pollen deposition on the stigma and/or removal from the anthers. Penstemon digitalis, therefore, offers an excellent opportunity to conduct experimental studies on the interactions between the bristle staminode and pollinating bees of varying size in terms of reproductive success. Our data indicate that bristle staminode presence and function are influenced by size-dependent selection on bee body size and associated pollen transporting attributes.
During the summers of 1995 and 1996 a study on the reproductive ecology of the bristle staminode of Penstemon digitalis was undertaken and included: (1) a detailed description of the floral biology; (2) classification of visiting bees into size categories; (3) estimation of flower visitation rates and flower visit duration of foraging bees; (4) estimation of the relationship between pollen deposition and seed production; (5) a staminode removal experiment to test the reproductive function of the bristle staminode from a male and female perspective; and (6) a staminode removal experiment to access the effects of differing bee sizes on components of male and female fitness.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Throughout the study, flower-visiting bees were collected, mounted, and identified. For each bee species, flower visitation rates were given a qualitative measure based on relative frequency of visitation observed over the 2-yr study. Values ranged from 1 for most frequent visitors to 5 for least frequent. The duration of each flower visit was estimated by observing individual flower visits and measuring the time spent foraging on the flower using a stop-watch accurate to 0.01 s. The duration of each visit was measured as the time between alighting to departure from the flower.
Sugar content of the nectar was determined at anthesis using thin-layer chromatography (TLC) (Stahl, 1969
). Ultraviolet (UV) reflectance patterns of flowers were examined by photographing flowers in sunlight using a Wratten 18A UV filter and TMAX 400 film (both from Kodak, Rochester, New York, USA) (Cabrera R. and Dieringer, 1992
; Dieringer and Cabrera R., 1994
). Several flowers were tagged and the floral life span and development of sexual phases noted. Floral life span was recorded as the time from anthesis to corolla loss. Male and female sexual phases were recorded as time of anther dehiscence and downward curving of style and stigma. The downward curving of the style and stigma is indicative of stigma receptivity and characteristic of the genus (Straw, 1956
).
Reproductive characteristics for the population were also recorded. The mean number of flowers and fruits produced per plant was estimated from 20 randomly tagged plants, same plant for each character. We estimated the number of pollen grains produced per unvisited flower and the number of seeds produced per open-pollinated fruit from an additional 21–25 randomly collected flowers or fruits, respectively, one per plant. The amount of pollen present in each flower was determined by squashing all the anthers of a flower in 0.5 mL of alcohol containing detergent. The number of grains in three aliquots per flower was counted using a hemacytometer, and the total number of grains per flower was estimated by accounting for dilution (Kearns and Inouye, 1993
; Dieringer and Cabrera R., 1994
).
Previous studies on the breeding system, determined from a series of experimental hand-crosses, found no autogamy, indicating that pollinators are required for fruit and seed production (Dieringer and Cabrera R., 2001
). A substantial amount of self-compatibility occurs in the population, e.g., of 22 self-pollinated flowers, 10 set fruit for an estimated 46% self-compatibility.
Previous studies indicated the staminode significantly influenced seed production (Dieringer and Cabrera R., 2001
). We decided that pollen deposited on stigmas would be a preferable variable to measure since premature excision of the style can negatively affect seed production data.
In 1995, the relationship between pollen deposition and resultant seed production was estimated using single bee visits to virgin flowers (Spears, 1983
; Dieringer, 1992
). Buds were bagged with fine mesh nylon prior to anthesis. The following day, bags were removed from flowers with recurved, receptive stigmas for pollinator visitation. A single, xenogamous visit was then permitted to occur. Immediately following a visit, flowers were tagged and rebagged until corolla drop and plants were tagged and numbered with flagging tape and flowers marked with colored thread. Bees used included: Bombus pennsylvanicus, B. impatiens, and Xylocopa virginica. A total of 22 single-visit crosses were performed. Following corolla loss (usually day 3), the turgid stigmas were carefully excised with a razor blade and fixed in alcohol. Fruits were allowed to mature in the field, then collected, and the total number of seeds counted.
In 1996, staminode removal experiments were conducted to test the effect staminodia have on components of female and male reproductive success, pollen deposition, and removal, respectively. Two sets of experiments were performed. The first experiment involved comparing pollen deposition and removal between flowers with and without staminodia that received open-pollination over their life span. Flowers buds were bagged and the morning of anthesis the bag was removed. Staminodia were removed from approximately half of these flowers and the other flowers functioned as the control (N = 17–34/treatment; 1 flower/plant). Following open-pollination, anthers and stigmas were collected and fixed in alcohol. Data from this first experiment were analyzed using t tests.
The second experiment compared pollen deposition and removal for flowers with and without staminodia that had received a single visit. Flowers were again bagged as buds and unbagged the morning of anthesis. Visiting bees were categorized into three size classes (Table 1). Sample sizes for each size category ranged from 20 to 36 for manipulated flowers (1 flower/plant) that had the staminode removed and from 19 to 37 for unmanipulated flowers (1 flower/plant) that had the staminode left intact as a control. Following the single visit, anthers and stigmas were collected and fixed in alcohol. For this second experiment, pollen deposition and pollen removal data were analyzed using a two-way ANOVA with treatment (staminode presence or removal) and bee size category (small, medium, and large) as main effects plus an interaction term. It is the interaction term that provides insight into how bees affect selection on the bristle staminode (Wilson and Thomson, 1996
). This second experiment specifically tested whether the staminode interacts with all bees equally or varies with bee size.
|
Pollen removed from visited flowers was estimated using the difference between residual pollen within the anthers of the visited flower and mean pollen production of unvisited flowers. The amount of pollen present in each flower was determined by squashing all the anthers of a flower in 0.5 mL of alcohol containing detergent. The number of grains in 3 aliquots/flower was counted using a hemacytometer, and the total number of grains per flower was estimated by accounting for dilution. Flower buds from 25 nearly mature buds (1 bud/plant) were collected and fixed in alcohol for calculation of pollen production from unvisited flowers.
For comparative purposes, the fate of pollen was measured in a pair of ratios following terminology outlined by Inouye et al. (1994)
. First we defined a stigma deposition ratio (S/W) where S equals the number of pollen grains deposited on a stigma from a single bee visit and W equals the mean number of seeds produced per fruit from flowers receiving open-pollination for a flower's life span. Second, we defined a pollen removal ratio (L/p) where L equals the number of pollen grains removed from a flower by a single bee visit and p equals the mean total amount of pollen produced by a virgin flower. For our experiments, these ratios represent pollen deposition and removal efficiencies, respectively (Inouye et al., 1994
). These ratios were then compared between flowers with and without staminodia. All statistical tests followed guidelines presented by Sokal and Rohlf (1981)
. Statistical tests were performed using the software package SAS (SAS Institute, 1985
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Small and medium bees were far more frequent visitors to flowers of Penstemon digitalis than large bees (Table 1). The most frequent bees were Hoplitis pilosifrons and Anthophora terminalis. Small and medium bees also foraged for a longer period of time on flowers than did large bees (Table 1). The longest foraging duration per flower visit for any bee species averaged over 1 min for Ceratina calcarata while the shortest averaged just over 1.5 s for A. terminalis.
Flowers produced nectar containing sucrose, fructose, and glucose as an attractant, although sucrose was the dominant sugar based on strength of the developed TLC spot. Xylocopa foraged for nectar by perforating the corolla base and, on occasion, Hoplitis was observed to forage for nectar through these same perforations. All other visiting bees, at some point, were observed to probe the corolla base, presumably foraging for nectar and were also observed to forage for pollen. Upon entering the flower, bees would pass directly over the staminode bristles. Both Ceratina and Osmia would frequently land on the labellum, climb onto the staminode bristles, and pollen forage by actively scraping the anthers, contacting the stigma inadvertently. Hoplitis entered the flower upright when foraging for nectar, but would the invert when foraging for pollen by possibly vibrating and bouncing the anthers. All other visiting bees groomed pollen from their bodies. Anthophora terminalis was the only bee that, at times, would approach but then reject a flower lacking a staminode. All other bees showed no obvious behavioral differences when foraging on flowers with the staminode removed.
The flowers that were white and had purple nectar guides were generally viable for 2–3 d based on the day of corolla drop, which was strongly dependent on ambient temperature. Ultraviolet reflectance photography showed the flowers were highly UV reflective except for the purple nectar guides. On the first day of anthesis, the anthers dehisced and the stigma was horizontally positioned while on the second day the stigma elongated and curved downward. The stigma remained in the curved position for days 2 and 3 and was considered to be receptive during this time.
Reproductive characteristics for the population are presented in Table 2. Plants produced a mean of 139.7 flowers, 74.4 fruits, 34 581.7 pollen grains/flower, and 60.6 seeds/fruit. These values were used in the calculation of S/W and L/p ratios presented below.
|
|
The staminode removal experiment for open-pollinated flowers showed that, during a flower's life span, only pollen deposition on stigmas was significantly affected (Table 3). Flowers without staminodia received about half as many grains on their stigmas than unmanipulated flowers. No difference was found between treatments for pollen removal, with approximately 99% of the pollen being removed from the anthers over a flower's life span (Tables 2 and 3).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Studies on Penstemon digitalis in Missouri recorded 15 taxa of bees and 1 taxon of wasp (Clinebell and Bernhardt, 1998
) visiting the flowers. In these Missouri populations, bumble bees were the predominant pollinators possibly reflecting differences between site diversity, area, and restorative efforts that support a greater diversity of pollinating bees. Mitchell and Ankeny (2001)
recorded halictid bees as the primary visitors to Ohio populations of P. digitalis. In studies on Penstemon in California, P. ellipticus was pollinated solely by Bombus and P. palmeri by Xylocopa, Bombus, and the megachilid Callanthidium, (Walker-Larsen and Harder, 2001
). Callanthidium, a medium bee, was the primary pollinator.
Breeding system studies in our study population clearly indicate that Penstemon digitalis is self-compatible and not autogamous (Dieringer and Cabrera R., 2001
). This explains the strong relationship between pollen deposition and seed production shown in Fig. 1 and justifies the use of pollen deposition on stigmas as a correlate of seed production. Interestingly, breeding system studies in Missouri populations displayed high levels of autogamy (Clinebell and Bernhardt, 1998
). Self-compatibility has also been demonstrated for P. tenuiflorus, P. hirsutus, P. pseudospectabilis, and P. penlandii and autogamy, to some degree, for populations of P. grandiflorus, P. tenuiflorus, P. hirsutus, and P. pseudospectabilis (Clinebell and Bernhardt, 1998
; Clements, Baskin, and Baskin, 1999
; Lange and Scott, 1999
; Tepedino, Sipes, and Griswold, 1999
).
Staminode function
Experimental staminode removal clearly demonstrated a significant role for the staminode over the flower's life span. The significant effect for pollen deposition confirms one of Straw's hypotheses (1956)
. Flowers with staminodia received 46% more pollen than flowers with stamondia removed. These results are consistent with those of seed production from previous experiments performed by Dieringer and Cabrera R. (2001)
and with studies on P. palmeri and P. ellipticus by Walker-Larsen and Harder (2001)
.
Our study detected no significant effect for total pollen removed. In the study by Walker-Larsen and Harder (2001)
, however, pollen removal differed between flowers with and without staminodia for Penstemon ellipticus but not for P. palmeri. The difference between species in pollen removal was attributed to the staminode's differing modes of action between the species. For P. ellipticus, the staminode is thought to function by restricting access to nectaries and its removal allows bees to move deeper into the corolla and changes their contact with the anthers.
For P. palmeri, the staminode is thought to act more as a lever or trigger mechanism causing the style to descend onto the bee's body. This mechanism seems fine for large and possibly medium bees with a mass sufficient to push the staminode downward, but does not explain the staminode's effect on small bees detected in this study. For both P. digitalis and Jacaranda acutifolia (Bignoniaceae), the stigma and anthers are placed in close proximity to one another with the staminode bristles lying directly below them (Dieringer and Cabrera R., 2001
). The bristles are critically important in keeping small pollen-foraging bees closer to the stigma, resulting in more frequent contact and pollen transfer. The precise arrangement of these structures indicates that placement of the bristles is important. This suggests that the filament portion of the staminode has one function and the bristle portion another. Interestingly, Penstemon ellipticus is pollinated solely by large bees, such as Bombus melanopygus, and possesses a staminode with only a few bristles at its distal end (Walker-Larsen and Harder, 2001
).
Studies by Wilson and Thomson (1991)
on Impatiens capensis (Balsaminaceae) categorized bees by behavior as either nectar- or pollen-foraging. In their studies, pollen-collecting bees (Apis and Dialictus) removed large amounts of pollen per visit but deposited fewer grains on stigmas compared to nectar-feeding bees (Bombus). Our results are similar in that large, nectar-feeding bees removed less pollen per flower visit, but differ in that small, pollen-collecting bees deposited large amounts of pollen. The difference is most likely caused by the staminode, which Impatiens lacks.
Interaction between bee size and staminode presence
Our data indicate that bristle staminode presence and function are influenced by size-dependent selection on bee body size and associated pollen transporting attributes. Staminode removal caused the greatest change in pollen-deposition efficiency,S/W ratios, for small bees. With the staminode present, small bees were capable of depositing high pollen loads with 61% of the pollen needed for full seed production being deposited in a single visit compared to 20% when removed.
Staminode removal caused the greatest change in pollen-removal efficiency, L/p ratios, for medium bees. With the staminode present, medium bees removed less pollen per visit with only 38% of the flower's pollen removed in a single visit compared to 60% when removed. Staminode presence, therefore, interacts with small bees by enhancing female reproductive success and medium bees by enhancing male reproductive success.
Together, papers presented by Harder and Thomson (1989)
, Harder (1990)
, and Wilson et al. (1994)
, present a model of selection on floral traits based on visitation rate. Selection is expected to favor floral traits that ensure stigmas receive sufficient pollen in a single pollinator visit, or at least only a few, to maximize seed production during periods of either sparse or abundant visitation. A single visit during periods of sparse visitation would ensure full seed production while during periods of abundant visitation high pollen loads deposited on stigmas simultaneously or in close succession would allow for competition between pollen grains as pollen tubes grow through the style. In contrast, selection is expected to favor floral traits that cause pollen to be dispensed progressively over multiple visits to enhance the probability of siring seeds on other flowers and/or plants only during periods of abundant visitation. Much of the pollen removed from anthers during a bee visit is unavailable for deposition on future stigmas due to grooming behavior of the bee or because pollen is dropped or distributed in areas of the bee's body that are inaccessible for contact with the stigma. If most of a flower's pollen were removed in a single visit, few offspring would be sired for the amount of pollen produced. Visitation to P. digitalis flowers is typically high (5.7 visits/h, Dieringer and Cabrera R., unpublished data, 2.57 visits/h, Mitchell and Ankeny, 2001
) and conforms to the assumptions of the above model.
Visits by the medium bee Anthophora terminalis, also present in the Missouri populations, resulted in more pollen removal per visit than large bees, but much less when the staminode was present. Visits by large bees resulted in the least amount of pollen removed whether the staminode was present or not. As mentioned by Walker-Larsen and Harder (2001)
, removal of the staminode may allow medium bees to thrust deeper into the corolla causing more of their body surface to scrape across the anthers and release more pollen. We expect the staminode to be under greater selective pressure from medium bees to minimize pollen dispensing than from large bees, thereby enhancing male reproductive success.
Large bees deposited very little pollen per flower visit thus requiring multiple visits to a flower to produce a full complement of seeds, yet they also removed the fewest pollen grains per visit, which should enhance male reproductive success. However, given their low visitation rate to flowers and low pollen deposition per visit, it is unlikely that they are responsible for a large proportion of offspring produced thereby contributing little to total reproductive success in our population. Nevertheless, when abundant, as in the Missouri populations (Clinebell and Bernhardt, 1998
), or when they are the only pollinator, as in P. ellipticus, we expect large bees to exert a strong selective force on staminode presence.
Studies by Wilson and Thomson (1996)
also examined the interactions between bees of differing size and floral traits for Impatiens pallida and Erythronium grandiflorum. No significant interaction terms were found for three floral form measures in I. pallida, nor for stigma exertion in E. grandiflorum. The lack of significant interactions suggests that bees of differing sizes did not differ in the selection gradients imposed on the flower traits examined.
Clinebell and Bernhardt (1998)
, in a study on five Midwest Penstemon species, recognized two size categories of visiting bees, large and small. They suggested that selection favors the two-tiered system based on the nectar- and pollen-foraging behavior of large and small bees, respectively. In their study, they categorized Anthophora terminalis as a large bee. Here, we suggest that it is the staminode that selects for the two-tiered pollinator system based on size-dependent selection of bee body size and associated pollen-transporting attributes.
Overall, our experimental studies on Penstemon digitalis have provided valuable insight into staminode function and its interaction with pollinating bees. Staminode presence enhances reproductive success through both female and male function but depends on the sizes of pollinating bees. More detailed studies into the exact roles of the filament portion versus the bristle portion of the staminode are warranted.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Banziger H. 1996 The mesmerizing wart: the pollination strategy of epiphytic lady slipper orchid Paphiodedilum villoscum (Lindl.) Stein (Orchidaceae). Botanical Journal of the Linnean Society 121: 59-90[CrossRef]
Cabrera R. L. G. Dieringer 1992 Reproductive biology of a population of Acourtia runcinata (Asteraceae: Mutisieae) at the northeastern limit of its range. American Midland Naturalist 128: 83-88[CrossRef][ISI]
Clements R. K. J. M. Baskin C. C. Baskin 1999 The comparative biology of the two closely-related species Penstemon tenuiflorus Pennell and P. hirsutus (L.) Willd. (Scrophulariaceae, section Graciles): II. Reproductive biology. Castanea 64: 299-309
Clinebell II R. R. P. Bernhardt 1998 The pollination ecology of five species of Penstemon (Scrophulariaceae) in the tallgrass prairie. Annals of the Missouri Botanical Garden 85: 126-136[CrossRef][ISI]
Darwin C. 1862 On the various contrivances by which orchids are fertilised by insects. John Murray, London, UK
Dieringer G. 1992 Pollinator effectiveness and seed set in populations of Agalinis strictifolia (Scrophulariaceae). American Journal of Botany 79: 1018-1023[CrossRef][ISI]
Dieringer G. L. Cabrera R 1994 Sexual selection of anther trichomes and sexual dimorphism in Ibervillea lindheimeri (Cucurbitaceae: Melothrieae). American Journal of Botany 81: 111-118[CrossRef][ISI]
Dieringer G. L. Cabrera R 2001 Pollination ecology of bristle staminodia in Penstemon digitalis (Scrophulariaceae) and Jacaranda acutifolia (Bignoniaceae). Biotam 12: 31-36
Endress P. K. 1984 The role of inner staminodes in the floral display of some relic Magnoliales. Plant Sytematics and Evolution 146: 269-282[CrossRef]
Galen C. 1989 Measuring pollinator-mediated selection on morphometric floral traits bumblebees and the alpine sky pilot Polemonium viscosum. Evolution 43: 882-890[CrossRef][ISI]
Galen C. 1996 Rates of floral evolution: adaptation to bumblebee pollination in an alpine wildflower, Polemonium viscosum. Evolution 50: 120-125[CrossRef][ISI]
Harder L. D. 1990 Pollen removal by bumble bees and its implications for pollen dispersal. Ecology 71: 1110-1125[CrossRef][ISI]
Harder L. D. J. D. Thomson 1989 Evolutionary options for maximizing pollen dispersal of animal-pollinated plants. American Naturalist 133: 323-344[CrossRef][ISI]
Herrera C. M. 1996 Floral traits and plant adaptation to insect pollinators: a devil's advocate approach. In D. G. Lloyd and S. C. H. Barrett [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 65–87. Chapman and Hall, New York, New York, USA
Inouye D. W. D. E. Gill M. R. Dudash C. B. Fenster 1994 A model and lexicon for pollen fate. American Journal of Botany 81: 1517-1530[CrossRef][ISI]
Kearns C. A. D. W. Inouye 1993 Techniques for pollination biologists. University Press of Colorado, Colorado, USA
Lange R. S. P. E. Scott 1999 Hummingbird and bee pollination of Penstemon pseudospectabilis. Journal of the Torrey Botanical Society 126: 99-106[CrossRef][ISI]
Mitchell R. J. D. P. Ankeny 2001 Effects of local conspecific density on reproductive success in Penstemon digitalis and Hesperis matronalis. Ohio Journal of Science 101: 22-27[ISI]
Mitchell T. B. 1962 Bees of the eastern United States, vol. II. Technical Bulletin Number 152. North Carolina Agricultural Experiment Station, Raleigh, North Carolina, USA
Polack K. M. 1900 Untersuchungen über die Staminodien der Scrophulariaceen. Osterreichische Botanische Zeitschrift 50: 33-41, 87–90, 123–132, 164–167 [CrossRef]
SAS Institute. 1985 SAS user's guide: statistics, version 5.0. SAS Institute, Cary, North Carolina, USA
Snow A. A. P. O. Lewis 1993 Reproductive traits and male fertility in plants: empirical approaches. Annual Review of Ecology and Systematics 24: 331-351[ISI]
Sokal R. R. F. J. Rohlf 1981 Biometry, 2nd ed. Freeman, San Francisco, California, USA
Spears E. E., Jr. 1983 A direct measure of pollinator effectiveness. Oecologia 57: 196-199[CrossRef][ISI]
Straw R. M. 1956 Adaptive morphology of the Penstemon flower. Phytomorphology 6: 112-119
Stahl E. 1969 Thin-layer chromatography. A laboratory handbook. (English translation). Springer-Verlag, New York, New York, USA
Tepedino V. J. S. D. Sipes T. L. Griswold 1999 The reproductive biology and effective pollinators of the endangered beardtongue Penstemon penlandii (Scrophulariaceae). Plant Systematics and Evolution 219: 39-54[CrossRef][ISI]
Torchio P. F. 1974 Mechanisms involved in the pollination of Penstemon visited by the masarid wasp, Pseudomasaris vespoides (Cresson). Pan-Pacific Entomologist 50: 226-234
Walker-Larsen J. L. D. Harder 2000 The evolution of staminodes in angiosperms: patterns of stamen reduction, loss, and functional re-invention. American Journal of Botany 87: 1367-1384
Walker-Larsen J. L. D. Harder 2001 Vestigial organs as opportunities for functional innovation: the example of the Penstemon staminode. Evolution 55: 477-487[CrossRef][ISI][Medline]
Waser N. M. 1983 The adaptive nature of floral traits: ideas and evidence. In L. Real [ed.], Pollination biology, 241–285. Academic Press, Orlando, Florida, USA
Wilson P. J. D. Thomson 1991 Heterogeneity among floral visitors leads to discordance between removal and deposition of pollen. Ecology 72: 1503-1507[CrossRef][ISI]
Wilson P. J. D. Thomson 1996 How do flowers diverge?. In D. G. Lloyd and S. C. H. Barrett [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 88–111. Chapman and Hall, New York, New York, USA
Wilson P. J. D. Thomson M. L. Stanton L. P. Rigney 1994 Beyond floral batemania: gender biases in selection for pollination success. American Naturalist 143: 283-296[CrossRef][ISI]
Young A. M. 1984 Mechanism of pollination by Phoridae (Diptera) in some Herrania species (Sterculiaceae) in Costa Rica. Proceedings of the Entomological Society of Washington 86: 503-518[ISI]
This article has been cited by other articles:
|
|
 |
|
  E. Guimaraes, L. C. di Stasi, and R. d. C. S. Maimoni-Rodella Pollination Biology of Jacaranda oxyphylla with an Emphasis on Staminode Function Ann. Bot., November 1, 2008; 102(5): 699 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Zorn-Arnold and H. F. Howe Density and seed set in a self-compatible forb, Penstemon digitalis (Plantaginaceae), with multiple pollinators Am. J. Botany, October 1, 2007; 94(10): 1594 - 1602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Wolfe, C. P. Randle, S. L. Datwyler, J. J. Morawetz, N. Arguedas, and J. Diaz Phylogeny, taxonomic affinities, and biogeography of Penstemon (Plantaginaceae) based on ITS and cpDNA sequence data Am. J. Botany, November 1, 2006; 93(11): 1699 - 1713. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
