| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
0 Geobotanical Institute, Swiss Federal Institute of Technology (ETH) Zürich, Zürichbergstrasse 38,8044 Zürich, Switzerland
Received for publication December 10, 1998. Accepted for publication May 25, 1999.
ABSTRACT
The plant Euphorbia cyparissias is commonly infected by rust fungi of the species complex Uromyces pisi. When infected, E. cyparissias is unable to flower, but instead is induced by the fungus to form pseudoflowers. Pseudoflowers are rosettes of yellow leaves upon which the fungus presents its gametes in a sweet-smelling fungal nectar. We hypothesized that the fungi, as they are heterothallic, are dependent on insect visitation to cross-fertilize their mating types. We confirmed that insects are required with an insect exclusion experiment. We further hypothesized that pseudoflowers of U. pisi interact with uninfected true host flowers through insects during their period of co-"flowering" in early spring. We conducted artificial array experiments in the field to test whether the two species share insects and whether they influenced each other's insect visitation. Insects moved between true flowers and pseudoflowers, but true flowers received more visits over all. Pseudoflowers and true flowers did not influence each other's visitation rates in mixtures. However, shorter visits were observed on pseudoflowers in mixtures than monocultures, suggesting that true flowers might be competitors for pseudoflowers. Further experiments are needed to determine whether the similarity of pseudoflowers to true flowers is adaptive.
Key Words: competition • facilitation • mating system • mimicry • pollination • pseudoflowers • rust fungi • spermatia
For over a hundred years studies have accumulated describing and discussing heterothallic rust fungi (fungi with more than one mating type) that present their haploid fungal gametes (spermatia) in a sweet-smelling, sugary nectar attractive to insects (Ráthay, 1882
; Buller, 1950
; Roy, 1994
). Since the experiments of Craigie (1927)
, it has generally been assumed that the insects attracted by these fungi aid in their sexual reproduction. However, only a handful of studies have actually proved that insects are required as gamete vectors for any rust fungi (Craigie, 1927, 1931
; Roy, 1993
). In the few studies in which insect-dependent fertilization was established, we can distinguish between nonsystemic (Craigie, 1927, 1931
) and systemic (Roy, 1993
) species of rust fungi. In nonsystemic infections, the spermatial lesions are very small and more than one mating type can be present on a leaf. In this case, fertilization of the fungus can occur when insects crawl over the leaf, or even by direct contact of hyphae of opposite mating types in nearby pustules (Craigie, 1927
), and gamete transfer among host plants is not necessary. In systemic infections, however, the spermatial lesions are usually generated by the same fungal genotype on the whole host plant, and these infections may contain only one mating type. In this case, insects are likely to be required to transfer gametes between infected host plants.
Systemic infections can also differ from nonsystemic infections by the degree to which they change host morphology. Systemic rusts of Puccinia monoica on Arabis holboellii inhibit the ability of their hosts to form functional flowers (Roy, 1993
). Instead, the leaves of the host become pale yellow, the host is induced to form a rosette on top of the stem, and the fungus presents sweet-smelling nectar on these leaves. Because these rosettes of infected leaves resemble true flowers, they are referred to as pseudoflowers (Roy, 1993
).
Here we present data on another pathogen-plant complex of systemic rust fungi that produce pseudoflowers on their hosts: fungi of the species complex Uromyces pisi (Pers.) Wint. on their host species Euphorbia cyparissias L. (Fig. 1). For over a 100 years this host-pathogen pair has been subject to much speculation concerning the function of nectar-producing spermatia (Ráthay, 1882
; Buller, 1950
). Today, it is generally assumed that they are produced to attract insects for fungal fertilization, but we do not know of any study that actually tests the role of insects for the fertilization of the fungus. We used an insect exclusion experiment to test the dependence of the fungus on insect visitation for sexual reproduction.
|
) and that the direction of selection can be influenced by the floral composition of the community because insects choose what to visit (Rathcke, 1983
; Waser, 1983
; Feinsinger, 1987
). Insect-mediated species interactions can lead to either competition or facilitation.
Competition for pollinators may cause selection for differences in flower characters because competition can be ameliorated when mutations arise that allow the use of another pollinator or the same pollinator at a different time (Robertson, 1895, 1924
; Levin and Anderson, 1970
; Mosquin, 1971
; Waser, 1978, 1983
; Feinsinger, 1987
; but also see Connell, 1980
). However, pollinator preferences can also drive the evolution of similarities. When pollinators prefer the more common flower phenotype (or species) to a rare one, this preference will select the co-occurring rare phenotype (or species) towards similarity to the most common flower species (Mosquin, 1971
; Straw, 1972
; Thomson, 1978
). This process is referred to as positive frequency-dependent selection and is important in the evolution of Müllerian floral mimicry.
Facilitation for pollinator visitation may cause selection for similarities in flower characters because co-flowering species can gain a selective advantage through their combined attractiveness to pollinators (Grant and Grant, 1968
; Macior, 1970
; Brown and Kodric-Brown, 1979
; Rathcke, 1983
; Feinsinger, 1987
; Gross, 1996
). Under facilitation, individual visitation rates to flowers in mixed patches should be higher than in single-species patches. Whether facilitation of visitation translates into selection for similarity or differentiation depends on the degree of improper pollen transfer (Rathcke, 1983
; Feinsinger, 1987
). Improper transfer of pollen between species reduces male fitness as the pollen is placed on the wrong species, and it can also reduce female fitness if the improper pollen reduces the probability of the proper pollen making contact with the stigma, or if the improper pollen is toxic. For facilitation to drive the evolution of similarity among flowers, the benefits of similarity in terms of pollinator attraction must outweigh the disadvantages of improper pollen transfer, or mechanisms must be in place that allow proper pollen transfer despite the visual similarity of the species.
In this study we used artificial array experiments to test, under controlled densities and frequencies, whether pseudoflowers of U. pisi and true flowers of E. cyparissias share visitors, and whether insect behavior can indicate competition or facilitation among the two species. If it is found that the species influence each other's visitation through insects, either in a positive or a negative way, then further experiments would need to be done to determine the influence of visitor behavior on fitness and selection of the two involved species.
MATERIALS AND METHODS
The host and its pathogen
The host plant, Euphorbia cyparissias L. (cypress spurge), is a Eurasian perennial that is very common throughout Switzerland, found from 200 to nearly 3000 m above sea level. It occurs mainly on poor and fairly dry soils, along forest edges, in sparse woods, and in dry alpine grasslands. Euphorbia cyparissias has also been introduced to eastern Canada and northeastern United States, where it occurs in similar habitats (Stahevitch, Crompton, and Wojtas, 1987
; Gassmann and Schröder, 1995
). It is self-compatible (Schürch, Pfunder, and Roy, in press
), but due to protogyny generally needs insect visits to set seeds. The plants have a mutualistic relationship with ants. Ants pollinate the flowers while feeding on extrafloral nectaries (Schürch, Pfunder, and Roy, in press
) and disperse the seeds, which bear an elaiosome, a food source for the ants.
Rust fungi in the species complex Uromyces pisi (Pers.) Wint. are common pathogens on E. cyparissias in Europe (Gäumann, 1959
). They produce pseudoflowers with nectar on their host. Uromyces pisi species possess a heteroecious life cycle, alternating from E. cyparissias to another host, a member of the Fabaceae (Fig. 2). Each species of the complex attacks only one species from the family of Fabaceae (Gäumann, 1959
).
|
). The fungus passes the winter in the roots of the host plant and grows with the host as it sprouts in the spring, drastically changing host morphology. Infected plants are hypertrophied, with enlarged cells and intercellular spaces where the mycelium of the fungus is present (Hartwich, 1955
). This makes infected leaves look succulent compared to the original leaf morphology of the host. Infected stems grow earlier in the season than uninfected stems, a result of enhanced concentrations of growth hormones (Pilet, 1952
). On top of the infected stem, the fungus induces its host to form pseudoflowers (Fig. 1). These pseudoflowers bear some resemblance to Euphorbia flowers in that they are both yellow and produce sweet-smelling nectar that is easily reached by insects without requiring specialized mouth parts.
Field experiments
Insect exclusion experiment
The insect exclusion experiment was conducted in Vicques, Cras de la Combe, in the Swiss Jura Mountains at ~500 m above sea level (co-ordinates 599 000/245 940), in a species-rich calcareous grassland. The site was oriented 144° south-southeast at roughly a 30° angle. The area in which the E. cyparissias population was studied was 2680 m2, with an average density of ~1 E. cyparissias stem/m2 (with a mean of ~2–3 stems per individual). However, the population was spatially clustered and in places reached much higher local densities of up to 16 stems/m2. About 20% of all stems were infected, but the proportion varied in space as well as time. Infected stems appeared ~1 mo earlier than uninfected ones, and partly dried up while some E. cyparissias flowers were still blooming. Nonetheless, overlap in nectar production and presentation of pseudoflowers and true flowers could be observed for ~1 mo.
The exclusion experiment was conducted in the early spring of 1997. Infected individuals of E. cyparissias were caged with small-meshed florist gauze nets (Kleen Test Products, division of Meridian Industries, Inc., Milwaukee, Wisconsin, USA) to exclude insects. The bags allow light, wind, and water to penetrate, but exclude even the smallest insects. If an individual plant produced more than one stem, all stems were included together into the same cage.
The first infected E. cyparissias stem was found on 7 March 1997. For the rest of the month, the field was visited at least twice a week, and all infected individuals that could be found before the fungus produced spermatia and nectar were caged or left uncaged as controls, marked with a flag, and numbered consecutively. During April, the field was visited at least once a week, but fresh stems of infected individuals were rare. One of the following four treatments was assigned systematically to each of the host individuals: (1) "self" (within plant) fertilization of caged individuals with paintbrushes, (2) hand outcross (among plant) fertilization of caged individuals with paintbrushes, (3) caged controls without treatment, and (4) uncaged controls with natural insect visitation.
For each outcross fertilization, three infected host individuals that were not included in the experiment otherwise were "visited" with a paintbrush and the sticky liquid nectar with spermatia was transferred back to the treated plant. For "self"-fertilization, spermatia were moved within a caged plant with a paintbrush. Each treatment application was repeated at two to five different days to each of the caged plants. The plants were harvested when the stems were either displaying aecia (fungal organs in which aeciospores as infective agents to the alternate hosts are produced) or were losing their leaves and drying up. The first stems were harvested on 2 May, the last on 20 May; after the harvest they were pressed and dried in a plant press. To collect data on fungal reproductive success, a binocular microscope (Wild, M5A, Heerbrugg) with 60–120x magnification was used. Fungal reproductive success was defined by the fungus' ability to produce aeciospores. Aeciospores are heterokaryotic spores that are distributed by wind to the Fabaceae hosts. The production of aeciospores is the result of successful fertilization and can therefore be compared to the production of seeds in plants. We counted the number of aecia, the fungal organs in which aeciospores are produced. We assume that more aecia also produce more aeciospores. 
2 tests were used to determine whether (a) the production of aecia on all individuals, (b) the mean proportions of aecia-bearing stems on the subset of aecia-bearing individuals, and (c) the mean proportion of aecia-bearing leaves on the subset of aecia-bearing stems were independent of the pollination treatment.
Artificial array experiment
To determine whether pseudoflower visitation may be facilitated by co-occurrence with true flowers of uninfected Euphorbia cyparissias, an artificial array experiment was conducted from 18 to 19 April in Vicques (the same field site as for the insect-exclusion experiment) and from 15 to 16 May in Ardez in the lower Engadine, southern Switzerland, at ~1410 m elevation. The Euphorbia population (140 m2) near Ardez is located in a dry calcareous grassland (co-ordinates 811 225/184 025), oriented at an ~20° angle and facing 190° south. The density of Euphorbia was much higher in Ardez than in Vicques. At the time of the experiment in Ardez we observed an average density of 53 E. cyparissias stems/m2, and about half (48%) of the stems were infected.
Florist pics (small plastic tubes) were used to arrange single stems of pseudoflowers and true flowers on two plots of 1 m2, located 5 m apart (Fig. 3A). Insect observations in the arrays were only conducted during sunny and calm weather conditions. To control density and frequency of the plots, all flowers were removed from these plots before setting out the arrays. Three kinds of artificial arrays were used: (1) a monoculture of ten pseudoflowers in spermatial stage (U. pisi) , (2) a mixture of five pseudoflowers in spermatial stage (U. pisi) and five true flowers of E. cyparissias, and (3) a monoculture of ten true flowers of E. cyparissias. In Vicques as well as in Ardez, two different plots 5 m apart (Fig. 3A) were chosen. Two observers were assigned randomly to the two plots, and both observers observed one array each during a time period of 20 min (t1–t4 in Fig. 3B). After each observation period, all tubes were removed and reordered for the next array. The experiment was analyzed as a factorial, blocked design with two blocking factors, a spatial block and a time block (Fig. 3B). A spatial block consisted of all arrays being observed once at one of the two plots and was designed to account for variation in visitation resulting from the observation plots being in locations 5 m apart. A time block consisted of all arrays being observed at both plots once (two spatial arrays) and was designed to account for variation in insect activity within and among the different times of observation.
|
). Recording the mixed array twice per spatial block gave us the opportunity to analyze only pseudoflowers or only true flowers in the two mixed arrays of Ardez and therefore to analyze the two species independently. The choice of which species was analyzed per mixed array was randomly assigned.
For each visitor, the visit to a pseudoflower or true flower was recorded, and the amount of time spent per visit on each flower or pseudoflower was measured with a stopwatch. To test whether pseudoflowers and true flowers actually shared insects, movements by insects within the arrays were also recorded and analyzed using Bateman's constancy index, which measures the degree of fidelity for insects to a particular species (Waser, 1986
). The particular insect visitors were identified as well as possible during the observations, and identifications were verified after the experiment by catching representatives. The data were separated into two categories: all insects and insects other than ants. Separating the ant data from the other insects is reasonable, because ants made up to ~40% of all insect visits in Vicques and ~10% in Ardez. From a different experiment we know that ants pollinate true flowers of E. cyparissias, but that they do not fertilize pseudoflowers of U. pisi (Schürch, Pfunder, and Roy, in press
).
To determine the kind of interaction between fungal pseudoflowers and true flowers as mediated by insects (e.g., competition, facilitation, or no interaction), the visitation rates and the mean duration per visit were calculated for true flowers and pseudoflowers separately, and the effect of the array (monocultures vs. mixtures) was tested. A mixed-model analysis of variance (ANOVA) was used with time blocks and spatial blocks included as random factors. Three-way interactions were excluded from the model, and two-way interactions were included and removed from the model if they were nonsignificant at P > 0.25. Synthetic denominators, as calculated in JMP, Version 3.1. (SAS, 1994
), were used to calculate F ratios in the mixed models. When the interaction of "species" (pseudoflowers or true flowers) by "array" (mixtures or monocultures) was significant, we analyzed the following a priori contrasts: U. pisi-mono (in monoculture) vs. U. pisi-mix (in mixtures) and E. cyparissias-mono (in monocultures) vs. E. cyparissias-mix (in mixtures). Visitation rates and mean duration per visit were analyzed for "all insects" and for "insects excluding ants." The model for visitation rates in Vicques excluding ants is not shown, because the whole-model test was not significant. The data were transformed only when necessary to meet the normality and homogeneity assumptions of ANOVA.
RESULTS
Insect exclusion experiment
The proportion of host individuals bearing fungal aecia were significantly different among treatments (2 = 46.74, 3 df, P < 0.0001; Table 1). When the fungus was hand outcrossed or left to natural visitation, it produced many more aecia than when it was "selfed" or prevented from visits by cages. Within the aecia-producing individuals we further observed the degree to which fertilization was successful: while the mean proportions of aecia-bearing stems on the subset of aecia-bearing individuals were not significantly different among treatments (2 = 7.29, 3 df, P = 0.0633; Table 1), the proportions of aecia-bearing leaves that could be counted on the subset of aecia-bearing stems were significant (2 = 14.344, 3 df, P = 0.0025; Table 1).
|
) could be calculated for the Ardez site where we had more than ten transitions between species. The calculated index was -0.02, a number very close to zero, indicating no preference for either pseudoflowers or true flowers by the insects flying between inflorescences in the mixtures. The interspecific movements were made by various species of Diptera (Calliphoridae and Syrphidae), and Hymenoptera of the family Halictidae, genus Lasioglossum. True flowers received more visits than pseudoflowers at both locations (Fig. 4). In monocultures in Vicques, the ratio of visits to true flowers vs. pseudoflowers was 5.0:1 for all insects and 5.6:1 for insects excluding ants. In Ardez, the monocultural plots were visited in a ratio of 2.5:1 for all insects and 2.3:1 for insects excluding ants, in favor of true flowers. Overall visitation to individual pseudoflowers and true flowers was not influenced by whether they occurred in a mixture or a monoculture (Table 2, Fig. 4A, B). These results did not change when ants were excluded (Table 3, Fig. 4A, B).
|
|
|
|
|
We addressed three questions in this study: (1) is insect visitation necessary for the fungi of U. pisi to reproduce, (2) do U. pisi and E. cyparissias share insect visitors, and (3) if visitors are shared, is competition or facilitation suggested by insect behavior?
The first question was answered unambiguously with an insect exclusion experiment: insects were required for the reproduction of U. pisi. Other studies of nonsystemic and systemic heterothallic rust fungi have shown a similar degree of fungal reproduction when insects were excluded (0–20%), or treated by hand "self"-fertilization (3–20%), hand outcross-fertilization (88–96%), and naturally insect-visited controls (98–100%) (Craigie, 1927
; Roy, 1993
; Schürch, Pfunder, and Roy, in press
). In all cases, natural insect-mediated fertilization as well as hand outcross-fertilization led to the most fungal reproduction. However, occasionally the fungus was able to reproduce under other treatment conditions such as self-fertilization or no fertilization. In our case, some degree of non-insect-mediated fertilization may have resulted from any of three causes: a low level of self-compatibility in the fungus, infections by two fungi of opposite mating type within one host, or errors in experimental work such as accidental caging of previously fertilized stems.
In our study, we found that the formation of aecia of U. pisi could be restricted to single stems or even single leaves on a host individual, depending on the treatment that we applied to the fungus. This can be explained by a study of Hartwich (1955)
, who showed that, although all stems of one Euphorbia individual come out of the same root, and therefore should be infected by the same systemic fungus, the mycelium can become separated by the growth of the host plant. This separation can lead to isolation of fungal parts in single stems or even single leaves of a host. Therefore, a visit to one infected stem or leaf does not necessarily lead to a dikaryotic mycelium in the whole infected plant, and the fertilization event can be locally restricted. If mycelial separation is common, this could mean that more insect visits to different stems or leaves on one host plant would also increase reproductive success for the fungus. Longer visitation periods on pseudoflowers could also lead to increased fungal reproductive success if the insects wander around more and thus make more fertilization contacts.
True flowers of the host plant E. cyparissias co-"flower" some of the time with fungal pseudoflowers. To test whether pseudoflowers and true flowers were visited by the same insects, we observed insect movements among single inflorescences in 50:50 mixtures. In the mixed arrays 42 insects moved between inflorescences, and of these, 17 transitions (40%) were interspecific movements between pseudoflowers and true flowers. Because the number of transitions observed was small, constancy indices are potentially unstable. Thus we do not dwell on them here, but our observations clearly showed that insects move between the two species.
Flowering species that share pollinators can interact with each other through them. To test whether pseudoflowers and true flowers interact through insects, we conducted an artificial array experiment in the field. Artificial arrays have the advantage that they can include natural communities. We were able to include two locations with different communities into our study. But there are limitations to the generality of our results. First, the effects of visitation rates on the reproductive success of pseudoflowers and true flowers cannot easily be quantified, and interspecific movements of insects can reduce reproductive success of the opponent species (Waser, 1978
; Roy, 1996
). Second, it was only possible to make observations on a limited subset of frequencies, densities, sites, and days. Because pollinators are known to respond to the overall frequency and density of flowers (Thomson, 1982
; Rathcke, 1983
; Feinsinger, 1987
), the results from the visitation experiment need to be interpreted within the context of the populations in which the arrays were set. At Vicques, the average density of Euphorbia stems in the population (infected and flowering) was 1 stem/m2, but the distribution of stems was patchy and stems often ranged up to 16 stems/m2. Thus, our density of ten inflorescences in monocultural arrays represented a reasonable compromise between average and local densities. However, true flowers were much more common than pseudoflowers in the population (4:1) than they were in our mixed arrays with five inflorescences each (1:1).
At Ardez, the average density of Euphorbia stems in the population (infected and flowering) ranged between 34 and 54 stems/m2. Thus, our density of ten individuals in monocultural arrays was much smaller than average natural densities. However, it was necessary to keep the density in the arrays the same as in Vicques if we wanted to make statistical comparisons between sites. Although the density of stems in our arrays was less than the population at large, the frequency of pseudoflowers and true flowers was approximately the same as in our mixed arrays (1:1).
At both sites, true flowers were visited more often than pseudoflowers. Insects were clearly able to differentiate between Euphorbia flowers and pseudoflowers. If insects are visiting flowers based only on the frequency of a species in a population (ignoring any innate predispositions), then we would expect visitation at Vicques to be 4:1 in favor of true flowers and 1:1 at Ardez. What we actually found was visitation in favor of true flowers at both sites; the mean visitation rate per individual and hour in monocultures was 5:1 (all insects) and 5.6:1 (no ants) in Vicques and 2.5:1 (all insects) and 2.3:1 (no ants) in Ardez. The fact that visitation was higher to true flowers than their frequency alone predicts suggests that the preference for true flowers was not purely frequency dependent.
When considering interactions with insect visitors, it is important to consider not just the number of visits, but also the duration of visits. Longer visitation times have been shown to benefit some plants through increased pollen deposition (Thomson and Plowright, 1980
; Galen and Plowright, 1985
; Thomson, 1986
; Harder, 1990
; Mitchell and Waser, 1992
) and may also benefit pseudoflowers by increasing the chance of fertilizing separated hyphae. Alternatively, longer visits might lower outcrossing rates by reducing the number of between-plant visits per unit time.
Our results showed that pseudoflowers in Ardez received shorter visits in mixed populations than in monocultures. If longer visits enhance the reproductive success of the fungus, our results indicate that true flowers of E. cyparissias might be serious competitors when visitation is a limiting resource. Further experiments are needed to determine whether pollinators are in limited supply and what the consequences of the behavior observed here are for phenotypic selection. If pollinators are limited, then competition could select for differences between species if the differences allow the species to partition the sparse resource, but this is not the only possible outcome. For example, a preference of insects for Euphorbia flowers could cause selection for pseudoflowers to become more like Euphorbia flowers, if there is variation in pseudoflower morphology and more flower-like pseudoflowers have higher fitness than those that are less Euphorbia-like.
In this study we have tested some, but not all, of the conditions required to show that the pseudoflowers produced by U. pisi mimic the flowers of Euphorbia. The rust fungus pseudoflowers are somewhat similar in appearance to Euphorbia flowers, the rust fungus requires insect visitation for reproduction, the two taxa have some overlap in "flowering" phenology, the same insect species visit both species, and the same individual insects fly between them. Furthermore, we have shown that these species interact through their insect visitors. Competition was suggested by the fact that overall visitation to the true flowers was higher, and the duration of visits was shorter on pseudoflowers in mixtures than on those in monocultures, although this difference in duration did not affect visitation rates. Further experiments are needed to determine whether the resemblance of pseudoflowers to Euphorbia flowers is actually adaptive; these studies must examine fitness and not simply visitation and should consider the consequences of variation in density and frequency throughout the flowering season.
FOOTNOTES
1 The authors thank H. Alexander, D. Brem, A. Leuchtmann, M. Parker, B. Rathcke, A.-B. Utelli, and D. Zuber for constructive comments on the manuscript; T. Steinger for statistical support; and D. Brada and O. Holzgang for their valuable help in the field. This work was funded by the Swiss National Science Foundation (NF 277-311-96). 
2 Author for correspondence (Fax.: +41 1 632 12 15; e-mail: pfunder{at}geobot.umnw.ethz.ch
).
3 E-mail: roy{at}geobot.umnw.ethz.ch
LITERATURE CITED
Brown, J. H., and A. Kodric-Brown. 1979 Convergence, competition, and mimicry in a temperate community of hummingbird-pollinated flowers. Ecology 60: 1022–1035. [CrossRef][ISI]
Buller, A. H. R. 1950 Researches on fungi, vol. VII. The sexual process in the Uredinales. University of Toronto Press, Toronto, Ontario, Canada.
Connell, J. H. 1980 Diversity and the coevolution of competitors, or the ghost of competition past. Oikos 35: 131–138. [CrossRef][ISI]
Craigie, J. H. 1927 Discovery of the function of the pycnia of the rust fungi. Nature 120: 765–767. [CrossRef]
———. 1931 An experimental investigation of sex in the rust fungi. Phytopathology 21: 1001–1040. [ISI]
Feinsinger, P. 1987 Effects of plant species on each other's pollination: is community structure influenced? Trends in Ecology and Evolution 2: 123–126.
Galen, C., and R. C. Plowright. 1985 The effects of nectar level on pollen carry-over in inflorescences of fireweed (Epilobium angustifolium: Onagraceae). Canadian Journal of Botany 63: 488–491.
Gassmann, A., and D. Schröder. 1995 Commentary: the search for effective biological control agents in Europe: history and lessons from leafy spurge (Euphorbia esula L.) and cypress spurge (Euphorbia cyparissias L.). Biological Control 5: 466–477. [CrossRef]
Gäumann, E. 1959 Die Rostpilze Mitteleuropas. Beiträge zur Kryptogamenflora der Schweiz Band 12: 1–1407.
Grant, K. A., and V. Grant. 1968 Hummingbirds and their flowers. Columbia University Press, New York, New York, USA.
Gross, C. L. 1996 Is resource overlap disadvantageous to three sympatric legumes? Australian Journal of Ecology 21: 133–143.
Harder, L. D. 1990 Pollen removal by bumblebees and its implications for pollen dispersal. Ecology 71: 1110–1125. [CrossRef][ISI]
Hartwich, W. 1955 Untersuchungen über die Entwicklung des Uromyces pisi (DC.) Otth. auf Euphorbia cyparissias L. Phytopathologische Zeitschrift 24: 73–96.
Levin, D. A., and W. W. Anderson. 1970 Competition for pollinators between simultaneously flowering species. American Naturalist 104: 455–467. [CrossRef][ISI]
Macior, L. W. 1970 The pollination ecology of Pedicularis in Colorado. American Journal of Botany 57: 716–728. [CrossRef][ISI]
Mitchell, R., and N. M. Waser. 1992 Adaptive significance of Ipomopsis aggregata nectar production: pollination success of single flowers. Ecology 73: 633–638. [CrossRef][ISI]
Mosquin, T. 1971 Competition for pollinators as a stimulus for the evolution of flowering time. Oikos 22: 398–402. [CrossRef][ISI]
Pilet, P. E. 1952 Etude physiologique du parasitisme de l’ Uromyces Pisi (Pers.) de By, sur l’ Euphorbia Cyparissias L. Experientia 9: 300–302. [ISI]
Ráthay, E. 1882 Untersuchungen über die Spermogonien der Rostpilze. In Denkschrift der Kaiserlichen Akademischen Wissenschaften Wien, Band XLVI, Wien, Germany.
Rathcke, B. 1983 Competition and facilitation among plants for pollination. In L. Real [ed.], Pollination biology. Academic Press, Orlando, Florida, USA.
Robertson, C. 1895 The philosophy of flower seasons, and the phenological relations of the entomophilous flora and the anthophilous insect fauna. American Naturalist 29: 97–117. [CrossRef]
———. 1924 Phenology of entomophilous flowers. Ecology 5: 393–407. [CrossRef][ISI]
Roy, B. A. 1993 Floral mimicry by a plant pathogen. Nature 362: 56–58. [CrossRef]
———. 1994 The use and abuse of pollinators by fungi. Trends in Ecology and Evolution 9: 335–339. [CrossRef]
———. 1996 A plant pathogen influences pollinator behavior and may influence reproduction of nonhosts. Ecology 77: 2445–2457. [CrossRef][ISI]
SAS. 1994 JMP version 3.1. SAS Institute, Cary, North Carolina, USA.
Schürch, S., M. Pfunder, and B. A. Roy. In press. Effects of ants on the reproductive success in Euphorbia cyparissias and associated pathogenic rust fungi. Oikos.
Shykoff, J. A., E. Bucheli, and O. Kaltz. 1997 Anther smut disease in Dianthus silvester (Caryophyllaceae): natural selection on floral traits. Evolution 51: 383–392. [CrossRef][ISI]
Snaydon, R. W. 1991 Replacement or additive designs for competition studies? Journal of Applied Ecology 28: 930–946. [CrossRef][ISI]
Stahevitch, A. E., C. W. Crompton and W. A. Wojtas. 1987 The biology of Canadian weeds. 85. Euphorbia cyparissias L. Canadian Journal of Plant Science 68: 175–191. [ISI]
Straw, R. M. 1972 A markov model for pollinator constancy and competition. American Naturalist 106: 597–620. [CrossRef][ISI]
Thomson, J. D. 1978 Effects of stand composition on insect visitation in two-species mixtures of Hieraceum. American Midland Naturalist 100: 431–440.
———. 1982 Patterns of visitation by animal pollinators. Oikos 39: 241–250. [CrossRef][ISI]
———. 1986 Pollen transport and deposition by bumblebees in Erythronium: influences of floral nectar and bee grooming. Journal of Ecology 74: 329–341. [CrossRef]
———, and R. C. Plowright. 1980 Pollen carryover, nectar rewards, and pollinator behavior with special reference to Diervilla lonicera. Oecologia 46: 68–74.
Waser, N. M. 1978 Interspecific pollen transfer and competition between co-occurring plant species. Oecologia 36: 223–236. [CrossRef][ISI]
———. 1983 Competition for pollination and floral character differences among sympatric plant species: a review of evidence. In C. E. Jones and R. J. Little [eds.], Handbook of experimental pollination biology, Van Nostrand Reinhold, New York, New York, USA.
———. 1986 Flower constancy: definition, cause, and measurement. American Naturalist 127: 593–603. [CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  A. Canto, C. M. Herrera, M. Medrano, R. Perez, and I. M. Garcia Pollinator foraging modifies nectar sugar composition in Helleborus foetidus (Ranunculaceae):An experimental test Am. J. Botany, March 1, 2008; 95(3): 315 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Fordyce The evolutionary consequences of ecological interactions mediated through phenotypic plasticity J. Exp. Biol., June 15, 2006; 209(12): 2377 - 2383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Kaiser Flowers and Fungi Use Scents to Mimic Each Other Science, February 10, 2006; 311(5762): 806 - 807. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lara and J. F. Ornelas Hummingbirds as vectors of fungal spores in Moussonia deppeana (Gesneriaceae): taking advantage of a mutualism? Am. J. Botany, February 1, 2003; 90(2): 262 - 269. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05 are given in boldface