THE GEOGRAPHICAL MOSAIC OF COEVOLUTION IN A PLANT–POLLINATOR MUTUALISM

Although coevolution is widely accepted as a concept, its importance as a driving factor in biological diversification is still being debated. Because coevolution operates mainly at the population level, reciprocal coadaptations should result in trait covariation among populations of strongly interacting species. A long-tongued fly ( Prosoeca ganglbaueri ) and its primary floral food plant ( Zaluzianskya microsiphon ) were studied across both of their geographical ranges. The dimensions of the fly's proboscis and the flower's corolla tube length varied significantly among sites and were strongly correlated with each other. In addition, the match between tube length of flowers and tongue length of flies was found to affect plant fitness. The relationship between flower tube length and fly proboscis length remained significant in models that included various alternative environmental (altitude, longitude, latitude) and allometric (fly body size, flower diameter) predictor variables. We conclude that coevolution is a compelling explanation for the geographical covariation in flower depth and fly proboscis length.     

Coevolution Between Food-Rewarding Flowers and Their Pollinators

The concept of coevolution was first developed by Darwin, who used it to explain how pollinators and food-rewarding flowers involved in specialized mutualisms could, over time, develop long tongues and deep tubes, respectively. He famously predicted that Angraecum sesquipedale, a long-spurred Malagasy orchid, must be pollinated by a hawkmoth with an exceptionally long tongue. Darwin’s idea of a coevolutionary “race” was championed by contemporary naturalists, including Alfred Wallace, and a hawkmoth fitting the expected tongue-length profile was eventually discovered in Madagascar during the early twentieth century. However, strong empirical support for the mechanism behind Darwin’s coevolutionary model has been forthcoming only in the past two decades. It is now established that selection often strongly favors plants with floral tubes that exceed the length of their pollinator’s tongues. There is also evidence that pollinators gain an energetic benefit from having tongues that enable them to consume most or all of the nectar in deep tubular flowers. Alternative explanations for the evolution of long pollinator tongues, such as evasion of predators that use flowers as ambush sites, are considered much less compelling and lack quantitative support. Another important advance in coevolution research has been the development of approaches that explicitly predict a geographical mosaic of coevolution. The expectation that coevolution can lead to geographical diversification and trait covariation among strongly interacting organisms is strongly supported by studies of long-proboscid fly and oil-bee pollination systems in South Africa. Macro- and microevolutionary studies of pollination systems suggest that coevolution can operate alongside other one-sided evolutionary processes, such as shifts, in shaping plant and pollinator traits.     

Morphological and genetic differentiation in Isodon umbrosus by altitudinal variation in bumblebee pollinator assemblages

The corolla tube length of the bumblebee-pollinated plant Isodon umbrosus shows conspicuous geographical variation, corresponding with the proboscis length of its bumblebee pollinators across its distributional range. We hypothesized that altitudinal variation in the pollinator assemblage is a principal factor mediating morphological and genetic differentiation among I. umbrosus populations. We examined determinants of the morphological and genetic differentiation of Isodon umbrosus by analyzing floral morphology and allozyme variation across the distributional range. A reanalysis of previous data confirmed that altitude was a good indicator of pollinator assemblages. Corolla tube length was highly variable among the 15 study populations, and genetic differentiation among the populations ( G _ST = 0.360) was also highly significant. The differentiation in corolla tube length was explained by altitudinal difference, a proxy of the difference in pollinator assemblages. Genetic differentiation among populations also tended to be affected by the same factor, but statistical support was weak. To better understand the mechanisms responsible for morphological and genetic differentiation in I. umbrosus , we need to investigate altitudinally different populations over a narrower geographical scale.     

The extremely long-tongued neotropical butterfly Eurybia lycisca (Riodinidae): proboscis morphology and flower handling

Few species of true butterflies (Lepidoptera: Papilionoidea) have evolved a proboscis that greatly exceeds the length of the body. This study is the first to examine the morphology of an extremely long butterfly proboscis and to describe how it is used to obtain nectar from flowers with very deep corolla tubes. The proboscis of Eurybia lycisca (Riodinidae) is approximately twice as long as the body. It has a maximal length of 45.6?mm (mean length 36.5?mm?±?4.1?S.D., N?=?20) and is extremely thin, measuring only about 0.26?mm at its maximum diameter. The proboscis has a unique arrangement of short sensilla at the tip, and its musculature arrangement is derived. The flower handling times on the preferred nectar plant, Calathea crotalifera (Marantaceae), were exceptionally long (mean 54.5?sec?±?28.5?S.D., N?=?26). When feeding on the deep flowers remarkably few proboscis movements occur. The relationship between Eurybia lycisca and its preferred nectar plant and larval host plant, Calathea crotalifera, is not mutualistic since the butterfly exploits the flowers without contributing to their pollination. We hypothesize that the extraordinarily long proboscis of Eurybia lycisca is an adaptation for capitalizing on the pre-existing mutualistic interaction of the host plant with its pollinating long-tongued nectar feeding insects.     

Temporal changes in calyx tube length of Clematis stans (Ranunculaceae): a strategy for pollination by two bumble bee species with different proboscis lengths

We examined the adaptive significance of a temporal decrease in the calyx tube length of Clematis stans, a dioecious species pollinated by Bombus diversus (long proboscis) and B. honshuensis (short proboscis). We compared visitation frequency, pollen removal, pollen deposition, and fruit set after a single visit among three flower stages, differentiated by calyx tube length. Bombus diversus frequently visited and removed significantly more pollen from long flowers. Bombus honshuensis visited and tended to remove more pollen from short flowers. Both pollinators deposited more pollen in short flowers, resulting in higher fruit set. These results indicate that size correspondence between the proboscis and the calyx tube enhances visitation frequency and pollen removal, but not pollen deposition. Because a single visit does not fertilize all ovules of a flower, multiple visits by two bumble bee species may increase seed production and genetic diversity of offspring. By temporally changing calyx tube length, C. stans can use two bumble bee pollinators and maintain specialized relationships with each. This strategy may be adaptive when the pollinator fauna fluctuates, and is economical because it eliminates costs required to produce different types of flowers. This constitutes a novel pattern of temporal specialization in flower-pollinator relationships.     

Pollination biology in the Snowy Mountains of Australia: Comparisons with montane Colorado,USA

Various aspects of the pollination biology of the alpine flora of Kosciusko National Park, NSW, were examined from late December 1983 until the end of March 1984, including flowering phenology, corolla tube lengths, flower colour, ultraviolet reflectance patterns, visitation rates to the flowers and proboscis lengths of the flower-visiting insects. An average of 5.3 species flowered in each of 13, 2 m×2 m montane plots and 5.6 species in the 13 alpine plots. The maximum number in flower simultaneously averaged 4.1 species in the montane and 3.3 in the alpine plots; flowering peaked in mid-January, Corolla tube lengths of the flora averaged 1.73 mm. The most common floral colour was white or predominantly white (40 species), followed by yellow (14 species). Only six of the 38 species (16%) examined had some type of reflectance pattern; the remaining species all absobed ultraviolet. Flies appeared to be the major pollinators. The insects collected in the study area comprised 60 species of Diptera, 33 species of Hymenoptera, and several species each of Lepidoptera and Coleoptera. On average, 14.4% of flowers watched during 379 observation periods (10 min each) were visited. On average, each plant species was visited by 6.4 species of flies, 2.4 species of bees, wasps or sawflies, one species of butterfly or moth and 0.3 species of beetles. Visitation rates increased over the growing season, and were significantly affected by ambient temperature (positively), light levels (positively) and wind speed (negatively). The maximum proboscis length for the 25 most common species of bees was 2.76 mm, but 18 of 51 species of flies had proboscis lengths longer than this. The mean proboscis length for all 25 species of bees was 1.68 mm, and for 51 species of flies was 2.31 mm. Proboscis lengths for flies were positively correlated with the average corolla length for the plant species they visited. For bees, however, the range in proboscis lengths was relatively small and did not show this pattern. There appear to be significant differences between the plant-pollinator community of alpine Australia and other alpine areas where bumblebees are common pollinators. These differences include shorter proboscis and corolla tube lengths, and perhaps an increased diversity and significance of flies as pollinators.     

Geographical and within-population variation in the globeflower–globeflower fly interaction: the costs and benefits of rearing pollinator’s larvae

Interspecific interactions can vary within and among populations and geographical locations, and this variation can influence the nature of the interaction (e.g. mutualistic versus antagonistic) and its evolutionary stability. Globeflowers are exclusively pollinated by flies whose larvae feed only on their seeds. Here we document geographical variability in costs and benefits in globeflowers in sustaining their pollinating flies throughout the range of this arctic-alpine European plant over several years. A total of 1,710 flower heads from 38 populations were analysed for their carpel, egg and seed contents. Individual and population analyses control for the confounding influences of variation in both: (1) population traits, such as fly density and egg distribution among flower heads; and (2) individuals traits, such as carpel and egg numbers per flower head. Despite considerable variation in ecological conditions and pollinator densities across populations, large proportions (range 33–58%) of seeds are released after predation, with a benefit-to-cost ratio of 3, indicating that the mutualism is stable over the whole globeflower geographical range. The stability of the mutualistic interaction relies on density-dependent competition among larvae co-developing in a flower head. This competition is revealed by a sharp decrease in the number of seeds eaten per larva with increasing larval number, and is intensified by non-uniform egg distribution among globeflowers within a population. Carpel number is highly variable across globeflowers (range 10–69), and flies lay more eggs in large flowers. Most plants within a population contribute to the rearing of pollinators, but some pay more than others. Large globeflowers lose more seed to pollinator larvae, but also release more seed than smaller plants. The apparent alignment of interests between fly and plant (positive relationship between numbers of seeds released and destroyed) is shown to hide a conflict of interest found when flower size is controlled for.     

Structure in seven bumblebee communities in eastern Finland in relation to resource availability

The structure of seven local bumblebee communities was studied in eastern Finland. The number of species in the study sites varied from 5 to 10, 11 being the number in the species pool. Most of the species (queens) had proboscis lengths of 8–11 mm. When floral resources were analysed on the basis of corolla tube depths, a group was found between 4 and 8 mm. In terms of flower utilization, species with similar proboscises had higher pairwise niche overlaps than species differing by their proboscis lengths.
Spacing between neighbouring species along the proboscis length dimension was not more uniform in the realized communities than what was expected on the basis of random draws from the local species pool. The results were the same for all species observed in the communities and for the abundant species only (frequency >5%), although competition among species in the latter group is expected to be more intense.
When the structure of bumblebee communities and of floral resources was expressed in terms of diversity of proboscis lengths and corolla tube depths, respectively, a positive correlation was found. Data suggest that species with similar proboscis tend to co-occur on fields with availability of appropriate corolla tubes.     

Intraspecific divergence and convergence of floral tube length in specialized pollination interactions

Floral tubes are often thought to be a consequence of adaptive specialization towards pollinator morphology. We explore floral tube length evolution within Tritoniopsis revoluta (Iridaceae), a species with considerable geographical tube length variation. We ask whether tube lengths of T. revoluta populations are associated with pollinator proboscis lengths, whether floral divergence occurs in the presence of different pollinators and whether floral convergence occurs between distantly related populations pollinated by the same pollinator. Finally, we ask whether tube length evolution is directional. Shifts between morphologically different pollinators were always associated with shifts in floral morphology, even when populations were very closely related. Distantly related populations had similar tube lengths when they were pollinated by the same pollinator. Shifts in tube length tended to be from short to long, although reversals were not infrequent. After correcting for the population-level phylogeny, there was a strong positive, linear relationship between floral tube length and pollinator proboscis length, suggesting that plants are functionally specialized on different pollinators at different sites. However, because tube length evolution in this system can be a bidirectional process, specialization to the local pollinator fauna is unlikely to result in evolutionary or ecological dead-ends such as canalization or range limitation.     

Altitudinal flower size variation correlates with local pollinator size in a bumblebee‐pollinated herb,Prunella vulgaris L. (Lamiaceae)

The influence of locally different species interactions on trait evolution is a focus of recent evolutionary studies. However, few studies have demonstrated that geographically different pollinator‐mediated selection influences geographic variation in floral traits, especially across a narrow geographic range. Here, we hypothesized that floral size variation in the Japanese herb Prunella vulgaris L. (Lamiaceae) is affected by geographically different pollinator sizes reflecting different pollinator assemblages. To evaluate this hypothesis, we posed two questions. (1) Is there a positive correlation between floral length and the proboscis length of pollinators (bumblebees) across altitude in a mountain range? (2) Does the flower–pollinator size match influence female and male plant fitness? We found geographic variation in the assemblage of pollinators of P. vulgaris along an altitudinal gradient, and, as a consequence, the mean pollinator proboscis length also changed altitudinally. The floral corolla length of P. vulgaris also varied along an altitudinal gradient, and this variation strongly correlated with the local pollinator size but did not correlate with altitude itself. Furthermore, we found that the size match between the floral corolla length and bee proboscis length affected female and male plant fitness and the optimal size match (associated with peak fitness) was similar for the female and male fitness. Collectively, these results suggest that pollinator‐mediated selection influences spatial variation in the size of P. vulgaris flowers at a fine spatial scale.     

Identification of effective pollinators of Primula sieboldii E. Morren in a wild habitat in Hiroshima,Japan

Primula sieboldii E. Morren is a heterostylous clonal herb that is widely distributed in Japan but in danger of extinction in the wild. The existence of pollinators in each habitat is imperative for its long-term survival, because seeds can be produced only by insect cross-pollination between different flower morphs. In this study, we identified the pollinators of P. sieboldii in a wild habitat in Hiroshima as those insects that we observed to (a) put the proboscis into a corolla tube, (b) deposit pollen grains on the proboscis, and (c) have a proboscis of appropriate length and width. Effective pollinators were identified from their contribution to pollination. In 2015 and 2016, the flower visitations of 232 and 558 insects, respectively, were recorded and 85 and 13 insects were captured. Two Bombylius species, B. major L. and B. shibakawae Matsumura, accounted for 90% of flower-visiting insects in both years. All 14 species that we captured were considered pollinators of P. sieboldii, because they had proboscises that were long enough to reach pollen and that had pollen grains deposited on them. The total visitation rate of “Bombyliidae” was the highest among all pollinator categories. The results of potential pollen transport per flower per hour, which was based on total pollen number and total visitation rate of each pollinator category, indicated that “Bombyliidae” species were the most effective pollinators of P. sieboldii in this habitat.     

Drinking with a very long proboscis: Functional morphology of orchid bee mouthparts (Euglossini,Apidae, Hymenoptera)

Neotropical orchid bees (Euglossini) possess the longest proboscides among bees. In this study, we compared the feeding behavior and functional morphology of mouthparts in two similarly large-sized species of Euglossa that differ greatly in proboscis length. Feeding observations and experiments conducted under semi-natural conditions were combined with micro-morphological examination using LM, SEM and micro CT techniques. The morphometric comparison showed that only the components of the mouthparts that form the food tube differ in length, while the proximal components, which are responsible for proboscis movements, are similar in size. This study represents the first documentation of lapping behaviour in Euglossini. We demonstrate that Euglossa bees use a lapping-sucking mode of feeding to take up small amounts of fluid, and a purely suctorial technique for larger fluid quantities. The mouthpart movements are largely similar to that in other long-tongued bees, except that the postmentum in Euglossa can be extended, greatly enhancing the protraction of the glossa. This results in a maximal functional length that is about 50% longer than the length of the food canal composing parts of the proboscis. The nectar uptake and the sensory equipment of the proboscis are discussed in context to flower probing.     

The least effective pollinator principle: specialized morphology despite generalized ecology

• The large body of work on the adaptation of plants to pollinators is still somewhat incomplete because most studies focus on one‐to‐one interactions. How will adaptation proceed in a multi‐pollinator environment? According to Stebbins' Most Effective Pollinator Principle, ‘the characteristics of the flower will be moulded by those pollinators that visit it most frequently and effectively.’
• To test this hypothesis, we studied the pollination biology of Pelargonium incrassatum (Geraniaceae) in the Namaqualand Region of Southern Africa. This species has a long floral tube and we expected its most important pollinator to have a long proboscis.
• Contrary to expectations, the most important pollinator was a short proboscid fly (a new species of Prosoeca), while Prosoeca peringueyi, which had a proboscis that matched the floral tube length, was a rare visitor. Consistent with the high degree of trait mismatching, we did not detect selection on tube length at most sites.
• The paradox of mismatching traits can be resolved by considering the strength of the trade‐off involved. Adaptation to the rare species can apparently occur without incurring the cost of reduced pollination by the abundant species. Generally, species may often evolve specialized morphology if they do not incur the cost of ecological specialization.

     

Geographic and within-population variation in the globeflower–globeflower fly interaction: the costs and benefits of rearing pollinators’ larvae

Interspecific interactions can vary within and among populations and geographic locations, and this variation can influence the nature of the interaction (e.g. mutualistic vs. antagonistic) and its evolutionary stability. Globeflowers are exclusively pollinated by flies, whose larvae feed only on their seeds. Here we document geographic variability in costs and benefits in globeflowers in sustaining their pollinating flies throughout the range of this arctic-alpine European plant over several years. A total of 1,710 flower heads from 38 populations were analysed for their carpel, egg and seed contents. Individual and population analyses control for the confounding influences of variation in both: (1) population traits, such as fly density and egg distribution among flower heads; and (2) individuals traits, such as carpel and egg numbers per flower head. Despite considerable variation in ecological conditions and pollinator densities across populations, large proportions (range 33–58%) of seeds were released after predation, with a benefit-to-cost ratio of 3, indicating that the mutualism is stable over the whole globeflower geographical range. The stability of the mutualistic interaction relies on density-dependent competition among larvae co-developing in a flower head. This competition is revealed by a sharp decrease in the number of seeds eaten per larva with increasing larval number, and is intensified by non-uniform egg distribution among globeflowers within a population. Carpel number is highly variable across globeflowers (range 10–69), and flies lay more eggs in large flowers. Most plants within a population contribute to the rearing of pollinators, but the costs are greater for some than for others. Large globeflowers lose more seed to pollinator larvae, but also release more seed than smaller plants. The apparent alignment of interests between fly and plants (positive relationship between numbers of seed released and destroyed) is shown to hide a conflict of interest found when flower size is controlled for.     

Bumblebee-pollination and temporal change of the calyx tube length in Clematis stans(Ranunculaceae)

Clematis stans is dioecious semi-arboreal, with pale purple–blue, nodding, tubulous flowers in a paniculate inflorescence. Both male and female flowers produce nectar from the base of the calyx tube during a flowering period of 3 or 4 days, and are pollinated by two bumblebee species, Bombus diversus and B. honshuensis, with different proboscis lengths. When the flowers open, four sepals constructing a calyx tube separate at the top and their respective tips gradually curl up, so that a tubular part shortens. Observations at two field sites showed that B. diversus (with a longer proboscis) most often visits the flowers with a longer calyx tube, and B. honshuensis (with a shorter proboscis) the flowers with a shorter calyx tube, i.e., later in the flowering period. By changing the calyx tube length, the flowers of C. stans accept the two bumblebee species with different proboscis length as pollinators and thus increase the chance of pollination for each flower. It was also found that the two bumblebee species prefer the male flowers to the female flowers, although the female flowers secrete more nectar as a reward than male flowers. This is likely because they visit the male flowers to collect pollen grains in addition to nectar. Electronic Publication     

Estimating attraction of Syrphidae (Diptera) to flowering plants with interception traps

Syrphidae with predaceous larvae are important predators of aphids and other insects and can be attracted and maintained in agricultural environments by the addition of flowering plants. Malaise interception traps baited with moveable flowering plants are a novel means of surveying for attractive species and can have the advantages of: (1) homogenising experimental site and plant quality, (2) portability, (3) continuous sampling, (4) capacity to simultaneously capture a broad range of insects (including pests) and (5) no requirement for additional sensory cues to be effective. Six of the 10 species of plants tested were relatively attractive (number of syrphids captured in flower-traps/numbers captured in no plant controls). While flower-traps captured more syrphid species than their associated controls, there were no differences between flower-traps and controls in the sizes (head height) or proboscis lengths of the flies collected. There were no significant relationships between relative attractiveness and flower width or depth or with plant height and floral area. Similarly, mean proboscis length of flies taken in flower-baited traps was not correlated with flower width or depth. The absence of the latter relationship may be due to the inability of an interception trap to distinguish between attraction and attraction-then-feeding.     

Contrasting covariation of above- and belowground invertebrate species across plant genotypes

1. Invertebrate species generally do not respond independently to genotypic variation in plants, giving rise to clusters of species that naturally associate with or avoid certain genotypes. This covariation causes coevolution to be diffuse rather than pairwise. Studies on this topic, however, have never considered the belowground invertebrate community, leaving a critical gap in our understanding. 2. We investigated the covariation among naturally colonising above- and belowground invertebrate species across six genetically distinct populations of the dune grass Ammophila arenaria. After having grown from seed in a common garden, plants were randomised in a single field site to exclude all but broad-sense genetic variation. 3. Strong positive covariation across genotypes among both above- and belowground invertebrates was detected, while correlations between these two groups were negative. This clustering of above- and belowground species matched well with order level taxonomy. Host range, trophic level and food type on the other hand did not correspond well with the clusters. Within the cluster of aboveground fauna, subsequent groupings were not related to any phylogenetic or ecological characteristic, although correlations within these subgroups were very high. We furthermore demonstrated significant differences in multiple invertebrate species occurrence between plant genotypes, in general as well as at the above- and belowground level. 4. The observed strong covariation suggests diffuse coevolution between A. arenaria and its associated invertebrate species. The trade-off between root and shoot invertebrates could however hamper directional selection on resistance to either group. 5. Our results clearly demonstrate the need for studies of plant-animal interactions to include the belowground fauna, as this might drastically alter our general conception of how plants and their associated animal communities interact and how these interactions shape the process of evolution.     

The effect of proboscis and corolla tube lengths on patterns and rates of flower visitation by bumblebees

Summary The rates at which bumblebees of different proboscis lengths forage on flowers of a series of corolla tube lengths were determined. The results indicate significant correlations between proboscis length and time spent by bees on flowers. Bumblebees of long proboscis length can forage significantly faster than bees of shorter proboscis length on flowers with long corolla tubes. There is also evidence which suggests that bumblebees of short proboscis length prefer and are more efficient on short corolla tubes. These results support the use of proboscis length as a morphological indicator of resource utilization in bumblebees.     

Evolution of mutualism between globeflowers and their pollinating flies

Plant/seed-eater pollinators mutualisms involve a plant pollinated by an insect whose larvae develop by eating a fraction of host-plant seeds. The outcome of the interaction therefore depends on the number of ovules fertilized by adult visits and the number of seeds destroyed by larvae. Among the very few cases of such mutualisms reported so far is the globeflower-globeflower flies mutualism, which is unique in that it involves several congeneric fly species (Chiastocheta genus) coexisting within a single host-plant species, Trollius europaeus. These species exhibit contrasted oviposition behaviors resulting in a more or less beneficial outcome for the plant. We designed an adaptive dynamics model to investigate how morphological traits of globeflower could affect the evolution of oviposition in its pollinating flies. Three fly traits (flower age at oviposition, clutch size and the level of avoidance of already parasitized flowers) and one plant trait (closed or open corolla) were examined. Whatever the shape of the flower, evolutionary branching occurs between early and late ovipositing flies, driven by strong competition among larvae within a fruit. Once this branching occurred, the closed shape of the corolla is likely to offer a better protection to eggs of early but not of late ovipositing flies. The difference in egg survival results in higher competition among early larvae and thus selects for decreased clutch size in early flies. This can be seen as a first step in the evolution of a mutualistic behavior. The prediction of our model fits field observations of fly behavior, giving theoretical support to the hypothesis of fly sympatric speciation within its host plant. Moreover, flower closed globe shape can be positively selected in globeflowers as it results in a reduction of parasitism strength. This last evolution therefore leads to a stable mutualism between globeflowers and globeflower flies.     

Functional decoupling between flowers and leaves in the Ameroglossum pernambucense complex can facilitate local adaptation across a pollinator and climatic heterogeneous landscape

Decoupling between floral and leaf traits is expected in plants with specialized pollination systems to assure a precise flower–pollinator fit, irrespective of leaf variation associated with environmental heterogeneity (functional modularity). Nonetheless, developmental interactions among floral traits also decouple flowers from leaves regardless of selection pressures (developmental modularity). We tested functional modularity in the hummingbird‐pollinated flowers of the Ameroglossum pernambucense complex while controlling for developmental modularity. Using two functional traits responsible for flower–pollinator fit [floral tube length (TL) and anther–nectary distance (AN)], one floral trait not linked to pollination [sepal length (SL), control for developmental modularity] and one leaf trait [leaf length (LL)], we found evidence of flower functional modularity. Covariation between TL and AN was ca. two‐fold higher than the covariation of either of these traits with sepal and leaf lengths, and variations in TL and AN, important for a precise flower–pollinator fit, were smaller than SL and LL variations. Furthermore, we show that previously reported among‐population variation of flowers associated with local pollinator phenotypes was independent from SL and LL variations. These results suggest that TL and AN are functionally linked to fit pollinators and sufficiently decoupled from developmentally related floral traits (SL) and vegetative traits (LL). These results support previous evidences of population differentiation due to local adaptation in the A. pernambucense complex and shed light on the role of flower–leaf decoupling for local adaptation in species distributed across biotic and abiotic heterogeneous landscapes.