Search

The Orchidaceae is characterised by the repeated evolution of sexual deception, one of the most specialised pollination strategies. In orchids, sexual deception involves long-range pollinator attraction via mimicry of female insect sex pheromones. At close range, visual signals involving colour mimicry, contrast to the background, and exploitation of pollinator sensory biases could attract pollinators, but remain largely untested. Here we focus on a remarkable system in which species from two only distantly related sexually deceptive orchid genera with strikingly different flowers (Drakaea livida and three species of Caladenia) share the same pollinator, males of the thynnine wasp Zaspilothynnus nigripes. We used spectral reflectance measurements and modelling to investigate pollinator perception of colour, including the first examination of overall colour patterns in flowers via colour pattern geometry analyses. Rather than closely matching the colours of female Z. nigripes, these orchids had strong chromatic and achromatic contrast against their backgrounds. For Caladenia, the sepals and petals show high contrast, while in D. livida, which has diminutive petals and sepals, it is the labellum that contrasts strongly against the background. Despite varying in colour, the Caladenia species all had strong within-flower contrast between a UV-bright central target (column and labellum) and a corolla of radiating stripes (petals and sepals). The colour pattern geometry analyses also indicated that the orchids’ overall colour patterns are highly conspicuous against their backgrounds. Contrast, UV, and target patterns could all enhance detection, and exploit pollinators’ innate preferences. Since colour contrast may function with a range of colours and floral forms, attracting pollinators via contrast rather than visual mimicry may be a critical but previously overlooked process facilitating the evolution of sexual deception.

Propagule dispersal is a crucial aspect of the survival and reproduction of sessile organisms, such as plants and fungi. As such, the colours of fleshy fruits serve as a visual cue for animal dispersers. However, little is known about how, or whether, specific traits of fungal fruiting bodies, such as colour or shape, attract animal dispersers, and additionally the identities of fungal dispersers are poorly understood. Globally, most truffle-like fungi are dull-coloured, subterranean, and likely have scents that are attractive to mammalian dispersers. In Aotearoa-New Zealand, however, brightly coloured truffle-like fungi that emerge from the forest floor have seemingly proliferated. This proliferation has prompted the hypothesis that they are adapted to dispersal by a bird-dominated fauna. In our study, we used the literature and citizen science data (GBIF) to explore whether colourful species occur at a higher proportion of the total truffle-like fungi flora in Aotearoa-New Zealand than elsewhere in the world. In addition, we tested for a relationship between biotic factors (avian frugivory and forest cover) and abiotic factors (precipitation, radiation, and temperature) and the prevalence of brightly coloured truffle-like fungi across the world. The most colourful truffle-like fungi are in three defined regions: Australia, South and Central America and the Caribbean, and Aotearoa-NZ. Potential dispersers and the environment both relate to the distribution of truffle-like fungi: we found that increasing levels of frugivory were associated with higher proportions of colourful truffle-like fungi. This finding provides new insights into drivers of certain fungal traits, and their interactions between birds and fungi. Unique ecosystems, such as Aotearoa-NZ's bird-dominated biota, provide fascinating opportunities to explore how plants and fungi interact with the sensory systems of animals., (© 2024. The Author(s).)

Artificial light at night (ALAN) is a growing conservation concern for seabirds, which can become disoriented and grounded by lights from buildings, bridges and boats. Many fledgling seabirds, especially Procellariiformes such as petrels and shearwaters, are susceptible to light pollution. The Hauraki Gulf, a seabird hotspot located near Tāmaki Makaurau/Auckland, Aotearoa-New Zealand's largest urban city, with a considerable amount of light pollution and regularly documented events of seabird groundings. We aim to identify the characteristics of locations especially prone to seabird groundings. We used an online database of seabirds taken to a wildlife rescue facility by the public to map 3 years of seabird groundings and test for correlations between seabird groundings and the natural night sky brightness. We found that areas with lower amounts of natural night sky brightness and greater light pollution often had a higher number of seabirds grounded. Further, we identified important seasonal patterns and species differences in groundings. Such differences may be a by-product of species ecology, visual ecology and breeding locations, all of which may influence attraction to lights. In general, seabird groundings correlate with the brightness of the area and are species-specific. Groundings may not be indicative of human or seabird population abundance considering some areas have a lower human population with high light levels and had high amounts of seabird groundings. These findings can be applied worldwide to mitigate groundings by searching and targeting specific brightly lit anthropogenic structures. Those targeted structures and areas can then be the focus of light mitigation efforts to reduce seabird groundings. Finally, this study illustrates how a combination of community science, and a concern for seabirds grounded from light attraction, in addition to detailed animal welfare data and natural night sky brightness data can be a powerful, collaborative tool to aid global conservation efforts for highly-at-risk animals such as seabirds., Competing Interests: The authors declare that they have no competing interests., (© 2022 Heswall et al.)

Urbanisation and anthropogenic alteration of ecosystems has led to conflict between humans and wildlife. Such conflict is often observed in apex predators. Although human-wildlife conflict has been extensively studied, male/female differences in behaviour are rarely considered. We investigated male/female differences in foraging behaviour of the predatory/scavenging brown skua Catharacta antarctica lonnbergi breeding on a New Zealand island nature reserve in proximity to farmland. These skuas are subject to culling, when perceived as a threat to livestock. As part of a long-term ecological study, we used high-resolution Global Positioning System (GPS) devices to characterise the space-use of foraging brown skuas. We also analysed stable isotopes of carbon (δ 13 C) and nitrogen (δ 15 N) from modern and archived blood samples to investigate possible changes in diet over the past ~30 years. Analysis of 100 GPS tracks collected from 2014 to 2016 demonstrated that males and females consistently visited different habitats. Males spent most of their time close to their breeding territory on the island nature reserve and females frequently visited a farmed island approximately two kilometres away. Consistent with this finding, we show that male and female skuas also differed markedly in their diets: males specialised on burrow-nesting white-faced storm petrels Pelagodroma marina (80%) with only a small proportion of sheep remains Ovis aries (<6%) contributing to their diet. In contrast, female diet comprised 27% white-faced storm petrels, other seabirds (18%) and a relatively large proportion of sheep remains (47%). Further, our data (186 blood samples from 122 individuals) show that this male/female difference in diet has persisted at least since 1987. Because females fed disproportionally on sheep remains, they may be more vulnerable to being culled by farmers. Importantly, our case study suggests that intersexual differences in diet and foraging patterns can have major implications for the reproduction and survival of apex predators that interact with farming. We strongly suggest that intersexual differences in behaviour should be considered when investigating human-wildlife conflicts., (© 2021 British Ecological Society.)

Deception has evolved in a range of taxa. When deception imposes costs, yet persists over generations, exploited species typically have traits to help them bear or minimise costs. The sexually deceptive orchids, Cryptostylis spp., are pollinated by tricking male haplodiploid wasps (Lissopimpla excelsa) into mating with flowers, which offer no reward and often elicit sperm wastage. We hypothesise that by attracting haplodiploid species, orchids have a pollinator ideally suited to withstand the costs of sexual deception-and a selective advantage compared to other orchids. Haplodiploid females can reproduce with or without sperm-albeit when spermless, females can only have sons. Through orchid deception and sperm wastage, deceived haplodiploid populations could become male biased, providing enough males to share between orchids and females. In this way, pollinator populations can persist despite high densities of sexually deceptive orchids. Here, we aim to broadly test this prediction using museum and digital records of the pollinator, L. excelsa, from sites with or without orchids. For robustness, we also analyse the sex ratio of a sister ichneumonid species that occurs in the same areas but is not deceived by orchids. We found that at sites with orchids, L. excelsa was significantly more male biased than at sites without orchids and significantly more male biased than the sister ichneumonid. This survey is the first to test the population-level effects of sexually deceptive orchids on their pollinator. It supports our prediction that orchid deception can drive male-biased sex ratios in exploited pollinators.

Passive acoustic monitoring (PAM) coupled with automated species identification is a promising tool for species monitoring and conservation worldwide. However, high false indications of presence are still an important limitation and a crucial factor for acceptance of these techniques in wildlife surveys. Here we present the Assemblage of Focal Species Recognizers-AFSR, a novel approach for decreasing false positives and increasing models' precision in multispecies contexts. AFSR focusses on decreasing false positives by excluding unreliable sound file segments that are prone to misidentification. We used MatlabHTK, a hidden Markov models interface for bioacoustics analyses, for illustrating AFSR technique by comparing two approaches, 1) a multispecies recognizer where all species are identified simultaneously, and 2) an assemblage of focal species recognizers (AFSR), where several recognizers that each prioritise a single focal species are then summarised into a single output, according to a set of rules designed to exclude unreliable segments. Both approaches (the multispecies recognizer and AFSR) used the same sound files training dataset, but different processing workflow. We applied these recognisers to PAM recordings from a remote island colony with five seabird species and compared their outputs with manual species identifications. False positives and precision improved for all the five species when using AFSR, achieving remarkable 0% false positives and 100% precision for three of five seabird species, and < 6% false positives, and >90% precision for the other two species. AFSR' output was also used to generate daily calling activity patterns for each species. Instead of attempting to withdraw useful information from every fragment in a sound recording, AFSR prioritises more trustworthy information from sections with better quality data. AFSR can be applied to automated species identification from multispecies PAM recordings worldwide., Competing Interests: The authors have declared that no competing interests exist.

Worldwide, anthropogenic change is causing biodiversity loss, disrupting many critical ecosystem functions. Most studies investigating the relationship between biodiversity and ecosystem functioning focus on species richness, predominantly within the context of productivity-related functions. Consequently, there is limited understanding of how other biodiversity measures, such as species evenness (the distribution of abundance among species), affect complex multitrophic functions such as pollination. We explore the effect of species evenness on the ecosystem function of pollination using a controlled experiment with selected plants and insects in flight cages. We manipulated the relative abundances of plant and pollinator species, while holding species richness, composition, dominance order, and total abundance constant. Then, we tested how numerical species evenness affected network structure and consequently, seed production, in our artificial communities. Contrary to our expectation, numerical dominance in plant communities increased complementarity in pollinator use (reduced pollinator sharing) among plant species. As predicted by theory, this increased complementarity resulted in higher seed production for the most dominant and rare plant species in our cages. Our results show that in a controlled experimental setting, numerical species evenness can alter important aspects of plant-pollinator networks and plant reproduction, irrespective of species richness, composition, and total abundance. Extending this understanding of how species evenness affects ecosystem functioning to natural systems is crucial as anthropogenic disturbances continue to alter species' abundances, likely disrupting ecosystem functions long before extinctions occur., (© 2019 by the Ecological Society of America.)

Aim: We test for spatial and climatic patterns of diversification in the Orchidaceae, an angiosperm family characterized by high levels of species diversity and rarity. Globally, does orchid diversity correlate with land area? In Australia, does diversity correlate with herbarium collecting effort, range size, or climate niche breadth? Where are Australia's orchids distributed spatially, in protected areas, and in climate space?, Location: Global, then Australia., Methods: We compared orchid diversity with land area for continents and recognized orchid diversity hotspots. Then, we used cleaned herbarium records to compare collecting effort (for Australian Orchidaceae vs. all other plant families, and also among orchid genera). Spatial and climate distributions were mapped to determine orchids' coverage in the protected area network, range sizes, and niche breadths., Results: Globally, orchid diversity does not correlate with land area (depauperate regions are the subantarctic: 10 species, and northern North America: 394 species). Australian herbarium records and collecting effort generally reflect orchid species diversity (1,583 spp.), range sizes, and niche breadths. Orchids are restricted to 13% of Australia's landmass with 211 species absent from any protected areas. Species richness is the greatest in three biomes with high general biodiversity: Temperate (especially southwest and southeast Australia), Tropical, and Subtropical (coastal northern Queensland). Absence from the Desert is consistent with our realized climate niche-orchids avoid high temperature/low rainfall environments. Orchids have narrower range sizes than nonorchid species. Highly diverse orchid genera have narrower rainfall breadths than less diverse genera., Main Conclusions: Herbarium data are adequate for testing hypotheses about Australian orchids. Distribution is likely driven by environmental factors. In contrast, diversification did not correlate with increases in range size, rainfall, or temperature breadths, suggesting speciation does not occur via invasion and local adaptation to new habitats. Instead, diversification may rely on access to extensive obligate symbioses with mycorrhizae and/or pollinators.

Two main mechanisms may buffer ecosystem functions despite biodiversity loss. First, multiple species could share similar ecological roles, thus providing functional redundancy. Second, species may respond differently to environmental change (response diversity). However, ecosystem function would be best protected when functionally redundant species also show response diversity. This linkage has not been studied directly, so we investigated whether native and exotic pollinator species with similar traits (functional redundancy) differed in abundance (response diversity) across an agricultural intensification gradient. Exotic pollinator species contributed most positive responses, which partially stabilized overall abundance of the pollinator community. However, although some functionally redundant species exhibited response diversity, this was not consistent across functional groups and aggregate abundances within each functional group were rarely stabilized. This shows functional redundancy and response diversity do not always operate in concert. Hence, despite exotic species becoming increasingly dominant in human-modified systems, they cannot replace the functional composition of native species., (© 2017 The Author(s).)

Sensory-based conservation harnesses species' natural communication and signalling behaviours to mitigate threats to wild populations. To evaluate this emerging field, we assess how sensory-based manipulations, sensory mode, and target taxa affect success. To facilitate broader, cross-species application of successful techniques, we test which behavioural and life-history traits correlate with positive conservation outcomes. We focus on seabirds, one of the world's most rapidly declining groups, whose philopatry, activity patterns, foraging, mate choice, and parental care behaviours all involve reliance on, and therefore strong selection for, sophisticated sensory physiology and accurate assessment of intra- and inter-species signals and cues in several sensory modes. We review the use of auditory, olfactory, and visual methods, especially for attracting seabirds to newly restored habitat or deterring birds from fishing boats and equipment. We found that more sensory-based conservation has been attempted with Procellariiformes (tube-nosed seabirds) and Charadriiformes (e.g. terns and gulls) than other orders, and that successful outcomes are more likely for Procellariiformes. Evolutionary and behavioural traits are likely to facilitate sensory-based techniques, such as social attraction to suitable habitat, across seabird species. More broadly, successful application of sensory-based conservation to other at-risk animal groups is likely to be associated with these behavioural and life-history traits: coloniality, philopatry, nocturnal, migratory, long-distance foraging, parental care, and pair bonds/monogamy., (© 2016 Cambridge Philosophical Society.)

Niche construction theory explains how organisms' niche modifications may feed back to affect their evolutionary trajectories. In theory, the evolution of other species accessing the same modified niche may also be affected. We propose that this niche construction may be a general mechanism driving the evolution of mutualisms. Drosophilid flies benefit from accessing yeast-infested fruits, but the consequences of this interaction for yeasts are unknown. We reveal high levels of variation among strains of Saccharomyces cerevisiae in their ability to modify fruits and attract Drosophila simulans. More attractive yeasts are dispersed more frequently, both in the lab and in the field, and flies associated with more attractive yeasts have higher fecundity. Although there may be multiple natural yeast and fly species interactions, our controlled assays in the lab and field provide evidence of a mutualistic interaction, facilitated by the yeast's niche modification., (© 2014 John Wiley & Sons Ltd/CNRS.)

Globally, dung beetles (Scarabaeidae: Scarabaeinae) are linked to many critical ecosystem processes involving the consumption and breakdown of mammal dung. Endemic New Zealand dung beetles (Canthonini) are an anomaly, occurring at high abundance and low diversity on an island archipelago historically lacking terrestrial mammals, except bats, and instead dominated by birds. Have New Zealand's dung beetles evolved to specialise on bird dung or carrion, or have they become broad generalist feeders? We test dietary preferences by analysing nitrogen isotope ratios of wild dung beetles and by performing feeding behaviour observations of captive specimens. We also use nitrogen and carbon stable isotopes to determine if the dung beetle Saphobius edwardsi will consume marine-derived carrion. Nitrogen isotope ratios indicated trophic generalism in Saphobius dung beetles and this was supported by behavioural observations where a broad range of food resources were utilised. Alternative food resource use was further illustrated experimentally by nitrogen and carbon stable isotope signatures of S. edwardsi, where individuals provided with decomposed squid had δ(15)N and δ(13)C values that had shifted toward values associated with marine diet. Our findings suggest that, in the absence of native mammal dung resources, New Zealand dung beetles have evolved a generalist diet of dung and carrion. This may include marine-derived resources, as provided by the seabird colonies present in New Zealand forests before the arrival of humans. This has probably enabled New Zealand dung beetles to persist in indigenous ecosystems despite the decline of native birds and the introduction of many mammal species.

Yeasts that invade and colonise fruit significantly enhance the volatile chemical diversity of this ecosystem. These modified bouquets are thought to be more attractive to Drosophila flies than the fruit alone, but the variance of attraction in natural yeast populations is uncharacterised. Here we investigate how a range of yeast isolates affect the attraction of female D. melanogaster to fruit in a simple two choice assay comparing yeast to sterile fruit. Of the 43 yeast isolates examined, 33 were attractive and seven repellent to the flies. The results of isolate-versus-isolate comparisons provided the same relative rankings. Attractiveness varied significantly by yeast, with the strongly fermenting Saccharomyces species generally being more attractive than the mostly respiring non-Saccharomyces species (P = 0.0035). Overall the habitat (fruit or other) from which the isolates were directly sampled did not explain attraction (P = 0.2352). However, yeasts isolated from fruit associated niches were more attractive than those from non-fruit associated niches (P = 0.0188) regardless of taxonomic positioning. These data suggest that while attractiveness is primarily correlated with phylogenetic status, the ability to attract Drosophila is a labile trait among yeasts that is potentially associated with those inhabiting fruit ecosystems. Preliminary analysis of the volatiles emitted by four yeast isolates in grape juice show the presence/absence of ethanol and acetic acid were not likely explanations for the observed variation in attraction. These data demonstrate variation among yeasts for their ability to attract Drosophila in a pattern that is consistent with the hypothesis that certain yeasts are manipulating fruit odours to mediate interactions with their Drosophila dispersal agent.

The extraordinary taxonomic and morphological diversity of orchids is accompanied by a remarkable range of pollinators and pollination systems. Sexually deceptive orchids are adapted to attract specific male insects that are fooled into attempting to mate with orchid flowers and inadvertently acting as pollinators. This review summarises current knowledge, explores new hypotheses in the literature, and introduces some new approaches to understanding sexual deception from the perspective of the duped pollinator. Four main topics are addressed: (1) global patterns in sexual deception, (2) pollinator identities, mating systems and behaviours, (3) pollinator perception of orchid deceptive signals, and (4) the evolutionary implications of pollinator responses to orchid deception, including potential costs imposed on pollinators by orchids. A global list of known and putative sexually deceptive orchids and their pollinators is provided and methods for incorporating pollinator perspectives into sexual deception research are provided and reviewed. At present, almost all known sexually deceptive orchid taxa are from Australia or Europe. A few sexually deceptive species and genera are reported for New Zealand and South Africa. In Central and Southern America, Asia, and the Pacific many more species are likely to be identified in the future. Despite the great diversity of sexually deceptive orchid genera in Australia, pollination rates reported in the literature are similar between Australian and European species. The typical pollinator of a sexually deceptive orchid is a male insect of a species that is polygynous, monandrous, haplodiploid, and solitary rather than social. Insect behaviours involved in the pollination of sexually deceptive orchids include pre-copulatory gripping of flowers, brief entrapment, mating, and very rarely, ejaculation. Pollinator behaviour varies within and among pollinator species. Deception involving orchid mimicry of insect scent signals is becoming well understood for some species, but visual and tactile signals such as colour, shape, and texture remain neglected. Experimental manipulations that test for function, multi-signal interactions, and pollinator perception of these signals are required. Furthermore, other forms of deception such as exploitation of pollinator sensory biases or mating preferences merit more comprehensive investigation. Application of molecular techniques adapted from model plants and animals is likely to deliver new insights into orchid signalling, and pollinator perception and behaviour. There is little current evidence that sexual deception drives any species-level selection on pollinators. Pollinators do learn to avoid deceptive orchids and their locations, but this is not necessarily a response specific to orchids. Even in systems where evidence suggests that orchids do interfere with pollinator mating opportunities, considerable further research is required to determine whether this is sufficient to impose selection on pollinators or generate antagonistic coevolution or an arms race between orchids and their pollinators. Botanists, taxonomists and chemical ecologists have made remarkable progress in the study of deceptive orchid pollination. Further complementary investigations from entomology and behavioural ecology perspectives should prove fascinating and engender a more complete understanding of the evolution and maintenance of such enigmatic plant-animal interactions., (© 2010 The Author. Biological Reviews © 2010 Cambridge Philosophical Society.)

Typically, floral colour attracts pollinators by advertising rewards such as nectar, but how does colour function when pollinators are deceived, unrewarded, and may even suffer fitness costs? Sexually deceptive orchids are pollinated only by male insects fooled into mating with orchid flowers and inadvertently transferring orchid pollinia. Over long distances, sexually deceptive orchids lure pollinators with counterfeit insect sex pheromones, but close-range deception with colour mimicry is a tantalising possibility. Here, for the first time, we analyse the colours of four sexually deceptive Cryptostylis orchid species and the female wasp they mimic (Lissopimpla excelsa, Ichneumonidae), from the perspective of the orchids' single, shared pollinator, male Lissopimpla excelsa. Despite appearing different to humans, the colours of the orchids and female wasps were effectively identical when mapped into a hymenopteran hexagonal colour space. The orchids and wasps reflected predominantly red-orange wavelengths, but UV was also reflected by raised bumps on two orchid species and by female wasp wings. The orchids' bright yellow pollinia contrasted significantly with their overall red colour. Orchid deception may therefore involve accurate and species-specific mimicry of wavelengths reflected by female wasps, and potentially, exploitation of insects' innate attraction to UV and yellow wavelengths. In general, mimicry may be facilitated by exploiting visual vulnerabilities and evolve more readily at the peripheries of sensory perception. Many sexually deceptive orchids are predominantly red, green or white: colours that are all potentially difficult for hymenoptera to detect or distinguish from the background.

Sexually deceptive orchids lure pollinators by mimicking female insects. Male insects fooled into gripping or copulating with orchids unwittingly transfer the pollinia. The effect of deception on pollinators has been considered negligible, but we show that pollinators may suffer considerable costs. Insects pollinating Australian tongue orchids (Cryptostylis species) frequently ejaculate and waste copious sperm. The costs of sperm wastage could select for pollinator avoidance of orchids, thereby driving and maintaining sexual deception via antagonistic coevolution or an arms race between pollinator learning and escalating orchid mimicry. However, we also show that orchid species provoking such extreme pollinator behavior have the highest pollination success. How can deception persist, given the costs to pollinators? Sexually-deceptive-orchid pollinators are almost exclusively solitary and haplodiploid species. Therefore, female insects deprived of matings by orchid deception could still produce male offspring, which may even enhance orchid pollination.

Spiders and their mating systems are useful study subjects with which to investigate questions of widespread interest about sexual selection, pre- and post-copulatory mate choice, sperm competition, mating strategies, and sexual conflict. Conclusions drawn from such studies are broadly applicable to a range of taxa, but rely on accurate understanding of spider sexual interactions. Extensive behavioural experimentation demonstrates the presence of sex pheromones in many spider species, and recent major advances in the identification of spider sex pheromones merit review. Synthesised here are the emission, transmission, structures, and functions of spider sex pheromones, with emphasis on the crucial and dynamic role of sex pheromones in female and male mating strategies generally. Techniques for behavioural, chemical and electrophysiological study are summarised, and I aim to provide guidelines for incorporating sex pheromones into future studies of spider mating. In the spiders, pheromones are generally emitted by females and received by males, but this pattern is not universal. Female spiders emit cuticular and/or silk-based sex pheromones, which can be airborne or received via contact with chemoreceptors on male pedipalps. Airborne pheromones primarily attract males or elicit male searching behaviour. Contact pheromones stimulate male courtship behaviour and provide specific information about the emitter's identity. Male spiders are generally choosy and are often most attracted to adult virgin females and juvenile females prior to their final moult. This suggests the first male to mate with a female has significant advantages, perhaps due to sperm priority patterns, or mated female disinterest. Both sexes may attempt to control female pheromone emission, and thus dictate the frequency and timing of female mating, reflecting the potentially different costs of female signalling and/or polyandry to both sexes. Spider sex pheromones are likely to be lipids or lipid soluble, may be closely related to primary metabolites, and are not necessarily species specific, although they can still assist with species recognition. Newer electrophysiological techniques coupled with chemical analyses assist with the identification of sex pheromone compounds. This provides opportunities for more targeted behavioural experimentation, perhaps with synthetic pheromones, and for theorising about the biosynthesis and evolution of chemical signals generally. Given the intriguing biology of spiders, and the critical role of chemical signals for spiders and many other animal taxa, a deeper understanding of spider sex pheromones should prove productive.

Floral traits such as odor, color, and morphology are important pollinator attractants. Variation in floral traits may influence floral constancy, the tendency of pollinating insects to visit flowers of only one type. We investigated for the first time variation in odor between floral morphs in heterostylous species. We analyzed inter- and intraspecific odor variation in the "pin" and "thrum" floral morphs of sympatric Primula elatior and P. farinosa (Primulaceae). Floral volatiles were sampled with headspace sorption. Quantitative analysis and chemical identification were performed by gas chromatography coupled to mass spectrometry. The species produced different floral bouquets. P. elatior emitted mostly limonene with small amounts of alpha-pinene, myrcene, and sabinene. P. farinosa produced benzaldehyde, 4-oxoisophorone (2,6,6-trimethyl-2-cyclohexene-1,4-dione), benzyl alcohol, and benzyl acetate. These interspecific differences may play a role in promoting floral constancy and maintaining species integrity. Conversely, no differences were detected between the scents of pin and thrum morphs within each species. Heterostyly relies on pollinators visiting both floral morphs. There may be stabilizing selection against divergences in traits that may cause pollinators to develop floral constancy to only one of the floral morphs.