Non-rewarding Flowers Research Articles

Bumble Bees (Bombus terrestris) Use Time-Memory to Associate Reward with Color and Time of Day.

Circadian clocks regulate ecologically important complex behaviors in honey bees, but it is not clear whether similar capacities exist in other species of bees. One key behavior influenced by circadian clocks is time-memory, which enables foraging bees to precisely time flower visitation to periods of maximal pollen or nectar availability and reduces the costs of visiting a non-rewarding flower patch. Bumble bees live in smaller societies and typically forage over shorter distances than honey bees, and it is therefore not clear whether they can similarly associate reward with time of day. We trained individually marked bumble bee (Bombus terrestris) workers to forage for sugar syrup in a flight cage with yellow or blue feeders rewarding either during the morning or evening. After training for over two weeks, we recorded all visitations to colored feeders filled with only water. We performed two experiments, each with a different colony. We found that bees tended to show higher foraging activity during the morning and evening training sessions compared to other times during the day. During the test day, the trained bees were more likely to visit the rewarding rather than the non-rewarding colored feeders at the same time of day during the test sessions, indicating that they associated time of day and color with the sugar syrup reward. These observations lend credence to the hypothesis that bumble bees have efficient time-memory, indicating that this complex behavior is not limited to honey bees that evolved sophisticated social foraging behaviors over large distances.

Bumblebees can detect floral humidity.

ABSTRACTFloral humidity, a region of elevated humidity in the headspace of the flower, occurs in many plant species and may add to their multimodal floral displays. So far, the ability to detect and respond to floral humidity cues has been only established for hawkmoths when they locate and extract nectar while hovering in front of some moth-pollinated flowers. To test whether floral humidity can be used by other more widespread generalist pollinators, we designed artificial flowers that presented biologically relevant levels of humidity similar to those shown by flowering plants. Bumblebees showed a spontaneous preference for flowers that produced higher floral humidity. Furthermore, learning experiments showed that bumblebees are able to use differences in floral humidity to distinguish between rewarding and non-rewarding flowers. Our results indicate that bumblebees are sensitive to different levels of floral humidity. In this way floral humidity can add to the information provided by flowers and could impact pollinator behaviour more significantly than previously thought.

Why Variation in Flower Color May Help Reproductive Success in the Endangered Australian Orchid Caladenia fulva

Caladenia fulva G.W. Carr (Tawny Spider-orchid) is a terrestrial Australian endangered orchid confined to contiguous reserves in open woodland in Victoria, Australia. Natural recruitment is poor and no confirmed pollinator has been observed in the last 30 years. Polymorphic variation in flower color complicates plans for artificial pollination, seed collection and ex situ propagation for augmentation or re-introduction. DNA sequencing showed that there was no distinction among color variants in the nuclear ribosomal internal transcribed spacer (ITS) region and the chloroplast trnT-trnF and matK regions. Also, authentic specimens of both C. fulva and Caladenia reticulata from the reserves clustered along with these variants, suggesting free interbreeding. Artificial cross-pollination in situ and assessment of seed viability further suggested that no fertility barriers existed among color variants. Natural fruit set was 15% of the population and was proportional to numbers of the different flower colors but varied with orchid patch within the population. Color modeling on spectral data suggested that a hymenopteran pollinator could discriminate visually among color variants. The similarity in fruiting success, however, suggests that flower color polymorphism may avoid pollinator habituation to specific non-rewarding flower colors. The retention of large brightly colored flowers suggests that C. fulva has maintained attractiveness to foraging insects rather than evolving to match a scarce unreliable hymenopteran sexual pollinator. These results suggest that C. fulva should be recognized as encompassing plants with these multiple flower colors, and artificial pollination should use all variants to conserve the biodiversity of the extant population.

Changing How Biologists View Flowers-Color as a Perception Not a Trait.

Studying flower color evolution can be challenging as it may require several different areas of expertise, ranging from botany and ecology through to understanding color sensing of insects and thus how they perceive flower signals. Whilst studies often view plant-pollinator interactions from the plant's perspective, there is growing evidence from psychophysics studies that pollinators have their own complex decision making processes depending on their perception of color, viewing conditions and individual experience. Mimicry of rewarding flowers by orchids is a fascinating system for studying the pollinator decision making process, as rewarding model flowering plants and mimics can be clearly characterized. Here, we focus on a system where the rewardless orchid Eulophia zeyheriana mimics the floral color of Wahlenbergia cuspidata (Campanulaceae) to attract its pollinator species, a halictid bee. Using recently developed psychophysics principles, we explore whether the color perception of an insect observer encountering variable model and mimic flower color signals can help explain why species with non-rewarding flowers can exist in nature. Our approach involves the use of color discrimination functions rather than relying on discrimination thresholds, and the use of statistical distributions to model intraspecific color variations. Results show that whilst an experienced insect observer can frequently make accurate discriminations between mimic and rewarding flowers, intraspecific signal variability leads to overlap in the perceived color, which will frequently confuse an inexperienced pollinator. This new perspective provides an improved way to incorporate pollinator decision making into the complex field of plant-pollinator interactions.

Learning in two butterfly species when using flowers of the tropical milkweed Asclepias curassavica: No benefits for pollination.

The ability of insect visitors to learn to manipulate complex flowers has important consequences for foraging efficiency and plant fitness. We investigated learning by two butterfly species, Danaus erippus and Heliconius erato, as they foraged on the complex flowers of Asclepias curassavica, as well as the consequences for pollination. To examine learning with respect to flower manipulation, butterflies were individually tested during four consecutive days under insectary conditions. At the end of each test, we recorded the number of pollinaria attached to the body of each butterfly and scored visited flowers for numbers of removed and inserted pollinia. We also conducted a field study to survey D. erippus and H. erato visiting flowers of A. curassavica, as well as to record numbers of pollinaria attached to the butterflies' bodies, and surveyed A. curassavica plants in the field to inspect flowers for pollinium removal and insertion. Learning improves the ability of both butterfly species to avoid the nonrewarding flower parts and to locate nectar more efficiently. There were no experience effects, for either species, on the numbers of removed and inserted pollinia. Heliconius erato removed and inserted more pollinia than D. erippus. For both butterfly species, pollinium removal was higher than pollinium insertion. This study is the first to show that Danaus and Heliconius butterflies can learn to manipulate complex flowers, but this learning ability does not confer benefits to pollination in A. curassavica.

Assessing the ecological significance of bee visual detection and colour discrimination on the evolution of flower colours

Bee pollinators interact with flowers in a complex signal-receiver system. Chromatic traits that allow reliable discrimination between rewarding and non-rewarding flowers have been proposed as an important feature of pollination syndromes: bee-pollinated flowers have spectral profiles that closely match the discrimination peaks of their pollinators across the visual spectrum. However, in the complexity of a natural environment, it may be hard for bees to even detect the presence of flowers. In particular, little is known about how discrimination and detection by bees may together contribute to pollinator-mediated selection on floral colour signals. We address here an unexplained feature of floral colour evolution: the extreme paucity of spectral patterns with pronounced changes in reflectance around 420–480 nm wavelength. We began by conducting experiments with honeybees in a Y-maze to determine their capacity to detect a stimulus rarely found in bee-pollinated flowers—one with a single sharp spectral reflectance change at 478 nm. We found bees to be poor at detecting this stimulus against a neutral background. We then conducted behaviourally-informed computer simulations that test how bee visual discrimination and detection interact, which yielded information about which flower colours most effectively facilitate cross-pollination. Finally, we identified from our previous work those bird-pollinated species whose floral colours had spectral characteristics similar to the stimulus used in the Y-maze experiment. These data demonstrate that plants can, and do, produce such spectra for pollinators other than bees. In combination, our results show that the interaction between colour discrimination and detection is important for understanding flower community assembly.

Trees as huge flowers and flowers as oversized floral guides: the role of floral color change and retention of old flowers in Tibouchina pulchra.

Floral color changes and retention of old flowers are frequently combined phenomena restricted to the floral guide or single flowers in few-flowered inflorescences. They are thought to increase the attractiveness over long distances and to direct nearby pollinators toward the rewarding flowers. In Tibouchina pulchra, a massively flowering tree, the whole flower changes its color during anthesis. On the first day, the flowers are white and on the next 3 days, they change to pink. This creates a new large-scale color pattern in which the white pre-changed flowers contrast against the pink post-changed ones over the entire tree. We describe the spectral characteristics of floral colors of T. pulchra and test bumblebees’ response to this color pattern when viewed at different angles (simulating long and short distances). The results indicated the role of different color components in bumblebee attraction and the possible scenario in which this flower color pattern has evolved. We tested bumblebees’ preference for simulated trees with 75% pink and 25% white flowers resembling the color patterns of T. pulchra, and trees with green leaves and pink flowers (control) in long-distance approach. We also compared an artificial setting with three pink flowers and one white flower (T. pulchra model) against four pink flowers with white floral guides (control) in short-distance approach. Bumblebees spontaneously preferred the simulated T. pulchra patterns in both approaches despite similar reward. Moreover, in short distances, pollinator visits to peripheral, non-rewarding flowers occurred only half as frequently in the simulated T. pulchra when compared to the control. Thefore, this exceptional floral color change and the retention of old flowers in T. pulchra favors the attraction of pollinators over long distances in a deception process while it honestly directs them toward the rewarding flowers at short distances possibly exploring their innate color preferences.

Differential response to colour tasks on resident and migratory hummingbirds: a field test

Foraging performance of different hummingbird species was quantified by experimentally simulating changes in the environment, and we tested the hypothesis that species of migratory hummingbirds are more likely than resident to adjust their behaviour to changes in the environment. Colour discrimination tasks were designed to explore how 120 recently captured hummingbirds perform when the expected rewarding floral colour is not profitable. Hummingbirds were presented with an array consisting of 36 randomly distributed artificial flowers of two contrasting colours, where 18 were empty and either pink, orange or red, and 18 were filled with a sugar solution and were either green, yellow or orange. Although a few individuals failed or gave up visiting the experimental array after few non-rewarding foraging bouts (28%), most hummingbirds foraged continuously (72%). Out of those that foraged continuously, 36% consistently visited non-rewarding flowers and 64% switched to the rewarding colour. On average, migrants visited proportionally more rewarding flowers and were more willing to forage than residents, except the poor performance of one migratory species. Both migratory and resident hummingbirds performed better than chance when foraging on red/orange arrays, but their foraging performance differed on pink/green and orange/yellow arrays. Our results showed that some individuals thrive on risk and novelty while others shrink from the same situation, and that these differences are due to intrinsic differences among individuals in their tendency to switch, or a differential sensitivity to different colour combinations, regardless of their migratory status.

Predatory pollinator deception: Does the orchid mantis resemble a model species?

Abstract Cases of imperfect or non-model mimicry are common in plants and animals and challenge intuitive assumptions about the nature of directional selection on mimics. Many non-rewarding flower species do not mimic a particular species, but attract pollinators through ‘generalised food deception’. Some predatory animals also attract pollinators by resembling flowers, perhaps the most well known, yet least well understood, is the orchid mantis Hymenopus coronatus. This praying mantis has been hypothesised to mimic a flower corolla and we have previously shown that it attracts and captures pollinating insects as prey. Predatory pollinator deception is relatively unstudied and whether this occurs through model mimicry or generalised food deception in the orchid mantis is unknown. To test whether the orchid mantis mimics a specific model flower species we investigated similarities between its morphology and that of flowers in its natural habitat in peninsular Malaysia. Geometric morphometries were used to compare the shape of mantis femoral lobes to flower petals. Physiological vision models were used to compare the colour of mantises and flowers from the perspective of bees, flies and birds. We did not find strong evidence for a specific model flower species for the orchid mantis. The mantis’ colour and shape varied within the range of that exhibited by many flower petals rather than resembling one type in particular. We suggest that the orchid mantis resembles an average, or generalised flower-like stimulus. Thus predatory pollinator deception in the orchid mantis is likely to function as a form of generalised food deception, as opposed to model mimicry.

Ancestral deceit and labile evolution of nectar production in the African orchid genusDisa

An outstanding feature of the orchid family is that approximately 30-40% of the species have non-rewarding flowers and deploy various modes of deception to attract pollinators, whereas the remaining species engage in pollination mutualisms based on provision of floral rewards. Here, we explore the direction, frequency and reversibility of transitions between deceptive and rewarding pollination systems in the radiation of the large African genus Disa, and test whether these transitions had consequences for diversification. By optimizing nectar production data for 111 species on a well-resolved phylogeny, we confirmed that floral deception was the ancestral condition and that nectar production evolved at least nine times and was lost at least once. Transitions to nectar production first occurred ca 17 million years ago but did not significantly affect either speciation or extinction rates. Nectar evolved independently of a spur, which was lost and gained multiple times. These results show that nectar production can be a highly labile trait and highlight the need for further studies of the genetic architecture of nectar production and the selective factors underlying transitions between deception and mutualism.

Temporal reward variability promotes sampling of a new flower type by bumblebees

Foraging bumblebees focus mainly on one flower species (their ‘major’), and occasionally visit other flowers (‘minors’) that are less rewarding. Minoring allows tracking of potential alternative resources that might become more profitable with time. This is expected to provide an adaptive benefit under unstable foraging conditions, but to reduce foraging efficiency when resources are steady. We tested how predictability of food sources affects minoring and foraging success, using Bombus terrestris workers that fed on sucrose solution in artificial flowers. Bees exposed to three colour-distinct flower types with fluctuating food rewards performed more minoring (visits to a fourth, nonrewarding flower type) than bees that encountered a stable reward schedule. This difference was mostly due to a higher frequency of flower type shifts in the fluctuating-reward treatment. Flowers of the highest-rewarding type were visited less frequently in the fluctuating-reward than in the stable-reward treatment. This suggests that the fluctuating-reward schedule limited the bees' ability to track the most profitable flower type. Intervals between successive visits were longer in the fluctuating-reward than in the stable-reward treatment, possibly because the fluctuating schedule required time-consuming neural processing. As expected, the number of minoring visits correlated negatively with the number of colour shifts and with foraging success in the stable-reward treatment. In the fluctuating-reward treatment, on the other hand, sucrose intake rates were independent of minoring and colour shift frequencies. We suggest that novelty seeking by foragers can evolve when food sources offer fluctuating rewards, which reduce foraging success but also the cost of exploring new resources.

Parallel evolution of angiosperm colour signals: common evolutionary pressures linked to hymenopteran vision

Flowering plants in Australia have been geographically isolated for more than 34 million years. In the Northern Hemisphere, previous work has revealed a close fit between the optimal discrimination capabilities of hymenopteran pollinators and the flower colours that have most frequently evolved. We collected spectral data from 111 Australian native flowers and tested signal appearance considering the colour discrimination capabilities of potentially important pollinators. The highest frequency of flower reflectance curves is consistent with data reported for the Northern Hemisphere. The subsequent mapping of Australian flower reflectances into a bee colour space reveals a very similar distribution of flower colour evolution to the Northern Hemisphere. Thus, flowering plants in Australia are likely to have independently evolved spectral signals that maximize colour discrimination by hymenoptera. Moreover, we found that the degree of variability in flower coloration for particular angiosperm species matched the range of reflectance colours that can only be discriminated by bees that have experienced differential conditioning. This observation suggests a requirement for plasticity in the nervous systems of pollinators to allow generalization of flowers of the same species while overcoming the possible presence of non-rewarding flower mimics.

How to cheat when you cannot lie? Deceit pollination in Begonia gracilis

Mimicry between rewarding and non-rewarding flowers within individuals has been accepted as a strategy favored by selection to deceive pollinators. It has been proposed that this mechanism relies on the exploitation of pollinator's sensory biases, but field evidence is still scarce. In this study, we describe the mechanism of deceit pollination in the monoecious herb Begonia gracilis, a species with exposed rewarding structures (pollen) and intersexual mimicry. Specifically, we test the role of mimicry and exploitation of sensory biases on the reproductive success of male (pollination visitation) and female flowers (probability of setting fruits). We show that pollinators' perception of the amount of reward provided by male flowers is influenced by the independent variation in the sizes of the androecium and the perianth. Large rewarding structures and small perianths were preferred by pollinators, suggesting a central role of the relative size of the rewarding structure on pollinators' foraging decisions. Hence, rewarding male flowers cheat pollinators by exploiting their sensory biases, a strategy followed by non-rewarding female flowers. We suggest that intersexual mimicry operates through the functional resemblance of male flowers' deceit strategy. Artificial manipulation of the flowers supports our findings in natural conditions. Overall, we propose that the continuous and independent variation in the size of the perianth and the reproductive organs among male and female flowers could itself be adaptive.

What We Think We Know vs. What We Need to Know About Orchid Pollination and Conservation: Cypripedium L. as a Model Lineage

While Darwin (1862, 1877) showed that reproductive success in orchid populations depended on adaptive floral morphology coupled with pollinator visitation a more recent review of the literature (Tremblay et al., 2005) confirmed that many out-breeding species are pollinator-limited because most orchid species showing low fecundity also lack rewards. The absence of rewards depresses both pollinator fidelity and the frequency of pollinator visits to an orchid population even though orchid flowers that lack rewards retain the same interlocking floral structures for precise pollinia removal and deposition found in related species that offer rewards. Using the genus, Cypripedium, as a model lineage of non-rewarding flowers this study also shows that the correlation between low fruit set in a Cypripedium sp. and its specific pollinator(s) is insufficient to predict specific frequencies of low fecundity. Annual rates of fruit set often vary broadly between populations of the same species and within the same population over several seasons. We speculate that fruit-set rates also decline when orchid demography and additional biotic and abiotic factors interrupt rates of pollinator activity (pre-zygotic) and fertilization/fruit maturation (post-zygotic). We suggest that that traditional field studies on pollination ecology and breeding systems be combined with data sets recording genetic variation and orchid flower demography in relation to seasonal variation in climate. We also propose that the same information be collected in regard to genetic variation, demography and phenology of populations of known orchid pollinators and co-blooming angiosperm species native to orchid habitats.

Honeybees choose "incorrect" colors that are similar to target flowers in preference to novel colors

How non-rewarding flowers are able to receive sufficient pollinator visits to enable successful reproduction is an intriguing question. This study investigates the psychophysics of honeybee perception when individual honeybees are presented with a choice of a known non-rewarding color that is perceptually similar to a previously rewarding (but not present) target color, or a novel color. In experiment one, bees were tested for color preferences to three different artificial blue flowers, two of which were very similar in color; there was no color preference for any of the stimuli. In a second experiment, bees were provided with absolute conditioning to one of the perceptually similar stimuli. In subsequent non-rewarded tests these bees could not discriminate between the similar colors but could discriminate the non-similar color. In a third experiment, bees provided with differential conditioning (target rewarded and distractor unrewarded) learned to reliably discriminate between the perceptually similar stimuli. Surprisingly, when these bees were presented with a task of choosing between the known non-rewarding flower and a novel flower, bees chose the non-rewarding flower. The finding reveals that there is a strong perceptual effect on bees to choose stimuli that are perceptually similar to known rewarding flowers. The results explain how non-rewarding floral mimics may gain sufficient pollinator visits to enable pollination.

Pollination and Facultative Ant-Association in the African Leopard OrchidAnsellia africana

The leopard orchid Ansellia africana (Orchidaceae) is an epiphytic species widely distributed across tropical Africa. The pollination ecology of A. africana was investigated by direct observation. Buds and stalks of A. africana exude droplets of extra-floral nectar, but mature flowers produce no nectar. The role of extra-floral nectar appears to be recruitment of foraging ants to tend the flowers resulting in a facultative ant-association between the orchid and gregarious ants. Four different ant species were found to forage on A. africana's inflorescences. Ant-tended inflorescences suffered significantly less damage by insects. Mature A. africana flowers are non-rewarding which suggests a deceit pollination mechanism being employed by this orchid. The pollinators of A. africana were observed to be wild solitary bees species in the genera Xylocopa, Amegilla (Apidae) and Gronocera (Megachilidae). Observations of bee visitors to A. africana revealed overall low-visitation rates but adequate fruit set despite the non-rewarding flowers. Solitary bees visiting A. africana were ‘deceived’ by the flowers, but attracted to the floral displays and scent. A. africana shows spatial and temporal manipulation of two distinct suites of insects, ants and solitary bees.

The timing of food‐deceptive flowers: a commentary on Internicola et al. (2008)

A recent article presents a study of pollinator visitation behaviour that is used to evaluate the selective pressure that pollinator visitation rate might have on the timing of the production of nonrewarding flowers. Here we take issue with the conclusions of the paper that there should be selection pressure for nonrewarding flowers to be available earlier in the season in order to avoid dissimilar sympatric rewarding species. Consideration of selection pressure must take into account temporal variation in total pollinator availability, pollinator longevity and unlearned response, and the stability of plant communities over time, as well as the learned responses of individual pollinators that the original study focused on. Learning alone would not necessarily select for early flowering by nonrewarders if temporal variation in pollinator numbers is strong or naïve pollinators consistently appear throughout the flowering season. Further, we argue that early flowering could simply be a natural corollary of longevity of flowers needed to combat negative frequency-dependent selection and low overall visitation rates by pollinators, rather than a trait that has been specifically selected to reduce temporal overlap with competing rewarding species.

Inadvertent social information in foraging bumblebees: effects of flower distribution and implications for pollination

Pollinators are confronted with spatial heterogeneity in floral resources. Acquiring information about their environment enables pollinators to exploit floral resources optimally, and hence strongly affects their fitness. Despite their evolutionary importance, spatial resource distribution and social information use have never been addressed in a plant–pollinator framework. We experimentally tested the effects of inadvertent social information (ISI) use (without or with an experienced demonstrator) in relation to spatial flower distribution (random or patchy) in foraging Bombus terrestris . We studied naïve bees' associative learning when confronted with rewarding and nonrewarding flowers that differed slightly in colour, reproducing the natural mimicry of deceptive species. Naïve bees were more efficient foragers with than without a demonstrator in the patchy flower distribution, suggesting that they used ISI from conspecifics, probably in the form of social attraction at the patch scale. In the random flower distribution, however, ISI did not improve performance of naïve bees, and could not replace individual sampling. Analyses of nonrewarding intraspecific transitions indicate that the efficiency of naïve bees in deceptive species pollination was lowered when rewarding and nonrewarding inflorescences were patchily distributed and especially when bees relied on ISI. Thus, spatial flower distribution and ISI availability interact in determining pollinator behaviour, which has important consequences for plant community pollination. This suggests that bumblebees are adapted to exploit patchily distributed resources socially, a characteristic that might be exploited by deceptive plant species. By coupling behavioural and community ecology, this study offers promising perspectives for cognitive ecology studies in plant–pollinator interactions.

Facultative autogamy in Cyrtopodium polyphyllum (Orchidaceae) through a rain-assisted pollination mechanism

Cyrtopodium includes ~42 species, among which is Cyrtopodium polyphyllum (Vell.) Pabst ex F. Barros that occurs in a rainforest in south-eastern Brazil. Its non-rewarding flowers, which attract Centridini bees by deceit, are rain-assisted self-pollinated, a phenomenon rarely found in orchids and other plant families. In addition, self-pollination has never been reported in Cyrtopodiinae and data on the pollination of South American orchids are scarce. Flowers were observed at different times of the day, on both sunny and rainy days, to record floral morphology, visitors and the effects of rainfall on flowers. On rainy days, water accumulates on the stigma and dissolves the adhesive substance of the stigmatic surface. A viscous drop thus forms, which contacts the pollinarium. When evaporation makes the viscous drop shrink, the drop moves the pollinarium with the anther onto the stigmatic surface and promotes self-pollination. Fruit set in natural habitat was low, with 2.4% at one study site, where a similar value (2.2%) was recorded in flowers self-pollinated by rain. In C. polyphyllum, facultative self-pollination assisted by rain is thus an important strategy that guarantees fruit set when pollinator’s visits are scarce, which is common in species pollinated by deceit.

Scientific approaches to Australian temperate terrestrial orchid conservation

This review summarises scientific knowledge concerning the mycorrhizal associations, pollination, demographics, genetics and evolution of Australian terrestrial orchids relevant to conservation. The orchid family is highly diverse in Western Australia (WA), with over 400 recognised taxa of which 76 are Declared Rare or Priority Flora. Major threats to rare orchids in WA include habitat loss, salinity, feral animals and drought. These threats require science-based recovery actions resulting from collaborations between universities, government agencies and community groups. Fungal identification by DNA-based methods in combination with compatibility testing by germination assays has revealed a complex picture of orchid–fungus diversity and specificity. The majority of rare and common WA orchids studied have highly specific mycorrhizal associations with fungi in the Rhizoctonia alliance, but some associate with a wider diversity of fungi. These fungi may be a key factor influencing the distribution of orchids and their presence can be tested by orchid seed bait bioassays. These bioassays show that mycorrhizal fungi are concentrated in coarse organic matter that may be depleted in some habitats (e.g. by frequent fire). Mycorrhizal fungi also allow efficient propagation of terrestrial orchids for reintroduction into natural habitats and for bioassays to test habitat quality. Four categories of WA orchids are defined by the following pollination strategies: (i) nectar-producing flowers with diverse pollinators, (ii) non-rewarding flowers that mimic other plants, (iii) winter-flowering orchids that attract fungus-feeding insects and (iv) sexually deceptive orchids with relatively specific pollinators. An exceptionally high proportion of WA orchids have specific insect pollinators. Bioassays testing orchid-pollinator specificity can define habitats and separate closely related species. Other research has revealed the chemical basis for insect attraction to orchids and the ecological consequences of deceptive pollination. Genetic studies have revealed that the structure of orchid populations is influenced by pollination, seed dispersal, reproductive isolation and hybridisation. Long-term demographic studies determine the viability of orchid populations, estimate rates of transition between seedling, flowering, non-flowering and dormant states and reveal factors, such as grazing and competition, that result in declining populations. It is difficult to define potential new habitats for rare orchids because of their specific relationships with fungi and insects. An understanding of all three dimensions of orchid habitat requirements can be provided by bioassays with seed baits for fungi, flowers for insects and transplanted seedlings for orchid demography. The majority of both rare and common WA orchids have highly specific associations with pollinating insects and mycorrhizal fungi, suggesting that evolution has favoured increasing specificity in these relationships in the ancient landscapes of WA.