Modern agriculture is overly-dependent on honeybees for pollination. Considering the current decline in honeybee populations worldwide, this is an ill-fated limitation. Alternative pollinators surely exist among the 30,000 other bee species, but unfortunately basic biological information for most of them is lacking. Unlike honeybees, most wild bees are solitary, not social. Many also have a specialized diet and are limited to collecting pollen from only a few plant species, even when others are available. For example, the bee genus Diadasia comprises ~30 species in North America. Each species uses only one of the following plant families as a pollen-host: mallows, cacti, morning glories, evening primroses, or sunflowers. DNA analyses suggest that the ancestor of Diadasia collected only mallow pollen, but at some point in the past a few switched their loyalties to one of the other four plant families. This project attempts to discover the mechanisms that 1) maintain these bees' specialized diets, and 2) mediate or constrain host-switching events. How are novel hosts 'chosen' by specialists? This project tests the hypothesis that specialist bees have a limited ability to recognize visual and odor cues of different flowers, and that they are therefore unable to recognize non-host flowers as potential pollen resources. The project also tests the prediction that the five families of host plants used by these bees share visual or chemical (i.e. scent) characteristics. First, analyses of the floral scent chemistry of host and co-blooming non-host plants will be conducted. Second, electroantennography will be used to discern which flower scent compounds can be recognized by specialist bees. And finally, behavioral trials will be conducted to assess the relative roles of visual and odor cues. Understanding these mechanisms will enhance the conservation of native bee and plant biodiversity, and provide necessary information for developing specialized crop pollinators.
Bees are by and large the only insects that purposefully gather pollen from flowering plants. There are over 20,000 species of bees worldwide. Most well-known among these are the honey bees and bumble bees, but they account for less than 1% of all bee species. What is more, honey bees and bumble bees share little in common with other bees in terms of their life histories. Relevant to this study, many bees are specialists; picky eaters who limit themselves to only a few flower species that may be blooming at any given time when collecting pollen. How do these bees distinguish between their ‘host plants’ and all of the other flowers that may be blooming at any given time? From studies of a handful of bee species, we know that they rely heavily on specific scent components of a flower, as well as the color and shape of the flower. For our study, we examined the visual and scent cues that may be attracting one particular genus of specialist bees (Diadasia) to their host plant. This genus was of special interest because, from previous research on the evolutionary history of Diadasia, it is known that the common ancestor to all species of this bee specialized on Mallow (Malvaceae) flowers. At some time in the past, certain individuals radiated, or switched hosts, onto Cactus (Cactaceae), Morning-Glory (Convolvulaceae), Sunflower (Asteraceae), and Farewell-to-Spring (Onagraceae) flowers. Out of all of the plants onto which these bees could have radiated, why did they pick these particular flowers? Was it something about the scent and visual appearance of the flowers, and was it related to how bees distinguish host plants from other flowers? We focused our research on just Cactus and Mallow flowers, and NSF provided funding specifically for the scent portion of our research. We first determined the chemical compounds that make up the floral scent of flowers within the cactus and mallow families that are visited by Diadasia. We compared this to the compounds making up the floral scent of co-blooming non-host flowers. We found that several compounds were in common to both mallow and cactus flowers that host Diadasia, including nonanal, decanal, trans-beta-ocimene, and trans-trans-alpha-farnesene. Interestingly these same compounds were found in non-host flowers; in other words they were not unique to the Diadasia host plants, and likely are not the sole attractor for Diadasia. However, we found that non-host plants also contained a large suite of additional compounds that were never found in host flowers. Since it was unknown whether a bee antenna was capable of actually recognizing any of these compounds, we next tested bee responses. By using an electro-antennographic machine, coupled to a gas chromatograph, we were able to assess the ability of a bee antenna to detect each individual compound produced by a flower. If the bee is capable of detecting a compound, an electrical response is generated by the antennae, and can be recorded, along with the compound that elicited that response. We found that two species of Diadasia (a cactus specialist and a mallow specialist) were capable of recognizing only a handful of the 50-100 compounds produced by their host flowers, but that chief among them were the common and abundant compounds, shared by both cactus and mallow flowers, mentioned earlier (nonanal, decanal, etc.). Moreover, Diadasia are capable of recognizing many compounds that do not occur in their host flowers, including linalool (a compound very common in many bee-pollinated flowers around the world), isoeugenol, and citronellol. Finally, we wanted to determine if the compounds that were recognizable by Diadasia acted as attractants or repellants in the field. We added small doses of compounds to host flowers, and monitored changes in the visitation rates of Diadasia before and after the addition of each compound. The compounds that naturally occurred in the host flowers (nonanal, decanal, trans-beta-ocimene, and alpha-alpha-farnesene) did not cause changes in visitation rates: the bees did not find these compounds to be repellant. However, the addition of compounds that occur in non-host flowers significantly decreased visitation rates. For example, the addition of linalool to mallow flowers caused bee visitation to drop from an average of 6 visits per 20 minutes to less than 2 visits per 20 minutes. Similar results were found for the addition of isoeugenol and citronellol. In conclusion: it appears that Diadasia are capable of detecting a wide array of chemical compounds, not only those that occur in their host plants, but also many that occur in non-hosts. However, Diadasia appear to be repelled by those compounds that are not naturally occurring in their host flowers. This may explain why they do not visit the many flowers that typically co-bloom with their hosts (to a Diadasia, these other flowers ‘stink’), and it also may explain how Diadasia switched hosts in the past.
