Search

Long-proboscid fly (LPF) pollination has been recorded in multiple plant lineages in South Africa. However in the large and morphologically diverse genus Erica, evidence for LPF pollination is limited to predictions based upon floral syndromes as well as anecdotal observation and limited citizen science records. Here we use breeding system experiments, pollinator exclusion, remote camera traps, spectral analysis and field observation in the Klein River Mountains, Western Cape Province, South Africa, to demonstrate this mode of pollination in the range-restricted taxon Erica aristata. This species has long narrow corolla tubes with narrow opening and contrasting white and crimson pink petals. Flowers were visited and pollinated exclusively by one long-proboscid fly species, Prosoeca rubicunda (Nemestrinidae) over four years at this study site. Breeding experiments showed that E. aristata is not self-compatible and pollinator exclusion experiments indicate that pollinator visits were required for seed-set. Low nectar volume and moderately high nectar sugar concentration were consistent with other long-proboscid fly-pollinated plant species in the Cape. This study represents the first confirmed instance of long-proboscid fly-pollination in the genus Erica.

In the Cape flora, pollination by small mammals is well known in the Proteaceae and has recently been documented in the Ericaceae. Erica lanuginosa , a Klein River Mountains (southern Overberg) endemic, has unusual pendulous, acorn-shaped flowers unlike those of a typical Erica as the lobes of the corolla are tightly pressed together at the tips to form a sharp closed “beak”. These unusual floral traits suggested the possibility of autonomous self-pollination in flowers that don't open (cleistogamy) or rodent-pollination. We used digital camera traps, mammal trapping and pollinator exclusions to test these alternative hypotheses. We show that Erica lanuginosa , is pollinated by small mammals, the second documented rodent-pollinated Erica species. Camera traps recorded fourteen images of three species of rodents visiting flowers on seven separate days to three different plants, with the majority of visits by Acomys subspinosus . PVC Sherman-type traps were used to capture rodents and examine their scats for the presence of pollen. Pollen tetrads typical of Ericaceae were observed in droppings of all three species. Flowers of Erica lanuginosa covered with pollinator exclosure bags set no seeds while the proportion of open, control flowers that set seed was high, indicating that pollinators are essential for seed set in E. lanuginosa . These data are consistent with rodent-pollination in E. lanuginosa and there is no evidence for cleistogamy in this species.

Bowl-shaped inflorescences, geoflory, dull-coloured flowers and winter flowering suggest that Protea foliosa is adapted for rodent-pollination. To test this hypothesis, rodents were trapped in a large P. foliosa population near Grahamstown and examined for the presence of pollen on their snouts and in their scats. Camera traps, modified for near focus (

In his landmark work on the pollination biology of South African plants in 1954, Stefan Vogel described the deposition of Habenaria epipactidea (= H. polyphylla ) pollinaria on the forelegs of the hawkmoth Hippotion celerio . The discovery of a large, well-pollinated population of H. epipactidea in the Eastern Cape allowed us to confirm the presence of this unusual pollen placement on a number of species of shorter-tongued hawkmoths. The long-tongued species Agrius convolvuli is likely to function as a nectar thief as the length of the tongue of this species relative to the nectar spur ensures that the forelegs are unlikely to come into contact with the viscidia. The legitimate hawkmoth pollinators removed a large proportion of pollinaria from the flowers and the majority of flowers had pollen deposited on their stigmas. Despite this, pollen transfer efficiency was relatively low at 8.4%. We also examined the scent production of H. epipactidea and show that volatile production increases markedly after 18:00, peaking at around 20:00. The scent of this species is dominated by methyl benzoate which makes up 99.4% of the floral scent at 20:00. The osmophores are located on the lateral petals and labellum, confirming Vogel's observations in the Soutpansberg (Zoutpansberg).

Autonomous self-pollination is surprisingly common among orchids and is thought to provide reproductive assurance when pollinators are scarce. During investigations of the reproductive biology of the orchid genus Eulophia, consistently high rates of capsule set were observed in Eulophia clavicornis var. clavicornis, E. c. var. inaequalis, E. c. var. nutans and E. tenella. A breeding system experiment showed that E. c. var. nutans is capable of autonomous self-pollination. Emasculated flowers of this taxon did not set fruit, suggesting that agamospermy is unlikely. The likely mechanism of autonomous self-pollination in these taxa was identified as the partial or complete absence of rostellum tissue, allowing contact between pollinia and stigmatic fluid, and thus for pollen tubes to grow from in situ pollinia to the ovules. In some individuals, basal flowers on an inflorescence possess intact rostellae and functional pollinaria, whereas distal flowers lack pollinia. Neither of these two flower classes set capsules. A few individuals of the otherwise outcrossing E. zeyheriana, which normally have well-developed rostellae, show evidence of autonomous self-pollination resulting from interrupted rostellae in these plants. Other outcrossing Eulophia species (E. speciosa and E. streptopetala) sometimes show high levels of fruit set, seemingly without insect visitation. However, investigations showed that these are pseudo-fruits lacking seeds and are a result of insect parasitism. Therefore, high levels of fruit set alone should not be used to infer autonomous self-pollination in orchids.

The shape of flowers frequently corresponds to the morphology of pollinators but some floral traits may also function to prevent non-pollinating flower visitors from stealing flower rewards. Despite the presence of such structures few studies have demonstrated their efficacy in limiting the nectar intake by nectar thieves. The flowers of Strelitzia reginae are regularly visited by sunbirds that do not effect pollination and act solely as nectar thieves. In this species, the nectary is covered by the convoluted bases of the petals (“nectar barriers”). In this study we investigate how non-pollinating sunbirds interact with these nectar barriers and whether nectar barriers play a role in limiting the amount of nectar sunbirds can steal. We quantified the volume of nectar that sunbirds consume while visiting flowers where nectar barriers were present and in flowers where these were experimentally removed. We found that sunbirds consume a median of 106.8 µl of nectar when visiting flowers with nectar barriers present and consumed a significantly greater volume of nectar (median = 158.03 µl) in flowers without nectar barriers. These results suggest that the convoluted petals that cover the nectary of S. reginae may function to reduce nectar theft but are likely to be more effective against insect nectar thieves. This is one of the first studies to quantitatively demonstrate the role of flower features that may function to limit nectar theft.

Little is known about the pollination biology of the large (c. 230 species) African orchid genus, Eulophia. Here we report the discovery of pollination by flower chafer beetles (Cetoniinae; Scarabaeidae) in two color forms of E. ensata and in E. welwitschii. Both species have congested, capitate inflorescences, traits that are generally associated with pollination by flower chafer beetles in Eulophia and other plant genera. Pollinarium reconfiguration, including pollinarium bending and anther cap retention, in these beetle-pollinated species is slow. Such slow reconfiguration is predicted by Darwin's hypothesis to be a mechanism that limits geitonogamous self-pollination by slow moving beetles. A breeding system experiment conducted on E. welwitschii showed that this species, like most others in the genus, is self-compatible, but is dependent on pollinators for fruit set. As all Eulophia species are non-rewarding, the basis of attraction of beetles to flowers of the study species seems to be the generalized resemblance of their inflorescence in terms of flower arrangement and color to the capitula of sympatric rewarding Asteraceae that are utilized as food or rendezvous sites (or both) by flower chafer beetles.

Mating success in plants depends largely on the efficiency of pollen dispersal. For hermaphrodite plants, self-pollination, either within or among flowers, can reduce mating opportunities because of pollen and ovule discounting and inbreeding depression. Self-pollination may be particularly detrimental in plants such as orchids and asclepiads that package each flower's pollen into one or more pollinia which, together with accessory structures, comprise a pollinarium. Darwin proposed that physical reconfiguration of pollinaria serves as a mechanism for reducing the likelihood of self-pollination. To be effective, the time taken for pollinarium reconfiguration would need to exceed that spent by a pollinator on a plant. We investigated pollinarium reconfiguration (including pollinarium bending, pollinium shrinking and anther cap retention) in 19 species and found a strong positive relationship between reconfiguration time and the duration of pollinator visits. Reconfiguration times were also consistently longer than pollinator visit times. These results provide strong support for Darwin's idea that this mechanism promotes cross-pollination.