Bees and butterflies tend to stand out when we think of insects that carry pollen. But some flies visit flowers with odors that people find unpleasant and support plant reproduction. In a study of a small South African flower, the main actors were little-known flies that often rest on rocks.
Annemarie Heiduk and colleagues investigated pollination in Ceropegia pulchellior, which grows in rocky grassland. They asked which insects carry pollen and whether insects are necessary for reproduction. The study also explored the roles of scent, color, and nectar. The paper was published online at the end of 2024 and included in a 2025 volume and issue.
Distinguishing Flower Visitors from Pollen Carriers
This plant grows in a restricted sandstone region of South Africa. Its stems extend along the ground and rocks, bearing dark reddish-purple flowers with pale streaks. Some related plants have flowers that imprison insects, but this species has open flowers without a pouch-like trapping mechanism.
Over several flowering seasons from 2014 to 2024, the researchers observed wild populations and cultivated plants. They collected insects visiting the flowers as well as flies resting on nearby rocks and examined whether pollen-carrying structures were attached to their bodies.
The insects observed visiting the flowers were all flies. Particularly important were members of the genus Cestrotus in the family Lauxaniidae. Known in English as “rocksitter flies,” these flies are found on rocks, and the genus had not previously been known to act as pollinators. Some members of the family Muscidae also carried pollen.
Rather than adhering as loose grains, this flower’s pollen is arranged in two masses joined by a clip-like structure. The clip attaches to a fly’s mouth, and pollen is transferred when a mass is inserted into the receiving opening of another flower. The collected flies carried not only complete structures but also clips missing one or both pollen masses. This offers evidence of a role in pollen transport beyond simply visiting flowers.
No Fruit Formed When Flies Were Excluded
To test dependence on insects, the researchers covered 15 plants with fine mesh and compared them with 15 plants that insects could visit freely. No fruit formed on the covered plants.
Even under natural conditions that allowed insect visits, however, many flowers did not become fruits. The mean of the fruit-set proportions calculated for individual plants was approximately 8% and approximately 5% in the two flowering seasons examined. Having insects that transport pollen is different from every flower setting fruit.
The researchers also examined what proportion of the pollen masses removed from flowers reached a flower’s receiving opening, finding substantial variation between measurement days. The authors suggest that the types of flies visiting, or the extent to which they visit flowers of the same species in succession, may differ from day to day.
Who Is the Dung-Like Odor Signaling To?
When compounds released into the air by flowers from four plants were collected and analyzed by GC–MS, the main components were fatty acids such as acetic acid, butyric acid, and propionic acid, together with p-cresol. Small amounts of indole were also present. The researchers describe the floral odor as resembling fresh herbivore dung.
These compounds are also found in decomposing organic matter. By contrast, some sulfur-containing compounds well known from carrion-mimicking flowers were not detected in this flower. The authors suggest that, rather than broadly attracting flies, this distinctive combination may provide a cue that attracts particular flies.
However, the study did not test attraction by removing the scent or presenting an artificial blend. Much also remains unknown about the food and egg-laying sites of Cestrotus flies, and what they perceive the flower to be when visiting it has yet to be established.
Measurements of flower color applied to an existing model of fly color vision suggested that the flowers may be difficult to distinguish chromatically from the rocks in the background. This, too, was not a test of vision in actual Cestrotus flies, but the authors regard it as evidence supporting the hypothesis that the flowers rely on scent rather than conspicuous color.
A Small Reward Beyond the Mimicked Odor
This flower does more than lure flies with scent. Examination of one flower from each of 56 plants found a very small amount of nectar, averaging approximately 0.07 microliters, in cavities near the center of the flower. The pollen-bearing structure is thought to attach to a fly’s mouth as it probes for this nectar.
The authors discuss the possibility that nectar prolongs the flies’ visits and exploration, increasing opportunities for pollen transfer. The flowers could attract flies with an odor suggestive of food or an egg-laying site while also providing an actual reward. This suggests a combination different from a completely rewardless “deceptive flower.” Experiments that alter the amount of nectar to test its effects remain a task for future work.
Summary
This study showed that rock-dwelling Cestrotus flies support reproduction in a flower restricted to a limited region. Four species not yet formally described were also identified among the collected flies, showing how tracing a relationship with a plant can lead to previously unknown diversity in the insects themselves.
A combination of dung-like compounds and a tiny amount of nectar creates a system that attracts particular partners and enlists them to carry pollen. This is the authors’ proposal. Investigating which parts of the odor the flies respond to, and whether related flowers also depend on the same flies, could provide a more concrete understanding of overlooked pollinators and floral diversification.
References
Heiduk, A., Gaimari, S., Whittington, A., Johnson, S. D., & Shuttleworth, A. (published online in 2024; volume and issue dated 2025). Rocksitter Flies (Lauxaniidae: Cestrotus) are Key-Pollinators of Ceropegia pulchellior—A Threatened and Localised South African Endemic With Foetid-Scented Flowers. Journal of Applied Entomology, 149(6), 888–900. DOI: 10.1111/jen.13392 / Full original paper.
The original paper states that the authors received no specific funding for the research, while its acknowledgments report support through a postdoctoral scholarship from South Africa’s NRF SARChI initiative. The authors declare no conflicts of interest. Original paper license: CC BY-NC-ND 4.0.
