Grape aroma brings fruit and wine to mind. Yet the small flowers that precede the fruit also contain volatile compounds. Are differences in these compounds connected to relationships with pollen-carrying insects?
A study published by Ahmed G. Darwish and colleagues in 2025 focused on muscadine grapes grown in the southeastern United States. It compared genotypes bearing female flowers with those bearing hermaphroditic flowers, which have both stamens and pistils, to find clues to floral chemistry and pollination mechanisms.
Extracting compounds from flowers of 16 genotypes
The study examined 16 genotypes: eight with female flowers and eight with hermaphroditic flowers. Hermaphroditic muscadine flowers can be fertilized by their own pollen, whereas female-flowered plants must receive pollen from other plants.
Collected flowers were pooled into three sets of samples per genotype, stored frozen, dried, and ground into powder. The powder was heated with saline at 60°C, and the gas accumulating above it was collected and analyzed by GC–MS.
This method captures compounds extractable from processed flower samples. It does not measure the scent that intact flowers release into the air in the vineyard. Comparisons based on analytical signals need to be distinguished from the odor intensity insects experience in nature.
Compounds abundant in hermaphroditic flowers or characteristic of female flowers
Hermaphroditic genotypes had, on average, more detected compounds per genotype and a higher analytical measure of total abundance. There were nevertheless substantial differences among genotypes within each floral type.
The researchers examined terpenes, a group of compounds involved in floral scent, in particular detail. Statistical comparisons of their overall combinations generally separated female and hermaphroditic flowers into different groups. There were exceptions, however, including hermaphroditic ‘Noble’ and female ‘Majesty’, so floral type did not determine everything.
Eleven compounds were selected as candidates contributing to the distinction between the two types. For example, β-elemene and β-bisabolene were characteristic of hermaphroditic flowers, whereas α-selinene was among those characteristic of female flowers. These candidates help distinguish types within this dataset; they were not confirmed to attract insects.
The flowers also differed visually. Inflorescences averaged about 44 florets in female genotypes and about 169 in hermaphroditic ones. The authors suggest that scent combinations, flower numbers, and rewards such as pollen may together make hermaphroditic flowers easier for insects to find.
What happened to the fruit after hand pollination?
In a separate experiment, the researchers used five plants each of the female cultivar ‘Darlene’ and the hermaphroditic cultivar ‘Floriana’. They selected 20 inflorescences on each plant, leaving half to be naturally pollinated and bagging the other half before hand-pollinating them with pollen from another cultivar, ‘Granny Val’. Stamens were also removed from the hand-pollinated hermaphroditic group to prevent self-pollination.
In Darlene, hand pollination increased berry number per cluster by about 60%. Individual berries were about 33% lighter, however, and the change in total cluster weight was small. This was therefore not a 60% increase in yield. Floriana showed no significant changes in these measures.
The authors interpret this as evidence that Darlene has fruit-setting potential not fully realized under natural pollination conditions. Only one cultivar of each floral type was tested, however. Hand pollination also involved bagging and, in Floriana, stamen removal, not simply an increase in pollen supply.
Whether scent attracted insects remains to be tested
The authors propose an explanation in which floral scent, appearance, and rewards such as pollen and nectar together shape flower–insect interactions. They also suggest that pollen presence and differences in floral structure may underlie the chemical differences between female and hermaphroditic flowers.
For α-selinene, characteristic of female flowers, they discuss a possible reduction in attraction because bee-repellent effects have been reported in other plants. This study did not present the compound to insects or observe visitation frequencies. Whether it has the same effect in muscadine, or how it acts in mixtures with other compounds, remains unconfirmed.
Compound identities were inferred by matching instrument signals to databases; the study did not include human sensory evaluation. The chemical differences and hand-pollination results do not by themselves demonstrate that scent causally determined pollination.
Summary
Female and hermaphroditic muscadine flowers differed in their compound combinations. In addition, hand pollination increased berry number in one female cultivar, suggesting that natural pollination conditions may not fully realize its fruit-setting potential.
Connecting these findings is the authors’ hypothesis that signals to insects involve not only scent but also flower numbers and pollen rewards. Which insects actually respond to the candidate compounds, and how? Testing those questions could provide basic information for understanding flower–pollinator relationships under changing environmental conditions and supporting crop pollination.
References
Darwish, A. G., Das, P. R., Olaoye, E., Gajjar, P., Ismail, A., Mohamed, A. G., Tsolova, V., Hassan, N. A., El Kayal, W., Walters, K. J., & El-Sharkawy, I. (2025). Untargeted flower volatilome profiling highlights differential pollinator attraction strategies in muscadine. Frontiers in Plant Science, 16, 1548564. DOI: 10.3389/fpls.2025.1548564 / Full text.
The research was supported by the Viticulture Advisory Council of the Florida Department of Agriculture and Consumer Services. The authors declared no commercial or financial relationships that could represent a conflict of interest. Original article: © 2025 Darwish et al., Creative Commons Attribution (CC BY).
