What comes to mind when you hear rose fragrance? Real flowers can have not only a classic rosy character but also fruity or lemon-like qualities. Are flowers that smell more typically rosy also more pleasant and more inviting to smell again?
Inès Adrar and colleagues compared people’s evaluations of 10 fresh rose cultivars with the compounds released by the flowers. The study was published online in late 2024 and appeared in a 2025 journal issue. Its aim was to investigate the complex mixtures in fresh-flower scent, which are difficult to capture using essential oils or individual fragrance molecules alone.
- Nineteen people evaluated flowers hidden from view
- Rose typicality was related to pleasantness and familiarity
- Collecting and analyzing emissions from fresh flowers separately
- Familiar rose ingredients did not explain everything
- We cannot say that adding one ingredient makes a scent more pleasant
- Mixing may change an ingredient’s role
- Positive associations with roses are not the same as improved mood
- Conclusion
- References
Nineteen people evaluated flowers hidden from view
Twenty people aged 18–35 from the Lyon region of France participated. One evaluated only two of the 10 cultivars and was excluded, leaving 19 in the analysis.
The 10 cultivars were selected to cover a wide range of scent characteristics. Flowers grown in Lyon rose gardens were placed in opaque containers so participants could not see them. Evaluations concerned scent alone, without the flowers’ color or shape.
Participants freely smelled each sample for 1 minute while rating eight attributes: rose typicality, pleasantness, desire to smell again, familiarity, intensity, fruitiness, lemon character, and floral character. Sample order was randomized separately for each participant.
Rose typicality was related to pleasantness and familiarity
The 10 cultivars differed in how typically rosy they smelled. Even after accounting for differences in how participants rated them, rose typicality was positively related to pleasantness, desire to smell again, and familiarity.
The correlation with pleasantness was about 0.25. There was an association, but it was not so strong that the two ratings were effectively the same. Rose typicality alone did not determine all of people’s liking.
In contrast, rose typicality had no statistically clear relationship with scent intensity. Flowers releasing a strong scent were not necessarily perceived as more typically rosy.
Fruitiness was negatively associated with rose typicality. Cultivars with a strong fruity character may have been somewhat further from these participants’ idea of a typical rose scent.
Collecting and analyzing emissions from fresh flowers separately
The researchers collected compounds released into the air by open flowers and analyzed them using gas chromatography–mass spectrometry, a method that separates compounds and examines molecular characteristics. They used 3–6 biological replicates per cultivar and detected 97 volatile compounds in total.
They measured what the flowers actually emitted. This was not the composition of an essential oil obtained by distillation or an extract made by crushing or heating petals.
Cultivars differed in both the total amount emitted and its composition. Those with higher total emissions tended to be rated as more intense and familiar. However, these chemical–perceptual comparisons used average values for each of the 10 cultivars. Repeated chemical analyses were not treated as a large number of independent cultivars.
Familiar rose ingredients did not explain everything
Geraniol, nerol, citronellol, and phenylethyl alcohol are often mentioned when discussing rose scent. Yet in this comparison of fresh flowers, these alone could not explain which cultivars smelled more typically rosy.
The authors focused on ionones and a group of fatty-acid-derived compounds called oxylipins. Some ionones can contribute to odor at very small quantities. Oxylipins include compounds associated with green-leaf impressions.
In this analysis, quantities and proportions of ionones emerged among the candidates associated with rose typicality. Several oxylipin compounds also became candidates for investigating that relationship.
One candidate associated with rose typicality was dihydro-β-ionone. The paper describes its odor in isolation as reminiscent of violet, wood, amber, and fruit, among other qualities. It is intriguing that studying roses brings other flowers and wood into the discussion.
Within a mixture, a compound may play a different role from the impression it gives when smelled alone. These findings offer clues for exploring such combinations. Experiments adding or removing compounds are needed to establish a formulation that enhances rose typicality.
We cannot say that adding one ingredient makes a scent more pleasant
The authors suggest that a pleasant rose scent may depend on a balance in the amounts and proportions of ionones, oxylipins, phenylethyl alcohol, and related compounds. Phenylethyl alcohol, abbreviated PEA in the paper, is a well-known rosy-smelling ingredient.
However, this explanation of the mixture is not yet an established recipe. The relationship between overall compound-group composition and pleasantness had a p value of 0.126 when tested using quantities and 0.051 using proportions, neither reaching the conventional 5% significance threshold.
Furthermore, correlations with individual compounds were not tested for statistical significance because there were only 10 cultivars and many compounds to examine. The researchers treated correlations exceeding their chosen coefficient thresholds as candidates for discussion.
It therefore cannot be said that three compound groups were demonstrated to increase pleasantness. The associations can narrow down candidates, but the next step is to add or remove compounds and test how people’s ratings change.
Mixing may change an ingredient’s role
Here, monoterpenes as a group were associated more with fruity and lemon-like impressions than with typical rose character. This group includes compounds that can smell rosy on their own.
Based on this finding, the authors argue that understanding a whole flower’s scent simply by adding together the impressions of familiar individual compounds has limits. If one compound enhances or masks another, the overall impression can change when they are mixed.
An important implication for recreating fresh-flower scents and breeding fragrant roses is that combinations involving other compounds deserve attention, not just the quantities of well-known rose ingredients. This comparison identifies candidates for experimental work on those combinations.
Positive associations with roses are not the same as improved mood
After completing all the scent ratings, participants freely described what the smell of roses brought to mind. Many responses involved positive emotions, cosmetics, or nature-related associations such as gardens.
This explored the memories and images participants associated with roses. It did not measure improvements in mood or reductions in stress before and after the experiment, and cannot be read as proof of health benefits from fresh-flower scent.
The sample also consisted of 19 young adults from the Lyon region and 10 selected cultivars. Expectations of a typical rose scent may vary with culture and experience, so testing people elsewhere and a wider range of cultivars is needed.
The research was supported by a research program of the Auvergne-Rhône-Alpes region. The acknowledgments also mention participation and support from rose breeder Meilland International and flower material from Lyon parks. The authors declared no competing interests.
Conclusion
When 10 fresh rose cultivars were evaluated without seeing the flowers, rose typicality was associated with pleasantness, desire to smell again, and familiarity. Chemical analyses identified combinations involving ionones and oxylipins, not just familiar rose ingredients, as candidates for further testing.
The study’s significance lies in broadening the focus from finding molecules that smell of roses to understanding how people perceive the mixture emitted by a fresh flower. The authors see correlations as a starting point for addition and omission experiments that could inform scent recreation, rose breeding, and research on the perception of complex odor mixtures.
References
Adrar, I., Hervé, M., Mantel, M., et al. (2025; published online in 2024). Why do we like so much the smell of roses: The recipe for the perfect flower. iScience, 28(2), 111635. DOI: 10.1016/j.isci.2024.111635. Full original paper. © 2024 The Author(s). Published by Elsevier Inc. The original paper is available under CC BY-NC 4.0.
