How to demand for perfect flawless roses takes its toll on workers and the planet
For more than a century, the global rose breeding industry has been waging war on nature in pursuit of the ideal flower. The casualties include wild species, pollinating insects, small-scale growers, a flower's most fundamental quality — its scent — and the genetic diversity that underpins the entire genus.
"The cause for the loss of fragrance in these flowers remains unknown, but it does not seem to conflict with the increase in vase life." — peer-reviewed scientific literature, 2022, on the rose breeding industry's centuries-long pursuit of the perfect cut flower
There is something philosophically peculiar about the rose's position in human culture. Of all the objects on earth to which we have assigned the meaning of beauty — of love, of perfection, of the inexpressible — we have chosen a flower that evolved specifically to attract insects. Its colour was not designed for human eyes. Its scent was not composed for human nostrils. The double flowers we prize most highly — the dense, multi-petalled globes of hybrid teas and English roses — are developmental abnormalities, the stamens and pistils transformed into additional petals by a genetic modification that renders the plant entirely unable to reproduce on its own. The bee has nowhere to land. The pollen, in many varieties, does not exist. The rose we have created in our image is, from the perspective of the organism itself, a catastrophe.
This is not a mere philosophical irony. It is the starting point for a story about what happens when a global industry dedicates itself to the production of cosmetic perfection, and what that dedication costs — in genetic terms, in ecological terms, in the terms of the wild species and wild places that have been diminished or destroyed to produce it, and in the human terms of growers in Kenya and Ecuador who pay royalties to breeders in Germany and France for the right to grow patented plants on their own land.
The story of rose breeding is, in one light, a story of remarkable human ingenuity: the creation, over three centuries of patient hybridisation and selection, of more than 30,000 cultivated varieties from a starting pool of fewer than ten wild species. But it is also a story of extraordinary narrowness — of an industry that has pursued a specific, commercially-defined idea of perfection so single-mindedly that it has, along the way, stripped fragrance from the world's most beloved flower, created genetic monocultures so uniform that they are critically vulnerable to disease, exported invasive species across continents, used radioactive cobalt and carcinogenic chemicals to force mutations in plant tissue, built a patent system that extracts royalties from subsistence-level growers in the developing world, and contributed to the decline of wild rose species that are irreplaceable reservoirs of the genetic diversity the industry now desperately needs to survive.
Thirty Thousand Varieties That Smell of Nothing
The easiest way to understand what the rose breeding industry has done to the rose is to visit a large commercial flower shop and find a red hybrid tea rose — the long-stemmed, cosmetically perfect variety that dominates the global market — and smell it.
Nothing. Or almost nothing: a faint, generic floral note that bears no resemblance to the complex, layered, deeply particular fragrance that made roses the symbolic heart of Western perfumery for centuries. Rosa damascena — the Damask rose, cultivated in Bulgaria and Turkey and the Middle East since the fourteenth century — produces a scent whose chemical complexity has been the subject of serious scientific analysis for decades: hundreds of distinct volatile compounds, including terpenoids, phenylpropanoids, and fatty acid derivatives, combining to produce something so multidimensional and alive that no synthetic substitute has been able to replicate it to the satisfaction of the perfume industry, which still pays premium prices for real rose essential oil. The old garden roses — albas, gallicas, centifolias, mosses — have scent profiles that range from pure rose to tea, fruit, myrrh, aniseed, and combinations that have no name in any language. They were grown for centuries partly because of those scents, which were considered among their most important qualities.
Modern commercial cut flower roses have, as a class, very little scent. This is not a secret. It is documented in the peer-reviewed scientific literature with a candour that reflects the degree to which the industry itself acknowledges the problem, even as it continues to breed for the traits that caused it. Breeding programmes for commercial cut roses have focused for centuries on longevity, vase life, stem length, uniformity, colour saturation, and resistance to the physical damage of transport. Fragrance has not only been deprioritised; it has been actively selected against, because some of the biochemical pathways that produce scent compounds also produce compounds that accelerate the ageing of the petals. A rose that smells strongly may not last long in a vase. A rose that lasts ten days in a vase may have traded away the metabolic resources for scent production in order to support the longevity of its petals. The commercial market has, consistently and over a very long period, preferred the latter.
The precise mechanism of fragrance loss in commercial roses is, remarkably, still not fully understood. A 2022 review in the International Journal of Molecular Sciences noted that "the cause for the loss of fragrance in these flowers remains unknown." What is known is that the genetic pathways responsible for scent compound biosynthesis — the terpenoid pathway, the phenylpropanoid pathway, the fatty acid derivative pathway — are present in modern roses but not fully expressed in most commercial cultivars. The biochemical machinery for fragrance exists but is largely switched off, the result of generations of selection pressure that favoured appearance and longevity over olfactory complexity. The rose still has the genes to smell. The breeding programme has instructed it not to use them.
The human cost of this transformation is not trivial. Fragrance is, in the specific biochemistry of roses, not merely a quality that gives pleasure. It is a function. Rose volatile compounds play roles in plant defence against certain insects and pathogens. They are essential to pollinator attraction. They are the basis of an essential oil industry worth hundreds of millions of euros annually. The Damask rose fields of Bulgaria's Rose Valley and Turkey's Isparta region — which together produce the majority of the world's rose essential oil — represent an agricultural and cultural system that depends on the existence of fragrant varieties, and which has been increasingly squeezed by competition from the labour-intensive but high-value synthetic fragrance industry on one side and the commercial floriculture industry on the other. The Bulgarian rose harvest, conducted by hand at dawn when volatile compound concentration is highest, is one of the world's most extraordinary agricultural spectacles. It is also, increasingly, an industry under economic pressure, as the perfume industry negotiates between the authenticity of natural rose oil and the reliability and scalability of synthetic alternatives.
Ten Wild Species, Thirty Thousand Cultivars, One Bottleneck
To understand the genetic crisis at the heart of the rose breeding industry, you need to understand how modern roses were made.
The genus Rosa contains between 140 and 180 wild species, distributed across the temperate zones of the Northern Hemisphere, from the Americas to Europe, the Middle East, and across Asia to Japan and China. These wild species represent an extraordinary reservoir of genetic diversity — millions of years of evolutionary adaptation to different climates, soils, disease pressures, and pollinators, encoded in a vast collection of traits: drought tolerance, cold hardiness, disease resistance, fragrance chemistry, growth habit, hip size and nutrition, and reproductive capacity. This diversity is the raw material from which all cultivated roses have been drawn.
The problem is that very little of it was actually used. Modern cultivated roses — the Rosa × hybrida group that encompasses hybrid teas, floribundas, grandifloras, and their relatives — derive their genetic heritage from introgressions of between seven and fifteen wild species, depending on whose analysis you accept. A 2025 study in Nature Plants, examining the genomics of 215 Rosa accessions across 84 species, confirmed that "the narrow genetic foundation of cultivated roses limits their further improvement." Most recent analysis has concluded that all modern roses trace back to, at most, a handful of founding wild ancestors, primarily from China — specifically Rosa chinensis and its relatives — with contributions from European and Middle Eastern species.
This narrow genetic base is the product of three centuries of hybridisation focused on specific, commercially-valued traits: recurrent (repeat) flowering, the ability to produce a second and third flush of blooms in a season; double flowers, with their dense, multi-petalled forms; a wide range of colours; and the particular formal elegance that reached its commercial apotheosis in the hybrid tea rose of the late nineteenth century. Breeders who wanted to introduce these traits could only use the species that carried them, and those species were a small subset of the full genetic range available. Everything else — the cold hardiness of arctic and subarctic species, the disease resistance encoded in centuries of wild co-evolution with fungal pathogens, the fragrance pathways of ancient garden varieties, the extraordinary hip nutrition of Rosa canina and Rosa rugosa — was not introduced into the commercial breeding pool, either because it was technically difficult to hybridise or because no one was paying for it.
The consequence is a global commercial rose industry built on a genetic monoculture of remarkable fragility. Modern hybrid tea roses are susceptible to a catalogue of diseases — black spot (Diplocarpon rosae), powdery mildew (Podosphaera pannosa), downy mildew (Peronospora sparsa), grey mould (Botrytis cinerea), rust (Phragmidium mucronatum) — at rates that reflect their limited immune genetic diversity. Managing these diseases requires, as discussed at length in the companion features in this series, an extraordinary chemical input regime. The roses are susceptible because they were bred for appearance and longevity, not survival. The pesticides are necessary because the roses cannot defend themselves. The genetic narrowness of the breeding pool and the chemical intensity of commercial floriculture are not separate problems; they are the same problem, viewed from two different angles.
Bombarding Plants With Radiation: The Gamma Garden's Dark History
If conventional hybridisation has narrowed the rose's genetic base by selecting from a small pool of wild species, the industry's attempts to artificially expand that pool have used methods whose risks and side effects have been systematically underplayed.
Mutation breeding — the deliberate induction of genetic mutations to create novel plant traits — has been a significant tool in commercial floriculture since the 1950s. Its history begins with the atomic age. After the Manhattan Project, the United States government and its scientific establishment were searching for "peaceful uses of atomic energy." Plant mutation breeding was one answer: exposing seeds, stem cuttings, or whole plants to ionising radiation to increase the natural mutation rate by, as the International Atomic Energy Agency describes it, one thousand to one million fold. Gamma rays from radioactive cobalt-60 sources became the most popular tool, applied in "gamma gardens" at national laboratories — large circular fields with a central radiation source, surrounded by concentric rings of plants at increasing distances, each receiving a different dose.
The FAO/IAEA Mutant Variety Database records 3,365 mutant plant varieties registered globally since the programme began, of which ornamental plants — flowers, including roses — form approximately 21 per cent. Roses have been among the most important targets of mutation breeding in floriculture, because the commercial market's demand for novel colours and forms provides a continuous commercial incentive for creating new variants, and because roses are vegetatively propagated, meaning a single successful mutant can be multiplied indefinitely from cuttings.
The method is not without consequences. Gamma radiation does not selectively mutate useful genes while leaving everything else intact. It bombards the entire genome with ionising radiation, causing DNA strand breaks, chromosome rearrangements, large deletions, and multiple simultaneous mutations across the genetic sequence. The desired mutation — a new colour, a modified flower form, a changed leaf structure — is one outcome among millions of random changes. The others accumulate silently in the plant's genome, creating what geneticists call "mutation load": a burden of deleterious changes that may not manifest immediately but can affect the plant's vigour, disease resistance, reproductive capacity, and long-term adaptation. A rose variety selected for a novel colour after gamma irradiation carries that colour mutation in a genome that has been comprehensively damaged by radiation. The breeder selects for the colour. The damage comes with it.
Chemical mutagens present a parallel set of problems. Ethyl methanesulfonate (EMS), dimethyl sulphate, sodium azide, and colchicine — a compound derived from the autumn crocus that interferes with cell division and is used to create polyploid plants with doubled chromosome sets — have all been used in rose breeding. Colchicine, in particular, deserves attention: it is classified as a probable carcinogen, and occupational exposure through skin contact and inhalation is a recognised health hazard for laboratory workers and breeders who handle it regularly. The tetraploid roses created using colchicine are staples of the commercial cut flower industry. The workers who produced them did so in contact with a substance whose carcinogenicity has been known for decades.
Sodium azide, another commonly used chemical mutagen, is highly toxic: it inhibits cytochrome c oxidase in the mitochondrial respiratory chain by the same mechanism as cyanide, and is a hazard to both the workers who handle it and the water systems into which it can leach from research and breeding facilities. The biochemical literature on plant mutation breeding acknowledges, in appropriately measured scientific language, that "its environmental optimization and biological safety need to be improved." This is a statement that the mainstream discussion of rose breeding rarely makes, and that consumer-facing accounts of the industry almost never make at all.
The Scent Engineers: CRISPR and the Blue Rose's Warning
The failure of conventional breeding to restore fragrance to commercial roses, combined with the commercial incentive to create novel traits, has driven the industry toward genetic engineering — and the history of genetically modified roses is a cautionary study in how the pursuit of a single trait can produce consequences that were not anticipated and cannot be easily recalled.
The blue rose has been a goal of the rose breeding industry for decades: roses do not naturally produce the blue and violet pigment delphinidin, and no amount of conventional hybridisation between existing varieties can produce a true blue flower, because the genetic pathway for blue pigmentation simply does not exist in the Rosa genome. In 2004, researchers at Suntory (a Japanese beverage company that had diversified into floriculture) and Florigene (an Australian plant biotechnology firm) announced they had produced a rose with a lavender-blue colour by introducing the gene for flavonoid 3',5'-hydroxylase from violet (Viola) into the rose genome — the first transgenic ornamental plant to reach commercial sale. The rose, named Applause, was eventually approved for sale in Japan, the United States, Canada, and Australia.
Applause is not blue. It is, at its most optimistic characterisation, mauve — a greyish-lavender that satisfies neither the horticultural ideal of a true blue rose nor the aesthetic expectations of most consumers who buy it. This is because the introduction of a single enzyme from a different plant into a rose genome does not straightforwardly produce the desired pigment; it interacts with the existing biochemistry of the rose petal in ways that are mediated by pH, co-pigmentation, vacuolar conditions, and other factors that are not fully understood or controllable. The gene was inserted. The anticipated trait did not reliably appear. The researchers sold the product anyway, at a premium price, on the basis that it was sufficiently novel.
More recently, CRISPR-Cas genome editing has entered rose breeding as the successor to both conventional hybridisation and transgenic approaches. CRISPR allows targeted modifications to specific genes with much greater precision than radiation or chemical mutagenesis, and without necessarily introducing foreign DNA — a distinction that has significant regulatory implications in some jurisdictions, since CRISPR-edited plants may not be classified as GMOs under current law. The potential applications are substantial: restoring fragrance by editing the regulatory elements that suppress scent pathways in commercial varieties, improving disease resistance by targeting susceptibility genes, modifying colour pathways, extending vase life. Research programmes at Wageningen University, the French National Research Institute for Agriculture, Food and Environment (INRAE), and multiple commercial entities are pursuing these goals.
The concern is not that these technologies are inherently dangerous in the narrow technical sense. It is that they accelerate a pattern already well-established in the industry: the engineering of individual traits, in isolation, without adequate consideration of what those traits are connected to in the broader biology of the plant, the ecosystem, or the communities that grow it. A rose edited for extended vase life may trade off against disease resistance in ways that do not manifest until the variety is in commercial production. A rose with CRISPR-restored fragrance may emit volatile compounds at concentrations or in proportions that attract different pollinators than expected, or none. Gene flow from edited varieties to wild relatives — a risk that has been specifically studied in roses — may introduce engineered traits into wild populations, with unpredictable ecological consequences.
A 2024 Japanese study examined gene flow from cultivated Rosa × hybrida to four wild rose species planted in the same field — Rosa multiflora, Rosa luciae, Rosa rugosa, and Rosa acicularis. The study found that the overlapping flowering periods and insect visits documented between cultivated and wild roses created conditions for pollen transfer between species. Seeds were collected from wild species and examined for the presence of cultivar-specific genetic markers. Gene flow from cultivated roses to wild relatives was confirmed. This is not a theoretical risk. It is a documented event in an experimental field planting with reasonable care taken to examine it. In commercial growing regions where cultivated roses are planted in proximity to wild Rosa populations — as occurs in parts of Kenya, Colombia, and across Europe — the same gene flow occurs without monitoring and without any mechanism for detection or recall.
The Invasive Export: Rosa Multiflora and the Legacy of a Rootstock
The rose breeding industry's most consequential unintended release into a natural ecosystem did not involve genetic engineering or CRISPR. It involved a species choice made for purely commercial reasons a century and a half ago, and it has reverberated through the ecology of North America ever since.
Rosa multiflora — the multiflora rose, native to East Asia — was introduced to the United States in the 1860s as rootstock for rose breeding programmes. Its vigorous growth, its hardiness, and its compatibility with a wide range of scion varieties made it commercially convenient as the universal platform onto which commercial rose cultivars were grafted. It was planted across the country as a matter of standard horticultural practice.
By the 1930s, the United States Department of Agriculture's Soil Conservation Service was actively promoting Rosa multiflora not just for rose breeding but for erosion control and as a living fence for livestock on farms across the Midwest and Northeast. Millions of plants were distributed at government expense. By the 1960s, conservationists were observing what the ecologists had not predicted: Rosa multiflora spreads aggressively, forming dense impenetrable thickets that outcompete native vegetation, eliminate wildlife habitat, and reduce farm productivity. Each plant can produce up to a million seeds annually, and the seeds remain viable in soil for many years. Birds eat the hips and disperse the seeds far beyond any intended planting range.
Today, Rosa multiflora is classified as a noxious weed in multiple US states, including Pennsylvania, Ohio, New Jersey, and Illinois, and is listed as a severe threat by conservation organisations across the eastern United States. It has spread across millions of acres of land, from roadsides to forests to protected natural areas. Its eradication, where attempted, requires sustained herbicide application or mechanical removal over multiple years, and is rarely fully successful. The Pennsylvania Department of Agriculture's list of invasive species ranks it among the most damaging. Massachusetts has banned its importation, distribution, and sale. Its spread cannot be reversed.
The rose breeding industry introduced Rosa multiflora because it was useful. It did not consider whether it was wise. The costs of that failure — the degraded wildlife habitat, the lost agricultural land, the decades of failed eradication effort — have been paid by the land, by the animals that depend on it, and by the farmers who inherited the problem. They were not paid by the breeders who made the choice.
The Wild Ancestors Are Disappearing
While the commercial rose industry has created an unprecedented number of cultivated varieties, the wild species from which they drew their genetic heritage are in decline — a decline that the industry has done relatively little to prevent, and from which it now urgently needs to be saved.
Of the 140 to 180 wild Rosa species, a growing proportion face habitat loss severe enough to threaten their long-term viability. Wild roses in Central Asia — the region that botanical evidence now suggests was the evolutionary origin point for many important ancestral forms — have been affected by land degradation, overgrazing, agricultural expansion, and climate change. Wild Chinese Rosa species, confirmed by the 2025 Nature Plants study as the primary diversity centres for the genus, are under pressure from urbanisation, afforestation with non-native species, and the loss of the meadow and scrub habitats they depend on. European wild roses, including Rosa canina and its relatives — which contributed cold hardiness and disease resistance genes to early hybrid development — are increasingly marginalised in intensively managed agricultural landscapes.
The irony is structural and severe: the commercial rose industry is built on a narrow genetic base extracted from wild species, has done little to conserve those wild species, and is now reaching a point where the further improvement of commercial varieties requires genetic diversity that can only be obtained from wild species that may no longer be there. Researchers at Wageningen University, in their 2019 roadmap for rose research, described this explicitly: "Fragrance is gaining new interest because of consumer demand and the improved understanding of its underlying genetics." Translation: the industry bred the scent out of the rose, and now wants it back. The wild species that might supply the necessary genes are among those under threat.
A research team at Kunming in China, which has spent three decades collecting and preserving more than 200 wild Rosa accessions, explicitly frames this work as conservation against an extinction threat: "Modern roses, admired for their vibrant colors and rich fragrances, originate from fewer than ten wild ancestors, leading to a narrow genetic base that limits breeding innovation." The founders of the Kunming collection understood that the genetic reservoir of the genus was at risk of being lost before it had been fully utilised. Their effort — one individual's personal commitment, not a systematic industry programme — has created China's largest Rosa germplasm garden, providing what the researchers call "a crucial foundation for research, breeding, and the introduction of new cultivars."
One individual's collection should not be the primary insurance policy for the genetic future of the world's most economically important ornamental plant. It is.
The Patent Regime: Who Owns the Rose?
The global rose breeding industry does not merely create new varieties. It owns them — through a system of Plant Breeders' Rights and utility patents that has, over the past four decades, transformed the commercial rose from a plant that growers could propagate freely into an intellectual property asset that generates royalties for European breeding companies every time a Kenyan farmer takes a cutting.
Plant Breeders' Rights — the intellectual property system that governs commercial plant varieties in most countries, based on the International Union for the Protection of New Varieties of Plants (UPOV) convention — grant the breeder of a new variety the exclusive right to sell and license its propagation for twenty years. For commercial cut flower roses, this means that a grower in Ecuador or Kenya who wishes to grow a patented variety must pay a per-stem royalty to the breeding company that developed it. The major breeding houses — Germany's W. Kordes' Söhne, France's Meilland International, the Netherlands' Dümmen Orange (itself a consolidation of multiple smaller breeders), Britain's David Austin Roses, and several others — collectively hold the intellectual property rights to most of the commercially grown rose varieties in circulation.
The royalty system is not, in principle, unreasonable. Plant breeding is expensive: Kordes breeds approximately 1.2 million new seeds annually, subjects several thousand to extended field trials over seven years, and introduces only five to seven new varieties per category per year that meet its standards. Recouping that investment requires a mechanism for collecting returns from the commercial use of successful varieties. The system exists to incentivise innovation, and without it, the private sector investment in rose breeding would decline.
In practice, however, the system creates a significant wealth transfer from growers — who are predominantly in the developing world, growing under economic pressures that are entirely invisible to the European breeding companies whose intellectual property they are using — to breeders in wealthy countries. A Kenyan flower farm growing a Kordes variety for the export market pays per-stem royalties to a German company for every stem it produces. Those royalties are calculated and enforced through a system that was designed in Europe, for European conditions, and that requires legal compliance infrastructure that many small growers in developing countries do not have the capacity to implement or negotiate around.
The consolidation of the breeding industry has intensified the concentration of this royalty flow. The 2015 creation of Dümmen Orange, through the merger of Terra Nigra and several other Dutch breeding companies, created one of the largest single players in the global rose variety market. An industry publication's report on the merger noted with telling frankness that "some growers will maybe be afraid that if more breeders will cluster the royalty prices will go up in the long run" — an observation that accurately reflects the market dynamics of a sector where a small number of breeders hold the IP rights to the varieties that most commercial growers must use.
David Austin Roses, the English company whose "English rose" varieties — blowsy, multi-petalled, heavily fragrant blooms in the style of old garden roses — have become enormously commercially significant, has been aggressive in pursuing IP violations globally. In 2025, the company's IP dispute over unauthorised propagation of its patented varieties moved forward in US federal courts, with the company's Brand Protection Manager noting that the courts' refusal to dismiss the case was "a very positive one not only for us but also for rights holders in the U.S." The company uses a combination of Plant Patents, Plant Variety Rights, and Trademarks — three overlapping IP instruments — to protect its varieties and the revenue they generate.
The asymmetry of this system is most visible at its extremes. A large commercial farm in Colombia growing David Austin varieties for the wedding flower market — the fastest-growing and highest-margin segment of the premium cut flower sector — has the legal and commercial infrastructure to enter into licensing agreements and pay royalties. A small-scale grower in Ethiopia attempting to propagate commercially successful varieties without the capital to enter formal licensing arrangements faces legal liability under intellectual property regimes they may not fully understand, enacted in countries they have never visited, by corporations whose revenues exceed the GDP of some of the districts where they farm.
What Has Been Lost and What Is Now Irretrievable
The catalogue of what the commercial rose breeding industry has lost or damaged in its pursuit of the perfect flower is not easily reduced to a single number or a single image. It is distributed across many systems simultaneously — genetic, ecological, cultural, economic, olfactory — and much of it is invisible precisely because it was lost before anyone thought to measure it.
The fragrance of the commercial rose is not fully gone — it persists in old garden varieties, in specialist fragrance rose collections, in the Bulgarian rose fields, in the gardens of rosarians who choose their plants by scent rather than stem length. But it has been systematically and deliberately bred out of the dominant commercial form of the flower, and restoring it will require either the reintroduction of genetic material from old varieties whose complex polyploid genomes make them technically difficult to use in breeding, or the deployment of CRISPR and other editing technologies that raise their own questions. The industry created the problem over centuries. The solution will not come quickly.
The wild rose species are not gone — the genus is large, and many species remain abundant in their native ranges. But the habitat loss and climate pressure that threaten the most genetically distinct and geographically marginal wild populations are ongoing, and the collections maintained by botanical gardens and individual enthusiasts are not a complete substitute for wild populations in functioning ecosystems. The genetic material held in the Kunming garden and in the Rosa gene banks at Wageningen and other institutions is an irreplaceable resource. Its preservation is inadequately funded and inadequately acknowledged.
The ecological damage from Rosa multiflora in North America cannot be undone. The species is established across millions of acres and will not be eradicated in any timeframe relevant to human planning. Its displacement of native plant communities, its reduction of agricultural and ecological value in the landscapes it has colonised, and the cost of the control efforts it has generated represent a permanent debt owed by the industry that introduced it.
The royalty flows from Kenyan and Ecuadorian and Ethiopian growers to European breeding companies represent a continuing transfer of wealth that is built into the structure of the global rose market and will not be changed by any reform that the breeding industry undertakes voluntarily. The UPOV convention is an international agreement. Its terms are not determined by the preferences of Kenyan growers.
The Scientists Who Are Trying to Put the Rose Back Together
Not all of the news is dark, and not all of the people working within or alongside the commercial rose breeding industry are content with the trajectory described above.
A growing community of rose researchers — centred at Wageningen University in the Netherlands, INRAE in France, the Chinese Academy of Agricultural Sciences, and several other institutions — has spent the past decade using the tools of modern genomics to reconstruct what was lost. The publication of multiple high-quality rose genome sequences, beginning with Rosa chinensis in 2018 and progressing through Rosa multiflora, Rosa rugosa, and the ancestral tea rose Rosa gigantea, has created the foundational resource for understanding, at molecular resolution, what the commercial breeding programme did to the rose's genetic architecture and what might be done to reverse it.
A 2024 Nature Communications paper on the genome of Rosa gigantea — the wild Chinese species identified as the primary ancestor of tea fragrance — mapped the specific biochemical pathways responsible for the tea scent that characterises some of the most complex rose fragrances, identified the genes involved in eugenol biosynthesis that attract specific pollinators, and explicitly framed the work as providing "a scientific foundation for enhancing rose fragrance via de novo domestication." Translated from scientific language: we now know, at molecular level, what the rose's scent is made of, how it is made, and how it might be restored.
The German breeding company Kordes has been operating, for several decades, a breeding programme specifically focused on disease resistance — selecting varieties that can survive without fungicide applications, and testing them rigorously before release. Its ADR (Allgemeine Deutsche Rosenneuheitenprüfung) certification system evaluates new varieties across multiple trial sites in northern Europe without any disease sprays, rejecting any cultivar that requires chemical support to survive. This is not altruism — disease-resistant varieties require less pesticide input from growers, which reduces their production costs and, increasingly, addresses regulatory pressure and consumer demand for lower-chemical floriculture. But the outcome is roses that are meaningfully less dependent on the chemical regimes described elsewhere in this series.
There is also, in the independent and heritage rose world, a counter-movement that has operated in parallel to commercial breeding for generations: the cultivation and preservation of old garden roses, species roses, and heritage varieties by enthusiasts, botanists, and small nurseries who have maintained the genetic and olfactory diversity that the commercial industry discarded. The collections at the Sangerhausen Rose Garden in Germany — the world's largest rose collection, holding approximately 9,000 varieties including many not found anywhere else — represent an irreplaceable living archive. Heritage rose societies in Britain, France, the United States, Australia, and elsewhere have documented and propagated varieties that commercial breeders would have allowed to disappear. The work is amateur in the best and oldest sense of the word: driven by love of the subject, conducted without commercial incentive, and producing a legacy of preserved genetic material that the scientific breeding community is now, belatedly, beginning to recognise as invaluable.
The Rose as It Was and the Rose as It Could Be
There is a particular kind of clarity that comes from standing in a garden of old roses in early summer — the Rosa gallica, the Rosa alba, the ancient damasks — and understanding what the commercial flower industry decided was not worth keeping. The scent alone is a revelation: dense, complex, shifting as you move from flower to flower, and entirely unlike the faint floral note of a hybrid tea purchased at a supermarket. These roses were not engineered. They were not irradiated or edited or bred for vase life. They evolved in the fields and gardens of centuries, shaped by the preferences of people who grew them because they gave pleasure in every dimension the plant is capable of, and not just the visual one.
The commercial industry's version of the rose — long-stemmed, odourless, cosmetically perfect, disease-susceptible, chemically dependent, genetically narrow, royalty-encumbered — is the product of a specific set of choices made over a specific period of time, for specific commercial reasons. Those choices are not permanent. They are not the definition of the rose. They are, in the most precise sense, what the industry decided to do with the rose in the service of a market that wanted something beautiful, uniform, and cheap.
The cost of those choices has been paid in lost fragrance and lost wild species and lost genetic diversity and invasive ecology and chemical contamination and the royalty payments of smallholder growers. It continues to be paid. Understanding that cost is the precondition for any serious discussion about whether a different set of choices is possible — about what a rose industry that treated the plant, its wild relatives, and the people who grow it with something approaching genuine respect would look like.
It would smell different, to begin with.
