Start a consultation

Support Tree TSAR: Sponsorship Opportunities

Rosaceae Juss.

Rose Family

Suggest a Correction

Family Overview

Rosaceae Juss., the Rose Family, is one of the defining families of the order Rosales and one of the most economically important radiations of flowering plants. Tree TSAR currently recognizes 93 accepted natural or stabilized hybridogenous genera and about 3,000 species, with additional intergeneric nothogenera being audited separately. The family ranges from perennial herbs such as strawberries and cinquefoils to shrubs, scrambling brambles, and large trees, and it includes roses, apples, pears, quinces, hawthorns, cherries, peaches, plums, apricots, almonds, rowans, whitebeams, and many less familiar lineages. Major sources differ in generic and species counts because several groups are rich in polyploids, apomicts, recurrent hybrids, and narrowly delimited microspecies. Plants of the World Online currently lists 99 accepted generic entities, including several nothogenera and segregate genera that Tree TSAR treats within broader genera (Phipps 2015; Xiang et al. 2017; Royal Botanic Gardens, Kew 2026).

Despite its diversity, the Rose Family has a recognizable structural theme. Leaves are usually alternate and stipulate, although both characters vary. Flowers are commonly radially symmetric, often five-parted, and frequently many-stamened. The floral organs arise from a hypanthium, a cup- or saucer-shaped structure whose form and degree of fusion with the carpels differ dramatically among lineages. One to hundreds of carpels may be present, and ovary position ranges from superior to inferior. These modifications help generate the family’s extraordinary fruit diversity: follicles in Spiraea and relatives, drupes in Prunus, pomes in apples, pears, and many Maleae, aggregate drupelets in Rubus, aggregate achenes on the enlarged receptacle of Fragaria, and rose hips in which achenes are enclosed by a fleshy hypanthium (Phipps 2015).

This variety is not the result of a simple progression from one fruit type to another. Nuclear phylogenomics shows that fleshy fruits arose repeatedly from dry-fruited ancestors and that several of the family’s major fruit architectures represent independent evolutionary experiments. Rosaceae therefore offer an unusually powerful system for studying how floral construction, genome duplication, dispersal ecology, and domestication interact through deep time (Xiang et al. 2017).

Classification and evolutionary history

Rosaceae have been recognized as a coherent family since Antoine Laurent de Jussieu published the name in 1789; the name is conserved and typified by Rosa. Historical systems nevertheless split portions of the family into segregate families or subfamilies centered on stone fruits, pome fruits, roses, or spiraeoid shrubs. Molecular systematics demonstrated that these familiar groups are parts of one natural family and led to the modern three-subfamily framework: Amygdaloideae, Dryadoideae, and Rosoideae. The large broad subfamily now called Amygdaloideae absorbs the traditional Maloideae and Spiraeoideae as well as the stone-fruit lineage; under the nomenclatural rules Amygdaloideae has priority for this expanded group (Potter et al. 2007; Phipps 2015; International Plant Names Index 2026).

Tree TSAR organizes Rosaceae below the subfamily level using tribes and, where they provide stable and informative structure, subtribes. This hierarchy is intentionally uneven rather than forcing every branch of the family into the same number of ranks. Amygdaloideae and Rosoideae contain several well-supported radiations for which tribal names provide useful evolutionary and navigational structure, while selected large or internally differentiated tribes are subdivided further. Maleae, for example, are represented by Malinae, Lindleyinae, and Vauqueliniinae; Agrimonieae by Agrimoniinae and Sanguisorbinae; and Potentilleae by Fragariinae and Potentillinae. Dryadoideae are sufficiently small and coherent that Tree TSAR does not impose an additional tribal subdivision. Gillenieae are retained separately from Maleae because current evidence leaves the precise boundary between those lineages open to alternative treatments. The resulting hierarchy therefore emphasizes strongly supported, biologically useful groups without implying that every named infrafamilial rank is equally immutable (Potter et al. 2007; Xiang et al. 2017; Wang et al. 2024).

The family itself is strongly supported as monophyletic. A Rosales-wide analysis based on nuclear and plastid loci recovered Rosaceae as sister to the remaining families of Rosales, while newer phylogenomic work shows that the deepest history of the order was rapid enough to leave extensive incomplete lineage sorting and signals of ancient hybridization. Within Rosaceae, however, the three principal subfamilies remain robust evolutionary units even where the exact order of their earliest divergences varies among datasets (Zhang et al. 2011; Xiang et al. 2017; Liu et al. 2026).

A large nuclear phylogenomic analysis estimated the crown of Rosaceae at approximately 102 million years ago, placing the family’s origin deep in the Cretaceous. By the Eocene, the fossil record shows a remarkably diverse rosaceous flora. The Okanogan Highlands of western North America preserve more than a dozen rosaceous taxa represented by leaves, flowers, fruits, wood, and pollen, including both extinct forms and lineages comparable with modern genera. The complexity of these Eocene assemblages indicates that much of the family’s structural and reproductive diversity was already established relatively early in its history (DeVore & Pigg 2007; Xiang et al. 2017).

Reticulation, apomixis, and genus boundaries

Hybridization and polyploidy are not peripheral features of Rosaceae; in several major radiations they are central to how diversity is generated and maintained. Apomixis, in which seeds can form without ordinary sexual fertilization of the embryo, is especially important in lineages such as Alchemilla, Amelanchier, Crataegus, Rubus, and the sorboid members of Maleae. Recurrent hybrid origin, polyploid stabilization, and partial sexuality can blur both species and genus boundaries, which is why different modern checklists may disagree even when they are working from much of the same biological evidence (Phipps 2015).

Tree TSAR treats those conflicts case by case rather than applying a blanket preference for splitting or lumping. In the former Sorbus complex, Tree TSAR adopts a broad Aria that includes the lineages historically segregated as Chamaemespilus and Torminalis. The phylogenetic evidence around the Torminalis boundary remains mixed enough that both broad and narrow treatments are defensible; Tree TSAR uses stability and predictive robustness as the tie-breaker. Conservation of the name Aria against Chamaemespilus and Torminalis makes the broad treatment nomenclaturally practical, and recent nomenclatural work has already transferred many affected species into Aria and Hedlundia (Mosyakin et al. 2022, 2025).

Under that treatment, Hedlundia is retained for stabilized evolutionary lineages combining Aria and Sorbus ancestry, whereas × Arsorbus is used for spontaneous or artificial intergeneric hybrids that have not formed independently stabilized lineages. The same integrative logic can support a different result elsewhere: Tree TSAR retains Mespilus apart from Crataegus because dense nuclear data allow it to be interpreted as a coherent sister lineage rather than an embedded segregate, and its morphology and historical identity provide additional explanatory value (Liston et al. 2021). The broader family treatment therefore aims for evolutionary coherence and stability rather than uniformity for its own sake.

Horticultural and Agricultural Uses

Few plant families combine horticultural familiarity with agricultural importance as completely as Rosaceae. Temperate orchards are dominated by rosaceous crops: apples and pears are pomes; peaches, nectarines, cherries, plums, apricots, and almonds belong to Prunus and share the drupe fruit plan; strawberries are enlarged receptacles bearing numerous achenes; and raspberries and blackberries are aggregate fruits composed of many drupelets. Quince and loquat add further pome-fruit crops, while regionally important species contribute fresh fruit, preserves, beverages, flavorings, and nuts. This concentration of economically important fruit types has made Rosaceae a major comparative system for studying domestication, fruit development, and crop evolution (Xiang et al. 2017; Soundararajan et al. 2019).

Many of the principal crops arose from temperate or seasonally cool environments, and their cultivation still reflects that history. Winter chilling, spring frost timing, water availability, rootstock adaptation, and disease pressure strongly influence orchard performance, while long generation times make breeding slower than in annual crops. Genomic resources have therefore become especially valuable. Reference genomes, pangenomic datasets, trait markers, and comparative maps are increasingly used to accelerate selection for fruit quality, flowering behavior, disease resistance, and tolerance of heat, drought, and other stresses (Soundararajan et al. 2019; Migicovsky 2025).

The ornamental side of the family is equally prominent. Rosa underpins one of the world’s largest ornamental breeding traditions and also supplies fragrance crops used for rose oil and rose water. Flowering cherries, crabapples, hawthorns, spiraeas, ninebarks, cotoneasters, pyracanthas, photinias, rowans, and whitebeams are widely planted for flowers, fruit display, foliage, wildlife value, and architectural form. Herbaceous rosaceous ornamentals such as Geum, Filipendula, and Potentilla broaden the family’s garden presence beyond woody plants. The same propensity for hybridization that complicates classification has been exploited deliberately in horticulture to create cultivars with novel combinations of flower color, fruit quality, disease resistance, and growth habit.

Wild relatives are increasingly important to the future of these crops. Commercial perennial fruit systems often rely on a relatively narrow range of clonally propagated cultivars, whereas wild populations continue to harbor alleles shaped by local climates, pathogens, soils, and ecological interactions. Conserving this diversity is therefore not separate from crop improvement: it is part of maintaining the genetic options available for future breeding (Migicovsky 2025).

Conservation Issues

Rosaceae are cosmopolitan, but conservation risk is not distributed evenly across the family. Many widespread roses, brambles, hawthorns, and cinquefoils remain common, while narrow island endemics, montane forest trees, high-elevation specialists, and crop wild relatives can be acutely vulnerable to habitat loss, altered fire regimes, grazing, overharvest, invasive pests, and climate change. For a family with major food crops, conservation also has an unusually direct genetic dimension: the loss of wild populations can remove adaptive variation that may later be needed for agriculture.

Wild apples illustrate that connection particularly clearly. Malus sieversii, the principal wild progenitor of the domesticated apple, survives in fragmented Central Asian fruit forests and contains valuable resistance and stress-tolerance diversity. Habitat degradation, agricultural expansion, climate change, pests, and crop-to-wild gene flow all threaten the ecological and genetic integrity of these populations. Recent syntheses emphasize that conserving wild apple forests in situ, while maintaining complementary ex situ germplasm, is essential both for biodiversity and for broadening the genetic base available to apple breeding (Tegtmeier et al. 2025; Migicovsky 2025).

Other lineages face very different pressures. High-Andean Polylepis woodlands form some of the world’s highest-elevation forests and have been fragmented by long histories of land use, grazing, burning, and wood extraction; climate change adds uncertainty to regeneration and range persistence. Effective conservation therefore depends not only on protected areas but also on restoration, community participation, and management tailored to local ecological conditions (Pinos 2020). In African montane forests, Prunus africana has been heavily harvested for medicinal bark and is listed in CITES Appendix II, making sustainable harvest, trade monitoring, and protection of mature seed-producing trees central conservation issues (CITES 2023).

Because many threatened rosaceous taxa are long-lived woody plants or clonally maintained crop relatives, conservation works best as a portfolio rather than a single technique. Protected wild populations preserve evolutionary processes and local adaptation; seed banks are valuable where storage biology permits; living collections, field genebanks, and clonal or cryogenic repositories secure material that cannot be represented adequately by conventional seed storage. Genomic sampling can help identify distinct conservation units, detect introgression, and ensure that ex situ collections capture rather than inadvertently narrow genetic diversity (Migicovsky 2025).

Major Clades

Modern Rosaceae are organized into three strongly supported subfamilies. Tree TSAR further organizes Amygdaloideae and Rosoideae into tribes and, where stable and informative, selected subtribes; the small, coherent Dryadoideae are left without an additional tribal subdivision. The Accepted Genera table carries that finer hierarchy and makes the deliberate asymmetry explicit. At family-page scale, however, the broad biological contrast among the three subfamilies is more informative than an exhaustive tribal catalogue (Potter et al. 2007; Xiang et al. 2017; Wang et al. 2024).

Amygdaloideae contains most of the family’s familiar woody fruit trees and a large radiation of shrubs, together with some herbs. It includes Prunus and its stone fruits; Maleae, with apples, pears, hawthorns, rowans, whitebeams, and their relatives; and spiraeoid lineages such as Spiraea, Physocarpus, and Neillia. Fruit types range from follicles and capsules to drupes and pomes. The broad Amygdaloideae concept replaces the older practice of separating groups such as Maloideae, Spiraeoideae, and Prunoideae as equivalent subfamilies (Phipps 2015).

Rosoideae include roses, brambles, strawberries, cinquefoils, avens, burnets, lady’s-mantles, and numerous related herbs and shrubs. One-seeded carpels are common, and the family’s diversity of aggregate fruits is especially conspicuous here. Polyploidy, apomixis, and reticulate evolution are recurrent themes in several large rosoid genera, contributing to both ecological success and persistent taxonomic difficulty (Phipps 2015).

Dryadoideae are the smallest of the three subfamilies and comprise Cercocarpus, Chamaebatia, Dryas, and Purshia in the Tree TSAR treatment. All four belong to the actinorhizal nitrogen-fixing radiation and can form root nodules with Frankia, an unusual trait within Rosaceae that links their ecology to the broader nitrogen-fixing fabid lineage. The group ranges from arctic-alpine Dryas to drought-adapted western North American shrubs and trees. Its small size belies its importance for understanding the early diversification and ecological breadth of the family.

Accepted Genera

The Accepted Genera table reflects Tree TSAR's current natural and stabilized hybridogenous genus backbone. It is not yet an exhaustive inventory of artificial or spontaneous intergeneric hybrids; additional nothogenera are being vetted for validity, parentage, and nomenclatural application.
Acaena Mutis ex L. (1771)
Adenostoma Hook. & Arn. (1832)
Agrimonia L. (1753)
Alchemilla L. (1753)
Amelanchier Medik. (1789)
Aremonia Neck. ex Nestl. (1816)
Argentina Hill (1756)
Aria (Pers.) J.Jacq. ex Host (1831)
Aronia Medik. (1789)
Aruncus L. (1758)
Cercocarpus Kunth (1824)
Chaenomeles Lindl. (1822)
Chamaebatia Benth. (1849)
Chamaebatiaria (Porter ex W.H.Brewer & S.Watson) Maxim. (1879)
Chamaecallis Smedmark (2014)
Chamaemeles Lindl. (1821)
Chamaerhodos Bunge (1829)
Cliffortia L. (1753)
Coleogyne Torr. (1851)
Comarum L. (1753)
Cormus Spach (1834)
Cotoneaster Medik. (1789)
Crataegus L. (1753)
Cydonia Daubenton & P.Daubenton (1753)
Dasiphora Raf. (1840)
Dichotomanthes Kurz (1873)
Dryas L. (1753)
Drymocallis Fourr. ex Rydb. (1908)
Eriobotrya Lindl. (1821)
Exochorda Lindl. (1858)
Fallugia Endl. (1840)
Farinopsis Chrtek & Soják (1984)
Filipendula Mill. (1754)
Fragaria L. (1753)
Geum L. (1753)
Gillenia Moench (1802)
Hagenia J.F.Gmel. (1791)
Hedlundia Sennikov & Kurtto (2017)
Hesperomeles Lindl. (1837)
Heteromeles M.Roem. (1847)
Holodiscus (K.Koch) Maxim. (1879)
Kageneckia Ruiz & Pav. (1794)
Kelseya (S.Watson) Rydb. (1900)
Kerria DC. (1818)
Leucosidea Eckl. & Zeyh. (1836)
Lindleya Kunth (1824)
Luetkea Bong. (1832)
Lyonothamnus A.Gray (1885)
Malacomeles (Decne.) Decne. (1882)
Malus Mill. (1754)
Margyricarpus Ruiz & Pav. (1794)
Mespilus L. (1753)
Micromeles Decne. (1874)
Neillia D.Don (1825)
Neviusia A.Gray (1858)
Oemleria Rchb. (1841)
Osteomeles Lindl. (1821)
Pentactina Nakai (1917)
Peraphyllum Nutt. (1840)
Petrophytum (Nutt.) Rydb. (1900)
Phippsiomeles B.B.Liu & J.Wen (2019)
Photinia Lindl. (1820)
Physocarpus (Cambess.) Raf. (1838)
Polylepis Ruiz & Pav. (1794)
Potentilla L. (1753)
Poteridium Spach (1846)
Poterium L. (1753)
Pourthiaea Decne. (1874)
Prinsepia Royle (1835)
Prunus L. (1753)
Pseudocydonia (C.K.Schneid.) C.K.Schneid. (1906)
Purshia DC. ex Poir. (1816)
Pyracantha M.Roem. (1847)
Pyrus L. (1753)
Rhaphiolepis Lindl. (1820)
Rhodotypos Siebold & Zucc. (1841)
Rosa L. (1753)
Rubus L. (1753)
Sanguisorba L. (1753)
Sibbaldia L. (1753)
Sibbaldianthe Juz. (1941)
Sibiraea Maxim. (1879)
Sieversia Willd. (1811)
Sorbaria (Ser. ex DC.) A.Braun (1860)
Sorbus L. (1753)
Spenceria Trimen (1879)
Spiraea L. (1753)
Spiraeanthus (Fisch. & C.A.Mey.) Maxim. (1879)
Stranvaesia Lindl. (1837)
Tetraglochin Poepp. (1833)
Vauquelinia Corrêa ex Bonpl. (1807)
Weniomeles B.B.Liu (2023)
Xerospiraea Henrickson (1986)

Additional Information

  • iNaturalist: Family taxon page with observations, photographs, maps, and community identifications. Link

  • Plants of the World Online: Kew family backbone with nomenclature, distributional data, and accepted genera. POWO currently recognizes 99 generic entities, including several nothogenera and segregates treated differently by Tree TSAR. Link

  • Flora of North America: Modern regional family treatment with morphology, subfamilies, tribes, horticultural context, and keys. Link

  • Flora of China: Extensive regional treatment of Chinese Rosaceae. It is valuable for morphology and species identification but explicitly notes that portions of its generic and subfamilial classification are traditional and require revision. Link

  • World Flora Online: Global taxonomic and nomenclatural portal for Rosaceae and its subordinate taxa. Link

  • International Plant Names Index: Nomenclatural record for Rosaceae Juss.; records publication in 1789, conserved status, and Rosa as the type. Link

  • Genome Database for Rosaceae: Specialized genomics, genetics, trait, marker, and breeding resource for major rosaceous crops and research species. Link

References and Further Reading

CITES (2023) CITES and timber: A guide to CITES-listed tree species — Prunus africana. Convention on International Trade in Endangered Species of Wild Fauna and Flora. https://cites.org/sites/default/files/timber_id_materials/files/CITES%20%20Timber%20-%20A%20guide%20to%20CITES-listed%20tree%20species%202023.pdf

DeVore ML, Pigg KB (2007) A brief review of the fossil history of the family Rosaceae with a focus on the Eocene Okanogan Highlands of eastern Washington State, USA, and British Columbia, Canada. Plant Systematics and Evolution 266: 45–57. https://doi.org/10.1007/s00606-007-0540-3

International Plant Names Index (2026) Rosaceae Juss., Genera Plantarum 334 (1789), nom. cons. https://www.ipni.org/n/30000200-2

Liston A, Weitemier KA, Letelier L, Podani J, Zong Y, Liu L, Dickinson TA (2021) Phylogeny of Crataegus (Rosaceae) based on 257 nuclear loci and chloroplast genomes: evaluating the impact of hybridization. PeerJ 9: e12418. https://doi.org/10.7717/peerj.12418

Liu L, Lv W, Soltis DE, Comes HP, Zhang Y, Shi Y, Li S, Wang M, Gao W, Li P (2026) Complex evolutionary history of Rosales mediated by extensive incomplete lineage sorting and hybridization. Molecular Phylogenetics and Evolution 225: 108716. https://doi.org/10.1016/j.ympev.2026.108716

Migicovsky Z (2025) Genomic resources for crop wild relatives are critical for perennial fruit breeding and conservation. American Journal of Botany 112(7): e70068. https://doi.org/10.1002/ajb2.70068

Mosyakin SL, Fedoronchuk MM, McNeill J (2022) (2886) Proposal to conserve the name Aria against Chamaemespilus and Torminalis (Rosaceae). Taxon 71(2): 480–481. https://doi.org/10.1002/tax.12705

Mosyakin SL, Fedoronchuk MM, McNeill J (2025) Simplifying the nomenclature of Sorbus sensu lato: new nomenclatural solutions in Aria and Hedlundia (Rosaceae). Ukrainian Botanical Journal 82(3): 206–224. https://doi.org/10.15407/ukrbotj82.03.206

Phipps JB (2015) Rosaceae. In: Flora of North America Editorial Committee (eds), Flora of North America North of Mexico, Vol. 9. Oxford University Press, New York and Oxford. https://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=10776

Pinos J (2020) Challenges and conservation implications of Polylepis woodlands in the Andean region: defining actions for sustainable management. Hacquetia 19(2): 143–153. https://doi.org/10.2478/hacq-2020-0001

Potter D, Eriksson T, Evans RC, Oh S, Smedmark JEE, Morgan DR, Kerr M, Robertson KR, Arsenault M, Dickinson TA, Campbell CS (2007) Phylogeny and classification of Rosaceae. Plant Systematics and Evolution 266: 5–43. https://doi.org/10.1007/s00606-007-0539-9

Royal Botanic Gardens, Kew (2026) Rosaceae Juss. Plants of the World Online. https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:30000200-2

Soundararajan P, Won SY, Kim JS (2019) Insight on Rosaceae family with genome sequencing and functional genomics perspective. BioMed Research International 2019: 7519687. https://doi.org/10.1155/2019/7519687

Tegtmeier R, Švara A, Gritsenko D, Khan A (2025) Malus sieversii: a historical, genetic, and conservational perspective of the primary progenitor species of domesticated apples. Horticulture Research 12(1): uhae244. https://doi.org/10.1093/hr/uhae244

Wang H, Li X-Y, Jiang Y, Jin Z-T, Ma D-K, Liu B, Xu C, Ge B-J, Wang T, Fan Q, Jin S-H, Liu G-N, Liu B-B (2024) Refining the phylogeny and taxonomy of the apple tribe Maleae (Rosaceae): insights from phylogenomic analyses of 563 plastomes and a taxonomic synopsis of Photinia and its allies in the Old World. PhytoKeys 242: 161–227. https://doi.org/10.3897/phytokeys.242.117481

Xiang Y, Huang C-H, Hu Y, Wen J, Li S, Yi T, Chen H, Xiang J, Ma H (2017) Evolution of Rosaceae fruit types based on nuclear phylogeny in the context of geological times and genome duplication. Molecular Biology and Evolution 34(2): 262–281. https://doi.org/10.1093/molbev/msw242

Zhang S-D, Soltis DE, Yang Y, Li D-Z, Yi T-S (2011) Multi-gene analysis provides a well-supported phylogeny of Rosales. Molecular Phylogenetics and Evolution 60(1): 21–28. https://doi.org/10.1016/j.ympev.2011.04.008