Invasive Species Compendium

Detailed coverage of invasive species threatening livelihoods and the environment worldwide


Cotoneaster horizontalis



Cotoneaster horizontalis (wall-spray)


  • Last modified
  • 06 November 2018
  • Datasheet Type(s)
  • Invasive Species
  • Host Plant
  • Preferred Scientific Name
  • Cotoneaster horizontalis
  • Preferred Common Name
  • wall-spray
  • Taxonomic Tree
  • Domain: Eukaryota
  •   Kingdom: Plantae
  •     Phylum: Spermatophyta
  •       Subphylum: Angiospermae
  •         Class: Dicotyledonae
  • Summary of Invasiveness
  • Cotoneaster horizontalis is a woody, perennial, deciduous or semi-evergreen, procumbent shrub with horizontally spreading branches. It is native to parts of China, and has been introduced widely to Europe, Nort...

Don't need the entire report?

Generate a print friendly version containing only the sections you need.

Generate report


Top of page
Cotoneaster horizontalis (rock cotoneaster); typical fruiting shoot.
TitleBranch with fruits
CaptionCotoneaster horizontalis (rock cotoneaster); typical fruiting shoot.
Copyright©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); typical fruiting shoot.
Branch with fruitsCotoneaster horizontalis (rock cotoneaster); typical fruiting shoot.©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); close-up of fruiting shoot.
TitleBranch with fruits
CaptionCotoneaster horizontalis (rock cotoneaster); close-up of fruiting shoot.
Copyright©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); close-up of fruiting shoot.
Branch with fruitsCotoneaster horizontalis (rock cotoneaster); close-up of fruiting shoot.©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); close-up of branch, showing leaves.
CaptionCotoneaster horizontalis (rock cotoneaster); close-up of branch, showing leaves.
Copyright©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); close-up of branch, showing leaves.
LeavesCotoneaster horizontalis (rock cotoneaster); close-up of branch, showing leaves.©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); mass of vegetative growth.
TitleSmothering habit
CaptionCotoneaster horizontalis (rock cotoneaster); mass of vegetative growth.
Copyright©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.
Cotoneaster horizontalis (rock cotoneaster); mass of vegetative growth.
Smothering habitCotoneaster horizontalis (rock cotoneaster); mass of vegetative growth.©Robert G. & Fiona J. Richardson - 2005. All Rights Reserved.


Top of page

Preferred Scientific Name

  • Cotoneaster horizontalis Decne.

Preferred Common Name

  • wall-spray

Other Scientific Names

  • Cotoneaster davidianus hort. ex Dippel
  • Cotoneaster microphyllus Diels
  • Cotoneaster symonsii (hort. ex Baker) Loudon ex Koehne

International Common Names

  • English: herringbone cotoneaster; rock cotoneaster; rockspray; rock-spray; wall cotoneaster
  • French: cotonéaster horizontal; cotonnier horizontal
  • Chinese: ping zhi xun zi; xiao ye ping zhi xun zi

Local Common Names

  • Germany: Fächer-Steinmispel; Fächer-Zwergmispel; Zwergmispel waagerecht wachsende
  • Italy: cotognastro orizzontale
  • Netherlands: vlakke dwergmispel

EPPO code

  • CTTHO (Cotoneaster horizontalis)

Summary of Invasiveness

Top of page

Cotoneaster horizontalis is a woody, perennial, deciduous or semi-evergreen, procumbent shrub with horizontally spreading branches. It is native to parts of China, and has been introduced widely to Europe, North America, Australasia and other parts of the world as an attractive garden plant with bright red berries which often persist into winter. Introduction has been deliberate for this purpose but because the species fruits freely, has been widely planted and its seeds are so easily dispersed by birds it is now found in many places as an environmental weed, especially in western Europe but increasingly so in parts of North America, Australia and New Zealand. In such places it is seen as smothering native plant communities, altering their structure and composition. Contact with the plant can in some cases lead to dermatitis, and the berries are reputedly poisonous to children.

Taxonomic Tree

Top of page
  • Domain: Eukaryota
  •     Kingdom: Plantae
  •         Phylum: Spermatophyta
  •             Subphylum: Angiospermae
  •                 Class: Dicotyledonae
  •                     Order: Rosales
  •                         Family: Rosaceae
  •                             Genus: Cotoneaster
  •                                 Species: Cotoneaster horizontalis

Notes on Taxonomy and Nomenclature

Top of page

The genus Cotoneaster includes 261 species in the temperate Old World, with many of these being apomictic aggregates (Mabberley, 2008). The genus is close to Crataegus, but its members have neither spines nor lobed leaves. Of the species C. horizontalis, Mabberley’s very brief description says “apomictic with herring-bone sprays of foliage”.

Lu and Brach (2003), in Flora of China, describe C. horizontalis and present a key for distinguishing the 59 species of Cotoneaster in China, of which 37 are endemic. These authors suggest that there are only about 90 species in the genus in the broad sense. They name two varieties of C. horizontalis: var. horizontalis, with sub-globose fruits, 5-7 mm in diameter, usually with 3 pyrenes (seeds); and var. perpusillus, with ellipsoid fruits, 3-4 mm in diameter, usually with 2 pyrenes. The same authors also mention that plants with variegated and somewhat larger leaves have been described as var. variegatus, and that C. atropurpureus and C. flinckii may be synonymous with C. horizontalis var. horizontalis. However, the Plant List (2013) gives C. horizontalis Decne. as the accepted name, with C. horizontalis var. horizontalis as one of six synonyms. C. horizontalis var. perpusillus, var. wilsonii and f. variegatus are recognized as infraspecific taxa, while C. atropurpureus Flinck & B. Hylm” and C. flinckii J.Fryer & B.Hylm” are recognized as separate species.  

Dickoré and Kasperek (2010) comment that the species is variable and its taxonomy seems poorly understood and blurred by the many putative minor variants of this and other species. Further studies are needed to clarify the taxonomy of the genus, complicated as it is by hybridization and apomixis (Lu and Brach, 2003).


Top of page

C. horizontalis is a shrub, deciduous or semi-evergreen, procumbent, usually to 50 cm tall, sometimes taller, with horizontally spreading, distichously much-branched stems (arranged in two opposite rows in the same plane, like a fish bone), the plant extending to 1-1.5 m. Branchlets blackish brown, round, initially strigose (covered in rigid straight hairs), hairless when old. Petiole 1-3 mm, pubescent; stipules readily detached, brown, narrow or lanceolate, 2-4 mm, puberulous; leaf blade suborbicular or broadly elliptic, rarely obovate, 6-14 × 4-9 mm, midvein raised abaxially and ± impressed adaxially, abaxially sparsely accumbent pubescent, adaxially glabrous, base cuneate, apex usually acute. Inflorescences 1- or 2-flowered. Pedicel short to nearly absent. Flowers 5-7 mm in diameter. Hypanthium (floral tube) campanulate, abaxially sparsely pubescent. Sepals triangular, 1-1.5 × 1-2 mm, apex acute. Petals erect, pink, reddish, or whitish, 3-4 × 2-3 mm, base shortly clawed, apex obtuse. Stamens about 12, shorter than petals. Ovary pilose apically; styles 2 or 3, free, not exceeding stamens. Fruit bright red, subglobose or ellipsoid, (3-)5-7 mm in diameter, pyrenes (‘seeds’) 3, rarely 2 (based on Lu and Brach, 2003). 

Plant Type

Top of page Broadleaved
Seed propagated
Vegetatively propagated


Top of page

The USDA-ARS (2015) lists C. horizontalis as native to temperate China and Taiwan, as well as Nepal but, according to Dickoré and Kasperek (2010), frequent misidentifications with other species of Cotoneaster have resulted in confusion over its native range and some records such as those from Tibet, Taiwan and Nepal may be of different species or be adventive.

In Japan, Australia, New Zealand, most of Europe, the USA and Canada, C. horizontalis has been introduced as an ornamental garden shrub and is now naturalized and often invasive in many areas (USDA-ARS, 2015). In Australia, the Queensland Government (2015) notes that C. horizontalis is sparingly naturalized in Victoria and southeastern South Australia, and possibly New South Wales. It is regarded as an environmental weed in Victoria and as a potential environmental weed in New South Wales, Tasmania and the Australian Capital Territory. The same source indicates that the species is naturalized in Washington and Oregon in northwestern USA. In British Columbia in western Canada, according to the Garry Oak Ecosystems Recovery Team (2015), C. horizontalis (rock cotoneaster) has escaped from gardens on southern Vancouver Island, the Gulf Islands, the Sunshine Coast and the lower Fraser Valley, and has been present in Garry oak (Quercus garryana) ecosystems in British Columbia since 2002.

Distribution Table

Top of page

The distribution in this summary table is based on all the information available. When several references are cited, they may give conflicting information on the status. Further details may be available for individual references in the Distribution Table Details section which can be selected by going to Generate Report.

Continent/Country/RegionDistributionLast ReportedOriginFirst ReportedInvasiveReferenceNotes


ChinaWidespreadNativeUSDA-ARS, 2015
-GansuPresentNativeUSDA-ARS, 2015
-GuizhouPresentNativeUSDA-ARS, 2015; USDA-ARS, 2015
-HubeiPresentNativeUSDA-ARS, 2015
-HunanPresentNativeUSDA-ARS, 2015
-JiangsuPresentNativeUSDA-ARS, 2015
-ShaanxiPresentNativeUSDA-ARS, 2015
-SichuanPresentNativeUSDA-ARS, 2015
-TibetPresentNativeUSDA-ARS, 2015disputed native range
-YunnanPresentNativeUSDA-ARS, 2015
-ZhejiangPresentNativeUSDA-ARS, 2015
JapanPresentIntroducedUSDA-ARS, 2015
NepalPresentNativeUSDA-ARS, 2015disputed native range
TaiwanPresentNativeUSDA-ARS, 2015disputed native range


EgyptPresent only in captivity/cultivationIntroduced Not invasive Khalifa and Loutify, 2006

North America

CanadaPresentPresent based on regional distribution.
-British ColumbiaLocalisedIntroducedGarry Oak Ecosystems Recovery Team, 2015
-OntarioPresentIntroducedUSDA-NRCS, 2015
USAPresentPresent based on regional distribution.
-CaliforniaPresent, few occurrencesIntroduced2007Jacobson and Zika, 2009First collection for California as an escape from cultivation: Del Norte County, pasture edge, 3 air km N of Lake Earl, elev. 8 m, 18 July 2007
-OregonPresentIntroducedUSDA-NRCS, 2015
-WashingtonPresentIntroducedUSDA-NRCS, 2015

South America

ArgentinaPresent, few occurrencesIntroducedGiorgis et al., 2011Cordoba mountains, 1825 m


AustriaWidespreadIntroducedNOBANIS, 2015very common, first recorded 1935 but introduced earlier
BelgiumLocalisedIntroduced1982 Invasive Dickoré and Kasperek, 2010; Verloove, 2013; NOBANIS, 2015first noted in the wild in 1982
Czech RepublicPresentIntroduced Invasive Dickoré and Kasperek, 2010
DenmarkPresent, few occurrencesIntroduced Not invasive NOBANIS, 2015rare
FranceWidespreadIntroduced Invasive Farille et al., 2010; Tela Botanica, 2015
GermanyPresentIntroduced Invasive Dickoré and Kasperek, 2010; NOBANIS, 2015
HungaryPresentIntroduced Invasive Dickoré and Kasperek, 2010
IrelandWidespreadIntroduced Invasive NOBANIS, 2015common, potentially invasive
ItalyLocalisedIntroduced Not invasive Celesti-Grapow et al., 2010naturalized in Lombardy and casual occurrences in Alto-Adige, Trentino and Veneto
LithuaniaPresent, few occurrencesIntroduced1992 Not invasive NOBANIS, 2015rare, first record 1992
LuxembourgPresentIntroduced Invasive Dickoré and Kasperek, 2010
NetherlandsPresent, few occurrencesIntroduced Not invasive Dickoré and Kasperek, 2010rare, potentially invasive
NorwayPresentIntroduced1982 Invasive NOBANIS, 2015first record 1982
PolandPresentIntroduced Invasive Dickoré and Kasperek, 2010
SwedenPresent, few occurrencesIntroduced1969 Invasive Hylmo, 1993rare
SwitzerlandPresentIntroduced Invasive Dickoré and Kasperek, 2010
UKWidespreadIntroduced Invasive BSBI, 2015
-Channel IslandsPresentIntroducedBSBI, 2015


AustraliaPresentIntroducedUSDA-ARS, 2015
-New South WalesPresentIntroducedQueensland Government, 2015Possibly naturalized, potential environmental weed
-South AustraliaPresentIntroducedQueensland Government, 2015Sparingly naturalized in the south-east
-TasmaniaPresentIntroducedQueensland Government, 2015potential environmental weed
-VictoriaPresentIntroducedQueensland Government, 2015Sparingly naturalized, environmental weed
New ZealandPresentIntroducedWebb et al., 1988; USDA-ARS, 2015

History of Introduction and Spread

Top of page

C. horizontalis is native to western China, where, according to Lu and Brach (2003), it is found in several provinces at altitudes of 1500 to 3500 m. From there, after it was recognized by European plant collectors in the late nineteenth century as a potentially attractive garden species, specimens were taken to western Europe and then on to North America and Australasia. In 1916 the Arnold Arboretum in Boston, USA, recognized it as having been “first raised in France forty years ago” (i.e., around the 1860s-1870s) and had been growing in the Arboretum for about twenty five years (Arnold Arboretum, 1916), continuing “this is an excellent plant for a large rock garden, and in Europe it is often trained to cover low walls, for which purpose it is well suited”. This and other favourable descriptions no doubt led to its wide adoption as a garden plant.

According to Pilkington (2011), C. horizontalis has been a highly popular garden plant in Britain since around 1889 when it was first introduced into cultivation. However, from 1940 onwards naturalized plants of this species began to be recorded from the wild in Britain, with subsequent rapid spread; in 1969, for example, sampling of 10-km grid squares centred around Oxfordshire recorded the species in 53, and 30 years later it was found in 961 squares, with the strongest populations apparently associated with light soils in southern England. It has also been observed in the wild in the Republic of Ireland and France (Pilkington, 2011). In Belgium the first observation of the plant in the wild was in 1982 and it is now classed as a ‘black list’ alien invasive weed with high environmental risk (Belgian Forum on Invasive Species, 2015).


Top of page
Introduced toIntroduced fromYearReasonIntroduced byEstablished in wild throughReferencesNotes
Natural reproductionContinuous restocking
France China 1860-1870s Horticulture (pathway cause) Yes Arnold Arboretum (1916)
New Zealand 1994 Horticulture (pathway cause) Yes New Zealand Plant Conservation Network (2015) first reported naturalized in 1994
UK 1889 Horticulture (pathway cause) Yes Pilkington (2011) first introduced into cultivation 1889, naturalized from about 1940 onwards
USA Europe 1880 Horticulture (pathway cause) Yes Slabaugh and Shaw (2008) first introduced into cultivation 1880

Risk of Introduction

Top of page

Although C. horizontalis is widely grown in many gardens in temperate climates and has escaped and naturalized in several countries, it could spread yet further in those countries, and might also become naturalized in countries in which it is at present confined to gardens. 

According to a risk assessment undertaken by Boer (2014), scores of 3 (high risk) were given to C. horizontalis and other small-leaved Cotoneaster species in Belgium for: (i) dispersion potential or invasiveness; (ii) colonization of high conservation value habitats; and (iii) adverse impacts on native species. There was also a medium risk (score 2) of alteration of ecosystem functions. Taking into account the situation in Belgium, the risk of establishment of C. horizontalis in neighbouring Netherlands was considered high. 


Top of page

The Flora of China (Lu and Brach, 2003) describes the native habitat of C. horizontalis as thickets, rocks, rocky slopes and dry mountain areas from 1500 to 3500 m. Similarly, where it has been introduced and become naturalized, it grows in rocky situations and, in urban areas, in cracks in stone walls, along roadsides, on house walls, along railways and in industrial areas (Dickoré and Kasperek, 2010).

In Belgium C. horizontalis has invaded calcareous grasslands considered biodiversity hotspots, and other high value habitats like sand dunes, as well as large urban centres (Piqueray et al., 2008). It is less common in eastern Belgium, however, where soils tend to be more acidic. In Britain too it invades limestone grassland, limestone pavement and cliffs, as well as railway banks, quarries, pavements, chalk pits and walls. It seems particularly fond of light, well-drained calcareous substrates (Pilkington, 2011). In Germany, C. horizontalis is found in rocky situations but also dry calcareous grassland with rocky outcrops such as in northern Hesse and in the Jura of Baden Württemberg and Bavaria (Dickoré and Kasperek, 2010).

In southeastern Australia, the species appears to have the potential to become an invasive weed in subalpine and alpine vegetation, particularly on rocky slopes (Queensland Government, 2015).

C. horizontalis is a shade intolerant shrub. In its native environment in China it seems able to colonize openings in the canopy regardless of slope aspect. In Mianning County in Sichuan Province it is an extremely abundant shrub on south-facing slopes (Jarvis and Liu, 1993).

Habitat List

Top of page
Terrestrial – ManagedCultivated / agricultural land Principal habitat Productive/non-natural
Rail / roadsides Secondary/tolerated habitat Natural
Urban / peri-urban areas Secondary/tolerated habitat Harmful (pest or invasive)
Urban / peri-urban areas Secondary/tolerated habitat Natural
Urban / peri-urban areas Secondary/tolerated habitat Productive/non-natural
Buildings Secondary/tolerated habitat Harmful (pest or invasive)
Terrestrial ‑ Natural / Semi-naturalNatural grasslands Secondary/tolerated habitat Harmful (pest or invasive)
Rocky areas / lava flows Principal habitat Natural
Scrub / shrublands Principal habitat Natural
Arid regions Principal habitat Natural
Coastal dunes Secondary/tolerated habitat Harmful (pest or invasive)

Hosts/Species Affected

Top of page

Piqueray et al. (2008) compared plots invaded by C. horizontalis with uninvaded plots in calcareous grassland in the vicinity of Dinant in southern Belgium and found that the species reduced the area of rocky or bare ground, leading to an increase in moss cover. Its presence also reduced the richness and diversity of the grassland, the four species whose incidence was most significantly reduced being Allium sphaerocephalon, Bromus erectus, Festuca lemanii and Lactuca perennis. Four other species were also affected, but not significantly so.

Biology and Ecology

Top of page


In Germany, both triploid (2n = 3x = 51) and tetraploid (2n = 4x = 68) individuals have been identified (Piqueray et al., 2008). In Britain, Pilkington (2011) describes C. horizontalis as an apomictic tetraploid, while Zika (2013) also reports the species as tetraploid.

Reproductive Biology

Although C. horizontalis is apomictic, producing seed without pollination and subsequent fertilization, it can also set seed after pollination. The dark red-pink flowers are bisexual, epigynous (the ovary is below the petals and sepals), borne in small clusters, pollinated by insects and abundant in spring (Pilkington, 2011).

Pilkington (2011) as well as Dickoré and Kasperek (2010) say that plants are self-layering; where tips of the aerial branches of C. horizontalis contact the ground they can root and form daughter plants.

Physiology and Phenology

C. horizontalis flowers May-June and fruits September-October in the northern hemisphere (Lu and Brach, 2003).

Germination usually takes between 1 and 18 months but seeds can sometimes remain viable for 5 years (Pilkington, 2011). Seeds are dormant when fresh and fruits contain inhibitory substances in the fleshy fruit coat which delay germination. Stratification in the soil at 5oC for 3-4 months, 1.5-3 hours in sulfuric acid or exposure to alternating temperatures of 15o and 24oC can break seed dormancy (Grbic and Skocajic, 2004). Germination rates for plants invading calcareous grasslands in Belgium were found to be around 30% (Frisson et al., 2008).

He (2011), in studies on the growth and development of C. horizontalis in its native China, found that additional nitrogen and/or potassium helped improve flowering rate. In calcareous grassland in Belgium, flowering was observed in every plant over 4 years of age (Piqueray, 2008), while the capacity for fruiting was noted in plants at 3 years (Frisson et al., 2008).


According to the Royal Horticultural Society (2015), C. horizontalis takes 10-20 years to attain its final height and spread. Piqueray et al. (2008) observed 14-year-old plants growing in the wild in invaded calcareous grasslands in Belgium. The USDA-NRCS (2015) describes the lifespan of this species as moderate.

Population Size and Structure

C. horizontalis invading calcareous grasslands in Belgium was observed to have a population density between 0.34 and 10 individual plants per hectare, with a high proportion of small young individuals in the most highly invaded areas (Frisson et al., 2008).


Jarvis and Liu (1993) describe species associated with C. horizontalis in its native environment among Pinus yunnanensis - sclerophyllous broadleaved forests in Sichuan Province of China. Piqueray et al. (2008) recorded species of plants growing close to C. horizontalis in calcareous grasslands in Belgium, indicating that it mostly occurred in complexes of xeric grasslands and calcareous rock; some plants were found at the edges or grasslands or under scrub.

Webber et al. (2012) observed insect visitors to C. horizontalis (and other plant species, both native and exotic) in Christchurch Botanic Gardens in New Zealand. They listed visits by six native insects, with the commonest being a ‘black hoverfly’, and visits by seven species of exotic (introduced) insects, with by far the commonest visitors being bumblebees, followed by honey bees (Apis mellifera) and ‘drone flies’. The behaviour of honey bees feeding on C. horizontalis flowers is described by Pflumm (1983). In Britain the plant attracts large numbers of wasps (Vespula vulgaris) as well as bees (Ian Popay, personal observation, 2014). In addition, the ants Formica fusca and Lasius alienus have been observed exclusively in early May feeding on pollen and nectar from the flowers (Carvalheiro et al., 2008).

One of the reasons for the popularity of the species as a garden plant is its small bright red fruits which make a bold garden display in autumn and into winter. According to Pilkington (2011) ripe fruits are eaten by thrushes and finches which then disperse the seeds in their droppings.

Environmental Requirements

C. horizontalis appears to perform best where it has been introduced to areas with cooler temperate climates like the northern USA, southern Canada and northern Europe. In North America the species does well on dry, rocky sites (Garry Oak Ecosystems Recovery Team, 2015). Similarly in Europe, it is best suited to dry, rocky ground, similar to its habitat in its native China. It has been found invading dry calcareous grasslands in Belgium (Piqueray et al., 2008), England (Pilkington, 2011; Larkin, 2012) and Germany (Dickoré and Kasperek, 2010).

Pilkington (2011) reports that the species grows particularly well in light, well-drained calcareous substrates, and is appreciated by gardeners for its toughness and tolerance of north and east-facing situations. It also seems to be tolerant of dry and nutrient-poor sites. According to the Royal Horticultural Society (2015), C. horizontalis tolerates chalk, clay, loam and sandy soils with a pH ranging from acid to alkaline. In experiments testing the salt tolerance of some ornamental shrubs, Marosz (2004) found that C. horizontalis was not affected by soil salinity of 12 mS cm−1 electrical conductivity.

In its native environment in China, C. horizontalis occurs at altitudes of 1500 to 3500 m. In the Alps of Europe, according to Dickoré and Kasperek (2010), it ascends to at least 1000 m.


Top of page
Cf - Warm temperate climate, wet all year Tolerated Warm average temp. > 10°C, Cold average temp. > 0°C, wet all year
Cs - Warm temperate climate with dry summer Tolerated Warm average temp. > 10°C, Cold average temp. > 0°C, dry summers
Cw - Warm temperate climate with dry winter Preferred Warm temperate climate with dry winter (Warm average temp. > 10°C, Cold average temp. > 0°C, dry winters)
Dw - Continental climate with dry winter Preferred Continental climate with dry winter (Warm average temp. > 10°C, coldest month < 0°C, dry winters)

Soil Tolerances

Top of page

Soil drainage

  • free

Soil reaction

  • acid
  • alkaline
  • neutral

Soil texture

  • heavy
  • light
  • medium

Special soil tolerances

  • infertile
  • saline
  • shallow

Natural enemies

Top of page
Natural enemyTypeLife stagesSpecificityReferencesBiological control inBiological control on
Agrilus sinuatus Herbivore Stems not specific Pilkington, 2011
Athrips rancidella Herbivore Leaves not specific Pilkington, 2011
Corythucha cydoniae Herbivore Leaves not specific Pilkington, 2011
Eriophyes pyri Herbivore Fruits/pods/Leaves not specific Pilkington, 2011
Erwinia amylovora Pathogen Whole plant not specific Obradovic and Ivanovic, 2013
Odocoileus virginianus Herbivore Whole plant not specific Fargione et al., 1991
Phytophthora ramorum Pathogen Growing point/Leaves not specific Bulajic et al., 2010

Notes on Natural Enemies

Top of page

Where C. horizontalis occurs in North America, browsing by white-tailed deer (Odocoileus virginianus) can cause severe damage (Fargione et al., 1991), although Dickoré and Kasperek (2010) say that the species is highly resistant to cutting and to grazing. According to Pilkington (2011), pests in the USA include the hawthorm lace bug (Corythucha cydoniae), the cotoneaster webworm (Athrips rancidella), scale and spider mites, sinuate pear tree borer (Agrilus sinuatus) and pear leaf blister mite (Phytoptus pyeri [Eriophyes pyri]).

Cotoneaster species are susceptible to fireblight (caused by Erwinia amylovora), but this is more likely to be a problem with larger and later flowering species, although its presence on C. horizontalis has been reported in Serbia (Obradovic and Ivanovic, 2013) and Turkey (Bastas and Sahin, 2012). Leaf spots and canker have been reported in the USA. The fungal disease Phytophthora ramorum has been found on ornamental specimens of C. horizontalis in Serbia (Bulajic et al., 2010).

Means of Movement and Dispersal

Top of page

Natural Dispersal

Self-layering, the rooting of arching tips of branches and establishment of new plants, can be a method of local spread of established C. horizontalis plants (Pilkington, 2011).

Vector Transmission (Biotic)

In many countries, and presumably in its native China, the bright red berries are attractive to birds that eat them and excrete the seeds, although most dispersal would be fairly localized to areas around the birds’ roosts (Verloove, 2013).

Accidental Introduction

As a popular garden plant, its further spread from cultivation seems inevitable. Viable propagules (berries, seeds) can easily be dispersed through improper disposal of cuttings and other garden waste.  

Intentional Introduction

C. horizontalis has already been introduced to many temperate countries as an attractive garden plant and its further spread in those countries is inevitable since it is readily available in garden centres and also as seeds from websites (Halford et al., 2011). It is much promoted in gardening journals and newspaper columns (see for example Dusoir, 2002). In several countries where C. horizontalis is recognized as an invasive alien, its planting in the wild is prohibited. The New Zealand Department of Conservation (2015), for example, regards the species as threatening forest and shrubland, as well as grasslands, cliffs and riverbeds and classes it as a Category C invasive weed, which is “potentially troublesome and should not be spread around”.

Pathway Vectors

Top of page
VectorNotesLong DistanceLocalReferences
Debris and waste associated with human activities Yes

Impact Summary

Top of page
Cultural/amenity Positive
Economic/livelihood Negative
Environment (generally) Negative
Human health Negative

Economic Impact

Top of page

Pilkington (2011) points out that in urban areas the species has the potential to damage buildings by rooting in crumbling mortar and cracks, and can be costly to eradicate from wildlife areas. Investigations into the costs and efficacy of techniques for controlling C. horizontalis and other invasive species in invaded ecosystems in Belgium found that invasive plant management is usually expensive and hard to implement but some techniques showed encouraging results (Mahy and Halford, 2009; Frisson et al., 2010).

Environmental Impact

Top of page

Impact on Habitats

C. horizontalis is one of the alien plant species invading lowland calcareous (chalk) grassland in the Wild Park Local Nature Reserve near Brighton in the South Downs of southern England (Larkin, 2012), and its impacts in calcareous grasslands in Belgium have been described by Piqueray et al. (2008). In northern Victoria, Australia, C. horizontalis has escaped from gardens in the Mt Buffalo National Park and spread to nearby alpine slopes (Queensland Government, 2015).

Impact on Biodiversity

Species richness of calcareous grasslands in Belgium is significantly lower on plots invaded by C. horizontalis (Frisson et al., 2008), so that locally it may pose a serious threat to biodiversity (Pilkington, 2011). Impact assessments carried out by Piqueray et al. (2008) indicated that the presence of C. horizontalis was associated with changes in both the structure and composition of calcareous grassland communities, caused by decreasing species richness and diversity and directly affecting grassland specialist species; they found significant reductions in invaded grasslands in populations of Allium sphaerocephalon, Bromus erectus, Festuca lemanii and Lactuca perennis. Four other species were also affected, but not significantly so. It is expected that these impacts will intensify over time as the C. horizontalis plants grow larger and cover more ground.

In the complex interrelationship between calcareous grasslands, black ants (Lasius alienus) and the silver-studded blue (Plebejus argus), a rare and declining heathland butterfly confined to small colonies in Britain, large areas of sparse grassland and bare ground are essential to encourage black ants, which in turn tend the butterfly’s larvae and chrysalises in spring, sometimes in their nests, and the adult butterfly as it emerges in summer. A 10-year study in Broadcroft Quarry in Dorset showed a loss in area of bare ground, the most suitable breeding habitat for the silver-studded blue, due to the spread of invasive plants such as bramble (Rubus fruticosus), Clematis, coarse grasses and the non-native C. horizontalis. These shaded out the main larval food plant, bird's-foot trefoil (Lotus corniculatus), and made conditions unsuitable for ants (de Whalley et al., 2006).

Threatened Species

Top of page
Threatened SpeciesConservation StatusWhere ThreatenedMechanismReferencesNotes
Plebejus argus (silver-studded blue butterfly)No DetailsEngland and WalesCompetition - shadingWhalley et al., 2006

Social Impact

Top of page

Weller and Ormerod (1996) reported contact dermatitis in a 43-year-old woman after contact with pruning waste of C. horizontalis. In addition, the fruits are said to be poisonous to humans; containing cyanogenic glycosides, the berries are toxic if eaten in large quantities, especially by children (HerbiGuide, 2015). The Royal Horticultural Society (2015) notes that ingestion of berries results in a mild stomach upset.

Risk and Impact Factors

Top of page Invasiveness
  • Proved invasive outside its native range
  • Has a broad native range
  • Pioneering in disturbed areas
  • Benefits from human association (i.e. it is a human commensal)
  • Long lived
  • Has high reproductive potential
  • Reproduces asexually
Impact outcomes
  • Altered trophic level
  • Ecosystem change/ habitat alteration
  • Reduced native biodiversity
  • Threat to/ loss of endangered species
  • Threat to/ loss of native species
Impact mechanisms
  • Causes allergic responses
  • Competition - shading
Likelihood of entry/control
  • Highly likely to be transported internationally deliberately
  • Difficult/costly to control


Top of page

Economic Value

As C. horizontalis is a very popular garden plant in many countries, it is widely raised and sold by nurseries (Piqueray et al., 2008; Halford et al., 2011; Pilkington, 2011).

Social Benefit

C. horizontalis is widely grown as an ornamental low-maintenance ground cover on banks and slopes, and as a hedge or screen against walls. It is considered very decorative, due to the orange and red autumnal colour of the leaves and the bright red fruits which make a bright and cheerful display in autumn and winter (Royal Horticultural Society, 2015).

When Sokkar et al. (2013) explored the antioxidant, anticancer and hepatoprotective properties of C. horizontalis, they found that ethanolic extracts of the branches contained β-carotene, ascorbic acid and lesser amounts of α-tocopherol and amygdalin. Toxicity testing showed that the ethanolic extracts could be considered as a potential source of natural antioxidant with hepatoprotective, hypolipidaemic and other properties.

Environmental Services

The flowers of C. horizontalis are highly attractive to bees and other insects as sources of nectar, while the berries attract birds and help sustain them through the winter (Pilkington, 2011). When grown against a wall, the plant provides sheltered nesting and roosting sites for birds.

Uses List

Top of page


  • Landscape improvement
  • Ornamental
  • Wildlife habitat


  • garden plant

Similarities to Other Species/Conditions

Top of page

Lu and Brach (2003) stressed that the taxonomy of the genus needs clarification, especially since species hybridize and many of the species are apomictic.  These two complications will continue to dog the identification of this and other Cotoneaster species in the future.

Dickoré and Kasperek (2010) describe the species of Cotoneaster found in Central Europe and mention several species with which C. horizontalis might be confused there. C. apiculatus is similar to C. horizontalis but has larger, almost circular, leaves. C. divaricatus has in the past been misidentified as C. horizontalis and young plants can be very similar. However it has a more erect habit, larger, more oblong leaves and cylindrical fruit. C. nitidus also resembles C. horizontalis in its regular distichous branches but the backs of the leaves are covered with short hairs lying flat on the surface, the inflorescences are 2-6 flowered, the peduncles reflexed and the fruits scarlet in colour. C. verruculosus is also similar but differs in its reduced indumentum, with bases of hairs giving a characteristic prickly appearance to the shoots. C. nitens is also similar in habit but is rarely planted in Central Europe and has black cylindrical fruit.

In New Zealand, C. conspicuus is similar to C. horizontalis but can be distinguished by its slightly arching branches (not in a herring-bone pattern), spreading rather than erect petals, and its scarlet rather than crimson fruits (Heenan et al., 1998).

Prevention and Control

Top of page


Halford et al. (2011) estimated that C. horizontalis was present in 53% of the sales catalogues of horticulturalists in southern Belgium and pointed out that as long as the species is used in the vicinity of calcareous grasslands, propagules are continuously provided, thus jeopardizing management actions. Pilkington (2011) also points out that C. horizontalis is sold by many garden centres and nurseries and adds that prevention measures should focus on discouraging its planting close to wildlife sites that support dry or rocky limestone grassland and other habitats. Piqueray et al. (2008) also stressed the need to prevent its commercialization and to intensively manage high-value habitats.

Public Awareness

In Belgium, Piqueray et al. (2008) suggest that information campaigns in high-risk areas would raise public awareness and early management of infestations. Dickoré and Kasperek (2010) say that the common practice of large-scale planting of mass produced non-native species, such as C. horizontalis, in the open landscape should be “considered a dangerous potential pool of new plant invaders and a deliberate threat to native flora” and should be discouraged. Pilkington (2011) recommends that the Landscape Institute (in Britain) and local authorities should be lobbied to discourage mass plantings in landscape schemes or new developments.

Halford et al. (2011) surveyed the attitudes of nurserymen (producers and sellers in horticultural federations) and private managers (landscape architects and garden contractors) to the possible introduction of codes of conduct on the use of invasive garden plants in Belgium and found that 61% of nursery men and 73% of private managers would be prepared to endorse such a code of conduct. Furthermore, 86% of gardeners would agree to buy their ornamental plants in nurseries espousing such a code.


Frisson et al. (2008) and Pilkington (2011) stress that, once established in wildlife areas, C. horizontalis is expensive and difficult to remove. Dickoré and Kasperek (2010) say that complete eradication of alien Cotoneaster species, other than on a very expensive, local and temporary scale, would “simply be impossible”.


Frisson et al. (2008) and Mahy and Halford (2009) tested both mechanical and chemical management techniques in calcareous grasslands, where highly selective techniques are needed for environmental reasons. They tried pulling, cutting, stump burning, spraying, cutting plus glyphosate application on stump or on stems. Efficiency, cost, feasibility and side-effects on ecosystems were estimated. Spraying and single cutting are not deemed efficient since resprouting occurs. The authors concluded that the best control method available was to cut and paint the stump with glyphosate.

Physical/Mechanical Control

The Tamar Valley Weed Strategy Working Group (2015) recommends grubbing cotoneasters at any time of the year, taking care to remove seedlings and small bushes using a mattock or spade. Pilkington (2011) also suggests care should be taken that fruits do not fall on the ground. The same author says that removing stumps and roots is vital as otherwise they can resprout, a comment echoed by the Garry Oak Ecosystems Recovery Team (2015). The latter group also suggests that cotoneaster plants should be removed when the soil is moist and that disturbed soil should be immediately replanted with desirable native species to prevent reinfestation.

Mechanical scraping was used in Broadcroft Quarry, Dorset, UK, to remove the rich topsoil and C. horizontalis roots to encourage the establishment and development of a less competitive, low nutrient-requiring calcareous flora (de Whalley et al., 2006).

The Garry Oak Ecosystems Recovery Team (2015) suggest that seedlings can be smothered with mulch or black plastic, but warn that burning is not effective for controlling regeneration. Jarvis and Liu (1993) point out that C. horizontalis regenerates quickly after fire.

Mahy and Halford (2009) found pulling plants from the soil manually was extremely difficult and ineffective.

Chemical Control

To kill small shrubs and control regrowth, the herbicides glyphosate or triclopyr can be applied as a wiper or by handheld sprayer when plants are actively growing between spring and autumn. Alternatively, these herbicides can be applied to cut stumps or to abraded bark (Pilkington, 2011; Garry Oak Ecosystems Recovery Team, 2015).


Top of page

Arnold Arboretum, 1916. Chinese cotoneasters. Bulletin of Popular Information, Arnold Arboretum, Harvard University. New series, 2(6):21-23.

Bastas KK; Sahin F, 2012. First report of fire blight disease caused by Erwinia amylovora on rockspray (Cotoneaster horizontalis) in Turkey. Plant Disease, 96(11):1690.

Belgian Forum on Invasive Species, 2015. Invasive species in Belgium. Cotoneaster horizontalis - rockspray.

Boer E, 2014. Risk assessment: Cotoneaster. Leiden, Netherlands: Naturalis Biodiversity Center, 19 pp.

BSBI, 2015. BSBI Maps Scheme: Hectad maps. Bristol, UK: Botanical Society of Britain & Ireland.

Bulajic A; Djekic I; Jovic J; Krnjajic S; Vucurovic A; Krstic B, 2010. Phytophthora ramorum occurrence in ornamentals in Serbia. Plant Disease, 94(6):703-708.

Carvalheiro LG; Barbosa ERM; Memmott J, 2008. Pollinator networks, alien species and the conservation of rare plants: Trinia glauca as a case study. Journal of Applied Ecology, 45(5):1419-1427.

Celesti-Grapow L; Pretto F; Carli E; Blasi C, 2010. The allochthonous and invasive vascular flora of the Italian regions (Flora vascolare alloctona e invasiva delle regioni d'Italia). Rome, Italy: Casa Editrice Universita La Sapienza, 207 pp.

Dickoré WB; Kasperek G, 2010. Species of Cotoneaster (Rosaceae, Maloideae) indigenous to, naturalising or commonly cultivated in Central Europe. Willdenowia, 40(1):13-95.

Dusoir R, 2002. How to grow: Cotoneaster horizontalis. The Telegraph (UK), 7 December 2002.

Fargione MJ; Curtis PD; Richmond ME, 1991. Resistance of woody ornamental plants to deer damage. Ithaca, New York, USA: Cornell University, 4 pp. [Cornell Cooperative Extension Fact Sheet 800.00.]

Farille MA; Zeller J; Jordan D; Charpin A; Garraud L, 2010. Cotoneasters escaped from gardens in Haute-Savoie. (Cotoneasters echappes de jardins en Haute-Savoie.) Le Monde des Plantes, 501:1-6.

Frisson G; Halford M; Delbart E; Mahy G, 2010. Can we restore natural habitats after plant invasion? Lessons from years of management. In: SER2010: 7th SER European Conference on Ecological Restoration, Avignon, France, 23-27 August 2010.

Frisson G; Piqueray J; Halford M; Mahy G; Vanderhoeven S, 2008. Cotoneaster horizontalis on calcareous grasslands in Belgium: from ornament to management. In: Poster session. 1st Meeting of the Working Group on Dry Grasslands in the Nordic and Baltic Region, Kiel, Germany, 28-30 August 2010.

Garry Oak Ecosystems Recovery Team, 2015. Cotoneaster species: cotoneasters. Victoria, British Columbia, Canada: Garry Oak Ecosystems Recovery Team Society, 4 pp.

Giorgis MA; Tecco PA; Cingolani AM; Renison D; Marcora P; Paiaro V, 2011. Factors associated with woody alien species distribution in a newly invaded mountain system of central Argentina. Biological Invasions, 13(6):1423-1434.

Grbic M; Skocajic D, 2004. Determination of intensity of seed inhibitory dormancy by lactuca test. In: Proceedings of an international scientific conference marking 75 years of the Forest Research Institute of the Bulgarian Academy of Sciences, Sofia, Bulgaria, 1-5 October 2003. Volume 2 [ed. by Rossnev, B.\Kitanova, S.\Alexandrov, A.\Raev, I.\Tsakov, H.\Dimitrov, V.\Grozeva, M.\Petrova, R.\Popov, G.\Grigorov, G.]. Sofia, Bulgaria: Forest Research Institute, 131-137.

Halford M; Heemers L; Mathys C; Vanderhoeven S; Mahy G, 2011. Socio-economic survey on invasive ornamental plants in Belgium. AlterIAS Project final report. Gembloux, Belgium: Gembloux Agro-Bio Tech, University of Liege, 31 pp.

He Q, 2011. Preliminary introduction studying of Cotoneaster horizontalis Dcne. in Nan Tian Hu of Feng Du County. MSc Thesis, China.

Heenan PB; Breitwieser I; Glenny DS; Lange PJ de; Brownsey PJ, 1998. Checklist of dicotyledons and pteridophytes naturalised or casual in New Zealand: additional records 1994-1996. New Zealand Journal of Botany, 36(2):155-162.

HerbiGuide, 2015. Cotoneaster.

Hylmo B, 1993. Wall-spray, Cotoneaster, in Sweden. (Oxbar, Cotoneaster, i Sverige.) Svensk Botanisk Tidskrift, 87:305-330.

Jacobson AL; Zika PF, 2009. California. Madrono, 56(1):64-65.

Jarvis DI; Liu HL, 1993. Vegetation patterns in the Pinus yunnanensis-sclerophyllous broadleaved forests, Mianning County, Sichuan Province, China. Journal of Biogeography, 20(5):505-524.

Khalifa SF; Loutify MH, 2006. Ornamental cultured plant collection. In: Proceedings of the First International Conference on "The strategy of botanic gardens", 10-12 May 2006, Dokki, Egypt. 61.

Larkin D, 2012. Wild Park, draft management plan 2012-2022. Brighton, UK: Brighton & Hove City Council, 46 pp.

Lu LD; Brach AR, 2003. 10. Cotoneaster Medikus, Philos. Bot. 1: 154. 1789. In: Flora of China, Volume 9. 85-108.

Mabberley DJ, 2008. Mabberley's plant-book: a portable dictionary of plants, their classifications and uses, Ed.3 [ed. by Mabberley, D. J.]. Cambridge, UK: Cambridge University Press, xviii + 1021 pp.

Mahy G; Halford M, 2009. Testing management practices on three invasive plants in the Walloon Region (Acer rufinerve, Cotoneaster horizontalis and Spiraea spp.) and sensitization of Department of Nature and Forests agents to the problem of biological invasions. Final report (Tests de methodes de gestion sur trois plantes invasives en Region wallonne (Acer rufinerve, Cotoneaster horizontalis et Spiraea spp.) et sensibilisation des agents DNF a la problematique des invasions biologiques - Rapport final). Gembloux, Belgium: Gembloux Agro-Bio Tech, 28 pp.

Marosz A, 2004. Effect of soil salinity on nutrient uptake, growth, and decorative value of four ground cover shrubs. Journal of Plant Nutrition, 27(6):977-989.

New Zealand Department of Conservation, 2015. Protecting and restoring our natural heritage - a practical guide. Appendix one: invasive weeds. Wellington, New Zealand.

New Zealand Plant Conservation Network, 2015. Cotoneaster horizontalis.

NOBANIS, 2015. Cotoneaster horizontalis (Rosaceae, Angiosperms). European Network on Invasive Alien Species (NOBANIS).

Obradovic A; Ivanovic M, 2013. Fire blight of ornamental plants. (Bakteriozna plamenjaca ukrasnih biljaka.) Biljni Lekar (Plant Doctor), 41(6):648-651.

Pflumm W, 1983. Pollen collection and pollination of cotoneaster (Cotoneaster horizontalis) by honeybees. (Zum Sammel- und Putzverhalten der Honigbiene auf einer abbluhenden Zwergmispel (Cotoneaster horizontalis).) Behaviour, 83(1/2):112-131.

Pilkington S, 2011. Wall cotoneaster, Cotoneaster horizontalis. York, UK: GB Non-Native Species Secretariat, 3 pp.

Piqueray J; Mahy G; Vanderhoeven S, 2008. Naturalization and impact of a horticultural species, Cotoneaster horizontalis (Rosaceae) in biodiversity hotspots in Belgium. Belgian Journal of Botany, 141(2):113-124.

Queensland Government, 2015. Weeds of Australia, Biosecurity Queensland edition. Queensland, Australia.

Royal Horticultural Society, 2015. Cotoneaster horizontalis, wall spray.

Slabaugh PE; Shaw NL, 2008. Cotoneaster Medik. Cotoneaster. In: The woody plant seed manual [ed. by Bonner, F. T. \Karrfalt, R. P.]. Washington, DC, USA: US Forest Service, 442-446. [USDA Forest Service Agriculture Handbook No. 727.]

Sokkar N; El-Gindi O; Sayed S; Mohamed S; Ali Z; Alfishawy I, 2013. Antioxidant, anticancer and hepatoprotective activities of Cotoneaster horizontalis Decne extract as well as alpha-tocopherol and amygdalin production from in vitro culture. Acta Physiologiae Plantarum, 35(8):2421-2428.

Tamar Valley Weed Strategy Working Group, 2015. Cotoneaster. Launceston, Tasmania, Australia.

Tela Botanica, 2015. Cotoneaster horizontalis Decne. Tela Botanica: Le reseau des botanistes francophones. Montpellier, France.

The Plant List, 2013. The Plant List: a working list of all plant species. Version 1.1. London, UK: Royal Botanic Gardens, Kew.

USDA-ARS, 2015. Germplasm Resources Information Network (GRIN). Online Database. Beltsville, Maryland, USA: National Germplasm Resources Laboratory.

USDA-NRCS, 2015. The PLANTS Database. Baton Rouge, USA: National Plant Data Center.

Verloove F, 2013. The genus Cotoneaster (Rosaceae) in the wild in Belgium: a preliminary overview. (Het genus Cotoneaster (Rosaceae) in het wild in Belgie: een voorlopig overzicht.) Dumortiera, 103:3-29.

Webb CJ; Sykes WR; Garnock-Jones PJ, 1988. Flora of New Zealand. Vol. IV. Naturalised pteridophytes, gymnosperms, dicotyledons. Christchurch, New Zealand: DSIR Botany Division, 1365 pp.

Webber CJ; Peterson AJ; Kelly D; Clemens J, 2012. Native and exotic flower visitors in the Christchurch Botanic Gardens and their contrasting plant preferences. New Zealand Natural Sciences, 37:37-49.

Weller R; Ormerod A, 1996. Contact dermatitis from cotoneaster. Contact Dermatitis, 34(6):433-434.

Whalley L de; Whalley B de; Green P; Gammon N; Shreeves W, 2006. Digging scrapes to enhance silver-studded blue Plebejus argus habitat at Broadcroft Quarry, Isle of Portland, Dorset, England. Conservation Evidence, 3:39-43.

Zika PF, 2013. Cotoneaster. Jepson eFlora [ed. by Jepson Flora Project]. Berkeley, California, USA: University of California and Jepson Herbarium.


Top of page

21/10/2014    Original text by:

Ian Popay, Consultant, New Zealand

Distribution Maps

Top of page
You can pan and zoom the map
Save map