Senecio glastifolius
(holly-leaved senecio)

Senecio glastifolius is an erect perennial herb, native to the Eastern and Western Capes of South Africa. It is sold commercially as an ornamental and is often planted in gardens. It is known to have been introduced to Madeira, the UK and California in the USA and is recorded as invasive in Australia and New Zealand. It grows in shrubland and rocky, damp areas such as riverbanks. Where it is invasive, it grows in disturbed areas, agricultural land, open woodlands and coastal areas. In suitable habitats it is an extremely aggressive weed, producing large numbers of wind dispersed seed, and spreading rapidly, forming dense stands and potentially outcompeting native biodiversity. Ecological niche modelling revealed that, if introduced, this species could become invasive across large areas of the world including the west coasts of Canada, USA and South America and parts of Europe and Africa. Early detection and removal of seedlings, often garden escapes, is likely to be the most effective method of control. In New Zealand, infestations are largely controlled by hand-pulling of plants; however some commercially available herbicides have proved effective. The sale of this species has been banned in Tasmania, Australia and the Taranaki region of New Zealand.

S. glastifolius is native to the Eastern and Western Capes of South Africa (USDA-ARS, 2017). In the Eastern Cape it occurs on the southern coast from George to Humansdorp (34°S, 21-26°E) (SANBI, 2002). Its altitudinal range is up to 330 m above sea level (Williams et al., 1999), though Germishuizen and Meyer (2003) reported plants occurring up to elevations of 1090 m.

This species has been introduced to Australia and New Zealand where it has become naturalized and invasive. In Western Australia, S. glastifolius is weedy in Albany, covering an area of about 100 km². Plants are spreading from this area and are now found up to 20 km away (Australian Government, 2003). It was originally recorded in Manjimup (Thompson, 2006), but it is no longer believed to be there (Beautrais, 2013). In New South Wales, S. glastifolius has been present in the Royal National Park in the vicinity of Bundeena, near Sydney, since 1991 (Australian Government, 2003). Beautrais (2013) mapped a now historic record about 20 km further north, dated 1953. In New Zealand, this species occurs primarily in coastal areas on the North Island but also in some inland areas and on the north coast of the South Island close to Nelson (Beautrais, 2013; Council of Heads of Australasian Herbaria, 2017). A particularly heavy infestation is located along State Highway 1 just north of Wellington.

S. glastifolius has also been introduced in the Scilly Isles, UK, where it was first recorded in 1971 (Stace, 1997; Preston et al., 2002), and in Tresco, Cornwall, UK (Parslow, 2010). Apparently it also occurs on rough ground in Guernsey (Beautrais, 2013). Hansen and Sunding (1993), Silva (2002) and Jardim and Sequeira (2008) record S. glastifolius as naturalized in Madeira. Roscoff Gardens on the Brittany coast in France holds images of S. glastifolius appearing as feral around the gardens, however its status is unverified. In the USA, INaturalist (2017) notes a population of S. glastifolius near Sentinel Dome in the Yosemite Valley, California. Its status is also unverified.

In a study of bioclimatic niche patterns for species in their home range of South Africa, Thuiller et al. (2005) modelled and mapped the global areas of potential habitat for S. glastifolius. These included:

North and South America:

• much of the western seaboard of the USA
• the coast of British Columbia, Canada
• the western seaboard of South America, but with southern Peru and northern Chile less acceptable
• the Sao Paulo area
• the Falkland Islands

Europe:

• Ireland
• western Scotland
• southwest England
• the Atlantic coasts of France and Spain
• Corsica

Africa:

• Ethiopia
• Kenya
• Rwanda
• Madagascar and South Africa, but both at lower risk

Asia/Pacific:

• New Guinea

Australia:

• the Western Australia coast from Cape Leewin to Albany                      
• coast of Tasmania
• southern coast of Victoria from Cape Howe, west to Adelaide

New Zealand:

• much of the coast and lowlands of the North Island
• much of the northern coasts of the South Island, and of the southern coasts, both east and west
• Stewart Island coasts

The threatened areas noted by Thuiller et al. (2005) include six global biodiversity hotspots, namely the Mediterranean Basin, California Floristic Province, Chocó-Darién Western Equator, Central Chile, Southwest Australia and New Zealand. With 48 South African species already invasive in Australia and 34 in New Zealand (Thuiller et al., 2005), further extension of the existing ranges of S. glastifolius in these areas seems inevitable. Further, increasing tourism increases invasion rates into Western Europe and the Mediterranean.

In Victoria, Australia, the climate-matching program, CLIMATE (Pheloung, 1996), was used by Agriculture Victoria (2017) to predict the potential distribution of S. glastifolius, based on the climate in overseas localities where it occurs. Results suggested that coastal habitat is the most suitable.

In New South Wales, Australia, Downey et al. (2010) include S. glastifolius as number 332 on a prioritized list of 340 naturalized plant species in New South Wales that have invasive potential. S. glastifolius is also on Australia's National Alert List for Environmental Weeds (Glanznig and Kessal, 2004), which has 28 species of concern. Weber and Panetta (2006) rated S. glastifolius as having a weediness score of 34 (the highest possible score is 47) and it is a terrestrial weed examined in the DEH (Department of Environment and Heritage) Alert List, so is recommended for containment or eradication.

In New Zealand, Beautrais (2013) used Species Distribution Modelling to describe how S. glastifolius is limited by its environment. The variables used were aridity, mean annual temperature range and minimum temperature of the coldest month. The suitable, though currently uninvaded areas are: "most of the northern half of the North Island and associated offshore islands [and] the entire western coast, ... the Wairarapa and southern Hawkes Bay, plus inland Whanganui and Manawatū ... [as well as] the coastal northern third" of the South Island (Beautrais, 2013).

Beautrais (2013) also predicted that the potential habitat for S. glastifolius extends along much of the southern coast of Australia and much of Tasmania, as well as the southern coast of South Africa. The author added that while the South African and New Zealand populations overlap in terms of their temperature range, there is almost no overlap in terms of humidity, so that in New Zealand the niche of S. glastifolius appears to have shifted to a more moist climate. Low winter temperatures probably prevent it occupying extremely dry areas of New Zealand, and wetter areas are not available in South Africa.

Further climate change is expected to impact on potential distributions. For example, Williams et al. (2012) stated, with respect to Queensland, Australia, that "under the current climate, only southeast Queensland contains suitable habitat for the holly-leaved senecio (S. glastifolius), whereas under future scenarios, suitable climatic space for this species contracts southward and the species may cease to exist in Queensland."

In South Africa, S. glastifolius occupies fynbos (fine-leaved heathland or shrubland) with species of the Restionaceae and Proteaceae families. It also occurs on rocky hillsides and in damp areas, e.g. riverbanks and marshy ground (Williams et al., 1999). Despite being within its native range, Wells et al. (1986) regard S. glastifolius as a potential problem plant for South Africa as it is common in the understory of plantation forests and on pastoral land in wetter, disturbed areas; it also spreads after disturbance, particularly fire.

In Albany, Australia, S. glastifolius occurs on grey sand and swampy loams on hillslopes (FloraBase, 2013). It now dominates the understorey vegetation of woodlands of jarrah (Eucalyptus marginata), blackbutt (E. pilularis) and marri (Corymbia haematoxylon) and open woodlands of Allocasuarina (Australian Government, 2003). It is noted as particularly invasive in open, damp areas. Potential habitat in Victoria is mainly coastal (Agriculture Victoria, 2017). It occurs in open areas, such as open woodland, shrubland, hillsides, coastal dunes, roadsides and near waterways (Australian Government, 2003).

In New Zealand, the habitat of S. glastifolius includes wastelands, hillsides, pasture, scrubland and river beds (Webb et al., 1988). It is particularly common on disturbed slopes above motorways (especially State Highway 1, north of Wellington) and in open dunes around the Manawatū coast.

In the Scilly Isles, UK, S. glastifolius occurs in open woodland near a rubbish tip (Stace, 1997).

There are no known effects of this species on crops in New Zealand, as the plant is restricted to wasteland and coastal areas. It does occur among scrub in extensively grazed farmland, where it may eventually come to restrict productivity of preferred forage plants.

Genetics

The base haploid chromosome number of the genus Senecio is 10 (Pelser et al., 2007), and species are generally 2n = 20 or 2n = 40, although some New Zealand populations have a chromosome number of 2n = 80 or 100 (Memory, 2012).

Using chloroplast DNA, Khan et al. (2013) reported S. giessii and S. sisymbrifolius as the closest relatives of S. glastifolius; S. windhoekensis and Erechtites diversifolia are also closely related. Of these, all are from South Africa or western Africa, except for E. diversifolia, which is from the South Island of New Zealand. Pelser et al. (2007), using the ITS region of the DNA, placed S. glastifolius with S. achilleifolius near S. windhoekensis, though Milton (2009) argues this is a misidentification and commonly associates S. glastifolius with S. giessii and E. diversifolia.

Reproductive Biology

Little is known about the reproductive biology of S. glastifolius. It appears to set seed readily in the wild, suggesting it is self-compatible and dispersed by the wind. Capitula produce up to 250 seeds each. In South Africa it is pollinated by bees, and is also visited by butterflies and birds.

Seed appears to germinate readily (within 2 weeks). Germination success ranges from 53-80% and appears to be higher at temperatures above 25°C (Williams et al., 1999). Fire followed by rains encourages germination (Australian Government, 2003). Seeds are viable for several years in soil seedbanks and slashing woodland, cultivation and fires all apparently release seeds from dormancy (Australian Government, 2003; FloraBase, 2013).

Individual plants of S. glastifolius live for at least 3-5 years, while the existence of a few very large plants persisting in sheltered situations, suggests that they can live for at least a decade. In Western Australia it grows between March and August, with germination over the same period, flowering from September to October and fruiting in November and December (FloraBase, 2013).

Flowering predominantly occurs in spring (September to October in New Zealand and South Africa, February to June in California) but occasional second flowerings have been recorded in the summer (January to February in the southern hemisphere).

In New Zealand, detached capitula do not usually result in any germinants (G Rapson, Massey University, Palmerston North, personal observation).

Associations

Williams et al. (1999) list a number of species associated with S. glastifolius, largely in the Wellington region of New Zealand. These tend to be common coastal scrub exotics known from Ulex europaeus, exotic grass and Chrysanthemoides monilifera habitats, along with occasional natives such as Ficinia nodosa, Muehlenbeckia australis and Spinifex sericeus.

In Australia, S. glastifolius is found in forests with Eucalyptus marginata, E. pilularis and Corymbia haematoxylon and in open woodlands of Allocasuarina (Australian Government, 2003).

Environmental Requirements

Beautrais (2013) found that acute cold stress (mean minimum daily air temperature for the coldest month in the range 0-3°C) could limit the distribution of S. glastifolius.

Westhuizen et al. (2010) noted damage by a common defoliating moth (Diota rostrata: Noctuidae, Arctiinae), which is widespread in southern Africa, but in no-choice trials it showed only limited feeding and no survival.

Williams et al. (1999) report herbivory in New Zealand by Nyctemera annulata, the native magpie moth, while Aglossa caprealis, the diurnal moth, feeds on the flowers.

A spittle bug has been seen on the plant in shrubland in New Zealand, but this does not appear to impede its growth in any way. Sheep apparently do browse foliage although it is potentially toxic, but cattle do not seem to graze it (Popay et al., 2010).

Environmental

• Amenity

General

• Botanical garden/zoo

Human food and beverage

• Honey/honey flora

Ornamental

• garden plant

S. glastifolius is extremely easy to detect during the flowering season, but is rather cryptic at other times, at least in New Zealand. Searching for new infestations or remnants is best done during the flowering period (early spring).

S. glastifolius is very conspicuous, at least in the New Zealand flora, because of its dense heads of bright pink capitula, which appear around October. No native plants can be confused with it, and based on its colour it most resembles the South African exotic S. elegans, a small succulent-leaved plant found on coastal dunes, which has ray florets of a much deeper purple. S. glastifolius is also rather like the exotic ragwort (S. jacobaea), which is also weedy in New Zealand (if biocontrol agents are absent), but this has yellow ray florets, lobed leaves, and is shorter and less woody.

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Prevention

Rapid Response

It is recommended that detected seedlings and adults are removed as quickly as possible, before flowering is completed.

Public Awareness

The public should be discouraged from planting S. glastifolius in gardens, as it is likely to escape.

Eradication

Eradication of S. glastifolius outside desired zones or natural ranges is strongly recommended and is being attempted in Albany, Australia.

In Western Australia, eradication was the aim of the Department of Environment and Conservation and was "largely achieved" after seven years, with the caveat that at least five years of follow-up, including in private gardens, is needed to ensure new seedlings are removed (Government of Western Australia, 2017).

In New Zealand, persistent hand-weeding over three or more years can eliminate small infestations.

Control

Physical/Mechanical Control

If early invaders of S. glastifolius are neglected in vulnerable areas then prolific flowering results in stands or populations that are so large and vigorous that spraying becomes the most appropriate control option. The preferred means of control in New Zealand is containment by hand-weeding of the isolated plants at the edges of its range, preventing new populations from developing.

S. glastifolius has been managed routinely in the Manawatū district of New Zealand for the past two decades by hand-pulling of individual plants during the flowering season (early spring), when plants can be easily located. In most environments, S. glastifolius is easily removed by pulling of the plant, holding the stem near its base. Due to the prickly nature of its leaves, including those that have died around the bases of the stems, gloves are recommended when hand-pulling, especially if there are more than fifty plants to pull. Once pulled, the stems are then folded back and partially broken to reduce the supply of resources from the disinterred rootstock to any developing seed and the plants are left out to dry. Seed set in this situation seems to be minimal, so there is little chance of the infestation spreading. In the most infested areas, e.g. dunes, this is a very easy and effective control mechanism and very little seed set, if any, takes place in plants treated in this way. If areas are protected from reinfestation, and are regularly treated in this way for 2-3 years, then eradication appears to be achievable.

However, grubbing or hand-pulling is not recommended by TERRAIN (2017) as the plants will regrow from root fragments, though this information may refer to ragwort (S. jacobaea). In Western Australia, pulled plants left on site continued to set seed (Australian Government, 2003), so FloraBase (2013) recommend that seed heads be burnt or otherwise destroyed to prevent seed ripening.

Movement Control

S. glastifolius should not be transported, or sold via garden centres, as Groves et al. (2005) rate it amongst the ten weediest species available for sale in Australia.

The Taranaki Regional Council, New Zealand, where S. glastifolius is not present, has banned it from sale and has an established plan to eradicate any invaders (TERRAIN, 2017). Similarly it is banned in Tasmania, Australia.

Biological Control

There are no known biological control agents for S. glastifolius (Froude, 2002). However, studies carried out by Brierley (1953) on the host range of chrysanthemum stunt virus found 39 plant varieties, including S. glastifolius, which were susceptible and developed recognizable symptoms.

Chemical Control

Atkinson (1980) tested a range of selective and non-selective herbicide sprays on S. glastifolius at the onset of flowering in the Gisborne District. Glyphosate was ineffective against small plants and seedlings, however 2,4-D+picloram and triclopyr all produced good results but not until four months after treatment.

Spraying with a commercially available herbicide, containing 100 g/L 2,4-D and 50 g/L dicamba appeared to be effective in trials (G Rapson, Massey University, Palmerston North, personal observation).

In Australia, FloraBase (2013) recommend applying a herbicide in autumn (before flowering) that contains clopyralid. Clopyralid (3,6-dichloro-2-pyridinecarboxylic acid) is a selective herbicide used for control of broadleaf weeds, especially thistles and clovers.

Monitoring and Surveillance (incl. remote sensing)

New invasion points could readily be detected by remote sensing, including drone technology, during the flowering season, as the flower colour and the density of the flowering heads makes the species easily identifiable. Surveillance by vehicle is also achievable where land is reasonably flat and road networks are sufficiently dense, as plants can be identified up to 0.5 km away.

Very little is known about the growth, ecophysiology, environmental preferences or demography of S. glastifolius.

10/11/17 Original text by:

Gillian l Rapson, Institute of Agriculture and Environment, Massey University, Palmerston North, New Zealand