Juncus ensifolius
(swordleaf rush)

Juncus ensifolius is a mostly pioneering or ruderal species of rush that readily establishes in disturbed wet soils, often from buried seeds.  Within its native range in western North America it is widespread and infrequent to common, typically a minor, secondary or at most co-dominant species in natural wetlands. Limited reports suggest that it behaves similarly in east Asia (Tachibana et al., 2001), where it is also native. Although it is recognized as having “potential for weediness” (Marr and Trull, 2002), it is not generally viewed as an aggressive invader within its native range and is widely used in wetland restoration and as a landscape plant. There are no published studies documenting its rate of growth, spread or dispersal.

Scattered occurrences in eastern North America are considered by some to be disjunct, rare, native populations worthy of tracking or protection, and by others to be introduced. It appears most likely that these populations result from combinations of accidental and deliberate human introductions and dispersal by animals (e.g. by migrating waterfowl).

In Europe, Australia, New Zealand and Hawaii, J. ensifolius has apparently become naturalized in the 20^th Century following deliberate (gardening) and accidental (as a contaminant in seed mixes and peat) introductions.  There appear to be no reports of it spreading outside its native range in Asia, but it has been introduced in constructed experimental wetlands in Israel. Occurrences in the introduced range are being monitored (both formally by government agencies, and informally by interested botanists), but there are no records of targeted control action. There is no evidence that J. ensifolius causes harm to human health, but there is some ambiguity as to whether it causes, or has the potential to cause, economic or environmental harm. It is recorded as a threat to several endangered plant species in Hawaii, but even in that case the situation is not entirely clear.

Juncus ensifolius occurs as a widespread, frequent to common native wetland plant from near sea level to subalpine elevations throughout western North America and along the Pacific Rim from California to Alaska, extending through the Aleutian Islands and the Kurile Islands of the Russian Far east to Hokkaido Island, Japan.  In western North America it can be found on both sides of the Rocky Mountains, extending eastward to the western edge of the Great Plains in Saskatchewan (where it is considered rare), South Dakota to eastern Colorado, and central Mexico.

There are disjunct locations along the remote shoreline of Hudson Bay in far northeastern Ontario and northwest Quebec, and also along the Delaware River in New York.  These occurrences were for many years considered native, and are still considered by some to be worth tracking or protecting (Oldham, 1999; Bouchard et al., 1983; New York Natural Heritage Program, 2015); the species is listed as endangered in New York (New York Natural Heritage Program, 2015). However, although human introduction to remote areas of Ontario and Quebec seems unlikely, most sources now consider the U.S. occurrences east of the Great Plains to be introduced (Brooks and Clemants, 2000; Weldy et al., 2015). It appears most likely that these populations result from combinations of accidental and deliberate human introductions and dispersal by animals (e.g. by migrating waterfowl). Other long-disjunct specimens have more recently been reported in Rhode Island and Vermont (BONAP, 2015; latter not verified with a herbarium specimen). J. ensifolius is widespread along disturbed roadsides and waterways in Wisconsin and has also been found at several locations in Michigan (Marr and Trull, 2002). In the Great Lakes region of the USA, the species has been described as “spreading but tough to eradicate” and ranked as “medium priority”, but did not require “early detection and rapid control” in Northern Wisconsin (NCWMA, 2014) and although it was tracked, it was not considered a “priority invasive” by the Great Lakes Indian Fish & Wildlife Commission (Falck et al., 2006), listed as a regulated or restricted species in Wisconsin (Wisconsin Department of Natural Resources, 2015), or prohibited or restricted in other Great Lakes states.

In Europe, Australia, New Zealand and Hawaii the species has apparently become naturalized in the 20^th Century following deliberate (gardening) and accidental (as a contaminant in seed mixes and peat) introductions. In Hawaii, Daehler (2005) did not rank it as a “disruptive invader” in upper montane environments, but it was later listed in a final rule under the Federal Endangered Species Act as posing a threat to several endangered plant species on Hawaii Island (US Fish and Wildlife Service, 2013). In Austria a 2002 government publication ranked its status as “impermanent” (non-invasive) and “so far without effects” (Essl and Rabitsch, 2002).

There appear to be no reports of spread outside the native range in Asia, but the species has been introduced in constructed experimental wetlands in Israel (Iasur-Kruh et al., 2010).

Risks of further introduction are high because the species is sold worldwide via the internet as an ornamental plant (e.g. Caribbean Garden, 2015; Le Jardin du Pic Vert, 2015) and because many founder populations are located in high traffic areas along roadways and trails.  Growing worldwide interest in wetland restoration and mitigation creates more opportunities for international introductions (e.g. Iasur-Kruh et al., 2010).

J. ensifolius is a facultative wetland species (Washington Native Plant Society, 2015) that also occurs on seasonally wet disturbed soils in upland sites.  Climatically, it is a cool temperate species that occurs marginally in boreal forests (i.e. in the southern boreal and along coastlines) and at high elevations in warmer climates (e.g. montane wetlands in Hawaii; alpine resorts in Australia).  In humid, coastal environments and cooler climates it has a broad ecological niche that encompasses estuaries, rainforest edges, roadsides and subalpine meadows, whereas in hotter, desert environments it is more narrowly restricted to riparian, wetland and sheltered habitats (e.g. adjacent to a waterfall). A characteristic feature of almost all habitats is that the soils are seasonally, but not continuously, inundated. Over most of its range, J. ensifolius is associated with disturbed soils, often proliferating in wet ditches, roadsides, and reservoir draw-down zones, or following events such as volcanic mudslides, flooding or forest fires. Although it does not tolerate shade, J. ensifolius is not strictly an early seral species, and appears to be sensitive to grazing. In riparian meadows of the US Rocky Mountains, for example, it is regarded as an indicator species of later successional stages with high stability, biotic integrity and minimal grazing disturbance (Winward, 2000; Jones, 2005).

There is no evidence that J. ensifolius currently causes significant economic damage to any crop plants, although it apparently occurs as a minor  contaminant in grass seed mixes (Piirainen, 2004) or commercial peat (Kirschner, 2002). It is recorded as a competitor of seven endangered species in Hawaii – see the 'Host Plants/Plants Affected' and 'Threatened Species’ tables for more information.

Genetics

Phylogenetic (cladistic) analysis of the Juncaceae performed using rbcL nucleotide sequences and chloroplast DNA (Drábková et al., 2003, 2004, 2006; Drábková, 2010) has confirmed that J. ensifolius is most closely related to other species of Juncus, subgenus Juncus, section Iridifolii, including J. oxymeris and J. xiphioides (also centred in the western US), and to species of section Ozophyllum (broadly distributed in eastern North America, Europe and the Far East).  J. ensifolius was found to differ from other species in lacking most of the anticodon domain of the trnF gene (Drábková et al., 2004). The chromosome number of J. ensifolius var. ensifolius, tested at disparate locations (Snogerup, 1963; Taylor and Mulligan, 1968; Harriman and Redmond, 1976; Probatova et al., 2004) was consistently 2n = 40. It is unclear whether the chromosome number of var. montanus (which differs from var. ensifolius in having 6 rather than 3 stamens) has been tested. 

Reproductive Biology

Juncus ensifolius reproduces both sexually and asexually (by means of rhizomes).  The flowers are wind-pollinated.  In Juncus species, flowers are bisexual. An individual flower opens only once for a few hours and very rarely for longer than 24 h (Buchenau, 1890, 1892, cited by Michalski and Durka, 2007). Plants grown at a botanical garden in Germany produced 1802 ± 324 (SD) pollen grains per flower and 49.0 ± 3.2 ovules per flower.  The pollen/ovule ratio of 37 ± 5 is unusually low for a wind-pollinated species, and the high rate of ovule production is thought to be advantageous in the open, ruderal and disturbed habitats favoured by this species (Michalski and Durka, 2010).  Many Juncus species appear to be capable of self-fertilization (Michalski and Durka, 2007; Huang et al., 2013), favouring higher energy investment in ovules. J. ensifolius may be one such species, as it does appear to produce prolific seeds, even in founder populations.

Juncus seeds are tiny, and those of J. ensifolius are small compared to other rushes (152,000/g). They are released when seed capsules shatter shortly after maturity (Hurd and Shaw, 1992).  J. ensifolius is common in soil seedbanks, and viable seeds are often found in mature forest vegetation where the species is absent from above-ground vegetation (Clark, 1991; Yearsley, 1993; Harmon and Franklin, 1995; Tu et al., 1998; Duncan, 2003; Reynolds and Cooper, 2011), suggesting that seeds can remain viable for decades to centuries.  Heating field-collected soil samples to 50^oC stimulated germination of buried viable seed compared to unheated soils, but no germination occurred in soils heated to 150^oC (Clark, 1991).  Light and chilling (30 days at 3-5^o C) are recommended for germination in the nursery (Hurd and Shaw, 1992), but other reports suggest that seeds will readily germinate within 2 weeks without presoaking or stratification if sown on the soil surface or in shallow water (Steinfeld, 2001).

J. ensifolius spreads and forms loose to dense turfs or swards by means of slender, horizontal rhizomes. New plants can become established from rhizome fragments deposited in moist soil. Disturbance or damage to the rhizomes stimulates the production of new shoots from dormant or adventitious buds.  In cultivation it is not as strongly rhizomatous as J. balticus, for example, and is usually sold and planted out as a bareroot seedling stock, rather than as a rhizome cutting with buds (Whitacre, 2014).

Physiology and Phenology

Klarich (1977) studied respiration and photosynthetic rates of J. ensifolius in water taken from 3 reaches of the Madison River, Wyoming, USA. The pH of the water was lowest < 7.6) and free carbon dioxide was highest (8 ppm) in the upper reach where J. ensifolius was most abundant (1.6% cover).  Lower in the river (pH 8, free CO₂ = 1 ppm) J. ensifolius was rare to absent. Photosynthetic rates, and the ratio of net photosynthesis to respiration were highest in carbon dioxide-rich waters from the upper reach of the river, while % light saturation levels were highest in the low carbon reach further down the river.  The rates of photosynthesis in J. ensifolius were much lower than those of the dominant macrophytes in the river, probably because they were true aquatic plants while J. ensifolius is only semi-aquatic and was thus ecologically marginal in the flowing river.

In an intertaxa laboratory comparison of 76 species of angiosperms (Wiley and Wilkins, 2006), the relative mean concentration of sulphur found in J. ensifolius (-0.94) was found to be very close to the mean of all species tested (-1.0), slightly lower than other wetland species tested (J. effusus -0.52; Carex nigra 0.39) but higher than most of the important food and forage grasses tested (typically -1.2 to -1.8). Other than these few studies, there does not appear to have been any physiological or experimental work done on J. ensifolius.

Within its native and introduced range, J. ensifolius most often grows in a cool temperate to southern boreal environment with a distinct winter dormant season, the length of which varies by several months depending on elevation, latitude and proximity to the ocean.  In moist, cool environments it is unlikely that there is a summer dormant period, although the wet soils in which it grows may experience a substantial draw-down in moisture levels from mid-summer to autumn. During this period, growth shifts from vegetative growth to seed maturation.  Typically in the Northern Hemisphere, flowering occurs in June and July (see Michalski and Durka, 2010, for other Juncus spp.) and seeds mature in August and September (Steinfeld, 2001), with capsules opening soon after maturation to shed the seed (Hurd and Shaw, 1992). In mild climates, it is likely that recharging of soil moisture in the fall initiates a period of root and possibly shoot growth prior to winter. 

Longevity

J. ensifolius is a perennial with rhizomatous growth and commonly reproduces vegetatively. Its longevity has not been studied, but like most rhizomatous perennials it probably lives for long periods of time (at least many decades).  Seeds appear to have extended longevity (decades to centuries) in the soil (e.g. Clark, 1991; Harmon and Franklin, 1986).

Population Size and Structure

Patches of J. ensifolius described in the literature range from single plants or small clumps to extended patches covering dozens to hundreds of square meters.  These populations probably result from a combination of sexual and vegetative propagation as seeds germinate from the seed bank or recent dispersal and genets then expand in size through rhizomatous growth.

Nutrition

Seedlings in nursery cultivation are fertilized once weekly with all-purpose plant food (15-30-15) (Tilley, 2011). There are no published studies of plant nutrition in the wild.

Associations

In Hokkaido, Japan, J. ensifolius was found to grow with Carex michauxiana subsp. asiatica, Agrostis scabra and J. effusus in a mire that was disturbed by human trampling (Tachibana et al., 2001).

On Amchitka Island in the Aleutians, it was found in shallow pools of water surrounded by Empetrum heath in association with J. triglumis and Eriophorum russeolum [E. chamissonis] (Shacklette et al., 1969).

In British Columbia, Cowan (1945) described a low elevation Juncus-Carex-Isoetes lakeshore mudflat plant community on Vancouver Island that was submerged from October-June and grew prolifically from June-August.  Dominant species were Ranunculus flammula, Carex lenticularis, C. viridula,J. supiniformis and Isoetes maritima.  In the same area, J. ensifolius was also common in a Carex sitchensis [C. aquatilis var. dives]-Oenanthe sarmentosa-Deschampsia cespitosa meadow that resulted from the gradual filling-in of former lakes and ponds. Brett et al. (2001) described a J. ensifolius – Alnus viridis [A. alnobetula] association on a moist floodplain with a fluctuating water table —J. ensifolius and other graminoids formed the understorey. 

At Mt. Rainier, Washington state (Dunwiddie, 1986), J. ensifolius grew in saturated, seasonally inundated soil adjacent to a pond in association with Carex spp. and Eriophorum polystachion [E. angustifolium]. Nearby, at Mt. St. Helens (Moral, 1999), J. ensifolius was the dominant herb in a primary successional wetland association with Salix sitchensis, J. bufonius, J. articulatus and Brachythecium moss on relatively dry pumice with high pH, and as a secondary species in wetter Salix sitchensis – Equisetum arvense and Typha latifolia – J. bufonius associations with lower pH.

In the US Intermountain region (Idaho, Montana, Utah, Wyoming, Colorado) J. ensifolius commonly occurs in sedge-willow associations adjacent to streambanks (Winward, 2000; Jones, 2005; Hough-Snee et al., 2013).  Typical associates include Carex utriculata, C. nebraskensis, C. aquatilis, Salix melanopsis, S. boothii, S. drummondiana,S. geyeriana, S. exigua, Agrostis scabra and Poa pratensis with the relative proportions of these species reflecting the degree of cattle and elk grazing disturbance.  Higher cover of J. ensifolius is considered indicative of low grazing pressure (Winward, 2000).

Outside its natural range, where J. ensifolius is naturalized or adventive, its most common associates are cosmopolitan weeds associated with highly disturbed habitats. Genera and species that have been repeatedly recorded as associates in Europe, Oceania and eastern North America  include Juncus, Plantago, Lotus, Phleum, Agrostis, Rumex, Ranunculus, Trifolium, Equisetum, and Carex (Adema and Mennema, 1976; Koch, 1991; Stolwijk and Zijlstra, 1995; Marr and Trull, 2002; Grøstad, 2003; US Fish and Wildlife Service, 2013; Atlas of Living Australia, 2015). 

Environmental Requirements

A seasonally fluctuating water table including temporary, but not year-round, inundation appears to be the most prominent feature of J. ensifolius habitats.  In humid climates, standing water does not appear to be obligatory – J. ensifolius is considered a facultative rather than an obligate wetland or aquatic species (Washington Native Plant Society, 2015).  Soils can be sandy or clayey, but are never well- or rapidly-drained. J. ensifolius appears to tolerate mildly saline soils (coastal and inland) and a wide range of soil pH.

Climatically its habitats can be described as cool, moist temperate.  In the south temperate to Mediterranean zone it appears to be mostly limited to high elevations (e.g. subalpine forests and parkland), whereas in the north temperate zone it is common near sea level.

Fungi: Uromyces junci-effusi (formerly Nigredo junci-effusi) is a rust in the family Puccinaceae that produces spots on the leaves of Juncus spp. in North America (Garrett, 1919). It is the most widely reported fungus on J. ensifolius (Farr and Rossman, 2015). Phaeosphaeria alpina is an ascomycete fungus that parasitizes the leaves, stems and foliar parts of a wide variety of grasslike plants in Europe and North America.  It was collected on J. ensifolius in an alpine meadow at Revelstoke, British Columbia (Shoemaker and Babcock, 1989). Bipolaris palousensis (also palousense) is a leaf spot or “blotch” fungus collected on J. ensifolius  foliage in Washington (1948) and Oregon (1957) (Sprague, 1961; Manamgoda et al., 2012).  Bipolaris are important pathogens of graminoids worldwide and this is the only Bipolaris reported on Juncus ensifolius; however, it remains a doubtful species as the existing specimen is in poor condition (Alcorn, 1991; Manamgoda et al., 2012). Cercospora juncina is another leaf spot fungus found on species of Juncus (Kasai, 1922), including J. ensifolius (Farr and Rossman, 2015). A species of Claviceps, a floral pathogen known as “ergot” that infects the ovaries of graminoids, has been found on J. ensifolius and several other Juncus spp. in Canada and California, but has not been identified at the species level because only conidia were present (Alderman et al., 2004).

A nematode (Gracilacus elongata [Paratylenchus elongatus]) has been described as parasitic on J. ensifolius in California (Abdel-Rahman and Maggenti 1988).

Although insects commonly feed on Juncus species, especially their seeds (Hurd and Shaw, 1992), J. ensifolius is not identified as a host plant on any insect-host plant database.

Muskrats (Ondatra zibethicus) are known to feed on the roots/rhizomes of J. ensifolius. Waterfowl (e.g. Canada geese – Branta spp.), deer, elk (Cervus canadensis), cattle and horses can cause significant damage by grazing.

Environmental

• Amenity
• Erosion control or dune stabilization
• Landscape improvement
• Revegetation
• Soil conservation
• Wildlife habitat

General

• Botanical garden/zoo
• Laboratory use
• Sociocultural value

Materials

• Fibre

Medicinal, pharmaceutical

• Traditional/folklore

Ornamental

• Propagation material
• Seed trade

J. ensifolius is a low-statured plant that would be difficult to detect remotely, although ditchlines and areas with wet soil and shallow standing water that can be identified from images should be targeted in surveys.  It can be readily identified visually by its combination of flattened leaves and dark, globular seedheads. Similar species can be readily distinguished as described in the ‘Similarities to Other Species’ section.

J. ensifolius is a very distinctive plant because of its dagger or sword-shaped leaves (folded and flattened at the base like those of an iris) and the fact that it almost always bears prominent clusters of densely globular dark, purplish-brown flowers and fruit below or at the tip of the foliage.  Throughout much of its native range and in Europe there are no other plants that combine these two features and it is unlikely to be confused with any other species.  However in the southwestern USA (California and adjacent states) and in east Asia there are several other rushes in the section Iridifolii (Kirschner et al., 1999) with flattened iris-like leaves that could possibly be confused with J. ensifolius. In most of these species the flowerheads are less evidently globular and paler in colour or the inflorescences are more diffuse, radiating outwards well above the leaves. J. xiphioides is a larger plant with spreading, brown inflorescences, whose range overlaps with J. ensifolius in California, Nevada and Arizona. J. oxymeris, which extends north from California to southern British Columbia, is also a larger plant with a more diffuse inflorescence with pointed fruit capsules (in British Columbia, although rare and at its northern limit, it is often confused with the more common J. ensifolius (Lomer, 2011)). J. macrandrus, endemic to California, has greenish-brown to greenish inflorescences that radiate star-like at some distance above the foliage. J. phaeocephalus, a tall rush that ranges from Baja California to southern Washington, and J. polycephalus, found in the southeastern U.S. outside the range of J. ensifolius, bear their fruit in open panicles with long stalks.  J. alatus, found in east Asia, and J. prismatocarpus, widely distributed through east Asia to New Zealand and Australia, also have loose, pale inflorescences. In New Zealand, J. ensifolius may be confused with the native J. prismatocarpus, which also has flattened, multitubular septate leaves (NZPCN, 2015).

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.

There is no available literature on the prevention and control of J. ensifolius invasion. Initial development of management practices, where management is necessary, must therefore be derived indirectly from other sources, including generic practices recommended for other invasive species.

Public Awareness

As online plant nurseries are selling J. ensifolius internationally, an initial step in building public awareness would be to contact growers offering this species for sale online and advise them that it is considered invasive and should not be sold to customers outside of its home range in western North America.  This step is probably most relevant to growers selling it to customers in Hawaii, where it is identified under an order in the Endangered Species Act as threatening endangered Hawaiian plant species (US Fish and Wildlife Service, 2013).

Physical/Mechanical Control

J. ensifolius has been shown to be sensitive to ungulate grazing in the western U.S., so potential control measures could involve repeated mowing or controlled grazing. Goats and sheep have been employed to control other Juncus spp., in New Zealand and Australia (Rolston et al. 1981; MLA, 2007), but unless very carefully regulated, these animals may cause greater environmental damage than they prevent. On wet ground, repeated mowing may not be feasible due to low trafficability and therefore may be only of limited value (Sellers and Farrell, 2015).

Movement Control

Collaborative intervention with nursery associations and native plant societies to reduce or prevent sales of this species outside of its native range are probably the most promising avenue for reducing further spread beyond existing occurrences. 

Biological Control

Although J. ensifolius has a number of natural enemies (see ‘Natural Enemies’ section), there are no reports of their use for biological control.

Chemical Control

There is no information available on the sensitivity of J. ensifolius to herbicides. In a recent trial, chemical formulations containing 2, 4–D amine were the most successful among 5 chemical products tested for control of J. effusus in Florida (Sellers and Farrell, 2015). High water tables and the potential for runoff may limit the use of chemical control agents for this species in sensitive environments.

Monitoring and Surveillance (Incl. Remote Sensing)

No information is available on this subject, but as a low-statured plant J. ensifolius will be more difficult to monitor remotely than larger invasive plants.  Most of the habitats where it grows do not have tree cover and may be suitable for low-level aerial photography.

Ecosystem Restoration

There is no published literature describing ecosystem restoration to address invasion by J. ensifolius.  Such action may have been undertaken in montane wetlands in Hawai’i Volcanoes National Park to restore endangered plants reportedly threatened by this species, but J. ensifolius is not listed by Belfield et al (2011) among the alien species subjected to control actions.

Three important gaps in our knowledge of J. ensifolius are:

(1) Invasiveness: Is it an aggressive species that actively interferes with native species and warrants direct control action or will it remain a minor associate of other wetland plants at most new locations?  In disturbed soils, does it disappear as succession proceeds?  Does it spread beyond ditches into undisturbed habitats? Assessments done to date have confirmed its presence at a growing number of sites but none have yet ranked it as a high priority for control.

In Hawai’i, it would be worthwhile clarifying the mechanisms by which J. ensifolius threatens each of the endangered species with which it was associated by the U.S. Fish and Wildlife Service (2013). All of these are much larger-statured plants than J. ensifolius, and at least a few of them do not appear to grow in cool, montane wetland (wet soil) situations. It is possible that presence of J. ensifolius represents a correlation with disturbed habitats rather than a cause of endangerment for these species.

Brooks and Clemants (2000) suggested that additional taxonomic work was needed to determine whether var. montanus warrants treatment as a separate species. From a plant invasions perspective, it would be of interest to determine whether both varieties exhibit invasive behaviour or whether the six-stamened variety is less weedy than the three-stamened var. ensifolius.

(2) Control measures: Treatments that kill the seed in contaminated peat, soil or hay and the efficacy of mowing or controlled grazing are two examples where more information about control measures would be useful. It may be possible to develop attractive sterile cultivars of this plant for the nursery trade.

(3) Distribution in Asia and other regions of the developing world:  Potential habitats exist near sea level in coastal temperate South America (Patagonia), southern Africa (equivalent climate to California) and temperate East Asia, and in cool, montane environments worldwide.

05/12/2015 Original text by:

Sybille Haeussler, University of Northern British Columbia, British Columbia, Canada