Rumex obtusifolius
(broad-leaved dock)

R. obtusifolius can be an invasive species on account of its prolific production of seeds, which can remain viable for a long time, its capability for vegetative reproduction, and adaptability to different environments. Mature plants are capable of withstanding unfavourable climatic conditions, such as severe cold and drought, thanks to the deep spreading root. It is primarily a weed of grassland, and in this habitat it benefits from agricultural practices which lead to disturbance or eutrophication. It reduces production from grassland and lowers the feeding value of the sward to animals.

Broadleaved
Herbaceous
Perennial
Seed propagated
Vegetatively propagated

R. obtusifolius is native to Eurasia but has been introduced to all continents. It is distributed mainly in temperate regions but can also be found inside the Arctic Circle and at high altitude in equatorial areas. Hultén (1950) describes it as a circumpolar plant and gives its northernmost limits as 68°N latitude.

In the UK, R. obtusifolius is specified as one of the five injurious weeds in the 1959 Weeds Act which require prevention from seeding, and its control can be enforced on an occupier of land (DEFRA, 2002). In the USA it is listed as an invasive weed (USDA-NRCS, 2002).

R. obtusifolius is present in cool temperate permanent grasslands throughout the world (Holm et al., 1977). On agricultural land it commonly grows in meadows and pastures, abandoned fields, field borders, hedgerows and orchards. It also occurs as a ruderal on roadsides, ditchbanks, and along riversides and streams, and can be found in woodland margins and forest clearings.

R. obtusifolius is not normally a weed of crops. However, Pino et al. (1998) report that in recent decades the species has seriously invaded irrigated lucerne (= alfalfa) (Medicago sativa) crops in Spain. It was also noted in lucerne crops in Tasmania (TAD, 1974) and New Zealand (Askarian et al., 1993).

Genetics

The chromosome number is 2n = 40. Hybridization between R. obtusifolius and other members of the subgenus Rumex occurs frequently (Cavers and Harper, 1964). The commonest hybrid is R. crispus x R. obtusifolius, described as R. acutus by Linnaeus and also as R. pratensis Mert & Koch. It can be recognized by intermediate leaves and tepals. This is the most fertile Rumex hybrid, and produces some viable seeds. Other hybrids, with R. longifolius, R. confertus, and R. patientia, are virtually sterile.

Physiology and Phenology

Germination of R. obtusifolius seeds can occur in any month of the year (Cavers and Harper, 1964), although it is concentrated in March-April and in August-October. The species shows very high sensitivity to light; germination of its seeds is controlled by phytochrome (Taylorson and Hendricks, 1972) and is more rapid in the presence of light (Milberg, 1997). Daylight or near-red light stimulates germination, whereas far-red light, increased by the presence of a leaf canopy, inhibits it (Takaki et al., 1981). Seeds can also germinate in the absence of light after an exposure to alternating temperatures (van Assche and Vanlerberghe, 1989); this accounts for the spring flush of germination when the temperature fluctuations in the soil are very large. Constant temperatures induce secondary dormancy (Roberts and Totterdell, 1981). Benvenuti et al. (2001) report that thermal optima for germination are between 20 and 25°C in light or in darkness. In their study seedlings did not emerge when seeds were buried at a depth of more than 8 cm. Seeds exhibit great within- and between-plant polymorphism, which enables germination under various environmental conditions (Honek and Martinkovà, 2002). Unlike seeds buried in the autumn in the soil that pass through annual dormancy/non-dormancy cycles, the seeds continuously shed from dry standing shoots of R. obtusifolius become germinable all year round, after a short period of burial.

In the seedling year most plants remain at the rosette stage and a long stout rootstock is produced, but flowering and seed set do not normally take place until the second year, unless competition from other plants is weak. A dense sward limits the juvenile growth and establishment of R. obtusifolius in grassland (Jeangros and Nösberger, 1990). The establishment of young plants is favoured by frequent mowing (Hongo, 1989). Adult R. obtusifolius plants in the close vicinity of the seedlings exert a negative influence on their survival (Makuchi and Kanda, 1980).

The plant is a hemicryptophyte and after overwintering regrowth from the rosette usually takes place in late February-March. New shoots are produced from buds on the underground stem. Inflorescences are first formed in early May and may be initiated until the first severe frost (Cavers and Harper, 1964). Flowering and seed set can occur twice in one growing season, first flowers appearing in May-June and the second, after cutting and regrowth of shoots, in August-September. Individual plants may live for at least 5 years. Inflorescences are produced every year under normal conditions and plants do not usually die after producing seed.

Reproductive Biology

Reproduction is amphimictic. Flowers have no nectar and are mainly wind-pollinated. Most individuals are highly self-fertile, although there is variation in this respect between plants.

Propagation is mainly by seeds, produced in very large numbers, varying from less than 100 to 60,000 per plant in a single year. Fruits can be blown for a considerable distance by wind, be carried floating on water, and be dispersed via the digestive system of birds and animals (Cavers and Harper, 1964). Spines on the fruit assist in long-distance distribution by animals, attached to fur or feathers, and by humans.

Seeds may remain viable for a long period when buried deeply in the soil; 83% germinated after burial for 21 years (Toole and Brown, 1946).

In a study carried out by Pino et al. (1995), R. obtusifolius plants showed a clonal growth system in which the roots branch after the first flowering and produce adventitious roots, which accumulate reserves over the next 2 years and separate from the original taproot in the fifth year. This 'phalanx' type of growth allows the plant to reproduce in areas where seedling establishment is inhibited by competition for space. It also has the capacity of vegetative propagation from fragments of root or underground stem split up by cultivation. This is a less important method of reproduction than through seeds; it can, however, significantly facilitate the spread of the species locally (Brock, 1972; Pino et al., 1995).

Environmental Requirements

Growth and development are promoted by climatic conditions typical for temperate regions. Cavers and Harper (1964) state that for the UK there does not seem to be any climatic limitation of the species. It is present inside the Arctic Circle in Northern Scandinavia, can be found at various altitudes and was recorded at 3200 m in Mexico.

R. obtusifolius can grow on a wide range of soils, except peat (Cavers and Harper, 1964). Ellenberg (1974) classifies it as indifferent to pH but its growth is inhibited on the most acid soils. It is a typically nitrophilous plant (Melzer et al., 1984) and shows high abundance on soils with high potassium concentrations (Humphreys et al., 1999). It is favoured by humid conditions, but mature plants with a well-established root system are capable of withstanding severe drought periods.

Associations

No mycorrhizal root symbionts have been recorded.

Cavers and Harper (1964) give a list of parasites and predators of R. obtusifolius, which includes 32 insect species and 12 fungi species. Shim et al. (2001) record 85 insect species associated with R. obtusifolius, a naturalized plant in Korea. In the UK the most studied invertebrate herbivore species is a chrysomelid beetle Gastrophysa viridula, suggested as a potential agent for biological control. In Japan, a similar capacity was shown for G. atrocyanea (Miyazaki and Naito, 1973). In natural conditions, grazing by herbivores does not result in the total destruction of R. obtusifolius plants. Of plant pathogens, a rust fungus Uromyces rumicis, specific to the genus, and the honey fungus Armillaria mellea, can cause particular damage to the weed.

The species which can be most readily confused with R. obtusifolius is R. crispus, especially as it occurs in similar habitats. Lamina of R. crispus leaves are narrower and have strongly wavy margins. The taproot in R. crispus is usually smaller and less branched in the upper part. Tepals of the flowers are heart-shaped with entire margins, whereas in R. obtusifolius they are triangular and have toothed margins. Young R. obtusifolius plants may be confused with R. conglomeratus on the account of morphological similarity of tepals with indistinct teeth and leaves without a pronounced heart-shaped base.

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.

Cultural Control

In Japan, Sakanoue et al. (1995) used mixed grazing by goats and cows on a permanent pasture, and after 3 years reported effective control of R. obtusifolius by goats through a process of defoliation, suppression of propagation, and population decrease. SAC (1986) recommend, as a preventive measure, maintaining a dense, well-managed grassland sward, which can minimise infestation since seedlings of R. obtusifolius are poor competitors and seed germination is inhibited in the presence of a dense leaf canopy. The choice of high-tillering and persistent grasses further reduces competition from R. obtusifolius. Moreover, excessive treading and poaching in pastures should be avoided.

In a study carried out in Austria, Poetsch and Krautzer (2002) obtained a significant reduction of seed germination within three weeks of composting of farmyard manure, and they state that it can therefore be seen as an efficient tool to interrupt the R. obtusifolius seed cycle on grassland farms. Storage in silage significantly reduces the viability of mature seeds (Masuda et al., 1984). Humphreys et al. (1997) report that after being subjected to 100 days in silage, seeds were non-viable. Humphreys et al. (1999) discuss the possibility of limiting the abundance of R. obtusifolius in grassland by maintaining moderate soil K concentration. Balanced fertilizer use, especially N and K, may prevent its spread (SAC, 1986).

Mechanical Control

Hand-pulling is not normally practicable and usually does not result in the complete removal of the root (SAC, 1986). Plants which have been pulled out should be burned to prevent seed dispersal. Dierauer and Thomas (1994) state that the only effective method is the removal of the entire root or cutting the root to at least 10 cm below ground. In grassland, the cutting frequency of the sward affects abundance of the weed. Courtney (1985), Niggli et al. (1993) and Hopkins and Johnson (2002) showed that under frequent defoliation (every 3-4 weeks) dock plants were less competitive and their herbage production was considerably lower than when cut every 6-7 weeks. The height of cut had no consistent effect on R. obtusifolius yields. Cavers and Harper (1964) suggest that a series of rotary cultivations may be effective for the elimination of the weed, however, the timing of land management practices has to be taken into account, as cultivation in late summer may promote the spread of the weed by shortening the length of time between germination and seed production (Weaver and Cavers, 1979). Ploughing followed by fallowing and repeated cultivation during spring and early summer is recommended by the SAC (1986) to exhaust the root reserves and provide control of young seedlings.

Chemical Control

Asulam is the most widely used herbicide for control of mature plants in grassland (Brock, 1972; Oswald and Haggar, 1976; SAC, 1986). Other recommended compounds include thifensulfuron, fluroxypyr, triclopyr, dicamba and mecoprop. Effective application of herbicides requires an adequate leaf area present. Spraying can be carried out in May before harvesting or in the following regrowth. For heavy infestations a follow-up spraying may be necessary 8 to 12 months later (SAC, 1986). Seedlings can be controlled by MCPA, MCPB, 2,4-D or mecoprop and are most readily killed at the two- to three-leaf stage. When the level of infestation is small and the plants are scattered, selective herbicide treatment with rope wick applicators can be used (SAC, 1986).

Biological Control

Gastrophysa viridula has been suggested as a possible agent of biological control (Speight and Whittaker, 1987). Although heavy grazing by this beetle reduces leaf area and overall biomass (Bentley and Whittaker, 1979) and affects the number and weight of seeds (Bentley et al., 1980), it is unlikely that G. viridula, acting alone, could be an effective control agent for the weed (Speight and Whittaker, 1987). In Tasmania the clearwing dock moth Chamaesphecia doryliphormis was introduced in 1997 at 2 sites to provide control of R. obtusifolius. However, there is no further information on the establishment and efficacy of this organism (TAS, 2003).

Studies involving fungal pathogens have not developed control methods, though the possibilities have been discussed for Uromyces rumicis (Schubiger et al., 1986) and Ramullaria rubella (Huber-Meinicke et al., 1989). Hughes et al. (1996) investigated the effects of infection of Armillaria mellea and A. ostoyae, which attack the root system, and reported that the taproots were extensively rotted. Moreover, there are complex interactions between the natural enemies. Hatcher and Paul (2000) showed that grazing by G. viridula reduces natural infection of R. obtusifolius by fungal pathogens, which attack leaves.

Integrated Control

A combination of different control practices and chemical techniques can be used to reduce infestation of mature plants. Repeated discing or rotary cultivation breaks up the taproots and the suppressed growth from root fragments allows more effective control by herbicides (SAC, 1986). An integrated programme of control involving single or repeated spraying with asulam or glyphosate, discing and sowing of grass-legume mixtures carried out in Poland is described by Biala (1998a, b) for heavily infested grassland.