Shrubs climbing. Branchlets terete, usually glabrous; prickles paired below leaves, sometimes sparsely scattered, curved, to 6 mm, stout, flat, gradually tapering to broad base. Leaves including petiole 5–10 cm; stipules pectinate, mostly adnate to petiole, margin glandular-pubescent or not; rachis and petiole pubescent or glabrous, glandular-pubescent, shortly prickly; leaflets (3–)5–9, obovate, oblong, or ovate, 1–5 × 0.8–2.8 cm, abaxially pubescent, adaxially glabrous, base rounded or cuneate, margin simply serrate, apex acute or rounded-obtuse. Flowers numerous in corymb, 1.5–4 cm in diam.; pedicel 1.5–2.5 cm, puberulous, glabrous, or glandular-pubescent, margin sometimes pectinate; bracts at base of pedicel, small. Hypanthium subglobose, glabrous. Sepals 5, deciduous, lanceolate, abaxially glabrous, adaxially pubescent, margin entire or with 2 linear lobes at middle. Petals 5, semi-double or double, white, pinkish, or pink (in some cultivated plants), fragrant, obovate, base cuneate, apex emarginate. Styles connate in column, exserted, slightly longer than stamens, glabrous. Hip red-brown or purple-brown, subglobose, 6–8 mm in diam., glabrous, shiny. 2n = 14*, 21.

Two varieties are recognized here. Plants of this species from Taiwan are usually called var. formosana Cardot (Notul. Syst. (Paris) 3: 263. 1916), which is characterized by small leaflets, 1–3 cm × 0.8–1.5 cm, but this taxon seems to fall within the overall range of variation for var. multiflora. Two other varieties are cultivated in China, but do not occur spontaneously: var. alboplena T. T. Yü & T. C. Ku (Bull. Bot. Res., Harbin 1(4): 12. 1981), which has white, double flowers, and var. carnea Thory (in Redouté, Roses 2: 67. 1821), which has pink, double flowers.

Rosa multiflora may be told from R. setigera, which it resembles, by a more trailing or arching habit, mostly 7 or 9 leaflets, 2-4 cm long, abundant, mostly white flowers in a pyramidal inflorescence, a glabrous style, and smaller fruit (Fernald 1950).

Thickets, scrub, slopes, river sides; 300--2000 m. Anhui, Fujian, S Gansu, Guangdong, Guangxi, Guizhou, S Hebei, Henan, Hunan, Jiangsu, Jiangxi, S Shaanxi, Shandong, Taiwan, Zhejiang [Japan, Korea].

Rosa multiflora (Multiflora Rose) introduced
(beetles feed on pollen; information is limited; observations are from MacRae)

Beetles
Buprestidae: Acmaeodera ornata fp (McR), Anthaxia flavimana fp (McR)

Barcode of Life Data Systems (BOLDS) Stats
Public Records: 4
Species: 16
Species With Barcodes: 1

Canada

Rounded National Status Rank: NNA - Not Applicable

United States

Rounded National Status Rank: NNA - Not Applicable

Rounded Global Status Rank: GNR - Not Yet Ranked

Preserve Selection and Design Considerations: Active control of multiflora rose is necessary mainly on agricultural land when it threatens to dominate pastures. It may also require management on preserve lands if found in old recovering pastures, as it can crowd out desirable grasses and other species.

Management Requirements: Mechanical Control: Repeated mowing will control the spread of multiflora rose, particularly where the grass cover is dense (Scott 1965, Fawcett 1980). Fawcett (1980) stated that mowing several times a year would prevent multiflora rose seedlings from becoming established. At the Woodborne Santurary in Pennsylvania, annual mowing in July helped control the spread of multiflora rose, but did not eradicate it (Stone 1982). Mowing can be difficult due to terrain, when the hedges become established in wooded and brushy pastures. It is also difficult, if not impossible, to mow when the individual clumps reach their mature size, which may exceed 10 ft. in height by 20 ft. in diameter (Doudrick 1987).

Hand cutting of established clumps is difficult and time consuming. Fawcett (1980) recommended use of a bulldozer to knock down large rose clumps but cautioned that further control would be necessary due to resprouting and because seeds will be spread and germinate readily on the disturbed soil. At Woodborne, a large hedge cutter was used to top cut ten foot high rose clumps. Following this, annual mowing has prevented the re-establishment of large clumps and kept the field open (Stone 1982, Davison 1987).

Burning: Burning has not, apparently, been tried for multiflora rose. However, it has been tested in southeastern Texas as a management practice for Mccartney rose (Rosa bracteata), another exotic pasture species in the southern U.S. Gordon and Scifres (1977) tested head fires at 2 to 3 month intervals starting in February, 1975. Fire intensity and fuel components varied seasonally; however, regardless of the date of the burning, topkill of Mccartney rose was greater than 90%. Regrowth was initiated within two weeks after burns, again, regardless of the date of the burning. The average cane elongation was about 4 cm per month and canopy cover replacement averaged 10 to 15% per month following burning. Burning in winter effectively reduced the rose canopy for short-term gains in brush control, and allowed native grasses to take advantage of the entire spring growing period. There were higher herbage yields following winter burns than other seasons. Scifres (1982) believes that multiflora rose response to burning would be similar.

Prescribed burning in combination with herbicides has also been evaluated for Mccartney rose in southeast Texas. Scifres (1975) found that mechanical methods such as raking and stacking were effective for initial removal of mature, dense and ungrazed stands of the rose, allowing access for subsequent treatment. The most effective herbicide treatment was 2,4,5-T plus picloram at 2 lb. per 100 gallons of water and a surfactant (0.5% V/V) applied to thoroughly cover the plant. The best time of treatment was in the fall, when Mccartney rose is actively growing, and resulted in 90% topkill rates. A prescribed burn 18 months later resulted in a high degree of control. This system (mechanical-chemical-burning) is most efficient when an adequate period of time for canopy replacement separates each phase, allowing for complete action of the herbicide. Prescribed burning removes the debris that remained after spraying and should reduce live Mccartney rose top growth by 75%. Periodic burning or respraying is probably necessary to prevent re-invasion of the rose (Scifres 1975).

Biological Control: The European Rose chalicid, Megastigmus aculeatus Swederus (Hymenoptera:Torymida), and rose rossette disease are potential biological control agents for multiflora rose.

M. aculeatus is a phytophagous wasp. The life cycle and distribution in North America has been summarized by Milliron (1949) and Balduf (1959). The adults are minute, weak flyers of limited lifespan. In May and June the long terebras of the female ovipositor pierces the still soft achene and deposits one egg in the soft, pulpy seed. The larvae subsequently develop during June and July, and consume the entire contents of the seed. After full growth in mid to late summer, the larvae undergo a long diapause and overwinter inside the now seedless achene. Pupation occurs in late April to June, and the adult emerges from the rose hip in early summer to renew the cycle. Populations are heavily female in number, suggesting that the majority of reproduction is parthenogenetic (Milliron 1949, Balduf 1959).

It is important to note that M. aculeatus adults are limited fliers, and do not appear to disseminate even locally through their own powers of flight (Balduf 1959). Their spread is dependent upon the use of rose seed, which explains the presence of these insects in nurseries on the East coast, where imported rose seed was used to start root stocks. Subsequent plantings, however, were done vegetatively, far from the nurseries where the plants were grown. It is possible that some of the large-scale plantings of multiflora rose throughout the Midwest are isolated from their chalicid limiting agent (Scott 19865). If true, this suggests that local reintroductions of M. aculeatus could be an effective control method for multiflora rose.

The rose rosette disease is another potential biocontrol agent for R. multiflora. Characteristic symptoms of the disease include abnormal floral development, a "witches broom" effect, and reddening of leaves and shoots (Doudrick et al. 1986). It was originally reported on wild native roses in the northwestern United States and Canada (Thomas and Scott 1953), and first showed up on multiflora rose at a Nebraska nursery in 1964 (Doudrick 1987). By the 1980's, rose rosette was widespread on multiflora rose in Kansas and Missouri (Crowe 1963), and the rose industry became concerned about the spread of the disease to ornamental roses. It is apparently spreading eastward and was first reported east of the Mississippi in southern Indiana and northern Kentucky in 1987 (Hindal et al. 1987). The disease is lethal to all roses, and Doudrick (1987) and Hindal (1987) reported the occurrence of entire fields in Missouri dominated by multiflora rose where 80-90% of the plants were dead or dying. However, the causal agent of the disease is unknown, and it is considered unsafe for use in a control management program for multiflora rose because of the potential threat to ornamental roses. Doudrick (1987) believed that the disease may have reached equilibrium status in Missouri, and that multiflora rose may begin to "bounce back" (i.e., most of the non-resistant genomes of R. multiflora have been attacked, leaving the more resistant ones). The natural spread of the disease may eliminate the need for active control of multiflora rose in some areas.

Chemical control: Plant growth regulators have been used to control multiflora rose in southwestern Virginia where it has been used as a safety barrier along highways. Of the four regulators tested in Spring 1977, chlorflurenol, maleic hydrazine, and MBR- 18337 effectively prevented fruit set and subsequent spread. The fourth regulator, gyloxime, did not give adequate control although it caused some fruit abscission after fruit set (Hipkins et al. 1980).

Various herbicides have been tested and found effective for control of multiflora rose. It is important to note that multiflora rose has the typical regenerative power of members of the rose family (Scott 1965), and control programs must be monitored and followed up if necessary by repeated herbicide application or used in conjunction with other control methods such as mowing or burning.

Glyphosate is effective against multiflora rose in a 1-2% V/V solution (Ahrens 1977, Lynn et al. 1979, Barbour and Meade 1980, Albaugh et al. 1977, Sherrick and Holt 1977, Fawcett et al. 1977). Although Reed and Fitzgerald (1979) reported glyphosate to be relatively ineffective, giving 25-75% stem kill over one season after a spring application, they did not follow-up their results to check for residual control the following year. Lynn et al. (1979) reported that a spring glyphosate treatment on R. multiflora showed increasing control over the growing season to complete control by the following spring. Treatments in the fall showed no results until the following spring, when effective control was realized (Lynn et al. 1979). Ahrens (1977) reported almost complete control of multiflora rose by the end of the second growing season after a late June application of either 1.5 or 3.0 lb/100 gal glyphosate, and noted that grasses growing underneath the roses were unaffected indicating that the spray on the rose overstory did not penetrate to the ground. Albaugh et al. (1977) found that the rate of application of glyphosate could be reduced to a 0.5% V/V solution for effective control with the addition of a surfactant.

2,4,5-T, 2,4-D, and picloram also give excellent control of multiflora rose (Sherrick and Holt 1977, Fawcett et al. 1977, Reed and Fitzgerald 1979). Sherrick and Holt (1977) reported excellent control with 2,4,5-T in a .5-1% V/V solution, and 2,4,5-T plus picloram, 2,4-D plus picloram, or picloram alone were also effective (all as foliar sprays). Ahrens (1977) found 2,4,5-T to be most effective when applied in late april as a dormant basal spray with 2,4-D in fuel oil at 7.5 + 7.5 lb/100 gal oil or alone as a foliar spray at 6 lb/100 gal water. Picloram was found to be relatively ineffective as a soil application. Reed and Fitzgerald (1979) also found erratic results using picloram in pellet form (soil application), with stem kills ranging from 25-100% over one growing season (they did not look for the effects of residual control the following spring, however). Barbour and Meade (1980) reported picloram pellets to be effective, studied over a three- year period, at 2,4, or 5 lb/A.

Other foliar sprays found to be effective against multiflora rose include dicamba (Sherrick and Holt 1977, Fawcett et al. 1979), triclopyr (Sherrick and Holt 1977, Reed and Fitzgerald 1970) and fosamine (Kmetz 1978, Ahrens 1979). Fosamine controls only woody species and is non-volatile, and may be suitable in situations where there is concern to protect herbaceous species (Fawcett 1982). Pelleted and granular treatments found adequate include tebuthiuron (Lynn et al. 1978, Link et al. 1981) while dicamba was not found adequate (Sherrick and Holt 1977, Fawcett et al. 1977, Ahrens 1977, Barbour and Meade 1980).

Management Programs: Multiflora rose has been declared a noxious weed in many states, including Kansas, Iowa, Missouri, Ohio, Pennsylvania, and West Virgina. It is mainly a threat to agricultural land, but has been reported to be a concern on at least two TNC preserves: the Spinn Prairie in Indiana and the Eldora Nature Preserve in New Jersey.

On the Spinn Prairie it occurs in small patches and monitoring may be necessary to determine if active control is necessary (Heitlinger 1987, McGrath 1987). At the Eldora Nature Preserve it is reportedly taking over old fields and there is concern about loss of habitat for some native moth species that feed on grasses in these areas (Davison 1987). No monitoring or management of multiflora rose has taken place at Eldora, but active control measures are considered necessary (Davison 1987).

Contact: Stewardship Director, The Nature Conservancy, Pennsylvania Field Office, 1218 Chestnut St., Suite 807, Philadelphia, PA 19107. (215) 925-1065.

Denny McGrath, Assistant Director, The Nature Conservancy, Indiana Field Office, 4200 N. Michigan Road, Indianapolis, IN 46208. (317) 923-7547.

A number of states where multiflora rose is a problem on agricultural land have cost share eradication programs whereby landowners can be reimbursed for a portion of the costs to control the plant on their property. These programs may also be available for preserve areas.

Contact: Iowa. Secretary of Agriculture, Iowa Dept. of Ag. and Land Stewardship, Wallace State Office Bldg, Des Moines, IA 50319.

Ohio. Larry Vance (614) 265-6610. Larry Summers (614) 265-6684. Ohio DNR. Div. of Soil & water Conservation, Fountain Square Bldg. E-2, Columbus, OH 43224.

Monitoring Programs: Heitlinger (1987) suggested monitoring of multiflora rose through the use of line intercept transects at the Spinn Prairie in Indiana to track its density and contraction/expansion. Contact: Denny McGrath, Ass't. Director, Indiana Field Office, The Nature Conservancy, 4200 N. Michigan Road, Indianapolis, IN 46208 (317) 923-7547.

Management Research Programs: Research is currently being conducted at West Virginia University on rose rosette as a control for multiflora rose. Contact:

Dr. Dale Hindal, Division of Plant & Soil Sciences, Dept. of Plant Pathology and Agricultural Microbiology, 401 Brooks Hall, West Virginia University, Morgantown, WV 26506. (304) 293-3911.

Dr. James Amrine, Dept. of Entomolgy, West Virginia University, Morgantown, WV 26506. (304) 293-6023.

Management Research Needs: Further research is needed in the area of biological control for multiflora rose. Both the phytophagous wasp Megastimus aculeatus and the Rose Rosette disease are potential biological control agents (see Management Procedures), but also represent a potential threat to ornamental roses.

In the case of M. aculeatus, the degree of host specificity is not fully understood. Milliron (1949) recognized two varieties of the wasp: a "light form" (M. aculeatus aculeatus) and a "dark form" (M. aculeatus nigroflavus). Milliron believed the dark form to be host specific on multiflora rose. However, Balduf (1959) recovered M. aculeatus nigroflavus from Rosa eglanteria and R. virginiana, both native roses. No further research has been conducted on M. aculeatus host specificity. A more promising control agent is the rose rosette disease. However, research to determine the causal agent of the disease has met with little success. Transmission of the disease is accomplished by an eriophyid mite, Phyllocoptes fructiphilus (Amrine et al. 1987). Symptoms of rose rosette, such as the witches broom and reddening of leaves, suggest a mycoplasma- like organism (MLO) as the causal agent, but the mite mouthpart (a sucking tube) is too small to suck up an MLO and also does not penetrate the phloem where an MLO would be found (Doudrick 1987). Other characteristics suggest a viral causal agent, but Doudrick et al. (1987) were unable to find anything resembling viral particles associated with diseased plants. Until more is known about the cause of rose rosette, it probably will not be employed in management programs for control of multiflora rose due to the threat to ornamental roses.

Other questions that may aid management of multiflora rose if carefully researched include the following. What are the germination requirements of multiflora rose and under what conditions are seeds least likely to germinate? How persistent is the rose in recovering grasslands that are no longer grazed? What are the effects of fire on seed viability and vegetative reproduction? How effective is fire in conjunction with herbicides or other control methods?

The preference is partial sun, mesic conditions, and fertile loamy soil. This rose species is very aggressive and hard to get rid of once it becomes established. At an open sunny site, this plant will develop into a shrub that produces numerous flowers, while at a shadier site with adjacent vegetation it becomes a climbing vine. The leaves are vulnerable to various kinds of foliar disease, particularly where there is an abundance of moisture and inadequate circulation of air.