Cogon grass stand. Photo by Kazuo Yamasaki.
The scientific name of cogon grass is I. cylindrica (L.) Beauv. (Poaceae) [9,25,43,64,72,105,173]. Some authorities recognize 5 varieties of cogon grass; according to that treatment, I. c. var. major (Nees) CE Hubb. is the entity found in North America [64,133]. Gabel (within ), does not recognize varieties of cogon grass.
Brazilian satintail and cogon grass are morphologically and genetically very similar, and their hybrids produce fertile offspring [57,133,165]. Hybridization, introgression, and overlapping morphological characters often cause taxonomic confusion between the 2 species, especially in North America. Some systematists consider the 2 species synonymous [25,57]. Hall  suggests that Brazilian satintail be classified as an infrataxon within I. cylindrica. Gabel [9,43] separates the taxa as 2 distinct species based upon continents of origin and morphological, cytological, and genetic attributes. This review treats Brazilian satintail and cogon grass as 2 distinct species.
Because there is little English-language literature currently available on Brazilian satintail, this review provides information mostly on cogon grass. Pertinent information on Brazilian satintail is included whenever possible. Given the taxonomic status of Imperata in North America, information included in this review may apply to both species; however, further research is needed to be certain the 2 taxa respond similarly to fire and control treatments.LIFE FORM:
|Brazilian satintail||Cogon grass|
|Alabama||Class A noxious weed||Class A noxious weed|
|Florida||Noxious weed||Noxious weed|
|Georgia||not listed||Noxious weed|
|Hawaii||not listed||Noxious weed|
|Massachusetts||Noxious weed||not listed|
|Minnesota||Prohibited noxious weed||Prohibited noxious weed|
|North Carolina||Class A noxious weed||Class A noxious weed|
|South Carolina||Plant pest||Plant pest|
|Vermont||Class A noxious weed||Class A noxious weed |
|Virginia||not listed||Highly invasive |
|U.S. Forest Service, Southern Region||Category 1||Category 1 |
Cogon grass is native to Korea, Japan, China, India, and tropical eastern Africa [37,64,105]. It is nonnative and invasive throughout other tropical regions of the world. In North America it occurs along the Gulf Coast from Mexico east to South Carolina [4,72]. In the United States it is most common in Mississippi, coastal Alabama, and Florida . Hitchcock  listed cogon grass as present in Oregon in 1950, although it has not been collected in Oregon for decades. There were 2 known locations of cogon grass introduction in the United States: 1 from Japan to Alabama in 1912, as packing material in a shipment of Unshu orange (Citrus reticulata) trees; and another from the Philippines to Mississippi in 1921, as a possible forage grass [33,44,142,143]. More than 1,000 acres (400 ha) of cogon grass were planted for livestock forage and soil stabilization in Florida the late 1930s and 1940s [32,133].
Grass Manual on the Web provides distributional maps of Brazilian satintail and cogon grass. It is commonly assumed that cogon grass is the more common of the 2 species; however Brazilian satintail or Brazilian satintail × cogon grass hybrid swarms may be misidentified as cogon grass [86,142]. Distributions of Brazilian satintail and Brazilian satintail × cogon grass hybrids may be more extensive in the Southeast than is currently known . The 2 species occur in similar habitats in the Southeast [37,64]. The following lists give biogeographic classifications where Brazilian satintail and cogon grass are known to be present or invasive. These lists may not be exhaustive.Brazilian satintail and cogon grass:
Details of Brazilian satintail communities of South America are also limited. In Peru, Brazilian satintail dominates montane savannas. Copperleaf (Acalypha spp.), muttonwood (Rapanea spp.), speedwell (Veronica spp.), and false-willow (Baccharis spp.) are dominant woody genera; bracken fern (Pteridium aquilinum) often codominates on Brazilian satintail grassland areas. Scott  lists associated plant species on Brazilian satintail-dominated savannas of eastern Peru.
Cogon grass occurs in southeastern pine and oak-pine communities that experience frequent fire (see Cogon grass in North America for further details). It is most common in the ground layer of mesic longleaf pine (Pinus palustris) savannas . Oaks including blackjack oak (Q. marilandica), turkey oak (Q. laevis), and southern red oak (Q. falcata) are often frequent in the overstory. Common shrub associates of cogon grass include persimmon (Diospyros virginiana), black highbush blueberry (Vaccinium fuscatum), dwarf huckleberry (Gaylussacia dumosa), and bitter gallberry (Ilex glabra). Common groundlayer associates include big bluestem (Andropogon gerardii), paintbrush (A. tenerius), Beyrich threeawn (Aristida beyrichiana), golden colicroot (Aletris aurea), and roundleaf thoroughroot (Eupatorium rotundifolium) . Cogon grass occurs in south Florida slash pine/firegrass (Pinus elliottii var. densa/Andropogon cabanisii) savannas of Everglades National Park. Brazilian peppertree (Schinus terebinthifolius) and silkreed (Neyraudia reynaudiana) are other nonnative invasive associates inventoried on Everglades savannas. Cogon grass dominates some grassland sites in the Everglades .
In Puerto Rico, cogon grass occurs in early seral bracatinga (Mimosa scabrella) forests. Leandra (Leandra australis) and cappel (Palicourea spp.) also occur in the overstory. Groundlayer associates include hemlock-rosette grass (Dichanthelium sabulorum) and flatsedge (Cyperus hermaphroditus) .In Southeast Asia, cogon grass dominates extensive grassland areas. It also dominates the ground layers of Khasia pine (Pinus kesiya), Chir pine (Pinus roxburghii), and other pine forests that experience frequent surface fires [37,97]. Sticky snakeroot (Eupatorium adenophorum), which is native to Southeast Asia, is commonly associated with cogon grass in both the Old and New Worlds . Jack-in-the-bush (Eupatorium odoratum) and broadleaf carpet grass (Axonopus compressus) are other common associates on Asian grasslands subject to frequent fire that also co-occur with cogon grass in the southeastern United States [80,82,103,144,172]. Cogon grasslands in Asia can become increasingly diverse with time since last fire . Eussen  and Tanimoto  provide further descriptions of cogon grassland associations of Southeast Asia.
Brazilian satintail is lesser known, and hence more poorly described, in the United States compared to cogon grass. It has slender, erect culms from 14 to 29 inches (36-74 cm) tall. Leaves are mostly basal and about 5 to 13 mm wide. The inflorescence is a 3- to 8-inch (7-20 cm), terminal panicle ; the fruit is a caryopsis. Brazilian satintail is rhizomatous, with a mat-like growth form [43,163].
Cogon grass grows to 3 feet (1 m) in height [25,105]. Leaves are mostly basal, growing from the rhizomes. Basal leaves are 0.4 to 0.8 inch (1-2 cm) wide . A few small upper leaves occur on the pedestal [25,105]. The leaves have a characteristic white midrib that is set off-center. Being high in silica , cogon grass leaves are coarse in texture [26,160]. The inflorescence is a dense, 4- to 8-inch (10-20 cm) panicle of paired spikelets. Spikelets are unawned with long (~12 mm), silky hairs [25,105]. The seeds are small (1-1.3 mm long) [74,75].
The root system is fibrous. Cogon grass rhizomes are "tough and scaly," with short internodes forming a dense underground mat. Cogon grass rhizomes develop in 2 stages: primary seedling rhizomes, and secondary rhizomes that sprout from seedling rhizomes . Rhizome and root depths vary with substrate. In central Florida, Gaffney  found cogon grass rhizomes were restricted to the top 4 to 6 inches (10-15 cm) of soil on a phosphate mine site, but grew down to 30 inches (80 cm) below ground on a clay settling pond site . In Southeast Asia, rhizomes typically occur 4 to 20 inches (10-40 cm) below ground and form dense, extensive layers. Some rhizomes grow as deep as 3 feet (1 m) [8,100]. Cogon grass's growth habit is loose to clumped, compacted aerial stems arising from the dense rhizome mat [35,43]. Dense stands may form monocultures [43,87].
Brazilian satintail and cogon grass are both nonnative, rhizomatous perennial grasses that are similar in appearance. They are primarily distinguished between one another by stamen numbers: Brazilian satintail usually has 1 stamen/flower, and cogon grass has 2 stamens/flower [43,64,106]. Other distinguishing characteristics include Brazilian satintail's relatively shorter spikelets (<3.5 mm) and narrower culm leaves (<5 mm) compared to cogon grass's spikelets and leaves . These characteristics overlap [43,86], however, and it is likely that the 2 grasses have been misidentified in the Southeast . Identification is further confounded by Brazilian satintail × cogon grass hybridization in the Southeast, the extent of which is unknown .RAUNKIAER  LIFE FORM:
Cogon grass reproduces from seed, rhizome expansion, and rhizome fragments [43,86]. Both seed and rhizome regeneration are important in its spread. Seed reproduction allows for long-distance dispersal and colonization, whereas rhizome spread is the primary means of population expansion [56,64]. Transported rhizome fragments also contribute to its long-distance dispersal and colonization .
Breeding system: Cogon grass is outcrossing [43,50,125,133]. Clonal populations show low or no fertility . Imperata cylindrica var. major shows considerable diversity in reproductive morphology and physiology in Asia ( and references therein),[124,146]. Studies in central and northern Florida suggested a high degree of genetic variability among cogon grass populations. Populations with low genetic diversity tended to have low seed viability, while populations with high genetic diversity had high seed viability. It is not known whether low seed viability was due to inability to outcross, poor environmental conditions, or other factors. The authors concluded that successful outcrossing was low in most cogon grass populations, but higher rates of genetic diversity and fecundity could be expected as southeastern populations expand and outcross .
Pollination: Cogon grass is pollinated by wind [94,133].
Flower production: Cogon grass flower production is highly variable. Some researchers report cogon grass as highly productive , but flowering is often sporadic, ranging from none to frequent flowering within and among populations [34,43,106,170]. In a common garden study using Malaysian collections, some cogon grass populations frequently produced flowers; others never produced flowers (but spread vegetatively); while most produced flowers only after mowing disturbance . Disturbances including nitrogen amendment, slashing, burning, defoliation, and grazing may trigger cogon grass flowering [43,63]. However, Shilling and others  found consistent flowering in 11 Florida cogon grass populations, none of which were disturbed. Field and greenhouse studies suggest that cogon grass flowering is not photoperiod-dependent .
Seed production and seed viability likewise vary widely among populations. A Florida study found that geographically isolated cogon grass populations did not produce seed, but plants within the population produced fertile seeds when cross-pollinated with pollen from another population . Saijise  found a mean of 700 seeds/panicle on cogon grass plants in the Philippines. A spikelet count in Florida showed a mean of 363 ± 47.5 spikelets/panicle. Actual production was higher because some spikelets had shattered prior to data collection . A Malaysian study found heavy flower production followed by low seed set . Preliminary investigations in Florida found flowers growing under stressful conditions rarely produced seed, so cogon grass has sometimes been labeled as a poor seed producer [37,124]. However, later research showed cogon grass can produce seed prolifically, even after disturbance [94,133]. Fire, tillage, mowing, and cold stress may stimulate cogon grass flower and seed production .
Seed/rhizome dispersal: Cogon grass seed is spread by wind. The seeds are small and light weight, with long, hairy plumes aiding wind carriage [43,94,133,164]. Cogon grass seeds may drift 15 miles (20 km) in open country . Shilling and others  showed that wind can disperse cogon grass spikelets up to 360 feet (110 m) from the parent plant. Cogon grass spread in Alabama from 1973 to 1985 was apparently due to northeasterly prevailing winds from the Gulf of Mexico blowing seeds up Interstate 65 [164,165].
Roads and road construction are important corridors for cogon grass dispersal [17,169]. Rhizomes are transported by machinery and fill dirt during construction [43,107]. Most long-distance dispersal of cogon grass is probably from inadvertent human transport of rhizomes and seeds . Willard and others [169,170] speculated that cogon grass spread in Florida was mostly from transporting soil contaminated with cogon grass propagules.
Seed banking: Cogon grass seed is short lived, generally remaining viable in the soil for about 1 year . Viability of seeds stored in a laboratory steadily decreased over 13 months . Field studies in Asia show a maximum seed life of 16 months [124,125].
Germination: Cogon grass seeds are not dormant and do not require stratification. They germinate 1 to 4 weeks after ripening [8,34,124,125,133]. Shilling and others  found that with 11 Florida cogon grass populations, seeds began germinating within 7 days of harvest, with 94% germination by day 14. Seed viability is variable. Seed collected from 9 sites in central Florida showed high variability in germination rate between sites, with viability ranging from 0% to 100% . An Alabama study found 80% to 95% seed viability ; another study found 0% germination in Mississippi and 20% in Florida in the same year . Across years in a single population, an Alabama study found 4% germination in 1970 and 70% germination in 1972 .
In the laboratory, cogon grass seed collected in Alabama germinated at temperatures from 77 °F to 95 °F (25 °C-35 °C), with best germination at 86 °F (30 °C). Light increased germination time and rate . A Philippine study also found high germination (>80%) in open areas . Light and soil fertility interactions may affect germination. In Florida, seeds germinated with light did not show an increased germination rate when fertilized with potassium nitrate solution; however, seed germinated in the dark had highest germination rate with addition of potassium nitrate .
Seedling establishment: Seedlings establish best on open, disturbed areas . In a greenhouse study conducted on seed bank samples collected over 2 years in Polk County, Florida, cogon grass seed emerged over a 3-month period. Seedling density averaged 1.9 ± 0.48 seedlings/m². There was no significant difference (P=0.78) in seedling emergence between collection years, but emergence differed significantly (P=0.001) with month of soil collection. Best emergence occurred in samples collected from April to June, particularly samples collected in May. Another emergence spike occurred in samples collected in December and January. Seedlings did not emerge from soil samples collected in other months . Cogon grass seedlings tend to emerge in clumps, reflecting the tendency of spikelets to disperse in clumps . Seedling mortality is generally high, with about 20% of emergents surviving to produce seed. Risk of mortality probably lessens when seedlings sprout rhizomes .
For established populations, asexual regeneration from rhizomes is cogon grass's primary method of expansion [7,43]. Kushwaha and others  reported that on old fields in India, cogon grass regenerated from mostly seed on recently burned, clipped, or abandoned plots, but regenerated only from rhizomes on 3- to 5-year-old fallows. On 2 study sites in Mississippi, cogon grass spread into longleaf pine savannas from infested roadsides. Spread was almost entirely from rhizomes. Rhizome spread slowed, but did not stop, as the populations expanded into interior savannas . Eussen  reported that cogon grass can produce 350 rhizomes in 6 weeks, and cover 4 m² in 11 weeks.
Regenerative capacity of cogon grass rhizomes is linked to stem age, length, thickness, and number of large buds. Only old ("2nd generation" or rhizomes arising from rhizome buds) rhizomes can sprout and grow roots . Rhizomes sprout readily after mowing, grazing, or burning removes top-growth . A low root:rhizome ratio aids in rapid regrowth after fire or mowing . In a growth chamber study, Ayeni and Duke  found old, large rhizome segments showed best stem sprouting and biomass gain compared to small, younger rhizome segments. Soerjani  found rhizome sprouting ability was not restricted by bud size, position on the node, internode length, or node diameter. In greenhouse and laboratory experiments, potted rhizomes buried deeper than 3 to 8 inches (8-20 cm) below the soil surface show poor sprouting ability [37,165].
Possibly because of low intrapopulation genetic diversity and inability to outcross, isolated cogon grass populations reproduce mostly or entirely by clonal expansion from rhizomes. Although rhizome growth is rapid, populations that reproduce mostly by cloning probably have lower overall rates of expansion compared to populations that reproduce from both seed and rhizomes. Overall rates of invasion probably increase when seed-reproducing cogon grass populations expand into and cross-pollinate with previously rhizome-expanding populations .
Growth: Ramet growth is considerably faster than seedling growth. In the greenhouse, Shilling and others  found plant height, leaf number, and biomass were significantly greater (P<0.001) in plants grown from broken rhizome fragments compared to seedlings. Rhizome fragments produced new secondary rhizomes within 4 weeks, while seedlings took 12 weeks to produce primary rhizomes. Cogon grass rhizomes can produce 350 shoots in 6 weeks and cover 4 m² in 11 weeks . In the greenhouse, cogon grass seedlings produced primary rhizomes 4 weeks after emergence . In Marion County, Florida, 3- to 4-month-old, wild seedlings were observed in the 5-leaf stage in October, and seedlings had formed roots and primary rhizomes. Secondary rhizomes were not yet present .
Growth may vary among cogon grass populations. In a greenhouse experiment, plants grown from rhizomes collected in Mississippi (2 populations of Philippine origin) were significantly smaller (P<0.05) than plants grown from Alabama rhizomes (2 populations of Japanese origin). In the growth chamber, ideal day/night temperatures and photoperiod across cogon grass populations were 84/73 °F (29/22 °C) and 16 hours, respectively .
Biomass of fully developed cogon grass stands is considerable. A New Guinea study found cogon grass's annual dry-matter production averaged 23 Mg/ha . In a Java field study, Soerjani  determined that undisturbed cogon grasslands contained approximately 3 to 6 million shoots/ha, 7 to 18 tonnes of leaves/ha, and 3 to 11 tonnes of rhizomes/ha.SITE CHARACTERISTICS:
Cogon grass tolerates a wide range of site conditions across its worldwide range. It is drought tolerant, and somewhat shade and salt tolerant . In its native lands of Asia and Africa, it grows on arid desert sands, river margins, and swamps . Describing cogon grass in Indonesia, Terry and others  wrote "unlike most other plants ... I. cylindrica can tolerate drought, waterlogging, fire, cultivation and short-term shade ... at a single site." Imperata cylindrica var. major, the variety in North America, commonly occupies a wide variety of habitats in Asia including grasslands, deforested areas, old fields, cultivated fields, riparian areas, and disturbed sites such as roadsides. Other varieties have narrower habitat requirements and are less ubiquitous in their native ranges . Hubbard  speculated that when Southeast Asian lands were still pristine, cogon grass may have been restricted to arid, relatively sterile, or heavy clay soils. In the United States, cogon grass is common on disturbed sites such as roadsides, mine spoils, pastures, agricultural lands, plantations, and early seral pine forests [43,90,173]. It also occurs on relatively undisturbed sites including wet and dry bottomland  and old-growth longleaf pine forests .
Soils: Cogon grass is sometime mistaken as an indicator of "degraded" lands with nutrient-poor soils. Although common on nutrient-poor soils (Ultisols and Oxisols) that native southeastern grasses cannot tolerate, it also occurs on soils of moderate to high fertility (Inceptisols and Andisols) [43,46,100]. Cogon grass tolerates a wide range of soil textures from coarse sands to heavy clays . Soils in cogon grass's native Asia are often highly leached, with low pH, fertility, and organic matter ; however, cogon grass is not limited to nutrient-poor soils in Asia . About 65% of cogon grass in Asia grows on strongly acidic soils (pH≤5.0) with a topsoil layer of 4 to 6 inches (10-15 cm) . Nigerian researchers report cogon grass growing on slightly acid to neutral soils . Best growth in North America occurs on moist, very strongly acid (pH 3.0-4.7) clay soils [43,64,124]; however, cogon grass often grows on clay soils of neutral pH in Florida [43,64]. On poor soils, cogon grass's ability to form monotypic stands in the southeastern United States is due in part to its ability to outcompete native herbs for space, light, water, and nutrients [12,26,38,48,86]. Cogon grass forms thick swards that cover thousands of hectares on abandoned phosphate mines dug in the heavy clay soils of Polk County, Florida .
Elevation: Worldwide, cogon grass is most common at elevations from sea level to 3,000 feet (1,000 m) elevation . Elevational ranges for cogon grass in the United States were not reported as of this writing (2005).
Climate: Cogon grass is native to regions of wet-tropical and subtropical Asia and Africa where annual rainfall averages between 40 to 100 inches (1,000-2,500 mm) ( and references therein),. Worldwide, cogon grass is most invasive in wet tropical and subtropical areas receiving 30 to 200 inches (750-5,000 mm) of annual rainfall . It tolerates hot temperatures but is sensitive to cold [164,165]. It is limited to latitudes below 45° in both hemispheres ( and references therein). Rhizomes cannot recover when subject to temperatures of approximately 14 °F (-10 °C). Cogon grass survived winter temperatures that dropped to 7 °F (-14 °C) in Alabama , but did not survive winter temperatures of 18 °F (-8 °C) in Mississippi .
Moisture regime: Cogon grass tolerates both xeric and flooded soils, but cannot tolerate soils that are waterlogged for long periods of time . Along the Nile River in Egypt, cogon grass is associated with high-moisture, high-salinity sites . It grows up to the edges of standing water in Florida , but does not invade continually flooded sites . In a greenhouse experiment, cogon grass germinants were intolerant of soil inundation and became increasingly tolerant of saturated soils as the plants matured. The authors concluded that soil inundation in early spring could limit cogon grass seedling establishment .SUCCESSIONAL STATUS:
Cogon grass is an early seral species. In both native and nonnative habitats, it depends on fire or other frequent disturbance to maintain dominance [37,82]. In tropical and subtropical ecosystems of Asia, cogon grass ordinarily declines and disappears with postdisturbance canopy closure [37,81,144]. It does best in full sun. Cogon grass cannot tolerate deep shade [15,100], but can survive in the moderate shade of savannas [63,64]. It establishes in forest gaps of all sizes. In longleaf pine/wiregrass (Aristida spp.) wet savannas of Grand Bay National Wildlife Refuge, Mississippi, cogon grass was experimentally seeded-in on small- (10 cm in diameter; 66% of full sunlight) to large-diameter (100 cm; 89% full sunlight) gaps created by herbicide spraying. Cogon grass germination averaged 40% across treatments. Seedling survivorship did not differ among gap sizes (P>0.05) . For information on cogon grass succession in fire-created gaps, see Discussion and Qualification of Plant Response.
In its native Southeast Asia, cogon grasslands are an early successional stage that develops following a stand-replacement event, usually fire [37,104,114]. Cogon grass is common in tropical old-field succession, with or without fire, but shifting (slash-and-burn) agriculture has greatly increased its occurrence in Asia and Africa. In its native lands, cogon grass has formed extensive swards in areas that were once forested [85,174]. In Asia, cogon grasslands succeed to tropical forest if succession is not interrupted by slash-and-burn agriculture [10,42,104]. Postdisturbance tropical forest development can be blocked by lack of sprouting trees, seed bank depletion, lack of off-site seed dispersal, and/or depletion of soil nutrients [42,100]. Several cycles of sand-replacement disturbance are usually needed for tropical forest-to-grassland conversion [37,100]. In northeastern India, cogon grass colonized burned fields for up to 6 years after burning. Importance value of cogon grass peaked at 74.5 in study plots at postfire year 3, when cogon grass was the most important plant species present. After 6 postfire years, Kashia pine and broadleaved trees began establishing and shading out cogon grass, sticky snakeroot, and other early successional herbs . Without frequent disturbances, cogon grass usually becomes less important as succession advances. On old fields in India, cogon grass was successionally eliminated on plots undisturbed for more than 5 years 
Cogon grass also occurs after nonanthropogenic disturbances. It was recorded as a pioneering species on Krakatau, Indonesia, 14 years after the 1883 volcanic eruption. The nearest point of possible seed dispersal was 25 miles (40 km) away (references in ). On coastal Japan, cogon grass dominates stabilizing sand dunes, becoming less common on either unstable or stable dunes . In Puerto Rico, cogon grass was 1 of the most frequent pioneering herbaceous species in bracatinga forests disturbed by hurricanes .SEASONAL DEVELOPMENT:
Worldwide, cogon grass shows variable phenological development depending upon climate and population genetics. Cogon grass flowers from May to June in its native Japan . Generally, populations growing in mediterranean climates tend to flower in spring and summer, while populations in tropical and subtropical areas (including Florida) tend to flower year-round . On foothill slopes of the western Himalayas in India, cogon grass germinated in March to April; grew vegetatively from May to June; flowered from July to mid-August; fruited from late August to early September; produced ripe fruit through September; and was dormant from October through December .
In Florida, cogon grass flowering peaks in late winter to spring (March-May) [25,173]. In a Polk County study, cogon grass flowered in November and December and again in March and April in both years of a 2-year study. Flowering time was consistent within a population, but varied across populations .Cogon grass rhizomes develop in spring at about 4 weeks of age, or the 3rd or 4th leaf stage of seedlings. Seedling rhizomes are initially vertical, growing horizontally by the 5th leaf stage [17,43].
Brazilian satintail sprouts from rhizomes after top-kill . Postfire seedling establishment is also likely. Mass flowering has been noted in Brazilian satintail following fires in Brazil [55,93].
Cogon grass sprouts from rhizomes after top-kill by fire [64,97,127,144]. It also establishes from seed, usually blown in from off-site . Regrowth from rhizomes is rapid [8,124], and frequent fire favors cogon grass over associated species worldwide [15,97,100,160,161]. Fire is so important to cogon grass's ecology that relative response to fire is one of the characteristics used to distinguish between its varieties .
Fuels: Descriptions of fuel characteristics and fuel loads in Brazilian satintail grasslands were not available as of 2005.
Cogon grass invasion changes fuel properties in pinelands of the southeastern United States. As a tall, rhizomatous grass on sites historically dominated by bunchgrasses, cogon grass produces more standing biomass and litter than native bunchgrasses. Thus, it increases fuel loads and horizontal and vertical continuity of fuels .
Fuel load estimates are needed for cogon grass-dominated sites in the United States. Fuel load measurements in native cogon grasslands may serve as a first step for estimating fuel loads in the southeastern United States. Pickford and others  conducted fuel sampling in burned and unburned forest-mangrove (Acacia mangium)/cogon grass stands in Java. They noted a "significant quantity" of dead, cured fuels that were created by and remained after burning, even in areas where cogon grass was green before the fire. They provide fuel loading and fire behavior estimates (based upon the BEHAVE fire behavior prediction system) for that community. Wibowo and others  provide fire behavior and severity information for a forest-mangrove/cogon grass community in West Java, Indonesia.
Fine fuels are the most important factor in ignition and spread of fire in Florida longleaf pine ecosystems , and cogon grass contributes a large fine fuel load. Observational  and anecdotal  accounts from Indonesia indicate that live cogon grass plants ignite and burn easily while still relatively green, and researchers in Indonesia note that cogon grass becomes very dry and flammable during the dry season . Cogon grass's fuel properties and abundant litter may alter fire behavior on invaded sites in Florida [86,87]. Cogon grass is high in silica content, so the litter decays relatively slowly. In an Australian study, cogon grass had the slowest decay rate of 3 grass species studied. Its half-life rate of decay exceeded the study period of 24 weeks .
On Florida sandhill longleaf pine savannas, Lippincott [86,87] compared fine fuel loads, fire behavior, and fire effects on uninvaded and cogon grass-invaded sites. Cogon grass produced significantly more persistent, standing dead biomass compared to sites with native understory vegetation (P<0.05), resulting in a greater fuel load on invaded sites. Fire mortality of young longleaf pines was greater on cogon grass sites, and postfire fuel accumulations were also greater on cogon grass sites. Average fire temperatures were higher on cogon grass sites and reached a maximum of 856 °F (458 °C) compared to a maximum of 604 °F (318 °C) on uninvaded sites . Such fires are severe enough to kill longleaf pine seedlings and saplings . See the Fire Case Study for additional details.
Even in frequently burned communities, cogon grass may alter fire characteristics by increasing fine fuel loads. Platt and Gottschalk  investigated the effects of cogon grass and silkreed (Neyraudia reynaudiana), another nonnative tropical grass, on fine fuel loads in south Florida slash pine savanna in Everglades National Park. The historical fire regime of the area is surface fires at 5- to 10-year intervals. Fuels are almost all fine: woody debris is rarely present except after hurricanes. Firegrass (Andropogon cabanisii) and other bunchgrasses native to the area tend to produce greatest biomass the first year following a fire; they also mass flower at that time. Productivity of native bunchgrasses decreases with time since fire. In contrast, cogon grass produces prodigious biomass nearly every year. Study plots were on prescribed underburn rotations of 10 years or less. Study design compared plots with a native ground cover of firegrass with areas that contained 1 of the 2 nonnative grasses. Total plant biomass (measured as g/484cm²) on plots with cogon grass was 1.7 times greater than on plots without cogon grass: a significant difference (P=0.03). Litter biomass was also significantly greater on plots with cogon grass (P=0.05) and was almost twice that on plots without cogon grass. Biomass of native plants was not different among plots with and without cogon grass .
Fire regimes: Little information is available on fire regimes where Brazilian satintail is native. Scott  investigated its occurrence in mixed muttonwood-copperleaf-Brazilian satintail-bracken fern savannas of montane eastern Peru. The study site was a tropical-humid forest area inhabited by Native Campa. He noted that the Campa practiced annual, dry-season burning around their village to maintain Brazilian satintail grassland. Areas where burning was abandoned succeeded to either bunchgrasses (in areas without sprouting woody species) or tropical forest. The origin of South American tropical savannas is unclear. Anthropogenic burning may be responsible. Hardpan soils over high water tables, wet climate, and a combination of anthropogenic burning, edaphic, and climatic factors are also suggested (numerous references cited in ). All researchers concede that regardless of their origins, South American savannas are currently maintained by frequent, intentional burning . For information on postfire succession on Brazilian satintail old fields of Peru, see Successional Status.
No information is currently available on how Brazilian satintail affects fire intervals and behavior in southeastern pinelands. Information is needed on the fire ecology of Brazilian satintail in the United States and elsewhere.
Worldwide, cogon grass is favored by frequent surface fire in pine (Pinus spp.) and other savannas and by very frequent (<10-year rotation) stand-replacement fire in grasslands. In its native Southeast Asia, cogon grass occurs in systems that experience frequent fire including farmlands, grasslands, and the understories of tropical and subtropical forests, especially pine forests [42,97,141]. Charcoal evidence of fires in Borneo date back to the Holocene (review by ), but knowledge of natural fire regimes where cogon grass is native is lacking. In Indonesia  and Australia  cogon grass can tolerate annual fires, and frequent fire maintains cogon grasslands, which are successionally replaced by shrubs and/or secondary tropical forest in the absence of fire [100,141,144,161]. Cogon grass fuels fires that help maintain subtropical and tropical savannas and forest-grassland mosaics in Southeast Asia [144,161]. Forest fires can result in the spread of cogon grass in these ecosystems . After large-scale fires in East Kalimantan in 1983 the area covered by cogon grassland expanded dramatically . Small-scale fires within subtropical and tropical forest of Nepal maintain uneven-aged forest-grassland mosaics . In subtropical Chir pine forests of Nepal and Pakistan, frequently burned slopes support cogon grass, several other grass species, and a variety of shrubs (Shrestha and Joshi 1997, cited in ).
Both fire and logging can increase establishment and spread of cogon grass, but this effect is "greatly enhanced" when these disturbances are combined . Fires in Borneo and other tropical areas tend to occur during El Niño-induced droughts, and cogon grasslands expand during these drought-fire cycles [114,172]. Logged areas tend to be more susceptible to fire, and when it occurs, fire is more severe in logged areas. In Borneo, logged forests showed less understory diversity after fire compared to unlogged forest. Logged and burned forests were mostly dominated by cogon grass and/or Jack-in-the-bush, whereas these species were present but not dominant after fire in unlogged forests .
Shifting agriculture (slash-and-burn) has shortened fire intervals in many tropical areas to the point that warm-wet-climate pines and other overstory trees can no longer regenerate, creating large-acreage swards of cogon grassland where cogon grass had formerly occupied only small patches within forest mosaics . Its spread in its native range in Southeast Asia is largely due to human clearing of tropical rain forests followed by frequent burning [38,46]. In northeastern India intervals between fires in Khasia pine forests have been shortened from 20- to 30-year intervals to 5-year intervals due to shifting agriculture. This has resulted in cogon grass dominance in fallow fields and cogon grass invasion into crops on cultivated lands . Cogon grass is successionally replaced by woody species in the absence of further fire [172,174]. Slashing and burning every 4 to 6 years tends to exclude species other than cogon grass and other herbaceous weeds, while succession to woody species occurs with 10- to 20-year slash-and-burn cycles . Repeated short-interval fires on cogon grasslands in Indonesia increase cogon grass abundance, reduce soil fertility, and increase soil erosion, ultimately making reversion to forest more difficult [51,52,53]. In Australia cogon grass has spread in tropical and subtropical regions where frequent burning occurs [13,115]. For example, it is an important component in an eucalyptus (Eucalyptus spp.) forests in northern Australia where Aborigines conduct frequent underburning .
Cogon grass in North America: There is potential for cogon grass to spread rapidly in warm-wet climate regions of the southeastern United States . Cogon grass's rapid spread in peninsular Florida and edges of the northern Gulf of Mexico is thought to have increased fire hazard on invaded sites, and cogon grass populations are expected to continue to spread in the Gulf Coast region . Cogon grass is well adapted to the subtropical pine ecosystems in that area (see Fire Adapations). Cogon grass may establish without or without fire [74,75], and can spread rapidly after fire [75,86,87].
Prior to the 20th Century, wildfires in the southeastern United States were most common in summer. Lightning strikes occur during the rainy season (May-October), with most lightning-ignited fires occurring in spring and summer [86,135]. Peninsular Florida's longleaf pine communities experienced frequent surface fires at 2- to 8-year intervals. These frequent fires maintained the savannas [24,101,123,140]. Southern Florida's slash pine forests probably had similar, but slightly longer (up to 15-year) intervals between surface fires. Reconstructing past fire regimes from fire scars has not been possible for southern Florida's pinelands . Florida was inhabited by Aboriginals for thousands of years, and seasonality and extent of Aboriginal burning in the area is uncertain .
Fire behavior: Cogon grass invasion in pinelands of the southeastern United States may shorten fire-return intervals and increase fire severity over prehistoric conditions [16,86,87]. A report by the Mississippi Exotic Pest Plant Council identified cogon grass as 1 of the 4 most serious nonnative invasives in the Southeast, based, in part, upon its potential to alter natural fire regimes .
Cogon grass invasion may increase fire's rate of spread and intensity on invaded vs. uninvaded sites [86,87]. Compared to native bunchgrasses, cogon grass produces a more continuous bed of fine fuels that is highly flammable when dry. A study in southern Florida pine sandhills found that cogon grass fuelbeds were more evenly distributed than fuelbeds dominated by native grasses, resulting in more horizontally continuous burned areas . On sites where cogon grass reaches maximum heights (see General Botanical Characteristics), it also increases vertical continuity of fuels, which may change the fire regime from surface to crown fire [16,87]. Observations in cogon grass sites in Mississippi indicated that flame heights were nearly twice those in sites dominated by wiregrass. Rates of fire spread were higher on cogon grass sites; however, maximum temperatures were lower (Grace, unpublished, cited by ). Ironically, cogon grass was once planted in firebreaks on the Withlacoochee State Forest, Florida .
The following table provides fire return intervals for plant communities and ecosystems where Brazilian satintail and cogon grass are important. For further information, see the FEIS review of the dominant species listed below. This list may not be inclusive for all plant communities in which Brazilian satintail and cogon grass occur. Find further fire regime information for the plant communities in which these species may occur by entering the species' names in the FEIS home page under "Find Fire Regimes".
|Community or Ecosystem||Dominant Species||Fire Return Interval Range (years)|
|shortleaf pine||Pinus echinata||2-15|
|shortleaf pine-oak||P. echinata-Quercus spp.||<10|
|slash pine||P. elliottii||3-8|
|slash pine-hardwood||P. elliottii-variable||<35|
|sand pine||Pinus elliottii var. elliottii||25-45 |
|South Florida slash pine||P. elliottii var. densa||1-15 [102,135,158]|
|longleaf-slash pine||P. palustris-P. elliottii||1-4 [102,158]|
|longleaf pine-scrub oak||P. palustris-Quercus spp.||6-10|
|pond pine||P. serotina||3-8|
|loblolly pine||P. taeda||3-8|
|loblolly-shortleaf pine||P. taeda-P. echinata||10 to <35 |
|cabbage palmetto-slash pine||Sabal palmetto-P. elliottii||<10 [102,158]|
Fire top-kills cogon grass and consumes much of its associated litter [64,111,133].DISCUSSION AND QUALIFICATION OF FIRE EFFECT:
Cogon grass sprouts from the rhizomes soon after fire [64,144]. Studies in Asia [49,97,136] and observations in Australia  show that burning increases cogon grass sprouting. A study in Java notes that cogon grass sprouted from "charred but obviously viable tillers" that "extended 4 to 6 inches (10-15 cm) above the soil surface" soon after fire. Forty days after fire cogon grass stem height averaged 4 feet (1.3 m) . Several reviews indicate that fire and other disturbances stimulate flowering in cogon grass (e.g., [43,64,133]). Postfire mass flowering has been noted in cogon grass in Thailand (Paisooksantivantana, cited in ).
Cogon grass also establishes from wind-blown, off-site seed after fire [97,136]. Since it has a short-term seed bank , seedling establishment from soil-stored seed is possible.
Cogon grass is favored by frequent fires [64,144,160,172]. Garrity and others  stated the "real distinguishing factor for its persistence is the intermittent occurrence of fire." Coster  called fire "the greatest help" to cogon grass spread in Asia. Reviews state that cogon grasslands remain stable  and become dense monocultures  when burned annually.
Fire increases nutritional value of cogon grass in the short term. For 3 months after fire in central Florida, cogon grass on burned sites had significantly higher nitrogen and phosphorus content, and lower fiber content, compared to unburned cogon grass .
In postfire succession in tropical forest ecosystems, cogon grass is more abundant on previously logged sites, where it is initially important, then declines as woody plants establish and assume dominance (e.g., [103,144,161]). A severe drought in Indonesia in 1982 and 1983 caused widespread wildfires across Borneo [103,172]. Woods  compared early postfire succession on logged and unlogged sites after a fire in Borneo in April and May 1983. Canopy losses were higher and postfire seedling regeneration was dominated by cogon grass and hilo grass on sites logged <2 years before fire, and by cogon grass and Jack-in-the-bush on sites logged 6 years before fire. Grasses and lianas were present but not dominant and canopy losses were less severe on sites that were not logged prior to the fire . Nykvist  found that grass biomass decreased with time on these sites, and mean grass biomass was 20 kg/ha at postfire year 5 and 3 kg/ha at postfire year 8.DISCUSSION AND QUALIFICATION OF PLANT RESPONSE:
|Treatment||Time since last fire||Rate of spread (± SD)|
|No treatment in plug-planted stands||5 years||0.5 ± 0.4 m/year|
|Fire exclusion in self-established stands||15 years||0.5 ± 0.4 m/year|
|Uncut plantation, self-established stands||4 years||0.6 ± 0.1 m/year|
|Prescribed burning, self-established stands||5 years||1.9 ± 0.9 m/year|
|Prescribed burning, self-established stands||1.5 years||2.6 ± 0.9 m/year|
|Clear-cut plantation, self-established stands||4 years||2.7 ± 0.4 m/year|
Cogon grass can establish in both large and small gaps, growing best if gaps are created by fire . On the Grand Bay National Wildlife Refuge, Mississippi, six 10 × 10-m areas of longleaf pine savanna were burned under prescription on 6 April 1998. Cogon grass was transplanted onto transplant study plots 2 days after burning or seeded into seed study plots 3 days after burning. Percentage photosynthetically active radiation was 78% and 30% on burned and unburned plots, respectively. At postfire month 2, survivorship did not differ between cogon grass transplants on burned and unburned plots (P=0.72), but growth (mean shoot length) of transplants was greater on burned vs. unburned plots (P=0.0007). For seedlings, emergence did not differ between burned and unburned plots; however, survival of germinants at postfire months 1 and 2 was significantly greater on burned vs. unburned plots (P<0.05) .
Early in its postfire recovery, cogon grass may allocate most of its biomass to rhizomes. Following mowing and prescribed burning in an old field in India, cogon grass regenerated entirely from rhizome sprouts. At postfire year 1, aboveground:belowground biomass ratio of cogon grass was 1:4, which was the lowest ratio of the 4 plant species studied on burned plots .
The Fire Case Study Imperata cylindrica in a Florida sandhill longleaf pine community provides information on fuel loads, prescribed fire use, and postfire response of juvenile longleaf pines on cogon grass-infested sites and uninfested sites.FIRE MANAGEMENT CONSIDERATIONS:
Cogon grass is extremely problematic for fire managers. It invades fire-adapted, warm-wet-climate ecosystems, reducing species diversity and ecosystem function. Cogon grass has the potential to shorten already short fire-return intervals to the point that native plant species cannot recover. Yet excluding fire from these fire-adapted ecosystems also results in loss of ecosystem diversity and function . Rapid accumulation of dense cogon grass litter, along with its spreading rhizome mass, makes unassisted recruitment of native warm-climate plant species unlikely on infested sites [86,87].
Fire alone cannot control cogon grass; in fact, burning with no further treatments will promote it (see Plant Response to Fire). Burning can help control cogon grass when it is part of an integrated control plan, however [62,160]. By removing cogon grass top-growth, the rhizomes are forced to utilize stored carbohydrates to produce new growth, thereby weakening the rhizomes. Removing cogon grass litter and standing dead biomass prior to other treatment often improves the success of other control measures. For example, tillage is more effective, and herbicide application to growing tissues more precise, if biomass is first removed . Johnson and Shilling  provide a contact list of managers and academics with experience using fire to control cogon grass.
Burning and allowing cogon grass regrowth, followed by tillage and herbicide treatment, is the most effective control measure for large, established infestations of cogon grass [35,133]. In a Florida study, burning was used to remove aboveground cogon grass biomass and prepare a bare soil study area on all plots. Postfire treatments were 2 herbicide sprayings, 2 diskings, or spraying/disking combinations. Imazapyr was applied 44 and 90 days after burning. Disking was done the day after burning and at postfire day 90. Measured 18 months after treatment, the most effective treatment was a 1st disking on postfire day 1, followed by spraying at postfire day 44, and a 2nd disking at postfire day 90. Compared to untreated plots, spraying alone provided 82% control, and disking alone provided 53% control. Disking followed by spraying without a 2nd disking resulted in 86% control .
After cogon grass suppression, establishment of native herbaceous species is needed for long-term control . Shilling and others  stated "if a replacement species does not fill the niche occupied by cogon grass after suppression then cogon grass will simply refill the niche."
Studies in southeast Asia show that although slash-and-burn treatments reduce rhizome biomass, they also encourage sprouting and seedling establishment. Slashing alone may produce more sprouts than slashing and burning [124,136]. However, Woods  found that logging and burning in combination resulted in greater postfire establishment of cogon grass than either disturbance alone. A combination of prescribed burning and mowing reduced cogon grass in infested pastures in Australia. In heavily infested pastures, burning was followed by reseeding to pasture grasses, then mowed repeatedly . A Malaysian manager reports that burning cogon grass early in the dry season reduces next-year fuel loads, while late dry-season fires tend to increase next-year fuel loads .Extensive, fire-created cogon grasslands can lower habitat quality and diversity in southeastern pinelands. Frequent fire on cogon grasslands in tropical Asia has reduced soil nitrogen and increased run-off . In Florida longleaf pine stands, gopher tortoise mounds provide fire refugia and disturbed seedbed sites for early seral herbs. Gopher tortoises have difficulty digging in cogon grass, preferring more open sites. Fewer gopher tortoise mounds in cogon grass-infested sites may affect postfire plant community succession .
|Common name||Scientific name|
|cogon grass||Imperata cylindrica|
|longleaf pine||Pinus palustris|
Lippincott, Carol L. 2000. Effects of Imperata cylindrica (L.) Beauv. (Cogongrass) invasion on fire regime in Florida Sandhill (USA). Natural Areas Journal. 20(2): 140-149. .
Lippincott, Carol L. 1997. Ecological consequences of Imperata cylindrica (cogongrass) invasion in Florida sandhill. Gainesville, FL: University of Florida. 165 p. Dissertation. .STUDY LOCATION:
Groundlayer vegetation differed on invaded and uninvaded plots. The ground layer on invaded sites was mostly nonnative cogon grass (Imperata cylindrica). Groundlayer vegetation on uninvaded sites was dominated by native bunchgrasses including pineland threeawn (Aristida stricta), pineywoods dropseed (Sporobolus junceus), narrowleaf silkgrass (Pityopsis graminifolia), and lopsided Indiangrass (Sorghastrum secundum). Summer farewell (Dalea pinnata) was a common forb associate .
Cogon grass was intentionally introduced into the Brooksville area in the early 1990s as a forage crop . It was probably introduced in the Citrus Tract from seed-contaminated road fill (Blanchard, cited in ). During the study period, cogon grass occurred in scattered swards in the Citrus Tract. Cogon grass patch size ranged from a few square meters to several hectares .PLANT PHENOLOGY
Longleaf pine juveniles were randomly tagged on burn and no-burn plots before prescribed burning. Fire spread was measured during burning. Fire severity was determined by measuring mortality of tagged juvenile longleaf pines and postfire growth rate of tagged juvenile longleaf pines that survived the fire. Mortality, height, and basal area of surviving pines were measured at postfire month 1. Height and basal area of surviving pines were measured again at postfire year 1 .
The natural fire regime of longleaf pine forests on Citrus Tract is surface fire at 2- to 8-year intervals, fueled primarily by pine needles, oak leaves, and perennial bunchgrasses including threeawns (Aristida spp.) and pineywoods dropseed (Sporobolus junceus). Lightning-ignited fires were most common during the spring and summer thunderstorm season .
The Florida Division of Forestry conducts regular prescribed burning on the Citrus Tract for forest and game management. The area in which study plots were located was last burned 4 years prior to study initiation . Just before this study's prescribed burn, mean moisture contents of live and dead fuels were similar on invaded and uninvaded plots (46.2% ± 10.7 and 42.3% ± 12.3, respectively). Mean heat of combustion was slightly higher for native grasses (18.40 kJ/g ± 0.20) compared to cogon grass (18.77 kJ/g ± 0.22). Prescribed fires were conducted early in the growing season (March and April) and ignited in mid-morning as backing fires. Midway through burning, wind shifts caused the 3rd and 4th fires to head. Treatment plot sizes and weather parameters were :
|Season||Fire type||Wind speed
|1||35 × 130||April 1995||backing||8||32||25|
|2||35 × 145||April 1995||backing||13||56-68||29|
|3||35 × 40||April 1995||backing/head||16||44-56||24|
|4||35 × 35||March 1996||backing/head||8||63||22|
Prefire fine fuel load was significantly less on native sandhill sites compared to cogon grass sites (P=0.04). From 0 to 0.49 m in height, fine fuel mass did not differ between uninvaded and cogon grass-invaded sites; however, fine mass of fine fuels from 0.50 to 1.50 m was higher (P<0.01) on cogon grass sites . Prefire fuel loads at 3 heights were :
Fine fuel biomass (g/m²)
|0 - 0.49||800||630|
|0.50 - 0.99||275||75|
|1.00 - 1.50||25||----|
Mean heat of combustion was slightly higher for native fuels (P<0.01). Instantaneous maximum fire temperature at 3 heights was measured with temperature-indicating paints on steel poles. There was a significant difference (P<0.05) in mean maximum temperature between prescribed fires in cogon grass (260.9 ±13.7 °C) and native sandhill (218.3 ± 14.5 °C) sites . Fuel load ratios and fire temperatures by height were :
|Aboveground fuel height||Fuel biomass ratio
Fire rate of spread was similar on invaded and uninvaded plots (P=0.75). Fireline intensity was also similar on invaded vs. uninvaded plots (P=0.22) :
|Site||Fire type||Rate of spread (m/s)||Intensity (kW/m)|
Fire mortality of longleaf pine juveniles was higher on cogon grass-invaded plots, and the postfire growth rate of surviving longleaf pine juveniles was decreased on invaded plots. Juvenile longleaf pine size classes were indicated by height. Longleaf pine mortality on invaded and uninvaded plots was [86,87]:
|Juvenile size class||Height (m)||
At postfire year 1, growth of surviving small juvenile longleaf pines was significantly less on invaded vs. uninvaded plots (P<0.01). Median increase of pines in the smallest size class was 21% vs. 50%, respectively . Poor growth in small longleaf pine juveniles was probably due to competitive interference by cogon grass, rather than direct fire effects to small longleaf pines . (See longleaf pine's Fire Ecology section for discussion on fire effects to longleaf pine juveniles). Height gains for medium- and large-sized longleaf pine juveniles were similar on invaded and uninvaded plots (P=0.86). For all size classes, stem diameter growth of longleaf pine juveniles was not significantly different on invaded and uninvaded plots .
Although cogon grass increases fire mortality of longleaf pine seedlings, mature longleaf pines may not be directly affected by cogon grass presence. In this study, growth of longleaf pines greater than 10.4 cm dbh was not slowed by cogon grass .FIRE MANAGEMENT IMPLICATIONS:
Cogon grass-invaded plots had 50% more fine fuel biomass than uninvaded plots prior to burning. Before burning, invaded plots had a significantly greater (P<0.01) fine-fuel load (µ=1,163 g/m² ± 285 g/m²) compared to uninvaded plots (µ=177 g/m² ± 297 g/m²). Structurally, fine fuels on invaded plots were generally taller than on uninvaded plots. Fine-fuel loads between 0.4 and 1.51 m in height were significantly greater on invaded plots (P<0.01). In contrast, fine-fuel loads less than 0.50 m in height were similar on invaded and uninvaded plots (µ=795 g/m² and 668 g/m², respectively; P<0.07). Horizontally, fine fuels were significantly (P=0.04) more continuous on invaded vs. uninvaded plots, with 3.0% bare ground on invaded plots and 0.3% bare ground on uninvaded plots. Fuels on invaded plots were distributed significantly higher above ground: 27% of total fuels were ≥0.5 m high on invaded plots compared to only 8% on uninvaded plots. Consequently, invaded plots produced significantly higher maximum fire temperatures compared to uninvaded plots (260.9 °C vs. 218.3 °C, P=0.40), and fires were more patchy on uninvaded plots. Fire temperatures in cogon grass reached a maximum of 458 °C on some strips at all aboveground heights measured. After fire, fine fuels accumulated more quickly on invaded plots [86,87].
These mortality data and instantaneous maximum temperature measurements at given points suggest that longleaf pine juveniles may succumb to cogon grass-fueled fires . Additional data on fire duration (e.g., Jacoby and others ) will help determine direct fire effects of cogon grass fuels on longleaf pine juveniles. Rapid growth out of the "grass stage" of growth gives juvenile longleaf pines protection from fire. However, this study showed that young longleaf pines in the 0.5-1 m height class are vulnerable to fire damage on cogon grass sites, which have more fuels and higher fire temperatures compared to sites with native bunchgrass fuels . Koskela and others  had similar findings in a Sumatran pine (Pinus merkusii)/cogon grassland of northern Thailand. As a juvenile, Sumatran pine has a "grass growth" stage similar to that of longleaf pine. Juvenile Sumatran pines were killed by frequent fires fueled by cogon grass .Cogon grass can alter longleaf pine community structure and consequently, its fire regime and level of diversity. As a fast-growing, rhizomatous grass that is supplanting slow-growing native bunchgrasses, cogon grass-invaded sites have higher fuels loads, greater horizontal and vertical fuel continuity, and potentially greater flame heights compared to sites with native herbaceous ground layers. In this study, overall understory plant diversity was lower on cogon grass-infested sites compared to uninfested sandhill sites . At postfire month 3, cogon grass sites had over 100% more fine fuels compared to uninvaded sites (P<0.01). Fuel accumulations at 6 and 14 postfire months were 86% and 50% more, respectively, on invaded compared to uninvaded plots. Rapid growth and nonbunching habit of cogon grass can increase fire severity, continuity, spread, and frequency in longleaf pine sandhill habitats, thereby increasing fire mortality of young longleaf pines and reducing habitat quality for native organisms adapted to longleaf pine/bunchgrass habitats .
Cogon grass is generally detrimental in wildlands and pastures in North America. It reduces habitat quality for wildlife that have evolved in pine/bunchgrass ecosystems [43,86]. It is little used as forage in the United States even though it was originally planted for that purpose. Pendleton  warned against cogon grass introduction in 1948:
Cogon grass "is anything but nutritious. Certainly its hazard as a potential weed for upland crops in the tropical and subtropical portions of the western hemisphere is a very much more serious threat to agriculture than the small amount of benefit it can possibly be as a forage. The writer feels very strongly that steps should be taken at once to completely eradicate this noxious weed from the western hemisphere"  (italics are Pendleton's).
Although cogon grass is weedy in wildlands where it is nonnative and in agricultural systems worldwide, it has a valuable ecological role as fuel and forage in grasslands where it is native. In Royal Bardia National Park, Nepal, for example, cured cogon grass helps fuel the natural fire cycle that maintains tropical forest-grassland mosaics. Live cogon grass provides cover for ground-nesting birds and forage for grazing animals. Cogon grass is an important component of the food web for threatened  animals in the Park including hispid hares, swamp deer, and tigers . In Borneo, the Banjarese historically burned open fields with cogon grass to create deer habitat and forage . Asian and African ranchers use cogon grass as cattle forage [30,35]. Cogon grass is expected to become less important forage in developing countries as it is replaced by plantings of more nutritious grass species .
Palatability: Cogon grass is relatively unpalatable and unnutritious for livestock and North American wildlife [40,41,43,58,86]. It is lower in nitrogen and higher in fiber and silica compared to native wiregrasses (Aristida spp.) of the Southeast [24,26,86]. The leaf blades are sharp and rough at the edges, discouraging animals from grazing . New spring growth and postfire sprouts are palatable to livestock for 3 to 4 weeks; however, plants become coarse and fibrous after that . In a rangeland study in subtropical Australia, cogon grass cover increased in response to cattle grazing at the expense of common carpet grass (Axonopus fissifolius), which is more palatable and nutritious . Stober  described cogon grass as unpalatable to domestic sheep in Malaysia; however, domestic sheep can learn to graze cogon grass .
As it becomes more common in the Southeast, cogon grass will affect grazing wildlife. Gopher tortoises, a federally threatened species , prefer native grasses and forbs to cogon grass . Three North American skipper butterflies graze cogon grass in the caterpillar stage .
Nutritional value: Nutritional studies of cogon grass in the Southeast are few. Lippencott  found that compared to native Florida sandhill herbs, cogon grass was higher in nitrogen and phosphorus and lower in fiber for the first 3 months after fire. By 6 postfire months, cogon grass provided less nitrogen and phosphorus, and by postfire month 14, it was higher in fiber compared to native herbs. Studies conducted on cogon grass in Asia are reported below.
Most sources claim that cogon grass forage quality declines quickly, is low in minerals (particularly phosphorus), and that cattle require nutritional supplements when grazing cogon grass ( and references therein, but see  for a contrasting viewpoint). In India, domestic goats on a native grassland mixture that included cogon grass showed poor weight gain . Asian ranchers have successfully raised cattle on cogon grass-legume pastures . Nutritional content of cogon grass from Asian sources (country not reported) was :
|Dry matter (%)||Crude protein (%)||Digestible protein (%)||Metabolizable energy (Mcal/kg)|
|1-14 days growth||27||10.4||6.7||2.36|
|85-98 days growth||35||8.5||5.1||2.18|
Analysis of cogon grass in Malaysia showed (as cited in ):
|Dry matter||Crude protein||Crude fat||Crude fiber||N-free extract||Ash|
|Digestible nutrients (%)||----||2.5||0.2||5.6||7.6||----|
A Thailand study suggests fire and repeated grazing reduce forage quality of cogon grass. A cogon grass sward was burned on 23 March 1978, then harvested every 3 weeks from April through October 1978. Mean dry-matter nitrogen content was 2.93% in April, declining to 0.56% in October. Phosphorus content declined from 0.90% to 0.37%, and in-vitro digestibility declined from 71% to 39% in the same time period. Nutrient values on an undisturbed 5-year-old cogon grass sward were generally lower: 0.66% nitrogen, 0.12% phosphorus, and 31.4% in-vitro digestibility .
Cover value: Cogon grass stands are poor habitat for most southeastern wildlife species . Cogon grass is about 3 times the height of native Florida grasses. Its height probably impedes movement of small animals. Ground-dwelling species can be displaced by cogon grass's dense cover . In central Florida, habitat quality of 2 keystone fossorial animals, gopher tortoises and scarab beetles, was reduced on cogon grass-invaded sites compared to uninvaded sites. Beetle populations were reduced approximately 76% on invaded sites. Threatened  gopher tortoise populations were too low to allow quantitative assessment; however, thick rhizome growth that deters burrowing, reduces the number of open areas for egg laying, and reduces herbaceous forage species, lowers gopher tortoise habitat quality. Southeastern pocket gophers, another keystone fossorial animal, were not affected by cogon grass presence .OTHER USES:
In traditional Asian folk medicine, cogon grass is used as a tonic, an emollient, an anti-inflammitory, and a fever-reducing agent (references in reviews by [48,90,91,92]). Chemical and pharmacological studies are underway to assess potential uses of cogon grass in modern medicine [90,91,92,112]. A cogon grass extract shows anti-insecticide properties against mosquitoes .
'Red Baron,' a cultivar of cogon grass with red leaves, is commercially available and planted as an ornamental despite cogon grass's federal and state listings as a noxious weed [27,171] an laws against its use . 'Red Baron' is usually described as infertile and nonspreading, but data are lacking to support the claim . Greenhouse studies suggest 'Red Baron' is at least capable of vegetative spread. After 3 months in the greenhouse, 'Red Baron' rhizomes produced a similar number of secondary rhizomes compared to Brazilian satintail rhizomes (x=5.7 and 5.1 rhizomes for 'Red Baron' and Brazilian satintail, respectively), and significantly more secondary rhizomes than Imperata cylindrica var. major, the "wild type" cogon grass (x=3.9 I. c. var. m. secondary rhizomes, p=0.05). Growth rates of the 2 cogon grass types and Brazilian satintail were similar . 'Red Baron' may be more shade tolerant than the wild type , and individual 'Red Baron' rhizomes may grow "aggressively" . Use of 'Red Baron' is not recommended in the United States for ecological reasons .IMPACTS AND CONTROL:
Cogon grass is identified as a large potential hazard to remnant old-growth longleaf pine stands due to its ability to invade undisturbed old-growth forests of the Southeast . In central Florida, Lippencott  found cover and diversity of native herbs, shrubs, palms, and small longleaf pines and oaks (Quercus incana, Q. laevis, Q. geminata) were significantly reduced on cogon grass sites compared to uninfested sites (P< 0.01) (see The Fire Case Study Imperata cylindrica in a Florida sandhill longleaf pine community. Yager  is conducting research on cogon grass rate of invasion into longleaf pine/bluestem and longleaf pine/shrublands in Mississippi. Preliminary data show an expansion rate ranging from 0 to 1.6 feet (0.5 m) over a 6-month period (April-Oct. 2002). Expansion rate is not significantly different between habitats . Spread of cogon grass outside the lower Coastal Plains states is unlikely because of cogon grass's poor tolerance to low temperatures [106,165].
Cogon grass allelopathy has been implicated in laboratory experiments [20,26,65,76,77,117]; however, such claims are based upon research using cogon grass extracts at concentrations that do not occur under field conditions. Reputed allelopathy of cogon grass awaits reciprocal transplant experiments in the field and/or greenhouse.
Silvicultural: Cogon grass competes with hardwood species for light, water, and nutrients. Cogon grass grows so tall and thick that it decreases growth and increases mortality of young trees . Allen and others  describe cogon grass as particularly problematic in peninsular Florida pinewoods. Lippencott's  Florida study suggests that pine and oak regeneration is reduced in cogon grass swards; however, she found that median basal area of mature pines was greater on cogon grass sites compared to uninfested sites. Basal area of oaks did not differ between sites (P<0.01). She postulated that mature longleaf pine was unaffected by understory cogon grass cover once longleaf pine reached a critical size (10.4 cm dbh or 85.1 cm² basal area in her study), but that longleaf pine regeneration would not occur until cogon grass was controlled .
Because cogon grass increases fire severity, wildfires in pine plantations infested with cogon grass may kill pine seedlings that are normally fire resistant .
Tropical ecosystems: Cogon grass is invasive in tropical and subtropical regions worldwide. Cogon grass is a troublesome agricultural weed in Asia and Africa [2,6,10,21,22]. Repeated agricultural burning has converted millions of acres of tropical forest to cogon grasslands . In a 1997 survey, Garrity and others  estimated that about 35 million acres (14 million ha) were dominated by cogon grass in Asia, mostly as a result of frequent fire on shifting agricultural lands. In Indonesia, for example, shifting agriculture has resulted in a type conversion of tropical forest to coarse tropical grasslands dominated by cogon grass and/or wild sugarcane (Saccharum spontaneum) [42,100]. Cogon grass coverage increased in Indonesia from 31.3 million acres (12.5 million ha) in 1996-1997 to 58 million acres (23.2 million ha) in 2000. Cogon grass coverage also increased in tropical forest understories . Van der Wall  demonstrated that it takes about 4 or 5 slash-and-burn cycles to effect a type conversion from tropical forest to cogon grassland.
Control: Cogon grass is not easy to kill. Its extensive rhizome system requires several treatments before control is effective [43,133]. Long-term control of cogon grass only works if the cogon grass monoculture is changed to a competitive, diverse plant community . Postcontrol revegetation needs to be accomplished quickly in order for a stable plant community to establish. Posttreatment inventory and spot-control treatments are needed to control new cogon grass infestations .
Prevention: The most efficient and effective method of managing invasive species is to prevent their invasion and spread . Preventing the establishment of nonnative invasive plants in wildlands is achieved by maintaining native communities and conducting aggressive surveying, monitoring, and any needed control measures several times each year. Preventing the introduction of cogon grass into uninfested areas, and early control of small infestations, should be a priority . Monitoring efforts are best concentrated on the most likely sites of cogon grass invasion: disturbed soil, roadsides, old fields, pine savannas, and burns. Uninvaded sites should be periodically surveyed to detect new invasions. The Center for Invasive Plant Management provides an online guide to noxious weed prevention practices.
Limiting seed and rhizome dispersal reduces chances for cogon grass spread. Regular cleaning of road equipment, and restrictions on long-range movement of equipment, can limit cogon grass spread. Removal of 'Red Baron' cultivars from the market may also reduce cogon grass spread .
Integrated management including burning, mowing, tillage, chemical, and/or cultural control will increase the likelihood of cogon grass control [35,69,133,166]. Successful control in Florida employs a combination of mechanical, chemical, and/or burning treatments. Johnson and Shilling  recommend beginning an integrated control program in late spring or summer, when cogon grass growth is peaking. Some treatments, like mowing and spot herbicide spraying, should be implemented year-round .
Physical/mechanical: Cogon grass can be controlled by mowing, disking, pressing, or a combination of these methods . Cogon grass is difficult to remove by hand cultivation due to its rhizomatous habit . Mechanical treatments may not be possible or practical on some wildland sites.
Mowing treatments alone cannot control cogon grass. In the Philippines, Sajise  found cutting treatments increased cogon grass sprouting compared to uncut control plots. Plants cut every 2 months produced more sprouts compared to plants cut monthly . Frequent, repeated mowing treatments, or mowing followed by reseeding and further mowing treatments, can reduce cogon grass . An Australian review recommends remowing before cogon grass flowers or reaches 12 inches (30 cm) in height .
Disking, or a combination of disking and other treatments, helps control cogon grass. Testing 11 populations in northern and central Florida, Shilling and others  found disking alone provided short-term control of cogon grass; however, disking and imazapyr application provided 96% control of cogon grass 1 year after treatments. Mowing in late spring to remove old growth and thatch, followed by disking 6 to 8 weeks later, when plants are 6 to 12 inches (20-30 cm) tall, was also effective . Plants should be disked before they reach 30 inches (75 cm) in height. A second disking is needed 2 to 3 weeks later if disking is the only control method used .
Another northern Florida study illustrates the importance of multiple treatments. One year after mechanical or mechanical/chemical treatments, 1 summer mowing and disking treatment had increased cogon grass density; 2 mowing and disking treatments provided "moderate" control; summer mowing and disking, fall imazapyr or glyphosate treatment, and a 2nd summer mowing and disking provided 100% control . In an Alabama study, Wilcut and others  found tilling cogon grass rhizomes to depths of 2 to 3 inches (5-8 cm) greatly reduced cogon grass sprouts. Rhizomes exposed to 23 °F to 25 °F (-5 °C to -4 °C) temperatures for 24 hours were also killed, suggesting that cold-weather tillage can maximize rhizome kill.
Small-farm operations in Asia commonly press or drum roll cogon grass stands, which breaks and lodges stems and leaves. In Indonesia, mechanical lodging is more effective than fire, mowing, or disking measures because cogon grass rhizomes are forced to grow through their own, artificially compacted, dense litter. Legume species tend to be better competitors for light after compaction treatments  (see Cultural control). Using roller/compactors, mechanical lodging may be effective as a part of an integrated control program on some sites.
Fire: See Fire Management Considerations.
Biological: There are no biological control agents currently approved for cogon grass control [116,134,148]. Cogon grass has several insects and fungal pathogens that infest it in Asia . Two fungal pathogens (Bipolaria sacchari and Drechslera gigantea) have shown potential as cogon grass control agents in greenhouse trials [177,178].
Cogon grass is widely used as livestock forage in Asia and Africa [40,41]. Domestic sheep grazing has been successfully used as a control measure in Malaysia ; however, there are no studies on using livestock to help control Brazilian satintail or cogon grass in the United States. Some ranchers in the Southeast use cogon grass swards as winter cattle range . Cattle weight gain is not as good on cogon grass compared to pasture grasses [40,41], but livestock grazing may be a useful part of an integrated Brazilian satintail/cogon grass control program if the primary management objective is weed control and not meat or milk production. Studies on how livestock use affects Brazilian satintail and cogon grass cover and reproduction are needed.
Chemical: Herbicides may provide initial control of a new invasion or a severe infestation, but used alone, they are rarely a complete or long-term solution to invasive species management . Herbicides are most effective on large infestations when incorporated into long-term management plans that include replacement of weeds with desirable species, careful land use management, and prevention of new infestations. Control with herbicides is temporary, as it does not change the conditions that allowed the invasion to occur (e.g. ). See The Nature Conservancy's Weed Control Methods Handbook for considerations on the use of herbicides in Natural Areas and detailed information on specific chemicals.
Imazapyr and glyphosate help control cogon grass [43,70,146,148,165,167]. Imazapyr is nonselective and has some soil residual activity. Glyphosate is also nonselective but is less residual, offering more flexibility in timing and species selection for posttreatment revegetation . Rhizomes must be killed for effective, long-lasting control [70,88,96]. Depending upon rhizome reserves, multiple herbicide applications and follow-up spot treatments are usually needed for complete rhizome kill and long-term control [26,96]. Fall applications are usually recommended because cogon grass transports carbohydrates and herbicides down into its rhizomes and roots at that time of year [70,88,96]. Shilling and others  found autumn applications of glyphosate or imazapyr provided better control than spring or summer applications. Winter applications can be effective if plants are still green. Analysis of rhizome total nonstructural carbohydrates (TNC) can show when carbohydrate allocation is directed toward rhizomes and therefore, when herbicide applications are likely to be most effective. In Florida, TNC content showed a small peak in December and January, and showed greatest gains in February and May. Rhizome TNC content was lowest in November . Twelve and sixteen months after Florida field trials, Gaffney  found December application was twice as effective as either September or January application. Imazapyr gave better control than glyphosate.
Follow-up herbicide applications in spring, prior to flowering, can suppress cogon grass seed production . Young plants can often be controlled using fewer applications and/or lower doses of herbicide than plants with well-developed rhizomes . Demers and Long  provide standard application rate recommendations for cogon grass. High concentrations of herbicides do not necessarily translate into higher rate of cogon grass kill compared to recommended rates of application. In a loblolly pine plantation study in Florida, Ramsey and others  report high rates of glyphosate or imazapyr actually inhibit translocation of the herbicide to rhizomes. They recommend half-dose applications, sprayed twice a year, to ensure good rhizome kill. Terry and others  found glyphosate application to newly burned cogon grasslands was not effective. Glyphosate is usually a foliar-applied and foliar-absorbed herbicide, and top-killed cogon grass lacked sufficient aboveground surface area to carry the herbicide to rhizomes .
Carefully timed herbicide treatments may control cogon grass for as long as 1 or 2 years, but without establishment of desirable native species, cogon grass can eventually reinfest treated areas [118,133] (see Cultural control below).
Altering soil nutrient status can increase the competitive ability of native species on sites with cogon grass. In a Mississippi study, addition of large amounts of phosphorus slowed rate of cogon grass invasion into longleaf pine savannas. Native plant species were seemingly not affected by the fertilizer. The mechanism of cogon grass inhibition was unknown, but the authors speculated that the addition of phosphorus could have caused changes to the soil microbial community, chemical changes in the soil, or a phosphorus toxicity reaction in cogon grass . Cogon grass forms mycorrhizal associations in Indonesia (Tjitrosemito and others 1994, as cited in  ); these associations help cogon grass obtain phosphorus, which is often limiting in tropical ecosystems. This competitive advantage is lost, however, where phosphorus is not limiting. Applying phosphorus on sites where it is already plentiful may help control cogon grass .Cultural control of cogon grass is effective, and is widely practiced in Asia and Africa [21,42,66]. Posttreatment revegetation gives best long-term control of cogon grass [23,43]. Because many legume species outcompete cogon grass for light and rapidly shade it out, legume plantings are particularly useful for cogon grass control . Managers and farmers in tropical countries successfully plant legume and other herbaceous species after using herbicides for cogon grass control [66,133,153]. In a Mississippi study, cogon grass invasion was negatively correlated with relative abundance of native legumes on unfertilized plots . Trees planted in cogon grass swards can shade cogon grass out when small areas around the trees are controlled: this method is used to control cogon grass in tropical orchards and plantations or to reclaim tropical forest that has converted to cogon grassland [15,42,66,67,83,84]. Fire may be excluded as part of the control method in orchards and plantations, using mechanical and herbicide controls on reinvading cogon grass [42,53].
On a phosphate mine site in central Florida, nonnative hairy indigo (Indigo hirsuta), nonnative perennial ryegrass (Lolium perenne), native bahia grass (Paspalum notatum), native switch grass (Panicum virgatum), or native partridge pea (Chamaecrista fasciculata) were planted after imazapyr or glyphosate treatments to control cogon grass. Imazapyr treatment followed by hairy indigo seedings gave best cogon grass control (100% control at posttreatment year 1), followed by imazapyr and switch grass or partridge pea plantings (86% control). Perennial ryegrass and bahia grass showed poor establishment . Shilling and others  found a significantly lower (p<0.001) density of cogon grass establishment in fields planted to bahia grass. Presence of bahia grass lowered cogon grass seedling density by 25% and lowered cogon grass ramet density by 37%. A combination of herbicides and revegetating with Bermuda grass and hairy indigo also gave good control of cogon grass 2 years after posttreatment seedings .Cultural plantings in cogon grass swards that have been burned but not otherwise treated does not provide control. Barnard  calls it "a failure and a waste of seed."
1. Agrawal, Arun K. 1990. Floristic composition and phenology of temperate grasslands of western Himalaya as affected by scraping, fire and heavy grazing. Vegetatio. 88: 177-187. 
2. Ahmad Faiz, M. A. 1998. Effects of herbicide mixtures, surfactants and spray volumes on the control of Imperata cylindrica (L.) Raeuschel. Journal of Rubber Research. 1(3): 179-189. 
3. Allen, Charles M.; Thomas, R. Dale. 1991. Brachiaria plantaginea, Imperata cylindrica, and Panicum maximum: three grasses (Poaceae) new to Louisiana and a range extension for Rottboellia cochinchinensis. SIDA. 14(4): 613. 
4. Allen, J. A.; Keeland, B. D.; Stanturf, J. A.; [and others]. 2001. A guide to bottomland hardwood restoration. Information and Technology Report USGS/BRD/ITR--2000-0011; Gen. Tech. Rep. SRS-40. Reston, VA: U.S. Department of the Interior, Geological Survey; Washington, DC: U.S. Department of Agriculture, Forest Service. 132 p. 
5. Anon. 1976. Effective and efficient control of Imperata cylindrica (lalang). Planters' Bulletin. 145: 83-84. 
6. Avav, T. 2000. Control of speargrass (Imperata cylindrica (L) Raeuschel) with glyphosate and fluazifop-butyl for soybean (Glycine max (L) Merr) production in savanna zone of Nigeria. Journal of the Science of Food and Agriculture. 80: 193-196. 
7. Ayeni, A. O. 1985. Observations on the vegetative growth pattern of speargrass (Imperata cylindrica (L.) Beauv.). Agriculture, Ecosystems and Environment. 13(3/4): 301-307. 
8. Ayeni, A. O.; Duke, W. B. 1985. The influence of rhizome features on subsequent regenerative capacity in speargrass (Imperata cylindrica (L.) Beauv.). Agriculture, Ecosystems and Environment. 13(3/4): 309-317. 
9. Barkworth, Mary E.; Capels, Kathleen M.; Long, Sandy; Piep, Michael B., eds. 2003. Flora of North America north of Mexico. Volume 25: Magnoliophyta: Commelinidae (in part): Poaceae, part 2. New York: Oxford University Press. 783 p. Available online: http://herbarium.usu.edu/webmanual/. 
10. Barnard, R. C. 1954. The control of lalang (Imperata arundinacea var. major) by fire protection and planting. Malaysian Forester. 17(3): 152-156. 
11. Bernard, Stephen R.; Brown, Kenneth F. 1977. Distribution of mammals, reptiles, and amphibians by BLM physiographic regions and A.W. Kuchler's associations for the eleven western states. Tech. Note 301. Denver, CO: U.S. Department of the Interior, Bureau of Land Management. 169 p. 
12. Boonitee, A.; Ritdhit, P. 1984. Allelopathic effects of some weeds on mungbean plants. In: Proceedings, 1st Pacific Weed Science Society conference; [Date of conference unknown]; Songkla, Thailand. 2: 401-406. 
13. Bowman, D. M. J. S. 1993. Evidence for gradual retreat of dry monsoon forests under a regime of aboriginal burning, Karslake Peninsula, Mellville Island, northern Australia. Proceedings of the Royal Society of Queensland. 102: 25-30. 
14. Brewer, J. Stephen; Cralle, Sean P. 2003. Phosphorus addition reduces invasion of a longleaf pine savanna (southeastern USA) by a non-indigenous grass (Imperata cylindrica). Plant Ecology. 167(2): 237-245. 
15. Brook, R. M. 1989. Review of literature on Imperata cylindrica (L.) Raeuschel with particular reference to South East Asia. Tropical Pest Management. 35(1): 12-25. 
16. Brooks, Matthew L.; D'Antonio, Carla M.; Richardson, David M.; Grace, James B.; Keeley, Jon E.; DiTomaso, Joseph M.; Hobbs, Richard J.; Pellant, Mike; Pyke, David. 2004. Effects of invasive alien plants on fire regimes. BioScience. 54(7): 677-688. 
17. Bryson, Charles T.; Carter, Richard. 1993. Cogongrass, Imperata cylindrica, in the United States. Weed Technology. 7(4): 1005-1009. 
18. Bussan, Alvin J.; Dyer, William E. 1999. Herbicides and rangeland. In: Sheley, Roger L.; Petroff, Janet K., eds. Biology and management of noxious rangeland weeds. Corvallis, OR: Oregon State University Press: 116-132. 
19. Calub, A. D.; Anwarhan, H.; Roder, W. 1997. Livestock production systems for Imperata grasslands. Agroforestry Systems. 36(1-3): 121-128. 
20. Casini, P.; Vecchio, V.; Tamantini, I. 1998. Allelopathic interference of itchgrass and cogongrass: germination and early development of rice. Tropical Agriculture. 75(4): 445-451. 
21. Chikoye, D.; Ekeleme, F. 2001. Growth characteristics of ten Mucuna accessions and their effects on the dry matter of Imperata cylindrica (L.) Rauesch. Biological Agriculture and Horticulture. 18: 191-201. 
22. Chikoye, D.; Manyong, V. M.; Ekeleme, F. 2000. Characteristics of speargrass (Imperata cylindrica) dominated fields in West Africa: crops, soil properties, farmer perceptions and management strategies. Crop Protection. 19: 481-487. 
23. Chikoye, D; Manyong, V. M.; Carsky, R. J.; Ekeleme, F.; Gbehounou, G.; Ahanchede, A. 2002. Response of speargrass (Imperata cylindrica) to cover crops integrated with handweeding and chemical control in maize and cassava. Crop Protection. 21: 145-156. 
24. Christensen, Norman L. 1981. Fire regimes in southeastern ecosystems. In: Mooney, H. A.; Bonnicksen, T. M.; Christensen, N. L.; Lotan, J. E.; Reiners, W. A., technical coordinators. Fire regimes and ecosystem properties: Proceedings of the conference; 1978 December 11-15; Honolulu, HI. Gen. Tech. Rep. WO-26. Washington, DC: U.S. Department of Agriculture, Forest Service: 112-136. 
25. Clewell, Andre F. 1985. Guide to the vascular plants of the Florida Panhandle. Tallahassee, FL: Florida State University Press. 605 p. 
26. Coile, Nancy C.; Shilling, Donn G. 1993. Cogongrass, Imperata cylindrica (L.) Beauv.: a good grass gone bad! Botany Circular No. 28. Gainesville, FL: Florida Department of Agriculture and Consumer Services, Division of Plant Industry. 3 p. 
27. Cole, James T.; Cole, Janet C. 2000. Ornamental grass growth response to three shade intensities. Journal of Environmental Horticulture. 18(1): 18-22. 
28. Coster, Ir. Ch. 1932. Some observations on the growth of "alang-alang" (Imperata cylindrica Beauv) and its examination. Tectona: Forest Research Institute. No. 26. 23 p. [English translation prepared by: Saad Publications, Translation Division No. 31458, Karachi, Pakistan; 1982]. 
29. D'Antonio, Carla M.; Vitousek, Peter M. 1992. Biological invasions by exotic grasses, the grass/fire cycle, and global change. Annual Review of Ecology and Systematics. 23: 63-87. 
30. Dano, Antonio M. 1990. Effect of burning and reforestation on grassland watersheds in the Philippines. In: Ziemar, R. R.; O'Loughlin, C. L.; Hamilton, L. S., eds. Research needs and applications to reduce erosion and sedimentation in tropical steepland: Proceedings of the Fiji symposium; 1990 June; [Meeting location unknown]. IAHS-AISH Publ. No. 192. [Publisher location unknown]: International Association of Hydrological Sciences: 53-61. 
31. Demers, Chris; Long, Alan. 2002. Controlling invasive exotic plants in north Florida forests. SS-FOR19. Gainesville, FL: University of Florida, Institute of Food and Agricultural Sciences, Florida Cooperative Extension Service. 9 p. Available online: http://edis.ifas.ufl.edu/pdffiles/FR/FR13300.pdf [2005, March 28]. 
32. Dickens, Ray. 1974. Cogongrass in Alabama after sixty years. Weed Science. 22(2): 177-179. 
33. Dickens, Ray; Buchanan, G. A. 1975. Control of cogongrass with herbicides. Weed Science. 23(3): 194-197. 
34. Dickens, Ray; Moore, G. M. 1974. Effects of light, temperature, KNO3, and storage on germination of cogongrass. Agronomy Journal. 66(2): 187-188. 
35. Dozier, Hallie; Gaffney, James F.; McDonald, Sandra K.; Johnson, Eric R. R. L.; Shilling, Donn G. 1998. Cogongrass in the United States: history, ecology, impacts, and management. Weed Technology. 12(4): 737-743. 
36. Ershad, D.; Abbasi, M. 1992. Studies in the rust fungi of Iran. Iranian Journal of Plant Pathology. 28: 23-26. 
37. Eussen, J. H. H. 1980. Biological and ecological aspects of alang-alang. In: In: Proceedings of Biotrop workshop on alang-alang; 1976 July 27-29; Bogor, Indonesia. Biotrop Special Publication No. 5. [Publisher location unknown]: [Publisher unknown]: 15-22. On file with: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT; RWU 4403 files. 
38. Eussen, Jacobus H. H.; Wirjahardja, Soemantri. 1973. Studies of an alang-alang (Imperata cylindrica (L.) Beauv.) vegetation. Biotrop Bulletin. 6: 2-24. 
39. Eyre, F. H., ed. 1980. Forest cover types of the United States and Canada. Washington, DC: Society of American Foresters. 148 p. 
40. Falvey, J. Lindsay; Hengmichai, Prakob; Pongpiachan, Puntipa. 1981. The productivity and nutritive value of Imperata cylindrica (L) Beauv. in the Thai highlands. Journal of Range Management. 34(4): 280-282. 
41. Flavey, J. L.; Gibson, T. A.; Andrews, A. C. 1984. Animal production from improved pastures in the Thai highlands. World Animal Review. 49: 13-18. 
42. Florece, Leonardo M.; Espaldon, Ma. Victoria; Galang, Celerino. 1997. Fire management, fire tolerance and biodiversity enhancement of grassland ecosystem: the use of Gliricidia sepium stem cuttings as a reforestation species. Imperata Project Paper 1997/11. Canberra, Australia: Australian National University, Centre for Resource and Environmental Studies. 14 p. 
43. Gabel, Mark Lauren. 1982. A biosystematic study of the genus Imperata (Gramineae: Andropogoneae). Ames, IA: Iowa State University. 90 p. Dissertation. 
44. Gaffney, James Frank. 1996. Ecophysiological and technological factors influencing the management of cogongrass (Imperata cylindrica). Gainesville, FL: University of Florida. 114 p. Dissertation. 
45. Garrison, George A.; Bjugstad, Ardell J.; Duncan, Don A.; Lewis, Mont E.; Smith, Dixie R. 1977. Vegetation and environmental features of forest and range ecosystems. Agric. Handb. 475. Washington, DC: U.S. Department of Agriculture, Forest Service. 68 p. 
46. Garrity, D. P.; Soekardi, M.; van Noordwijk, M.; de la Cruz, R.; Pathak, P. S.; Gunasena, H. P. M.; So, N. van; Huijun, G.; Majid, N. M. 1997. The Imperata grasslands of tropical Asia: area, distribution, and typology. Agroforestry Systems. 36(1-3): 3-29. 
47. Georgia Exotic Pest Plant Council. 2002. Proposed exotic pest plant species for Georgia, [Online]. Southeast Exotic Pest Plant Council (Producer). Available: http://www.gaeppc.org/exotalk1.html [2003, August 25]. 
48. Ghosal, Shibnath; Kumar, Yatendra; Chakrabarti, Dilip K.; Lal, Jawahar; Singh, Sushil K. 1986. Parasitism of Imperata cylindrica on Pancratium biflorum and the concomitant chemical changes in the host species. Phytochemistry. 25(5): 1097-1102. 
49. Goldammer, J. G.; Penafiel, S. R. 1990. Fire in the pine-grassland biomes of tropical and subtropical Asia. In: Goldammer, J. G., ed. Fire in the tropical biota: ecosystem processes and global challenges. Berlin: Springer-Verlag: 53-64. 
50. Grace, James B.; Smith, Melinda D.; Grace, Susan L.; Collins, Scott L.; Stohlgren, Thomas J. 2001. Interactions between fire and invasive plants in temperate grasslands of North America. In: Galley, Krista E. M.; Wilson, Tyrone P., eds. Proceedings of the invasive species workshop: The role of fire in the control and spread of invasive species; Fire conference 2000: the first national congress on fire ecology, prevention, and management; 2000 November 27 - December 1; San Diego, CA. Misc. Publ. No. 11. Tallahassee, FL: Tall Timbers Research Station: 40-65. 
51. Grist, Peter; Menz, Ken. 1996. Burning in an Imperata fallow/upland rice farming system. Imperata Project Paper 1996/7. Canberra, Australia: Australian National University, Centre for Resource and Environmental Studies. 17 p. 
52. Grist, Peter; Menz, Ken. 1997. On-site effects of Imperata burning by Indonesian smallholders: a bioeconomic model. Bulletin of Indonesian Economic Studies. 33(3): 79-96. 
53. Grist, Peter; Menz, Ken. 1999. Evaluation of fire versus non-fire methods for clearing Imperata fallow. In: Menz, K.; Magcale-Macandog, D.; Rusastra, I. W., eds. Improving smallholder farming systems in Imperata areas of southeast Asia: alternatives to shifting cultivation. ACIAR Monograph Series No. 52. Canberra, Australia: Australian Centre for International Agricultural Research: 25-34. 
54. Grodzki, Leocadio; Boeger, Maria Regina Torres. 2001. Characterization of the pioneer vegetation on the bracatinga (Mimosa scabrella Benth.) agroforestry system in the Colombo municipality, PR. Floresta. 31(1-2): 93-98. 
55. Haddad, Claudia R. B.; Valio, I. F. M. 1993. Effect of fire on flowering of Lantana montevidensis Briq. Journal of Plant Physiology. 141: 704-707. 
56. Hakansson, Sigurd. 1982. Multiplication, growth and persistence of perennial weeds. In: Holzner, W.; Numata, M., eds. Biology and ecology of weeds. The Hague: Dr. W. Junk: 123-135. 
57. Hall, David W. 1978. The grasses of Florida. Gainesville, FL: University of Florida. 498 p. Dissertation. 
58. Hartemink, Alfred E. 2001. Biomass and nutrient accumulation of Piper aduncum and Imperata cylindrica fallows in the humid lowlands of Papua New Guinea. Forest Ecology and Management. 144: 19-32. 
59. Hartemink, Alfred E.; O'Sullivan, J. N. 2001. Leaf litter decomposition of Piper aduncum, Gliricidia sepium and Imperata cylindrica in the humid lowlands of Papua New Guinea. Plant and Soil. 230(1): 115-124. 
60. Hennessy, D. W.; McLennan, D. J.; Williamson, P. J.; Morris, S. G. 1998. Changes in characteristics of pastures in the coastal subtropics when grazed by cattle during years of low rainfall. Australian Journal of Experimental Agriculture. 38(8): 813-820. 
61. Hitchcock, A. S. 1951. Manual of the grasses of the United States. Misc. Publ. No. 200. Washington, DC: U.S. Department of Agriculture, Agricultural Research Administration. 1051 p. [2nd edition revised by Agnes Chase in two volumes. New York: Dover Publications, Inc.]. 
62. Hobbs, Richard J.; Huenneke, Laura F. 1992. Disturbance, diversity, and invasion: implications for conservation. Conservation Biology. 6(3): 324-337. 
63. Holm, LeRoy G.; Plocknett, Donald L.; Pancho, Juan V.; Herberger, James P. 1977. The world's worst weeds: distribution and biology. Honolulu, HI: University Press of Hawaii. 609 p. 
64. Hubbard, C. E. 1944. Imperata cylindrica: Taxonomy, distribution, economic significance and control. Imperial Agricultural Bureaux Joint Publication No. 7. Oxford, UK: Imperial Forestry Bureau; Aberystwyth, UK: Imperial Bureau of Pastures and Forage Crops. 63 p. 
65. Inderjit; Dakshini, K. M. M. 1991. Investigations on some aspects of chemical ecology of cogongrass, Imperata cylindrica (L.) Beauv. Journal of Chemical Ecology. 17(2): 343-352. 
66. Ivens, G. W. 1983. The natural control of Imperata cylindrica: Nigeria and northern Thailand. Mountain Research and Development. 3(4): 372-377. 
67. Jackson, Rudolph E. A. 1989. The effect of Imperata cylindrica on seed germination and seedling growth of Leucaena leucocephala and Gliricidia sepium. Nitrogen Fixing Tree Research Reports. 7: 28-29. 
68. Jacoby, P. W.; Ansley, R. J.; Trevino, B. A. 1992. Technical note: an improved method for measuring temperatures during range fires. Journal of Range Management. 45(2): 216-220. 
69. Johnson, E. R. R. L.; Gaffney, J. F.; Shilling, D. G. 1999. The influence of discing on the efficacy of imazapyr for cogongrass [Imperata cylindrica (L.) Beauv.] control. In: A glance to the past, a vision for the future: Proceedings of the 52nd annual meeting of the Southern Weed Science Society; 1999 January 25-27; Greensboro, NC. In: Proceedings, Southern Weed Science Society. Raleigh, NC: Southern Weed Science Society; 52: 165. Abstract. 
70. Johnson, Eric R. R. L.; Shilling, Donn G. 2004. Fact sheet: Cogon grass--Imperata cylindrica (L.) Palisot, [Online]. In: Weeds gone wild: Alien plant invaders of natural areas. Plant Conservation Alliance, Alien Plant Working Group (Producer). Available: http://www.nps.gov/plants/alien/fact/imcyl.htm [2005, March 28]. 
71. Jose, Shibu; Cox, Joseph; Miller, Deborah L.; [and others]. 2002. Alien plant invasion: The story of cogongrass in southeastern forests. Journal of Forestry. 100(1): 41-44. 
72. Kartesz, John T. 1999. A synonymized checklist and atlas with biological attributes for the vascular flora of the United States, Canada, and Greenland. 1st ed. In: Kartesz, John T.; Meacham, Christopher A. Synthesis of the North American flora (Windows Version 1.0), [CD-ROM]. Chapel Hill, NC: North Carolina Botanical Garden (Producer). In cooperation with: The Nature Conservancy; U.S. Department of Agriculture, Natural Resources Conservation Service; U.S. Department of the Interior, Fish and Wildlife Service. 
73. Kearl, Leonard C. 1982. Nutrient requirements of ruminants in developing countries. Logan, UT: Utah State University, Agricultural Experiment Station, International Feedstuffs Institute. 381 p. 
74. King, Sharon E.; Grace, James B. 2000. The effects of gap size and disturbance type on invasion of wet pine savanna by cogongrass, Imperata cylindrica (Poaceae). American Journal of Botany. 87(9): 1279-1286. 
75. King, Sharon E.; Grace, James B. 2000. The effects of soil flooding on the establishment of cogongrass (Imperata cylindrica), a nonindigenous invader of the southeastern United States. Wetlands. 20(2): 300-306. 
76. Koger, Clifford H.; Bryson, Charles T. 2004. Effect of cogongrass (Imperata cylindrica) extracts on germination and seedling growth of selected grass and broadleaf species. Weed Technology. 18(2): 236-242. 
77. Koger, Clifford H.; Bryson, Charles T.; Byrd, John D., Jr. 2004. Response of selected grass and broadleaf species to cogongrass (Imperata cylindrica) residues. Weed Technology. 18(2): 353-357. 
78. Koskela, J.; Kuusipalo, J.; Sirikul, W. 1995. Natural regeneration dynamics of Pinus merkusii in northern Thailand. Forest Ecology and Management. 77: 169-179. 
79. Kuchler, A. W. 1964. United States [Potential natural vegetation of the conterminous United States]. Special Publication No. 36. New York: American Geographical Society. 1:3,168,000; colored. 
80. Kushwaha, S. P. S.; Ramakrishnan, P. S.; Tripathi, R. S. 1981. Population dynamics of Eupatorium odoratum in successional environments following slash and burn agriculture. Journal of Applied Ecology. 18(2): 529-535. 
81. Kushwaha, S. P. S.; Ramakrishnan, P. S.; Tripathi, R. S. 1983. Competitive relationships of the plants of Imperata cylindrica established from rhizomes and from seeds. Tropical Plant Science Research. 1(1): 53-57. 
82. Kushwaha, S. P. S.; Ramakrishnan, P. S.; Tripathi, R. S. 1983. Population dynamics of Imperata cylindrica (L.) Beauv. var. major related to slash and burn agriculture (jhum) in north eastern India. Proceedings of the Indian Academy of Sciences. 92(4): 313-321. 
83. Kuusipalo, Jussi; Adjers, Goran; Yusuf, Jafarsidik; Otsamo, Antti; Tuomela, Kari; Vuokko, Risto. 1995. Restoration of natural vegetation in degraded Imperata cylindrica grassland: understorey development in forest plantations. Journal of Vegetation Science. 6(2): 205-210. 
84. Kuusipalo, Jussi; Hadi, Tjuk S.; Otsamo, Antti. 1995. Environmental restoration through reforestation of Imperata cylindrica grasslands. FRI Bulletin Rotorua. 192: 213-214. 
85. Liangzhong, Zeng; Whelan, Robert J. 1993. Natural reforestation of abandoned farmland: the role of soils. Australian Geographer. 24(2): 14-25. 
86. Lippincott, Carol L. 1997. Ecological consequences of Imperata cylindrica (cogongrass) invasion in Florida sandhill. Gainesville, FL: University of Florida. 165 p. Dissertation. 
87. Lippincott, Carol L. 2000. Effects of Imperata cylindrica (L.) Beauv. (Cogongrass) invasion on fire regime in Florida Sandhill (USA). Natural Areas Journal. 20(2): 140-149. 
88. Lippincott, Carol; McDonald, Sandra. 1996. Imperata cylindrica. In: Randall, John M.; Marinelli, Janet, eds. Invasive plants: Weeds of the global garden. Handbook #149. Brooklyn, NY: Brooklyn Botanic Garden: 88. 
89. Maruyama, K.; Miura, S. 1981. Studies on the soil-vegetation system in the west Niigata coastal sand dune, with special reference to the comparison of affected and controlled areas by wind blown sand. In: Bulletin of the Niigata University Forests. No. 14. Niigata, Japan: The Niigata University Forests: 43-78. 
90. Matlack, Glenn R. 2002. Exotic plant species in Mississippi, USA: critical issues in management and research. Natural Areas Journal. 22(3): 241-247. 
91. Matsunaga, Kimihiro; Ikeda, Masato; Shibuya, Masaoki; Ohizumi, Yasushi. 1994. Cylindol A, a novel biphenyl ether with 5-lipoxygenase inhibitory activity, and a related compound from Imperata cylindrica. Journal of Natural Products. 57(9): 1290-1293. 
92. Matsunaga, Kimihiro; Shibuya, Masaoki; Ohizumi, Yasushi. 1994. Cylindrene, a novel sesquiterpenoid from Imperata cylindrica with inhibitory activity on contractions of vascular smooth muscle. Journal of Natural Products. 57(8): 1183-1184. 
93. Matsunaga, Kimihiro; Shibuya, Masaoki; Ohizumi, Yasushi. 1995. Imperanene, a novel phenolic compound with platelet aggregation inhibitory activity from Imperata cylindrica. Journal of Natural Products. 58(1): 138-139. 
94. McDonald, S. K.; Shilling, D. G.; Okoli, C. A. N.; Bewick, T. A.; Gordon, D.; Hall, D.; Smith, R. 1996. Population dynamics of cogongrass, Imperata cylindrica. In: In: Weed science meets the press; 1996 January 15-17; Charlotte, NC. In: Proceedings, Southern Weed Science Society. Champagne, IL: Southern Weed Science Society: 49: 156. Abstract. 
95. Meguro, M. 1969. Fatores que regulam a floracao em Imperata brasiliensis Trin. floracao em Imperata brasiliensis Trin [Gramineae). Ciencias e Letra da Universidade de Sao Paulo: Serie Botanica. 24: 103-106. [Boletim da Faculdade de Filosofia]. 
96. Miller, James H. 2003. Nonnative invasive plants of southern forests: A field guide for identification and control. Gen. Tech. Rep. SRS-62. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 93 p. Available: hhtp://www.srs.fs.usda.gov/pubs/gtr/gtr_srs062/ [2004, December 10]. 
97. Mishra, B. K.; Ramakrishnan, P. S. 1983. Secondary succession subsequent to slash and burn agriculture at higher elevations of north-east India. I. -- Species diversity, biomass and litter production. Acta Ecologica. 4(2): 95-107. 
98. Mitchell, B. A. 1964. Periodical cropping of Imperata cylindrica for paper pulp. Malaysian Forester. 27(1): 22-45. 
99. Mohsen, Zohair H.; Jawad, Abdul-Latif M.; Al-Saadi, May; Al-Naib, Ala. 1995. Anti-oviposition and insecticidal activity of Imperata cylindrica (Gramineae). Medical and Veterinary Entomology. 9(4): 441-442. 
100. Murniati. 2002. From Imperata cylindrica grasslands to productive agroforestry. Aula, Wageningen: Wageningen University. 172 p. Dissertation. 
101. Myers, Ronald L. 1990. Scrub and high pine. In: Myers, Ronald L.; Ewel, John J., eds. Ecosystems of Florida. Orlando, FL: University of Central Florida Press: 150-193. 
102. Myers, Ronald L. 2000. Fire in tropical and subtropical ecosystems. In: Brown, James K.; Smith, Jane Kapler, eds. Wildland fire in ecosystems: Effects of fire on flora. Gen. Tech. Rep. RMRS-GTR-42-vol. 2. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 161-173. 
103. Nykvist, Nils. 1996. Regrowth of secondary vegetation after the `Borneo fire' of 1982-1983. Journal of Tropical Ecology. 12(2): 307-312. 
104. Ohtsuka, Toshiyuki. 1999. Early stages of secondary succession on abandoned cropland in north-east Borneo Island. Ecological Research. 14: 281-290. 
105. Ohwi, Jisaburo. 1965. Flora of Japan. Washington, DC: Smithsonian Institution. 1067 p. 
106. Patterson, D. T.; Flint, E. P.; Dickens, Ray. 1980. Effects of temperature, photoperiod, and population source on the growth of cogongrass (Imperata cylindrica). Weed Science. 28(5): 505-509. 
107. Patterson, D. T.; McWhorter, C. G. 1980. Distribution and control of cogongrass (Imperata cylindrica) in Mississippi. In: In: Proceedings, 33rd annual meeting of the Southern Weed Science Society; 1980 January 15-17; Hot Springs, AR. [Champaign, IL]: Southern Weed Science Society: [Page unknown]. Abstract. 
108. Peet, Nicholas B.; Watkinson, Andrew R.; Bell, Diana J.; Sharma, Uday R. 1999. The conservation management of Imperata cylindrica grassland in Nepal with fire and cutting: an experimental approach. Journal of Applied Ecology. 36: 374-387. 
109. Peet, Robert K.; Allard, Dorothy J. 1993. Longleaf pine vegetation of the southern Atlantic and Gulf Coast regions: a preliminary classification. In: Hermann, Sharon M., ed. The longleaf pine ecosystem: ecology, restoration and management: Proceedings, 18th Tall Timbers fire ecology conference; 1991 May 30 - June 2; Tallahassee, FL. Tallahassee, FL: Tall Timbers Research, Inc: 45-81. 
110. Pendleton, Robert L. 1948. Cogon grass, Imperata cylindrica, in the western hemisphere. Journal of the American Society of Agronomy. 40(11): 1047-1049. 
111. Pickford, Stewart; Suharti, Mieke; Wibowo, Ari. 1992. A note on fuelbeds and fire behavior in alang-alang (Imperata cylindrica). International Journal of Wildland Fire. 2(1): 41-46. 
112. Pinilla, Veronique; Luu, Bang. 1999. Isolation and partial characterization of immunostimulating polysaccharides from Imperata cylindrica. Planta Medica. 65: 549-552. 
113. Platt, William J.; Gottschalk, Robert M. 2001. Effects of exotic grasses on potential fine fuel loads in the groundcover of south Florida slash pine savannas. International Journal of Wildland Fire. 10: 155-159. 
114. Potter, Lesley M. 2002. Forests and grassland, drought and fire: the island of Borneo in the historical environmental record (post-1800). Advances in GeoEcology. 34: 339-356. 
115. Pressland, A. J. 1982. Fire in the management of grazing lands in Queensland. Tropical Grasslands. 16(3): 104-112. 
116. Pusat Penelitian Karet; Natural Resources Institute. 1996. Imperata management for smallholders: an extensionist's guide to rational Imperata management for smallholders. [Jakarta] Indonesia: Indonesian Rubber Research Institute, Sembawa Research Station; United Kingdom: Natural Resources Institute. 56 p. 
117. Rajan, A.; Lovett, J. V.; Andrews, A. C. 1988. Phytotoxic activity in lalang (Imperata cylindrica) rhizosphere in relation to growth of the plant. Journal of Plant Protection in the Tropics. 5(2): 115-125. 
118. Ramsey, Craig L.; Jose, Shibu; Miller, Deborah L.; Cox, Joseph; Portier, Kenneth M.; Shilling, Donald G.; Merritt, Sara. 2003. Cogongrass [Imperata cylindrica (L.) Beauv.] response to herbicides and disking on a cutover site and in a mid-rotation pine plantation in southern USA. Forest Ecology and Management. 179: 195-207. 
119. Raunkiaer, C. 1934. The life forms of plants and statistical plant geography. Oxford: Clarendon Press. 632 p. 
120. Rawat, Gopal S.; Wikramanayake, Eric D. 2001. Himalayan subtropical pine forests (IM0301), [Online]. In: Terrestrial ecoregions of the world. World Wildlife Fund (Producer). Available: http://www.worldwildlife.org/wildworld/profiles/terrestrial/im/im0301_full.html [2005, March 28]. 
121. Raymond, Oliver. 1992. Letter to the editor. International Journal of Wildland Fire. 2(3): [Pages unknown]. 
122. Riswan, Soedarsono; Hartanti, Lies. 1995. Human impacts on tropical forest dynamics. Vegetatio. 121: 41-52. 
123. Robbins, Louise E.; Myers, Ronald L. 1992. Seasonal effects of prescribed burning in Florida: a review. Misc. Publ. No. 8. Tallahassee, FL: Tall Timbers Research, Inc. 96 p. 
124. Sajise, Percy Eres. 1972. Evaluation of cogon (Imperata cylindrica (L.) Beauv.) as a seral stage in Philippine vegetational succession: I. The cogonal seral stage and plant succession; II. Autecological studies on cogon. Ithaca, NY: Cornell University. 152 p. Dissertation. 
125. Santiago, A. 1966. Studies in the autecology of Imperata cylindrica (L.) Beauv. In: Proceedings of the 9th international grassland conference; [Meeting date unknown]; Sao Paulo, Brazil. Sao Paulo, Brazil: [Agricultural Secretariat]: 499-502. On file with: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT. 
126. Saxena, K. G.; Ramakrishnan, P. S. 1983. Growth and allocation strategies of some perennial weeds of slash and burn agriculture (Jhum) in northeastern India. Canadian Journal of Botany. 61: 1300-1306. 
127. Saxena, K. G.; Ramakrishnan, P. S. 1984. Herbaceous vegetation development and weed potential in slash and burn agriculture (Jhum) in N. E. India. Weed Research. 24: 135-142. 
128. Scott, Geoffrey A. J. 1977. The role of fire in the creation and maintenance of savanna in the Montana of Peru. Journal of Biogeography. 4: 143-167. 
129. Shaltout, K. H.; El-Sheikh, M. A. 1993. Vegetation-environment relations along water courses in the Nile Delta region. Journal of Vegetation Science. 4: 567-570. 
130. Sharma, Brij M.; Okafor, Augustine N. 1987. Contribution to the ecology of speargrass (Imperata cylindrica (L.) P. Beauv.). Ekologia Polska. 35(3-4): 767-774. 
131. Sheley, Roger; Manoukian, Mark; Marks, Gerald. 1999. Preventing noxious weed invasion. In: Sheley, Roger L.; Petroff, Janet K., eds. Biology and management of noxious rangeland weeds. Corvallis, OR: Oregon State University Press: 69-72. 
132. Shiflet, Thomas N., ed. 1994. Rangeland cover types of the United States. Denver, CO: Society for Range Management. 152 p. 
133. Shilling, Donn G.; Bewick, T. A.; Gaffney, J. F.; McDonald, S. K.; Chase, C. A.; Johnson, E. R. R. L. 1997. Ecology, physiology, and management of cogongrass (Imperata cylindrica). Publication No. 03-107-140. Gainesville, FL: University of Florida. 128 p. 
134. Singh, C. M.; Angiras, N. N.; Kumar, Suresh. 1993. Perennial weed management in non-cropped situations. In: Proceedings of the Indian Society of Weed Science international symposium; 1993 November 18-20; Hisar, India. In: Proceedings, Indian Society of Weed Science. Hisar, India: Indian Society of Weed Science; 1: 379-387. 
135. Snyder, James R.; Herndon, Alan; Robertson, William B., Jr. 1990. South Florida rockland. In: Myers, Ronald L.; Ewel, John J., eds. Ecosystems of Florida. Orlando, FL: University of Central Florida Press: 230-274. 
136. Soerjani, Mohamad. 1970. Alang-alang, Imperata cylindrica (L.) Beauv. (1812), pattern of growth as related to its problem of control. Biotrop Bulletin. 1: 4-87. 
137. Stanturf, J. A.; Conner, W. H.; Gardiner, E. S.; Schweitzer, C. J.; Ezell, A. W. 2004. Recognizing and overcoming difficult site conditions for afforestation of bottomland hardwoods. Ecological Restoration. 22(3): 183-193. 
138. Stickney, Peter F. 1989. Seral origin of species comprising secondary plant succession in Northern Rocky Mountain forests. FEIS workshop: Postfire regeneration. Unpublished draft on file at: U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Fire Sciences Laboratory, Missoula, MT. 10 p. 
139. Stober, S. 1993. Weed control by integration of sheep in permanent tree crops in West Malaysia. In: Thomas, J. M., ed. International federation of organic agriculture movements: International conference proceedings; 1993 July 5-9; Dijon, France. Quetigny Cedex, France: Association Colloque IFOAM: 213-218. 
140. Streng, Donna R.; Glitzenstein, Jeff S.; Platt, William J. 1993. Evaluating effects of season of burn in longleaf pine forests: a critical literature review and some results from an ongoing long-term study. In: Hermann, Sharon M., ed. The longleaf pine ecosystem: ecology, restoration and management: Proceedings, 18th Tall Timbers fire ecology conference; 1991 May 30 - June 2; Tallahassee, FL. No. 18. Tallahassee, FL: Tall Timbers Research, Inc: 227-263. 
141. Supriana, Nana; Ruswandy, Hamdan. 1986. Effect of forest fire on undergrowth species composition. A case study in the Simincak Forest Complex, South Tapanuli. Forestry Research Bulletin. 1(2): 1-8. 
142. Tabor, Paul. 1949. Cogon grass, Imperata cylindrica (L) Beauv., in the southeastern United States. Agronomy Journal. 41: 270. 
143. Tabor, Paul. 1952. Comments on cogon and torpedo grasses: a challenge to weed workers. Weeds. 1: 374-375. 
144. Tanimoto, Takeo. 1981. Vegetation of the alang-alang grassland and its succession in the Benakat District of South Sumatra, Indonesia. Bulletin of the Forestry and Forest Products Research Institute. 314: 11-19. 
145. Taylor, Dale L. 1980. Fire history and man-induced fire problems in subtropical south Florida. In: Stokes, Marvin A.; Dieterich, John H., technical coordinators. Proceedings of the fire history workshop; 1980 October 20-24; Tucson, AZ. Gen. Tech. Rep. RM-81. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station: 63-68. 
146. Terry, P. J.; Adjers, G.; Akobundu, I. O.; Anoka, A. U.; Drilling, M. E.; Tjitrosemito, S.; Utomo, M. 1997. Herbicides and mechanical control of Imperata cylindrica as a first step in grassland rehabilitation. Agroforestry Systems. 36(1-3): 151-179. 
147. Tu, Mandy. 2002. Weed notes: Imperata cylindrica `Red Baron' (Japanese blood grass), [Online]. In: The invasive species initiative--invasives news. The Nature Conservancy (Producer). Available: http://tncweeds.ucdavis.edu/moredocs/impcyl01.pdf [2005, April 22]. 
148. Tu, Mandy; Hurd, Callie; Randall, John M., eds. 2001. Weed control methods handbook: tools and techniques for use in natural areas. Davis, CA: The Nature Conservancy. 194 p. 
149. U.S. Department of Agriculture, Animal and Plant Health Inspection Service. 2004. APHIS: Federal noxious weed list, [Online]. In: Pest detection and management programs: Noxious weeds. Plant Protection and Quarantine (Producer). Available: http://www.aphis.usda.gov/ppq/weeds [2005, September 14]. 
150. U.S. Department of Agriculture, Forest Service, Southern Region. 2001. Regional invasive exotic plant species list, [Online]. In: Regional Forester's list and ranking structure: invasive exotic plant species of management concern. In: Invasive plants of southern states list. Southeast Exotic Pest Plant Council (Producer). Available: http://www.se-eppc.org/fslist.cfm [2003, August 25]. 
151. U.S. Department of Agriculture, Natural Resources Conservation Service. 2008. PLANTS Database, [Online]. Available: http://plants.usda.gov/. 
152. U.S. Department of the Interior, Fish and Wildlife Service. 2004. Endangered and threatened wildlife and plants; review of species that are candidates or proposed for listing as endangered or threatened; annual notice of findings on resubmitted petitions; annual description of progress on listed accounts: 50 CFR Part 117, [Online]. In: Federal Register 69(86): 24876-24904. U.S. Department of the Interior, Fish and Wildlife Service (Producer). Available: http://endangered.fws.gov/candidates/2003_CNOR.pdf [2005, May 12]. 
153. Udensi, Udensi E.; Akobundu, I. Okezie; Ayeni, Albert O.; Chikoye, David. 1999. Management of cogongrass (Imperata cylindrica) with velvetbean (Mucuna pruriens var. utilis) and herbicides. Weed Technology. 13(2): 201-208. 
154. van der Wal, Hans. 1999. Chinantec shifting cultivation: InTERAcTIVE landuse: a case-study in the Chinantla, Mexico, on secondary vegetation, soils and crop performance under indigenous shifting cultivation. Heelsum, NL: Treemail Publishers. 162 p. 
155. Varner, J. Morgan, III; Kush, John S. 2004. Remnant old-growth longleaf pine (Pinus palustris Mill.) savannas and forests of the southeastern USA: status and threats. Natural Areas Journal. 24(2): 141-149. 
156. Velayuthan, A.; Lim, Cheo Yu. 1986. Sheep farming--another tool for weed control under oil palm/rubber plantations by using a cheaper management technique. Planter. 62: 319-332. 
157. Virginia Department of Conservation and Recreation, Division of Natural Heritage. 2003. Invasive alien plant species of Virginia, [Online]. Virginia Native Plant Society (Producer). Available: http://www.dcr.state.va.us/dnh/invlist.pdf [2005, June 17]. 
158. Wade, Dale D.; Brock, Brent L.; Brose, Patrick H.; Grace, James B.; Hoch, Greg A.; Patterson, William A., III. 2000. Fire in eastern ecosystems. In: Brown, James K.; Smith, Jane Kapler, eds. Wildland fire in ecosystems: Effects of fire on flora. Gen. Tech. Rep. RMRS-GTR-42-vol. 2. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 53-96. 
159. Wade, Dale D.; Lunsford, James D. 1989. A guide for prescribed fire in southern forests. Technical Publication R8-TP-11. Atlanta, GA: U.S. Department of Agriculture, Forest Service, Southern Region. 56 p. 
160. Walsh, S. R. 1954. Blady grass and its control by mowing on the Atherton Tableland. Queensland Agricultural Journal. 79: 325-333. 
161. Wibowo, A.; Suharti, M.; Sagala, A. P. S.; Hibani, H.; van Noordwijk, M. 1997. Fire management on Imperata grasslands as part of agroforestry development in Indonesia. Agroforestry Systems. 36(1-3): 203-217. 
162. Wibowo, Ari; Suharti, Mieke dan; Pickford, Stewart G. 1991. Fuel characteristics and fire behaviour in alang-alang under Acacia mangium plantation in Depok, West Java. Forestry Research Bulletin. 544: 1-7. 
163. Wiggins, Ira L. 1980. Flora of Baja California. Stanford, CA: Stanford University Press. 1025 p. 
164. Wilcut, J. W.; Truelove, B.; Davis, D. E. 1985. Cogongrass and torpedograss troublesome in coastal area. Highlights of Agricultural Research. 32(3): 9. 
165. Wilcut, John W.; Dute, Roland R.; Truelove, Bryan; Davis, Donald E. 1988. Factors limiting the distribution of cogongrass, Imperata cylindrica, and torpedograss, Panicum repens. Weed Science. 36(5): 577-582. 
166. Willard, T. R.; Shilling, D. G. 1990. The influence of growth stage and mowing on competition between Paspalum notatum and Imperata cylindrica. Tropical Grasslands. 24: 81-86. 
167. Willard, Thomas R.; Gaffney, James F.; Shilling, Donn G. 1997. Influence of herbicide combinations and application technology on cogongrass (Imperata cylindrica) control. Weed Technology. 11(1): 76-80. 
168. Willard, Thomas R.; Shilling, Donn G.; Gaffney, James F.; Currey, Wayne L. 1996. Mechanical and chemical control of cogongrass (Imperata cylindrica). Weed Technology. 10(4): 722-726. 
169. Willard, Tommy R.; Hall, David W.; Shilling, Donn G.; Lewis, James A.; Currey, Wayne L. 1990. Cogongrass (Imperata cylindrica) distribution on Florida highway rights-of-way. Weed Technology. 4(3): 658-660. 
170. Willard, Tommy Ray. 1988. Biology, ecology and management of cogongrass [Imperata cylindrica (L.) Beauv.]. Gainesville, FL: University of Florida. 113 p. Dissertation. 
171. Wolfe, June, III; Zajicek, J. M. 1998. Are ornamental grasses acceptable alternatives for low maintenance landscapes? Journal of Environmental Horticulture. 16(1): 8-11. 
172. Woods, Paul. 1989. Effects of logging, drought, and fire on structure and composition of tropical forests in Sabah, Malaysia. Biotropica. 21(4): 290-298. 
173. Wunderlin, Richard P. 1998. Guide to the vascular plants of Florida. Gainesville, FL: University Press of Florida. 806 p. 
174. Wyatt-Smith, J. 1949. Natural plant succession. Malaysian Forester. 12(3): 148-152. 
175. Yadav, B. P. S.; Gupta, H. K.; Gupta, J. J. 1998. Growth pattern in goats on native pasture of northeastern India. Range Management and Agroforestry. 19(2): 133-138. 
176. Yager, Lisa; Jones, Jeanne; Miller, Deborah. 2003. Cogongrass (Imperata cylindrica) encroachment into upland pine communities on Camp Shelby Training Site (CSTS), Mississippi. In: Proceedings, annual meeting of the Ecological Society of America; 2003 August 3-8; Savannah, GA. In: [Ecology]. Washington, DC: The Ecological Society of America; 88: 367. Abstract. 
177. Yandoc, Camilla B.; Charudattan, Raghavan; Shilling Donn G. 2004. Suppression of cogongrass (Imperata cylindrica) by a bioherbicidal fungus and plant competition. Weed Science. 52: 649-653. 
178. Yandoc, Camilla B.; Charudattan, Raghavan; Shilling, Donn G. 2005. Evaluation of fungal pathogens as biological control agents for cogongrass (Imperata cylindria). Weed Technology. 19(1): 19-26. 
179. Youtie, Berta; Soll, Jonathan. 1990. Diffuse knapweed control on the Tom McCall Preserve and Mayer State Park. Unpublished report (prepared for the Mazama Research Committee) on file at: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT. 18 p.