26. CAREX Linnaeus, Sp. Pl. 2: 972. 1753; Gen. Pl. ed. 5, 420. 1754.
Peter W. Ball & A. A. Reznicek
Herbs, perennial, cespitose or not, rhizomatous, rarely stoloniferous. Culms usually trigonous, sometimes round. Leaves basal and cauline, sometimes all basal; ligules present; blades flat, V-shaped, or M-shaped in cross section, rarely filiform, involute, or rounded, commonly less than 20 mm wide, if flat then with distinct midvein. Inflorescences terminal, consisting of spikelets borne in spikes arranged in spikes, racemes, or panicles; bracts subtending spikes leaflike or scalelike; bracts subtending spikelets scalelike, very rarely leaflike. Spikelets 1-flowered; scales 0–1. Flowers unisexual; staminate flowers without scales; pistillate flowers with 1 scale with fused margins (perigynium) enclosing flower, open only at apex; perianth absent; stamens 1–3; styles deciduous or variously persistent, linear, 2–3(–4)-fid. Achenes biconvex, plano-convex, or trigonous, rarely 4-angled. x = 10.
Species ca. 2000 (480 in the flora): worldwide.
Carex is one of the largest genera of vascular plants. Almost worldwide in distribution, it is absent from tropical lowlands except for a few species in southeast Asia. The genus is poorly represented also in sub-Saharan Africa. The flora area is species rich, almost comparable to eastern Asia. Carex is most commonly associated with moist to wet habitats, usually with water not more than 50 cm deep in the growing season. Species of Carex are often dominant or co-dominant in such habitats, including arctic tundra. Carex is often common also in montane grasslands, montane rocky habitats, and forests. In eastern North America, many species occur in deciduous or mixed forests, where they are major constituents of the understory. In eastern forests, 20 or more species of Carex may be found within a few hectares.
All Carex are perennial, but a few species may fruit in their first year and not persist (C. bebbii, C. crawfordii, C. sychnocephala, C. viridula, and no doubt others). A few species, such as C. adusta, are short-lived, with individuals living only three to five years. All species have rhizomes, though in densely cespitose species the rhizomes are often very short and inconspicuous, leading some authors to consider them absent. In some, the rhizomes are elongated, and the plants may form extensive swards. A few species, such as C. stricta and some other tussock forming species and some species that occur on sand dunes, have rhizomes that can grow vertically.
Shoots of sedges vary greatly in their longevity. In many species, all aboveground shoots are annual. In others, individual shoots may live more than one season. In such species, shoots may be vegetative in the first year and flower in a subsequent year. In species with annual flowering stems, only bladelesss basal sheaths occur, clothing the base of the stem. Those stems are termed aphyllopodic (from the absence of leaf blades on the basal sheaths of the stem) or lateral (because they are formed lateral to previous year’s shoot). In species where the flowering stem is produced from the apical meristem of a vegetative shoot from the previous year, the base of the stem is clothed in leaves of the previous year, in varying states of decay. Those stems are termed phyllopodic or central. This distinction is usually clear in regions with cold winters; it may be blurred in subtropical and tropical areas. In all cases, fertile stems die after fruiting.
In most Carex, the only aboveground stems are those bearing inflorescences. Leaves may be basal or distributed along the stem. Vegetative shoots usually have only basal leaves, with the stemlike aboveground portion being composed only of overlapping leaf sheaths. In a few species, the vegetative shoots are true stems with nodes and internodes. True vegetative stems are characteristic of Carex sect. Ovales, though they are often developed only after fruiting. They are also found in sect. Holarrhenae. In Carex sect. Carex, they are well developed and exceed the fertile stems. In a few species, such as C. assiniboinensis, C. chordorrhiza, C. limosa, and C. mackenziei, some of the vegetative shoots are short, while others differentiate into elongate, leafy stolons that function in vegetative reproduction. In some members of sect. Ovales, the vegetative shoots are very leafy and elongate. In a few species, such as C. tribuloides, C. projecta, C. longii, C. ozarkana, and several others, the vegetative shoots may overwinter and produce new shoots at the nodes; in many species of sect. Ovales, and in sections Carex and Holarrhenae, those stems are strictly annual.
Leaves of Carex are typically linear and have a ligule at the junction of the blade and the sheath. The ligule is mostly fused to the blade, with a narrow, entire or erose-ciliate free portion. Sheaths are often differentiated, with the front (the side opposite the blade) being thin, translucent, and sometimes dotted or veined. When veined, the disintegration and tearing of the sheath front may leave a regular pattern of veins described as ladder-fibrillose. In some wetland species, the sheaths are spongy with large air cavities between the cross veins. Upon drying, the collapse of these air cavities results in the cross veins becoming very prominent, a condition termed septate-nodulose. In some species, such as C. sprengelii, the veins in the sheaths are heavily sclerified and persist as fibrous tufts at the base of the plant and along the rhizome after the leaves decay. These basal fibrous tufts should not be confused with ladder-fibrillose sheaths. Blades are usually V- or M-shaped in cross section. The V-shaped leaves are keeled with a midvein prominent on the abaxial surface and M-shaped leaves have a midvein prominent on the abaxial surface and two mid-lateral veins prominent on the adaxial surface. Both may become essentially flat at maturity. Some species have leaves that are trough-shaped or involute, or even bristlelike. In these, the midvein usually is not keeled and may be no more pronounced than other veins; prominent mid-lateral veins are also lacking. In many species with narrow leaves, the ultimate tip becomes triangular in cross section. Some forest understory species have very broad, flat, short leaves, with short sheaths and may lie essentially flat on the ground.
Inflorescence and flower structures of Carex and other genera in tribe Cariceae (Cymophyllus and Kobresia in the flora, Schoenoxiphium and Uncinia from outside the flora) are difficult to interpret; it is by no means clear whether apparently similar structures in different sections of Carex and in the other genera are homologous (A. A. Reznicek 1990). Staminate flowers consist of three or fewer stamens subtended by a single scale. The arrangement and development of the stamens can be interpreted as each stamen originating from a single flower, the staminate flower therefore being a condensed inflorescence (D. L. Smith and J. S. Faulkner 1976). This interpretation is considered by some authors to be unnecessarily complex and based on inconclusive evidence (T. V. Egorova 1999), so here the staminate flower is regarded as a single flower. The pistillate flower of Carex is believed to be derived by reduction from an inflorescence unit similar to those found in Kobresia and Schoenoxiphium, namely a branch in the axil of a scalelike bract bearing pistillate flower(s) proximally and staminate flowers distally (D. L. Smith and J. S. Faulkner 1976). The first leaf of the axillary branch is modified to a spathelike prophyll. In Schoenoxiphium and Kobresia the prophyll encloses the branch, but the edges are free. In Carex, Cymophyllus, and Uncinia the staminate flowers have been lost and the edges of the prophyll are united to form a perigynium, which has an orifice at the tip through which the style projects. In most species of Carex, the axillary branch (rachilla) is apparently absent, although some minute trace of the rachilla is probably always present (A. A. Reznicek 1990), and in a few species it is well developed and shortly extruded. The flowers are arranged in distinct condensed inflorescences (spikes) that resemble the primary inflorescences (spikelets) of most other genera of Cyperaceae. The spikes are unisexual or bisexual: in bisexual spikes either the distal flowers are staminate and the proximal pistillate (androgynous) or the distal flowers are pistillate and the proximal staminate (gynecandrous).
Inflorescence structure and organization present many problems of interpretation (see summaries by D. L. Smith and J. S. Faulkner 1976; A. A. Reznicek 1990; T. V. Egorova 1999). Inflorescence organization was one of the more important characters used by G. Kükenthal (1909) and others to define subgenera in Carex. Kükenthal divided Carex into four subgenera, which are often referenced in discussions of the classification and evolution of the genus. Subgenus Primocarex consisted of all species in which the inflorescence consisted of a single spike. In different sections of the subgenus, the inflorescence was unisexual, or bisexual and androgynous, or gynecandrous. In subg. Vignea, the spikes were usually bisexual, sessile, and arranged in spikes to elaborately branched panicles. In most sections of subg. Vignea, the styles were bifid and the inflorescence branches lacked prophylls. In subg. Carex (subg. Eucarex), the spikes were most frequently unisexual, the terminal and sometimes some distal lateral spikes being staminate, and the lateral or proximal lateral spikes pistillate; bisexual spikes occurred occasionally. The lateral spikes were peduncled or subsessile and bore a minute, tubular prophyll at the base of the peduncle. The inflorescences were usually spicate or racemose, although more complex inflorescences occurred, especially in Asia. Subgenus Indocarex usually had bisexual spikes that were usually peduncled and usually arranged in branched panicles and had a prophyll that resembled a perigynium at the base of an inflorescence branch. Many deviations from these generalizations exist. For a table summarizing differences among the subgenera as defined by Kükenthal see A. A. Reznicek (1990).
Although the classification and phylogeny of Carex have been the subjects of much debate, the evidence does not allow definitive conclusions. Historic accounts of previously published classifications and phylogenetic interpretations, based primarily on morphologic and anatomic evidence, are available (T. V. Egorova 1999; A. A. Reznicek 1990). The most widely referenced and most recent monograph of the genus is that by G. Kükenthal (1909). His division of the genus into four subgenera has been heavily criticized by many authors, particularly regarding subg. Primocarex. One view maintained that that subg. Primocarex was artificial and that all single-spiked species were derived by reduction from the other three subgenera or even from the genus Uncinia (V. I. Kreczetowicz 1936). Presenting similar criticisms E. Nelmes (1952) distributed sections of subg. Primocarex to other subgenera of Carex and Uncinia, although without making the necessary nomenclatural changes. Despite criticisms Kükenthal’s proposed subdivision of Carex into four subgenera is still widely accepted, albeit with some modifications to meet various criticisms of subg. Primocarex (A. O. Chater 1980; T. V. Egorova 1999). In addition to not accepting subg. Primocarex, T. Koyama (1961b, 1962) did not accept subg. Indocarex and distributed the sections of the subgenus to various parts of subg. Carex. One or two additional subgenera have been proposed as segregates of subg. Carex (e.g., T. V. Egorova 1999); these have not gained general acceptance, and DNA sequence investigations (e.g., E. H. Roalson et al. 2001) do not support their recognition.
In North America, most authors have followed K. K. Mackenzie’s (1931–1935) arrangement of the genus, in which he did not recognize subgenera and instead divided the North American Carex into 71 sections. The sections were narrowly defined, for the most part consisting of groups of species that were very similar morphologically. Mackenzie appears to have arranged sections using much the same criteria although he never explained his arrangement. He started with a group of sections that consisted of single-spiked species, included in G. Kükenthal’s subg. Primocarex, and referred to these sections as “primitive” in his synopsis of the genus. Many single-spiked species, however, were associated with sections Kükenthal had included in the other subgenera. Mackenzie next placed a group of sections together that Kükenthal had included in subg. Vignea, and then a group of sections included by Kükenthal in subgenera Primocarex, Indocarex, and Carex. Thus Mackenzie’s arrangement met many of the criticisms of Kükenthal’s arrangement, but it cannot be applied to the genus on a worldwide basis because few species of subg. Indocarex occur in North America.
The arrangement of the genus followed below is a modified version of K. K. Mackenzie’s system; the sections largely follow his delimitations, except as noted under particular sections. The sections are arranged roughly in a sequence that groups together first those sections usually included in subg. Vignea, including a few species and sections placed by G. Kükenthal in subg. Primocarex (sects. a–q); then sections usually included in subg. Carex, plus sections from subg. Primocarex that are probably related to them (sects. r–jjj); then a group of unispicate sections included by Kükenthal in subg. Primocarex (sects. kkk–qqq); and finally the only North American section included by Kükenthal in subg. Indocarex (sect. rrr). Subgenera are not recognized; although it is possible to assign the multi-spiked sections to subgenera, relationships of a number of unispicate sections are very unclear.
Recent investigations of the genus and related genera utilizing DNA sequences have both clarified and confused the classification and phylogeny of the genus (J. R. Starr et al. 1999; A. C. Yen and R. G. Olmstead 2000, 2000b; E. H. Roalson et al. 2001). All investigations have confirmed the monophylly of subg. Vignea as modified by most recent authors. Subgenus Carex is also generally monophyletic, except that the few species of subg. Indocarex that have been investigated are ambiguously grouped with the bulk of subg. Carex, as are a few species of subg. Primocarex. The bulk of subg. Primocarex, together with Cymophyllus, Uncinia, and Kobresia plus a few species of subg. Carex, form another monophyletic group, with Schoenoxiphium as a sister group. The investigations differ in their placement of subg. Vignea; some place it as a sister group to subg. Carex, while others place it as a sister group to the whole of the tribe Cariceae. Sectional groupings have not been thoroughly investigated, with a few exceptions, and, to date, these investigations have not clarified sectional relationships within the genus. The investigations generally do not support the morphologically based hypothesis (V. I. Kreczetowicz 1936) that most single-spiked species of Carex are recently derived from multi-spiked species, but they have confirmed that a small number of unispicate species are closely related to multi-spiked species.
Numerous interspecific hybrids have been reported in Carex, though few have been produced experimentally. Most hybrids involve species in the same section; some intersectional hybrids are known. Most hybrids are sterile or have very low fertility and most are rare or uncommon. In a few sections, hybrids are much more frequent and may be locally common, in particular in sects. Ceratocystis, Phacocystis, and Vesicariae, where hybrids may be partially fertile. An extensive review of interspecific hybridization involving North American species of Carex was published by (J. Cayouette and P. M. Catling 1992), who accepted reports of 253 hybrids in North America and an additional 73 between species that occur in North America, although to date, hybrids have been found only in Eurasia.
Considering the size and widespread occurrence of Carex, the rather limited economic utility of the genus is surprising. Species of Carex are important constituents of many peat deposits that are exploited commercially. Species of Carex are often important components of moist to wet habitats used as forage for livestock and for herbivorous wildlife. Montane grasslands may also have significant biomass in species of Carex utilized by domestic animals and wildlife. Many species of Carex occurring in eastern Canada, especially the long-rhizomatous ones, are of high forage value (P. M. Catling et al. 1994). Their results are probably of general application as a wide diversity of Carex sections were included. In recent years, species of Carex have greatly increased in popularity in North America as ornamentals. A variety of Eurasian, as well as a few native species, are available commercially. Some of these species have recently become locally naturalized (e.g., Carex pendula); the frequency of naturalization is likely to increase.
For additional information about Carex, see P. M. Catling et al. (1990) and J. M. Bernard and L. Soukupová (1988).
Treatments of Carex do not strictly follow the usual sequence of characteristics; the sequence reflects their complex morphology and the fact that descriptions generally refer to the stage when fruit is developed but not fully ripe.
Bernard, J. M., and L. Soukupová, eds. 1988. Studies in wetland carices of the temperate zone. Aquatic Bot. 30: 1–168. Catling, P. M., A. R. McElroy, and K. W. Spicer. 1994. Potential forage value of some eastern Canadian sedges (Cyperaceae: Carex). J. Range Managem. 47: 226–230. Catling, P. M., A. A. Reznicek, and W. J. Crins, eds. 1990. Systematics and ecology of the genus Carex (Cyperaceae). Canad. J. Bot. 68: 1405–1472. Cayouette, J. and P. M. Catling. 1992. Hybridization in the genus Carex with special reference to North America. Bot. Rev. (Lancaster) 58: 351–438. Chater, A. O. 1980. Carex. In: T. G. Tutin et al., eds. 1964–1980. Flora Europaea. 5 vols. Cambridge. Vol. 5, pp. 290–323. Egorova, T. V. 1999. Sedges (Carex L.) of Russia and Adjacent States within the Limits of the Former USSR. St. Petersburg and St. Louis. Hermann, F. J. 1970. Manual of the Carices of the Rocky Mountains and Colorado Basin. Washington. [Agric. Handb. 374.] Hudson, J. H. 1977. Carex in Saskatchewan. Saskatoon. Kreczetowicz, V. I. 1936. Are the sedges of subgenus Primocarex Kük. primitive? Bot. Zhurn. S.S.S.R. 21: 395–425. Kükenthal, G. 1909. Cyperaceae–Caricoideae. In: H. G. A. Engler, ed. 1900–1953. Das Pflanzenreich…. 107 vols. Berlin. Vol. 38[IV,20]. Nelmes, E. 1952. Facts and speculations on phylogeny in the tribe Cariceae of the Cyperaceae. Kew Bull. 1951: 427–436. Reznicek, A. A. 1990. Evolution in sedges (Carex, Cyperaceae). Canad. J. Bot. 68: 1409–1432. Roalson E. H., J. T. Columbus, and E. A. Friar. 2001. Phylogenetic relationships in Cariceae (Cyperaceae) based on ITS (nrDNA) and trnT-L-F (cpDNA) region sequences: Assessment of subgeneric and sectional relationships in Carex with emphasis on section Acrocystis. Syst. Bot. 26: 318–341. Robertson, A. 1984. Carex of Newfoundland. St. John’s. Smith, D. L. and J. S. Faulkner. 1976. The inflorescence of Carex and related genera. Bot. Rev. (Lancaster) 42: 53–81. Starr, J. R., R. J. Bayer, and B. A. Ford. 1999. The phylogenetic position of Carex section Phyllostachys and its implications for phylogeny and subgeneric circumscription in Carex (Cyperaceae). Amer. J. Bot. 86: 563–577. Yen, A. C. and R. G. Olmstead. 2000. Molecular systematics of Cyperaceae tribe Cariceae based on two chloroplast DNA regions: ndhF and trnL intron-intergenic spacer. Syst. Bot. 25: 479–494. Yen, A. C. and R. G. Olmstead. 2000b. Phylogenetic analysis of Carex (Cyperaceae): Generic and subgeneric relationships based on chloroplast DNA. In: K. L. Wilson and D. A. Morrison, eds. 2000. Monocots: Systematics and Evolution. Melbourne. Pp. 602–609.