Flora of North America species comparison

The Poaceae or grass family includes approximately 700 genera and 11,000 species (Chen et al. 2006). The two grass volumes in this series treat 10 subfamilies, 25 tribes, 236 genera, and 1373 species. Of these, all the subfamilies, 22 tribes, 136 genera, and 892 species are native to the Flora region; 2 tribes, 78 genera, and 290 species have become established in the region. The remaining taxa include ornamental species; species grown for research purposes; species that, if introduced to the region, would pose a threat to important agricultural species; and a few waifs, i.e., species that have been found in the region but have not become established. Most of the waifs are species that were found on ballast dumps near ports around the turn of the last century.

Grasses constitute the fourth largest plant family in terms of number of species. Nevertheless, the family is clearly more significant than any other plant family in terms of geographic, ecological, and economic importance. Grasses grow in almost all terrestrial environments, including dense forests, open deserts, and freshwater streams and lakes. There are no truly marine grasses, but some species grow within reach of the highest tides.

In addition to being widely distributed, grasses are often dominant or co-dominant over large areas. The importance of such areas to humans is reflected in the many words that exist for grasslands, words such as meadow, palouse, pampas, prairie, savanna(h), steppe, and veldt. Not surprisingly, given their abundance and prevalence, grasses are of great ecological importance as soil stabilizers and as providers of shelter and food for many different animals.

The economic importance of grasses to humans is almost impossible to overestimate. The wealth of individuals and countries is dependent on the availability of such sources of grain as Triticum (wheat), Oryza (rice), Zea (corn or maize), Hordeum (barley), Avena (oats), Secale (rye), Eragrostis (tef), and Zizania (wild rice). Most countries invest heavily in research programs designed to develop better strains of these grasses and the many other grasses that are used for livestock, soil stabilization, and revegetation. Developing improved grasses for recreation areas, such as playing fields, golf courses, and parks, is also a major industry in many parts of the world; increasing recognition of the aesthetic value af grasses is reflected in their prominence in horticultural catalogs.

There are, of course, grasses that are considered undesirable, at least in some parts of the world, but even the most obnoxious grasses may be well-regarded over a portion of their range. For instance, Bromus tectorum (cheatgrass) is a noxious, fire-prone invader of western North American ecosystems; it is also welcomed as a source of early spring feed in some parts of the Flora region. Cynodon dactylon (bermudagrass) is listed as a noxious weed in some jurisdictions; in others it is valued as a lawn grass.

Although grasses are widespread and often dominant in open areas, all evidence points to an origin of the family in forests, most likely in the Southern Hemisphere, at least 55-70 mya (Grass Phylogeny Working Group 2001). Recent evidence from phytoliths (isolated silica bodies commonly produced inside the epidermal cells of grasses and some other plants) embedded in fossil coprolites strongly suggests that grasses evolved earlier in the Cretaceous than previously thought (Prasad et al. 2005). Living representatives of the three earliest lineages of the grass family, together comprising about 30 species, are perennial, broad-leaved plants of relatively small stature, native to tropical or subtropical forests in South America, Africa, southeast Asia, some Pacific Islands, and northern Australia. The major diversification of the family probably occurred in the mid-Cenozoic, and was associated with climatic changes that produced more open habitats. All major lineages of the grass family were present by the middle of the Miocene (Jacobs et al. 1999), and C4 photosynthesis in grasses had evolved by then, as well.

Molecular and morphological data unequivocally support a single origin for the Poaceae (Grass Phylogeny Working Group 2001). The caryopsis, a single-seeded, usually dry and indehiscent fruit with the pericarp usually strongly adherent to the seed, and the laterally positioned, highly differentiated embryo are unique to grasses. Beyond the three early-diverging lineages (Anomochlooideae Potztal, Pharoideae, and Puelioideae L.G. Clark et al.), the great diversity of grasses can be divided into two major lineages: the BEP clade (Bambusoideae, Ehrhartoideae, and Pooideae); and the PACMCAD clade (Panicoideae, Arundinoideae, Chloridoideae, Micrairoideae Pilger, Centothecoideae, Aristidoideae, and Danthonioideae, i.e., the PACCAD clade of volume 25 plus the Micrairoideae, support for recognition of which was obtained after publication of that volume). Relationships among the BEP grass lineages remain uncertain, and some evidence points to the Pooideae as being more closely related to the PACMCAD clade than to the Bambusoideae or Ehrhartoideae. The PACMCAD clade includes all known C4 or warm-season grasses.

The closest relatives of the Poaceae lie within a group of six families, all native primarily to the Southern Hemisphere: Joinvilleaceae Toml. & A.C. Sm., Ecdeiocoleaceae D.F. Cutler &c& Airy Shaw, Restionaceae R. Br., Centrolepidaceae Endl., Anarthriaceae D.E Cutler &c& Airy Shaw, and Flagellariaceae Dumort. (Poales Small, sensu stricto). Joinvilleaceae, Ecdeiocoleaceae, and Poaceae constitute a three family clade, with Ecdeiocoleaceae probably being closer than Joinvilleaceae to the Poaceae (Bremer 2000; Bremer 2002; Michelangeli et al. 2003). Rudall et al. (2005), based on a study of reproductive structures in the Ecdeiocoleaceae, suggest that the grass caryopsis may represent "one end of a transformation series embodied by the reduced gynoecial structure and indehiscent fruit of other Poales such as Flagellaria and Ecdeiocolea" (p. 1441).

(Discussion copyrighted by Flora of North America; reprinted with permission.)

The Bambuseae include about 80-90 genera and around 1400 species. They are most abundant in Asia and South America, but are also found in Africa, Australia, Central America and North America. One genus, Arundinaria, is native to the Flora region, where it is represented by 3 species. Many other genera and species are cultivated in the region for their ornamental value. These often persist for decades; some have become established beyond the original planting. Identification of these introduced bamboos is hampered by the lack of taxonomic studies in their countries of origin, particularly studies of vegetative features, and the large number of taxa that have not yet been described. Identification is further hindered by their infrequent flowering.

Woody bamboos differ from all other grasses in the division of vegetative growth into two phases. Growth commences with the production of unbranched new culms which bear protective leaves called culm leaves (or culm sheaths) at their nodes. Later in the same season, or in the following season, buds at the nodes of the new culms develop into branches, which produce foliage leaves with photosynthetic blades. The leaves on the culms and foliage branches are homologous, each consisting of a sheath, adaxial ligule, and blade, but the culm leaves have less well-developed blades and quickly become non-photosynthetic or fall away, while the foliage leaves have more persistent, photosynthetic, well-developed blades.

Another feature characteristic of some, but not all, bamboos is the 'pseudospikelet', which differs from a determinate spikelet in its indeterminate growth from lateral buds in the axils of the basal glumelike bracts. This can lead to extensive ramification of the pseudospikelet into a capitate cluster [see McClure (1966) for a more detailed discussion].

Woody bamboos are also distinguished by perennial lignified culms which often have complex branch complements at the nodes, and often by cyclical flowering at intervals of up to 150 years. In cyclical flowering, an entire population or even species will flower in a given year, after which the parents usually die, regeneration being through slow-growing and vulnerable seedlings.

Most woody bamboos are tropical or subtropical in their distribution, but about 25 genera are found in temperate regions, mainly in eastern Asia. Two genera are native to North America; Arundinaria is the only genus native to the Flora region. It is thought to have crossed the Bering Strait from eastern Asia, and may represent the sole remnant of a much larger pre-glacial bamboo population. The other genus, Otatea (McClure & E.W. Sm.) C.E. Calderon & Soderstr., is probably a relatively recent entrant into Mexico from South America.

The taxonomy of woody bamboos has developed slowly because of the scarcity of flowering material, and the distribution of the species in predominantly inaccessible or less-developed parts of the world. Concentration on floral characters in earlier classification systems has made taxonomy and identification even more difficult. Cytological information has been of little value beyond the separation of two major Asian groups, one tropical with 2n = 56-72 and the other temperate with 2n = 48. There appears to be more variation among American bamboos.

Molecular data have thrown doubt upon the phylogenetic validity of many earlier attempts to classify the woody bamboos by floral characters alone. Nevertheless, they are illuminating interesting patterns of evolution and dispersal. A wider range of morphological characters is increasingly being studied and applied. Because many species have only been described in recent years, while many cultivated bamboos have been grown under misapplied or speculative names, the taxonomy of bamboos will require continued study for many years to come. Fieldwork, herbarium study, and laboratory investigations will all be necessary, along with extensive international collaboration. It is highly likely that a substantial proportion of forest bamboos will become extinct before they have been properly documented.

The American Bamboo Society lists over 450 taxa of bamboo, representing 240 species in 40 genera, as being commercially available in the United States. Although bamboos are increasingly widely cultivated in the Flora region, they are most common in the coastal and southern states. Most of the cultivated species are Asian in origin, but in recent years numerous Central and South American taxa have been introduced. Most introduced taxa will persist indefinitely without cultivation. In favorable climates, many will spread beyond the original plantings to become naturalized, especially those with long rhizomes. Wider dispersal occurs when sections of the rhizome are transported from one location to another, as may happen during floods. Clump-forming bamboos may eventually spread through seed dispersal.

Bamboos are multipurpose plants of immense utility to mankind. Traditional uses, such as building, basketry, and fodder, have been supplemented by industrial-scale paper-pulp production, the canning of edible shoots, and production of advanced board products, laminates, and cloth.

This treatment is limited to a full treatment of the native genus, Arundinaria, and descriptions and representative illustrations for the three genera and seven species thought to have become established in the Flora region.

(Discussion copyrighted by Flora of North America; reprinted with permission.)