Research Report | SR132

ORIGIN, BIOGEOGRAPHICAL MIGRATIONS AND DIVERSIFICATIONS OF TURFGRASSES James B Beard1

Executive Summary Whether a turfgrass species is characterized as Primitive ancestral grasses are now proposed native or naturalized to North America has been to have appeared during the Late Cretaceous based on world-wide simplistic observations between 65 and 96 mya (million years ago) in focused on where the greatest genetic diversity Gondwanan Africa. The ancestral Pooideae are occurred, termed center-of-origin. Research infor- estimated to have migrated to the steppes of mation as to dating and locations of subsequent Laurasian Eurasia during the Eocene ~ 38 to

migration and diversification has been minimal 47 mya. Taxonomic divergence of the base C3 due to a lack of needed research technologies. Pooideae group appears to have been initiated in

Intercontinental migration of grasses has been Europe ~ 26 to 33.5 mya. The base C4 Pooideae assumed to have been unlikely due to oceanic apparently arose in Africa ~ 30 to 33 mya, followed separation. Recent development of paleobotanical by migration to West Gondwana South America studies using ultrastructural electron microscopic and to East Gondwana India and Australia. techniques and stable carbon isotope dating instrumentation and research procedures, plus Diversification led to the emergence of an ancient molecular phylogenetic research and cladistic Poeae group known as the fine-leaf fescues biogeographic analysis of large data sets are (Festuca) in central-Europe during the mid- clarifying our understanding of migration patterns Miocene ~ 13 mya. Subsequent migration occurred and dating of multiple secondary centers-of-origin via the mountains of central and eastern Asia, for grasses. across the Bering Land Bridge into North America, southward via the western mountains, over the Pre-anthropological intercontinental migration of Panamanian Land Bridge, and across the Andes grasses is now being documented as more sig- to Patagonia. This occurred between ~ 3.8 and 10 nificant than previously assumed. Global heating/ mya, which is before the anthropological effects cooling phases and shifts in sea level appear to of humans. Clearly, the fine-leaf fescues are native have facilitated oceanic island hopping. At varying to North America, migrating eastward after the times, these bridge migrations may have involved ice age glacial melt ~ 1 mya. Similar investigations the northern Europe to North America route with other turfgrasses may clarify the migration via Greenland, Bering Bridge between Asia and routing and diversification dating of other turf- North America, southeast Asia to Australia via grass species. Thus, the current status of turfgrass New Guinea, South America to North America via species in terms of native or anthropological Caribbean and later via Panama Bridge, and Africa- naturalization is presented in this paper and sum- Europe via the Gibraltar Straits. Also, the potential marized in the following table. for transoceanic migrations is being reconsidered.

1 Professor Emeritus, Texas A&M University and International Sports Turf Institute, College Station, Texas, 77840; e-mail isti@ neo.tamu.edu. This electronic version is being published by the Michigan State University Press in a book titled Turfgrass His- tory and Literature. Copyright 2012, James B Beard.

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 1 Summary of native and naturalized categories of 24 turfgrasses found in North America, based on current knowledge.

Native/Naturalized Categories Grass Genus and Species Turfgrasses Common Name

Native, based on biogeo- Festuca rubra red fescues graphical, phenotypic, and phylogenetic studies. Festuca trachyphylla hard fescue Festuca ovina sheep fescue

Native, based on phenotypic Axonopus compressus broadleaf carpetgrass and genetic assessments.* Axonopus fissifolius common carepetgrass

Bouteloua curtipendula side-oats gramagrass

Bouteloua gracilis blue gramagrass

Buchloe dactyloides American buffalograss

Stenotaphrum secundatum St. Augustinegrass

Possible native, based on Agrostis capillaris colonial bentgrass phenotypic assessments and allied species biogeographical, Agrostis stolonifera creeping bentgrass phylogenetic studies.* Agrostis canina velvet bentgrass

Poa pratensis Kentucky bluegrass

Possible naturalized, following Eremochloa ophiuroides centipedegrass anthropological introduction.* Lolium multiflorum annual ryegrass

Lolium perenne perennial ryegrass

Paspalum vaginatum seashore paspalum

Poa trivialis rough bluegrass

Schedonorus arundinaceus tall fescue

Zoysia japonica Japanese zoysiagrass

Zoysia matrella ssp. matrella manila zoysiagrass

Zoysia pacifica mascarene zoysiagrass

Undocumented, needs Cynodon dactylon ssp. dactylon dactylon bermudagrass research Poa annua annual bluegrass *Needs further biogeographical and phylogenetic research to confirm.

2 Origin, Biogeographical Migrations and Diversifications of Turfgrasses Introduction

Understanding the relatively recent research in persisted in recent times if human disturbance had grass species evolution and development has not occurred. Disturbed environments can have dictated revision of how native and naturalized long-term or short-term effects on plant com- turfgrass species are viewed, especially in North munity composition depending on the degree and America. The supporting research for this updat- duration of the disturbance event. Disturbance can ing is reviewed herein.2 be a natural aspect of plant ecosystems as caused by drought, fire, disease, insect, glaciation, heat Native Turfgrass Species. Webster’s Third stress, freeze stress, volcanic smoke, soil deposi- New International Dictionary defines native as tion, soil salt level, animal pressures, etc. Defining something indigenous to a particular locality. a native plant based on a fixed point in time, such Indigenous is defined as native and as originating as pre-Holocene, pre-Neolithic, or pre-European or developing or produced naturally in a particular colonial, is unrealistic as it assumes no non-human land or region or environment. The emotive as- disturbances would have occurred subsequently. sociated terminologies relative to native that have It also fails to acknowledge humans as a compo- been used include naturalized, alien, exotic and nent of the ecosystem. Historically, the species non-native. Criteria for classifying native spe- composition of most plant communities has been cies that have been utilized or proposed include a continuous, dynamic process of transitional historical, geographical, economical, behavioral, diversification and recovery. It is likely the natural sociological and/or political considerations. The distribution of plant communities still would latter four involve value judgments based on an be different even if human intervention had not arbitrary frame time that results in controversy occurred. Similarly, species move geographically due to the diversity of human preferences and in response to environmental changes caused both needs (53). Thus, the native aspects of turfgrasses naturally and by human disturbances. Thus native addressed herein will rely on time and geographic sites should not be viewed as fixed geographical dimensions documented by sound, hard science. areas that existed prior to human activity. Also, There are proponents who consider the plant spe- the premise stated by some individuals that native cies present in America when Europeans arrived as turfgrasses inherently possess lower cultural the natives that have persisted from time imme- requirements, resource inputs, and/or water use morial free from human disturbances. However, rates, plus better human benefits than non-native recent research indicates there are woodland areas species is not valid (53). that have developed subsequent to anthropologi- Application of the native terminology for grasses cal disturbance during the late-Holocene at least has varied over the past century, and previously 4,000 years ago (80). These mound-building, early- emphasized the effects of tectonic processes American populations were active in domesticated that resulted in assumed geographic barriers crop propagation involving intensive land use by such as oceanic separation of continents. Some fire to remove forested areas. Such anthropological grass taxonomy and flora books include a geo- impacts on plant species disturbance and reinva- graphic description that proposes a continent(s) sion by intermediate succession species have been or region(s) where each species is native. In the recently documented for regions in southeastern early 1900s there were North American texts on and in north east central North America (33, 80), grass taxonomy that referred to grass species, such and others probably will be found. as the Agrostis, Festuca and Poa, as natives. Then by One can only speculate as to the grass species the mid-1900s most grass taxonomic, flora and that would have been naturally introduced and other texts listed such species as native to Eurasia.

2Assistance of Michigan State University Library personnel in the Turfgrass Information Center is acknowledged, including literature acquisition by Peter Cookingham and Michael Schury, and artwork by Aaron Tomak. Also, critical reviews of the section by Jeff V. Krans, Emeritus, Mississippi State University, William A. Meyer, Rutgers University, Robert B. Shaw, Texas A&M University, and Robert C. Shearman, Emeritus, University of Nebraska, are appreciated.

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 3 This change in the center-of-origin concept was The traditional grass taxonomic system is based based on plant exploration that indicated the on morphological similarities, especially the geographic region thought to possess the greatest inflorescence. A more recent classification system morphological diversity. Unfortunately, the true is cladistics, or phylogenetics systematics in center-of-origination for a species may no longer which groups or clades are organized based on exist botanically due to severe, extended drought relative dating of common ancestors with shared as occurred in Africa or due to ice age glacial diversity of anatomical, physiological, biochemical presence as in the Northern Hemisphere. Also, the and morphological features. Basically a clade may oceanic separation of continents is not proving to encompass all descendant species of an ancestor. be the migration barrier once assumed (48, 85). Two major grass clades are BEP encompassing the Bambusoideae, Ehrhartoideae, and Pooideae The development of molecular phylogenetic subfamilies and PACCMAD encompassing the techniques during the past two decades, involving Panicoideae, Arundinoideae, Chloridoideae, DNA sequencing instrumentation and unique phe- Centothecoideae, Micrairoideae, Aristidoideae, netic and cladistic analyses via computer programs and Danthonioideae subfamilies (15, 32, 42, 43, 77). for large data sets, have drastically changed our

understanding of when and where diversifications The BEP clade contains exclusively C3 grass

of grass genera, subgenera, species and subspe- species, while the PACCMAD clade includes C4 cies occurred. Thus, an update of the native or species, as well as all known C grasses. The latter naturalized status of the major turfgrass species is are mainly distributed within the Chloridoideae, addressed herein. Aristidoidae, and Panicoideae subfamilies. The crown node for the BEP clade is estimated to be ~ Naturalized Turfgrass Species. The term 10 million years older than that of the PACCMAD naturalized turfgrass species is defined herein as clade (17). Numerous subclades become evident a plant that is introduced, becomes established from the cladistic applications (2, 15, 25, 26, 39, and persists without direct human inputs. The 48, 51, 58, 62, 67, 85, 86), resulting in taxonomic introduction may be unintended or intended. An classifications that more nearly depict evolution introduced, heterozygous species may diversify of the grass family. Cladistic classification of the and naturalize as a local ecotype that is well monocotyledons indicates the Joinvilleaceae are adapted, functionally valuable, and non-invasive. a sister group to the Poaceae within the Subclass Also, multiple independent origins of similar Commelinidae (35, 51, 86, 89). genotypes have occurred in widely separated but similar environments, as has been documented Plant Evolution Phases. The geophysical for Agrostis capillaris, Agrostis stolonifera, and Poa annua aspects of grass species evolution involve a (a) (57). The plant community composition is inher- primitive ancestral center-of-origination, (b) early ently dynamic, with preservation of the status quo diversification, (c) migration or dispersal, and (d) on a long-term basis unlikely. It is natural to have species secondary centers-of-origin or diversifica- both migration/introduction and naturalization tion in order to adapt and survive changes. These of species. Anthropological related naturaliza- changes include environmental (temperature, tion contributes to species diversity of plant water, irradiation, and carbon dioxide), edaphic communities. (soil texture, pH, nutrients, and salts), and biotic (animal grazing, human activities, and pests). Taxonomy. The grass taxonomy employed herein There also are continuing intraspecies diversifica- is that of Barkworth et al. (6, 96, 97). The clas- tion regions in response to current natural and sification system for grasses is Kingdom - Plantae, human environmental disruptions. Accordingly, Phylum - Magnoliophyta (syn. Division), Class micro centers-of-diversification occur involving -Liliopsida (syn. Monocotyledonae), Subclass - small areas where mutations and hybridization are Commelinidae, Superorder-Poanae, Order - Poales, significant as represented by the presence diploid, and Family-Poaceae (syn. Gramineae). Subfamily triploid, tetraploid, hexaploid, and even octoploid names end in -oidea, tribes in -eae, and subtribes plant species (101). in -inae.

4 Origin, Biogeographical Migrations and Diversifications of Turfgrasses Previously a sometimes ill-defined center-of- land surface area. This was not always the case. origination had been considered phenotypically to Early evolutionary history of primitive ancestral be a region where the greatest diversity of a grass grasses and their broad migration patterns are species occurs. More recently, either confirma- thought to have been influenced by shifting posi- tion or major adjustments in dating and location tions of land masses (27) and seas that created and have been made possible by more sophisticated even destroyed environments where the ancestral paleobotany using ultrastructural electron mi- grasses originated or migrated. Plate tectonic theo- croscopic techniques plus stable carbon isotope ry has been advanced by new geological evidence dating instrumentation and research procedures. and contributes to concepts of angiosperm- Actually, the secondary species centers-of-origin Poaceae distributions. The supercontinent Pangea and/or diversification may be more important than formed ~ 500 mya. Its configuration as it existed the center-of-origination (48), especially if those in the medial Cretaceous prior to continental ancestors in the center-of-origination became breakup is shown in Figure 1. Pre-continental extinct as a result of adverse climatic changes, separation of Pangaea with possible ancestral grass such as the severe drought that occurred in Africa. migrations also are shown. Also, key secondary centers-of-diversification for Pangaea split into Laurasia, now North America temperate C3 grasses were formed after subsidence of the last glacial age. and Eurasia, and into Gondwana, now South America, Africa, Antarctica and Australia, dur- Current data indicate certain morphological ing the late-Mesozoic era ~ 160 mya. A land diversifications of grasses have occurred at least connection between Gondwanan-Africa and 23 mya (million years ago) after the origin of Laurasian-Europe was present in the western a particular character (51). Diversification is a Mediterranean region. The early Laurasian land dynamic ongoing process with multiple secondary connection between Europe and North America is micro-centers-of-origin. A grass population has the termed the North Atlantic Land Bridge. Gondwana potential for further differentiation that leads to was characterized by mild temperatures and a better genetic adaptation to a specific environment huge range of flora and fauna for millions of years. by natural selection. The early West Gondwana formation may have allowed South America to be more directly land Early Plate Tectonics. Grass-dominated ecosys- accessible to primitive ancestral plant migration tems now cover more than one-third of the earth’s from Africa than from Laurasian North America.

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 5 Figure 1. Depiction of the probable relative positions of the Laurasian and Gondwanan land masses in the medial Cretaceous ~ 100 mya (adapted from Smith, Briden and Drewry, 1973), and possible broad migration patterns of primitive ancestral grasses (PA), plus later migration of cool-season ancestral grasses (CA) and warm-season ancestral grasses (WA).

CA

PA

WA

6 Origin, Biogeographical Migrations and Diversifications of Turfgrasses Intercontinental Plant Migration. The move- an indirect land route from Asia via the Bering ment of plant species between continents and Land Bridge, and a more recently formed land directionally across large continental areas is route from South America via the Panamanian referred to as migration. Transcontinental disper- Land Bridge (Table 1). sals have been found to occur earlier and more frequently than was once assumed (48, 85). Wind, Note that moderate biogeographic plant migration birds, animals and/or floating debris may have by oceanic island-hopping probably occurred via contributed to such dispersals. Species distribu- birds, wind and/or flotation for a significant time tions and multiple centers-of-diversification tend following geologic dated continent separation. to be influenced by conditions existing since the A narrow gap interspersed with volcanic islands Pleistocene. Direct intercontinental plant migra- between Africa and South America presumably tion of primitive ancestral grasses was possible could have allowed primitive flora migrations up up to or following specified dating as extensively into the Paleocene (58, 72). It has been proposed reviewed by Raven and Axelrod (72) (Table 1). that tropical grass subfamilies may have continued Patterns of potential intermittent ancestral plant to migrate until continental separation reached a migration to North America (72) may have in- water gap of 745 miles (1,200 km) during the first volved a Laurasian direct land route from Europe, half of the Tertiary ~ 50 mya (28, 72).

Table 1. Summary of when possible early intercontinental plant migration patterns for primitive an- cestral grasses could have occurred (72).

Event Specific Continental Event Dating Occurrence Contact Event (millions of years ago)

Last Africa and India ~ 100 mya - late-Lower Cretaceous

Last Africa and South America ~ 90 mya - early-Upper Cretaceous

Last Africa and Eurasia, prior to Miocene event ~ 63 mya - early-Paleocene

Last Europe and North America ~ 49 mya - mid-Eocene

Last South America to Australia via Antarctica ~ 38 mya - mid-Eocene

First India contacts Asia ~ 45 mya - early-Eocene

First reestablished between Africa and Eurasia ~ 17 mya - mid-Miocene

First South America and North America ~ 5.7 mya - late-Miocene

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 7 Major island-hopping of animals and grasses is biogeographic analyses propose a dating range of probably represented by southeast Asia - Australia Poaceae at Late Cretaceous ~ 65 to 96 mya (16, 17, via New Guinea ~ 15 mya, and by South America - 26, 35, 58). North America via an intermittent land connection ~ 10 to 12 mya (72). Migration from Asia-Europe to Both diversifications and migrations of grasses North America via the Bering Land Bridge has oc- have been influenced by changing regional envi- curred since ~ 55 mya (72). The ancestral Bovidae ronmental, edaphic and biotic influences. Included migrated to North America from Europe-Asia via were changes in tectonic plate processes, ice the Bering Land Bridge (73). There also was a age glaciation, volcanic activity, tropical domi- land uprising to form the Gibraltar Strait Bridge nance, sea level, fire, water availability, favorable between Europe and Africa ~ 5 mya by which temperatures, seasonality, competitive flora and Bovidae migrated to Africa. Oceanic island-hop- fauna. Note that ice age climatic change during the ping between South America and North America Pleistocene was particularly disruptive in terms of may have occurred from ~ 45 mya. Mountain–hop- grass species diversification and migration in the ping from Europe to Asia to North America to Northern Hemisphere. South America may have occurred starting at least Diversification. Phytogeographical analyses sug- 10 mya for Agrostis, Festuca, and Poa (39, 48, 72). gest early differentiation of the lineages for tribes Primitive Ancestral Grasses. Definitive of the Poaceae may have started by the end of the evidence is limited concerning the continental Upper Cretaceous (26, 58). The Pooideae probably center-of-origination for primitive ancestral diversified initially in temperate environments at grasses. Fossil records for the grass family Poaceae high latitudes and high altitudes in the tropics within the angiosperms and their ancestors are (58). From ancestral origination in fringe-forested very incomplete. Based on paleobotanical evi- regions, there appears to have been further evolu- dence, including silicified plant tissues termed tion to savannas, and then to temperate steppes in the case of the Pooideae. The C cool-season an- phytoliths and stable carbon isotope studies, 3 primitive ancestral grasses appeared rather late cestral grasses are estimated to have arisen during in the earth’s history at somewhere between 55 the Eocene ~ 44 to 55 mya (30, 95). The Pooideae and 70 mya (43, 48, 51). It has been proposed are estimated to have migrated to the steppes of and supported by phylogenetic analyses that the Eurasia ~ 38 to 47 mya (17). Grass pollen fossils are Arundinoideae was the ancestor from which the more abundant during this epoch (64). cool-season monophyletic Pooideae arose (28, 52). The C4 physiological characteristics of warm- Available evidence suggests the primitive ances- season ancestral grasses evidently appeared as at tors of Poaceae occurred in the tropical forests least 12 distinct events and locations starting in and fringes of Gondwanan-Africa (17, 28, 58, 86). the Oligocene ~ 30 mya and subsequently thru 15 mya (16, 17, 43, 51, 54, 86). The C warm-season An extensive review of pollen fossil records by 4 Muller (64) indicates basal Poaceae were present ancestral grasses of the Panicoideae and the in the Paleocene epoch ~ 55 to 60 mya in central Chloridoideae are estimated to have originated to eastern Africa and east-central South America. ~ 26 mya and ~ 29 mya, respectively (17). The emergence of C grasses was associated with a Macro-fossil plant remains from a formation 4 drop in atmospheric carbon dioxide (CO ) levels in western Tennessee suggest the presence of a 2 common ancestor for the Poaceae under tropical, (16, 37) and/or seasonally drier climates (28, 56). seasonally dry conditions ~ 55mya (30). Macrofossils of Chloridoideae grass leaves from Kansas indicate C4 Kranz leaf anatomy existed This is a significant finding, relative to native in the Miocene ~ 5 to 7 mya (97). Also, evidence

grass species evolution in North America. The indicates that C4 grasses expanded in the Americas crown node of the BEP & PACCMAD clades has during the late-Miocene (56). been estimated phylogenetically at ~ 57 mya (17). Phytoliths preserved in coprolites from central Early grasses eventually became significant compo- India revealed five extant Poaceae subclades nents of plant communities in the dryer savannas, present during the Upper Cretaceous ~ 65 to 70 steppes, and prairies. The increasing dominance mya (70). Molecular phylogenetic and cladistic of grasses was probably aided by periodic

8 Origin, Biogeographical Migrations and Diversifications of Turfgrasses lightning-induced fires that destroyed most The same chaff-like, light-weight seed character- tree and shrub competitions, while the grasses istics also aid migration in flowing water down survived via basal tillering. Actually the dry grass streams and rivers, especially in association with biomass provided fuel for the fires. eroded soil. Stenotaphrum secundatum seed has an inflorescence structure that enhances flotation on It is postulated that since Gondwana phytogeo- ocean currents between islands and along coastal graphically separated into the present major shorelines (79). continents, numerous distinct grass diversifica- tions have occurred on four separate continents. The presence of a chemical germination inhibitor The original ancestral grasses have now evolved in the seed is another mechanism by which some into a large family of angiosperms with ~ 40 tribes, species of grass seeds extend the duration for ~ 800 genera, and ~ 11,000 species (26). potential migration. Poa pratensis is an example with modern cultivars having seed germination Migration. Grasses had eventually dispersed to delays of 3 to 4 months (10) following maturation. all continents ~ 60 million years after their ances- Certain earlier ancestral grasses probably had tral Gondwanan-Africa origin (17). Phylogenetic longer germination inhibition periods due to a studies suggest the first migration within the higher inhibitor content and/or greater resistance Pooideae from Eurasia to North America oc- to inhibitor degradation. curred ~ 38 mya via the North Atlantic Land Bridge when there was a global warming trend Some grass species have what is termed hard seed, (17). Subsequently there is evidenced that other in which the seed coat is impermeable to water BEP taxa migrated from Europe via Asia to North needed to initiate germination. Typical turfgrass America by the Bering Land Bridge (48). species with hard seeds include Cynodon dactylon, Eremochloa ophiuroides, and Zoysia japonica (10). Modes of seed dispersal away from the parent Germination can be improved by mechanical or plant can occur by animals, gravity, wind and acid scarification to open the seed coat barrier. water. Certain grass species possess characteris- Either the presence of a hard coat or an inhibitor tics that allow either internal or external transport in seed could facilitate distance dispersal of such by animals. Internal mechanisms can occur via the grass species in the digestive systems of birds digestive tract, primarily of large grazing animals and animals, especially herbaceous-grass feeding (66, 71). The ingestion and internal passage of animals (24). viable seeds through the digestive systems of ani-

mals has been documented for C3 grasses Agrostis Grazing Mammals and Turfgrasses. The an- stolonifera, Festuca rubra, Festuca elatior, Poa annua, Poa cestral species for some turfgrass species evolved

compressa, and Poa pratensis (34) and for C4 grasses in association with herbivorous grazing mammals. Axonopus fissifolius (19, 65), Buchloe dactyloides, Cynodon Early, now extinct browsing ungulates known as dactylon, and Paspalum notatum (19). Condylarths evolved while feeding on ancestral

C3 grass terrestrial ecosystems during the early Certain chaff-like grass seeds facilitate external Paleocene ~ 65 mya (8, 42). Subsequently, during transport on the fur coats and hairy hides of mobile the early-Eocene ~ 55 mya, early herbivorous ru- animals. Seeds with distinct awns, hairy lemmas, minant Artiodactyls arose. The Bovid subfamilies and/or sharp-pointed florets allow penetration in exhibited a rapid diversification~ 23 mya (63). The animal hair and resist easy withdrawal (24). Bovidae family, with a distinct tooth structure, Many grasses adapted for turfgrass purposes have arose in the late- Miocene ~ 5 to 9 mya. Evolving very small, light-weight seeds that can migrate via in Eurasia were cattle, antelope and bison of the wind. For example, the approximate number of subfamily Bovinae, plus goats, sheep, water buffalo seeds per pound (0.45 kg) for Agrostis capillaris > 5 and musk-ox of the subfamily Caprinae (63, 95). million, Cynodon dactylon > 1.5 million, Poa pratensis > Early grasses persisted as a minor component 1 million, and fine-leafFestuca > 400,000 (10). Large of a forested plant community for a long time. dust storms from Africa to the Americas may have Eventually, ancient grass-dominated ecosystems furthered intercontinental dispersal of certain developed in certain open regions ~ 5 to 6 mya as a grass seeds. result of significant changes to a drier environment

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 9 that caused a decline in tree populations (91). bedding composed primarily of grasses. Later During this time, certain herbivorous mammals colonists traveling via boat to the New World, also were evolving with mouth parts adapted to primarily from Europe, brought grasses for agrar- grazing more closely, and also with high-crowned ian forage and pasture purposes. These introduced teeth and/or constantly-growing teeth needed for grasses had been selected for more than 400 years grazing on grasses with high phytoliths or silici- based on a higher yield or biomass production fied fibers (61). Consequently, there was selection needed for feeding herbivorous domesticated toward grass species in certain regions that were animals (81). These early colonists preferred the structurally adapted to survive severe defoliation taller-growing, high-yielding, introduced grasses (88). for agrarian purposes. The European forage and pasture grasses were less desirable as turfgrasses, While open-habitat grasses exhibited taxonomic especially in early times of laborious manual mow- diversification in North America at least 34 mya, ing. Many European-introduced grass genotypes phylogenetic studies suggest they did not become naturalized in North America through intraspe- ecologically important until formation of the cies diversification and natural selection over a savanna-like Great Plains 23 to 27 mya (90). A period of 300 to 400 years. These grasses became major expansion of C grasses occurred during the 4 interspersed with pre-colonial native grass com- late Miocene ~ 6 to 8 mya during a major climate munities, especially in eastern North America. For change to a dryer environment and lower atmo- example, a good germplasm source for turfgrass spheric CO levels (23, 37, 56). Intercontinental 2 selection is in old cemeteries that have been distribution of both C and C grasses stabilized 3 4 mowed regularly, but typically not irrigated or during the late Miocene ~ 6 mya (61). fertilized. Clarification as to the specific genetic Carbon dating of herbivorous mammalian sources from which many contemporary turfgrass skeletons and attached grass remnants by paleo- cultivars were derived will require molecular botanical studies and DNA sequencing confirm phylogenetic research. a co-migration and even co-evolution among a Modern Diversification. Globally, there are ~ number of grazing mammals and certain peren- 40 grass species utilized as turfgrasses to varying nial grasses (16, 24). The interrelationship of degrees. Most are heterozygous with inflorescence low-growing grasses and herbivorous mammals fertilization by wind-aided cross-pollinated. The continued over an estimated period of ~ 10 to 20 result is a wide range in diversity with many grass million years (8). The result was a number of grass ecotypes that have specific adaptation potentials to species having (a) leaves with basal meristems, the numerous global variations in the environmen- (b) shoots with short basal internodes, and (c) tal, edaphic and biotic components contributing prostrate, creeping growth habits via lateral to the broad array of ecosystems. Furthermore the stems termed stolons and rhizomes. Evolution of diversification process for individual grass species these morphological adaptations in some grasses is ongoing with new genotypes emerging and some allowed subsequent use for turfgrass purposes genotypes possibly becoming extinct if poorly involving frequent, low mowing. adapted to the specific environment in certain Some centers-of-diversification for temperate ecosystems (81). grasses are of more recent post-Pleistocene de- To the untrained observer a turfgrass community, velopment ~ 2 to 1 mya due to glacial effects from such as a lawn, may appear to lack diversity. This the last ice age. Examples are the North American is because the basic leaf characteristics are more northeast and midwest regions and the higher similar in shape, orientation and color compared altitude meadowlands of Central Europe. to the broad-leaf dicotyledonous species such as Human Dispersal. Intercontinental human grass trees, shrubs, and flowers. Actually the genetic dispersals may have occurred even earlier than diversity is substantial in most mature turfgrass New World colonial times. Over 600 years ago lawn communities. It is a key to the capability of world voyagers and global explorers may have grasses to recover and survive as perennials under dumped from their ships at various world-wide wide seasonal variations in temperature, water, locations sweepings containing straw and hay shade, traffic, and pest stresses.

10 Origin, Biogeographical Migrations and Diversifications of Turfgrasses II, III VI VIII - ** Er IX, X ; VII. I IX I X Stenotaphrum I ** ** ; VI. Zoysia ; V. ; V. Cynodon V, VII ; IV. ; IV. . Solid lines depict preanthropological migration and migration . Solid lines depict preanthropological ** V IV Bouteloua ** I and ** ** IX Buchloe Schedonorus arundinaceus IV,V,VII ; X. ** ; III. ** I II Lolium Axonopus ; II. II III * * IV VI Agrostis * I IV (fine-leaf); IX. * VIII , and ** III IV Poa Paspalum ; VIII. Broad global depiction of the current understanding and probable post-Pangea centers-of-origination**, major migration major migration centers-of-origination**, post-Pangea and probable understanding global depiction of the current Broad II, III (fine-leaf), VIII Figure 2. Figure I. Designation: evolved. extant turfgrasses which from of early grasses centers-of-origin/diversification secondary and key routes, Festuca emochloa human introduction. dash lines represent

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 11 COOL-SEASON TURFGRASSES

The primitive ancestral center-of-origination for for the major species utilized in the United States. cool-season grasses (Pooideae) is thought to have Included are the fine-leaf fescues, bluegrasses, been from the tropical forest margins of Africa bentgrasses, ryegrasses and broad-leaf tall fescue. at higher altitudes (28). The unique evolutionary feature possessed by Pooideae was adaptation Fine-Leaf Fescue. The early species center-of- to cold climates that allowed migration to the origin for fine-leaf fescues Festuca( species) has temperate steppes of Eurasia (28). Subsequently been reported to be in central-Europe during the the Pooideae may have emerged under a tree- mid-Miocene ~ 13 mya based on phenotypic and shrub dominant, temperate ecosystem with biogeographical analyses (48). Festuca is reported sufficient precipitation for sustained growth and to be an ancient group and a main evolutionary possibly survival during occasional short periods line in diversification of the Poeae Tribe (19). Two of droughty conditions. Taxonomic divergence main Festuca clades are evident from phenetic and of the base Pooideae group appears to have been phylogenetic analyses, being fine-leaf and broad- initiated after a global super-cooling period ~ 26 leaf , that had split into diverging lineages (29, to 33.5 mya (78). Diversification also was probably 31, 93, 94). Within the fine-leafFestuca clade arose characterized by a cool temperate climate, plus the F. ovina and F. rubra groups (94). The latter is reasonably good precipitation and soil fertility. smaller in number of subgroups (69) and is yet Cool-season grass diversification in Eurasia may to be fully defined due to the paraphyletic nature have been greater in the mountainous areas where (2, 6, 10). The F. ovina complex is the oldest of shade from trees was reduced. Plant explorers3 the turfgrass-types (31), followed in lineage by F. report greater diversity within the Festuca, Poa and trachyphylla and F. rubra ssp. rubra. Based on phe- Agrostis species in Europe-Asia at altitudes above netic distance analysis F. ovina and F. trachyphylla 3,000 feet (915 m). are distinct, but closely related (2).

The cool-season turfgrasses discussed herein are Fine-leaf fescues probably diversified in cool, continental, open forest regions, possibly in taxonomically classified in the Pooideae, with C3 physiology (96). Within the Pooideae there are mountainous areas, and are included in the Festuca two supertribes, including Poodae, and seven subgenus (1, 2, 6, 7). Diversification of theFestuca tribes, including Poeae. The Pooideae includes ~ genus resulted in a major expansion, with ~ 400 3,300 species. The Poeae encompass ~ 115 genera known species (6). Their extant occurrence ranges and ~ 2,500 species (6), and occur with a high spe- from temperate to alpine to polar regions of both cies frequency percentage in the higher latitudes hemispheres (2). Chewing’s fescue (F. rubra ssp. and higher altitudes where summer and winter commutata), hard fescue (F. trachyphylla), sheep temperatures are colder (92). The cool-season fescue (F. ovina), slender-creeping red fescue (F. turfgrasses occur in the temperate climates of rubra ssp. littoralis), and strong-creeping red fescue both the northern and southern hemisphere and (F. rubra ssp. rubra) are the fine-leafFestuca species in mountains of the tropics, with an optimum now used as turfgrasses in the United States (10, temperature range of 60 to 75°F (16 to 24°C) (10, 11). These five fine-leafFestuca are seed propagated and largely cross-pollinated. The taxonomic 11). Migration bridges of certain C3 cool-season perennial grasses evidently occurred by mountain- classification of Chewing’s fescue as traditionally hopping. The likely migration route for the freeze used is open to question (2). The fine-leaf fescues tolerant Agrostis, Festuca and Poa species into migrated outward throughout much of Europe, the South American Andes was via the North into central and eastern temperate Asia (25), and American, Asian and European mountains (18, into Africa. Eventually they moved across the 48). In North America these three genera typically Bering Land Bridge into North America and mi- have their largest range of species diversity in the grated southward into Central America and over western mountainous regions. Geophysical aspects the Panamanian Land Bridge into South America of extant cool-season turfgrass will be addressed (Figure 2). The basic route in the Americas was

3 Personal communication with William A. Meyer of Rutgers University, New Jersey, USA. 12 Origin, Biogeographical Migrations and Diversifications of Turfgrasses north-to-south via the western mountains of Land Bridge to North America (72) is probable for North America to the Andes to Patagonia over a the bluegrasses (Figure 2), especially as many Poa period ~ 3.8 to 10 mya. This migration route has species have similar temperature responses includ- been documented among the Festuca species by ing good to excellent freeze stress tolerance (2, 9). biogeographical analyses (48). There also were Poa pratensis has a circumpolar distribution (85), outward migrations into the temperate climatic with a capability to develop numerous ecotypes regions of the continent. These fescues are used to thrive in distinctly different habitats (93). as a turfgrass in the cooler portions of the cool More than 60 species of Poa are now considered climatic regions of the United States, being espe- native to North America (6). Poa has a high spe- cially adapted to shaded and low-nitrogen fertility cies frequency percentage in regions with a July conditions (10, 11). mid-summer isotherm below 75° F (24°C) (47). Chloroplast-DNA phylogenetics and cladistic Most turfgrass-type fine-leaf fescue species analyses for biogeographical events revealed at (Festuca) are native to North America, having least six groups of Poa species had independently persisted for more than three centuries. There are colonized secondary centers-of-diversification in at least 37 Festuca species that are native to North North America, including Kentucky bluegrass and America (6, 7), including the red, hard, and sheep Texas bluegrass. The Poa are found extensively in fescues now used as turfgrasses. the western mountain ranges of North America Bluegrass. The early species center-of-origination (41), and the P. pratensis complex has been termed for the bluegrasses (Poa) based on phenotypic and native to the Rocky Mountains (50). The extant cladistic analyses (85) is temperate, open Eurasia. Kentucky bluegrasses of eastern United States The Poa genus is classified in thePoeae tribe (96), may be of an integrated ancestry with post-glacial reinvasion from the west or south and European with diversification being expansive to~ 500 known species. Global distribution studies reveal colonist introductions (22). the Poa species have the highest relative specific The Poa species utilized in the United States as differentiation in regions of high latitude and high turfgrasses resulting from diversification/migration altitude (47). Poa pratensis evolved and diversified include Kentucky bluegrass (P. pratensis), rough in geographical regions characterized by a cool bluegrass (P. trivialis), and Texas bluegrass (P. arach- continental climate and probably in fringe-open nifera) (10, 11). Poa pratensis is a facultative apomictic meadow areas interspersed in forests on relatively and highly polyploid. Kentucky bluegrass is widely fertile soils. Poa trivialis is thought to have evolved used as a turfgrass in the cool climatic regions of in the cooler, northerly, forest regions of Europe the United States under minimum shade, while under relatively wet climatic and soil conditions rough bluegrass is best adapted in moist to wet, on clayey soils. The diploid P. trivialis ssp. trivi- shaded conditions in the cooler portions of the alis has naturalized in North America (6). The cool-humid climatic region (10, 11). bluegrass species are propagated by seed and vegetatively by lateral stems such as rhizomes or Texas bluegrass is unique among the turfgrass- stolons. type Poas in adaptation to warm climates and is native to the southern Great Plains (85). It is best Poa annua is a ubiquitous, cosmopolitan grass that adapted to moist, fertile soils and bottom lands. is very well adapted to anthropological habitats. Available evidence suggests P. arachnifera or its an- It is considered an annual weed in some turfgrass cestor migrated north from South America, as two ecosystems, especially the bunch-type annual very close sister species occur there (85). Recently bluegrass (P. annua var. annua). The stoloniferous hybrid cultivars of P. pratensis x P. arachnifera have perennial creeping bluegrass (P. annua var. reptans) been developed for turfgrass use. tends to dominate when under a long-term man- aged component in certain turfgrass situations To conclude, a number of Poa species are possible (11). Poa annua is thought to have arisen from North American natives, including P. arachnifera hybridization between P. infirma and P. supina (6). and P. pratensis. Biogeographical analyses/molecular phylogenetic research with a range of Poa species A biogeographic migration pattern similar to the similar to that for the fine-leaf fescues are needed fine-leafed fescues through Asia via the Bering

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 13 for proper documentation, which should include P. conditions under which creeping bentgrass origi- annua and P. trivialis. nated in Eurasia. The extant bentgrass cultivars tolerate close, frequent mowing while sustaining Bentgrass. Based on phenotypic evidence the good shoot density (10, 11). early diverged species center-of-origination for the bentgrasses (Agrostis species) was in south-central The turfgrass-type Agrostis species have been Mediterranean Eurasia (84). The Agrostis genus is of viewed by some as being introduced to the United ancient origin, and has been morphologically clas- States due to oceanic separation from the early sified in the Poeae, with ~ 200 known species (6, center-of-origination. By inference from recently 96, 97) including at least 21 native North American available evidence, specifically from the fine-leaf species. The Agrostis probably diversified in an fescue research, the turfgrass-type bentgrasses environment characterized by fringe-forest zones would be considered possible natives to the of partial shade, and minimal moisture stress on United States. Hopefully, needed clarifying poorly-drained, clayey soils. Mountainous regions biogeographical analyses/molecular phylogenetic favored the origination of polyploid forms of investigations will be accomplished. Agrostis (84), with most adapted to narrow habitats (7). Diversification resulted in three bentgrass Ryegrass. The early species center-of-origination turfgrasses as now utilized in the United States, for the ryegrasses (Lolium species) based on ge- including colonial bentgrass (A. capillaris , syn. A. netic diversity was the open, temperate, southern tenuis), creeping bentgrass (A. stolonifera, syn. A. Eurasian region along the Mediterranean Sea. The palustris), and velvet bentgrass (A. canina) (10, 11). Lolium genus is classified in the Poeae (96), with diversification limited to~ 5 known species that Low levels of interspecific hybridization may occurred starting ~ 2 mya (6, 48). It has been pro- occur between A. stolonifera and both A. capillaris posed that the Lolium genus and Schedonorus pratensis and A. castellana, but are not likely for A. canina evolved from a common ancestor (74, 79), and split (13). Allotetraploids A. capillaris and A. stolonifera ~ 2.8 mya (48). Extant Lolium turfgrass cultivars

have the A2 subgenome in common (49, 85). tend to have limited shade adaptation and their Phylogenetic analyses and divergence time esti- center-of-origin and diversification probably were mates indicate these two species diverged from a characterized by open areas. common ancestor ~ 2.2 mya (76). Cluster analysis shows the primary cultivated species A. capillaris Propagation is by seed being cross-pollinated, and A. stolonifera are more similar to each other with most mowed turfgrass cultivars being than the diploid A. canina is to either (4). Agrostis bunch-types. The turfgrass-type species used in clavata has been suggested as a progenitor of A. the United States include perennial ryegrass (L. capillaris and A. stolonifera based on MITE - display perenne) and annual ryegrass (L. multiflorum) (10, marker clustering and ploidy prediction (3). Also 11). They are interfertile and naturally intergrade genetic marker studies suggest A. canina may have (6). The turfgrass-type perennial ryegrasses are contributed to the evolution of A. stolonifera as the utilized in the warmer portion of the cool climatic likely maternal parent (74). Turfgrass-type Agrostis region of the United States and for winter over- species are propagated primarily by seed, being seeding of warm-season turfgrasses. They are best largely cross pollinated. Agrostis stolonifera also can adapted to mild winters and cool-moist summers, be propagated vegetatively by stolons. and to fertile soils (10, 11).

A mountainous migration pattern of Agrostis spe- Migration of the Lolium was south into northern cies from Europe to North America (72) probably Africa, especially in mountainous regions, into occurred similar to that documented for the fine- northern Europe, and also eastward into Asia and leaf fescues (Figure 2). The turfgrass-type Agrostis India (Figure 2). Transcontinental pre-Holocene have good to excellent freeze stress tolerance (9, migration research is lacking. Lack of freeze stress 11). Northern salt marsh and lakeside populations tolerance (9, 10, 11) may have prevented migration of A. stolonifera are considered native to North to North America via the Bering Land Bridge. America (6). The species is well adapted to tem- Thus, based on the available phenotypic evidence porary flooding and wet soil sites, reflecting the the ryegrasses utilized for turfgrass purposes

14 Origin, Biogeographical Migrations and Diversifications of Turfgrasses in the United States are probably naturalized by seed being primarily cross-pollinated, with tall grasses first introduced by human trans-Atlantic fescue behaving as a bunch-type when regularly movement in the mid-1600s for forage and pas- mowed as a turfgrass. ture utilization in domestic animal agriculture. Confirmation is needed via biogeographical analy- Migration of S. arundinaceus from Europe extended ses/molecular phylogenetic research. eastward into Asia and south into Africa, espe- cially via the mountainous regions (Figure 2). Tall Fescue. The early species center-of- Potential transcontinental migration to the United origination for broad-leaf Schedonorus was in States prior to Holocene human activities has not mild-temperate Eurasia, separating from fine-leaf been properly investigated. Northward dispersal Festuca during the Miocene ~ 15 mya (48, 93). The in Europe may have been limited due to a lack of Schedonorus genus is classified in theSchedonorus freeze stress tolerance (9, 10, 11). Tall fescue is best subgenus (96, 07), with the turfgrass-type being adapted for turfgrass use in the transitional and allopolyploid tall fescue (Schedonorus arundinaceus, warmer-cool climatic regions of the United States syn. Festuca arundinacea) (6). Tall fescue probably under regular mowing at moderately high heights diversified under summers characterized by higher (10, 11). temperatures and less rainfall. S. arundinaceus is younger in lineage than the turfgrass-type fine-leaf Based on the limited available information, tall Festuca, and based on cluster analysis is closely fescue is considered a naturalized species thought related to Schedonorus pratensis (6, 99). Biological to have been first introduced into the United clock estimates indicate the Asian-American clade States by transcontinental human activities in the of broad-leaf Schedonorus apparently diverged from mid-1800’s for forage and pasture use in domestic the fine-leafFestuca clade ~ 12 mya (48). Certain animal production. Biogeographical analyses/ hexaploid turfgrass-type tall fescues, such as molecular phylogenetic research are needed for Rebel II and Bonanza, are clustered in a subgroup confirmation. according to cluster analysis (99). Propagation is

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 15 Related Events Related levels lowered levels lowered 2 South America isolated from Africa South America isolated from Africa from separate and Australia India, Antarctica, climate mild to tropical North Europe isolated from America-Greenland and climate tropical seas retreat of mammals spread rapid mammals herbivorous early India contacts Asia warm climate and seas flood the land Antarctica from separates Australia mountains form Alps and Rocky modern mammals emerge climate cools and becomes dryer widespread forests atmospheric CO frequency fire increased Himalayas and Andes rise sea level falls, Bovidae and Asia evolve in Europe of mammals, with most diversity age golden climate cools and becomes dryer position continents in present Bovidae Africa invade Land forms Panamanian Bridge covers northern lands and sea levellast ice age falls humans evolve & develop hunting skills of ice age of the ice sheets/glaciers withdrawal climate warms and sea level rises by humans development of agriculture rise of human civilization                                                           grass species grass 3 grasses to Africa grasses 3 warm-season grasses evolve warm-season grasses cool-season grasses arise cool-season grasses 3 4 Major Plant/Grass Events extinction of many species of African flora species of African flora of many extinction flowering plants evolution of nongrass pollen fossils reports of grass reports pollen fossils definitive of grass successful flowering plants become more plants appear extant primitive C - Poaceae evidence pollen fossils increasing of grass and macrofossils pollen fossils abundant grass diversification of C cover much of land and grasses trees of C migration primitive C in Europe fine-leafed fescues emerge - Festuca become ubiquitous globally grasses form globally grasslands temperate large emerge and grasses of grasslands expansion Plains on US Great in Africa forests many savannas replace Plains species of US Great shift in grass fauna and flora present use developed in last 800 years turfgrass diversification/selectionnatural of cultivars areas in mowed turfgrass                                             - 9.7 9.7 2.7 17.7 17.7 18.9 34.1 21.9 0.01 0.01 2.59 45.9 45.9 Approximate Approximate lions of years) Duration (mil Duration - 0 - 0.01 0 - 0.01 2.6 - 5.3 2.6 0.01 - 2.6 - 2.6 0.01 5.3 - 23.0 5.3 - 23.0 33.9 - 55.8 65.5 - 99.6 65.5 - 99.6 23.0 - 33.9 23.0 55.8 - 65.5 99.6 - 145.5 - 145.5 99.6 Cumulative Cumulative Time* (mil Time* lions of years) A representation of key events in grass origin and evolution during the latter portion of earth geological history, portion of earth geological history, during the latter origin and evolution in grass events of key A representation Epoch Lower/Early Cretaceous Upper/Late Cretaceous Paleocene Eocene Oligocene Miocene Pliocene Pleistocene Holocene (ongoing) Table 2. Table knowledge. based on current Geological Society of America (2009). The time scale is from geologic *The

16 Origin, Biogeographical Migrations and Diversifications of Turfgrasses WARM-SEASON TURFGRASSES

The primitive ancestral center-of-origination for Locations and dating for species centers-of-origin the warm-season grasses is thought to have been and geophysical migrations are increasingly based Gondwanan-Africa (28, 72), with West Gondwana on phenotypic, molecular phylogenetic, and allied migration to South American and East Gondwanan cladistic assessments. Fossil types of evidence migration to India and to Australia via Antarctica. are limited as fossilization was restricted in dry Key synapomorphies include Kranz leaf anatomy savanna climates where warm-season grasses and chloridoid bicellular microhairs on the leaf diversified. Geophysical characteristics of extant

epidermis (55). The C4 photosynthetic pathway warm-season turfgrasses are addressed for the characteristic of warm-season grasses evolved in major species utilized in the United States, includ-

Africa from C3 ancestral grasses at a later time in ing bermudagrass, zoysiagrass, St Augustinegrass, the earth’s history, and involved multiple bio- centipedegrass, seashore paspalum, carpetgrass,

chemical and histological C4 events (16, 17, 32, 51, American buffalograss, and gramagrass.

82). The first evidence of 4C photosynthesis in the Chloridoideae occurred ~ 33 to 30 mya (15, 17). Bermudagrass. The primary center-of-origina- tion for the bermudagrasses (Cynodon species) is Fossil records indicate the ancestral species in open, southeastern Africa (38). Subsequently of warm-season Poaceae originated in West early diversification was associated with graz- Gondwana, now Africa, mainly in savanna eco- ing pressure from large herds of herbivorous systems (72). Phylogenetic analyses suggest the African, hoofed mammals. Based on biostematic monophyletic Panicoideae and Chloridoideae and phylogenetic molecular research, the Cynodon are derived from the ancestral polyphyletic species secondary centers-of-origin/diversification Arundinoideae (32, 52). The PACCMAD clade include South Africa, India, Afghanistan, China, exhibited phylogenetic diversifications dur- and Australia (45). ing the mid-Miocene through late-Quaternary (16). Phylogenetic molecular studies reveal the The ubiquitous, cosmopolitan C. dactylon is now Chloridoidae may date as early as the late-Oligo- found throughout the warm-temperate, subtropi- cene ~ 31 mya (16, 17), but major global migration cal, and tropical climatic regions of the world (45). may not have occurred until the Pliocene (67). The Cynodon genus is classified in the Cynodonteae and Chloridinae, with limited diversification of Species of C4 turfgrass-types tend to have multiple secondary centers-of-origination in geographical ~ 9 known species and 10 varieties (6, 46, 67, 96). regions closer to the equator under warm climatic Cluster analysis of DNA amplification products conditions. Twenty-four shifts in phylogenetic has confirmed this earlier taxonomic classification. diversification rates have been identified among Evidently the genetic diversity of C. dactylon var. dactylon was generated internally (45). It thrives the C4 grass lineages, with 19 during the Miocene and 5 during the Pliocene (16). under soil disturbance on sunny sites. Low- growing Cynodon species used as turfgrasses in the Most warm-season turfgrasses are taxonomically United States include dactylon bermudagrass (C. classified within either the Chloridoideae or dactylon var. dactylon), a sterile hybrid bermuda- Panicoideae (96). The Chloridoideae has eight grass (C. dactylon x C. transvaalensis), and a natural tribes including Cynodonteae and Zoysieae, while sterile triploid hybrid (Cynodon x magennisii) (10, the Panicoideae encompasses two tribes includ- 11). Molecular phylogenetic analysis has identified ing the Paniceae. Extant warm-season turfgrasses cluster groups broadly based on relative leaf width occur mostly in the warm, tropical and subtropical (20). climates, with an optimum temperature range of 80 to 95°F (27 to 75°C). Most are prone to chill The presumed geophysical migration of the stress in the 52 to 58°F (11 to 15°C) range, with re- ancestral grass was via the East Gondwana land sultant low temperature discoloration and winter connection. Subsequent pre-Holocene migration dormancy (10, 11, 12). Extant Chloridoideae typi- was outward around Africa, and into southern cally are adapted to survive at mean temperatures Europe, Asia, and Australia (Figure 2). Propagation for the coldest months above 50° F (10° C) (64). is vegetatively by stolons and rhizomes, and

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 17 by seed for C.dactylon being cross-pollinated. close genetic relationship. Also used as a turfgrass The climate under which bermudagrass species is hybrid zoysiagrass (Z. japonica x Z. pacifica) evolved was subtropical with typically dry sum- cv. Emerald. Cluster analyses of RAPD revealed mers that resulted in a deep extensive root system, groupings that differ from the classical taxonomic low evapotranspiration rate, extensive lateral stem species classification (98). development, and superior drought resistance. The modern Cynodon turfgrass cultivars have East Gondwana migration of Zoysia primitive retained these water-conserving characteristics, ancestors probably occurred from Africa to but most lack shade adaptation (10, 11). India, although a West Gondwana migration of the Zoysiinae to South America cannot be ruled Unintended trans-Atlantic human movement of out entirely without additional investigations C. dactylon via ships of the Spanish conquistadors (67). Following the separation of India from probably occurred in the 1600’s to both North Africa and eventual contact of India with Asia, and South America (100), followed by further the Zoysia ancestors migrated to southeast Asia dispersal/naturalization and intraspecies diversifi- (72). Pre-Holocene migration and secondary cation in the warm climatic regions. The dactylon centers-of-origin/diversification forZoysia species bermudagrass (C. dactylon var. dactylon) introduced appear phenotypically to have been limited to the into the United States was readily dispersed, and Australasian region, extending north up into Japan became naturalized. The questions are when, by and Korea, south into northeast Australia, and to what means, and if extant native C. dactylon were the south Pacific Islands extending to Polynesia already present in the United States as a result of (Figure 2). The sandy coastal regions and salt earlier natural migration from Africa, possibly via marshes of the southwest Pacific have concentra- South America. In contrast, African bermudagrass tions of Z. matrella, including the Malay peninsula, (C. transvaalensis) was probably a human intro- New Guinea, Philippines, and southeast China duced species from South Africa to the United (39). Propagation has been primarily vegetative States, with minimal dispersal/naturalization. by rhizomes and stolons, with limited seed avail- Biogeographical analyses/molecular phylogenetic ability for Z. japonica. Seed germination among research are needed to clarify these dispersal Zoysia species ranges from poor to non-existent, aspects. which may have impaired their migration and diversification. Zoysiagrass. Phenotypically, the zoysiagrasses (Zoysia species) are recognized to have a post- Zoysiagrasses are used in the United States ancestral center-of-origin under warm-tropical, primarily for turfgrass purposes in the humid- open conditions in southeast China. Their primi- subhumid transitional temperature zone between tive ancestral grass center-of-origination probably the cool and warm climatic regions. The Zoysia occurred in Africa, where it is assumed they species diversified in open habitats under a subsequently became extinct during a severe arid tropical climate characterized by hot, humid period (72). The Zoysia genus is classified in the summers of southeast China with a large amount Cynodonteae and Zoysiinae (68), with diversifica- of precipitation distributed throughout the year. tion being relatively small with ~ 11 known species As a consequence, the zoysiagrasses evolved with (7, 67, 96). Shoots of the turfgrasstype Zoysia spe- a short root system that continues to be expressed cies possess a high silica content, which suggests in modern turfgrass-type cultivars which con- that herbivorous grazing mammals may not have tributes to a significant reduction in dehydration been as significant in their diversification and -mi avoidance and resultant less drought resistance gration. Based on molecular phylogenetic analyses than bermudagrass (10, 11). Zoysiagrass cultivars the Zoysia genus arose later than the Cynodon (15). used as turfgrasses are characterized by a very- Zoysia species now utilized for turfgrass purposes slow establishment rate. in the United States include Japanese zoysiagrass (Z. japonica), manila zoysiagrass (Z. matrella var. Japanese zoysiagrass was present in the United matrella), and to a lesser extent mascarene zoysia- States by the late-1800’s. It probably was inten- grass (Z. pacifica, syn. Z. matrella var. tenuifolia) (7, tionally introduced by man from southeast China 10, 11). These three readily intercross suggesting a and subsequently has naturalized. Biogeographic

18 Origin, Biogeographical Migrations and Diversifications of Turfgrasses analyses/molecular phylogenetic research are subtropical southeastern and south-central United needed to clarify the migratory/introduced status. States, extending from Florida to the southern coastal region of Texas (40). It is best adapted to St Augustinegrass. Evidently the Stenotaphrum sandy coastal sites and to moist, well-drained, genus is a paleotropic offshoot ofPaspalidium (79). fertile soils plus a relatively high mowing height The post-ancestral center-of-origination for the St (10, 11). The yellow-stigma Gulf Coast Group has Augustinegrasses (Stenotaphrum species), based on good shade adaptation, whereas the purple-stigma phenotypic diversity, suggests coastal southeastern group typically has both poor shade adaptation Africa since 6 of these 7 known species occur in and freeze stress tolerance. that region (79). The extensive occurrence of S. secundatum in the tropical and subtropical Gulf of Phenotypic evidence indicates St Augustinegrass Mexico Caribbean regions of the Americas sug- is a naturally diversified species, that has become gests a secondary center-of-origin/diversification, native to the United States with two secondary possibly via the West Gondwanan bridge from centers-of-origin/diversification. Biogeographical Africa via South America. The Stenotaphrum genus analyses/molecular phylogenetic research are is classified in the Paniceae and Setariinae, with needed to determine the early intercontinental diversification limited to~ 7 known species (6, migration routing and timing. 96). Centipedegrass. An East Gondwanan migration The Stenotaphrum species most probably migrated of the primitive ancestors to the Eremochloa species outward into tropical regions around the primary may have occurred from Africa to India. Then and secondary species centers-of-origin/diversi- from India to Asia following the separation of fication, especially in sandy coastal and stream India from Africa and eventual contact with Asia swale regions (79). Most propagation is vegetative (72). The post-ancestral center-of-origination for by stolons, although seed of some genotypes are centipedegrass (Eremochloa species), based on phe- viable being cross pollinated while others are notypic genetic diversity, is open temperate and self-sterile (79). The inflorescence is adapted for subtropical southeast Asia (40). The Eremochloa sub-regional island dispersal of seed by ocean cur- genus is classified in the Paniceae and Setariinae, rents (79). Transcontinental migration to southeast with diversification limited to~ 11 known species Asia and northeast Australia probably occurred (6, 96). E. ophiuroides is the only species of centipe- prior to human movements via the coastal areas degrass now utilized as a low-growing turfgrass along the Indian Ocean and then southwest via species in the United States (10, 11). tropical oceanic island hopping (79) (Figure 2). Migration of Eremochloa species appears to have One species dominates turfgrass usage in the been outward into tropical southeast and central United States, that being S. secundatum (10, 11). China (44), and to southeast Asia and to Australia Floral stigmas and stolon internodes of the (Figure 2). Propagation is by seed being a sexually Florida-centered, coarse-textured S. secundatum reproducing diploid with some self-incompatibili- group are purple, while the narrower-leafed Texas- ty, and vegetatively by stolons (44). centered Gulf Coast group have yellow stigmas and green stolon internodes (60). The purple The Eremochloa species evolved in an open tropical stigma types are triploids and tetraploids both climate with hot, humid summers and a large forming seeds that are highly sterile (60). This amount of precipitation distributed throughout would have limited their migration primarily to much of the year. As a consequence, centipede- Florida. The yellow stigma types are diploids that grasses evolved with a short root system that form seeds with a reasonable germination capabil- continues to be expressed in modern turfgrass- ity and considerable heterozygosity. This allowed type cultivars (11). Centipedegrass as utilized selection for improved freeze stress tolerance that for turfgrass purposes has a very-low nitrogen resulted in wider adaptation and migration along requirement (10, 11). the Gulf Coast and northward. Government records indicate one species of cen- The S. secundatum species of St Augustinegrass is tipedegrass was intentionally introduced into the used for turfgrass purposes in warm tropical and United States in 1916. Whether there were natural

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 19 intercontinental migrations to the Americas before Indies. This ecotype has become a naturalized Holocene human activities need investigation via species, persisting primarily in the brackish, sandy biogeographical analysis/molecular phylogenetic coastal regions of southeastern and south-central research. Until needed research is accomplished E. United States. ophiuroides should be considered as introduced into the United States. It has become naturalized to Carpetgrass. A post-ancestral, distinct second- the humid and subhumid southeastern and south- ary center-of-origin for the carpetgrasses (Axonopus central regions of the United States (40). species) is phenotypically considered to have been tropical Brazil and extending into the Caribbean Seashore Paspalum. The primitive ancestral region of the Americas (14, 40, 67). Their primitive center-of-origination for the Paspalums is thought ancestral grass and related geophysical aspects to be West Gondwana-African, with migration to may have involved origination in Africa and South America prior to continental plate separa- migration to east and central Brazil in South tion of Pangea. Thus, South America was a key America (14). The Axonopus genus is classified center-of-origin, typically in brackish areas and in the Paniceae and Setariine, with significant salt marshes. However, the species center-of- diversification indicated by~ 100 species (6). The origin for certain fine-leaf turfgrass-typePaspalum two low-growing, turfgrass-type species in use are species based on phenotypic evidence is tropical common carpetgrass (A. fissifolius, syn. A. affinis), and subtropical southern Africa, with subsequent and tropical or broadleaf carpetgrass (A. compressus) secondary diversification centers in coastal South (10, 11). They prefer open sites and wet soils that America and southeastern North America, based are periodically flooded. on phylogenetic molecular research and cluster analysis (59). The Paspalum genus is classified in Pre-Holocene migration of Axonopus species pre- the Paniceae and Setariine, with ~ 320 known sumably was outward regionally from the African species (6, 96) with seven cluster groups identified ancestral center-of-origination, and then from (59). There are very-coarse, intermediate, and fine- secondary centers-of-origin in South America, leaf species and ecotypes. The fine-leafP. vaginatum and eventually southeast Asia, and southeastern is the only species group utilized as a turfgrass in United States (67) (Figure 2). Propagation is the United States (36). vegetatively by stolons and by seed being a diploid with cross pollination and some self-sterility (21). Pre-Holocene migration of the fine-leafP. vaginatum Phenotypic observations reveal a wide-spread, expanded primarily around the tropical and extant global distribution in the tropical- subtropical coastal and stream swales of southern subtropical climatic regions of the Americas, Africa, that were characterized by brackish, marsh Southeast Asia-Pacific Islands, and Africa. In terms ecosystems including tidal flooding (Figure 2). of turfgrass use, common carpetgrass is found in Propagation is vegetative via rhizomes and stolons, the southeastern United States with limited use, and also can be by seed being a diploid with cross while tropical carpetgrass is widely used in south- pollination and considerable self-sterility (21). It is east Asia. Common carpetgrass is best adapted to most valuable as a turfgrass on soils with high salt acidic, moist to wet soils (10, 11). levels. Carpetgrass is native to the southeastern United Early introduction of fine-leafP. vaginatum into the States and the American tropics and subtrop- United States probably was by human activity via ics (6, 41). Biogeographical analyses/molecular ships to southeastern Atlantic coastal ports during phylogenetic research would elucidate the early the 1700’s (36), as well as into the coastal West diversification and migration of carpetgrass.

20 Origin, Biogeographical Migrations and Diversifications of Turfgrasses SEMI-ARID WARM/COOL-SEASON TURFGRASSES

The United States Great Plains extends from (87). The environmental causes of the shift to the Rocky Mountains eastward to the forested Buchloe and Bouteloua species probably were related region and from the Gulf of Mexico northward to to the northern Great Plains being covered with the coniferous forests of central Montana. This glacial ice plus soil changes in the southern por- extensive, temperate grassland has exhibited tions related to moraine sands and silts deposited radical, multiple changes in flora. Fossil evidence by flooding from large rivers flowing from the indicates species in the Stipeae, such as the now melting glaciers. Also there was a bison invasion infrequently found Piptochaetium, occurred in the ~ 0.5 to 1.2 mya (61) and expansion that formed Pleiocene ~ 23 to 27 mya (89), followed by species large dominate grazing herds where the horse had of Hordeum and Bromus (87). formerly been prominent (24). The present grasslands of the North American Several key warm-season, low-growing grasses Great Plains are relatively recent ecosystems that of Buchloe and Bouteloua species emerged dur- emerged savanna-like during the Miocene-Pliocene ing the Pleistocene. Their ancestral grass origin transition ~ 5 to 7 mya (5). Aridity increased may have been West Gondwanan, in Africa, during the Pliocene including a distinct drought with geophysical aspects involving continental season that resulted in a dry, flammable biomass migration to South America prior to significant of grass. This natural fuel combined with a smooth Pangea separation (72). Subsequent migration to rolling topography of the plains and active presumably occurred to Central America where winds favored fires. Lightning ignited the fires in the two low-growing semi-arid genera evolved earlier times and later humans employed fire to within a relatively comparable time frame (15 ). preserve the grasslands. Humans used grass fire Severe defoliation by the herbivorous grazing to harvest food, drive animals, reduce pests, clear bison resulted in diversification with tolerance land, provide openings for travel, and both as to subsequent mowed turfgrass usage. The two defensive and offensive strategies. An additional genera contributed to what has been termed the biotic factor was emergence of numerous browsing shortgrass prairie of the Great Plains (5, 40). and grazing mammals on the Great Plains during the late-Pleistocene. American buffalograss Buchloe( dactyloides) evolved in a secondary diversification center in Mexico Wheatgrass. Subsequently the tall-growing relatively recently (87), and then migrated north- wheatgrasses (Agropyron and Pascopyrum species) ward into the US Great Plains of North America migrated from the U.S. western mountain areas (Figure 2). Buchloe has only one known species into the temperate, dry northern Great Plains (87). and is classified in the Chlorideae (6, 96). It has

Previously they had migrated via the Bering Land C4 physiology, is usually dioecious, and is best Bridge from the open Euro-Mediterranean region adapted to the open, upland, semi-arid western to Asiatic China and eastward. They were key third of the Great Plains. Propagation is by seed components of what has been called the tallgrass and vegetatively by stolons.

prairie (5, 40). Both genera are C3 cool-season grasses classified in the Triticeae (96). There are~ The gramagrasses (Bouteloua species) evolved at a 15 Agropyron species and 1 Pascopyrum species (6, 96, secondary diversification center in open northern 97). Included were crested wheatgrass (A. cristatum) Mexico. Included were blue gramagrass (B. gracilis) and western wheatgrass (P. smithii), that have been and the apomictic side-oats gramagrass (B. cur- used as minimal maintenance, unirrigated turf- tipendula), that became major components of the grasses in the northern portion of the semi-arid short-grass prairie. The Boutelous genus is classified region (10, 11). Western wheatgrass is best adapted in the Cynodonteae and Boutelouinae, with ~ 40 known species and C physiology (6, 67, 68, 96). to alkaline, mesic meadow-like conditions (7). 4 They also exhibited a northward migration into American Buffalograss and Gramagrass. the United States (Figure 2). Propagation is by Another major shift in grass species dominating seed and vegetatively by rhizomes. the Great Plains occurred during the Pleistocene

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 21 The three grass species have been utilized as environments of the Great Plains. However, the turfgrasses primarily under unirrigated conditions high density, more aggressive turfgrass species, in the Great Plains region of the United States (10, such as Kentucky bluegrass and bermudagrass, 11). They are best adapted to semi-arid climates tend to dominate these three species in higher where the precipitation is not sufficient to sup- precipitation regions (10, 11). Also, the semi-arid port most tree and shrub species (10, 11). They environments in which these genera evolved tend to become dormant during extensive sum- resulted in genotypes that were not subjected to mer droughts, their drought survival is achieved pathogen pressures common to humid regions and primarily by a dormancy escape mechanism, rather consequently did not result in strong selection for than an inherent dehydration tolerance component resistance to such diseases. of drought resistance. These semi-arid, warm- season grasses have a relatively moderate shoot American buffalograss and blue and side-oats density. gramagrasses were originally naturalized species to the United States, that also may be termed Perennial turfgrasses with a higher water require- native species. ment are less favored in unirrigated, semi-arid

SUMMARY COMMENTS

Certain groups advocate that only plants and is appropriate and fortunate that some individu- grasses native to a particular region should be als are interested in preserving and propagating grown in that landscape. A premise for this con- native grass stands in various climatic regions, it is cept is that native grasses have superior adaptation unwise and even detrimental to promote laws or compared to any introduced, naturalized grasses. other means that force all individuals to use only There are naturalized grasses as well adapted or native grasses in landscape plantings throughout even better adapted to certain climatic-soil regions a city or region. Advocating only the use of native of the world than the native grasses. Actually plants in what is essentially an artificial non- neither category is genetically stable, rather they native environment of urban structures, concrete, continue to diversify. asphalt, and disturbed lands is not logical. Many plants arbitrarily identified as native to a local The term native has been used as an arbitrary con- natural environment may not be adapted to a cept to access whether a species belongs naturally nearby urban ecosystem, while certain natural- at a specific site. However, it is a questionable ized plant species may be better adapted and approach to separate a natural process from the more capable of enhancing the functional human presence and influence of human activity. While it quality-of-life in urban areas.

22 Origin, Biogeographical Migrations and Diversifications of Turfgrasses REFERENCES: 1. Aiken, S.G., M.J. Dallwitz, C.L. McJannet, and between Agrostis stolonifera and related Agrostis under L.L. Consaul. 1996 onwards. Festuca of North field conditions. Crop Science. 43:240-246. America: descriptions, illustrations, identification 14. Black, G.A. 1963. Grasses of the genus Axonopus, a and information retrieval. http://delta-intkey.com/ taxonomic treatment. Advancing Frontiers of Plant festuca. Science. 5:1-186. 2. Aiken, S.G., M.J. Dallwitz, C.L. McJannet, and L.L. 15. Bouchenak-Khelladi, Y., N. Salamin, V. Savolainen, Consaul. 1997. Biodiversity among Festuca (Poaceae) F. Forest, M. van der Bank, M.W. Chase, and T.R. in North America: diagnostic evidence from DELTA Hodkinson. 2008. Large multi-gene phylogenetic and clustering programs, and an INTKEY package trees of the grasses (Poaceae): progress towards for interactive, illustrated identification and complete tribal and generic level sampling. information retrieval. Canadian Journal of Botany. Molecular Phylogenetics and Evolution. 47:488-505. 75:1527-1555. 16. Bouchenak-Khelladi, Y., G. A. Verboom, T. R. 3. Amundsen, K.L. 2009. Origin and evolution of Hodkinson, N. Salamin, O. Francois, G. N. cultivated Agrostis spp. PhD dissertation, George Chonghaile, and V. Savolainen. 2009. The origins Mason University, Fairfax, Virginia, USA. 144 pages. and diversification of C grasses and savanna 4. Amundsen, K. and S. Warnke. 2011. Species adapted ungulates. Global Change Biology. 15:2397- relationships in the genus Agrostis based on flow 2417. cytometry and MITE-display molecular markers. 17. Bouchenak-Khelladi, Y., G.A. Verboom, Crop Science. 51:1224-1231. V. Savalainen, and T.R. Hodkinson. 2010. 5. Axelrod, D.I. 1985. Rise of the grassland biome, Biogeography of the grasses (Poaceae): a central North America. The Botanical Review. phylogenetic approach to reveal evolutionary history 5(2):163-201. in geographical space and geological time. Botanical 6. Barkworth, M.E., K.M. Capels, S.Long, and M.B. Journal of the Linnean Society. 162:543-557. Piep (coeditors). 2003. Flora of North America, 18. Burkart, A. 1975. Evolution of grasses and Magnoliophyta: Commelinidae (in part): Poaceae. grasslands in South America. Taxon. 24(1):53-66. Oxford University Press, New York City, New York, 19. Burton, G.W. and J.S. Andrews. 1948. Recovery and USA. 25 (part 2): 809 pp. viability of seed of certain southern grasses and 7. Barkworth, M.E., L.K. Anderton, K.M. Capels, S. lespedezas passed through the Bovine digestive Long , and M.B. Piep (coeditors). 2007. Manual tract. Journal of Agricultural Research. 76:95-103. of Grasses for North America. Intermountain 20. Caetano-Anollés, G., L.M. Callahan, P.E. Herbarium and Utah State University Press Logan, Williams, K.R. Weaver, and P.M. Gresshoff. 1995. Utah, USA. 630 pp. DNA amplification fingerprinting analysis of 8. Barnard, C. and O.H. Frankel. 1964. Grass, grazing bermudagrass (Cynodon): genetic relationships animals, and man in historic perspective. (In between species and interspecific crosses. Grasses & Grasslands, ed. C. Barnard). Macmillan Theoretical Applied Genetics. 91:228-235. & Company Ltd., London, UK. p. 1-12. 21. Carpenter, J.A. 1958. Production and use of seed 9. Beard, J.B. 1966. Direct low temperature injury in sea-shore paspalum. Journal of the Australian of nineteen turfgrasses. Quarterly Bulletin of Institute of Agricultural Science. 24:252-256. the Michigan Agricultural Experiment Station. 22. Carrier, L. and K.S. Bort. 1916. The history of 48(3):377-383. Kentucky bluegrass and white clover in the 10. Beard, J. B. 1973. Turfgrass: Science and Culture. United States. Journal of the American Society of Prentice-Hall, Inc., Englewood Cliffs, New Jersey, Agronomy. 8(4):256-266. USA. 658 pp. 23. Cerling, T.E., Y. Wang, and J. Quade. 1993.

11. Beard, J.B and H.J. Beard. 2005. Beard’s Turfgrass Expansion of C4 ecosystems as an indicator of Encyclopedia for Golf Courses, Grounds, Lawns, global ecological change in the late Miocene. Sports Fields. Michigan State University Press, East Nature. 361:344-345. Lansing, Michigan, USA. 513 pp. 24. Chapman, G.P. 1996. The Biology of Grasses. CAB 12. Beard, J. B and P. O. Cookingham. 2007. William International, Wallingford, Oxon, UK. 273 pp. J. Beal — Pioneer applied botanical scientist 25. Chen, X., S.G. Aiken, M.J. Dallwitz, and P. and research society builder. Agronomy Journal. Bouchard. 2003. Systematic studies of Festuca 99:1180-1187. (Poaceae) occurring in China compared with taxa 13. Belanger, F.C., T.R. Meagher, P.R. Day, K. Plumley, in North America. Canadian Journal of Botany. and W.A. Meyer. 2003. Interspecific hybridization 81:1008-1028. Origin, Biogeographical Migrations and Diversifications of Turfgrasses 23 26. Clark, L.G., W. Zhang, and J. F. Wendel. 1995. A 39. Goudswaard, P.C. 1980. The genus Zoysia phylogeny of the grass family (Poaceae) based on (Gramineae) in Malesia. Blumea 26:169-175. ndhF sequence data. Systematic Botany. 20(4):436- 40. Gould, F. W. 1975. The Grasses of Texas. Texas 460. A&M University Press, College Station, Texas, 27. Clayton, W.D. 1975. Chorology of the genera of USA. 653 pp. Gramineae. Kew Bulletin. 30:111-132. 41. Gould, F.W. and R.B. Shaw. 1983. Grass Systematics. 28. Clayton, W.D. 1981. Evolution and distribution of Texas A&M University Press, College Station, grasses. Annals of the Missouri Botanical Garden. Texas, USA. 397 pp. 68:5-14. 42. Grass Phylogeny Working Group. 2000. A phylogeny 29. Clayton, W.D., R. Cerros-Tlatilpa, M.S. Kinney, of the grass family (Poaceae) as inferred from M.E. Siqueiros-Delgado, H.L. Bell, M.P. Griffith, eight character sets. (In Grasses: systematics and and N.F. Refulio-Rodiguez. 2007. Phylogenetics of evolution. S.W.L. Jacobs and J.E. Everett). CSIRO, Chloridoideae (Gramineae): a preliminary study Melbourne, Australia. p. 3-7. based on nuclear ribosomal internal transcribed 43. Grass Phylogeny Working Group. 2001. Phylogeny spacer and chloroplast trnL-F sequences. (In and subfamilial classification of the grasses Monocots: Comparative Biology and Evolution— (Poaceae). Annals of the Missouri Botanical Poales, eds. J.T. Columbus, E.A. Fariar, J.M. Porter, Garden. 88(3):373-457. L.M. Prince and M.G. Simpson). Rancho Santa Ana 44. Hanna, W.W. and G.W. Burton. 1978. Cytology, Botanic Garden, Claremont, California, USA. pp. reproductive behavior, and fertility characteristics 565-579. of centipedegrass. Crop Science. 18:835-837. 30. Crepet, W.L. and G. D. Feldman. 1991. The earliest 45. Harlan, J.R. and J.M.J. de Wet. 1969. Sources of remains of grasses in the fossil record. American variation in Cynodon dactylon (L.) Pers. Crop Science. Journal of Botany. 8:1010-1014. 9:774-778. 31. Darbyshire, S.J. and S.I. Warwick. 1992. Phylogeny 46. Harlan, J.R., J.M.J. de Wet, K.M. Rawal, M.R. of North American Festuca (Poaceae) and related Felder, and W.L. Richardson. 1970. Cytogenetic genera using chloroplast DNA restriction site studies in Cynodon L.C. Rich. (Gramineae). Crop variation. Canadian Journal of Botany. 70:2415-2429. Science. 10:288-291. 32. Davis, J.I. and R.J. Soreng. 1993. Phylogenetic 47. Hartley, W. 1961. Studies on the origin, evolution, structure in the grass family (Poaceae), as and distribution of the Gramineae. IV. The genus determined from chloroplast DNA restriction site Poa L. Australian Journal of Botany. 9:152-161. variation. American Journal of Botany. 80:1444-1454. 48. Inda, L.A., J.G. Segarra-Moragues, J. Müller, P. M. 33. Delcourt, P.A., H.R. Delcourt, C.R. Ison, W.E. Peterson, and P. Catalán. 2008. Dated historical Sharp, and K.J. Gremillion. 1998. Prehistoric human biogeography of the temperate Loliinae (Poaceae, use of fire, the eastern agricultural complex, and Pooideae) grasses in the northern and southern Appalachian oak-chestnut forests: paleoecology of hemispheres. Molecular Phylogenetics and Cliff Palace Pond, Kentucky. American Antiquity. Evolution. 46:932-957. 63(2):263-278. 49. Jones, K. 1956. Species differentiation inAgrostis . 34. Dore, W.G. and L.C. Raymond. 1942. Pasture Part II. The significance of chromosome pairing Studies XXIV. Viable seeds in pasture soil and in the tetraploid hybrids of Agrostis canina subsp. manure. Scientific Agriculture. 23(2):69-79. Montana Hartmn., A. tenuis Sibth. and A. stolonifera L. 35. Doyle, J.J., J.I. Davis, R.J. Soreng, D. Garvin, and Journal of Genetics. 54:377-393. M.J. Anderson. 1992. Chloroplast DNA inversions 50. Keck, D.D. 1965. Poa. (In California Flora, A. Munz). and the origin of the grass family (Poaceae). University of California Press, Berkeley, California, Proceedings of the National Academy of Sciences. USA. p. 1482-1490. USA. 89:7722-7726. 51. Kellogg, E. A. 2000. The grasses: A case study in 36. Duncan, R.R. and R.N. Carrow. 2000. Seashore macroevolution. Annual Review of Ecology and Paspalum The Environmental Turfgrass. Ann Arbor Systematics. 31:217-238. Press, Chelsea, Michigan, USA. 281 pp. 52. Kellogg, E.A. and C.S. Campbell. 1987. Phylogenetic 37. Ehleringer, J.R., R.F. Sage, L.B. Flanagan, and R.W. analyses of the Gramineae. (In Grass Systematics Pearcy. 1991. Climate change and the evolution of and Evolution. ed. T.R. Soderstron, K.W. Hilu, C.S. C photosynthesis. Trends Ecological Evolution 4 Campbell and M.E. Barkworth). Smithsonian Inst. 6(3):95-99. Press, Washington, D.C., USA. pp. 310-322. 38. Forbes, I., Jr. and G.W. Burton. 1963. Chromosome 53. Kendle, A.D. and J.E. Rose. 2000. The aliens have numbers and meiosis in some Cynodon species and landed! What are the justifications for ‘native only’ hybrids. Crop Science. 3:75-79.

24 Origin, Biogeographical Migrations and Diversifications of Turfgrasses policies in landscape plantings? Landscape and 67. Peterson, P. M., J. T. Columbus, and S. J. Pennington. Urban Planning. 47:19-31. 2007. Classification and biogeography of new world 54. Kingston, J.D., B.D. Marino, and A. Hill. grasses: Chloridoideae. Aliso. 23:580-594. 1994. Isotopic evidence for Neogene hominid 68. Peterson, P.M., K. Romaschenko, and G. Johnson. paleoenvironments in the Kenya Rift Valley. 2010. A classification of the Chloridoideae Science. 264:955-959. (Poaceae) based on multi-gene phylogenetic trees. 55. Krans, J.V., J.B Beard, and J.F. Wilkinson. 1979. Molecular Phylogenetics and Evolution. 55:580-598.

Classification of 3C and C4 turfgrass species based 69. Piper, C.V. 1906. North American species of Festuca.

on CO2 compensation concentration and leaf Contribution from the U.S. National Herbarium, anatomy. HortScience. 14:183-185. Washington, D.C. 10:1-48. 56. Latorre, C., J. Quade, and W.C. McIntosh. 1997. 70. Prasad, V., C. A.E. Strömberg, H. Alimohammadian,

The expansion of C4 grasses and global change in and A. Sahni. 2005. Dinosaur coprolites and the the late Miocene: Stable isotope evidence from the early evolution of grasses and grazers. Science. Americas. Earth and Planetary Science Letters. 310:1177-1180. 146:83-96. 71. Quinn, J.A., D.P. Mowrey, S.M. Emanuele, and 57. Levin, D.A. 2001. The recurrent origin of plant races R.D.B. Whalley. 1994. The “Foliage is the Fruit” and species. Systemic Botany. 26(2):197-204. hypothesis: Buchloe dactyloides (Poaceae) and the 58. Linder, H.P. 1987. The evolutionary history of the shortgrass prairie of North America. American Poales/Restionales — a hypothesis. Kew Bulletin. Journal of Botany. 81(12):1545-1554. 42(2):297-318. 72. Raven, P.H. and D.I. Axelrod. 1974. Angiosperm 59. Liu, Z.W., R.L. Jarret, R.R. Duncan, and S. biogeography and past continental movements. Kresovich. 1994. Genetic relationships and variation Annals of the Missouri Botanical Garden. 61:539-673. among ecotypes of seashore paspalum (Paspalum 73. Rezaei, H.R., S. Naderi, I.C. Chintauan-Marquier, vaginatum Swartz) determined by random amplified P. Taberlet, A.T. Virk, H.R. Naghash, D. Rioux, M. polymorphic DNA (RAPD) markers. Genome. Koboli, and F. Pompanon. 2010. Evolution and 37:1011-1017 taxonomy of the wild species of the genus Ovis 60. Long, J.A. and E.C. Bashaw. 1961. (Mammalia, Artiodactyla, Bovidae). Molecular Microsporogenesis and chromosome numbers in St. Phylogenetics and Evolution. 54:315-326. Augustinegrass. Crop Science. 1:41-43. 74. Rotter, D., K.V. Ambrose, and F.C. Belanger. 2010. 61. MacFadden, B.J. 2000. Origin and evolution of the Velvet bentgrass (Agrostis canina L.) is the likely grazing guild in Cenozoic New World terrestrial ancestral diploid maternal parent of allotetraploid mammals. (In Evolution of herbivory in terrestrial creeping bentgrass (Agrostis stolonifera L.). Genetic vertebrates: perspectives from the fossil record. ed. Resource Crop Evolution. 57:1065-1077. H. Sues). Cambridge University Press, Cambridge, 75. Rotter, D., K. Amundsen, S.A. Bonos, W.A. Meyer, England, UK. pp. 223-244. S.E. Warnke, and F.C. Belanger. 2009. Molecular 62. Mathews, S., R.C. Tsai, and E. Kellogg. Phylogenetic genetic linkage map for allotetraploid colonial structure in the grass family (Poaceae): evidence bentgrass. Crop Science. 49:1609-1619. from the nuclear gene phytochrome B. American 76. Rotter, D., A.K. Bharti, H.M. Li, C. Luo, S.A. Bonos, Journal of Botany. 87:96-107. S. Bughrara, G. Jung, J. Messing, W.A. Meyer, 63. Matthes, C.A. and S.K. Davis. 2001. Molecular S. Rudd, S.E. Warnke, and F.C. Belanger. 2007. insights into the evolution of the Family Bovidae: Analysis of EST sequences suggests recent origin A nuclear DNA perspective. Molecular Biology and of allotetraploid colonial and creeping bentgrass. Evolution. 18(7):1220-1230. Molecular Genetics and Genomics. 278:197-209. 64. Muller, J. 1981. Fossil pollen records of extant 77. Sanchez-Ken, J.G., L.G. Clark, E.A. Kellogg, and angiosperms. The Botanical Review. 47:1-140. E.E. Kay. 2007. Reinstatement and emendation of 65. Neto, M.S., R.M. Jones, and D. Ratcliff. 1987. subfamily Micrairoideae (Poaceae). Systematic Recover of pasture seed ingested by ruminants. Botany. 32:71-80. 1. Seed of six tropical pasture species fed to 78. Sandve, S.R. and S. Fiellheim. 2010. Did gene family cattle, sheep and goats. Australian Journal of expansions during the Eocene-Oligocene boundary Experimental Agriculture. 27:237-246. climate cooling play a role in Pooideae adaptation 66. Ortmann, J., W.H. Schacht, and J. Stubberndieck. to cool climates? Molecular Ecology. 19:2075-2088. 1998. The “Foliage is the Fruit” hypothesis: complex 79. Sauer, J.D. 1972. Revision of Stenotaphrum adaptations in buffalograss Buchloe( dactyloides). (Gramineae: Paniceae) with attention to its America Midland Naturalist. 140(2):252-263. historical geography. Brittonia. 24:202-222.

Origin, Biogeographical Migrations and Diversifications of Turfgrasses 25 80. Scharf, E. 2010. Archaeology, land use, pollen and Chloridoideae) from the Miocene with Kranz restoration in the Yazoo Basin. Veget. History anatomy. Science. 233:876-878. Archaeobotany. 19:159-175. 93. Torrecilla, P. and P. Catalán. 2002. Phylogeny 81. Scholz, H.1975. Grassland evolution in Europe. of broad-leaved and fine-leavedFestuca lineages Taxon. 24(1):81-90. (Poaceae) based on nuclear ITS sequences. 82. Sinha, N.R. and E.A. Kellogg.1996. Parallelism and Systematic Botany. 27:241-251.

diversity in multiple origins of C4 photosynthesis in 94. Torrecilla, P., J.A. Rodriguez, D. Stanick, and P. grasses. American Journal of Botany. 83:1458-70. Catalán. 2003. Systematics of Festuca L. Sects. 83. Smith, A.G., J.C. Briden, and G.E. Drewry. 1973. Eskia Willk., Pseudatropis Kriv., Amphigenes (Janka) Phanerozoic world maps. (In Organisms and Tzvelev., Pseudoscariosa Kriv., and Scariosae Hack. Continents through Time. ed. N.F. Hughes.) based on analyses of morphological characters and Palacontol. Assian. London, Special Paper DNA sequences. Plant Systematics and Evolution. Palacontol. 12:1-43. 239:113-139. 84. Sokolovskaya, A.P. 1938. A caryo-geographical study 95. Vislobokova, I.A. 2008. The major stages in the of the genus Agrostis. Cytologia (Tokyo). 8:452-467. evolution of Artiodactyl communities from the 85. Soreng, R. J. 1990. Chloroplast-DNA phylogenetics Pliocene-early middle Pleistocene of northern and biogeography in a reticulating group: study in Eurasia: Part 1. Paleontological Journal. Poa (Poaceae). American Journal Botany. 77(11):1383- 42(3):297-312. 1400. 96. Watson, L. and M.J. Dallwitz. 1992. The Grass 86. Soreng, R.J. and J.I. Davis. 1998. Phylogenetics and Genera of the World. CAB International, character evolution in the grass family (Poaceae): Wallingford, Oxon, UK. 1024 pp. simultaneous analysis of morphological and 97. Watson, L. and M.J. Dallwitz. 1992 onwards. chloroplast DNA restriction site character sets. The The grass genera of the world: descriptions, Botanical Review. 64:1-85. illustrations, identification, and information 87. Stebbins, G. L. 1975. The role of polyploid retrieval; including synonyms, morphology, complexes in the evolution of North American anatomy, physiology, phytochemistry, cytology, grasslands. Taxon. 24(1):91-106. classification, pathogens, world and local distribution, and references. Version 23rd April 88. Stebbins, G.L. 1981. Coevolution of grasses and 2010. http://delta-intkey.com/grass. herbivores. Annals of the Missouri Botanical Garden. 68:75-86. 98. Weng, J.H., M.J. Fan, C.Y. Lin, Y.H. Liu, and S.Y Huang. 2007. Genetic variation of Zoysia as revealed 89. Stevenson, D.W. and H. Loconte. 1995. Cladistic by random amplified polymorphic DNA (RAPD) analysis of monoct families. (In Monocotyledons: and isozyme pattern. Plant Production Science. Systemics and Evolution, eds. P.J. Rudall, P.J. 10(1):80-85. Cribb, D.F. Cutler, and C.J. Humphries). Royal Botanic Gardens, Kew, England, UK. pp. 543-578. 99. Xu, W.W. and D.A. Sleper. 1994. Phylogeny of tall fescue and related species using RFLPs. Theoretical 90. Strömberg, C.A.E. 2005. Decoupled taxonomic and Applied Genetics. 88:685-690. radiation and ecological expansion of open- habitat grasses in the Cenozoic of North America. 100. Wu, Y. 2011. Cynodon (In Wild Crop Relatives: Proceedings of the National Academy of Science. Genomic and Breeding Resources, Millets and 102:11980-11984. Grasses. ed. C. Kobe). Springer-Verlag, Berlin, Germany. 53-71. 91. Strömberg, C.A.E. 2011. Evolution of grasses and grassland ecosystems. Annual Review of Earth and 101. Zhukovsky, P.M. 1968. New centres of the origin Planetary Sciences. 39:517-544. and new gene centres of cultivated plants including specifically endemic micro-centres of species 92. Thomasson, J.R., M.E. Nelson, and R.J. closely allied to cultivated species. Botanicheskii Zakrzewski. 1986. A fossil grass (Gramineae: Zhurnal. 53:430-460.

Michigan State University | College of Agriculture & Natural Resources | AgBioResearch MSU is an affirmative-action, equal-opportunity employer.

SR132 1P-Web-1-16-13-VO 26 Origin, Biogeographical Migrations and Diversifications of Turfgrasses