Index | Search | Home | Table of Contents

Taylor, N.L. 1990. The true clovers. p. 177-182. In: J. Janick and J.E. Simon (eds.), Advances in new crops. Timber Press, Portland, OR.

The True Clovers*

Norman L. Taylor

  7. Table 1
  8. Table 2
  9. Table 3
  10. Fig. 1


The true clovers (Trifolium) include about 248 leguminous species, many of which are important forage', throughout the world. They are located in three main centers of diversity: Eurasia, South Africa, and the Americas. The four most important species in the United States are red clover (Trifolium pratense L.), white clover (T. repens L.), crimson clover (T. incarnatum L.) and alsike clover (T. hybridum L). In addition to their value for livestock feed, they fix elemental nitrogen, and thus are important soil improvers, and help control soil erosion The more important species in the U.S. were introduced with the early settlers from Europe, and have been adapted to U.S. conditions by natural selection, breeding, and introduction of genes for various characteristics, the most important of which is probably disease resistance. This report summarizes the more important activities in introduction, preservation, evaluation, and enhancement of the clovers in the United States in the last two decades.

In a review of genetic vulnerability and germplasm resources of the clovers in 1976, Taylor et al. (1977) concluded that the trend in number and diversity of cultivars of the cultivated species was downward, perhaps because of emphasis on other crops and narrowing of gene bases by breeding. However, because the cultivated clovers are mostly cross pollinated, heterozygous, and heterogeneous, they were not considered to be pa particularly vulnerable at that time. Germplasm resources were thought to be in a more hazardous position because the many accessions of these species were haphazardly assembled and were not systematically evaluated. It was, recommended that evaluation and use of accession should be established by a committee of experts knowledgeable in clover germplasm. It is pleasing to report that many of the recommendations of the report have been acted upon, and that germplasm resources are now in a much better state.


Of most importance was the establishment of a Clover and Special Purpose Legume (CSPL) Advisory Committee to rate germplasm activities. A Trifolium curator was appointed and the seed stocks available in the United States were surveyed and assembled at the Curator location and at the Southern Regional Plant Introduction Station for the annual, and at the Northeastern Plant Introduction Station for the perennial legumes. Seeds of about 215 taxa of clovers were assembled at Lexington, Kentucky. The committee concluded that adequate numbers of accessions of the cultivated species were available, but accessions of some perennial species were rare. Plant exploration was initiated to increase the number of accessions of rare taxa- The first of these explorations was in Greece, Italy, and Crete in 1977 (Gillett and Smith 1985). Thirty-six species represented by 291 accessions were added to the collection as a result of the exploration. The second exploration was in Romania in 1984 (Taylor and Rumbaugh 1986) to collect perennial species. Forty-two accessions were collected representing 13 different species. P.I. numbers were assigned and chromosome numbers determined (Taylor and Rumbaugh 1986). However, no new species were added to the collection, and it was determined that Romania is not a prime center of diversity of Trifolium. Accessions of species were added that could contribute genes different from those previously available. The CSPL Advisory Committee concluded that the prime center for future collection of new perennial species is in the dinaric alps of Yugoslavia and Albania. Because it is difficult to enter Albania, an exploration trip to Yugoslavia was proposed for the late summer of 1988 and was completed shortly before this conference. In addition to these formal exploration efforts, the curator and others have collected in the United States and foreign countries. One of the prime centers for diversity is Turkey, where clovers were collected as a part of the Medicago exploration trip in the 1960s. Domestic species also have been collected, particularly in the Western United States and by informal germplasm exchange. The clovers of Ethiopia, South Africa, South America, New Zealand and Australia also have been added to the curator and P.I. station collections.

The clovers of the eastern United States recently have come under considerable scrutiny because of near extinction of several species. These include T. stoloniferum Muhl., T. reflexum, L., T. virginicum Small ex Small and Vail., T. carolinianum Michx., T. polymorphum Pair. ex Lam. and Pair., and T. bejariense Moric. These species, once prevalent in the eastern United States, are now only rarely seen. T. stoloniferum (running buffalo clover), thought to be possibly extinct, was listed in the Federal Register as an endangered species (Jacobs 1987), and a recovery team has been formed. Small colonies of this clover now have been found in West Virginia, Kentucky, and Indiana. A second stolonifereous species, as yet not named, has been discovered in Tennessee and Virginia (Taylor 1988).


Preservation of the introduced germplasm is a continuing problem that the CSPL advisory committee recognized early. The committee recommended that all Trifolium seed stocks for long- and short-term storage be maintained at below freezing temperature (-5 to -18°C), which was shown by Rincker (1974) to be effective for T. pratense and other seeds. The curator collection at Kentucky has been maintained at these temperatures since about 1974, and the regional plant introduction stations have purchased equipment for below-freezing storage that may be in place at this writing. Storage at above-freezing temperatures has resulted in gradually declining seed viability, necessitating periodic seed increases. About two-thirds of the species of the genus are self-pollinated (Taylor et al. 1980) and increase of these seeds is relatively easy because isolation is not required. Most of the self-pollinated species are annuals, maintained at Experiment, Georgia, and periodic seed increases have been made at that location. Recently, however, a contract was issued to increase a number of accessions in southern California. The increase procedure requires establishing at least 50 plants of each accession which are mostly, if not entirely, self-pollinated and thus do not need to be isolated. For the cross pollinated, and particularly for the perennial species, the procedure is more difficult. The former procedure was to allow cross pollination among accessions without isolation, but this resulted in loss of genetic identity A more recent procedure is to conduct such increases under cages in which bees are used for pollination. Contracts recently have been issued to increase seeds of accessions of T. medium L. and T. ambiguum Bieb. in northern locations where plants are established, allowed to over-winter for vernalization, and produce seed the following year. Seed increases are expensive but necessary to rejuvenate accessions with seeds of low viability and to increase seed when, as is usually the case, only small amounts of seed were originally introduced.


Prior to official evaluation efforts, examination of species of Trifolium in the Kentucky collection revealed that annual species generally have simple tap roots and low chromosome numbers. They are usually self pollinating and many were introduced from a Mediterranean type climate. Perennial species generally are taprooted, stoloniferous, or rhizomatous, possess higher chromosome numbers than annuals, are mostly cross-pollinating and do not have specific climate-habitat relationships. Species introduced from Eurasia are more numerous and more diverse in base chromosome number (x = 5 to 8) than are those from other origins. Only species with diploid chromosome numbers of 16 or higher are stoloniferous or rhizomatous. Rhizomatous species, mostly cross-pollinated, were introduced from Eurasia, North and South America, but not from Africa, and not often from Mediterranean climates. Self- and cross-pollinated species occur in all regions. Different flower colors and leafmarks are not associated with origin, climate, or other morphological and physiological characteristics (Taylor et al. 1979).

Systematic evaluation of the many accessions of the cultivated species has begun under contract with the Plant Introduction Service. Descriptor and evaluation priorities have been established for T. pratense, L., T. repens L., T. incarnatum L., T. vesiculosum Savi, T. subterraneum L. and T. hirtum L. Evaluations that are underway include disease and nematode resistance for T. pratense, isoflavone content in T. subterraneum and T. hirtum, and determination of chromosome numbers among accessions of T. medium and T. ambiguum. It is expected that evaluation efforts will proceed as rapidly as funds are made available, and personnel can be found with expertise to conduct the evaluations. An unanticipated problem is that the researchers may perceive evaluation as a service rather than a research activity.


Enhancement activities long underway for the major cultivated species have led to the release of a large number of cultivars, germplasms, and gene markers (see Taylor 1985, Clover Science and Technology for a review, of these activities). In terms of "New Crops" to which this paper is primarily addressed, introduction efforts have resulted in some cultivar and germplasm releases (Table 1). Perhaps the most notable of these introductions in recent years is T. vesiculosum (arrowleaf clover), introduced from Italy in 1963. Three different introductions varying in maturity date resulted in three cultivars, 'Amclo', 'Meechee', and 'Yuchi' (Miller and Wells 1985). Knight and Watson (1977) estimated the acreage of arrowleaf clover at 200,000 ha in 1980. Rose clover (T. hirtum) is another introduction that achieved a fairly important position in United States agriculture. According to Love (1985), seed of the 'Wilton' cultivar developed in Australia was certified in California in 1949. It has found a significant place in California range sowings, but estimates of acreage are not available. Other introductions of clover that are of some value in the United States are listed in Table 1. Several other introduced clovers have become naturalized in the United States and occasionally make significant contributions to pasture and hay crops. Little or no breeding has been done with these species, which include ball clover (T. nigrescens), hop clovers (T. campestre Shreb., T. aureum L., and T. dubium Sibth.), lappa clover (T. lappaceum L.), cluster clover (T. glomerulatuum L.), bigflower clover (T. michelelianum Savi.) and rabbitfoot clover (T. arvense L.) (Knight 1985). About 10 native range species clovers of Western United States are locally important as forage for sheep, cattle, and deer (Crampton 1985).

Much of the interest in plant introduction has stemmed from the possibility of hybridizing perennial wild clovers with the cultivated species, and then to develop new crops for agriculture. Barriers to interspecific hybridization are quite difficult to overcome among species closely related to red clover (Taylor et al. 1980, Rubtsov and Komkova 1983). Only four species have been hybridized with red clover, and only one of these was fertile (Table 2). The first hybrid was a mating of T. pratense (2n = 2x = 14) with T. diffusum L. (2n = 2x = 16), an annual species. The hybrid was sterile at the diploid level but fertile as an amphidiploid (Taylor et al. 1963). Another annual species, T. pallidum L., (2n =2x = 16) was successfully hybridized with tetraploid T. pratense (2n = 4x = 28) to produce a sterile sesquidiploid (Armstrong and Cleveland 1970). In the third hybrid, embryo rescue was necessary for the hybridization of red clover with the perennial T. sarosiense Hazsl. (2n = 6x = 48) (Phillips et al. 1982), but unfortunately the hybrid is sterile and has remained so even when doubled with colchicine. Tetraploid T. pratense (2n = 4 x = 28) also has been crossed with T. medium (2n = 10x = 80) to produce a sterile hybrid (Merker 1984).

Among species closely related to red clover, T. sarosiense has been hybridized with T. medium and T. alpestre L. has been hybridized with T. heldreichianum (Gib. and Belli) Hausskn. and T. rubens L., all perennial species (Quesenberry and Taylor 1976, 1977, 1978) (Table 2). The transfer of genes from these species to red clover might be possible if the sterility of the T. sarosiense x T. pratense hybrid could be overcome. The perennial species possess perennial tap roots and/or rhizomes and are resistant to several virus diseases, but the annual species seem to possess few useful characteristics.

It is possible to construct a diagram showing natural affinities based on successful crosses among species closely related to red clover (Fig. 1). According to this diagram and Table 1, red clover is most closely related to the annual species, T. diffusum, and somewhat less closely related to T. pallidum, T. sarosiense and T. medium. Trifolium sarosiense is closely related to T. medium and T. alpestre, which in turn are closely related to T. rubens and T. heldreichianum.

Interspecific hybridization of white clover has been more successful than with red clover. Hybrids that have been produced are listed in Table 3. White clover has been hybridized with the annual species T. xerocephalum (Frenzl.) at the diploid level and with T. nigrescens (Viv.) both at the diploid and tetraploid levels to produce partially fertile hybrids. White clover also has been hybridized with the perennial species, T. occidentale (Coombe) and T. uniflorum L. and with T. ambiguum at the tetraploid level (Table 3). The most promising of these hybrids involves T. ambiguum in which one hybrid plant is partially fertile (Williams and Verry 1981) and F2 populations and backcrosses to white clover have been produced. The transfer of perenniality and virus disease resistance from T. ambiguum to T. repens has proved to be extremely difficult. Research continues using colchicine to double the chromosome number of the hybrid, irradiating of the F1 to aid in transfer of genes to white clover by backcrossing, and evaluating virus resistance in backcross populations.

In summary, although much remains to be accomplished in accessing the genetic variation available in the clovers, research is underway, and the wide diversity exhibited in the germplasm resources of the clovers is rapidly being cataloged and preserved for future generations.


*The investigation reported in this paper (No. 88-3-201) is in connection with a project of the Kentucky, Agricultural Experiment Station and is published with approval of the Director.
Table 1. Recent introduction of species of Trifolium that have resulted in cultivar and or germplasm releases.

SpeciesCommon nameLongevityz Origin USA releases Referencesy
T. vesiculosum Savi arrowleaf A Italy Amclo, Meecher Yuchi cultivars (1)
T. ambiguum Bieb. kura P USSR, Asia minor C-2 and ARS-2678 germplasms (2, 3)
T. medium L. zigzag P Europe C-1 and M-1 germplasms (4, 5)
T. alexandrinum L. berseem A Syria, Egypt Bigbee cultivar (6)
T. resupinatum L. Persian A Asia Minor Abon cultivar (7)
T. hirtum L. rose A Asia Minor Wilton cultivarx (8)
T. fragiferum L. strawberry P Europe, Asia Minor Salina and Fresa cultivars (9, 10)
zA = annual; P = perennial
y(1) Miller & Wells 1985; (2) Townsend 1975; (3) Rumbaugh 1988; (4) Townsend 1971; (5) Taylor et al. 1982; (6) Knight 1985; (7) Weihing 1962; (8) Love 1985; (9) Peterson et al. 1962; (10) Baltensperger et al. 1982.
xReleased in Australia and seed certified in California.

Table 2. Interspecific crosses among species closely related to red clover.

Female Male Somatic chromosome no. of hybrid Fertility Referencey
T. pratense (2n=14)T. diffusum (2n=16)15 sterile (1)
T. pratense(2n=28)T. diffusum (2n=32) 30 fertile (2)
T. pratense(2n=28)T. pallidum(2n=16) 22 sterile (2)
T. sarosiense(2n=48)T. alpestre (2n=32) 40 partial (3)
T. medium(2n=72)T. sarosiense (2n=48)z 58-60 partial (4)
T. alpestre(2n=16)T. heldreichianum (2n=16) 16 partial (5)
T. alpestre(2n=16)T. rubens (2n=16) 16 partial (5)
T. sarosiense (2n=48)T. pratense (2n=14) 31 sterile (6)
T. medium (2n=80)T. pratense (2n=28) 54 sterile (7)
zIncludes reciprocal.
y(1) Taylor et al. 1963; (2) Armstrong and Cleveland 1970; (3) Quesenberry and Taylor 1978; (4) Quesenberry and Taylor 1977; (5) Quesenberry and Taylor 1976; (6) Phillips et al. 1982; (7) Merker 1984.

Table 3. Interspecific crosses involving Trifolium repens.

Female Male Somatic chromosome no. of hybrid Fertility Referencez
T. repens (2n=32) T. nigrescens (2n=16) 24 Partial (1)
T. repens (2n=32) T. nigrescens (2n=32) 32 Partial (2)
T. repens (2n=32) T. occidentale (2n=32)32 Partial (3)
T. repens (2n=32) T. xerocephalum (2n=16) 24 Partial (4)
T. uniflorum (2n=32) T. repens (2n=32) 32 Partial (4, 5)
T. repens (2n=32) T. ambiguum (2n=48) (7)
T. repens (2n=32) T. ambiguum (2n=32) 32 Partial (6)
T. repens (2n=32) T. uniflorum
x T. occidentale
32 Partial (5)
T. repens (2n=32) T. occidentale
x T. uniflorum
32 Partial (5)
z(1) Chen and Gibson 1970; (2) Brewbaker and Keim 1953; (3) Gibson and Beinhart 1969; (4) Kazimerski and Kazinierska 1968; (5) Gibson et al. 1971; (6) Williams and Verry 1981; (7) Yamada and Fukasoka 1985.

Fig. 1. Crossing affinities among species closely related to red clover.

Last update August 26, 1997 by aw