Tải bản đầy đủ - 0trang
CHAPTER 3. RED CLOVER BREEDING AND GENETICS
N. L. TAYLOR AND R . R . SMITH
it is the leading legume in forage production, and it maintains a significant
position in forage production in the United States. Red clover is adapted to a
wide range of soil types, pH levels, and environmental conditions. This plasticity
has enabled red clover to retain its usefulness for hay, silage, pasture, and soil
improvement in much of the temperate region of the world.
A. ECONOMIC IMPORTANCE
The overall economic importance of red clover to world agriculture is difficult
to assess. Red clover is consumed by animals in the form of hay, silage, or
pasture fodder, and is marketed to the consumer in the form of meat, fiber, and
dairy products. Its introduction to central and northern Europe and the United
States had a profound impact on civilization and agriculture by providing a
stabilized supply of feed to livestock (Piper, 1924; Gras, 1940; Fergus and
Red clover is generally grown with timothy, Phleumprutense L. In the United
States, red clover is grown for hay, silage, pasture, and soil improvement on
about 5.4 million hectares. According to United States Department of Agriculture Crop Production and Agricultural Statistics Reports, the annual yield is
approximately 21.5 million metric tons, or about 4.0 metric tons per hectare.
Maximum yields using improved varieties should easily be twice these values.
The average value received in 1976 per metric ton of hay was $66.44. Therefore,
the economic worth of red clover is approximately 1.4billion dollars annually, or
$265 per hectare. This does not consider the value of the red clover as pasture.
The value of the annual red clover seed crop in the United States is between 23
and 27 million dollars.
Red clover is a nitrogen-fixing legume, thus contributing to the supply of
nitrogen available in the soil for subsequent crops. It is estimated that red clover
provides between 125 and 200 kg of nitrogen per hectare (Rohweder et ul.,
B. ORIGIN AND DISTRIBUTION
Red clover is thought to have originated in southeastern Europe and Asia
Minor. Local ecotypes probably developed over 2000 years ago, but documentation is limited. A review of the earliest references to the utilization of red clover
as forage is provided by Fergus and Hollowell (1960) and Pieters and Hollowell
(1937). Red clover was probably introduced to northern Europe during the fifteenth century A.D. and into North America in the late seventeenth or early
eighteenth century A.D. Today, it is an important forage legume throughout the
RED CLOVER BREEDING AND GENETICS
temperate regions of Europe, the Soviet Union, Australia, New Zealand, Argentina, Chile, Canada, Japan, Mexico, Columbia, southern Canada, and eastern
and central United States.
Red clover is a species of the genus Trifolium, which contains 8 subgenera or
sections (Hossain, 1961). The subgenus Trifolium, containing about 70 species,
was divided into 17 subsections by Zohary (1971, 1972). Within T . pratense,
two main types of clover are grown in the United States: medium or double-cut,
and mammoth or single-cut. Mammoth usually flowers later, is taller, yields
more in the first growth, and produces less aftermath growth than medium.
Double-cut red clover, contrary to its name, may produce several flushes of
growth per year, depending on the length of the growing season. Most red clover
grown in the United States is of the medium type.
A . RELATED SPECIES AND CHROMOSOME NUMBERS
Red clover has a diploid chromosome number of 14 ( n = 7). Classified by
Zohary (1971) as closely related to red clover are species with base numbers of 8.
Trifolium difisum Ehrh. and T . pallidum Waldst. and Kit., two annual species,
have a diploid number of 16. A closely related perennial is T . noricum Wulf.,
which also has 16 diploid chromosomes. Somewhat more distantly related are T.
medium L. (zigzag clover, 2n = 64-80), T . alpestre L. (2n = 16), T . rubens L.
(2n = 16), and T . heldreichianum (h= 16). Another species sometimes considered to be a form of T . medium is T . sarosiense Hazsl. (2n = 48). Two other
annual species perhaps more distantly related are T . hirtum All. and T . cherleri
L. (2n = 10).
B . EVOLUTION
Probably the best evidence on the evolution of red clover is provided by
interspecific hybridization. Although many interspecific crosses have been attempted, the only verified hybrids involving red clover were obtained with the
diploid annuals T. difisum and T . pallidum. The diploid hybrid of T. pratense
x T . difisum was sterile, but the amphidiploid of 4x T . pratense x 4x T .
diffusum was fertile (Taylor et al., 1963). Pollination of 2n T . pratense with 2n
T . pallidum yielded only shriveled seeds, but the cross of 4x T. pratense x 2n T .
pallidum resulted in a sterile triploid hybrid (Armstrong and Cleveland, 1970).
N. L. TAYLOR AND R. R. SMITH
Meiotic analyses of diploid and tetraploid hybrids of T. pratense X T . diffusum suggested that the chromosomes of the two species differ by a series of
complex structural interchanges. This is based on the presence of chain multivalents in the diploid and amphidiploid hybrids (Schwer and Cleveland, 1972a,b).
Analyses of chromosome pairing in the triploid hybrid of 4x T. pratense X 2x T .
pallidurn revealed mostly bivalents and univalents, but the presence of multivalents suggested some homology between the two species. These results apparently indicate that red clover is more closely related to annuals than to perennials,
and to T. difisum than to T . pallidum. Cytological analyses of interspecific
hybrids suggest that speciation has resulted from a complex series of structural
interchanges causing chromosome differentiation and eventual loss of one
chromosome pair in T. pratense.
Crosses among perennial species related to red clover have been reported by
Maizonnier (1972) and Quesenberry and Taylor (1976, 1977, 1978); they are
summarized in Table I. Observation at metaphase- 1 of diploid hybrids indicated
good pairing, with most PMCs having eight bivalents. Pollen stainability was
about 50%, indicating that the chromosomes of the diploid species differ by
small structural changes. Although crossability barriers among these species
were weak, barriers after fertilization were strong, ranging from embryo abortion
to poor germination and low F2 survival. Analysis of meiosis of polyploid hybrids also showed a low frequency of multivalents. Fertility of pollen from hybrids
was 70-85%, and no major barriers to hybridization were exhibited except for
Interspeciflc Hybrids of Perennial Trifolium Species in Section Trifolium Zoh
Species and chromosome
sar (48) x alp (16)
med (80) x sar (48) & rec.
med (64)x sar (48) & rec.
alp(16) x held(16)&rec.
alp (16) x rub (16)
med (72) x sar (48)
sar (48) x alp (32)
"Species abbreviations: sar = T . sarosiense Hazsl.; alp = T. alpesfre L.; med = T. medium L.;
held = T. heldreichianum Hausskn.; rub = T . rubens L.
Male and female fertility as measured by cross-pollination among hybrid plants.
'Refers to literature cited: (1) Maizonnier (1972); (2) Quesenberry and Taylor (1976); (3) Quesenberry and Taylor (1977); (4) Quesenberry and Taylor (1978).
RED CLOVER BREEDING AND GENETICS
Trifolium pratense produced no hybrids with the perennial species, indicating
that little gene exchange has occurred. Hybridization may have been involved in
the evolution of the perennial species, but the presence of F, sterility and the
absence of natural polyploids among the species most closely related to red
clover indicate that little hybridization has occurred in the more recent evolution
of T. pratense.
A. FLORAL STRUCTURE
Red clover has the complete leguminous flower, consisting of calyx, corolla,
ten stamens, and a pistil. The calyx tube terminates in five lobes, or teeth. Five
petals unite at the base to form a corolla tube consisting of a standard petal, two
wing petals, and two keel petals. Nine stamens and the one stigma unite to form
the sexual column. The tenth anther is free. Up to 125 or more flowers are
situated on peduncles to form a single head or, in rare cases, multiple heads.
Flowers usually are reddish-pink but vary from white to deep reddish-purple.
B. POLLEN DEVELOPMENT AND TRANSFER
Anthers of red clover consist of four microsporangia. Prior to dehiscence the
walls between the microsporangia break down, forming a common cavity on
each side of the anther. Dehiscence of the anther occurs before the petals have
reached their full size. Pollen fills the space between the fused keel petals and is
held in place until disturbed (Hindmarsh, 1963).
Transfer of pollen generally is by bees, but the flowers of red clover are so
large that honeybees (Apis mellifera L.) will avoid them if other nectar sources
are available. Honeybees may collect pollen, however. Bumblebees (Bombus
spp.) are efficient pollinators of red clover. Alkali bees (Nomia melanderi Ckll.)
and leaf cutter bees (Megachile rofundata [F.]) also have been used for red
clover pollination. When red clover is pollinated by a bee, the sexual column
protrudes from the interior of the flower, with the pistil extending slightly beyond
the stamens. When the weight of the bee is removed, the sexual column returns
to its original position. Each flower may be pollinated several times, as long as
the stigma remains receptive.
C. SEED SET AND DEVELOPMENT
The stage of bloom for optimum seed set appears to be when flowers are about
half open. Stigma receptivity and pollen viability continue for about 10 days
N. L. TAYLOR AND R. R. SMITH
under greenhouse and field conditions. In liquid nitrogen, pollen may be stored
up to 26 weeks without loss of viability (Engelke and Smith, 1974). The length
of time between pollination and fertilization of the egg cell in diploid red clover
is between 28 and 35 hours, and between 17 and 26 hours in tetraploids (Mackiewicz, 1965). Seeds are physiologically mature 14 days after pollination and
are dry enough for harvesting at about 21 days. Two ovules occur in each ovary,
but, except for some bred strains, one usually aborts (Van Bogaert, 1958; Schieblick, 1966; Bingefors and Quittenbaum, 1959). In rare cases, up to four ovules
per ovary occur (Povilaitis and Boyes, 1959).
D. MECHANISMS CONTROLLING CROSS-FERTILIZATION
Red clover is a self-incompatible cross-pollinated species that under natural
conditions produces very few self-seed. However, self-compatible, autogamous
genotypes have been found (Diachun and Henson, 1966). The mechanism of
self-incompatibility (reviewed by Fergus and Hollowell, 1960) is the one-locus,
gametophytic S-allele system in which plants with the same S-alleles are prevented from selfing by slow growth of pollen tubes through styles. The system
also prevents cross-fertilization of plants that have the same S-allele genotype.
Details of the S-allele systems in red clover have been investigated by inbreeding via pseudo-self-compatibility (PSC), a term used to indicate the production of
self-seed by normally self-incompatible genotypes. Leffel (1963) showed that
selfing heterozygous (S,S,) clones produced a 1- 1 ratio of homozygotes (SJ,
and/or SzS2) to heterozygotes (S,Sz). However, the progeny of one clone significantly deviated from this ratio, producing fewer homozygotes than expected.
Johnston et af. (1968) also found a deficiency of homozygotes in I, (selfed)
progenies of 15 heterozygous (S,S2) clones. Detailed examination of three
families indicated that two possessed both homozygous classes (S ,S and S4S2)as
well as the heterozygous class (S J2), and one possessed only one homozygous
class and the heterozygous class. Consequently, it appears that the S-allele system in red clover deviates in some plants from the classical system. The reason
for the deviation is unknown.
An extensive series of S-alleles exists in red clover, and much research has
centered on the means by which such a large number are generated. Irradiation
induces mutations to self-compatibility (Sf)rather than to new S-specificities (de
Nettancourt, 1969). Denward (1963) observed what were thought to be new
S-alleles among I, sib plants from selfed (I,) clones, but he could not be certain
that they were not contaminants. Johnston et al. (1968) found no changes in 1,’s
of three red clover families. Anderson et al. (1974a) reported one changed
S-specificity in eight I2 families. Pandey (1970) speculated that the principle
method of generation of new S-alleles during inbreeding was intracistronic re-
RED CLOVER BREEDING AND GENETICS
combination in the structural cistrons of the S-gene complex, with a minor role of
point mutations and deletions. Regulatory genes in the heterozygous condition
during outbreeding suppress recombination. Inbreeding produces homozygous
recessive regulatory genes, thus eliminating suppression of recombination. Experimental verification of this hypothesis has not been obtained, in part because
of the low frequency of recombinants.
IV. Heritability and Gene Action
A. MORPHOLOGICAL CHARACTERS
Polyphylly, leaves with five leaflets, is conditioned by two recessive genes, f
and n (Simon, 1962). Polyphylly results when either or both of these genes are
present in the homozygous condition. Artemenko (1972) developed a population
with four to nine leaflets on 89% of the plants.
One of the first reports of the inheritance of male sterility was that of Smith
(1971), who described a male-sterile plant with shriveled and dark-orange anther
sacs. The male-sterile condition is controlled by a single recessive gene designated m s l m s l . A second gene for male sterility isolated by Taylor et al. (1978~)
was designated mszmsz. The two loci interacted to give a duplicate recessive
epistatic type of gene action. Macewicz (1976a,b) reported a male-sterile type in
which the anthers were either incompletely developed or entirely absent. Male
sterility was controlled by the complementary action of a recessive gene, rf,, and
a dominant one, Rf,, in a sterile cytoplasm. Restorers of fertility were readily
detected in the population. Shcheglov and Zvyagina (1975) and Zvyagina (1973)
reported cytoplasmic male sterility induced by colchicine. Whittington ( 1958)
described a male-sterile type resulting from asynapsis. Since the original source
was also female-sterile, no genetic information was obtained.
Liang (1965) reported that leafmarking is inherited on a monofactorial basis
with developmental and suppressor genes active. Annual habit of growth was
reported to be controlled by a single recessive gene in some cases, and by two or
more in others (Strzyzewska, 1974).
B. PHYSIOLOGICAL CHARACTERS
Inheritance of a gibberellin-responsive dwarf mutant in red clover was described by Smith (1974). The dwarf characteristic is controlled by one gene,
dw,,and is recessive to normal growth.
Nutman (1968) described two independent recessive host genes, n and d, each
of which prevents nitrogen fixation in the nodules of red clover. Each factor is
N . L. TAYLOR A N D R. R . SMITH
specific in its ineffectiveness of nitrogen fixation for specific strains of the
bacteria Rhizobium trifolii. The two factors, n and d, are independent and
nonallelic to factors i l and ie reported earlier by Nutman (1954, 1957).
Graham and Newton (1971) reported internal necrosis of crown pith tissue of
red clover, which they called internal breakdown (IB). Selection studies by
Zeiders et al. (1971) have shown that resistance or susceptibility to IB can be
readily obtained, suggesting that few genes control this character.
C . PEST RESISTANCE
Disease resistance in red clover generally is controlled by one to a few genes.
Inheritance of resistance to southern anthracnose caused by Colletotrichum
trifolii B. & E. is conditioned by one recessive gene, according to Athow and
Davis (1958). However, resistance to most diseases in red clover is conditioned
by dominant genes. For northern anthracnose, caused by Kabatiella caulivora
(Kirch.) Kavak., two (Sakuma et al., 1973) or more than three (Smith and
Maxwell, 1973) dominant genes are involved. Resistance to powdery mildew
caused by Erysiphe polygoni DC is dominant and, for five races of the fungus, is
monogenic. In two other races, resistance seems to be controlled by two genes,
and in a third race, inheritance of resistance varies among red clover clones
(Hanson, 1966; Stavely and Hanson, 1967). Inheritance of resistance to rust
caused by Uromyces rrifolii var. fallens Arth. is controlled by a single dominant
gene (Diachun and Henson, 1974a,b). However, this source of resistance cannot
be used for cultivar development because it is linked with a seedling lethality
factor (Engelke et al., 1977). Other types of resistance to rust are inherited in a
quantitative manner (Engelke et al., 1975). Although earlier investigations indicated that resistance to crown rot caused by Sclerotinia trifoliorum Eriks. is
heritable, no detailed inheritance investigations have resulted. Autotetraploid
cultivars are more resistant to crown rot than are comparable diploids (Vestad,
1960). The effect of induced tetraploidy differs by genotype, suggesting that
dosage effects of genes for resistance may be important.
Resistance to virus in red clover also appears to be qualitatively inherited.
Diachun and Henson (1974a,b) reported three types of resistance to bean yellow
mosaic virus, each controlled by a different dominant gene: necrotic local lesion
(hypersensitive) reaction; resistance to mottling and systemic necrosis; and resistance to general mottling, controlled by a gene that appears to be epistatic to the
gene for hypersensitive reaction. Khan et al. (1978) determined that the resistance to red clover vein mosaic virus was controlled by a single dominant gene,
Resistance to the stem nematode (Ditylenchus dipsaci) was reported by Nordenskiold (1971) to be regulated by two dominant genes. One of the genes is
closely linked to the S-locus (self-incompatibility).