Download The Patterns of Evolution and Ornamental Plant Breeding

The Patterns of Evolution and Ornamental Plant Breeding
Wolfgang Horn
TU München
Chair for Floriculture Crops and Horticultural Plant Breeding
D-85350 Freising
Germany
Keywords: Begonia, Primula, interspecific hybridisation, polyploidy, review, history
Abstract
A review is presented of the origin and development of sexually and asexually
propagated ornamental crop species, with special emphasis on Primula malacoides
and Begonia semperflorens-cultorum.
INTRODUCTION
It is considered fitting at this meeting to look back on the origin of our ornamentals,
and to follow their pathways during development, though there have been reviews of this
topic in the past (Darlington, 1963; Horn, 1968). I should like to begin with some citations worthy of note.
• Interest in the patterns of evolution extends beyond the historical, because they are
also the methods currently used by plant breeders (Allard, 1960)
• The development of cultivated plants is the most important point of contact between plant breeding and evolution
• Plant breeding since Mendel is evolution directed and expedited by man
• Origin of cultivated plants is evolution under conscious or unconscious assistance
of human beings.
• Plant breeders are applied evolutionists (even if they have rarely noticed the fact)
(Simmonds, 1976).
• Ornamental crop evolution – genetic change in crop populations – is probably at
least as rapid now as ever it has been and, in some crops, much more rapid.
There are many unanswered questions concerning the origin and domestication of cultivated species. A. v. Humboldt stated 200 years ago that we knew nothing about the
original sources of our most useful plants. Then came Darwin and genetics. Experimental
evidence for the evolution of cultivated plants has long employed plant-breeding researchers. The history of most ornamentals started in the 17th and 18th century when botanists, plant collectors, missionaries, medical doctors on board of ships brought back new
plants from the East and from the New World to Europe. Ornamental plants, existing in
an enormous number of species, most of them little, but many much, improved. It is advantageous that their improvement being mostly recent and often very well recorded. Ornamental plants, therefore, are especially well suited to study in which way cultivated
plants have arisen from wild species.
GENERAL ASPECTS
In one way or the other, cultivated plants must have arisen from wild progenitor
species. I will not go too much in detail, but since the days of Darwin and the early days
of genetics it has became clear that the main patterns in which they evolved are:
• Mutations, which are the primary source of genetic variability and supply the raw
materials for selection and recombination. Gene mutation is the basic factor for
plants to evolve. New genes (better: alleles) arise continuously and may enable the
plant to meet new environmental conditions more successfully.
• Recombination; the combination of variants provides further variability. It has
certainly played and is still playing an important role in developing new ornamentals. Since Mendel it is known that new recombinants arise in F2 and later generations. During the early development of ornamentals, strains carrying heritable
Proc 21st IS on Breeding Ornamentals, Part II
Eds: G. Forkmann & S. Michaelis
Acta Hort 651, ISHS 2004
19
variations were grown near each other and natural hybrids occured, and greater
numbers of combinations of variants resulted. A large number of interaction effects arise from the many gene combinations and result in unforeseen character
modifications. Progeny from species hybrids that are heterozygous for a large
number of genes produces an immense variation. Where asexual propagation is
possible, interspecific hybridisation permits the development of new types even if
they are sterile. Variation by gene recombination will have a determinative effect
only if it is combined with selection. Survival of the fittest means of the best
adapted, and genetic adaptation is accomplished primarily by gene recombination.
• Selection shapes the raw material that is sifting for superior, adapted, biologically
fit, desired genotypes. Steady selection pressure could slowly and gradually alter
characters by accumulating small additive effects of many genes. Sorting out the
superior types is accomplished by both human and natural selection and is often a
combination of both. Selection is the most difficult part of plant breeding. Man selected severely among variants, molding them into something of greater use to
him. Variants useful to man often would have expired in natural conditions but for
human interference. Under cultivation mutant types have been favored because
they appealed to the curiosity of man. For centuries, selection was the only
method practised for the improvement of plants, and the natural seedlings of selected plants were the only material from which a choice was possible. Selection
thus operated on the maternal side only, the pollen was derived by chance from
other individuals, and there was no selection of the paternal parent (Crane & Lawrence, 1952). When selecting for ornamental breeding aims, aesthetic value, production costs, postproduction longevity, quality and marketability are important.
• Isolation is effected by reproductive barriers, control of mating and spontaneous
gene exchange.
• Drift produces rapid changes of gene frequencies in small populations.
Selection, isolation and genetic drift are mentioned here in passing only, a closer
look at mutations and recombination plus polyploidy will follow. Regarding mutations
one can differentiate between nuclear, extra-nuclear mutations and polyploidy. As to recombination, variants are combined through intra-specific hybridization possibly followed by auto-polyploidy, or through inter-specific hybridization possibly followed by
allo-polyploidy.
There are quite a number of ornamentals where only mutations transformed a wild
species into a plant cultivated in our greenhouses. Examples may be found amidst foliage
plants such as Ananas comosus, Aglaonema, several Ficus spp., Hedera helix, which are
still near the wild species and often modified by bud sports only.
Numerous variegated plants are grown in floriculture, and are examples of extranuclear mutation. One group of variegations develop from plastid mutations. Cells with
mutant plastids give rise to white or yellow cell lineages. By forming stable periclinal
chimeras the plastid mutations are preserved. Mutant plastids may be transmitted by the
eggs into the next generation.
It is relatively easy to develop variability by introducing new types through hybridization within or between species, and by changing the chromosome number.
Polyploidy has certainly played and is still playing an important role. New
tetraploids arise and are produced from diploid and from tetraploid × diploid crosses, apparently via unreduced gametes. It is evident that the accident of the unreduced gamete
may be of considerable importance to the plant breeder, in the first place because it is the
commonest way in which polyploids arise, and secondly because it is the means whereby
selected genes may be transferred from the diploid to the tetraploid condition. As will be
shown later on polyploids have some advantages but genetics of autopolyploids is much
more complicated than that of diploids or allopolyploids (amphidiploids). The degree of
heterozygosity is usually much greater. In tetraploids often some dosage effects are observed.
20
EVOLUTION OF ORNAMENTAL PLANTS
Asexually and Sexually Propagated Crops
In asexually propagated ornamental species all patterns of evolution are found; interspecific crosses followed by polyploidy, and somatic mutations – bud mutations or
sports – are of special importance in nearly all genera (Table 1). In Narcissus e.g. we find
all four patterns. The origin of the garden forms (Wylie, 1952) through selection, interspecific crosses and polyploidy is shown in Fig. 1. In Rhododendron simsii cultivars from
bud sports have great importance (Table 2). Interspecific crosses and polyploidy had a
great impact on the development of cultivated Begonia spp. as well as of Impatiens hawkeri hybrids (New Guinea Impatiens) and Kalanchoe hybrids (Fig. 2, 3, 4).
It is quite obvious that sports play no role whatsoever in sexually propagated crops
(Table 3). In this group, mutation and recombination within species and in several cases
autopolyploidy are of importance. In contrast, Begonia × tuberhybrida has been developed from interspecific crosses (Fig. 4, Table 4) in the 19th century, and mutations such
as double flowers (firstly observed in 1873), white and yellow flower colour (1874/75) as
well as polyploidy (1877) produced horticulturally valuable characters. Since abt. 1880
tetraploidy has made sexual propagation (F1-hybrids) possible after clonal propagation
was the rule. The history of Primula praenitens, the Chinese primrose, following its introduction to Europe since 1820 is not easy to trace in regard to the appearance of definite
mutations. Differences found can be assigned to a limited number of genes through genetic investigations (Table 5). Certain gene mutations appear to have occured more than
once (Crane and Lawrence, 1952). The years from 1860 to 1885 saw the production of
numerous cultivars, developed by variation within a single species. There are diploid and
tetraploid forms of the Chinese primrose in cultivation.
Evolution within One Species: Primula malacoides
The fairy primrose is a fine example how from a diploid pale lilac flowering wild
species with a low variability a tetraploid pot plant species with a great variability has
been developed. Of Chinese origin, it was first described in France in 1886, and the first
seeds were offered in Great Britain in 1908. Since 1919 breeding took place in the USA,
Switzerland and especially in Germany. Once in cultivation several horticulturally important mutants were isolated: white and pink flowers (1914/15), rounded petals (1923),
compact growth (1927) and large tetraploid flowers (1924). Many diploid and, mainly,
tetraploid cultivars have been developed. Flower traits are given in Table 6.
Allele dosage effects have been observed regarding flower colour (Table 7).
Somewhat surprisingly tetraploids produced much more seeds per flower than diploids
(Table 8). This is characteristic for tetraploids from unreduced gametes (meiotic origin) in
contrast to colchiploid tetraploids (mitotic origin). Already Navashin (1926) and
Greenleaf (1941, 1942) have observed such differing fertility of meiotically and mitotically originated polyploids.
In P. malacoides in diploid crosses unreduced gametes occured in very low frequencies in both sexes, while in crosses between different ploidy levels much more unreduced gametes are formed, and differences between genotypes have been found (Table 9).
In 4x × 2x crosses they are produced relatively often, and in the males only, and in 2x × 4x
in somewhat lower frequency, and in the females only. In 2x × 2x crosses only 5 ‰
tetraploids occurred (Table 10). In crosses between different ploidy levels, there are combinations, where 4x × 2x produced most polyploids and vice versa, though on the average,
4x × 2x produced more. The high frequency of 4x progeny from 2x × 4x crosses may be
caused by an advantage of unreduced gametes in such crosses, and possibly by a better
viability of tetraploid embryos while triploid embryos are rarely viable (Skiebe 1966). In
4x × 2x crosses, however, more 3x than 4x types originated. Remarkably, all diploids
from 2x × 4x crosses are matrocline. They came from haplo-diplo- and diploparthenogenesis.
It has been observed very often in different genera and species, that the rate of
21
polyploid progeny from such crosses varies with the genotype of partners and from year
to year. Skiebe (1966) e.g. has found in the fairy primrose in one year, 51 % tetraploids
and in another year 80 % from identical crosses. On the other hand, in some 4x × 2x
crosses up to 98 % tetraploids originated, depending on the genotype of the seed parent.
Similarly, in Cyclamen polyploids from unreduced gametes have been very successful.
Evolution Including Interspecific Crosses: Begonia semperflorens
Fine examples of the origin of an ornamental plant by interspecific crosses and
polyploidy can be found in the genus Begonia. The fibrous-rooted begonia (B. × semperflorens-cultorum, B. cucullata var. hookeri) came to Germany from Brazil by chance in
1821. It was not until 1870 when its good outdoor performance as bedding plant was detected and gene mutations regarding new flower and leave colours (1879, 1888) were
selected, the production of cultivars was intensified in France and Germany. 1890 a dwarf
mutant was found. Following the introduction of B. schmidtiana (white flowering) in the
late 1870ies and crosses between both species, which have the same chromosome number, fast progress was made. Their diploid hybrid was nearly sterile and reproduced in
Germany as cv. ‘Erfordia’ (syn. Blütenmeer, Rosamunde) regularly from 1894 until 1964
by crossing the parental species (Skiebe, 1966). Therefore, it was the very first F1 hybrid
cultivar in plant breeding. Then in the late 1890ies allo-tetraploid forms have originated
in Germany and France (“B. semperflorens gracilis”, Fig. 5) and the resulting truebreeding cultivars surpassed the diploids (Zeilinga, 1962; Skiebe, 1966, 1970). The first
tetraploid F1-hybrid cultivars have been produced in Germany by Benary seedgrowers
(G. Besoke), from 1909 until 1961 (‘Primadonna’ and ‘Feuerzauber’). From 1922 onwards also diploid hybrid cultivars have been developed.
From crosses between diploid and tetraploid the first triploid hybrid cultivars
originated (Benary 1934 ‘Rosa Tausendschön’, syn. Rosalinde, Rosa Media), which were
nearly sterile but showing exceptional performance, and with long lasting flowers. As
Skiebe (1966, 1970) could demonstrate (Fig. 6), triploids originating in the F2 of the cross
B. semperflorens × schmidtiana and in the B1F1 of the backcross to B. schmidtiana. Tetraploids were found in the B1F2 of the backcross to B. semperflorens and after
crossing the mentioned triploids. In this way Skiebe resynthesized the fibrous-rooted begonia from old genotypes.
OUTLOOK
The plant breeder can learn from nature. Plant breeding since Mendel is evolution
directed and controlled by man. Mendelian variation, interspecific hybridization and
polyploidy are the methods of transforming wild species into cultivated plants. In fact,
ornamental crop evolution is probably at least as rapid now as ever it has been and, in
some crops, much more rapid. Even nowadays breeders accelerate plant evolution under
domestication, mainly by recombination. Transformations and transposon-induced
changes are the new way of mutation. Transformations and somatic hybridisation are new
ways of recombination. Techniques like embryo rescue and in vitro pollination have permitted hybridizations till now impossible. We also realize that the progress in molecular
biology led to an advance of evolutionary biology, and molecular data offer very easy and
reliable methods for the reconstruction of relationships between our cultivated plants and
their progenitors. So in different ways new lanes have been opened for plant breeding and
phylogenetic evolutionary research.
Literature Cited
Allard, R.W. 1960. Principles of plant breeding. John Wiley & Sons, New York, London.
Crane, M.B. and Lawrence, W.J.C. 1952. Genetics of garden plants. McMillan, London.
Darlington, C.D. 1963. Chromosome botany and the origins of cultivated plants. Allen &
Unwin, London.
Greenleaf, W.H. 1941. Sterile and fertile amphidiploids: Their possible relation to the
22
origin of Nicotiana tabacum. Genetics 26: 301-324.
Greenleaf, W.H. 1942. Genic sterility in tabacum-like amphidiploids of Nicotiana. Journ.
Genet. 43: 69-96.
Heursel, J. 1999. Azalea’s, oorsprong, veredeling en cultivars. Lannoo, Tielt/Terra,
Warnsveld, Belgium.
Horn, W. 1968. Genetische Ursachen der Variation bei Zierpflanzen.
Gartenbauwissenschaft 33: 317-333.
Jahr, W. 1970. Polyploidie bei Fliederprimeln. Tagungsbericht 101: 121-133. Deutsche
Akad. Landwirtschaftswiss., Berlin.
Navashin, M. 1926. Variabilität des Zellkerns bei Crepis-Arten in Bezug auf die
Artbildung. Z. Zellforsch. Mikrosk. Anatomie 4: 171-215.
Seyffert, W. 1959. Untersuchungen über interallele Wechselwirkungen. III.
Naturwissensch. 46: 271-272.
Simmonds, N.W. 1976. Evolution of crop plants. Longman, London.
Skiebe, K. 1958. Die Bedeutung von unreduzierten Gameten für die Polyploidiezüchtung
bei der Fliederprimel Primula malacoides. Züchter 28: 353-359.
Skiebe, K. 1966. Die züchterische Entwicklung von Begonia semperflorens-cultorum in
Deutschland. Züchter 36: 168-171.
Skiebe, K. 1970. Polyploidie bei Begonia semperflorens cultorum. Tagungsbericht 101:
109-119. Deutsche Akad. Landwirtschaftswiss., Berlin.
Wylie, A.P. 1952. The history of garden narcissi. Heredity 6: 137-156.
Zeilinga, A.E. 1962. Cytological investigations of hybrid varieties of Begonia semperflorens. Euphytica 11: 126-136.
23
Tables
Table 1. Patterns of evolution - Asexually propagated ornamental species.
Bud
mutations
Mutation and recombination
within spp.
between spp.
Polyploidy
Auto-
Allo-
Alstroemeria
x
x
x
Begonia cheimantha
x
x
x
Chrysanthemum
x
x
x
Dahlia
x
x
x
Dianthus caryophyllus
x
Erica hybrids
x
Euphorbia pulcherrima
x
Freesia
x
x
x
Gladiolus
x
x
x
x
x
x
x
x
Hibiscus rosa-sinensis
Hyacinthus
x
x
(x)
x
x
Impatiens NG
(x)
x
x
Kalanchoe
(x)
x
x
Lilium hybrids
(x)
x
(x)
Narcissus
x
x
x
Pelargonium
x
(x)
x
Rhododendron simsii
x
x
(x)
Rosa
x
x
x
Tulipa
x
x
x
24
(x)
x
Table 2. Rhododendron simsii hybrids – cultivars from bud sports.
Source cv.
Year of origin
No. of sports
Year
Avenir
1911
9
1936-1969
Hellmut Vogel
1967
25
1972-1994
Knut Erwen
1934
10
1958-1998
Mme. Petrick
1880
17 (35)
1911-1948
Paul Schäme
1890
15 (21)
1928-1960
Vervaeneana
1886
1887-1928
Table 3. Patterns of evolution – Sexually propagated ornamental species.
Mutation and recombination
within spp.
Ageratum
Antirrhinum
Begonia fibrous-rooted
Begonia tuberous-rooted
Callistephus
Campanula persicifolia
Coreopsis
Cyclamen
Gaillardia
Impatiens walleriana
Lathyrus odoratus
P. malacoides
P. obconica
Petunia
Primula praenitens
Rudbeckia
Tagetes erecta
Tagetes patula
Viola wittrockiana
x
x
(x)
x
x
x
x
x
x
x
x
x
between spp.
Polyploidy
Auto-
Allo-
x
(x)
x
x
x
x
x
x
x
x
(x)
x
x
x
x
x
x
x
x
x
x
x
x
x
x
25
Table 4. Origin of tuberous begonia (sect. Huszia).
Species / colour
1st flower
Great Britain
No. of chromosomes
B. cinnabarina
red
1948
2n = 26
B. boliviensis
red
1867
2n = 28
B. pearcei
yellow
1866
2n = 26
B. veitchii
red
1867
2n = 28
B. davisii
red
1876
2n = 28
B. dregei
white
1836
2n = 26
South Africa (sect. Augustia)
Table 5. History of Primula praenitens gene mutations arisen in cultivation.
Year
Introductions from China
Stellata
Allele
1820–1826
abt. 1821
ch
Purplish-crimson flowers
1824
G
White flowers
1827
D
Fertile double magenta and white
1838
M
Crimson forms figured
1846
B
Large eyes recorded
1864
A
Coral flowers
1906
K
Ivy leaf
1907
iv
Maple leaf
1929
mp
Dazzler type
1933
dz
Auto-tetraploids
1909
(extracted from Crane and Lawrence, 1952)
26
Table 6. Flower traits in 2x and 4x Primula malcoides.
2x wild type
2x cvs
4x cvs
no. flowers (main shoot)
41.4
36.5-68.5
48.4-58.9
no. umbels
4.5
3.7-5.6
4.9-6.2
flower size
1.1
1.7-1.9
2.1-2.9
distance between umbels (1/2)
2.1
2.2-3.7
2.4-3.2
Distance : flower size (1/2)
2:1
1.3-2.0 : 1
0.8-1.6 : 1
flowering period [d]
17.8
19.1-26.1
26.1-31.2
Table 7. Polyploidy in Primula malacoides – Allele dosage effects and flower colour.
Genotype
Relative concentrations
HCC
Petunidin
Malvidin
bbbb
1.00
traces
820
bbb+
.63
.37
822
bb++
.34
.66
724
b+++
.08
.92
025
++++
traces
1.00
27
27
Table 8. Fertility of diploids and tetraploids in Primula malacoides.
Cross
No. flowers
No. seeds
total
No. seeds
per flower
2x B1F1.1
150
2883
19
2x B1F1.2
90
2871
32
4x B1F1.1
150
16080
107
4x B1F1.2
90
3784
42
2x F1 × F1
240
4631
19
4x F1 × F1
240
21415
89
(Data from Skiebe 1958)
Table 9. Primula malacoides – number of unreduzed gametes in 2x cultivars.
Cross
Plants
tested
2x gametes
eggs
pollen
per plant
2x x 2x
4489
3
3
.0013
2x x 4x
97
61
0
.6280
4x x 2x
39
0
51
1.3080
Table 10. Frequencies of polyploids in Primula malacoides – results from crosses.
Plants
tested
No. of
% 4x
2x
3x
4x
2x x 2x
8476
8471
1
2x x 4x
146
15
31
100
68.4
4x x 2x
47
0
9
38
80.9
28
4
.005
Figures
Fig. 1. Origin of the garden forms of narcissi (Wylie, 1952).
Fig. 2. Origin of Impatiens Neuguinea cultivars. Origin Madagascar, introduced to Paris
1927
29
Fig. 3. Development of Kalanchoe cultivars.
Fig. 4. Origin of cultivated Begonias.
30
Fig. 5. Origin of Begonia semperflorens (Zeilinga, 1962).
Fig. 6. Polyploidy in Begonia semperflorens (Skiebe, 1970).
31