- - -
- - -
Dépôt Institutionnel de l’Université libre de Bruxelles / Université libre de Bruxelles Institutional Repository
Thèse de doctorat/ PhD Thesis Citation APA:
Vekemans, X. (1992). Evolution of plant breeding systems: armeria maritima Mill.(Willd.) as a study case (Unpublished doctoral dissertation). Université libre de Bruxelles, Faculté des sciences, Bruxelles.
Disponible à / Available at permalink : https://dipot.ulb.ac.be/dspace/bitstream/2013/212934/3/5c0d6a0b-cafb-48be-a64f-03fb08c09476.txt
(English version below)
Cette thèse de doctorat a été numérisée par l’Université libre de Bruxelles. L’auteur qui s’opposerait à sa mise en ligne dans DI-fusion est invité à prendre contact avec l’Université ([email protected]).
Dans le cas où une version électronique native de la thèse existe, l’Université ne peut garantir que la présente version numérisée soit identique à la version électronique native, ni qu’elle soit la version officielle définitive de la thèse.
DI-fusion, le Dépôt Institutionnel de l’Université libre de Bruxelles, recueille la production scientifique de l’Université, mise à disposition en libre accès autant que possible. Les œuvres accessibles dans DI-fusion sont protégées par la législation belge relative aux droits d'auteur et aux droits voisins. Toute personne peut, sans avoir à demander l’autorisation de l’auteur ou de l’ayant-droit, à des fins d’usage privé ou à des fins d’illustration de l’enseignement ou de recherche scientifique, dans la mesure justifiée par le but non lucratif poursuivi, lire, télécharger ou reproduire sur papier ou sur tout autre support, les articles ou des fragments d’autres œuvres, disponibles dans DI-fusion, pour autant que :
Le nom des auteurs, le titre et la référence bibliographique complète soient cités;
L’identifiant unique attribué aux métadonnées dans DI-fusion (permalink) soit indiqué;
Le contenu ne soit pas modifié.
L’œuvre ne peut être stockée dans une autre base de données dans le but d’y donner accès ; l’identifiant unique (permalink) indiqué ci-dessus doit toujours être utilisé pour donner accès à l’œuvre. Toute autre utilisation non mentionnée ci-dessus nécessite l’autorisation de l’auteur de l’œuvre ou de l’ayant droit.
--- English Version ---
This Ph.D. thesis has been digitized by Université libre de Bruxelles. The author who would disagree on its online availability in DI-fusion is invited to contact the University ([email protected]).
If a native electronic version of the thesis exists, the University can guarantee neither that the present digitized version is identical to the native electronic version, nor that it is the definitive official version of the thesis.
DI-fusion is the Institutional Repository of Université libre de Bruxelles; it collects the research output of the University, available on open access as much as possible. The works included in DI-fusion are protected by the Belgian legislation relating to authors’ rights and neighbouring rights.
Any user may, without prior permission from the authors or copyright owners, for private usage or for educational or scientific research purposes, to the extent justified by the non-profit activity, read, download or reproduce on paper or on any other media, the articles or fragments of other works, available in DI-fusion, provided:
The authors, title and full bibliographic details are credited in any copy;
The unique identifier (permalink) for the original metadata page in DI-fusion is indicated;
The content is not changed in any way.
It is not permitted to store the work in another database in order to provide access to it; the unique identifier (permalink) indicated above must always be used to provide access to the work. Any other use not mentioned above requires the authors’ or copyright owners’ permission.
UNIVERSITE LIBRE DE BRUXELLES
FACULTE DES SCIENCES
FACULTE DES SCIENCES APPLIQUEES Section Interfacultaire d'Agronomie
Laboratoire de Génétique et d'Ecologie Végétales
EVOLUTION OF PLANT BREEDING SYSTEMS:
ARMERIA MARITIMA Mill. (Willd.) AS A STUDY CASE.
Directeur: Thèse présentée pour l'obtention du titre de C. Lefèbvre Docteur en Sciences Agronomiques
au Grade scientifique
par Xavier VEKEMANS
Année académique 1991-1992
UNIVERSITE LIBRE DE BRUXELLES
FACULTE DES SCIENCES
FACULTE DES SCIENCES APPLIQUEES Section Interfacultaire d'Agronomie
Laboratoire de Génétique et d'Ecologie Végétales
EVOLUTION OF PLANT BREEDING SYSTEMS:
ARMERIA MARITIMA Mill. (Willd.) AS A STUDY CASE.
Directeur: Thèse présentée pour l'obtention du titre de C. Lefèbvre Docteur en Sciences Agronomiques
au Grade scientifique
par Xavier VEKEMANS
Année académique 1991-1992
Xavier Vekemans.
Titre de la t h è s e annexe:
Quelle influence sur réiaboration d e s programmes d'amélioration et
sélection d e s végétaux pourrait avoir le s u c c è s du génie moléculaire d e s
s y s t è m e s d'autoincompatibilité multialléliques.
CONTENTS CHAPTERl
INTRODUCTION
1.1 THE THEME OF THIS THESIS 2
1.2 ORIGIN OF THE DIVERSITY IN BREEDING SYSTEMS:AN INTRODUCTION 4
1.2.1 Inbreeding dépression 5
1.2.2 Genetic détermination of breeding Systems 7 1.2.3 Pollination ecology 8
1.2.4 Sex allocation 9 CHAPTER2
ARMERIA MARITIMA AS A STUDY CASE 11
2.1. THE PATTERN OF BREEDING SYSTEM VARIATION IN ARMERIA MARITIMA 12
2.2. ORGANIZATION OF THE THESIS 15 CHAPTER3
A MODEL FOR THE BREAKDOWN OF A NON-HETEROSTYLOUS DI- ALLELIC SELF-INCOMPATIBILITY SYSTEM 16
3.1 INTRODUCTION 17
3.2 MODELS FOR THE EVOLUTION OF SELHNG 20 3.3 MODEL AND ASSUMPTIONS 22
3.4 PARTIAL BREAKDOWN OF THE INCOMPATIBIUTY REACTION 25
3.4.1 Introduction 25
3.4.2 Conditions for the spread of self-compatible Ib mutants. ...27 3.4.3 Study of the initial rate of increase of self-compatible Ib mutants 30
3.4.4 Study of the equilibrium frequencies of self-compatible Ib mutants 32
3.4.5 Study of the effect of asymmetrical breakdown between morphs 33
3.4.6 Discussion 33
3.5 BREAKDOWN OF A DIMORPHIC INCOMPATIBIUTY SYSTEM BY RECOMBINATION WITHIN THE SUPERGENE 35
3.5.1 Introduction 35
3.5.2 Recombination within self-incompatible individuals 37 3.5.3 Recombination within partially self-compatible
individuals 37 3.5.4 Discussion 38
3.6. EVOLUTION AND BREAKDOWN OF HETEROSTYLY 41 3.6.1. Partial self-fertility in heterostylous populations 41 3.6.2. The évolution of heterostyly 42
CHAPTER 4
PARTIAL BREAKDOWN OF SELF-INCOMPATIBILITY IN DIMORPHIC METALUCOLOUS POPULATIONS OF ARMERU MARITIMA. 45
4.1 INTRODUCTION 46
4.2 MATERIALS & METHODS 48
4.3 RESULTS 53
4.3.1 Estimations of the outcrossing rate 53 4.3.2 Stability of self-fertility 56
4.3.3 Fruit-set under legitimate and illegitimate crosses 56 4.3.4 Inbreeding dépression 57
4.3.5 Pattems and levels of genetic variation 58 4.4 DISCUSSION 62
4.4.1 Estimations of the selfing rate 62
4.4.2 Metallicolous populations and the model of breakdown of di-allelic S.I. Systems 68
4.4.3 Population genetic structure 70
4.4.4 Is the selfing ability of metallicolous populations a relict characteristic or not? 72
CHAPTER5
EVOLUTION OF THE BREEDING SYSTEM IN ARCTIC AND AMERICAN POPULATIONS OF ARMEIUA MARITIMA 74
5.1 INTRODUCnON 75
5.2. MATERIALS AND METHODS 76 5.3RESULTS 81
53.1 Flower morphology 81
5.3.2 Biomass allocations to reproductive functions 82 5.3.3 Pollen/Ovule ratios 85
5.3.4 Pattems and levels of isozyme variation 87
5.3.5 Variation in phenolic compounds 89
5.4 DISCUSSIONS 89
5.4.1 The évolution of monomorphism 89
5.4.2 Evolution of outcrossing within monomorphic populations 96
5.4.3 Conclusions 103 CHAPTER 7
C^£N Jï^RÀL C^)NCLUS1 ^)NjS»»»»» •••>»••••»••••••••••»•••••*••>•• ••••••••••••••••••••••••••••••••^••••••••••••••^ lOS 7.1 THE INFLUENCE OF GENETIC AND ECOLOGICAL
FACTORS ON THE DIVERSITY OF PLANT BREEDING SYSTEMS 106
7.1.1 Genetic factors 106 7.1.2 Ecological factors 107
7.1.3 What about the rôle of constraints on the diversity of breeding Systems? 108
7.2 SELECTION THINKING AND THE EVOLUTION OF
BREEDING SYSTEMS 110
BOX 0.1. Summary of numerical parameters used in the model.
Symbol Meaning
5 1-d = Probability that a selfed ovule results in a progeny individual that survives to the next génération, relative to the probability for a non-selfed ovule
S Selfing rate of fully self-compatible phenotypes Çic and aC) Sbl Selfing rate of the partially self-compatible phenotype ACIb Sb2 Selfing rate of the partially self-compatible phenotype acib
Pl Probability of illegitimate cross-fertilization between phenotypes^ and C assuming partial breakdown of the incompatibility reaction
P2 Probability of illegitimate cross-fertilization between phenotypes a and c assuming partial breakdown of the incompatibility reaction
FL Critical size of the compatible pollen pool below which the realized female fertility déclines linearly
a 0.5/FL = size of the compatible pollen pool for self-incompatible individuals (0.5) relative
to the critical size below which the realized female fertility déclines linearly
Box 0.2. Summary of the parameters used to describe the genetic variation within and among populations.
PLP : percentage of loci polymorphic, that is with the frequençy of the most common allele being less than 0.99
K : average number of alleies per locus per population
Hq : observed heterozygosity (number of hetero2ygotes/total sample size), averaged overloci
F/5 : inbreeding coefficient, réduction in heterozygosity of an individual due to nonrandom mating within its subpopulation (departure from Hardy-Weinberg genotypic proportions)
Fgj : fixation index, réduction in overall heterozygosity due to differentiation in allele frequencies among the subpopulations (Wahlund effect)
FjY : overall inbreeding coefficient, réduction in overall heterozygosity including the contributions of both previous factors
Fg : inbreeding coefficient expected at equilibrium under selfing at a rate (1-/) and computed as Fg = ( l - /)/(! + /); AF = Fjs-Fe
Hj : gene diversity in the total population (here category of breeding system) Hg : mean gene diversity within subpopulations (here single populations) DsT ' mean gene diversity among subpopulations (with = i/j+D^j).
Ggf : proportion of gene diversity between subpopulations relative to the overall diversity (G^t = D^tIHj)
RsT ' inter-populational gene diversity relative to intra-populational diversity,
computed as/?57' = DJHs, with D;„ = Npop 1)57/(Npop-1)
CHAPTER 1
INTRODUCTION
INTRODUCTION 2
1.1 THE THEME OF THIS THESIS
Diversity is a universal attribute of natural populations of plants and animais.
Population genetics in dealing with genetic diversity, tiy to détermine the amoimt of variation existing in natural populations and to explain it in terms of its origin, maintenance and evolutionaiy significance (Hartl, 1988). This thesis is about the origin, maintenance and evolutionary significance of diversity in plant breeding Systems^ This is a particularly exciting topic as breeding Systems have a great impact on the organization of genetic variability within and among
populations, on the one hand, and show themselves a great amount of variation, on the other hand. Thèse aspects are outlined below.
(a) Impact on the organization of genetic varlability
The breeding System is one of the most important factors in shaping the genetic composition of populations (Hamrick, 1982), and hence their potential for evolutionary changes. Breeding Systems influence the amount, nature and organization of genetic variation of populations by determining the pattem of genetic recombination among individuals. For instance to exempliiy extrême situations, in a dioecious species, the following mating pattem is observed: each zygote is formed from the fusion of two gamètes, one coming from one maie individual and the other coming from one female individual; when in a cleistogamous species, all zygotes will be produced from the fusion of two gamètes from the same individual. The offsprings issued from thèse two species will show drastic différences in terms of mean heterozygosity.
^ Throughout this thesis I shall be using the terni "breeding Systems" in a vcry broad sensé to
include all aspects of sex expression that affect the relative genetic contributions to the next
génération of individuals within a species. Other terms are also commonly used with more or less
similar meaning, namely: sexual Systems, mating Systems, sometimes even genetic Systems. In
french one would use the less ambigous term "système de reproduction".
Table 1.1 Some common types of sex distribution within and between fiowers.
and within and between genêts of seed plants. Anthers aie designated i , and gynoeda are designated 9. Hermaphrodites are designated ^ .
Distribution of sex organs
within a within a Angiosperm
Name flower plant Breeding System species(%) dioecy <}or9 <;or? xenogamous (out-
OTJSsing) 4
gynodioecy ^ , d or 2 ^ or 9 xenogamous, geitonogamous, autogamous
7
monoecy 6 or 9 aDogamous, some
sel£ng,some Crossing
5
gynomonoecy ^ or 9 allogamousand 3
autogamous
hermaphrodity i allogamousand 72
autogamous 72
(other) 9
100
(from R i c h a r d s , 1 9 8 6 )
Table 12. Qassification of flowering plant senial systems^^Çadapted from Bawa & Beach, 1981).
A. Systems based on the spatial distribution of maie and female reproductive organs.
L Sezually monomorphic Systems diaracterized by only one gender dass of individuals L Eermaphroditism : Fiants bear only bisexual flowers.
2. Monoeeîsm : Rants bear maie and female flowers.
Z.Andromonoecîsm : Plants bear bisezual and maie flowers.
4. Cynomonoecîsm : Plants bear bisexual and female flowers.
n . Sezually dimorphic spedes characterired by only two gender classes of individuals.
1. Dîoeeîsm : Plants bear either maie or female flowers.
2. Cynodîoecîsm : Plants bear either female or bisemal flowers.
Z.Androdîoecîsm : Plants bear either maie or bisexual flowers.
B. Systems based on the temporal distribution of maie and female organs (Dichogamy).
L Protandry : Pollen removed from the anthers before stigmas attain receptivity.
2. Frotogyny. Stigmas become réceptive before anthers release pollen.
C Systems based on the spatial séparation of maie and female organs.
L Eerkogamy : Pollen présentation is spatially separated from pollen receipt, all individuals are of the same type.
2. Eeterostyly : Two or threc types of individuals bear différent fonns of flowers with respect to style and stamen length. Generally assodated with self-inccmpatîbility.
D. Systems based on the présence or absence of self-incompatibility alleles.
L Self'incompaxibilîty : Plants polymorphic with respect to the présence of self-incompatibllity alleles; pollinations involving pollen and sdgma sharing the same self-incompan'blility alleles, including self-pollinations, resuit in no fruit s e t
2. Self-compatiblU'Uy : Plants monomorphic and without the présence of self-incompatibility alleles;
all pollinations, including self-pollinations. resuit in fruit set.
* In addition to tfae Systems described below, tbere odst other Systems such as cieistogazny and various fonns of apomoda.
^ tfae Systems described below are not mutuaDy cœiusive.
INTRODUCTION 3 (b) Variation in breeding Systems.
A wide range of variation in breeding Systems (cf. Tables 1.1. and 1.2.) is found in flowering plants and the alternatives can be summarized as follows (Richards, 1986):
- Hermaphroditism versus unisexuallty: unisexuates are obligate outcrossers while hermaphrodites can often self-fertilize.
- Self-poUination versus cross pollination: flower morphology and temporal events in hermaphrodites détermine to some extent the amount of pollen transferred from anthers to stigmas within flowers (autogamy) and between flowers (allogamy). Outcrossing (xenogamy) occurs by allogamous pollen transfer between différent plant individuals.
- Self-fertilization versus cross fertilizatlon: self-pollination may not give rise to self-fertilization in the présence of mechanisms of rejection of selfed pollen (self-incompatibility Systems).
- Sexuality versus asexuality: végétative reproduction and production of unfertilized viable seeds (agamospenny) are différent alternatives to sex.
(c) Sélection acting on breeding System variation.
Variation in breeding Systems often comprises an important genetic
component available for sélection. Examples of highly genetically controlled mating Systems include dioecy (Westergaard, 1958 in Bawa, 1980), self- incompatibility Systems (de Nettancourt, 1977) and gynodioecy^ (Kheyr- Pour, 1980; Van Damme, 1983). Hence, breeding Systems themselves do evolve. This raises the interesting and controversial problem as to décide at what level, between-individual or between-population, sélection is acting
(Maynard Smith, 1989). Indeed the breeding System influences not only the fitness of the individual, that is its compétitive reproductive ability, but also the evolutionaiy potential of the population (through its action on the level of genetic variability).
See Table 12. for a classification of plant breeding Systems.
INTRODUCTION 4 The overall aim of the thesis is to seek evolutionaiy explanations for the
diversity in plant breeding Systems. This question will be kept in mind while studying spécifie evolutionaiy pathways in our biological model, the species complex Ameria maritima (Mill.) Willd. The choice of this model rely on
preliminary investigations showing a unique pattem of breeding System variation within the species complex (Baker, 1966; Lefèbvre, 1970; Lefèbvre & Vemet,
1989; see also Chapter 2).
The knowledge of the functioning of the breeding system of a species, together with a deep understanding of its conséquences on population genetic structure, are now considered as critical matter by the plant biologist or the agronomist. In plant breeding, for instance, knowing the mode of reproduction of a crop is essential to the choice of an appropriate breeding scheme and multiplication procédure. It is also important when trying to maintain any improved material in stabilized conditions. In genetic conservation management, sampling stratégies are based on the pattem of genetic diversity within and among natural
populations, which is under the control of the breeding system (Jain, 1975).
During the last two décades, the interest in studies of breeding Systems has been stimulated by two developments (Brown et aL, 1985; Hedrick, 1990). Firstly, the refinement of the procédures for foUowing mating events by the use of isozyme markers and DNA fingerprinting, secondiy major theoretical advances in several topics such as procédures for mating system estimation, effect of nonrandom mating on the organization of genetic variability, and évolution of breeding Systems. Both aspects will be illustrated in this thesis. The next section will give an overview of the main factors involved in the évolution of breeding Systems.
1.2 ORIGIN OF THE DIVERSITY IN BREEDING SYSTEMS:
AN INTRODUCTION
The difficulty in studying the évolution of breeding Systems arises as Maynard Smith (1989) quoted, "from the complexity of the phenomena that have to be explained, and the bewildering array of facts that have to be borne in mind".
Here, I would like to illustrate the effects of several important factors on the diversity of breeding Systems. Two types of factors are generally distinguished:
(1) genetic factors, like the genetic basis of sex détermination or the inbreeding
INTRODUCTION 5 dépression; (2) ecological factors, like poUination ecology, sex allocation^, sexual sélection, seed dispersai, prédation and some others biotic or abiotic factors.
Two of each will be discussed.
IJLl Inbreeding dépression
The term inbreeding dépression is used to designate the harmfùl effects usually accompanying close inbreeding in plants or animais. Thèse effects are expressed by an obvious décline in viabili^, fertiliQr and compétitive abili^, in other words by a lower fîtness of inbred lines. Inbreeding dépression is seen as the most important sélective pressure against the évolution of selfîng gènes in self-
compatible hermaphroditic populations. Two main théories have been proposed to account for inbreeding dépression, namely the overdominance and the partial dominance hypothèses (Charlesworth & Charlesworth, 1987b). The
overdominance hypothesis is based on a superiority of hétérozygotes at fitness- determining loci over both homozygotes (hétérozygote advantage). As
inbreeding tends to lower the percent of heterozygous loci in the progeny, the resulting décline in fitness is obvious. The second hypothesis, partial dominance, is concemed with the occurrence of partially récessive deleterious alleles in the génome. Thèse alleles are produced by mutation and, despite their négative effect on fîtness, are maintained at low frequencies in natural populations
because of their récessive nature (mutation-sélection balance). The réduction in fitness observed at the population level as a conséquence of this phenomenon is called the mutational load ("fardeau génétique" in french). When individuals become homozygotes for such alleles, as promoted by inbreeding, their relative fitness will be reduced.
The way by which inbreeding dépression affects the évolution of breeding Systems dépends greatly on its genetic détermination, which can vaiy itself. This has been illustrated by modeling the effect of various selfing rates on the level of inbreeding dépression expected in natural populations assuming either
hypothesis: overdominance vs partial dominance (Charlesworth & Charlesworth, 1987b, 1990;Charlesworth et aL, 1990). Inbreeding dépression aiways increases
3 The allocation of resources between maie and female reproductive functions.
k{S)
0.0 -i • i ' 1 ' 1 ' 1 ' I
0 . 0 0.2 0 . 4 0 . 6 0 . 8 1 . 0 S
Fig. 1.1. Relationshîp between the relative înbreeding dépression, k(S), and the selfîng rate, S, for varions values of the sélection coefficient, s, and of the degree of dominance of deleterious alleles, A, tmder the partial
dominance model (&om Charlesworth & Charlesworth, 1987b).
Fig. 1.2. Rates of change in frequency of modifier alleles changing the selfing rates i^lp) under the partial dominance model with multiplicative interactions of the fîmess loci . The modifîers were introduced into starting
populations having différent selfîng lates denoted by S
(from Charlesworth & Charlesworth, 1990b).
INTRODUCTION 6 with the selfing rate when referring to the overdominance modeH while the reverse is true with the partial dominance model. Moreover, within the last model, a large variation in the rate of decrease of inbreeding dépression with selfing rate is found in relation with the degree of dominance of deleterious alleles and the sélection intensity (see Fig. 1.1.). However, the relative importance of thèse différent types of genetic déterminations of inbreeding dépression that actually occurs in natural populations is still unknown. The scarce expérimental data reviewed by Charlesworth & Charlesworth (1987b) seem to favour the partial dominance hypothesis, with approximately the same contribution from mildly deleterious and highly deleterious (sub-lethals) mutations.
Another point, relevant to the effect of inbreeding dépression on the diversity of breeding Systems, is worth mentioning hère. Campbell (1986) and Holsinger (1988b) have pointed out that evolutionary changes in selfmg rate should be accompanied not only by (1) a change in the level of inbreeding dépression (see above), (2) a change in the frequencies of selfing gènes, but also by (3) some degree of genetic association^ between selfing gènes and loci responsible for inbreeding dépression ("fitness loci"). The last factor can have a dramatic effect on the prédictions of the models for the évolution of selfing (Holsinger, 1991).
One of the conséquences of this association has been illustrated by Charlesworth et ai (1990). They show that under certain circumstances, gènes inducing a large increase in selfing rate will be advantageous while gènes inducing only minor increases in selfing rate will be selected against (Fig. 1.2.). So, for a same average level of inbreeding dépression in the population, the harmful effect of inbreeding will prevent the évolution of selfing in one case and in the other will not.
* Assuming symmetrical overdominance, that is, both homozygotes have the same fîtness while the hétérozygote has a higher fitness.
^ The existence of thèse genetic associations can be intuitively understood as follows: in a population practicing mixed selfing and outcrossing "an individual that is heterozygous at one locus is more likely to have been produced through outcrossing than is a randomly chosen individual .„ [given that] it is also more likely to be heterozygous at a second locus... This implies that an individual that is heterozygous at a locus determining the selfing rate is more likely to be heterozygous at fitness-determining loci than is a randomly chosen individual in the population.
Thus, the génotype at the selfing locus will not be independcnt of the fitness" (Holsinger, 1988b).
INTRODUCTION 7 In short, three parameters associated with the phenomenon of inbreeding
dépression have an influence on the évolution of plant breeding Systems: the genetic détermination of inbreeding dépression, the average level of inbreeding dépression of the population, and the associations between fitness and mating System loci.
1.2.2 Genetic determinatioii of breeding Systems
The nature of the genetic détermination of a particular sexual phenotype may have an impact on its further évolution. An good example refers to the évolution of maie sterility (subsequently noted as MS) in plant populations. The
conditions for the spread of MS mutants in a population, and for their
maintenance, will dépend greatly on the mode of inheritance of MS: nuclear or cytoplasmic (Lewis, 1941; Charlesworth & Charlesworth, 1978). With nuclear inheritance, a MS mutant must be more than twice as fecund as an
hermaphrodite in order to spread, while a ratio slightly superior to one is
sufficient with cytoplasmic détermination. On the other hand, under the last, no stable polymorphism (mixed frequencies of male-steriles and hermaphrodites, that is gynodioecy) is expected.
Extensive empirical studies have found some gynodioecious Systems involving both nuclear and cytoplasmic gènes (Kheyr-Pour, 1980; Stevens & Richards,
1985; Van Damme & Van Delden, 1982; Van Damme, 1983) and it seems that nuclear-cytoplasmic inheritance of MS applies to many gynodioecious species (Ross, 1978; Charlesworth, 1981). Theoretical studies have shown that, under particular assumptions, nuclear-cytoplasmic MS may reach a stable polymorphic equilibrium (Delannay et aL, 1981; Charlesworth, 1981; Gouyon et al, 1991).
Interestingly, stable limit cycles, that is consistent fluctuations of the proportion of maie stériles in gynodioecious populations can be maintained without
invoking any ecological cause (Gouyon et al, 1991).
Beside thèse equilibrium prédictions, a dynamic model involving altemation between colonization stages with a cytoplasmic inheritance of MS and
establishment stages with a mostly nuclear détermination bas been proposed
(Gouyon & Couvet, 1985; see also Frank, 1989). This model elegantly suggests
favourable conditions for the spread of MS (conditions under cytoplasmic
détermination are easier to fulfil) together with a feedback mechanism
controUing the frequencies of male-sterile individuals.
INTRODUCTION 8
1 . 2 3 P<riliiiation ecdogy
TWo aspects of pollination ecology are relevant to a discussion on the factors influencing breeding System diversity.
The first one is the type of pollination System. Qassical studies in pollination ecology have led to the description of pollination syndromes, sets of characters that represent adaptations to particular types of pollinators (Wyatt, 1983). Some corrélations between particular syndromes and breeding System have been observed. One example is the évolution of dichogamy in insect-pollinated plants.
Dichogamous plants poUinated by bees and Aies are generally protandrous while protogyny is encountered mostly when beetles and wasps are the main
pollinators (Bawa & Beach, 1981; Wyatt, 1983). One explanation involves the foraging pattems of pollinators, with bees and Aies generally visiting individual inflorescences form bottom to top while beetles and wasps are moving
downwards (Faegri & van der Pijl, 1979). As most plants show acropetal maturation, flowers at advanced stages are localized to the bottom of inflorescences. Thus in both cases, protandiy in upward and protogyny in downward poUinated plants, it is the présentation of mature stigmas to the newly-arrived pollinator (supposedly covered with pollen from another individual) that is maximized (Wyatt, 1983). Différent hypothèses can account for this observation. First, the avoidance of self-pollination, or of interférence between the functions of pollen receipt and pollen dispatch (Lloyd & Webb,
1986) may be favoured when pollination of flowers at the "female stage" occurs in the first place. Second, a sexual sélection argument, compétition between individuals for maie (pollen) success may have selected for a mechanism by which mature stigmas remove foreign pollen from the pollinator while the flowers at a maie stage cover the pollinator with new pollen and send them off (Chamov, 1982).
Another example of the influence of pollination type on the évolution of breeding Systems concems the distribution of selfing rates among wind- pollinated and among insect-pollinated plants. Aide (1986) has shown that anemogamous plants show a bimodal distribution of selfing rate (mostly selfed and mostly outcrossed species), while entomogamy is characterized by a
continuons distribution. Most genetic models, however, predict a bimodal distribution independently of pollination type (Schemske & Lande, 1985;
Charlesworth et aL, 1990). The discrepancy between observed and theoretical
INTRODUCTION 9 distributions, in entomogamous plants, may be related to the greater
unpredictability of animais as pollen vectors leading to lower rate of response to sélection on the mating System (Schemske & Lande, 1986).
The second important aspect of poUination ecology is the variation in poUinator tvailability. When poUinators are scarce such as at margins of a species' range, or when plant density is too low to attract effîciently poUinators, outcrossing events become rarer and the average seed-set (percentage of ovules ripening into seeds) of the population may be dramatically reduced. Thèse conditions are favourable to the évolution of autogamy if flowers are self-compatible and floral structures are such that within-flower selfmg in the absence of poUinators is occurring (Richards, 1986). This factor has been commonly invoked to explain the évolution of the selfmg habit in plants (Arroyo, 1975; Baker, 1966; Solbrig &
RoUins, 1977; Schoen, 1982a; Piper et al, 1986; Barrett et al, 1989) and is known as the reproductive assurance hypothesis (Jain, 1976). Similarily, the rôle of poUinator limitation in the évolution of apomixis has also been suggested (Bennett & Elgar, 1987).
1.2.4 Sex allGcation
Most flowering plants are hermaphrodites (Table 1.1.). Based on this
observation Chamov et al (1976) have developed a theory, subsequently called the sex allocation theory, that tries to predict when hermaphroditism as opposed to dioecy would be an evolutionary stable strategy (E.S.S.). The theory is based on the foUowing assumptions (see Goldman & WUlson, 1986): (1) the total reproductive effort of an individual is resource limited and hence its total reproductive success also; (2) the trade-off hypothesis, that is the total
reproductive allocation is divided between maie and female functions and an increased allocation to one of thèse nécessitâtes a decrease to the other; (3) there is a genetic variation in sex allocation. Based on thèse assumptions they pointed out that the stabUity of hermaphroditism versus dioecy would dépend primarily from the shape of the trade-off between maie and female reproductive success within an hermaphrodite. More precisely several authors have shown that the ESS breeding System would dépend on the relationship between reproductive success via either the maie or the female function, and the
proportion of resources devoted to each (Chamov, 1979 and 1982; Charlesworth
& Charlesworth, 1981). If thèse relationships are less than proportional, i.e.
INTRODUCTION 10 foUow a law of diminishing retums, then hermaphroditism is selected for. If they are more than proportional then dioecy is favoured.
The observed corrélation between dioecy and animal-dispersed seeds (Bawa, 1980; Givnish, 1980) has been interpreted in terms of the sex allocation theoiy.
The authors argued that under animal dispersai, a larger seed crop attracts a disproportionate number of dispersai agents and thus results in a more than proportional relationship between female allocation and female fitness. Thèse are precisely the conditions favouring dioecy over hermaphroditism.
My aim was not to give a complète account of the theory of sex allocation but to point out that factors like the resource limitation of maie and female
reproductive success can also influence the diversity of breeding Systems.
o m o z c '< in
CHAPTER 2
ARMERIA MARITIMA
AS A STUDY CASE
Fig. 2.1. Scanning électron micrographs of the stigmatic papillae of (a) COB and (b) PAP morphs, and of the pollen of (c) œ o
and (d) PAP morphs (from L. Devos). The white line in the bottom indicates 100 and 50 \m for stigma and pollen, respcctively.
Armeria maritima 12
2.1. THE PATTERN OF BREEDING SYSTEM VARIATION IN ARMERIA MARITIMA
Armeria maritima (Mill.) Willd. is a polymorphic and widespread perennial species, found through much of the northem hémisphère in open coastal and inland habitats. A botanical description of the plant is given in Table 2.1. Within the species important variations in the breeding System have been reported by Baker (e.g. 1966) and his work on Armeria maritima has become a classic model about the évolution of plant breeding Systems. He showed that in most of Europe the mating pattem of A. maritima is sporophytically controUed by a dimorphic self-
incompatibility System: one morph, hereafter called COB, is heterozygous at the incompatibility locus having "cob" stigma (allele C) and coarsely reticulate pollen (allele A); the other morph, called PAP, is homozygous with a papillate stigma (allele c) and finely reticulate pollen (allele a). Normally only intermorph crosses are fertile so that homo^gous PAP (acac) and heterozygous COB (ACac) morphs are expected to be found in a one to one ratio. The pollen and stigma ^ e s are illustrated in Fig. 2.1.
Table 2.1. Botanical description of Armeria maritima subsp. maritima (after Woodell & Pale, unpublished)
Perennial hcrb with a short-branchcd, woody root-stock. Leavcs 10-150 x 0.4-15 mm, linear, flat, acute or blunt, l-(3) veined at least at the base, glabrous-ciliate with veiy narrow scarious margins.
Scape 10-560 mm, crect, leafless, shortiy pubescent, rondy, glabrow, with its growing point covered by a tubular scarious sheath, 3-21 mm, which is a prolongation of the outer involucral bracts.
Outer involucral bracts 2-15 (-25) mm, more or Icss grecn on the back, acuminate or mucronate;
inncr involucral bracts wholly scarious, blunL Head a capitulum, 10-25 (-30) mm in diamcter;
fiorets usually bracteate in groups of 2-4 sequentially developing in bracteate cymose spikelets.
Spikelet bracts 4.5-10 x 33-9 mm, blunt with a wide scarious margin surrounding a green central portion; florct bracts much smaller, clliptic-oval, scarious. Calyx 4.5-9 mm, stalked, funnel-shaped, pubescent, (sometimes only on the 10 ribs). Calyx-teeth 5, acute, less than half the calyx length, joined by a scarious pleat Corolla of five petals, joined only at the base, 7-11 x 2.5-4.5 mm, normally reddish-purple, but vaiying through pale pink to white, occasionally deep red-purple.
Stamens, 5, joined to the base of the petals, their filaments, 5-8 mm, broadened below. Styles 5,
filiform, free to near the base, pubescent below. Fruit single-seeded, endosed in the calyx by a
persistent corolla, dehiscing transversely above or irregularly below.
F i g n n î ^ SchcmcloiQustimtethcapcrmtionofwithm-motphincompatibility
in thc thrift. Arment mtritima. Upid i* bomc cxtcmal to thc pollen ; n i n in the findy rcticuUtcd papillate grain (below), and is ablc to adhère to thc unooth cob ftigma. but I c u well to the papillate itigma. In thc coaraciy rcticulatcd cob g n i n , the Upid is suniwn in the tcctum apertures, whert it can only makc contact with
the papillate »ti5ina.(from R i c h a r d s , 1 9 8 6 )
Oimorphic population T y p e A : 5 0 A C j c Type B:50ac.ae
lOOAc^c' Monomorphic population
FicX.S.Sdinnc ibewinf tht functloniai ef Uw brccdini lyittn in Armtria. Letnid: ^^^mm natuni fertile combinatiom; — — expérimental fertile comfainationi; — . - — itcrile rambination.
