Citation
Phylogenetic and population genetic analysis of Fusarium oxysporium f. sp. cubense, the causal agent of fusarium wilt on banana

Material Information

Title:
Phylogenetic and population genetic analysis of Fusarium oxysporium f. sp. cubense, the causal agent of fusarium wilt on banana
Creator:
Koenig, Rosalie Lynn, 1962-
Publication Date:
Language:
English
Physical Description:
x, 113 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Alleles ( jstor )
Diseases ( jstor )
DNA ( jstor )
Enzymes ( jstor )
Fusarium ( jstor )
Genotypes ( jstor )
Haplotypes ( jstor )
Mitochondrial DNA ( jstor )
Pathogens ( jstor )
Population genetics ( jstor )
Bananas -- Diseases and pests ( lcsh )
Dissertations, Academic -- Plant Pathology -- UF ( lcsh )
Fusarium oxysporum -- Evolution ( lcsh )
Fusarium oxysporum -- Genetics ( lcsh )
Plant Pathology thesis, Ph. D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 105-112).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Rosalie Lynn Koenig.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
027620582 ( ALEPH )
37229272 ( OCLC )

Downloads

This item has the following downloads:


Full Text









PHYLOGENETIC AND POPULATION GENETIC ANALYSES OF FUSARIUM OXYSPORUM F. SP. CUBENSE,
THE CAUSAL AGENT OF FUSARIUM WILT ON BANANA














BY

ROSALIE LYNN KOENIG














A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1997































Copyright 1997

by

Rosalie Lynn Koenig














I dedicate this research to my father and mother who always stressed the

importance of education, and to my husband Tom and daughter Amaleah, who put up with me during this stressful period of my life.













ACKNOWLEDGMENT S


I would like to thank Dr. Corby Kistler for his support, patience, understanding and encouragement through all stages of my graduate studies.

I would like to thank Dr. Dave Mitchell for his guidance, counsel, encouragement and personal commitment to excellence and graduate students. Many thanks to my committee members Drs. Randy Ploetz and Eduardo Vallejos for their input and cooperation. Additional thanks to Ulla Benny, Gerald Benny and Lyndel Meinhardt who always lended their technical expertise and support.

Special thanks to Dr. Brian Bowen for his expertise in genetic analysis; he always made time for me and provided clear explanations of complex subject matters. I thank the NSEP for funding the field work in Honduras. I thank Drs. Maricio Rivera and Phil Rowe for their support in Honduras. Thanks to my many friends and classmates who shared their dreams and aspirations.

















iv














TABLE OF CONTENTS

page

ACKNOWLEDGMENTS............................................ iv

LIST OF TABLES............... ....................... .......... vii

LIST OF FIGURES................................................ viii

ABSTRACT ........................... ........................... ix

CHAPTERS

1 INTRODUCTION ...................................1

The Host..................... .............................. 4
Genetic Characterization of Fusarium oxysporum f sp. cubense.............8
Clonality and its Consequences in the- Study of Population Genetics of
Asexually Reproducing Organisms ............................... 11

2 FUSARI9M OXYPSORUM F. SP. CUBENSE CONSISTS OF A SMALL NUMBER OF DIVERGENT AND GLOBALLY
DISTRIBUTED CLONAL LINEAGES ............... 15

Introduction................................................. 15
Materials And Methods ......................................... 17
Results .................................................... 30
Discussion ................................................42

3 GENETIC VARIATION IN TWO HONDURAN FIELD POPULATIONS OF FUSARJUA'IOKYSPORUM F. SP. CUBENSE ...... 47

Introduction ................................................ 47
Materials and Methods ............................ ............. 50
Results. .................................................... 60
Discussion ................................................. 65

4 SUMMARY AND CONCLUSIONS ........................ .......73



V









APPENDICES

A EXPLORING THE POTENTIAL OF USING POLYMERASE CHAIN REACTION METHODOLOGIES IN PHYLOGENETIC STUDIES
OF FUSARIUM OXYSPORUMF. SP. CUBENSE .................... 81

In tro d u ctio n . . . . . . 8 1
R esults and D iscussion ........................................... 82

B ADDITIONAL INFORMATION FROM PHYLOGENETIC STUDY ....... 91

C ALLELIC DATA SCORED AS PRESENCE OR ABSENCE FOR EACH PROBE ENZYME COMBINATION; ISOLATE NAME AND
VEGETATIVE COMPATIBILITY GROUP INCLUDED AS A
REPRESENTATIVE FOR EACH UNIQUE RESTRICTION
FRAGMENT LENGTH POLYMORPHISM HAPLOTYPE ........... 94

D LIST OF ISOLATES OF F. OXYSPORUMF. SP. CUBENSE, THEIR VEGETATIVE COMPATIBILITY GROUP, HAPLOTYPE AND
L IN E A G E ................................................. 100

REFERENCES .................................................. 105

BIOGRAPHICAL SKETCH ................................. ......... 113























vi













LIST OF TABLES

Table pagg

2-1: List of isolates, their vegetative compatibility groups (VCG), clonal lineages,
the cultivar, and geographical regions from which they were collected ....... 18 2-2: Similarity matrix of simple matching coefficients based on restriction fragment
length polymorphisms for selected isolates of Fusarium oxysporum f. sp.
cubense, F. oxysporum f sp. lycopersici and F. oxysporum f. sp. niveun .... 35 2-3: Clonal lineages of Fusarium oxysporum f sp. cubense isolates, their
geographical distributions and corresponding vegetative compatibility
groups (V C G ) ................................................. 36

2-4: A comparison of allelic data for isolates in FOC I, FOC II and FOC VII ........ 3,9 2-5: Clone-corrected measurements of gametic disequilibrium among pairs of
alleles in a world-wide collection of Fusarium oxysporum f. sp. cubense ...... 41 3-1. Genotypes of banana planted in Field I ................................. 51

3-3. Field design of banana plants in rows of the race 2 disease screening plot ....... 55 3-4. Isolates of Fusarium oxysporum f. sp. cubense collected in Field 1 and their
incidence on different host genotypes ................................ 63
















vii













LIST OF FIGURES

Figure p~ge

2-1: DNA of isolates was digested with Eco RV restriction enzyme and probed
w ith clo ne 187 .. .. .. .. .. .. . . . 26

2-2: Frequency distribution of RFLP haplotypes among the 165 isolates
representing a world collection of Fusarium oxysporum f. sp. cubense ....... 31 2-3: Midpoint rooted 50% majority rule consensus tree representing 500
bootstrap replicates ............................. .... ......... 32

2-4: Mitochondrial DNA haplotypes of selected Fusarium oxysporum f sp.
cubense isolates ................................................ 40

3-1: mtDNA haplotype of selected Honduran field isolates ..................... 61

3-2: HindIII digested DNA of selected Honduran field isolates that were probed
with clone pEY10 to generate DNA fingerprints ...................... 64

3-3: Incidence and distribution of mtDNA haplotypes of Fusarium oxysporum f.
sp. cubense on host genotype ...................................... 66



















viii













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

PHYLOGENETIC AND POPULATION GENETIC ANALYSES
OF FUSARI1UM OXYSPORUM F. SP. CUBENSE,
THE CAUSAL AGENT OF FUSARIUM WILT ON BANANA By

Rosalie Lynn Koenig

May 1997

Chairperson: Dr. H. C. Kistler
Major Department: Department of Plant Pathology

A worldwide collection of Fusarium oxysporum f. sp. cubense was analyzed using anonymous, single-copy, restriction fragment length polymorphism (RFLP) loci. Of the 165 isolates examined, 72 RFLP haplotypes were identified. Consistent with strict clonal reproduction, individuals with identical haplotypes were geographically dispersed, and clone-corrected tests of gametic disequilibrium indicated significant nonrandom association among pairs of alleles. Parsimony analysis divided haplotypes into two major branches. Ten clonal lineages were identified based on coefficients of similarity between 0.94-1.00. The two largest lineages had pantropical distribution. Isolates comprising one lineage (FOC VII) may represent either an ancient genetic exchange between individuals in the two largest lineages or an ancestral group. The two largest lineages (FOC I and II) and a lineage from East Africa (FOC V) are genetically distinct; each may have acquired the ability to be pathogenic on banana independently.

ix








Populations of the pathogen were sampled from two disease screening fields in

Honduras. Field 1 was planted to 10 genotypes of banana and 97 isolates were recovered from host tissue. Isolates recovered in Field I comprised three mtDNA haplotypes, and their incidences and distributions varied with the host genotype sampled. DNA fingerprints indicated that the isolates represented five clones. Field 2 was planted to a single genotype of banana. The 59 isolates collected from host tissue represented a single mtDNA haplotype and clone. The frequencies with which the pathogen clones infected the host were dependent upon host genotype. Two races of the pathogen (race 1 and 2) have been reported in Honduras. Since the cultivar Highgate is susceptible to race I but resistant to race 2, isolates recovered from the cultivar Highgate were assumed to represent race 1. Isolates within a lineage associated with race 1 isolates were preferentially recovered from Highgate and the Fundaci6n Hondurefia de Investigacio'n Agricola (FHIA) hybrids. Isolates representing a second lineage were recovered at a low frequency from Highgate, suggesting that these two lineages differ in their ability to infect bananas susceptible to race 1. Assessing the pathogen populations and determining the lineages of isolates infecting particular banana breeding lines could be useful for banana breeding programs.















x













CHAPTER 1
INTRODUCTION


The genus Fusarium (Link) is comprised of fungi which have evolved from an ancient ancestor whose geological age has been postulated to date back at least 200 million years (Snyder, 1981). Whether or not this is an accurate estimate of the age of this important fungal group, the genetic diversity present in members of the genus Fusarium is the result of evolutionary processes which have developed over a long time span.

Fusarium oxysporum (Schlecht.) Synd. & Hans. is an ubiquitous, soil borne fungus that is comprised of strains that are ecologically adapted to diverse environments and which have the ability to utilize an array of substrates in the soil (Burgress, 1981). The reproductive success of strains within this species depends on the ability to produce a large number of propagules, in the form of microconidia or macroconidia, as well as resistant resting spores, referred to as chlamydospores, that can ensure long-term survival during adverse environmental conditions. The morphological characteristics of these propagules also have been used by Wollenweber and Reinking (1935) to include the species in section Elegans. No teleomorph have been described for the species.

Plant pathogenic isolates ofF oxysporum cause diseases known as vascular wilts and constitute a small portion of the strains within this species. However, because of their importance to agriculture, Snyder and Hansen (1940) developed the concept offormae speciales to recognize plant pathogenic strains ofF. oxysporum that were morphologically





2

indistinguishable from saprophytic strains of the species but had the ability to cause disease on specific hosts. Over 70 formae speciales have been described (Kistler, 1997). Although formae speciales are not recognized by the International Code of Botanical Nomenclature, they are useful to plant pathologists because they provide informal groupings of morphologically indistinguishable isolates. Isolates within a forma specialis may be pathogenic on a specific host, such as F. oxysporum f sp. pisi, which causes a wilt disease on peas, Pisum sativum L., or may have a broader host range, such as 1. oxysporum f. sp. vasinfectum (Atk.) Snyd. & Hans., which is pathogenic on plants in the families Malvaceae, Solanaceae and Fabaceae (Windels, 1991). Two assumptions that have been made about isolates within a forma specialis are that 1) they are genetically more similar to each other than to isolates of different formae speciales, and 2) that formae speciales are monophyletic groupings (Kistler, 1997).

Fusarium wilt of banana, commonly referred to as Panama disease, is caused by F. oxysporuni Schlechtend.: Fr. f sp. cubense (E.F. Sm.) W.C. Synder & H. N. Hans.

This fungus systemically colonizes the xylem tissue of susceptible banana cultivars, disrupting normal water uptake (Beckman, 1987). This disruption eventually results in wilt and the subsequent death of the plant. Studies described by Beckman (1990) indicate that conidia in aqueous solutions are readily taken up into the vascular elements and carried in the transpiration stream to sites of entrapment at vessel endings. Subsequent secondary colonization occurs in susceptible cultivars by the dispersal of conidia through the vascular system. Resistant cultivars prevent spread by the formation of occluding gels and tyloses that wall off the lumena of the infected vessels.






3

To date, three races of the pathogen, 1, 2 and 4, have been described which are pathogenic on banana; these races are differentiated by the banana cultivars Gros Michel, Bluggoe, and Cavendish, respectively. Historically, this pathogen has caused large scale disease epidemics when susceptible banana cultivars have been grown in the presence of the pathogen (Ploetz, 1994).

The first recording of Fusariumn wilt of banana was made in southeast Queensland in 1874, on the cultivar Silk, known locally as Sugar (Bancroft, 1976). The history of this disease generally paralleled the dissemination of susceptible genotypes to new areas. The edible bananas are asexually propagated, typically using rhizomes or sucker plants. Occasionally, infested rhizomes were brought to new plantations and this is likely how the pathogen was widely distributed. Interestingly, one region where the pathogen has been found only recently is the vast area of Oceania where susceptible cultivars have been grown for millennia (Shivas and Philemon, 1996).

Prior to the late 1950s, monoculture of the race 1 -susceptible cultivar Gros Michel resulted in severe and wide-spread epidemics in the export banana trade. The disease became economically less important in export production once the trade converted to race 1-resistant Cavendish cultivars. However, approximately 20-40 years after Cavendish cultivars were planted widely in a number of subtropical regions, they too began to succumb to the disease. Outbreaks on Cavendish cultivars were reported in the Canary Islands in 1926 (Ashby, 1926), the 1940s in South Africa, the 1950s in Australia, and the 1960s in Taiwan (Ploetz et al., 1990). In the late 1970s outbreaks on Cavendish were observed in the Mindinao province of the Philippines. However, in contrast to outbreaks in the subtropics, disease outbreaks in the Philippines occurred in localized patches that






4

tended not to spread (Ploetz et al., 1990). All isolates in VCG 0122 are from the Philippines and they all have been recovered from Cavendish cultivars.

In the late 1970s Su et al. (1977) conclusively demonstrated that isolates capable of causing disease on Cavendish were different from race I and 2 isolates. These isolates, which also cause disease on race 1- and race 2- susceptible cultivars, were designated race 4.

In addition to economic losses in the export trade, Panama disease is a serious

problem on locally consumed dessert bananas and cooking bananas (Ploetz, 1992; Ploetz et al., 1992; 1994; 1995). Only 14% of the bananas that are produced in a given year are exported. The remaining 86% (65 million metric tons in 1992) are consumed locally (FAO, 1993) and are often staple foods for poor, subsistence farmers in the tropics (INIBAP, 1993). In total, the impact of Fusarium wilt on these bananas is far greater than on bananas produced for export.

The Host

The F. oxysporum f. sp. cubense: banana pathosystem likely developed from a process of co-evolution between the host and pathogen, resulting in complementary evolutionary changes in both organisms. Bananas belong to the order Zingiberales, which consists of eight families of tropical and subtropical rhizomatous perennial plants that inhabit moist forested areas or invade disturbed sites (Kress, 1990). Bananas belong to the family Musaceae, which contain three genera (Musa, Musella, and Ensete) whose natural range extends from West Africa to the Pacific. Edible bananas are the most important cultivated crop of this family. Two additional minor crops of economic





5

importance are abacd (M. textilis), a fiber crop, and cultivated forms of Ensete ventricosum, a food and fiber plant of upland Abyssinia.

Edible bananas have been domesticated from two species, M acuminata and M balbisiana. Simmonds and Shepherd (1955) used 15 phenotype characters and scored each on the scale of I (accuminata expression) to 5 (balbisiana expression) to develop a system to classify edible bananas. From this, the genetic constitution of any edible banana can be specified using a genome formula indicating the contributions from M accurninata

(AA) and M balbisiana (BB).

Edible M. acuminata cultivars are either diploid (AA) or triploid (AAA), and their center of origin includes Southern Thailand, Sumatra, Java, Borneo, Burma, and Malaysia, which is also the primary center of diversity for the species. This species is most diverse in peninsular Malaysia and the equatorial islands where the wild species is native. Six subspecies of M. acuninata are recognized. Although the banana clone Gros Michel has phenotypic characters that suggest descent from M acunzinata subsp. malaccensis (Stover, 1990), the parents of most clones are not known.

Simmonds (1962) argues that edibility in bananas arose in Malaysia through parthenocarpy in M. acurninata diploids followed by triploidy (AAA). Through the process of domestication, man has selected for both parthenocarpy and sterility, which are genetically independent characteristics. Additionally, domestication has led to selection of triploidy because it contributes to high plant vigor and fruit growth rate, and confers gametic sterility (Simmonds, 1962).

In contrast, M balbisiana has a center of origin in a region including the countries of India, Burma, and Northern Thailand. It is less genetically diverse than M acuminata.





6

It is presumed that edibility did not arise in M balbisiana because no naturally occurring parthenocarpic clones have been identified. Similarly, no purely M. balbisiana triploids or tetraploids are known. M balbisiana is naturally cross fertile with a number of Musa species, including M. acuninata, but it is reproductively quite isolated (Simmonds, 1962). The hybrid groups (AB, AAB, ABB) are peripheral, being diversified in India and, to a lesser extent, in Indochina and eastern Malaysian areas from which wild M acuninata is locally absent but in which M balbisiana is native (Simmonds, 1962). The transport by humans of AA-type bananas into areas occupied by wild M. balbisiana is inferred to have provided the opportunity for interspecific hybridization (Simmonds, 1962). Characteristics such as plant hardiness, drought resistance, disease resistance to nematodes and yellow Sigatoka, and starchiness of fruit that are found in interspecific hybrid bananas are believed to be contributed from M balbisiana (Simmonds, 1962). Interspecific hybrids between M balbisiana and M. acuminata occur in nature because the species are sympatric over a large area of southeast Asia and introgression occurs often.

A secondary center of diversity for bananas is East Africa. In East Africa,

mixtures of genotypes are grown for local use and particular dietary needs. The East African baking and beer type bananas (AAA) are unique to the region and are often found in mixed plantings with cultivars of exotic origin, such as ABBs and various dessert types (Ploetz et al., 1990). Inland cultivation of banana in East Africa is based primarily on triploid clones (AAA) but the coastal areas have much more diverse populations. Coastal East Africa is the only location outside of Southeast Asia where there is an assemblage of edible diploids cultivars and extensive local diversification. The most striking






7

accumulations of unique clones are in upland East Africa and the Pacific Islands, where it appears that subsistence farmers actively preserved these clones (Simmonds, 1962).

Murdock's (1959) discussion of plant introduction in relation to African history suggests that Malaysian food plants entered Africa in the hands of Indonesian migrants about 2000 B.P. Bananas, as well as taro, yams, breadfruit, coconut, some rices and sugarcane, are adapted to moist climates and became staple foods of the wetter parts of Central Africa (Simmonds, 1962). Madagascar appears to be the principal point of entry of bananas into Africa. Simmonds' (1962) view is that the early African cultivars reached Madagascar from Malaysia late in the first half of the first millennium AD. From West Africa, Portuguese travelers took plants to the Canary Islands early in the fifteenth century, and a least one clone traveled to Hispaniola in 1516. Simmonds (1962) believed that this represented the first of many introductions made into the New World. However, Langdon (1993) suggests, based on linguistic, botanical, and archeological data, that banana was present in the New World before the Spaniards arrived.

Presently, the only location with a high level of genetic diversity of M acurninata is Papua New Guinea and surrounding islands. Here, the indigenous people have preserved an intact population of primitive cultigens, a relic of an early stage in the evolution of the group. Jarret (1992) used 98 phylogenetically informative restriction fragment length polymorphisms (RFLPs) to characterize eight species representing three sections of Musa from Papua New Guinea, as well as one species of Ensete. The study found only a distant relationship between M balbisiana and other species of Musa; M. balbisiana was not in the same clade as the other species within the section Musa (also






8

referred to as Eumusa). The diploid land races from this section clustered with M. acunfitiata. All accessions of M acurninata shared a large number of alleles.

Genetic Characterization of Fusarium oxvsvorurn f. sp. cubenise

Cladistic analyses of the phylogenetic relationships among isolates of a fungal

species are based on the concept that members of a species have evolved from a common ancestor (Scotland, 1992). This ancestor possessed primitive genetic characteristics which may have changed through the evolutionary history of the organism. Derived character changes, those changes that are different from the ancestral type, are the result of evolutionary processes which consist of both genetic (mutation, recombination, and transposon rearrangements) and spatial (migration, divergence, and convergence) forces.

Evolutionary studies of organisms aim to determine phylogeny by examining the genetic relatedness among members of a group. Studies may involve closely related organisms, such as fungal isolates of a given species, or organisms more distantly related, such as all fungal isolates within a location. However, many evolutionary studies are conducted at the genus or species level. These types of studies rely on identifying genetic markers that discern among individuals sampled in the study. Prior to the development of molecular techniques, researchers primarily relied on phenotypic traits to develop evolutionary theories. However, the development of molecular techniques, such as isozyme analysis, restriction fragment length polymorphism analysis, random amplified DNA analysis and gene sequencing, have made it possible to objectively assess traits at the genotypic level.

Isolates ofF oxysporuni have been characterized using morphological,

biochemical and genetic markers (Kistler, 1997). Additionally, vegetative compatibility





9

groups (VCGs) based on work on Fusarium moniliforme also have been used to differentiate similar isolates of oxysporumn (Puhalla and Spieth, 1985). Isolates ofF. oxysporumn which share identical alleles at the loci governing heterokaryon incompatibility, commonly referred to as het or vic, are vegetatively compatible. Conventionally, this is determined via the ability of nitrate-nonutilizing (nit) auxotrophic mutants to complement one another for nitrate utilization (Leslie, 1993). Since F. oxysporum has no known sexual stage, it is not possible to determine genetically the number of loci involved in vegetative incompatibility. However, genetic studies performed on F. mnoniliforme (teleomorph- Gibberellaffuijikuroi) indicate that a minimum of 10 vic loci govern vegetative compatibility in this closely related species (Puhalla and Spieth, 1985). Unfortunately, the relationship among isolates comprising different VCGs is impossible to discern based solely on this technique because a single change of alleles at any of the vic loci will generate a new VCG.

Vegetative compatibility group diversity varies among the known formae speciales of F. oxysporumn. For example, among 115 isolates ofF. oxysporum f sp. lycopersici, Elias and Schneider (1991) identified one major and two minor VCGs, as well as a number of single-member VCGs. In contrast, a worldwide collection of 52 isolates ofF. oxysporum f sp. vasinfectum isolates comprised 10 VCGs and isolates belonging to distinct races of the pathogen were never in the same VCG (Fernandez et al., 1994). Currently, at least 17 VCGs have been described for F. oxysporum f sp. cubense (Ploetz, 1994). The majority of VCGs appear to have limited geographical distribution. However, the majority of isolates belong to a few major VCGs that are geographically widespread and contain more than one race of the pathogen (Ploetz, 1994).






10

Until the advent of tissue culture, the rhizome of the banana plant was used to

propagate banana. It is likely that the worldwide movement of planting stock (rhizomes and associated soil particles) from region to region is responsible for much of the geographical distribution of the pathogen outside its region of origin. In terms of evolution, the diversity of the pathogen in areas outside of its center of origin will depend on the number of years the pathogen has been present in the new location and whether single or multiple clonal lineages were introduced. For example, one might expect the populations ofF. oxysporum f. sp. cubense in East Africa to be more genetically variable than populations in the USA because the crop was introduced into East Africa thousands of years prior to the introductions to Florida.

Many earlier genetic characterizations of various formae speciales ofF. oxysporum focused on a small number of isolates with limited geographical distribution. These studies gave an incomplete picture of genetic diversity when more extensive testing using a larger number of isolates with a large geographical distribution was performed. For example, when 170 isolates of the pathogen F. oxysporum f. sp. dianthi were collected from 42 cultivars at 40 different sites in Israel, all of the isolates belonged to the same VCG (Katan et al., 1989). However, when isolates of the same pathogen were sampled from a number of geographical locations, including Italy, Spain, The Netherlands, the United States and Germany, eight VCGs were identified (Aloi and Baayen, 1993). The latter researchers found not only single races within single VCGs but also multiple races within single VCGs (Aloi and Baayen, 1993).

Isolates ofF. oxysporun f sp. cubense also have been grouped based on

electrophoretic karyotype, random amplified polymorphic DNA (RAPD) analysis and the






II
production of volatiles in culture. Boehm et al. (1994) proposed two groupings of isolates ofF oxysporuwn f sp. cubense based on similarities in chromosome number and genome size. In general, these groupings agreed with RAPD data (Bentley et al., 1995, Sorensen et al., 1993) and groups determined when the formation of aldehydes by different isolates were examined (Brandes, 1919; Moore et al., 1991, 1993; and Stover, 1959).

Clonality and Its Consequences in the Study of Population Genetics of Asexually Reproducing Organisms

Fusarium oxysporum is considered to be an asexual fungus based on the fact that, although many researchers have searched for a sexual stage in both the laboratory and field, none has been reported. For an asexually reproducing organism like F oxysporunl, it is generally assumed that isolates within a VCG are genetically similar and represent clonal populations. Many asexually reproducing populations consist of a number of clonal lineages with wide geographical distribution. Kohli et al. (1992), using a DNA fingerprinting probe and mycelial compatibility, detected 39 clones of Sclerotinia sclerotioruni on canola from three provinces in Canada. The most widely distributed clone was found in all three provinces. Overall, they found that clones of S. sclerotiorum were distributed over large geographic areas and field populations were composed of more than one clone.

Patterns of genetic variability in natural populations of fungi may reflect strict clonality, panmixia or a mixture of clonality and mating (Anderson and Kohn, 1995). In asexually reproducing organisms, there is presumably no contribution to genotypic variability through meiotic recombination or introgression. In general, genotypic variability in asexual fungi may be generated through mutation, random genetic drift,





12

parasexuality, somatic recombination and transposable elements. Although mutation rates are generally low per locus, mutations are persistent and over time they generate much variability, summed over all loci (McDonald, et al., 1989). Tibrayrenc et al. (1990) established criteria that characterize asexually reproducing organisms. They include over-represented, ubiquitous multilocus genotypes in a population; the absence of recombinant genotypes; correlation between two independent sets of genetic markers; and the presence of linkage disequilibrium. Genetic analyses of a large collection of isolates using neutral, single-copy, polymorphic loci ideally can test for the absence of recombination.

Recently there has been an increased interest in genetically characterizing field populations of plant pathogenic fungi. Many field studies have been conducted which have characterized the genetic diversity of fungal populations within a defined geographical location. The concept of a fungal population has been described by McDonald and McDermott (1993) as a group of individuals sharing a common gene pool and present in a limited geographical area. Population genetic analysis involves the identification of genetic markers that are unambiguous and informative [polymorphic] (McDonald and McDermott, 1993). These types of markers have been used to characterize populations of a number of fungal species, including Cryphonectria parasitica, Fusarium oxysporum, Sclerotinia sclerotiorum, Phytophthora infestans, Mycosphaerella graminicola, Cochhiobolus carbonum, Phytophthora megaspernia f sp. glycinea, Magnaporthe grisea and Stagonospora nodorum (Appel and Gordon, 1994; Goodwin et al., 1992; Koh et al., 1994; Kohli et al., 1992; Kohli et al.,1995; Leonard and Leath, 1990; Levy et al., 1991; Levy et al., 1993; McDonald and Martinez, 1990;






13
McDonald et al., 1994; McDonald et al., 1995; Sujkowski, et al., 1994; Whisson et al., 1992).

Population genetic studies of Fusarium oxysporuni f sp. melonis have addressed genetic structure in both agricultural and native soils and also have examined the genetic relationships among pathogenic and nonpathogenic isolates by characterizing their VCGs and mitochondrial haplotypes (Appel and Gordon, 1994, Gordon and Okamoto, 1992; Gordon et al., 1992,). Based on parsimony analysis of the mtDNA data, nonpathogenic isolates from Maryland were more closely related to F. oxysporum f sp. melonis than to the California nonpathogens (Appel and Gordon, 1994). Based on mtDNA, both pathogen VCGs were more closely related to a nonpathogen VCG than to each other (Appel and Gordon, 1994). Essentially all diversity in mtDNA was found among the nonpathogenic strains of the species, whereas pathogenic strains were represented by only a single mtDNA haplotype at any one location (Gordon and Okamoto, 1992; Jacobson and Gordon, 1990b) indicating that pathogenic strains are less genetically diverse than non pathogenic strains. In contrast, a study of the genetic relatedness among 120 strains ofF. oxysporum in native and cultivated soils consisted of 23 different mtDNA haplotypes, of which the five most common haplotypes accounted for 78% of the isolates. Isolates representing these haplotypes were found in both native and agricultural soils (Gordon et al., 1992). Seventy two percent of the isolates found in cultivated soil were associated with the same mtDNA haplotype as one or more of the isolates in the native soil (Gordon et al., 1992).

Host populations can act as powerful selective forces on pathogen populations and vice versa (McDonald et al., 1989). For example, the rice blast fungus, Magnaporthe






14

grisea, can be divided into a large number of pathotypes. Pathotypes are analogous to races in that specific pathotypes of a pathogen are virulent on certain genotypes of the host. Using a repetitive fingerprinting probe that serves as a genealogical index among rice blast isolates, Levy et al. (1993) examined the population structure and associated virulence properties among isolates in a resistance-breeding nursery in Colombia. Sampling the nursery population enabled the recovery of 39 pathotypes among 151 isolates. Using the fingerprinting probe called MGR 586, they could distinguish six distinct clonal lineages, each with a generally non-overlapping cultivar range and an associated set of several related pathotypes. Eighty-five percent of the pathotypes were lineage specific. Their results support the view that cultivar specificity and pathotype evolution have developed within constraints imposed by the genetic background of each lineage (Levy et al., 1993).

The overall objectives of this dissertation research were to determine the

phylogenetic relationships among isolates ofF. oxysporum f. sp. cubense and to determine the level of genetic variability and its distribution in field populations of the pathogen on banana. An RFLP study of a world collection of isolates was undertaken to assess the phylogenetic relationships among isolates. A study which characterized isolates by their mitochondrial haplotype and distinguished between different clones by utilizing a DNA fingerprinting probe was undertaken to assess the level and distribution of genetic variability in two Honduran field populations. Both of these studies should contribute to the overall understanding of the genetic structure and relationships among isolates ofF. oxysporum f, sp. cubense.













CHAPTER 2
FUSARIUMOXYPSORUMF. SP. CUBENSE CONSISTS OF A SMALL NUMBER OF
DIVERGENT AND GLOBALLY DISTRIBUTED CLONAL LINEAGES Introduction

Fusarium wilt of banana, commonly referred to as Panama disease, is caused by Fusarium oxysporumn Schlechtend.: Fr. f. sp. cubense (E.F. Sm.) W.C. Synder & H. N.

Hans. Isolates ofF. oxysporum f. sp. cubense previously have been characterized by morphology or biochemical and genetic markers. On morphological bases, F. oxysporumn was included in section Elegans by Wollenweber and Reinking (1935). No teleomorph has been described for the species.

Vegetative compatibility groups (VCGs) have been used to categorize isolates of F. oxysporum, including Fusarium oxysporum f. sp. cubense. Based on data from other fungi, isolates that share identical alleles at the loci governing heterokaryon incompatibility, commonly referred to as het or vie, are vegetatively compatible (Leslie, 1993). Conventionally, this is determined by the ability of nitrate-nonutilizing (nit) auxotrophic mutants to complement one another for nitrate utilization (Correll et al., 1987). Currently, at least 17 VCGs have been described for this forma specialis (Ploetz and Correll, 1988; Ploetz, 1994; Ploetz et al., 1997). The majority of isolates belong to two major VCGs that have pantropical distribution and contain more than one race of the pathogen. Minor VCGs were found to have a more limited geographical distribution (Ploetz and Correll, 1988; Ploetz, 1990). For an asexually reproducing organism like F.

15





16

oxysporurn, it is generally assumed that isolates within a VCG are genetically similar and represent clonal populations (Appel and Gordon, 1994; Jacobson and Gordon, 1991; Tantaoui et al., 1996).

In addition to characterizing isolates by VCG, isolates ofF. oxysporum f. sp. cubense have been grouped based on electrophoretic karyotype, randomly amplified polymorphic DNA (RAPD) analysis and the ability to produce volatile organic compounds in culture. Boehm et al. (1994) proposed two groupings of isolates ofF. oxysporum f sp. cubense based on similarities in chromosome number and genome size. Group I comprised isolates in VCGs 0124, 0125, 0124/0125, 01210 and 01214. Group II comprised isolates in VCGs 0120, 0121, 0122, 0123, 0129 and 01213. In general, these groupings agreed with RAPD data (Bentley et al.,1995; Sorensen et al., 1993). When the presence or absence of RAPD bands were treated as binary data and subjected to phenetic analysis based on the unweighted pair group method with arithmetic mean (UPGMA), isolates in VCGs 0120, 0121, 0122, 0126, 01210, 01211 and 01212 formed one group and isolates in VCGs 0123, 0124, 0124/0125 and 0125 formed a second group.

Similar major groups of isolates were evident when differentiation was based on the formation of aldehydes in culture. Brandes (1919) noted that certain isolates of the pathogen produced these odorous compounds when grown on steamed rice, whereas others did not; the latter isolates were classified as variety inodoratum. Stover (1959) examined a larger set of isolates from tropical America and the Caribbean and noted that production of odorous compounds was a consistent and repeatable trait. Those isolates which produced the aldehydes were referred to as cultivar Odoratum whereas, those which did not were referred to as cultivar Inodoratum. More recently, Moore et al. (1991;





17

1993) analyzed the production of these compounds with high pressure liquid chromatography. Isolates in VCGs 0120, 0129 and 01211 produced characteristic volatile profiles, whereas isolates in VCGs 0123, 0124, 0124/0125, 0125 and 0128 did not.

Restriction fragment length polymorphisms (RFLPs) also have been employed to determine the genetic relationships among isolates ofF. oxysporum. These markers are ideally suited to genetic diversity studies because of the following characteristics: i) most are selectively neutral; ii) polymorphisms tend to be more numerous compared to other types of markers, such as isozymes; iii) they are reproducible; and iv) those identifying random single-copy loci avoid problems that are associated with linkage of markers.

In this study, probes from F. oxysporum f. sp. lycopersici that correspond to

single-copy, anonymous loci (Elias et al., 1993) were used to identify polymorphic alleles in F. oxysporum f sp. cubense. The objectives of the study were to determine whether isolates within VCGs ofF. oxysporum f. sp. cubense were clonally derived and if clonal lineages correlated with previously determined VCGs. Phylogenetic relationships among the various VCGs ofF. oxysporum f sp. cubense and F. oxysporum from other hosts were also characterized to determine if host specificity is monophyletic.

Materials And Methods

Fungal Isolates

One hundred and sixty-five isolates representing worldwide distribution and 17

VCGs ofF. oxysporum f. sp. cubense from the collection located at the Tropical Research and Education Center in Homestead, Florida, were selected for analysis (Table 2-1). Additionally, three isolates ofF. oxysporum f. sp. lycopersici and a single isolate of





18

Table 2-1: List of isolates, their vegetative compatibility groups (VCG), clonal lineages, the cultivar, and geographical regions from which they were collected


VCG' Isolate CultivarV Origin and collectors Lineage

0120 IC2 Cavendish Icod de los Vinos, Canary Islands,d II
22425 Cavendish Wamuran, Queensland, Australia,g II
ORT2 Cavendish La Orotava, Canary Islands,d II
0-1220 Mons Queensland, Australia,c II
GAL2 Cavendish Las Galletas, Canary Islands,d II
C2 Cavendish Canary Islands,f II
ADJ2 Cavendish Adeje, Canary Islands,d II
Cl Cavendish Canary Islands, f II
22424 Lady Finger Moorina, Queensland, Australia,g II
0-1222 Mons Queensland, Australia,c II
0-1219 Mons Queensland, Australia,c II
A2 Mons Mari Australia,f II
ADJ1 Cavendish Adeje, Canary Islands,d II
STGM1 Gros Michel Costa Rica,i II.
3 S1 Highgate Honduras,i IP
PAJ1 Cavendish Pajalillos, Canary Islands,d II
ORT1 Cavendish La Orotava, Canary Islands,d II
GALl Cavendish Las Galletas, Canary Islands,d II
BUEl Cavendish Buenavista, Canary Islands,d II
NW Williams Natal, South Africa,f II
NH Williams Natal, South Africa,f II
NB Cavendish Natal, South Africa,f II
F9127 Grand Naine South Africa,g II
15638 x Malaysia,a II
FCJ7 Lacatan Jamaica,q II
Pacovan Pacovan Bahia, Brazil,n II
MGSA1 SH3142 South Africa,s II
SA6 Cavendish Transvaal, South Africa,b II
SA4 SH3362 Natal, South Africa,b Ir
SA3 Williams Tansvaal, South Africa,b II

0121 GM Gros Michel Taiwan,h
9130 Cavendish Taiwan,g III
0-1124 x Taiwan,c IIIx
H1 Cavendish Taiwan,e III
ML Cavendish Taiwan,h III
TBR Cavendish Taiwan,h III

0122 Ph3 Cavendish Philippines,l VI
Ph6 Cavendish Philippines,l VI
P79 Cavendish Philippines,h VI
LAP Cavendish Philippines,h VI
SABA Saba Philippines,h VI
PW3 Cavendish Philippines, m VI






Table 2- 1 --continued 19

VCG" Isolate Cultivar' Origin and collector Lineage

PW6 Cavendish Philippinesm VI
PW7 Cavendish Philippinesm VI

0123 DAVAO Silk Philippinesh VII
TI Gros Michel Taiwanf VII
PhL2 Latundan Philippinesi VII
Ph12 Latundan Philippinesl VII
9129 Latundan Taiwang VII
JLTH4 Klue namwa Smoeng Hwy 1269, Thailandv x
JLTH5 Klue namwa Smoeng Hwy 1269, Thailandv

0124 A36 x Brazilk
GMB Gros Michel Braziln
Maca Maca Braziln
STPAI Pisang Awak Burundii
STD2 Highgate Hondurasi
BLUG Bluggoe Hondurash
Sx Tetraploid 1242 Bodles, Jamaica, i
FCJ2 Bluggoe Jamaicaq
FCJ3 x Jamaicaq
FCJ8 x Jamaicaq
FCJ9 Tetraploid 1242 Jamaicaq
STJ2 Grande Naine Jamaicaj
MW43 Harare Chitipa, Karonga, Malawib
MW45 Harare Chitipa, Karonga, Malawib
NM47 Harare Chesenga, Malawib Ix
MW50 Harare ChitipaKaronga, Malawib I
MW52 Sukali Karonga South, Malawib I
MW58 Harare Karonga, Malawib I
NM64 Harare Kaporo North, Malawib I
MW67 Kholobowa Thyolo, Blantyre, Malawib I
MW69 Kholobowa Thyolo, Blantyre, Malawib I
MW71 Kholobowa Mulanje, Blantyre, Malawib I
MW78 Harare Vinthukutu, Karonga, Malawib I
STN2 Bluggoe Corinto, Nicaraguai I
STN5 Bluggoe Corinto, Nicaraguai I
STN6 Bluggoe Corinto, Nicaraguaj I
STN7 Bluggoe Corinto, Nicaraguaj I
STPA2 Pisang Awak Tanzaniai I
B 1 Burro(Bluggoe) FloridaUSAb 1
0124 B2-1 Burro FloridaUSAb I
JCB I Burro FloridaUSAb I
YLTH2 Klue namwa Smoeng Hwy 1269, Thailandv I
JLTH7 Klue namwa Smoeng Hwy 1269, Thailandv I
JLTH 15 Klue namwa Chai Yo Hwy, Thailandv I






Table 2-1 --continued 20

VCGu Isolate Cultivarv Origin and collector' Lineagze

0124/0125
MW9 Zambia Kaporo, Malawi,r I
MWI 1 Harare Kaporo, Malawi,r I
M1W39 Harare Chitipa, Karonga, Malawi,b I
MW53 Sukali Karonga, Malawi,b i
MW56 Zambia Karonga, Malawi,b I
MW 60 Zambia Karonga, Malawi,b I
M W6 I Harare Vinthukutu, Karonga, Malawi,b I
NMW63 Harare Karonga, South, Malawi,bI
MIW66 Kholobowa Thyolo, Blantyre, Malawi,bI
MW70 Kholobowa Thyolo, Blantyre, Malawi,bI
MW 86 Mbufu Chitipa, Karonga, Malawi,bI
JLTIi Klue namwa Ban Nok, Thailand,vI
JLTH16 Klue namwa Ban Nok, Thailand,v I
JLTH 17 Klue namwa Ban Nok, Thailand,v JX
JLTH 18 Klue namwa Ban Nok, Thailand,v JX
JLTH1 9 Klue namwa Ban Nok, Thailand,v I

0125 A4 Lady Finger Australia,f 1
8606 Lady Finger Currumbin, Queensland,g I
8611 Lady Finger Currumbin, Queensland,g 1
22468 Lady Finger Tomewin, Queensland,g I
22479 Ducasse Bowen, Queensland, Australia,g Py
22600 Lady Finger Murwillumbah, New S. Wales, Australia,g 1
22417 Lady Finger Rocksberg, Queensland, Australia,g 1
22541 Lady Finger Murwillumbah, New S. Wales, Australia,g I
0-1223 Mons Queensland, Australia,c I
1 5X Williams Bodles, Jamaicaji I
STPA3 Pisang Awak Ugandaji I
JLTH20 Klue namwa Ban Nok, Thailand,v 1

0126 51 Highgate Honduras,i 1I
STA2 Highgate H-ondurasji 1I
STM3 Maqueno Hondurasji I1
STB2 Highgate Hondurasji II

0128 22993 Bluggoe South Johnstone, Queensland,g 1
22994 Bluggoe South Johnstone, Queensland, Australia,g I
A47 Bluggoe Comores Islands~j Ix

0129 N5331 Cavendish Yandina, Queensland, Australia,g 11
0-122 1 Mons Queensland, Australia,c 11
N5443 Cavendish Doonan, Queensland, Australia,g II
8627 Cavendish North Arm, Queensland, Australia,g 11
22402 Cavendish Wamuran, Queensland, Australia,g 1I
8604 Cavendish North Arm, Queensland, Australia,g I1






Table 2-1--continued 21

VCGu Isolate Cultivarv Origin and collector"' Lineage

01210 A2-1 Apple Florida, USA,b IV
A4-1 Apple Florida, USA,b IV
CSB Apple Florida, USA,b IV
JC14 Apple Florida, USA,b IV
A15 Apple Florida, USA,b IV,
A3-I Apple Florida, USA,b IV
JC8 Apple Florida, USA,b IV
F2 Apple Florida, USA,b IV
F3 Apple Florida, USA,b IV
JC1 Apple Florida, USA,b IV
GG1 Apple Florida, USA,b IV

01211 13721 x xa IXx
SH3142 SH3142 Queensland, Australia,g IX

01212 STNP1 Ney Poovan Pemba Island, Zanzibar, Tanzania,i VIII
STNP2 Ney Poovan Tenguero Station, Tanzania,i VIII
STNP4 Ney Poovan Bukava Station, Tanzania,i VIII

01213 1-1 Cavendish Taiwan,u III
2-1 Cavendish Taiwan,u III
6-2 Cavendish Taiwan,u III
5-1-1 Cavendish Taiwan,u III
4-2-1 Cavendish Taiwan,u III
4-1-1 Cavendish Taiwan,u III
2-2 Cavendish Taiwan,u III
ES2-1 Cavendish Taiwan,u

01214 MW2 Harare Misuku, Karonga, Malawi,r V
MW40 Harare Misuku Hills, Karonga, Malawi,b V
MW41 Mbufu Misuku Hills, Karonga, Malawi,b V
MW42 Harare Misuku Hills, Karonga, Malawi,b V
MW44 Harare Misuku Hills, Karonga, Malawi,b V
MW46 Harare Misuku Hills, Karonga, Malawi,b V
MW48 Harare Misuku Hills, Karonga, Malawi,b V
MW51 Harare Misuku Hills, Karonga, Malawi,b V
MW89 Harare Misuku Hills, Karonga, Malawi,b V


01215 CR1 Gros Michel Isolona,Costa Rica,b II
CR2 Gros Michel Hamburgo,Rio Reventazon,Costa Rica,b II
CR4 Gros Michel Hamburgo,Rio Reventazon,Costa Rica,b II
CR5 Gros Michel Hamburgo,Rio Reventazon, Costa Rica,b II






Table 2-1--continued 22

VCGU Isolate Cultivarv Origin and collectors Lineage

0120/01215
INDO20 Musa spp. Jatesari, East Java,Indonesia,g III
INDO15z Musa spp. Jatesari, East Java,Indonesia,g III
INDO 18Z Musa spp. Jatesari, East Java,Indonesia,g III

Fusarium oxysporumn f sp. lycopersici
SC548 Homestead Bradenton, Florida
SC626 Oristano Italy
SC761 Sunny Bradenton, Florida

Fusarium oxysporum f sp. niveum
0082 CS85-4 Florida,w


uVegetative compatibility groups (VCGs) were assigned using nitrogen metabolism (nit) mutants according to the protocols of Cove (1976) as modified by Puhalla (1985).

vCultivars are inter- and intraspecific diploid or triploid hybrids ofMusa acuminata (AA) and Musa balbisiana (BB). The ploidy levels and constitutions of the cultivars are as follows: AAA = Gros Michel, Highgate, Mons (Mons mari), Cavendish, Dwarf Cavendish, Grande Naine, Williams, Lacatan; AA = SH3142 and SH3362 (synthetic clones); AAB = Lady Finger, Pacovan, Prata, Silk, Latundan, Maquefio; ABB = Saba, Bluggoe, Harare, Kholobowa, Pisang awak, Klue namwa, Ducasse, Mbufu, Burro, Zambia; AB = Ney poovan, Sukali; AAAA = Tetraploid 1242

missing data made it impossible to determine coefficient of similarity but based on all other data it is presumed to be in the lineage indicated.

Y= unique isolate that had no lineage affinity based on defined criteria; the lineage assigned is based on coefficients of similarity to the most closely related isolates. = isolates analyzed with this designation do not correspond to any isolates currently held in the Homestead collection.

WCollector or original source: a, American Type Culture Collection; b, R.C. Ploetz, Homestead, Florida; c, Paul E. Nelson, Fusarium Research Center, University Park, Pennsylvania; d, J.H. Hemrnandez, Tenerife, Canary Islands; e, S.C. Hwang, Taiwan Banana Research Institute, Pingtung; f, B. Manicom, Nelspruit, South Africa; g, K. Pegg, Brisbane, Australia; h, S. Nash Smith, Alameda, California; i, R.H. Stover, La Lima, Honduras; j, IFRA, Montpellier, France (via R.C. Stover); 1, A.M. Pedrosa, Philippines; m, N.I. Roperos, Philippines; n, Z. J. M. Cordeiro, EMBRAPA, Cruz das Almas, Brazil (via E.D. Loudres); q, J. Ferguson-Conie, Banana Board, Kingston, Jamaica; r, B. Braunworth, Oregon State University, Corvallis; u, Tsai-young Chuang, National Taiwan University, Taipei; v, J. Leslie, Kansas State University, Manhattan; w, F. Martin, USDA, ARS, Salinas, CA.





23

F. oxysporurn f sp. niveum were analyzed to compare results with genotypes representing different formae speciales.

DNA Isolation

All cultures were derived from single microconidia and were stored at 4C on

strips of Whatman filter paper (Correll et al., 1987). Paper strips were plated onto potato dextrose agar (PDA) (Difco, Detroit, MI). After approximately 7 days of growth at 270C, a 5-mm3 block was excised from the margin of colonies on PDA and transferred to 4-L Erlenmeyer flasks containing at least 100 ml of potato dextrose broth (Difco 24 g/L). After 7 to 10 days of growth in still culture, the contents of the flask were filtered through sterilized cheese cloth to collect mycelium. The mycelium was placed into 13-ml plastic tubes, frozen at -80'C and lyophilized for at least 12 hours.

Buffer for DNA extraction was comprised of a 1:1.0.4 volume of the following solutions: Buffer A (0.3 M Sorbitol, 0.1 M Tris, 20 mM EDTA at pH 7.5), Buffer B (0.2 M Tris at pH 7.5, 50 mM EDTA and 0.2 mM CTAB) and 5% Sarkosyl. Ten ml of the extraction buffer were mixed with approximately 0.5 g of ground mycelial powder, and the tubes were placed in a 65C water bath for 30 minutes. The contents of the tubes were then shaken and 1 ml of solution was transferred to each of 10, sterile, 1 .5-ml microcentrifuge tubes. Five hundred microliters of chloroform:octanol (24:1) solution were added to each tube. The solution was mixed using a vortex shaker for approximately

2 minutes before centrifugation for 10 minutes at 12,000 g in a microcentrifuge at room temperature. Unless noted, all further centrifuge steps were done in a microcentrifuge at room temperature. The supernatant was transferred to sterile 1.5-ml tubes and treated with 5 al of 20 mg RNase A (Sigma Chemical Co., St. Louis, MO)/ml of solution for 30






24

minutes at 37C. Following RNase treatment, 5 pl of 20 mg proteinase K (Sigma)/ml of solution were added to the tubes, and the tubes were incubated for 20 minutes at 37C. Approximately one volume isopropanol was added, and the tubes were centrifuged for 15 minutes at 12,000 g. The isopropanol then was discarded, and 100 1 of ice-cold 70% ethanol were added before centrifugation for five minutes. The ethanol was discarded, and the DNA sample was air-dried for at least 30 minutes in a laminar flow hood. At least 100 pla of TE buffer (10 mM Tris, 1 mM EDTA at pH 7.4) were added to the tubes, and they were placed in a water bath at 65C until the DNA pellet was dissolved.

Samples which were difficult to bring into solution were subjected to a LiCI

treatment. Three hundred microliters of ice-cold 4M LiCI solution were added to each tube, and the tubes were placed on ice for 30 minutes before centrifugation at 12,000 g for 10 minutes at 4C. The supernatant was transferred to a sterile 1.5-ml tube containing 600 ll of isopropanol. This solution was mixed, and the tubes were kept at room temperature for 30 minutes. After centrifugation at 12,000 g for 10 minutes at 4C, the supernatant was discarded and 100 pl of ice-cold 70% ethanol were added to the tubes. After centrifugation at 12,000 g for 5 minutes, the ethanol was discarded and the DNA was airdried. TE buffer (100 l) was added to the tubes and they were placed in a water bath at 650C until the pellet was dissolved.

The concentration of DNA in the samples was estimated by running 3 p1 of each sample on an agarose gel along with DNA fragments (bacteriophage lambda DNA digested with PstI) of a known concentration and making visual comparisons of their relative fluorescence in the presence of ultraviolet (UV) light and 0.5 !ag of ethidium bromide/ml of solution.






25

Southern Blotting and Hybridization

Approximately 10 ig of DNA were digested with at least 10 units of either Dral, EcoRV or HaeIII restriction enzymes (Bethesda Research Laboratories, Gaithersburg, MD) and incubated for at least 3 hours. These restriction enzymes were chosen based on their insensitivity to DNA methylation and their ability to digest DNA consistently. Restriction fragments were separated by electrophoresis in 0.7% agarose for EcoRV- and DraI-digested DNA or 1.5% agarose for HaeIII-digested DNA in TBE buffer at pH 7.0. Gels were run at either 30 volts for approximately 16 hours or 40 volts for approximately 12 hours. Bacteriophage lambda DNA digested with either PstI or HindIII was included on each gel and used to calculate the molecular mass of restriction fragments obtained from F. oxysporuni f sp. cubense DNA. Ethidium bromide (10 mg/ml) was dissolved in the agarose gel at a concentration of 1 p11-100 ml, and the digested DNA was illuminated by UV irradiation and photographed. The DNA was transferred to Nytran membranes (Schleicher & Schuell, Inc., Keene, NH) using the capillary transfer method (Sambrook et al., 1989). The DNA transfer proceeded for at least 12 hours, and the DNA was immobilized by UV cross linking (UV 254-nm cross linker, model FB UVXL 1000, Fischer Scientific, Pittsburgh, PA).

To reduce the incidence of repeatedly scoring similar regions of the genome or hypervariable regions, clones containing single-copy DNA sequences, obtained from Talma Katan (The Volcani Center, Bet Dagan, Israel), were utilized (Elias, et al., 1993). Clones were considered to be single-copy based on the criteria defined by Elias et al. (1993) as hybridizing to only a single DNA fragment in any of the isolates tested using at least one of three restriction enzymes (Figure 2-1).






26
1 4 5 6 71 0 11 12 13 14 15 16 17 18 19





W eight ::.:.::::::::::::::.::.:::

Kb .....


N .. ::.:.:. :... K ii..:.:....:, -,::.F.f ++:...:::::: :>
65




1 XA 201 001 0
2 ADJ 0.2 .010.00 ...
........... .

........... ...A .. .. .. .. ..C
.. .....



Isolate VCG Polymorphic State
IGAL2 0120 01000
2 ADJ2 0120 01000
3 01222 0120 01000
4 F9127 0120 01000
5 Pacovan 0120 01000
6 GM 0121 00100
7 SABA 0122 01000
8 FCJ9 0124 01000
9 MW60 0124/25 01000
10 CVA 0124/25 01000
11 S? 0125 01000
12 JLTH1 0124/25 01000
13 JLTH21 0125 01000
14 22994 0128 01000
15 A47 0128 01000
16 MW2 01214 10010
17 1-2 01213 10000
18 22507 0129 01000
19 SI 0126 01000

Figure 2-1: Single-copy sequences of selected Fusarium oxysporum f. sp. cubense isolates. DNA of isolates was digested with Eco RV restriction enzyme and probed with clone 187. Polymorphic states indicate the presence (1) or absence (0), respectively of the
6.5, 5.1, 4.6, 4.2, or 3.4 Kb bands








27
DNA for each clone was labeled using random hexamer primers to incorporate fluorescein- 12-dUTP following the procedures provided by the manufacturer (Dupont NEN "Renaissance," E. I. du Pont de Nemours & Co. Inc., Boston, MA). DNA labeling, hybridization and detection followed the procedures provided by the manufacturer (Dupont, Boston, MA). Prehybridization, hybridization and washing were performed at 65C using a Hybaid hybridization oven (Dot Scientific Inc., Flint, MI). Membranes were placed between acetate sheets and exposed to X-ray film for at least 5 hours.

Mitochondrial DNA ofF. oxysporum f sp. cubense isolate 3S1 (VCG 0120) was isolated following the procedures of Kistler and Leong (1986) and labeled as described above. The mitochondrial DNA profiles of a subset of isolates were obtained by digesting approximately 10 gg of total DNA with at least 10 units of the restriction enzyme HaeIII and probing with the mitochondrial DNA of isolate 3S 1. Restriction fragments were separated by electrophoresis in 1.5% agarose in TBE buffer at pH 7.0. Gels were run at 30 volts for approximately 22 hours. Southern blotting and hybridization followed the procedures described above.

Data Analyses

Initially, a subset of 38 geographically widespread isolates ofF. oxysporum f. sp. cubense that represented all 17 VCGs was used to determine if a particular probe-enzyme combination was polymorphic. Only polymorphic loci were considered informative for phylogeny determinations. If polymorphisms were detected in the subset, then all 165 isolates were analyzed for that probe-enzyme combination. If all isolates within the subset were monomorphic, it was assumed that the entire collection was monomorphic for that







28

probe-enzyme combination. The different restriction size fragments generated for each combination of probe and enzyme were considered to be alleles at a single RFLP locus, and their presence or absence was scored for each isolate. RFLP patterns for each combination of probe and enzyme were combined to assign an RFLP haplotype to each isolate.

The data were analyzed by a cladistic approach based on parsimony analysis using the computer program PAUP Version 3.1.1 (Swofford, 1993), and by a phenetic approach using distance matrix methods (UPGMA clustering, Sneath and Sokal, 1973) and the neighbor-joining algorithm of Phylip Version 3.5c (J. Felsenstein, University of Washington). For parsimony analysis, phylogenies were derived by using the heuristic search option, and the degree of support was evaluated using 500 bootstrap replicates. In addition, coefficients of similarity based on simple matching were calculated for those isolates in which data were available for every RFLP loci scored, based on the formula described by Sneath and Sokal (1973). Isolates were arbitrarily considered to be within the same clonal lineage based on coefficients of similarity ranging from 0.94 to 1.00. Since many of the isolates had identical multilocus haplotypes, only a single isolate was used to represent each haplotype in data analyses.

To determine whether this collection of isolates provided evidence for clonal

reproduction, the gametic disequilibrium coefficient (D) was calculated among pairs of alleles at different loci by methods described by Weir (1990). Clone-corrected allele frequencies, using only a single representative for each haplotype, were employed for the calculations. Also, to avoid the potential problem of repeatedly scoring similar regions,







29

only data from a single restriction enzyme digestion were used for each probe in the analyses. Nine hundred and eighty-eight pairwise comparisons were performed to test for disequilibrium between multiple alleles at nine loci. A test for the significance of the disequilibrium coefficient between each pair of alleles at two loci was formulated with the chi-square statistic






where n was the number of individuals in the sample and D!,Vwas the maximum likelihood estimator for the coefficient of disequilibrium between alleles u and v.

The observed allele frequencies for the loci were Pt and P, respectively

(McDonald et al., 1994; Weir, 1990). The chi-square statistic had one degree of freedom and the pairs of loci that showed significant departure from random expectations (P <0.05) were considered to be in disequilibrium. A test for significance of the disequilibrium coefficient across all alleles for each pair of loci was formulated with the chi-square test statistic as described by McDonald et al. (1994).

The isolates tested in the study were assigned previously to a VCG, and one to several representative isolates within each of the examined VCGs were pathogenic on at least one and as many as eight different banana cultivars (Ploetz, personal communication).

Although one sample of DNA was isolated from each isolate, many isolates were tested at least once and as many as three times for each probe enzyme combination examined. Additionally, a subset of 24 F. oxysporumr f sp. cubense and 2 F. oxysporum f







30

sp. lycopersici isolates tested here were separately analyzed using DNA sequencing data (O'Donnell, personal communication). Similar phylogenetic relationships were observed using this separate analysis, further validating the groups resolved in this analysis.

Results

In this study, 38 isolates ofF. oxysporum f. sp. cubense were screened for

polymorphisms using 19 probe-enzyme combinations. Only six of the 19 probe-enzyme combinations were monomorphic among the 38 selected isolates, indicating a high degree of genetic diversity among the isolates. The entire collection of 165 isolates was then scored for polymorphisms using the 13 probe-enzyme combinations that were found to be informative during the initial screening of isolates. A multilocus RFLP haplotype was assigned to each isolate based on the allelic data for all probe-enzyme combinations. Only 72 distinct multilocus haplotypes were detected among the 165 isolates, 50 of which were represented by a single isolate (Figure 2-2). The five most common haplotypes represented 45% of the isolates. The median number of alleles per locus was three, and if three alleles were present at each locus, theoretically 1.16 x 109 (3 possible haplotypes could exist for this collection ofF. oxysporum f. sp. cubense. However, the majority of single-isolate haplotypes found were the result of one to a few allelic differences from a more common haplotype.

To determine relatedness among isolates, the 72 RFLP haplotypes were subjected to phenetic and cladistic analyses. Both types of analyses produced trees with similar branching patterns. The 50% majority rule bootstrap consensus tree generated by PAUP 3.1.1. is presented in Figure 2-3. A dichotomy with strong bootstrap support (99%) was









31








30





25 2015
am
0 E z
10





5






LID LO N LO r I
Haplotype













Figure 2-2: Frequency distribution of RFLP haplotypes among the 165 isolates representing a world collection of Fusarium oxysporum f sp. cubense










32


18 EUN-CS85-4
SFOL SC626 F oxysporum
4) 1 L5(90) 14fOL SC548 l sp lycopersici
L FOL SC761 and
1 VCG 0122 Ph3 F oaysporun
VCG 0122 LAP f sp. nivean
8(96)] VCG 0122 PW7 FOCVI
VG 0122 P79
VCG 0122 SABA01214 MW VCG 01214 MW40 VCG 01214 MW41
4(69) VOCG 01214 MW42 f 01VCG 01214 MW48 FOCV
VCG 01214 MW44
GVCG 01214 MW46 V 0VCG2 01214 MW51
5 VCGVCG 01214 MW89 VCG 0120 C2.. S VCG 0120 F9127
VCG 0120 PAJ1 FOC II
VCG 0120 IC2
2 VCG 0120 Pacovan
04 VCG 0121 GM IFOCX
1-- VCG 0121 F9130 I
SVCG 0121 H1 5(58) VCG 01213 6-2
5(8 VCG 01213 1-2 FOC II
-I- VCG 01213 4-1-1 VCG 0121 0-1124 VCG 0120/01215 INDO20 2
3 VCG 0123 DAVAO FOC VI
14(74) VCG 0123 Phl2 F
1 VCG 0123 T1 VCG 0123 PhL2
15 VCG 0123 JLTH4 FOC X
1 VCG 0123 JLTH5
VCG 0120 MGSAI VCG 0L20 SA4 .. ... VCG 0120 SA3 FOC II
7(83) VCG 0126 STA2
_VCG 0126 STB2
5 VCG 0126 S1
VCG 0129 N5443
VCG 01210 A2-1 VCG 01210 GG 1 FOC IV
,, VCG 01210 A15
VCG 01211 13721 4 FOC IV
VCG 01210 JC1 FOC IV A VCG 01211 SH3142 r4FOCa VOG 01213 ES2-1 A VCG 0124 A36 ......... VCG 0124 GMB
-... VCG 0124 Maca ....... VCG 0124 STJ2 VCG 0124 MW43 VpG 0124 MW47
VCG 0124 JLTH15 VCG 0124 MW64
VCG 0124 STN5 FOC I
VCG 0124 STN7 VCG 0124/25 MW9 13(99) VCG 0124/25 MW11
VCG 0124/25 MW53 VCG 0124/25 MW63 VCG 0124/25 MW86 nVCG 0125 8606 VCG 0125 22479 VCG 0125 22541
VCG 10128 A47
4 83 VCG 01212 STNP1 FOC VIII
VCG 01212 STN?4
2 VCG 0124/25 JLTH16 VCG 0124/25 JLTH17 FOCI
VCG 0124/25 JLTH18 VCG 0124/25 JLTH19

Figure 2-3: Midpoint rooted 50% majority rule consensus tree representing 500 bootstrap
replicates; One isolate represents each of the 72 RFLP haplotypes of Fusarium oxysporum
f. sp. cubense, three isolates ofF oxysporum f sp. lycopersici and one isolate E
oxysporum f. sp. niveum. Branch lengths are indicated on each branch and bootstrap
values are in parenthesis. Tree length= 351; Consistency Index= 0.214; Homoplasy
Index = 0.786; Retention Index = 0.705.








33

observed on the midpoint-rooted tree among the 72 haplotypes representing the 165 F. oxysporuni f sp. cubense isolates and four isolates from other formae speciales. Isolates ofF. oxysporum f. sp. cubense belonging to VCGs 0124, 0124/0125, 0125, 0128, and 01212 formed one main phylogenetic branch, while isolates belonging to VCGs 0120, 0121, 0122, 0123, 0126, 0129, 01210, 01211, 01213, 01214, 0120/01215 and 01215, as well as the isolates ofF. oxysporum f sp. lycopersici and f. sp. niveum, were found on the second main branch.

Isolates comprising these two branches could be further divided into eight major clades which have moderate to strong bootstrap support (values greater than 70%). Within one of the branches, isolates in VCGs 0124, 0124/0125, 0125, and 0128 formed one clade. Isolates in VCG 01212 were genetically similar but distinct (83% bootstrap support) (see below) from these isolates. Within the other branch, six clades were identified. Isolates in VCGs 0122 (96% bootstrap support), 0123 (74% bootstrap support), 0126 (83% bootstrap support), 01214 (97% bootstrap support), and isolates of F. oxysporum f sp. lycopersici (100% bootstrap support) each formed their own clade. Isolates in VCGs 0121, 01213, and three isolates in 0120/01215 formed a clade of weak support (58%), and therefore could not be confidently differentiated from isolates in VCGs 0120, 0123 (two isolates) 0129, 01210, 01211, 0120/01215 and the single isolate ofF. oxysporum f sp. niveum.

All clades with strong bootstrap support are comprised of isolates that have identical or nearly identical multilocus haplotypes and are referred to here as clonal lineages. Additionally, many of the isolates which could not be resolved using bootstrap








34

analysis shared nearly identical multilocus haplotypes with other isolates. To further understand the genetic relationships among unresolved isolates, a simple matching coefficient of similarity for comparison was used (Table 2-2). Isolates with coefficients of similarity ranging from 0.94 to 1.00 were considered to be within a clonal lineage. This range reflects apparent natural groups (Figure 2-3); isolates within a lineage possess either small or no genetic differences. In cases where isolates did not fall within this range for all pairwise comparisons, isolates were included in the lineage if they shared values near or within the specified range with the majority of isolates comprising the lineage. These isolates are marked with an asterisk in Table 2- 1. Two isolates, ES2-1 and GM, could not be assigned to lineage based on these criteria.

A similarity matrix, which includes coefficients of similarity for selected isolates representing the major RFLP haplotypes and VCGs, is presented (Table 2-2). In general, VCGs aligned with single clonal lineages; exceptions to this were isolates in VCGs 0 123. Table 2-3 lists each lineage (with the prefix FCC), the number of isolates represented, the VCG of each lineage, and its geographical distribution. Seventy-four percent of the isolates studied were represented by Lineages FCC I, FCC 11 and FCC 111. Each of these three lineages contain more than one VCG, with lineages FCC I and FCC 11 having a pantropical distribution. Isolates in FCC IV through X each belong to a single VCG and represent one to a few geographical regions.

Coefficients of similarity between isolates of FCC I and FCC II ranged from

0.66 to 0.74. By comparison, two isolates of F. oxysporlim f. sp. lycopersici had coefficients of similarity ranging from 0.55 to 0.71 compared to isolates in FCC I and 0.64

















Table 2-2: Similarity matrix of simple matching coefficients based on restriction fragment length polymorphisms for selected isolates of Fusarium oxysporum f sp. cubense, F. oxysporum f sp. lycopersici and F. oxysporum f sp. niveum.




0




C0 () t C9 C C) t C) C C) C) to t, C, C 0 0 C) 0 C) C)

VCG 0120 9127 I
VCG 0120 IC2 0.98 1
VCG 0121 F9130 0. 82 0.85 1
VCG 0121 HI 0.86 0.88 096
VCG 0122 SABA 0.91 0 88 0 82 0 86 1
VCG 0123 DAVAO 0.78 0 75 0.74 0.75 0.75 1
VCG 0123 TI 08 0M78 074 078 0.78 093 I
VCG 0123 Phl2 0.76 0.76 0.73 0.76 0.74 0.94 0.92 1
VCG 0123 PhL2 0.81 0.79 0(75 0.79 0.79 0.94 0(99 0.93 1
VCG 0123 JLTH4 0.78 0075 0.72 0075 0.8 088 0.93 0 87 0.94 1
VCG 0124 MW43 0.73 071 0.62 0.66 0.75 074 079 073 0.8 0.81 1
VCGO0124MW64 0.72 0.69. 0,61 065 0.74 0.75 0.78 0.74 0.79 0.8 0.99 I
VCG0124/0125 MV11 0.74 0.72 0. 64 0.67 076 073 078 0.72 0.79 0.85 096 0.95 1
VCG 0125 8606 0.71 0.68 06 064 073 0.74 0.76 073 0.78 079 0.98 099 0.94 1
VCG 0126STA2 096 0.96 0.84 085 0.87 076 0.79 0.75 0.8 0.76 072 0.71 073 0.69 1
VCG 0126 SI 0.94 094 0.81 0.82 085 0.74 0.79 0.75 0.8 076 0.72 0.71 073 069 0.98 1
VCG 0129 N5443 096 0.99 0.86 087 0.87 0.74 076 0.75 0.78 0.74 069 0.68 0.71 0.67 0.95 0.93 1
VCG 01210 A2-1 0.89 092 091 0.92 87 07 .81 0.84 0.82 085 0.81 072 0.71 0.73 0.69 0.93 0.91 0.91 1
VCG01211 SH3142 0.91 0.93 0.85 0.86 0.86 0.75 0.75 0.74 0.76 0.78 0.71 0.69 0.69 0.68 0.89 0.87 0.92 0.89
VCG01212 STNP4 0.73 0.71 0.6 0.64 0.75 072 0.76 071 0.78 0.79 0.95 0.94 0.92 0.93 0.72 0.72 0.69 0.69 0.71 1
VCG01213 1-2 0.86 0.88 0.94 0.98 0.84 075 0.8 0.79 0.81 0.78 0.68 0.67 0.69 0.66 0.85 0.85 0.87 0,92 086 0.66 1
VCG01214MW2 0.76 0.74 0.75 074 0.79 0.75 08 0,74 0.79 0.8 0.68 0.67 0.69 0.66 078 0.75 0.75 0.8 0.76 068 0_74
VCG 01214 MW41 0.76 0.74 0.78 0.74 0.79 0.78 08 0.74 0.79 0.8 0.68 067 0.69 0.66 0.78 0.75 0.75 0.8 0.79 0.71 0.74 0.95
VCG 01214 MW51 079 0.76 075 074 0.79 0.75 0.8 074 0.79 0.8 0.68 0.67 0.69 0.66 08 0.78 0.78 0.8 0.79 0.68 0.74 0.98 0.95 1
VCG01214MW89 0,8 0.78 074 0.78 0.82 0.79 084 0.78 0.82 0.84 0.72 0.71 0.73 069 0.79 0.76 0.76 0.81 0.8 0.72 0.78 0.94 0.94 0.94 1
VCG012001215Indo20 0.87 0.89 0.95 0.99 085 0.76 0.79 078 0.8 0.76 067 0.66 0.68 0.65 0.86 0.84 0.88 0.93 087 065 0.99 0.75 0.75 0.75 079 I
SC626 lycopersici 071 073 0.76 075 075 069 0.72 06 0. 0.73 0 74 069 0.68 0.71 0.67 0.72 0.72 0.74 0.76 075 0.67 075 0.66 0.68 0.66 067 0.74 1
SC761 lycopersici 0.64 066 072 071 0 .71 06 062 6 0641 0.64 065 06 0.59 0.61 0.58 065 065 0.9 0. 0. 0. 0 67 0.69 068 062 071 0.64 0.66 0.61 0.62 0.69 088
CS85-4 niveum 074 0.74 075 076 0.76 08 085 79 0.8 .86 0.85 0.8 079 079 0.78 075 0.75 075 0.8 076 078 0.79 0.81 079 0.81 0 8 0 78 075 0.71 1






36

Table 2-3: Clonal lineages of Fusarium oxysporuni f sp. cubense isolates, their geographical distributions and corresponding vegetative compatibility groups (VCG).


No. of VCG
Lineage Isolatesz Geographic Distribution Represented

FOC I 65 Australia, Brazil, Burundi, 0124,0124/25,
Comores Islands Honduras 0125, and
Jamaica, Malawi, Nicaragua, 0128
Tanzania, Thailand, Uganda,
and the United States (Florida)

FOC II 43 Australia, Brazil, Canary Islands, 0120, 0126,
Costa Rica, Honduras, 0129, and
Jamaica, Malaysia, South Africa, 01215
and Taiwan

FOC III 15 Indonesia and Taiwan 0121, 01213,
and
0120/01215

FOC IV 11 United States (Florida) 01210
FOC V 9 Malawi 01214
FOC VI 8 Philippines 0122
FOC VII 5 Philippines, Taiwan 0123
FOC VIII 3 Tanzania 01212
FOC IX 2 Australia 01211
FOC X 2 Thailand 0123


ZIsolates ES2-1 in VCG 01213 and GM in VCG 0121 did not align with any lineage based on coefficient of similarity data






37

to 0.74 compared to isolates in FOC II. The single isolate ofF. oxysporum f sp. niveurn had coefficients of similarity ranging from 0.75 to 0.81 compared to isolates in FOC I and from 0.73 to 0.76 compared to isolates in FOC II. Thus, the two largest lineages ofF. oxysporuni f sp. cubense each are more genetically similar to the F. oxysporun f sp. niveun isolate than to each other. Similarly, they are roughly as genetically distinct from each other as either is to the F. oxysporun f. sp. lycopersici isolates.

All isolates in VCG 01214 comprise FOC V. Isolates in this lineage formed a clade exhibiting the longest branch length compared to all other clades representing F. oxysporuni f sp. cubense isolates. In fact, its length was comparable to the branch lengths of the clades containing the isolates from the other formae speciales (Figure 2-3). Additionally, isolates comprising this lineage had coefficients of similarity which did not align closely to any other lineage. For example, isolates in FOC V had coefficients of similarity ranging from 0.62 to 0.73 compared to isolates in FOC I, from 0.72 to 0.81 compared to isolates in FOC II and from 0.61 to 0.69 compared to the F. oxysporun f. sp. lycopersici isolates. This group had a large number of lineage-specific alleles at several RFLP loci which accounted for its relative lack of similarity to other F oxysporum f sp. cubense isolates. All of the isolates in VCG 01214 have a very limited geographical distribution. Isolates in this group also do not form chlamydospores (R.C. Ploetz, unpublished) as other F oxysporum isolates do, which is usually a defining trait for the species.

FOC VII consisted of the majority of isolates in VCG 0123. FOC X consisted of two additional isolates in VCG 0123. Similar to FOC V, isolates comprising these lineages had coefficients of similarity which did not align closely to any other lineage.






38

Surprisingly, the RFLP multilocus haplotypes of isolates within these lineages consisted of alleles similar to those in both FOC I and II, in addition to some lineage-specific alleles (Table 2-4). A representative FOC VII isolate (PhL2) shared 43% (6/14) of polymorphic alleles with isolates in FOC II and 36% (5/14) of these alleles with isolates in FOC I. By contrast, the isolates in FOC I displayed only 29% (4/14) allelic similarity with isolates in FOC II. With the exception of FOC V, isolates in the other lineages more closely aligned with isolates in either FOC I or II. Unlike isolates in other lineages, the multilocus haplotype of isolates in FOC VII and X appear to represent a combination of alleles from FOC I and II.

The mitochondrial DNA haplotypes of a subset of 55 isolates representing the

major VCGs were examined to determine their genetic relationships based on an additional independent genetic marker. Isolates within the same VCG shared identical mitochondrial DNA haplotypes. Isolates could be divided into three major groups based on visual assessment of similar though not identical, mitochondrial RFLP patterns (Figure 2-4). Isolates in VCGs 0123, 0124, 0124/0125, 0125, 0128, 01212, and 01214 formed one group. Isolates in VCGs 0120, 0121, 0122, 0129, 01213, 0120/01215 and 01215 formed a second group. Isolates in VCGs 0126 and 01210 formed a third group. In general, these groupings aligned with those based on RFLP analysis of single-copy loci, although more groups could be resolved using the latter method.

Measures of gametic disequilibrium were performed for alleles at nine loci (Table 2-5). Even though many of the individual comparisons were not significant, 34 of the 36 pairwise comparisons among alleles at different RFLP loci showed significant





39
Table 2-4: A comparison of allelic data for isolates in FOC I, FOC II and FOC VII.
Isolate (VCG)y Allelesz


A36 (VCG 0124) 11111111111111
IC2 (VCG 0120) 22122122221222
PhL2 (VCG 0123) 12132123331113


YIsolates A36 and IC2 represent the two largest RFLP haplotypes in FOC I and FOC II, respectively, and PhL2 represents an isolate in FOC VII.

zAlleles found in the most common haplotypes were given the number 1. Alleles in the second most common haplotype, if different from haplotype 1, were given the number 2. Alleles in isolate PhL2, if different than those in the two most common haplotypes were given the number 3.






Molecular Weight 40
Kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 23.1

9.9 g ,v
6.7


4.3 h30
...........,.


2.3 M022
2.0






IslteVG tNAGo
... .. ...... ..... .














1 ORTZ 0120 2
2 NW 0120 2
3 Ph3 0122 2I
4 DAVAO 0123 1
5 S? 0124 1
6 MW60 0124/25 1
7 8606 0125 1
8 STNP4 01212 1
9 22993 0128 1
10 DAVO 0123 1
11 STM3 0126 3
12 N5443 0129-- 2
13 TBR 0121 2
14 ES2-1 01213 2
15 JLTIH7 0124 1
16 SC548 F oxysporum other
f sp.
lycopersici


Figure 2-4: Mitochondrial DNA haplotypes of selected Fusarium oxysporum f sp. cubense isolates.






41

Table 2-5: Clone-corrected measurements of gametic disequilibrium among pairs of alleles in a world-wide collection of Fusarium oxysporum f sp. cubense.


Probe / Enzyme Combinations


120 162 177 187 204 228 260 261
Hae III Dra I EcoRV EcoRV Hae IlI EcoRV EcoRV EcoRV

7 13/35 4/10 4/18 5/30 5/15 7/30 5/20 22/40
Dra 1 77.7 (24) 25.1 (5) 56.8 (10) 174.4 87.4 (8) 50.0 (24) 183.1 86.6
(20) (12) (28)

120 8/14 17/21 9/42 11/21 9/42 13/28 14/56
Hae 11I 31.33(6) 68.3(12) 122.6 142.4 121.1 87.5(18) 184.3
(30) (12) (30) (42)

162 4/6 8/12 4/6 5/12 5/8 6/16
Dra I 4.91 (2)* 37.7 5.49 (2)* 128.7 (5) 29.3 23.3
(5) (3) (7)

177 5/18 4/9 7/18 2/12 9/24
EcoRV 33.0(10) 73.09 78.1 (10) 37.7 (6) 54.6(14)
(4)

187 6/18 17/36 6/24 6/48
EcoRV 40.5 126.2 63.54 65.2 (35)
(10) (25) (15)

204 7/18 6/12 11/24
Hae I11 51.3(10) 48.2(6) 85.8(14)

228 8/24 7/48
EcoRV 53.7 (15) 57.2 (35)

260 16/32
EcoRV 190.6
(21)

Note:
All tests had at least 68 multilocus haplotypes in the comparisons and a maximum of n=72. The first row in each box represents the number of significant chi-square tests (P<0.05) between individual alleles at different RFLP loci over the total number of tests made for each pairwise comparison; the second row in each box represents the results of a chisquare test for the significance of association between all alleles at the two loci. The number in the parenthesis is the degrees of freedom for the test. All are significant at P<
0.05 and P<0.01 except those noted with *






42
nonrandom associations at the 1% level. Many individual allele combinations were not present in the population, but the chi-square test for these combinations was not significant; based on the overall allelic frequency, the expected number of these combinations was small. In contrast, for the most common alleles the large number of nonrandom associations was indicated by significantly larger or smaller numbers of observations when compared to the expected frequency of these combinations.

Discussion

Several lines of evidence support the concept that F. oxysporum f. sp. cubense has a clonal population structure in line with criteria established by Tibayrenc et al. (1990). A unifying feature of clonally reproducing organisms is widespread geographic distribution of a few successful clones. Even though this study identified 72 multilocus haplotypes in a worldwide collection ofF. oxysporum f sp. cubense, the five most common haplotypes accounted for nearly half of the isolates. Additionally, the two most common haplotypes were found on all five continents sampled in this study, indicating the pantropical distribution of a small number of genotypes.

Further evidence of clonal reproduction is the absence of recombinant genotypes. Significant gametic disequilibrium for alleles at 34 of 36 loci tested supported nonrandom association between alleles of different loci. In addition, the strong correlation between independent genetic markers (VCG, mitochondrial and multilocus RFLP haplotype) also are indicative of a clonally reproducing organism (Tibayrenc et al., 1990).

This study confirms that in phylogenetic analysis ofF. oxysporum f sp. cubense using parsimony, VCG is a strong predictor of cladistic groupings. Further differentiation into lineages may be done based on coefficients of similarity of RFLP haplotypes. At least 17 VCGs have been described for F. oxysporum f sp. cubense (Ploetz and Correll,






43

1988; Ploetz, 1990; Ploetz et al., 1992; Ploetz, 1994; Ploetz et al., 1997), and representatives from each of these VCGs were included in this study. With the exception of two isolates in VCG 0123, all isolates within a VCG were in the same clade and clonal lineage. The correlation between VCG and RFLP patterns has been observed previously in F. oxysporum formae speciales, including albedinis, conglutinans, dianthi, gladioli lycopersici, melonis, pisi, raphani, and vasinfectum (Elias and Schneider, 1991; Elias et al., 1993; Kim et al., 1992; Kistler et al., 1987; Manicom et al., 1990, Manicom and Baayen, 1993; Mes et al., 1994; Namiki et al., 1994; Tantaoui et al., 1996; Whitehead et al., 1992).

Although VCGs are good indicators of genetic similarity among the individuals comprising them, they do not provide information regarding the genetic similarity to individuals in different VCGs. In fact, this study shows that isolates belonging to different VCGs could have identical or nearly identical RFLP haplotypes. With two exceptions, the entire collection of isolates consisted of only ten distinct clonal lineages (Table 2-2). Clonal lineages provide a conservative system for grouping similar isolates, and the coefficient of similarity provides a numerical value to assess genetic relationships among isolates representing different lineages. This is in contrast to VCG groupings where, for an asexually reproducing organism such as F. oxysporum, it is impossible to determine the quantitative differences among individuals in different VCGs. Thus, the use of clonal lineages rather than VCG is proposed to genetically characterize similar isolates ofF. oxysporum f sp. cubense. In most instances, VCGs can be used to predict lineage.

Many of the isolates used in this study also have been classified based on their

electrophoretic karyotype (Boehm et al., 1994) and RAPD profile (Bentley et al., 1995). Based on their electrophoretic karyotype, Boehm et al. (1994) divided 118 isolates into






44
two major groups. Group I contained isolates from VCGs 0 124, 0124/0125, 0125, 012 10 and 0 1214 and was characterized by high chromosome number and large relative genome size [39.9-58.9 megabase pairs (Mbp)]. Group II contained isolates from VCGs 0120, 0121, 0122, 0123, 0129 and 01213, which had correspondingly fewer chromosomes and smaller genome sizes (32.1-44.9 Mbp). Using RAPD analysis, Bentley et al. (1995) similarly found that 54 isolates, representing 11I VCGs, could be divided into two major groups. Group I contained isolates in VCGs 0120, 0121, 0122, 0126, 01210, 01211 and

0 1212 while group II contained isolates in VCGs 0 123, 0124, 0124/0125 and 0 125. Cluster analysis indicated that VCG 0 1212 was distinct from the other VCGs in group I and 0 123 was distinct from group II.

Although the results presented here corroborate most of the broad conclusions made previously, this study provides additional and sometimes disparate conclusions regarding the affinities of some of these isolates. The bootstrap 50% majority rule consensus tree showed strong support for more than two clades among isolates ofFE oxysporurn f sp. cubense. The midpoint rooted tree divides isolates into two major groups. One group is comprised of isolates in five VCGs, which represent two significant clades. Isolates in this major group are remarkably homogenous, and the branch lengths that separate isolates are minimal. In contrast, the second branch encompasses isolates representing eight lineages, I1I VCGs, a large number of significant clades, as well as isolates belonging to other formae speciales of F oxysporum. Isolates representing the second group had more variable branch lengths compared to isolates in the first major branch.

With the exception of VCGs 0 122, 0123, 0126, 01210, 01212, and 0 1214, the relationship of isolates in 10 of the 17 VCGs correspond to those defined by previous






45

studies (Bentley et al., 1995; Boehm et al., 1994). Unlike previous investigations, isolates in VCGs 0122, 0126, 01212 and 01214 each formed individual clades with bootstrap values greater than 70%. Additionally, because this study used numerous, independent clones, had a large sample size, and provides bootstrap support for the clades, it gives a greater resolution of the genetic relationships among isolates ofF. oxysporum f. sp. cubense than do previous studies.

A number of the clonal lineages described here are phylogenetically distinct. Some isolates ofF. oxysporum f. sp. cubense are as genetically dissimilar to one another as they are to other formae speciales ofF. oxysporum (niveum and lycopersici). One interpretation of these results is that isolates belonging to the dissimilar groups acquired their ability to be pathogenic on bananas independently.

FOC V contains isolates from the Misuku Hills in Malawi, a relatively small area (approximately 400 square kilometers) on the country's northern border with Tanzania (Ploetz et al., 1992). All isolates in FOC V are in VCG 01214, and this is one of the few VCGs which has not been found in Southeast Asia, the center of origin of banana. Due to numerous lineage-specific alleles, FOC V is distant from all other lineages. One hypothesis is that this lineage ofF oxysporum f sp. cubense may have evolved independently of other members of the taxon in East Africa.

Alternatively, isolates within FOC V could have arisen by a founder effect. Bananas probably first arrived at the island Madagascar in the later half of the first millennium A.D. and from there moved to the coastal and then interior regions of the African continent (Ploetz et al., 1992). Diverse genotypes of banana are now found in East Africa, many of which are found nowhere else. It is possible that the pathogen was moved from Southeast Asia on the bananas introduced to Africa, and, as a result of






46
mutation and selection or through adaptation to endemic bananas, isolates in VCG 01214 may have diverged from their Asian progenitors.

FOC VII and X contain isolates in VCG 0123. The RFLP haplotype of isolates

belonging to these groups carry an assortment of alleles from the two major lineages (FOC I and II) as well as a number of lineage-specific alleles. Additionally, isolates comprising this clade are quite heterogenous; the five isolates comprising FOC VII belong to two significant sister groups. Also, all of the isolates in VCG 0123 fall neither into the same lineage nor in a single clade. Based on this information, this group may provide evidence of an ancient genetic exchange between individuals in FOC I and II. Alternatively, it may represent an ancestral group possessing primitive character states found in FOC I and FOC II.

In conclusion, isolates ofF. oxysporum f. sp. cubense represent a genetically diverse group of organisms, many of which are distantly related. Previous studies on other formae speciales indicate that many are genetically diverse. However, this study is the first based pm RFLP data of nuclear DNA to present evidence that a forma specialis of F oxysporum may be polyphyletic. The implied independent origin of pathogenicity to banana in some of the lineages has practical implications for work on this disease. Much effort is devoted towards developing cultivars of banana which are resistant to Fusarium wilt. Clearly, new hybrids should be screened against isolates representing the two most common lineages of the pathogen (FOC I and II). Ideally, breeding programs could screen new hybrids against isolates from each clonal lineage to increase the probability of developing cultivars that resist genetically distinct populations of the pathogen.













CHAPTER 3
GENETIC VARIATION IN TWO HONDURAN FIELD POPULATIONS OF FUSARIUM OXYSPORUM F. SP. CUBENSE Introduction

Fusariumu oxysporum Schlechtend, Fr. forma specialis cubense (E.F. Sm.) W.C.

Synder & H. N. Hans., is the causal agent of Fusarium wilt of banana, a disease which has been responsible for significant yield losses in banana. This disease had a devastating effect on the export banana trade in Central America during the early part of this century (Stover, 1962).

Bananas were introduced to Honduras well over 200 years ago and commercial trade began about 1876. Until the late 1950s, the race 1 susceptible banana cultivar Gros Michel was the most widely planted commercial cultivar in Honduras and elsewhere (Stover, 1962). However, disease epidemics in many commercial banana plantations were exacerbated by an increased prevalence of race I isolates due to the perennial monoculture production of this susceptible, clonally derived cultivar. Eventually, Cavendish cultivars, which are resistant to race I and 2, were used to replace Gros Michel for export production.

Three races of the pathogen that are pathogenic on banana have been reported in prior work. Only two races of the pathogen have been identified in Honduras (Stover, personal communication), and these isolates belong to only three vegetative compatibility groups (VCGs) (0120, 0124 and 0126) (Ploetz, 1990). Race I isolates occur in all three



47






48

of these VCGs, while race 2 isolates appear to be limited to VCG 0124. In a number of subtropical countries (Australia, South Africa, Canary Islands and Taiwan), many commercial plantations of Cavendish cultivars succumbed to Fusarium wilt shortly after they were planted (Ploetz, 1990). More recently, isolates in VCGs 01213 and 01216 have caused disease on Cavendish cultivars in the tropics. Isolates capable of causing disease on Cavendish cultivars were identified by Su et al. (1977) as race 4, a new race of the pathogen. Race 4 isolates also are pathogenic on all cultivars susceptible to race 1 and race 2 but presently have a more limited geographical distribution than races 1 and 2. There have been no reports of race 4 in Central America.

Fusarium wilt has been a problem in most major banana-producing regions of the world and the occurrence of the pathogen in areas where bananas are not indigenous is most likely a consequence of human dissemination of infested rhizomes. Bananas are rhizomatous, perennial plants that are grown in subtropical and tropical climates. Cultivated bananas are commonly sterile diploid, triploid, or tetraploid clones derived from Musa acuminata or interspecific hybridization between M. acuminata and M balbisiana (Simmonds, 1962). Since most edible bananas do not produce viable seed, they must be asexually propagated, typically using rhizomes or sucker plants. As such, they are usually also accompanied by soil and associated microorganisms; it is through the movement of infested plant material that the pathogen has been introduced to nonendemic regions. For example, isolates that have been introduced to the Americas on infested plant material comprise only a limited number of VCGs and clonal lineages compared to those found in southeast Asia, the center of origin of M. acurninata (Ploetz and Correll, 1988; Chapter 2).






49

A study of a world collection of F oxysporum f. sp. cubense isolates representing 17 vegetative compatibilty groups (VCGs) revealed that this pathogen is comprised of at least 10 clonal lineages (Chapter 2). The greatest diversity of VCGs and lineages are present in Malesia (Ploetz, 1994; Chapter 2), which is also the region of origin of the host (Simmonds, 1962).

Although the level of genotypic diversity that was represented in a worldwide

collection of F. oxyspornrn f sp. cubense isolates has been described previously (Bentley et al, 1995; Chapter 2), the amount of diversity found in individual field populations of the pathogen is unknown. Similarly, factors that may affect the levels of diversity within a field have not been explored for F. oxysporuni f. sp. cubense. McDonald et al. (1995) discussed several factors which may cause pathogen populations to undergo significant changes in genetic structure over time. For a perennial, tropical crop such as banana, cropping practices such as rotation, field burning, and fungicide application may affect the pathogen population. However, the use of host disease resistant genotypes also likely plays a major role. Similarly, the reproductive strategy and mutational rate of the pathogen, as well as the frequency of immigration of new genotypes, may also have a profound effect on the genetic composition of a field population.

The objectives of this study were to: i) determine the amount of genotypic

diversity in field populations of the pathogen in Honduras; and ii) determine whether specific genotypes of the pathogen preferentially infect specific host genotypes.






50

Materials and Methods

Description of Sampling Procedure

Sampling ofF. oxysporum f sp. cubense isolates was conducted in two adjacent disease screening fields (Fields 1 and 2) at the banana breeding station operated by the Fundaci6n Hondurefia de Investigaci6n Agricola (FHIA) in La Lima, Honduras. Historically, Field 1 had been planted to cultivar Highgate, which is susceptible to race 1, and had a high incidence of plants with Fusarium wilt based on visual assessment of symptoms. Therefore, it was an optimal location to evaluate the resistance to race 1 in seven of FHIA's tetraploid hybrids (Rivera, pers. comm.). Prior to planting the trial, inoculum levels were artificially increased in the field using a number of procedures. First, freshly infected pseudostem and rhizome tissue from Highgate plants infected with F. oxysporum f sp. cubense was uniformly spread throughout the field. Subsequently, the entire field was planted to the race 1-susceptible cultivar Maquefio. Prior to planting the Maquehio rhizomes, approximately 250-gram (g) chunks of symptomatic pseudostem and rhizome tissue from Highgate were placed in the bottom of each hole. After a year, 94% of the resulting plants exhibited internal symptoms of Fusarium wilt (vascular discoloration of rhizomes and pseudostems). Prior to planting the disease screening trial, infected Maquefio plants were incorporated into the soil. Additionally, prior to planting each experimental hybrid, 250 g of symptomatic Maquefio tissues were placed in each planting hole.

The experiment was arranged in a randomized complete block design with nine treatments (cultivars) and seven replications (Table 3-1). The banana plants used in the experiments were derived from tissue culture. The treatments were planted in the same





51

Table 3-1. Genotypes of banana planted in Field 1


Hybrid Genotype

FHIA 1: AAB x AA = AAAB
Dwarf Prata x 3142 (Pisang Jari Buaya derivative)

FHIA 2: AAA x AA = AAAA
Williams*x 3397 (3142 (AA) x 3217 (AA); both are Pisang Jari Buaya derivatives)

FHIA 3: ABB x BB = ABBB; ABBB x AA = AABB; AAB or ABB x AA = AAA,
AAB, or ABB
Cardaba x BB = ABBB x 2741 (AA) = 3386 [selected triploid x 3320 (AA)]

FHIA 6: AAB X AA = AAAB
Maquefio x 3437 (Black sigatoka resistant Mura acuminata subsp. burimanica)

FHIA 15: AAB x AA = AAAB
Same parentage as FH1A 6, but this selection is faster to ratoon

FHIA 17: AAA x AA = AAAA
Highgate x 3362 (3142 x 3217)

FHIA 23: AAA x AA = AAAA
Same parentage as FH1A 17 but different selection

Highgate: AAA (race 1 susceptible control (race 2 resistant)) Williams: AAA ( race 1 and race 2 resistant control) Bluggoe: ABB (race 2 susceptible control (race 1 resistant))

Note: With the exception of FHIA 3, all FHIA hybrids are either interspecific or intraspecific tetraploids obtained by performing crosses on Musa acumninata (AA) and M. balbisiana (BB) parental lines. Highgate, Williams and Bluggoe are banana cultivars used as host differentials to differentiate among the three races of Fusarium oxysporum f sp. cubense.
* It is questionable as to whether or not Williams was actually the parent of this cross, due to the cultivar Williams extreme sterility.






52

rows in which the cultivar Maquei'io had been planted. Each plot contained three plants spaced at 2.5 m, and the entire experimental area encompassed approximately 6500 in2. The field consisted of seven rows; three of these contained the seven replications of the experiment planted in a serpentine fashion throughout the plot. Experimental rows were separated from one another by a single row of the race 2-susceptible cultivar Bluggoe (Table 3-2).

The first samples were collected in February 1995 using a nondestructive method, which consisted of removing the first few outer layers of pseudostemn tissue from symptomatic and asymptomatic plants (Ploetz, 1992). A second set of samples were collected from the same planting in September 1995 using a destructive sampling method, which consisted of excising inner tissue of both the rhizome and pseudostem. When collecting symptomatic tissue, attempts were made to obtain samples displaying necrotic vascular strands. Since sampling occurred during two periods, some plants were sampled twice. Samples were obtained from every plant showing visible disease symptoms in all treatments and also from the symptomatic race 2 susceptible Bluggoe plants in the adjacent rows. Approximately one of every 12 asymptomatic plants in the entire field also were sampled to determine incidence of infection on these plants.

Field 2 was planted to the cultivar Bluggoe in an effort to increase inoculum levels for a future field trial to evaluate hybrids for resistance to race 2. The field was approximately 3500 m' and consisted of eight rows (Table 3-3). Seventy-six percent of the plants showed visible symptoms of Fusariumn wilt during an initial assessment of the field. Symptomatic plants were fairly evenly distributed throughout the field, and, based on this distribution, a systematic sampling strategy was followed in which one of every three plants was sampled following a serpentine pattern through the plot. Samples were






5 3

Table 3-2. Experimental design of banana plants in rows of Field 1.


Row Number'


2 3 4 5 6 7
B F15 B F15 B F2 B
B F15 B F15 B F2 B
B F15 B F15 B F2 B
B F6 B F3 B F I B
B F6 B F3 B F I B
B F6 B F3 B F I B
B F23 B W B F6 B
B F23 B W B F6 B
B F23 B W B F6 B
B F17 B H B F6 B
B F17 B H B F6 B
B F17 B H B F6 B
B F I B F17 B F17 B
B F I B F17 B F17 B
B F1 B F17 B F17 B
B H B F23 B F2 B
B H B F23 B F2 B
B Hz B F23 B F2 B
B W B F6 B F15 B
B W B F6 B F15 B
B W B F6 B F15 B
B F2 B Hz B H B
B F2 B Hz B H B
B F2 B H B H B
B F3 B F3 B F23 B
B F3 B F3 B F23 B
B F3 B F3 B F23 B
B W B F17 B F3 B
B W B F17 B F3 B
B W B F17 B F3 B
B F17 B F23 B W B
B F17 B F23 B W B
B F17 B F23 B W B
B F15 B F15 B F1 B
B F15 B F15 B F I B
B F15 B F15 B F1 B
B F6 B W B F2 B
B F6 B W B F2 B
B F6 B W B F2 B
B F2 B F2 B W B
B F2 B F2 B W B





54
Table 3-2--continued.
Row Numbery


1 2 3 4 5 6 7

B F2 B F2 B W B
B Fl B Fl B F6 B
B Fl B Fl B F6 B
B F1 B Fl B F6 B
B F3 B F2 B H B
B F3 B F2 B H B
B F3 B F2 B H B
B Hz B H B F17 B
B Hz B H B F17 B
B H B H B F17 B
B F23 B F1 B F1 B
B F23 Bz Fl B F1 B
B F23 B Fl B Fl B
B F23 B F6 B F15 B
B F23 B F6z B F15 B
B F23 B F6 B F15z B
B F17 B F3 B F3 B
B F17 B F3 B F3 B
B F17 B F3 B F3 B
B W B F15 B F23 B
B W B FI5 B F23 B
B W B F15 B F23 B

Y' Each letter or letter/numeral combination represents a single plant; bold type indicates plants that were sampled for the presence of Fusarium oxysporum f. sp. cubense; B= cultivar Bluggoe, H= cultivar Highgate, W= cultivar Williams, F l =FHIA 1, F2=FHIA 2 F3=FHIA 3, F6=FHIA 6, F 15=FHIA 15, F 17=FHIA 17, F23=FHIA 23

indicates multiple samples were obtained from these plants






55

Table 3-3. Field design of banana plants in rows of the race 2 disease screening plot.


Plant Number Row Number'

1 2 3 4 5 6 7 8

D D D D D D D D
2 D D D D D D D H
3 H D D D D D D D
4 D D D D D D D D
5 D D D D D D D D
6 H D D D D D D D
7 H D D D D D D D
8 H D D D D D D D
9 D D D D D D H
10 H D D D D D D H
I I H D H D D D H H
12 H D D D H H D D
13 D H D H D H D D
14 H H D D D H D H
15 H H D D D D D D
16 D H D D D D H H
17 D H D D D D D D
18 D H D D H D D D
19 H H D D D D D D
20 H D D H D D D D
21 H D D D D D D D
22 H D D D D D D D
23 H D D H D D D D
24 H D D D D D D D
25 H D D D D H D D
26 H D H D D D D D
27 H D D D H D D H
28 H H D D D D D D
29 D D D D D D D D
30 D H D D D D D D
31 H H D D D D D D
32 D H D H D H D H
33 D H D D D D D D
34 H D D D D H H H
35 H H D D D D H D
36 H D D D D D D H
37 H D D H H H D D
38 D D D D D D D H
39 H H D D H H D D
40 D


'The entire field planted to the race 2 susceptible cultivar Bluggoe. Each letter represents an individual plant. H = asymptomatic plants and D= symptomatic plants based on visual assessment-, samples were obtained from plants in bold type. Total plants = 312; Total D = 238 (76%); Total H = 74 (24%).






56
collected in March 1995 and consisted of destructive sampling of inner rhizome and pseudostem tissues. Sampling included both symptomatic and nonsymptomatic tissue. Fungal Isolations

Tissue samples were allowed to air dry for 3 to7 days prior to transport to Florida. In Florida, each sample was cut into approximately eight, 1 -cm pieces, which each contained at least one discolored and presumably infected vascular element. The tissue pieces were surface disinfested by placing them in 70% ethanol for 10 seconds, followed by a 2-min soak in 10% bleach, and a final rinse for at least 1 minute in sterile water. Pieces were air dried in a laminar flow hood and then placed into two petri dishes to provide four pieces of tissue in each dish. One of two media were poured over the tissue pieces. Medium 1 consisted of amended water agar, streptomycin sulfate (10 mg/L; Sigma Chemical Company, St. Louis, Mo.), rifampicin (12 mg/L; Sigma) and 10 A1 of danitol/L of medium (Chevron Corporation, San Francisco, CA.). Medium 2 consisted of amended potato dextrose agar (PDA) containing streptomycin (10 mg/L), rifampicin (12 mg/L), danitol 10 gl/L/L, and terigitol NP-10 (100 /A/L; Sigma). Medium 1 allowed for quick growth of Fusarium spp. while suppressing the growth of other microorganisms. However, identification ofF. oxysporum isolates was difficult on this medium because of poor mycelial growth and spore formation. Medium 2 allowed for better identification of F. oxysporum colonies but the additional nutrients were also more conducive to the growth of other microorganisms. Plates were incubated at room temperature, and within 3 days mycelial growth was observed on tissue pieces. Putative colonies of Fusariunm spp. were selected based on visual growth characteristics and the formation of microconidia or macroconidia. Agar sections (1 mm2) were excised from these colonies and placed on the






57
reciprocal medium (from medium 1 to 2 and vice versa). Colonies that were confirmed to be F. oxysporum based on mycelium pigmentation and growth characteristics, microconidia or macroconidia formation, phialide anatomy, or chlamydospore formation were streaked onto plates containing PDA. Within 14 hours, four single, germinated spores were picked off using a compound microscope and placed with a sterile needle on plates containing PDA; at least one single spore isolate was collected for each sample. DNA Extraction, Southern Blotting and Hybridization

Procedures for DNA extraction have been described previously (Chapter 2). Approximately 10 gg of DNA were digested with at least 10 units of either HaeIII or HindIII restriction enzymes (Bethesda Research Laboratories, Gaithersburg, MD) and incubated for at least 3 hours at 37C. Restriction fragments were separated by electrophoresis in 0.7% agarose for HindIII-digested DNA or 1.5% agarose for HaeIIIdigested DNA in TBE buffer at pH 7.0. Gels were run at 30 volts for approximately 24 hours. Bacteriophage lambda DNA digested with either PstI or HindIII was included on each gel and used to calculate the molecular mass of restriction fragments obtained from F. oxysporum f sp. cubense DNA. The DNA of each isolate was visualized by incorporating 100 ng of ethidium bromide/ml of solution into the agarose gels and then illuminating the gels using UV irradiation. The DNA was transferred to Nytran membranes (Schleicher & Schuell, Inc., Keene, NH) using the capillary transfer method (Sambrook et al., 1989). The DNA transfer proceeded for at least 12 hours, and the DNA was immobilized by UV crosslinking (UV 254-nm crosslinker, model FB UVXL 1000, Fisher Scientific, Pittsburgh, PA).






58
Mitochondrial DNA (mtDNA) was isolated from Honduran isolate 3 S1 (VCG 0120) following a procedure described previously (Kistler and Leong, 1986). Total mtDNA was labeled using random hexamer primers to incorporate fluorescein- I 2-dUTP. DNA labeling; hybridization and detection followed the procedures provided by the manufacturer (Dupont NEN "Renaissance," E. I. du Pont de Nemours & Co. Inc., Boston, MA). Prehybridization, hybridization and washing were performed at 65oC using a Hybaid hybridization oven (Dot Scientific Inc., Flint, MI). Membranes were placed between acetate sheets and exposed to x-ray film for 15 to 30 minutes.

Preliminary work on a world-wide collection ofF. oxysporum f sp. cubense

isolates revealed that mitochondrial haplotypes were correlated with VCG and isolates tested within a VCG had identical mitochondrial haplotypes (Kistler and Momol, 1990, Chapter 2). Isolates belonging to the three VCGs found in Honduras (0120, 0124 and 0126) could be differentiated based on their unique mitochondrial profiles, obtained by probing blots of total HaellI-digested DNA with mtDNA isolated from Honduran isolate 3S1. Using this method, a mitochondrial haplotype was assigned to each isolate. Honduran isolates which had mitochondrial profiles identical to isolates in VCG 0120 were designated mitochondrial haplotype (mthap) 0120, those which had mitochondrial profiles identical to isolates in VCG 0124 or VCG 0126 were designated mthap 0124 and mthap 0126, respectively.

Similarly, F. oxysporum f sp. cubense isolates were differentiated based on their DNA fingerprint by digesting total DNA with the restriction enzyme HindIII and probing with clone pEYI0 (Kistler et al, 1991). The clone pEY10 hybridizes to a nuclear dispersed, middle repetitive DNA sequence (Benny and Kistler, unpublished) and contains






59
an open reading frame encoding a polypeptide with 51% amino acid identity to the transposase of the Magnaporthe grisea transposon Pot3 (Farman et. al., 1996; Pettway and Kistler, unpublished). All previously tested isolates from a world-wide collection of F. oxysporunm f sp. cubense had unique DNA fingerprints using pEY10 (Kistler et al., 1991). The DNA labeling, hybridization and detection procedures followed those described above.

Data Analysis

DNA fingerprints were scored by visual assessment. Only bands that were of strong intensity and consistently present were scored. Similar patterns were assigned a number based on their frequency of occurrence in the field; the most common pattern was assigned the number 1. Two chi-square analyses were performed. In the first test, observed percent infected and noninfected of each FHIA hybrid and the two controls, Highgate and Williams, were obtained and compared against the null expectation of no differences among cultivars. The expected number of infected plants for this test was 6. 1, which was obtained by summing the number of isolates obtained from the FHIA hybrids and the cultivars Highgate and Williams and dividing by 189 the total number of plants in the experiment, excluding the cultivar Bluggoe. This number was then divided by 9, which represents the total number of cultivars in the experiment. The expected number of noninfected plants was 14.8, and this number was obtained as described above but based on the total number on noninfected plants. The chi-square analysis had 8 degrees of freedom. The second test determined whether the pathogen genotypes obtained from infected plants were randomly distributed among the different host genotypes. In this test, the observed frequency of each pathogen genotype in the field and the host genotype from






60
which they were obtained were compared against the null expectation of no differences among the host genotypes. The expected frequency for each genotype was estimated by dividing the total number of isolates obtained from each cultivar by three. For example, a total of 16 F. oxysporurn f sp. cubense isolates were obtained from the cultivar Highgate, and the expected number of isolates obtained for each mtDNA haplotype was 5.3. The chi-square analysis had 16 degrees of freedom.

Field I was sampled twice and isolates were collected from both symptomatic and asymptomatic tissue during each period. Duplicate isolates were collected from symptomatic tissue of treatments 104-2, 106-3, 208-2, 306-2, 308-2, 408-1, and 408.3. Additionally, duplicate samples were obtained from one Bluggoe plant. Field 2 was sampled once and duplicate isolates were analyzed for four of the Bluggoe plants. Additionally, many of the isolates from both fields were duplicated on different gels to validate the results.

Results

Ninety-seven isolates were collected from Field 1 (the race 1 -screening field). Of these 97 isolates, 12 isolates were obtained from plants that were symptomatic during both sampling periods and represent multiple samples taken from the same plant. A mtDNA haplotype was assigned to each isolate by probing blots of HaeJII-digested DNA from each isolate with labeled mtDNA isolated from Honduran isolate 3S1I (Figure 3 -1). Isolates collected from the field were assigned a mtDNA haplotype as described in the materials and methods. The incidence and distribution of pathogen mtDNA haplotypes varied with host genotype (Table 3-4). The Highgate and Cavendish cultivars; used as susceptible and resistant experimental controls had the highest and lowest incidence of






61
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Molecular
Weight SWma
Kb









11
W
4 .5 ...... ..... .















Isolate/Host Cultivar mtDNA Haplotype

2 1-5-4Bluggoe 0124
3 1-3-63 Bluggoe 0124
4 1-3-62 Bluggoe 0124
5 1-1-52 Bluggoe 0124
6 27-5-11 Bluggoe 0124
7 2-8-26 Bluggoe 0124 ... .. ..........










8 1-3-51 Bluggoe Unknown type
9 1-3-3 Bluggoe 0124
10 106-2 Highgate 0120
11 601-2 FHIA 6 0126
12 308-1 Highgate 0126
............::: :: :::::::::::::

....... ..... i" : iii~ i i i i :: !i ........ .:. .. .....ii:!i!!i:!ii:i:::
























13 707-3 FHIA 15 0126
14 502-1 FHIA 17 0126
..........~~~~~~ ~~~~....... .::: .:: .... ::i :ii::::i:::~ii!::!,i~i














Isolate/15 304-3 FHIA 15 0126ost Cultivar tDNA Haplotype
1 y- 1-20 Bluggoe Unknown type









16 1-508-1 FHIA 2Bluggoe 0126
17 707-3-63 BluggoeFHIA 15 0124
18 408-3-62 Bluggote 0126
5 1-1-52 Bluggoe 0124
6 2Y-5-11 Bluggoe 0124


















Figure 3-1: mtDNA haplotype of selected Honduran field isolates.
7 2-8-26 Bluggoe 0124









8Isolates with 1-designation were1-3-51collected from the cultivar Bluggoe Unknowin Field 1.type
9Isolates with 2-designation were collected from the cultivar1-3-3 Bluggoe 0124in Field 1.
10FHIA hybrids designated according to their lot number in Field 1.20
11 601-2 FHIA 6 0126
12 308-1 Highgate 0126
13 707-3 FHIA 15 0126
14 502-1 FHIA 17 0126
15 304-3 FHIA 15 0126
"i6 108-1 FHIA 2 0126
17 707-3 FHIA 15 0124
18 408-3 Highgate 0126
19 409-3 FHIA 6 0126

Figure 3 1: mtDNA haplotype of selected Honduran field isolates. Isolates with 1 -designation were collected from the cultivar Bluggoe in Field 1. Y Isolates with 2-designation were collected from the cultivar Bluggoe in Field 1. SFHIA hybrids designated according to'their plot number in Field 1.






62
infection, respectively. No isolates were obtained from the Cavendish cultivar Williams, which is resistant to both race I and race 2 of the pathogen. In contrast, 17 isolates were collected from the Highgate plants. Fourteen of the 17 isolates had 0126 mtDNA haplotypes, whereas one each had 0120 and 0124 mtDNA haplotypes. One isolate collected from the cultivar Highgate had a previously uncharacterized mtDNA haplotype for F. oxysporum f sp. cubense.

Percent infection for the seven hybrid bananas ranged from less than 1% for FHIA

1 to 57% for FHIA 15 (Table 3-4). The observed percent infection of each cultivar was obtained and compared against the null expectation of no differences among cultivars with a series of chi-square tests. The differences in infection on the cultivars were highly significant (p < .001) indicating that cultivars are not equally likely to be infected. This suggests differences in host resistance genes. The distribution of pathogen genotypes varied on the different host genotypes. The overwhelming majority of isolates collected from the hybrids had 0126 mtDNA haplotypes. Isolates with 0120 and 0124 mtDNA haplotypes were recovered from some of the hybrids but at a lower frequency compared to isolates with 0126 mtDNA haplotypes. In contrast, 96% of the isolates collected in the same field but obtained from the race 2 susceptible cultivar Bluggoe, had the 0124 mtDNA haplotype (Table 3-4).

The clone pEY1 0 was used to generate DNA fingerprints of the isolates (Figure 3-2). Isolates which had identical DNA fingerprints were presumed to represent the same clone. Clone frequency and distribution on host genotypes were examined in the two field populations. The nuclear DNA fingerprints of the isolates with 0126 mtDNA haplotypes showed little variation. Only two patterns, which were 99% similar, were






63

Table 3-4. Isolates of Fusarium oxysporum f sp. cubense collected in Field 1 and their incidence on different host genotypes

Number of Isolates with mtDNA Haplotypes


Cultivar or Hybrid 0120 0124 0126 Total Infections
Isolates' %Z

Highgate (race 1
susceptible) 1 1 14 16 76
Williams (race 1 &
2 resistant) 0 0 0 0 0

Bluggoe (race 2
susceptible) 0 26 1 27 11

FHIA 1 0 1 0 1 <1

FHIA 2 1 1 7 9 43

FHIA 3 0 0 2 2 14

FHIA 6 0 0 8 8 38

FHIA 15 0 1 11 12 57

FHIA 17 0 0 3 3 14

FHIA 23 0 1 3 4 19

XFor plants where more than one isolate was collected, only one is represented if the mtDNA haplotypes were identical.
Y Total percentage of plants in Field 1 from which isolates were recovered Chi-square analysis: p < .001; df= 8






1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 64


Molecular ......
W eight .... ......
Kb

5.07
4 .7 .
*: i-.', :' i
4.5 K .1



2.1 1.9
1.7


1,16
1.09 H



Isolate/Host Cultivar mtDNA Haplotype
1 2x-3-18 Bluggoe 0124
2 102-3 FHIA 6 0126
3 2-7-37 Bluggoe 0124
4 707-2 FHIA 15 0126
5 503-2 Highgate 0126
6 1-5-66 Bluggoe 0124
7 503-3 Highgate 0124
8 707-3 FHIA 15 0124
9 309-1 FHIA 2 0124
10 208-1 Highgate 0126
11 1-1-59 Bluggoe 0124
12 304-1 FHIA 15 0126
13 208-2 Highgate 0126
14 2-8-16 Bluggoe 0124
15 1-1-29 Bluggoe 0124
16 1-5-4 Bluggoe 0124

Figure 3-2: HindlII digested DNA of selected Honduran field isolates that were probed with clone pEY10 to generate DNA fingerprints. x Isolates with 2-designation were collected from the cultivar Bluggoe in Field 2. Y FHIA hybrids designated according to their plot number in Field 1. Isolates with 1-designation were collected from the cultivar Bluggoe in Field 1.






65
observed, and they represented the same clonal lineage. The two isolates with 0120 mtDNA haplotypes, which infected the cultivars Highgate and FHIA 2, had identical nuclear DNA fingerprints that were distinct from isolates representing other mtDNA haplotypes. With the exception of a single isolate, all of the isolates with 0124 mtDNA haplotypes, which infected the cultivar Bluggoe and FHIA hybrids 1,2,15 and 23, had identical nuclear DNA fingerprints that were distinct from isolates representing other mtDNA haplotypes. An isolate collected from the cultivar Highgate with a 0124 mtDNA haplotype had a unique nuclear DNA fingerprint.

In contrast to Field 1, which contained ten different genotypes of banana, Field 2 was entirely planted to the cultivar Bluggoe. Fifty-nine isolates ofF. oxysporum f sp. cubense were collected in this field, and all of them shared the same 0124 mtDNA haplotype (Table 3-3). Similarly, all tested isolates had a nuclear DNA fingerprint identical to those isolates in Field 1 which had a 0124 mtDNA haplotype and were recovered from Bluggoe and FHIA hybrids 1, 2, 15 and 23.

The observed frequency of each pathogen genotype in the entire field and the host genotype from which they were obtained were compared against the null expectation of no differences (no host selection) with a series of chi-square tests (Figure 3-3). The differences in clone frequencies on the various host genotypes were highly significant. Clones of the pathogen did not randomly infect banana plants representing the various host genotypes. Instead clones were preferentially distributed on specific host genotypes.

Discussion

Clonal diversity was low in the two field populations of F. oxysporum f sp.

cubense; only five clones were recovered from the 10 genotypes of banana. Although the






66











30

25

20
S Chi-square analysis: p < 001; df=16
*mthaplotype 0120
15 a mthaplotype 0124
0,
gmthaplotype 0126 E 10
z
5

0



)Observed frequency of m thaplotype 0r-20 = .02 in Field 1.



Observed frequency of mt haplotype 0124 = .38 in Field 1.
Observed frequency of mt haplotype 0126 = .60 in Field 1.









Figure 3-3: Incidence and distribution of mtDNA haplotypes ofFusarium oxysporum f.
s. cubense on host notess in Field 1.
-~ T 44
-= L LL LL LL 7 T I LL LL LL F
H ost c ultiva r


Observed frequency of m thaplotype 0120 = .02 in Field 1.
Observed frequency of mt haplotype 0 124 = .3 8 in Field 1.
Observed frequency of mt haplotype 0 126 = .60 in Field 1.









Figure 3 -3:i Incidence and distribution of mtDNA haplotypes of Fusariuni oxysporum f. sp. cubense on host genotypes in Field 1.






67
number of clonal genotypes in these two populations was low, there appeared to be a relationship between the level of genotypic diversity in the pathogen and host populations. By increasing the number of genotypes of banana in a field, more genotypes of the pathogen were detected. In fact, the number of different clones of the pathogen which were recovered in each field was proportional to the number of host genotypes planted in the field. For example, all of the isolates recovered from host tissues in Field 2, which was planted to a single cultivar, belonged to a single mtDNA haplotype and represented a single clone. In contrast, isolates recovered from host tissues in Field 1, which was planted to 10 different genotypes of banana, comprised three mtDNA haplotypes and five clones. Each mtDNA haplotype was represented by only one or two clones.

These findings were consistent with field studies which compared the genetic diversity of nonpathogenic F. oxysporum in native soils to F. oxysporum f sp. melonis isolates. In contrast to populations of native, nonpathogenic Fusarium oxysporum, pathogen populations ofF. oxysporum f. sp. melonis were less diverse based on their VCGs and mtDNA haplotypes (Gordon and Jacobson, 1990a and b; Gordon and Okamoto, 1992). In contrast, field populations of other asexually reproducing fungi, including Stagonospora nodorum and Magnaporthe grisea as well as sexually reproducing fungi including Sclerotinia sclerotiorum, Mycosphaerella graninicola and some populations of Phytophthora infestans, contained a larger number of genotypes, indicating higher levels of clonal diversity (Goodwin et al., 1992; Kohli et al.,1995; Kohn, 1995; Levy et al., 1993; McDonald et al., 1994; McDonald et al., 1995).

The recovery of different clones ofF. oxysporum f sp. cubense in these two fields was dependent on the genotype of the host. Chi-square analyses were performed to test






68
whether isolates of the pathogen were randomly recovered from the different banana genotypes; these showed a significant association between a mtDNA haplotype and its incidence on specific host genotypes. In fact, isolates with certain mtDNA haplotypes were more frequently and sometimes almost exclusively recovered on certain host genotypes (Figure 3-3). For example, 99% of the isolates recovered from the race 2-susceptible cultivar Bluggoe in Fields 1 and 2 had a 0124 mtDNA haplotype and represented a single clone based on their identical DNA fingerprints. In contrast, 87% of the isolates recovered from Highgate had a 0126 mtDNA haplotype, representing two nearly identical clones.

The occurrence of a specific genotype of the pathogen associated with virulence on a specific host genotype is likely the consequence of linkage disequilibrium found in clonally reproducing organisms (Tibayrenc et al., 1991). Since sexual reproduction does not occur in these organisms, the entire genome is effectively linked (Anderson and Kohn, 1995). Therefore, alleles at different loci, such as those for mt haplotypes, VCG, virulence, are repeatably associated. Additionally, in these field populations it appears that the host genotype drives selection for particular clones of F. oxysporum f sp. cubense. In fields planted to a susceptible cultivar, genotypes of the pathogen which are virulent on the host genotype may have greater fitness. They may, for example, produce more propagules. The pattern of introducing strains ofF. oxysporum f sp. nielonis to agricultural soils and the correspondingly intense selection pressure on the susceptible crop, has been described previously (Gordon and Okamoto, 1992). Such selection pressure undoubtably has an even more significant effect on the pathogen populations in perennial, monocultural production systems, such as banana plantations, since susceptible






69
host plants are exposed to the pathogen for even longer times. In these systems, one would expect a large increase in the number of propagules of virulent pathogen genotypes when susceptible cultivars are grown.

In contrast to certain plant pathogens, such as rusts, where new pathotypes appear annually and breeders are continually challenging the pathogens by introducing cultivars with new combinations of resistant genes, resistance genes to Fusarium wilt in Cavendish cultivars have provided stable, relatively long-term protection. In fact, Cavendish cultivars have been planted in Central America for at least 40 years and no substantial reports of Fusarium wilt on these cultivars have been made. However, in a limited number of subtropical and tropical countries, race 4 isolates of the pathogen, which are capable of causing disease on Cavendish cultivars, do occur. The asexual nature of this fungus may preclude rapid change in the pathogen population. However, lack of genetic change due to recombination in populations of the pathogen actually may be detrimental in commercial banana plantations where monoculture production of clonally propagated banana cultivars are typical. In these cropping systems, when a virulent pathotype evolves, lack of recombination in the pathogen can serve to stably perpetuate the rapid selection and growth of the most virulent pathotype.

This study, in combination with information from Chapter 2, indicates that race 1 isolates belong to more than one distinct lineage. The study of a world collection ofF. oxysporum f. sp. cubense found that isolates in VCG 0124 belonged to one clonal lineage, while isolates in VCG 0120 and 0126 belonged to a second lineage. Isolates in these two lineages were as genetically dissimilar to one another as they were to isolates ofF. oxysporun f. sp. niveuni and f. sp. lycopersici isolates. In the present study, the DNA





70
fingerprint of isolates representing these two lineages were distinct, further substantiating their genetic disimilarity. Race I isolates are known to occur in all three of the VCGs found in Honduras (Ploetz and Correll, 1988) and the present results showing mitochondrial haplotypes corresponding to these three VCGs, recovered from diseased Highgate plants (Table 3-4), support this observation. The definition of race assumes that all race I isolates share identical virulence characteristics corresponding to susceptibility in the cultivar Gros Michel (and Highgate) and resistance in the cultivar Cavendish. However, in this study, the frequency with which genotypes representing these two lineages were isolated differed in the race 1 -susceptible cultivar Highgate and the FHIA hybrids. Isolates in the lineage represented by mthap 0126 preferentially infected the cultivar Highgate and most of the FHIA hybrids. In contrast, isolates with mthap 0124, which represent a second phylogenetic lineage, were recovered only at a low frequency from these same host genotypes. Because isolates with mthap 0124 were abundantly present in the field and were frequently recovered from Bluggoe plants, the basis for the race 1 phenotype may be different in these two lineages. This may imply genetic differences in the virulence determinants of these two nominally race I lineages. This of course assumes that isolates with identical mthap and DNA fingerprints in field I have the identical pathogenic phenotype.

Similar situations have been observed in other fungi. In a survey of Magiiaporthe grisea pathotypes from the United States, Levy et al. (1991) found that isolates of a single pathotype had significantly different DNA fingerprints, and this suggested that the pathotypes were derived from independent clonal lineages. In F. oxysporurm f sp. cubense, isolates in VCG 0126 that cause disease in Highgate may possess genetic factors






71
that make them better adapted than those in VCG 0 120 and 0 124 to the ecological and environment factors in Honduras.

In this study, the majority of isolates recovered from Highgate had mthap 0126, whereas the majority of isolates recovered from Bluggoe had mthap 0124 (Table 3-2). Isolates representing other haplotypes were recovered on these host genotypes, but at low frequencies. Additionally, similar results were observed for some of the FHIA hybrids. The presence of a low rate of mixed infections may provide one explanation for this observation. During the initial selection of isolates from the infected tissue, occasionally multiple samples were taken from a single plant when fungal colonies displayed different growth characteristics. Of these samples, only one plant was infected with isolates having both 0 124 and 0 126 mtDNA haplotypes. Additionally, five plants from Field 1, each representing different genotypes of banana, were sampled multiple times by excising different xylem elements within a single plant. A total of 45 samples were obtained and all of the sampled plants were infected by isolates representing a single mtDNA haplotype. Thus, it appears that mixed infections occur, albeit at a low frequency.

Banana breeders in Honduras as well as other regions are involved in developing cultivars of banana with resistance to all races of this pathogen. Because of the polyploid nature of bananas and the sterility of many of the important edible cultivars, it is impossible to do traditional plant breeding and disease screening. Additionally, in Situi disease screening in regions where the pathogen is nonendemic is limited to isolates found within the country. Since isolates belonging to only three VCGs and two clonal lineages have been described in Honduras, yet over 17 VCGs and 10 clonal lineages have been described worldwide, breeding for resistance using endemic populations in Honduras may






72
not result in resistance to populations of the pathogen in other locations (Ploetz, 1995; Stover and Buddenhagen, 1986). Ideally, new genotypes should be screened against isolates that encompass the range of genetic diversity present in the entire formae speciales. Therefore, there is a need to assess the range of variation in the pathogen populations for virulence on particular breeding lines (Milgroom and Fry, in press). Using both the mtDNA haplotype and a nuclear DNA fingerprinting probe to estimate genetic diversity within a field site can improve in situ screening for Fusarium wilt. In this pathosystem there is a relatively long-term interaction between host and the pathogen that may not be assessed in artificial inoculation tests (Buddenhagen, 1990). Using molecular techniques for assessing the pathogen population at a given site and determining the lineage of isolates infecting particular host genotypes can increase the effectiveness of disease screening and assist in the evaluation of new breeding lines.













CHAPTER 4
SUMMARY AND CONCLUSIONS


This research has provided a genetic characterization and phylogenetic analysis of isolates of F. oxysporurn f sp. cubense representing both a world collection and population from two fields. Prior investigations, which genetically characterized isolates based on VCGs, RAPIs and electrophoretic karyotyping, provided only limited information on the genetic relationships among isolates (Bentley et aL, 1995; Boehm et al., 1994; Ploetz and Correll, 1988; Sorensen et al., 1993). However, the present research, which utilized numerous, independent DNA clones to identify polymorphic loci among a worldwide collection of isolates, has provided a comprehensive assessment of the genetic relationships among isolates. Parsimony analysis of the data using bootstrapping also provided a statistical evaluation of the RFLP data as well as an unbiased assessment of significant groupings among the isolates.

In addition, this research represents the first attempt to determine the amount of genotypic diversity present in field populations of F. oxysporumf. sp. cubense based on DNA fingerprinting. The study of two Honduran field populations provides information regarding the amount of genetic variability present in a defined location. By identifying the mtDNA haplotypes of individual isolates and determining the extent of clonality in the population using a DNA fingerprinting probe, this study assessed the genetic diversity of





73






74
the collected isolates and demonstrated how the genotype of the host can affect the pathogen populations.

Both studies demonstrated the importance of choosing an appropriate DNA

marker in phylogenetic and population genetic investigations. An important benefit to using RFLPs in phylogenetic analysis is that the number of RFLP markers available for use is large because any cloned, low-copy-number piece of DNA can be used as a probe and several restriction enzymes can be assayed to identify RFLPs with each probe (Michelmore and Hulbert, 1987). One important characteristic of the RFLP markers used in this phylogenetic analysis is that they were likely selectively neutral. This is in contrast to several markers that have been employed in traditional studies, such as pathogen virulence genes or resistance to fungicides, which are under strong selection pressure in agricultural systems and may provide a biased estimate of genetic diversity (McDonald and McDermott, 1993).

Other molecular markers that are commonly used in phylogenetic analyses of fungi include isozymes, RAPDs, the nuclear and mitochondrial rRNA genes and internal transcribed spacers (ITS), and mtDNA maps. Isozymes have been used to characterize fungal isolates but their disadvantage is that the possible number of enzymes that are informative for an organism is limited and many loci are monomorphic and, thus, noninformative. (Michelmore and Hulbert, 1987).

Analysis ofF. oxysporum f sp. cubense isolates using RAPDs was attempted

initially in this research. However, this technique proved to be unreliable for phylogenetic studies. The largest obstacle to using them was that polymorphisms could not be scored unambiguously without cloning the putative polymorphic bands and then performing an






75
RFLP analysis. Even when RAPDs were used that could be excised with an appropriate restriction enzyme, results were difficult to interpret (see Appendix I). Two problems associated with methods based on polymerase chain reaction (PCR), such as RAPDs, are that 1) nonamplification can result from base pair substitutions at any of the different nucleotides complementary to the primers at the ends of the amplified fragment, and 2) results may be affected by changes in environmental conditions and at times are not reproducible (Rosewich & McDonald, 1994).

Mitochondrial DNAs and, in certain organisms, mitochondrial plasmids have been used to characterize isolates ofF. oxysporum f sp. melonis, f sp. conglutinans, and f sp. cubense (Jacobson and Gordon, 1990; Kistler et al., 1987; Kistler and Momol, 1990). Two disadvantages of using mtDNA for phylogenetic analyses is that the genome is small, so consequently fewer polymorphisms may be found, and the rate of evolution of mtDNA may be different than that of nuclear DNA. Mitochondrial DNA was used in both the phylogenetic and population studies. In the phylogenetic study, groupings of isolates based on mtDNA haplotypes were congruent with groups based on the RFLP analysis; however, fewer groups could be distinguished based on mtDNA analysis. In the population studies, a mtDNA haplotype was used to group similar isolates, which were then subjected to DNA fingerprint analysis to determine whether isolates were clonally derived.

Sequencing of both the nuclear and mitochondrial rRNA genes as well as ITS regions have been described for fungi and used in phylogenetic studies (Bruns et al., 1991). An underlying assumption of sequence analysis is that the phylogeny of the region is a good indicator of the phylogeny of organisms (Bruns et al., 1991). Another approach






76

to identifying sequence differences in the ITS region is through an RFLP analysis using the PCR-amplified region. In this study, a number of primers were used to amplify ITS regions ofF oxysporum f sp. cubense isolates. Several four- and six-base restriction enzymes were used to digest the amplified products (Appendix II). No polymorphisms were observed using the restriction enzymes studied, and this methodology was discontinued because it did not prove useful for a phylogenetic analysis.

A central point of discussion in both studies was the overwhelming evidence of clonality in this organism. With the exception of isolates in VCG 0123, the genetic evidence which resulted from this research is consistent with the hypothesis that F. oxysporuni f sp. cubense is a clonally reproducing organism. Based on criteria established by Tibayrenc et al. (1990), a unifying feature of clonally reproducing organisms is widespread geographic distribution of a few successful clones. The RFLP study identified 72 multilocus haplotypes in a worldwide collection ofF. oxysporurn f, sp. cubense. The five most common haplotypes accounted for nearly half of the isolates, while the two most common haplotypes were found on all five continents sampled in this study, indicating the pantropical distribution of a small number of genotypes. Additional evidence of clonal reproduction is the absence of recombinant genotypes. Significant gametic disequilibrium for alleles at 34 of 36 loci tested supported nonrandom association between alleles of different loci. Finally, the strong correlation between independent genetic markers (VCG, mitochondrial and multilocus RFLP haplotype) also are indicative of a clonally reproducing organism (Milgroom, 1996; Tibayrenc et al., 1990).

The Honduran field population study also indicated that clonal reproduction accounted for the occurrence of a specific genotype of the pathogen associated with






77
virulence on a specific host genotype. This observation is likely the consequence of linkage disequilibrium found in clonally reproducing organisms where alleles at different loci, such as mtDNA haplotypes, VCG, virulence, are observed to be repeatedly associated. The clonal structure in agricultural populations of plant pathogenic fungi also has been found in some populations of Scierotinia sclerotiorum, Fusarium oxysporumn, Phytophthora infestans and Magnaporthe grisea (Appel and Gordon, 1994; Goodwin et al., 1992; Gordon et al., 1992; Kohli et al., 1995; Kohn, 1995; Levy et al., 1991).

A significant finding of the phylogenetic analysis was that a number of the clonal lineages are phylogenetically distinct. Host specialization, such as the ability to cause disease on a particular host, appears to be polyphyletic in F. oxysporumn f sp. cubense. Based on coefficients of similarity, isolates in the two largest lineages ofF. oxysporum f sp. cubense are genetically more similar to the F oxysporum f sp. niveum isolate than to each other. Similarly, they are roughly as genetically distinct from each other as either is to the isolates of F. oxysporum f. sp. lycopersici. An interpretation of these results is that isolates belonging to the dissimilar groups acquired their ability to be pathogenic on bananas independently. This is the first reporting of a potential polyphyletic acquisition of pathogenicity on a particular host for F. oxysporum f. sp. cubense. Appel and Gordon (1994) observed a close relationship between nonpathogenic strains ofF. oxysporum and F. oxysporum f. sp. melonis isolates in VCG 0131, suggesting an independent origin as a melon pathogen separate from FE oxysporum f. sp. melonis isolates in VCG 0134. However, this study was based on mtDNA haplotype data and interpretation may be complicated by the possibility of horizontal transfer of mtDNA.






78

Interestingly, the results of the R-FLP study indicate that isolates in VCG 0 1214, which comprise lineage FOC V, are genetically distant from other F oxysporuni f. sp. cubense isolates. One explanation is that this group could have originated in East Africa, a region outside the host's center of origin. Another explanation is that the genetic differences in these isolates could have arisen by a founder effect. Founder effects result when small groups of individuals are isolated from larger populations. It is possible that over a period of time, isolates in VCG 0 1214 were subjected to different selection forces which altered gene frequency in this group. Additionally, mutations which occur in the founding population could be fixed in the successive generations due to the clonal nature of this organism.

The data indicate that isolates in VCG 0123, which comprise both lineages FC VII and X, are phylogenetically distinct. This group may provide evidence of an ancient genetic exchange between individuals in lineages FCC I and II, or they may represent an ancestral group possessing primitive character states.

There are numerous practical applications of this research; however, only the most important aspects are highlighted. The phylogenetic study helps clarify the genetic relationships among the isolates which represent different races of the pathogen. Isolates which represent race 1 of the pathogen occur in the two largest and divergent lineages (FCC I and II) in the collection. However, data from the Honduras field study revealed that isolates which have been considered race I based on reaction to the host differential may not share the same host infectivity. In field tests, isolates representing these two lineages were not recovered at the same frequency from susceptible host tissue, even though isolates representing FCC I and FCC II were assumed to be evenly distributed






79
throughout the field. Isolates which represent race 2 appear to be limited to lineages FOC I and VIII, which are associated with one main branch on the midpoint rooted tree, and FOC V, which is only distantly related to other F oxy.sporuni f. sp. cubense isolates (Figure 2-2). In contrast, isolates that represent race 4 appear to be limited to lineages FOC 11, 111, and VI, which form a second branch on the midpoint rooted tree. Therefore, these two races appear to have evolved only in certain lineages of the pathogen. Race 4 isolates occur in many of the VCGs that are present in tropical areas, such as Central America, the Caribbean, and South America, but where race 4 has not yet been reported. If the ability to cause disease on Cavendish cultivars is controlled by a single gene it is possible that race 4 isolates could evolve in regions where it has not yet been found. However, it would most likely evolve in those isolates representing VCGs that presently contain race 4 isolates. Additionally, because race 4 isolates have a wide host range, it is important to identify and develop new sources of disease resistance. These new sources of disease resistance must then be incorporated into commercially acceptable banana cultivars.

Banana breeders worldwide are involved in developing cultivars of banana with resistance to all races ofF oxysporzrni f sp. cubense but are faced with a number of obstacles. Because of the polyploid nature of bananas and the low level of fertility of important edible cultivars, it is impossible to use many of the desirable commercial clones as parental lines for breeding work. Additionally, the results of container disease screening experiments do not necessarily correlate in field studies. Stover and Buddenhagen (1986) pointed out that results from disease screening tests, which relied on






80

seedlings, small rhizomes or in i'dro meristemn plants grown in containers and subjected to artificial inoculation, did not always correlate with field tests.

Additionally, in sitit disease screening in regions where the pathogen is not endemic is limited to isolates found within the country. The Honduras field study demonstrated that the mtDNA haplotype and a nuclear DNA fingerprinting probe can aid in situ screening for resistance to Fusarium wilt. Using molecular techniques for assessing the pathogen population at a given site and determining the lineage of isolates infecting particular host genotypes can increase the effectiveness of screening programs and assist in the evaluation of new breeding lines. To date, disease screening programs primarily have relied on visual assessment to determine disease incidence on new breeding material. Using molecular techniques to characterize isolates of F. oxysporurn f. sp. cubense infecting banana breeding lines would enable breeders to determine host resistance and susceptibility based on the lineage of the pathogen. This could aid in directing breeding programs toward developing resistance to all lineages of the pathogen rather than the three races that are known to infect bananas. This would provide a more comprehensive approach to~breeding for resistance to this pathogen. Finally, the optimal disease screening methodology requires that the isolates that are being used encompass the range of genetic diversity present in the entire forma specialis. This could be accomplished by screening breeding materials with representatives of each of the 10 lineages of F. oxysporum f. sp. ciubense described in this research. Ideally, this would occur in a field in which all lineages of the pathogen could be introduced.













APPENDIX A
EXPLORING THE POTENTIAL OF USING POLYMERASE CHAIN REACTION METHODOLOGIES IN PHYLOGENETIC STUDIES OF FUSARIUMOXYSPORUM F. SP. CUBENSE Introduction

Randomly amplified polymorphic DNAs (RAPDs) have been used as molecular markers in a number of fungal species to distinguish among genetically different isolates, to aid in genetic mapping and for DNA fingerprinting. Randomly amplified polymorphic DNAs can be generated quickly using small quantities of DNA which do not need to be highly purified. However, RAPD techniques are prone to artifactual variation, which can result in erroneous data (Ellsworth et al., 1993). The possibility of artifactual variation prevents the direct use of the technique in genetic diversity studies aimed at describing the genetic relatedness among isolates. Similarly, the inability to determine directly if bands of similar molecular size represent similar amplification products presents a problem in scoring individuals in genetic diversity studies.

Sequencing of both the nuclear and mitochondrial rRNA genes, as well as the

internal transcribed spacer region (ITS) region, have been described for fungi and used in phylogenetic studies (Bruns et al., 1991). The ITS region and intergenic spacer of the nuclear rRNA repeat units evolve the fastest compared to other regions of the rRNA genes and may vary among species and populations (White et al., 1990). The ITS primers make use of the conserved regions of the 18S, 5.8S and 28S rRNA genes to amplify the


81






82
noncoding regions between them (White et al., 1990). The primers are based on sequences from Saccharomyces cerevisiae, Dictyosteliurn discoideun and Stylonicha pustulata (Dams et al., 1988). A diagram of these primers and their location along the nuclear rDNA region of Saccharomyces cerevisiae is provided by White et al., (1990).

Restriction fragment length polymorphisms and RAPD markers are advantageous in phylogenetic studies because they are presumed to be random traits, not subject to natural selection. In this study, the ITS regions were amplified and subjected to restriction enzyme digestion to determine if polymorphisms could be detected. Additionally, RAPD fragments were generated to provide quick identification of putative polymorphic loci. Polymorphic amplification products were labeled and used as probes for Southern hybridization, enabling the confirmation and description of each polymorphism. The objective of this study was to identify polymorphic regions of the genome that could be used in phylogenetic analysis of a large subset of a world collection of Fusarium oxysporun f sp. cubense.

Results and Discussion

ITS Region Analysis

A number of ITS primers described by White et al. (1990), were used in this study to amplify the total genomic DNAs from a subset of isolates representing a large number of VCGs and a pantropical distribution (Table A-i). The amplification protocol for this study, as well as the RAPD analysis, are provided in Table A-2. Internal transcribed spacer primers 2 and 5 were used to amplify the ITS I region of the nuclear rRNA genes and primers ITS4 and 5 were used to amplify the ITS 1, ITS2 and the 5.8S rDNA regions in F. oxysporum f. sp. cubense isolates (for primer sequences see Table A-3). The






83
resultant amplified products were similar for all isolates. These products were then subjected to restriction enzyme digestion using enzymes HaeIII, Sau3AI, and EcoRI. All of the tested isolates had exactly the same restriction pattern for all of the enzymes tested. RAPD Analysis

The identification of potential polymorphic regions of the genome was

accomplished using RAPD analysis following protocols described previously (Welsh and McClelland, 1990; Williams et al., 1990). A number of oligonucleotide primers were screened for their ability to amplify bands in F. oxysporum f sp. cubense (Table A-4). Many of the primers generated heterogenous products which could be used to differentiate isolates. The majority of primers generated numerous bands of varying intensity. A limited number of polymorphic bands were further tested for their potential in identifying polymorphisms using Southern analysis. Southern analysis provided a method for distinguishing whether a missing band observed in a RAPD profile was the result of the presence of multiple alleles, deletions or a problem with the amplification procedure. Randomly amplified polymorphic DNA fragments were generated using 10-base oligonucleotide primers which contained internal hexameric palindromes that may be cut by single restriction endonucleases specific to the sequence (Kit F, Operon Technologies, Inc.). Fragments were subjected to agarose gel electrophoresis, and putative polymorphic regions were identified. Polymorphic amplified products were excised from low melting point agarose gels and labeled with a nonradioactive digoxigenin (dig)-1 1-dUTP using the random primed method (Boehringer Mannheim, Genius System).

Genomic DNA isolated from a subset of 30 isolates representing 15 VCGs and a pantropical distribution (Table A-1) was digested with restriction enzymes with






84
recognition sequences corresponding to those found in the internal hexameric sequences of the oligonucleotide primer. Standard protocols were used for restriction enzyme digestions, blotting, and hybridizations as described in chapters 1 and 2.

When genomic DNAs were amplified with an oligonucleotide containing a BamnHI recognition sequence (GAGGATCCCT-primer F2), a 0.68-kilobase (kb) amplification product was present in isolates 0-1224, STDI, BLUG, S?, B2-1, MW65, STN3, MW1 1, CVA, STNP5, 22994 and STNP 1, but absent in the remaining isolates listed in Table A-1. This fragment, designated BLUG-2-1, was excised from a gel containing the amplification product from isolate Blug in VCG 0124 and labeled as described above. Genomic DNAs from the same isolates were digested with BamHI and probed with BLUG-2-1. A 0.68-kb fragment that hybridizes to BLUG-2-1 was present only in those isolates that amplified a .68-kb fragment by PCR. The same probe has been used on a total of 80 isolates ofF. oxysporum f sp. cubense. The probe appears to be specific for DNAs from isolates in VCGs 0124, 0124/25, 0125, 0128 and 01212.

When genomic DNAs were amplified with an oligonucleotide containing a EcoRV recognition sequence (GGGATATCGG-primer F8), a 0.95-kb fragment was present in isolates 0-1224, STD1, BLUG, S?, B2-1, MW65, STPN3, MW11, CVA, 22994 and STNP1, but absent in isolate STNP5 in VCG0125 and the remaining isolates described in Table A-1. This fragment, designated BLUG-8-1, was excised from a gel containing the amplification product from isolate Blug in VCG 0124 and labeled as described above. Genomic DNAs from the isolates were digested with EcoRV and probed with BLUG-8-1. The probe hybridized to a monomorphic 0.95-kb DNA fragment in all isolates. Additionally, the probe hybridized to a 2.1-kb DNA fragment in isolates 0-1224, STD 1,






85
BLUG, S?, B2-1, MW65, STPN3, MWI 1, CVA, 22994 and STNPL. The PCR amplification using the 10-mer revealed a polymorphism not detected in Southern hybridizations with restricted DNA, which suggests that the polymorphism is due to a point mutation at the F8 primer sequence, but not in the EcoRV recognition site. With the exception of isolate STNP5, the 2.1-kb fragment appears to be specific for DNAs from many isolates in VCGs 0124, 0124/25, 0125, 0128 and 01212.

When genomic DNAs were amplified with an oligonucleotide containing a EcoRI recognition sequence (CCGAATTCCC-primer F5), a 1.6-kb amplification product was present in some isolates, and either absent or less intensively amplified in other isolates. This fragment, designated SH3142-5-1, was excised from a gel containing the amplification product from isolate SH3142 and labeled as described above. Genomic DNAs from the isolates were digested with EcoRI and probed with SH3142-5-1. A 1.6kb fragment was clearly present in isolates Ph6, MW40, ORT2, 1-1, 0-1219, STGM1, F9127, 15638, FCJ7, Pacovan, F9130, DAVAO, A2-1, STB-2, and SH3142. A 1.6-kb PCR amplified fragment was also clearly visible in these isolates. Isolates 0-1224, STD1, S?, STN3, B2-1, MW1 1, CVA, STNP5, 22994 and STNP1, which contained a very faint

1.6-kb amplified product, weakly hybridized to the 1.6-kb labeled probe. In the same isolates, the probe hybridized more strongly to a 4.6-kb fragment. No hybridizations were observed in isolates BLUG and MW65, which did not contain a 1.6 amplification product. A monomorphic 3.0-kb fragment was present in isolates that contained either a 1.6-kb or

4.6-kb fragment.

Informative polymorphic single-copy and multi-copy loci can be identified quickly by PCR using random 10-base oligonucleotide primers. However, a correlation between a






86
polymorphism observed through PCR amplification and a similar polymorphism observed when the DNA of the same isolates is subjected to Southern analysis, does not always occur. This poses a problem for directly interpreting the polymorphisms observed from RAPD analysis in a phylogenetic study.

Polymorphic products generated from single primer amplification may be used as probes for Southern hybridization. Restriction fragment length polymorphisms identified through Southern hybridization can be used to discern genetic relationships among isolates of the world collection. However, some RAPD bands may represent multiple-copy sequences and therefore are not ideal in phylogenetic studies. However, probes that appear to be lineage specific, such as BLUG-2- 1, may provide a quick method for determining the VCG of new isolates and for genetic studies of field populations.

In conclusion, using RAPD analysis for the determination of phylogenetic

relationships among isolates is difficult because in many cases interpretation of the data is ambiguous. Clark and Lanigan (1993) have set forth criteria that should be met when using RAPD analysis to estimate nucleotide divergence of closely related taxa. One criterion identified is that polymorphic bands must be shown to behave as Mendelian factors. Since F. oxysporum f. sp. cubense does not have a known sexual state, Mendelian inheritance can not be proven. Another criterion is that allelism of bands must be ascertained by Southern blotting or segregation analysis. For F. oxysporum f. sp. ciibense, segregation analysis is not an option and Southern blotting in two of the three examples described above did not provide definitive answers to the allelic nature of the polymorphisms present in the isolates.






87

Table A-1. Isolates ofFusarium oxysporumn f sp. cubense used in RAPD analysis and presence or absence of major RAPD bands.


Primer F-2 Primer F-5 Primer F-8
Isolate VCG 0.68-kb band 1.6-kb band 0.95-kb band

0-1219 0120 +
STGM1 0120 +
F9127 0120 +
15638 0120 +
FCJ7 0120 +
Pacovan 0120 +
ORT2 0120 +
F9130 0121 +
Ph6 0122 +
DAVAO 0123 +0-1224 0124 + +* +
STD1 0124 + +* +
BLUG 0124 + +
S? 0124 + +* +
B2-1 0124 + +* +
MW65 0124 + +
STN3 0124 + +* +
MW11 0124/25 + +* +
CVA 0124/25 + +* +
STNP5 0125 + +*
STB-2 0126 +
22994 0128 + +* +
N5331 0129 +
A2-1 01210 +
Al 01210 +
SH3142 01211 +
STNP-1 01212 + +* +
1-1 01213 +
MW40 01214 +


* Weak band present





88
Table A-2. Amplification protocol for RAPD and ITS analysis


25 ng template DNA 15 ng primer DNA (Operon Technologies, San Jos6, CA)
0.3 p1 10mM dATP 0.3 /l 10mM dCTP 0.3 4 10mM dTTP 0.3 l 10mM dGTP
0.2 U Taq polymerase (Boehringer Mannheim, Indianapolis, IN)
1 l 10 mM MgCI,
2.5 ul 10Ox Taq polymerase buffer sterile water to volume of 25 4l



Amplifications were performed in a Gene Machine II thermal cycler. The machine was programmed for one cycle for 30 seconds at 940C followed by 35 cycles of 1 minute at 94C, 1 minute at 350C, and 2 minutes at 72C, followed by a final cycle of 10 minutes at 720C.





89

Table A-3. Primer sequence for the ITS regions


ITS Primerx Sequence

2 GCTGCGTTCTTCATCGATGC
4 TCCTCCGCTTATTGATATGC
5 GGAAGTAAAAGTCGTAACAAGG

"Primers obtained from Dr. Kerry O'Donnell with the sequences based on White et al. (1990).






90
Table A-4. List of oligonucleotide primers tested on Fusarium oxysporum f sp. cubense isolates

Amplification
Operon Primer Recognition Sequence (Yes/No)


BI GTTTCGCTCC Y
B2 TGATCCCTGG Y
B3 CATCCCCCTG Y
B4 GGACTGGAGT Y
B5 TGCGCCCTTC Y
B6 TGCTCTGCCC Y
B7 GGTGACGCAG Y
El CCCAAGGTCC Y
E2 GGTGCGGGAA Y
E3 CCAGATGCAC Y
E5 GTGACATGCC Y
E6 AAGACCCCTC Y
E7 AGATGCAGCC Y
Ell GAGTCTCAGG Y
E 12 TTATCTCCCC Y
E13 CCCGATTCGG Y
E 14 TGCGGCTGAG Y
E15 ACGCACAACC Y
E16 GGTGACTGTG Y
E17 CTACTGCCGT Y
E18 GGACTGCAGA Y
E19 ACGGCGTATG Y
E20 AACGGTGACC Y
Fl ACGGATCCTG Y
F2 GAGGATCCCT Y
F4 GGTGATCAGG Y
F5 CCGAATTCCC Y
F6 GGGAATTCGG Y
F7 CCGATATCCC N
F8 GGGATATCGG Y
FIO GGAAGCTTGG Y
Fll TTGGTACCCC Y
F12 ACGGTACCAG Y
F13 GGCTGCAGAA Y
F14 TGCTGCAGGT N
F15 CCAGTACTCC N
F16 GGAGTACTGG Y
F17 AACCCGGGAA Y
F18 TTCCCGGGTT Y
F19 CCTCTAGACC N
F20 GGTCTAGAGG Y




Full Text

PAGE 1

3+
PAGE 2

&RS\ULJKW E\ 5RVDOLH /\QQ .RHQLJ

PAGE 3

, GHGLFDWH WKLV UHVHDUFK WR P\ IDWKHU DQG PRWKHU ZKR DOZD\V VWUHVVHG WKH LPSRUWDQFH RI HGXFDWLRQ DQG WR P\ KXVEDQG 7RP DQG GDXJKWHU $PDOHDK ZKR SXW ZLWK PH GXULQJ WKLV VWUHVVIXO SHULRG RI P\ OLIH

PAGE 4

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

PAGE 5

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

PAGE 6

$33(1',&(6 $ (;3/25,1* 7+( 327(17,$/ 2) 86,1* 32/<0(5$6( &+$,1 5($&7,21 0(7+2'2/2*,(6 ,1 3+
PAGE 7

/,67 2) 7$%/(6 7DEOH SDJH /LVW RI LVRODWHV WKHLU YHJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*f FORQDO OLQHDJHV WKH FXOWLYDU DQG JHRJUDSKLFDO UHJLRQV IURP ZKLFK WKH\ ZHUH FROOHFWHG 6LPLODULW\ PDWUL[ RI VLPSOH PDWFKLQJ FRHIILFLHQWV EDVHG RQ UHVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV IRU VHOHFWHG LVRODWHV RI )XVDULXP R[\VSRUXP I VS FEHQVH ) R[\VSRUXP I VS O\FRSHUVLFL DQG ) R[\VSRUXP I VS QLYHXP &ORQDO OLQHDJHV R )XVDULXP R[\VSRUXP I VS FXEHQVH LVRODWHV WKHLU JHRJUDSKLFDO GLVWULEXWLRQV DQG FRUUHVSRQGLQJ YHJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*f $ FRPSDULVRQ RI DOOHOLF GDWD IRU LVRODWHV LQ )2& )2& ,, DQG )2& 9,, &ORQHFRUUHFWHG PHDVXUHPHQWV RI JDPHWLF GLVHTXLOLEULXP DPRQJ SDLUV RI DOOHOHV LQ D ZRUOGZLGH FROOHFWLRQ RI )XVDULXP R[\VSRUXP IVS FXEHQVH *HQRW\SHV RI EDQDQD SODQWHG LQ )LHOG )LHOG GHVLJQ RI EDQDQD SODQWV LQ URZV RI WKH UDFH GLVHDVH VFUHHQLQJ SORW ,VRODWHV RI )XVDULXP R[\VSRUXP I VS FXEHQVH FROOHFWHG LQ )LHOG DQG WKHLU LQFLGHQFH RQ GLIIHUHQW KRVW JHQRW\SHV

PAGE 8

/,67 2) ),*85(6 )LJXUH SDJH '1$ RI LVRODWHV ZDV GLJHVWHG ZLWK (FR 5) UHVWULFWLRQ HQ]\PH DQG SUREHG ZLWK FORQH )UHTXHQF\ GLVWULEXWLRQ RI 5)/3 KDSORW\SHV DPRQJ WKH LVRODWHV UHSUHVHQWLQJ D ZRUOG FROOHFWLRQ RI )XVDULXP R[\VSRUXP I VS FEHQVH 0LGSRLQW URRWHG b PDMRULW\ UXOH FRQVHQVXV WUHH UHSUHVHQWLQJ ERRWVWUDS UHSOLFDWHV 0LWRFKRQGULDO '1$ KDSORW\SHV RI VHOHFWHG )XVDULXP R[\VSRUXP I VS FEHQVH LVRODWHV PW'1$ KDSORW\SH RI VHOHFWHG +RQGXUDQ ILHOG LVRODWHV +LQGOOO GLJHVWHG '1$ RI VHOHFWHG +RQGXUDQ ILHOG LVRODWHV WKDW ZHUH SUREHG ZLWK FORQH S(
PAGE 9

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ 3+
PAGE 10

3RSXODWLRQV RI WKH SDWKRJHQ ZHUH VDPSOHG IURP WZR GLVHDVH VFUHHQLQJ ILHOGV LQ +RQGXUDV )LHOG ZDV SODQWHG WR JHQRW\SHV RI EDQDQD DQG LVRODWHV ZHUH UHFRYHUHG IURP KRVW WLVVXH ,VRODWHV UHFRYHUHG LQ )LHOG FRPSULVHG WKUHH PW'1$ KDSORW\SHV DQG WKHLU LQFLGHQFHV DQG GLVWULEXWLRQV YDULHG ZLWK WKH KRVW JHQRW\SH VDPSOHG '1$ ILQJHUSULQWV LQGLFDWHG WKDW WKH LVRODWHV UHSUHVHQWHG ILYH FORQHV )LHOG ZDV SODQWHG WR D VLQJOH JHQRW\SH RI EDQDQD 7KH LVRODWHV FROOHFWHG IURP KRVW WLVVXH UHSUHVHQWHG D VLQJOH PW'1$ KDSORW\SH DQG FORQH 7KH IUHTXHQFLHV ZLWK ZKLFK WKH SDWKRJHQ FORQHV LQIHFWHG WKH KRVW ZHUH GHSHQGHQW XSRQ KRVW JHQRW\SH 7ZR UDFHV RI WKH SDWKRJHQ UDFH DQG f KDYH EHHQ UHSRUWHG LQ +RQGXUDV 6LQFH WKH FXOWLYDU +LJKJDWH LV VXVFHSWLEOH WR UDFH EXW UHVLVWDQW WR UDFH LVRODWHV UHFRYHUHG IURP WKH FXOWLYDU +LJKJDWH ZHUH DVVXPHG WR UHSUHVHQW UDFH ,VRODWHV ZLWKLQ D OLQHDJH DVVRFLDWHG ZLWK UDFH LVRODWHV ZHUH SUHIHUHQWLDOO\ UHFRYHUHG IURP +LJKJDWH DQG WKH )XQGDFLQ +RQGXUHD GH ,QYHVWLJDFLQ $JUFROD )+,$f K\EULGV ,VRODWHV UHSUHVHQWLQJ D VHFRQG OLQHDJH ZHUH UHFRYHUHG DW D ORZ IUHTXHQF\ IURP +LJKJDWH VXJJHVWLQJ WKDW WKHVH WZR OLQHDJHV GLIIHU LQ WKHLU DELOLW\ WR LQIHFW EDQDQDV VXVFHSWLEOH WR UDFH $VVHVVLQJ WKH SDWKRJHQ SRSXODWLRQV DQG GHWHUPLQLQJ WKH OLQHDJHV RI LVRODWHV LQIHFWLQJ SDUWLFXODU EDQDQD EUHHGLQJ OLQHV FRXOG EH XVHIXO IRU EDQDQD EUHHGLQJ SURJUDPV [

PAGE 11

&+$37(5 ,1752'8&7,21 7KH JHQXV )XVDULXP /LQNf LV FRPSULVHG RI IXQJL ZKLFK KDYH HYROYHG IURP DQ DQFLHQW DQFHVWRU ZKRVH JHRORJLFDO DJH KDV EHHQ SRVWXODWHG WR GDWH EDFN DW OHDVW PLOOLRQ \HDUV 6Q\GHU f :KHWKHU RU QRW WKLV LV DQ DFFXUDWH HVWLPDWH RI WKH DJH RI WKLV LPSRUWDQW IXQJDO JURXS WKH JHQHWLF GLYHUVLW\ SUHVHQW LQ PHPEHUV RI WKH JHQXV )XVDULXP LV WKH UHVXOW RI HYROXWLRQDU\ SURFHVVHV ZKLFK KDYH GHYHORSHG RYHU D ORQJ WLPH VSDQ )XVDULXP R[\VSRUXP 6FKOHFKWf 6\QG t +DQV LV DQ XELTXLWRXV VRLO ERUQH IXQJXV WKDW LV FRPSULVHG RI VWUDLQV WKDW DUH HFRORJLFDOO\ DGDSWHG WR GLYHUVH HQYLURQPHQWV DQG ZKLFK KDYH WKH DELOLW\ WR XWLOL]H DQ DUUD\ RI VXEVWUDWHV LQ WKH VRLO %XUJUHVV f 7KH UHSURGXFWLYH VXFFHVV RI VWUDLQV ZLWKLQ WKLV VSHFLHV GHSHQGV RQ WKH DELOLW\ WR SURGXFH D ODUJH QXPEHU RI SURSDJXOHV LQ WKH IRUP RI PLFURFRQLGLD RU PDFURFRQLGLD DV ZHOO DV UHVLVWDQW UHVWLQJ VSRUHV UHIHUUHG WR DV FKODP\GRVSRUHV WKDW FDQ HQVXUH ORQJWHUP VXUYLYDO GXULQJ DGYHUVH HQYLURQPHQWDO FRQGLWLRQV 7KH PRUSKRORJLFDO FKDUDFWHULVWLFV RI WKHVH SURSDJXOHV DOVR KDYH EHHQ XVHG E\ :ROOHQZHEHU DQG 5HLQNLQJ f WR LQFOXGH WKH VSHFLHV LQ VHFWLRQ (OHJDQV 1R WHOHRPRUSK KDYH EHHQ GHVFULEHG IRU WKH VSHFLHV 3ODQW SDWKRJHQLF LVRODWHV RI ) R[\VSRUXP FDXVH GLVHDVHV NQRZQ DV YDVFXODU ZLOWV DQG FRQVWLWXWH D VPDOO SRUWLRQ RI WKH VWUDLQV ZLWKLQ WKLV VSHFLHV +RZHYHU EHFDXVH RI WKHLU LPSRUWDQFH WR DJULFXOWXUH 6Q\GHU DQG +DQVHQ f GHYHORSHG WKH FRQFHSW RLIRUPDH VSHFLDOHV WR UHFRJQL]H SODQW SDWKRJHQLF VWUDLQV RI ) R[\VSRUXP WKDW ZHUH PRUSKRORJLFDOO\

PAGE 12

LQGLVWLQJXLVKDEOH IURP VDSURSK\WLF VWUDLQV RI WKH VSHFLHV EXW KDG WKH DELOLW\ WR FDXVH GLVHDVH RQ VSHFLILF KRVWV 2YHU IRUPDH VSHFLDOHV KDYH EHHQ GHVFULEHG .LVWOHU f $OWKRXJK IRUPDH VSHFLDOHV DUH QRW UHFRJQL]HG E\ WKH ,QWHUQDWLRQDO &RGH RI %RWDQLFDO 1RPHQFODWXUH WKH\ DUH XVHIXO WR SODQW SDWKRORJLVWV EHFDXVH WKH\ SURYLGH LQIRUPDO JURXSLQJV RI PRUSKRORJLFDOO\ LQGLVWLQJXLVKDEOH LVRODWHV ,VRODWHV ZLWKLQ D IRUPD VSHFLDOLV PD\ EH SDWKRJHQLF RQ D VSHFLILF KRVW VXFK DV ) R[\VSRUXP I VS SL VL ZKLFK FDXVHV D ZLOW GLVHDVH RQ SHDV 3L VXP VDWLYXP / RU PD\ KDYH D EURDGHU KRVW UDQJH VXFK DV ) R[\VSRUXP I VS YDVLQIHFWXP $WNf 6Q\G t +DQV ZKLFK LV SDWKRJHQLF RQ SODQWV LQ WKH IDPLOLHV 0DOYDFHDH 6RODQDFHDH DQG )DEDFHDH :LQGHOV f 7ZR DVVXPSWLRQV WKDW KDYH EHHQ PDGH DERXW LVRODWHV ZLWKLQ D IRUPD VSHFLDOLV DUH WKDW f WKH\ DUH JHQHWLFDOO\ PRUH VLPLODU WR HDFK RWKHU WKDQ WR LVRODWHV RI GLIIHUHQW IRUPDH VSHFLDOHV DQG f WKDW IRUPDH VSHFLDOHV DUH PRQRSK\OHWLF JURXSLQJV .LVWOHU f )XVDULXP ZLOW RI EDQDQD FRPPRQO\ UHIHUUHG WR DV 3DQDPD GLVHDVH LV FDXVHG E\ ) R[\VSRUXP 6FKOHFKWHQG )U I VS FEHQVH () 6Pf :& 6\QGHU t + 1 +DQV 7KLV IXQJXV V\VWHPLFDOO\ FRORQL]HV WKH [\OHP WLVVXH RI VXVFHSWLEOH EDQDQD FXOWLYDUV GLVUXSWLQJ QRUPDO ZDWHU XSWDNH %HFNPDQ f 7KLV GLVUXSWLRQ HYHQWXDOO\ UHVXOWV LQ ZLOW DQG WKH VXEVHTXHQW GHDWK RI WKH SODQW 6WXGLHV GHVFULEHG E\ %HFNPDQ f LQGLFDWH WKDW FRQLGLD LQ DTXHRXV VROXWLRQV DUH UHDGLO\ WDNHQ XS LQWR WKH YDVFXODU HOHPHQWV DQG FDUULHG LQ WKH WUDQVSLUDWLRQ VWUHDP WR VLWHV RI HQWUDSPHQW DW YHVVHO HQGLQJV 6XEVHTXHQW VHFRQGDU\ FRORQL]DWLRQ RFFXUV LQ VXVFHSWLEOH FXOWLYDUV E\ WKH GLVSHUVDO RI FRQLGLD WKURXJK WKH YDVFXODU V\VWHP 5HVLVWDQW FXOWLYDUV SUHYHQW VSUHDG E\ WKH IRUPDWLRQ RI RFFOXGLQJ JHOV DQG W\ORVHV WKDW ZDOO RII WKH OXPHQD RI WKH LQIHFWHG YHVVHOV

PAGE 13

7R GDWH WKUHH UDFHV RI WKH SDWKRJHQ DQG KDYH EHHQ GHVFULEHG ZKLFK DUH SDWKRJHQLF RQ EDQDQD WKHVH UDFHV DUH GLIIHUHQWLDWHG E\ WKH EDQDQD FXOWLYDUV *URV 0LFKHO %OXJJRH DQG &DYHQGLVK UHVSHFWLYHO\ +LVWRULFDOO\ WKLV SDWKRJHQ KDV FDXVHG ODUJH VFDOH GLVHDVH HSLGHPLFV ZKHQ VXVFHSWLEOH EDQDQD FXOWLYDUV KDYH EHHQ JURZQ LQ WKH SUHVHQFH RI WKH SDWKRJHQ 3ORHW] f 7KH ILUVW UHFRUGLQJ RI )XVDULXP ZLOW RI EDQDQD ZDV PDGH LQ VRXWKHDVW 4XHHQVODQG LQ RQ WKH FXOWLYDU 6LON NQRZQ ORFDOO\ DV 6XJDU %DQFURIW f 7KH KLVWRU\ RI WKLV GLVHDVH JHQHUDOO\ SDUDOOHOHG WKH GLVVHPLQDWLRQ RI VXVFHSWLEOH JHQRW\SHV WR QHZ DUHDV 7KH HGLEOH EDQDQDV DUH DVH[XDOO\ SURSDJDWHG W\SLFDOO\ XVLQJ UKL]RPHV RU VXFNHU SODQWV 2FFDVLRQDOO\ LQIHVWHG UKL]RPHV ZHUH EURXJKW WR QHZ SODQWDWLRQV DQG WKLV LV OLNHO\ KRZ WKH SDWKRJHQ ZDV ZLGHO\ GLVWULEXWHG ,QWHUHVWLQJO\ RQH UHJLRQ ZKHUH WKH SDWKRJHQ KDV EHHQ IRXQG RQO\ UHFHQWO\ LV WKH YDVW DUHD RI 2FHDQLD ZKHUH VXVFHSWLEOH FXOWLYDUV KDYH EHHQ JURZQ IRU PLOOHQQLD 6KLYDV DQG 3KLOHPRQ f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f WKH V LQ 6RXWK $IULFD WKH V LQ $XVWUDOLD DQG WKH V LQ 7DLZDQ 3ORHW] HW DO f ,Q WKH ODWH V RXWEUHDNV RQ &DYHQGLVK ZHUH REVHUYHG LQ WKH 0LQGLQDR SURYLQFH RI WKH 3KLOLSSLQHV +RZHYHU LQ FRQWUDVW WR RXWEUHDNV LQ WKH VXEWURSLFV GLVHDVH RXWEUHDNV LQ WKH 3KLOLSSLQHV RFFXUUHG LQ ORFDOL]HG SDWFKHV WKDW UQ

PAGE 14

WHQGHG QRW WR VSUHDG 3ORHW] HW DO f $OO LVRODWHV LQ 9&* DUH IURP WKH 3KLOLSSLQHV DQG WKH\ DOO KDYH EHHQ UHFRYHUHG IURP &DYHQGLVK FXOWLYDUV ,Q WKH ODWH V 6X HW DO f FRQFOXVLYHO\ GHPRQVWUDWHG WKDW LVRODWHV FDSDEOH RI FDXVLQJ GLVHDVH RQ &DYHQGLVK ZHUH GLIIHUHQW IURP UDFH DQG LVRODWHV 7KHVH LVRODWHV ZKLFK DOVR FDXVH GLVHDVH RQ UDFH DQG UDFH VXVFHSWLEOH FXOWLYDUV ZHUH GHVLJQDWHG UDFH ,Q DGGLWLRQ WR HFRQRPLF ORVVHV LQ WKH H[SRUW WUDGH 3DQDPD GLVHDVH LV D VHULRXV SUREOHP RQ ORFDOO\ FRQVXPHG GHVVHUW EDQDQDV DQG FRRNLQJ EDQDQDV 3ORHW] 3ORHW] HW DO f 2QO\ b RI WKH EDQDQDV WKDW DUH SURGXFHG LQ D JLYHQ \HDU DUH H[SRUWHG 7KH UHPDLQLQJ b PLOOLRQ PHWULF WRQV LQ f DUH FRQVXPHG ORFDOO\ )$2 f DQG DUH RIWHQ VWDSOH IRRGV IRU SRRU VXEVLVWHQFH IDUPHUV LQ WKH WURSLFV ,1,%$3 f ,Q WRWDO WKH LPSDFW RI )XVDULXP ZLOW RQ WKHVH EDQDQDV LV IDU JUHDWHU WKDQ RQ EDQDQDV SURGXFHG IRU H[SRUW 7KH +RVW 7KH ) R[\VSRUXP I VS FEHQVHn EDQDQD SDWKRV\VWHP OLNHO\ GHYHORSHG IURP D SURFHVV RI FRHYROXWLRQ EHWZHHQ WKH KRVW DQG SDWKRJHQ UHVXOWLQJ LQ FRPSOHPHQWDU\ HYROXWLRQDU\ FKDQJHV LQ ERWK RUJDQLVPV %DQDQDV EHORQJ WR WKH RUGHU =LQJLEHUDOHV ZKLFK FRQVLVWV RI HLJKW IDPLOLHV RI WURSLFDO DQG VXEWURSLFDO UKL]RPDWRXV SHUHQQLDO SODQWV WKDW LQKDELW PRLVW IRUHVWHG DUHDV RU LQYDGH GLVWXUEHG VLWHV .UHVV f %DQDQDV EHORQJ WR WKH IDPLO\ 0XVDFHDH ZKLFK FRQWDLQ WKUHH JHQHUD 0XVD 0XVHOOD DQG (QVHWHf ZKRVH QDWXUDO UDQJH H[WHQGV IURP :HVW $IULFD WR WKH 3DFLILF (GLEOH EDQDQDV DUH WKH PRVW LPSRUWDQW FXOWLYDWHG FURS RI WKLV IDPLO\ 7ZR DGGLWLRQDO PLQRU FURSV RI HFRQRPLF

PAGE 15

LPSRUWDQFH DUH DEDF£ 0 WH[WLOLVf D ILEHU FURS DQG FXOWLYDWHG IRUPV RI (QVHWH YHQWULFRVXP D IRRG DQG ILEHU SODQW RI XSODQG $E\VVLQLD (GLEOH EDQDQDV KDYH EHHQ GRPHVWLFDWHG IURP WZR VSHFLHV 0 DFXPLQDWD DQG 0 EDOELVLDQD 6LPPRQGV DQG 6KHSKHUG f XVHG SKHQRW\SH FKDUDFWHUV DQG VFRUHG HDFK RQ WKH VFDOH RI DFFXPLQDWD H[SUHVVLRQf WR EDOELVLDQD H[SUHVVLRQf WR GHYHORS D V\VWHP WR FODVVLI\ HGLEOH EDQDQDV )URP WKLV WKH JHQHWLF FRQVWLWXWLRQ RI DQ\ HGLEOH EDQDQD FDQ EH VSHFLILHG XVLQJ D JHQRPH IRUPXOD LQGLFDWLQJ WKH FRQWULEXWLRQV IURP 0 DFFXPLQDWD $$f DQG 0 EDOELVLDQD %%f (GLEOH 0 DFXPLQDWD FXOWLYDUV DUH HLWKHU GLSORLG $$f RU WULSORLG $$$f DQG WKHLU FHQWHU RI RULJLQ LQFOXGHV 6RXWKHUQ 7KDLODQG 6XPDWUD -DYD %RUQHR %XUPD DQG 0DOD\VLD ZKLFK LV DOVR WKH SULPDU\ FHQWHU RI GLYHUVLW\ IRU WKH VSHFLHV 7KLV VSHFLHV LV PRVW GLYHUVH LQ SHQLQVXODU 0DOD\VLD DQG WKH HTXDWRULDO LVODQGV ZKHUH WKH ZLOG VSHFLHV LV QDWLYH 6L[ VXEVSHFLHV RI 0 DFXPLQDWD DUH UHFRJQL]HG $OWKRXJK WKH EDQDQD FORQH *URV 0LFKHO KDV SKHQRW\SLF FKDUDFWHUV WKDW VXJJHVW GHVFHQW IURP 0 DFXPLQDWD VXEVS PDODFFHQVLV 6WRYHU f WKH SDUHQWV RI PRVW FORQHV DUH QRW NQRZQ 6LPPRQGV f DUJXHV WKDW HGLELOLW\ LQ EDQDQDV DURVH LQ 0DOD\VLD WKURXJK SDUWKHQRFDUS\ LQ 0 DFXPLQDWD GLSORLGV IROORZHG E\ WULSORLG\ $$$f 7KURXJK WKH SURFHVV RI GRPHVWLFDWLRQ PDQ KDV VHOHFWHG IRU ERWK SDUWKHQRFDUS\ DQG VWHULOLW\ ZKLFK DUH JHQHWLFDOO\ LQGHSHQGHQW FKDUDFWHULVWLFV $GGLWLRQDOO\ GRPHVWLFDWLRQ KDV OHG WR VHOHFWLRQ RI WULSORLG\ EHFDXVH LW FRQWULEXWHV WR KLJK SODQW YLJRU DQG IUXLW JURZWK UDWH DQG FRQIHUV JDPHWLF VWHULOLW\ 6LPPRQGV f ,Q FRQWUDVW 0 EDOELVLDQD KDV D FHQWHU RI RULJLQ LQ D UHJLRQ LQFOXGLQJ WKH FRXQWULHV RI ,QGLD %XUPD DQG 1RUWKHUQ 7KDLODQG ,W LV OHVV JHQHWLFDOO\ GLYHUVH WKDQ 0 DFXPLQDWD

PAGE 16

,W LV SUHVXPHG WKDW HGLELOLW\ GLG QRW DULVH LQ 0 EDELVLDWLD EHFDXVH QR QDWXUDOO\ RFFXUULQJ SDUWKHQRFDUSLF FORQHV KDYH EHHQ LGHQWLILHG 6LPLODUO\ QR SXUHO\ 0 EDOELVLDQD WULSORLGV RU WHWUDSORLGV DUH NQRZQ 0 EDOELVLDQD LV QDWXUDOO\ FURVV IHUWLOH ZLWK D QXPEHU RI 0XVD VSHFLHV LQFOXGLQJ 0 DFXPLQDWD EXW LW LV UHSURGXFWLYHO\ TXLWH LVRODWHG 6LPPRQGV f 7KH K\EULG JURXSV $% $$% $%%f DUH SHULSKHUDO EHLQJ GLYHUVLILHG LQ ,QGLD DQG WR D OHVVHU H[WHQW LQ ,QGRFKLQD DQG HDVWHUQ 0DOD\VLDQ DUHDV IURP ZKLFK ZLOG 0 DFXPLQDWD LV ORFDOO\ DEVHQW EXW LQ ZKLFK 0 EDOELVLDQD LV QDWLYH 6LPPRQGV f 7KH WUDQVSRUW E\ KXPDQV RI $$W\SH EDQDQDV LQWR DUHDV RFFXSLHG E\ ZLOG 0 EDOELVLDQD LV LQIHUUHG WR KDYH SURYLGHG WKH RSSRUWXQLW\ IRU LQWHUVSHFLILF K\EULGL]DWLRQ 6LPPRQGV f &KDUDFWHULVWLFV VXFK DV SODQW KDUGLQHVV GURXJKW UHVLVWDQFH GLVHDVH UHVLVWDQFH WR QHPDWRGHV DQG \HOORZ 6LJDWRND DQG VWDUFKLQHVV RI IUXLW WKDW DUH IRXQG LQ LQWHUVSHFLILF K\EULG EDQDQDV DUH EHOLHYHG WR EH FRQWULEXWHG IURP0 EDOELVLDQD 6LPPRQGV f ,QWHUVSHFLILF K\EULGV EHWZHHQ 0 EDOELVLDQD DQG 0 DFXPLQDWD RFFXU LQ QDWXUH EHFDXVH WKH VSHFLHV DUH V\PSDWULF RYHU D ODUJH DUHD RI VRXWKHDVW $VLD DQG LQWURJUHVVLRQ RFFXUV RIWHQ $ VHFRQGDU\ FHQWHU RI GLYHUVLW\ IRU EDQDQDV LV (DVW $IULFD ,Q (DVW $IULFD PL[WXUHV RI JHQRW\SHV DUH JURZQ IRU ORFDO XVH DQG SDUWLFXODU GLHWDU\ QHHGV 7KH (DVW $IULFDQ EDNLQJ DQG EHHU W\SH EDQDQDV $$$f DUH XQLTXH WR WKH UHJLRQ DQG DUH RIWHQ IRXQG LQ PL[HG SODQWLQJV ZLWK FXOWLYDUV RI H[RWLF RULJLQ VXFK DV $%%V DQG YDULRXV GHVVHUW W\SHV 3ORHW] HW DO f ,QODQG FXOWLYDWLRQ RI EDQDQD LQ (DVW $IULFD LV EDVHG SULPDULO\ RQ WULSORLG FORQHV $$$f EXW WKH FRDVWDO DUHDV KDYH PXFK PRUH GLYHUVH SRSXODWLRQV &RDVWDO (DVW $IULFD LV WKH RQO\ ORFDWLRQ RXWVLGH RI 6RXWKHDVW $VLD ZKHUH WKHUH LV DQ DVVHPEODJH RI HGLEOH GLSORLGV FXOWLYDUV DQG H[WHQVLYH ORFDO GLYHUVLILFDWLRQ 7KH PRVW VWULNLQJ

PAGE 17

DFFXPXODWLRQV RI XQLTXH FORQHV DUH LQ XSODQG (DVW $IULFD DQG WKH 3DFLILF ,VODQGV ZKHUH LW DSSHDUV WKDW VXEVLVWHQFH IDUPHUV DFWLYHO\ SUHVHUYHG WKHVH FORQHV 6LPPRQGV f 0XUGRFNfV f GLVFXVVLRQ RI SODQW LQWURGXFWLRQ LQ UHODWLRQ WR $IULFDQ KLVWRU\ VXJJHVWV WKDW 0DOD\VLDQ IRRG SODQWV HQWHUHG $IULFD LQ WKH KDQGV RI ,QGRQHVLDQ PLJUDQWV DERXW %3 %DQDQDV DV ZHOO DV WDUR \DPV EUHDGIUXLW FRFRQXW VRPH ULFHV DQG VXJDUFDQH DUH DGDSWHG WR PRLVW FOLPDWHV DQG EHFDPH VWDSOH IRRGV RI WKH ZHWWHU SDUWV RI &HQWUDO $IULFD 6LPPRQGV f 0DGDJDVFDU DSSHDUV WR EH WKH SULQFLSDO SRLQW RI HQWU\ RI EDQDQDV LQWR $IULFD 6LPPRQGVn f YLHZ LV WKDW WKH HDUO\ $IULFDQ FXOWLYDUV UHDFKHG 0DGDJDVFDU IURP 0DOD\VLD ODWH LQ WKH ILUVW KDOI RI WKH ILUVW PLOOHQQLXP $' )URP :HVW $IULFD 3RUWXJXHVH WUDYHOHUV WRRN SODQWV WR WKH &DQDU\ ,VODQGV HDUO\ LQ WKH ILIWHHQWK FHQWXU\ DQG D OHDVW RQH FORQH WUDYHOHG WR +LVSDQLROD LQ 6LPPRQGV f EHOLHYHG WKDW WKLV UHSUHVHQWHG WKH ILUVW RI PDQ\ LQWURGXFWLRQV PDGH LQWR WKH 1HZ :RUOG +RZHYHU /DQJGRQ f VXJJHVWV EDVHG RQ OLQJXLVWLF ERWDQLFDO DQG DUFKHRORJLFDO GDWD WKDW EDQDQD ZDV SUHVHQW LQ WKH 1HZ :RUOG EHIRUH WKH 6SDQLDUGV DUULYHG 3UHVHQWO\ WKH RQO\ ORFDWLRQ ZLWK D KLJK OHYHO RI JHQHWLF GLYHUVLW\ RI 0 DFXPLQDWD LV 3DSXD 1HZ *XLQHD DQG VXUURXQGLQJ LVODQGV +HUH WKH LQGLJHQRXV SHRSOH KDYH SUHVHUYHG DQ LQWDFW SRSXODWLRQ RI SULPLWLYH FXOWLJHQV D UHOLF RI DQ HDUO\ VWDJH LQ WKH HYROXWLRQ RI WKH JURXS -DUUHW f XVHG SK\ORJHQHWLFDOO\ LQIRUPDWLYH UHVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV 5)/3Vf WR FKDUDFWHUL]H HLJKW VSHFLHV UHSUHVHQWLQJ WKUHH VHFWLRQV RI 0XVD IURP 3DSXD 1HZ *XLQHD DV ZHOO DV RQH VSHFLHV RI (QVHWH 7KH VWXG\ IRXQG RQO\ D GLVWDQW UHODWLRQVKLS EHWZHHQ 0 EDOELVLDQD DQG RWKHU VSHFLHV RI 0XVD 0 EDOELVLDQD ZDV QRW LQ WKH VDPH FODGH DV WKH RWKHU VSHFLHV ZLWKLQ WKH VHFWLRQ 0XVD DOVR

PAGE 18

UHIHUUHG WR DV (XPXVDf 7KH GLSORLG ODQG UDFHV IURP WKLV VHFWLRQ FOXVWHUHG ZLWK0 DFXPLQDWD $OO DFFHVVLRQV RI 0 DFXPLQDWD VKDUHG D ODUJH QXPEHU RI DOOHOHV *HQHWLF &KDUDFWHUL]DWLRQ RI )XVDULXP R[\VSRUXP I VS FEHQVH &ODGLVWLF DQDO\VHV RI WKH SK\ORJHQHWLF UHODWLRQVKLSV DPRQJ LVRODWHV RI D IXQJDO VSHFLHV DUH EDVHG RQ WKH FRQFHSW WKDW PHPEHUV RI D VSHFLHV KDYH HYROYHG IURP D FRPPRQ DQFHVWRU 6FRWODQG f 7KLV DQFHVWRU SRVVHVVHG SULPLWLYH JHQHWLF FKDUDFWHULVWLFV ZKLFK PD\ KDYH FKDQJHG WKURXJK WKH HYROXWLRQDU\ KLVWRU\ RI WKH RUJDQLVP 'HULYHG FKDUDFWHU FKDQJHV WKRVH FKDQJHV WKDW DUH GLIIHUHQW IURP WKH DQFHVWUDO W\SH DUH WKH UHVXOW RI HYROXWLRQDU\ SURFHVVHV ZKLFK FRQVLVW RI ERWK JHQHWLF PXWDWLRQ UHFRPELQDWLRQ DQG WUDQVSRVRQ UHDUUDQJHPHQWVf DQG VSDWLDO PLJUDWLRQ GLYHUJHQFH DQG FRQYHUJHQFHf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f $GGLWLRQDOO\ YHJHWDWLYH FRPSDWLELOLW\

PAGE 19

JURXSV 9&*Vf EDVHG RQ ZRUN RQ )XVDULXP PRQLOLIRUPH DOVR KDYH EHHQ XVHG WR GLIIHUHQWLDWH VLPLODU LVRODWHV RI ) R[\VSRUXP 3XKDOOD DQG 6SLHWK f ,VRODWHV RI ) R[\VSRUXP ZKLFK VKDUH LGHQWLFDO DOOHOHV DW WKH ORFL JRYHUQLQJ KHWHURNDU\RQ LQFRPSDWLELOLW\ FRPPRQO\ UHIHUUHG WR DV KHW RU YLH DUH YHJHWDWLYHO\ FRPSDWLEOH &RQYHQWLRQDOO\ WKLV LV GHWHUPLQHG YLD WKH DELOLW\ RI QLWUDWHQRQXWLOL]LQJ QLWf DX[RWURSKLF PXWDQWV WR FRPSOHPHQW RQH DQRWKHU IRU QLWUDWH XWLOL]DWLRQ /HVOLH f 6LQFH ) R[\VSRUXP KDV QR NQRZQ VH[XDO VWDJH LW LV QRW SRVVLEOH WR GHWHUPLQH JHQHWLFDOO\ WKH QXPEHU RI ORFL LQYROYHG LQ YHJHWDWLYH LQFRPSDWLELOLW\ +RZHYHU JHQHWLF VWXGLHV SHUIRUPHG RQ ) PRQLOLIRUPH WHOHRPRUSK *LEEHUHOOD IXMLNXURLf LQGLFDWH WKDW D PLQLPXP RI YLH ORFL JRYHUQ YHJHWDWLYH FRPSDWLELOLW\ LQ WKLV FORVHO\ UHODWHG VSHFLHV 3XKDOOD DQG 6SLHWK f 8QIRUWXQDWHO\ WKH UHODWLRQVKLS DPRQJ LVRODWHV FRPSULVLQJ GLIIHUHQW 9&*V LV LPSRVVLEOH WR GLVFHUQ EDVHG VROHO\ RQ WKLV WHFKQLTXH EHFDXVH D VLQJOH FKDQJH RI DOOHOHV DW DQ\ RI WKH YLH ORFL ZLOO JHQHUDWH D QHZ 9&* 9HJHWDWLYH FRPSDWLELOLW\ JURXS GLYHUVLW\ YDULHV DPRQJ WKH NQRZQ IRUPDH VSHFLDOHV RI ) R[\VSRUXP )RU H[DPSOH DPRQJ LVRODWHV RI ) R[\VSRUXP I VS O\FRSHUVLFL (OLDV DQG 6FKQHLGHU f LGHQWLILHG RQH PDMRU DQG WZR PLQRU 9&*V DV ZHOO DV D QXPEHU RI VLQJOHPHPEHU 9&*V ,Q FRQWUDVW D ZRUOGZLGH FROOHFWLRQ RI LVRODWHV RI ) R[\VSRUXP I VS YDVLQIHFWXP LVRODWHV FRPSULVHG 9&*V DQG LVRODWHV EHORQJLQJ WR GLVWLQFW UDFHV RI WKH SDWKRJHQ ZHUH QHYHU LQ WKH VDPH 9&* )HUQDQGH] HW DO f &XUUHQWO\ DW OHDVW 9&*V KDYH EHHQ GHVFULEHG IRU ) R[\VSRUXP I VS FEHQVH 3ORHW] f 7KH PDMRULW\ RI 9&*V DSSHDU WR KDYH OLPLWHG JHRJUDSKLFDO GLVWULEXWLRQ +RZHYHU WKH PDMRULW\ RI LVRODWHV EHORQJ WR D IHZ PDMRU 9&*V WKDW DUH JHRJUDSKLFDOO\ ZLGHVSUHDG DQG FRQWDLQ PRUH WKDQ RQH UDFH RI WKH SDWKRJHQ 3ORHW] f

PAGE 20

8QWLO WKH DGYHQW RI WLVVXH FXOWXUH WKH UKL]RPH RI WKH EDQDQD SODQW ZDV XVHG WR SURSDJDWH EDQDQD ,W LV OLNHO\ WKDW WKH ZRUOGZLGH PRYHPHQW RI SODQWLQJ VWRFN UKL]RPHV DQG DVVRFLDWHG VRLO SDUWLFOHVf IURP UHJLRQ WR UHJLRQ LV UHVSRQVLEOH IRU PXFK RI WKH JHRJUDSKLFDO GLVWULEXWLRQ RI WKH SDWKRJHQ RXWVLGH LWV UHJLRQ RI RULJLQ ,Q WHUPV RI HYROXWLRQ WKH GLYHUVLW\ RI WKH SDWKRJHQ LQ DUHDV RXWVLGH RI LWV FHQWHU RI RULJLQ ZLOO GHSHQG RQ WKH QXPEHU RI \HDUV WKH SDWKRJHQ KDV EHHQ SUHVHQW LQ WKH QHZ ORFDWLRQ DQG ZKHWKHU VLQJOH RU PXOWLSOH FORQDO OLQHDJHV ZHUH LQWURGXFHG )RU H[DPSOH RQH PLJKW H[SHFW WKH SRSXODWLRQV RI ) R[\VSRPP I VS FEHQVH LQ (DVW $IULFD WR EH PRUH JHQHWLFDOO\ YDULDEOH WKDQ SRSXODWLRQV LQ WKH 86$ EHFDXVH WKH FURS ZDV LQWURGXFHG LQWR (DVW $IULFD WKRXVDQGV RI \HDUV SULRU WR WKH LQWURGXFWLRQV WR )ORULGD 0DQ\ HDUOLHU JHQHWLF FKDUDFWHUL]DWLRQV RI YDULRXV IRUPDH VSHFLDOHV RI ) R[\VSRUXP IRFXVHG RQ D VPDOO QXPEHU RI LVRODWHV ZLWK OLPLWHG JHRJUDSKLFDO GLVWULEXWLRQ 7KHVH VWXGLHV JDYH DQ LQFRPSOHWH SLFWXUH RI JHQHWLF GLYHUVLW\ ZKHQ PRUH H[WHQVLYH WHVWLQJ XVLQJ D ODUJHU QXPEHU RI LVRODWHV ZLWK D ODUJH JHRJUDSKLFDO GLVWULEXWLRQ ZDV SHUIRUPHG )RU H[DPSOH ZKHQ LVRODWHV RI WKH SDWKRJHQ ) R[\VSRUXP I VS GLDQWKL ZHUH FROOHFWHG IURP FXOWLYDUV DW GLIIHUHQW VLWHV LQ ,VUDHO DOO RI WKH LVRODWHV EHORQJHG WR WKH VDPH 9&* .DWDQ HW DO f +RZHYHU ZKHQ LVRODWHV RI WKH VDPH SDWKRJHQ ZHUH VDPSOHG IURP D QXPEHU RI JHRJUDSKLFDO ORFDWLRQV LQFOXGLQJ ,WDO\ 6SDLQ 7KH 1HWKHUODQGV WKH 8QLWHG 6WDWHV DQG *HUPDQ\ HLJKW 9&*V ZHUH LGHQWLILHG $ORL DQG %DD\HQ f 7KH ODWWHU UHVHDUFKHUV IRXQG QRW RQO\ VLQJOH UDFHV ZLWKLQ VLQJOH 9&*V EXW DOVR PXOWLSOH UDFHV ZLWKLQ VLQJOH 9&*V $ORL DQG %DD\HQ f ,VRODWHV RI ) R[\VSRUXP IVS FEHQVH DOVR KDYH EHHQ JURXSHG EDVHG RQ HOHFWURSKRUHWLF NDU\RW\SH UDQGRP DPSOLILHG SRO\PRUSKLF '1$ 5$3'f DQDO\VLV DQG WKH

PAGE 21

SURGXFWLRQ RI YRODWLOHV LQ FXOWXUH %RHKP HW DO f SURSRVHG WZR JURXSLQJV RI LVRODWHV RI ) R[\VSRUXP IVS FEHQVH EDVHG RQ VLPLODULWLHV LQ FKURPRVRPH QXPEHU DQG JHQRPH VL]H ,Q JHQHUDO WKHVH JURXSLQJV DJUHHG ZLWK 5$3' GDWD %HQWOH\ HW DO 6RUHQVHQ HW DO f DQG JURXSV GHWHUPLQHG ZKHQ WKH IRUPDWLRQ RI DOGHK\GHV E\ GLIIHUHQW LVRODWHV ZHUH H[DPLQHG %UDQGHV 0RRUH HW DO DQG 6WRYHU f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f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f ,Q DVH[XDOO\ UHSURGXFLQJ RUJDQLVPV WKHUH LV SUHVXPDEO\ QR FRQWULEXWLRQ WR JHQRW\SLF YDULDELOLW\ WKURXJK PHLRWLF UHFRPELQDWLRQ RU LQWURJUHVVLRQ ,Q JHQHUDO JHQRW\SLF YDULDELOLW\ LQ DVH[XDO IXQJL PD\ EH JHQHUDWHG WKURXJK PXWDWLRQ UDQGRP JHQHWLF GULIW

PAGE 22

SDUDVH[XDOLW\ VRPDWLF UHFRPELQDWLRQ DQG WUDQVSRVDEOH HOHPHQWV $OWKRXJK PXWDWLRQ UDWHV DUH JHQHUDOO\ ORZ SHU ORFXV PXWDWLRQV DUH SHUVLVWHQW DQG RYHU WLPH WKH\ JHQHUDWH PXFK YDULDELOLW\ VXPPHG RYHU DOO ORFL 0F'RQDOG HW DO f 7LEUD\UHQF HW DO f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f DV D JURXS RI LQGLYLGXDOV VKDULQJ D FRPPRQ JHQH SRRO DQG SUHVHQW LQ D OLPLWHG JHRJUDSKLFDO DUHD 3RSXODWLRQ JHQHWLF DQDO\VLV LQYROYHV WKH LGHQWLILFDWLRQ RI JHQHWLF PDUNHUV WKDW DUH XQDPELJXRXV DQG LQIRUPDWLYH >SRO\PRUSKLF@ 0F'RQDOG DQG 0F'HUPRWW f 7KHVH W\SHV RI PDUNHUV KDYH EHHQ XVHG WR FKDUDFWHUL]H SRSXODWLRQV RI D QXPEHU RI IXQJDO VSHFLHV LQFOXGLQJ &U\SKRQHFWULFL SDUDVLWLFD )XVDULXP R[\VSRUXP 6FOHURWLQLD VFOHURWLRPP 3K\WRSKWKRUD LQIHVWDP 0\FRVSKDHUHOOD JUDPLQLFROD &RFKOLREROXV FDUERQXP 3K\WRSKWKRUD PHJDVSHUPD I VS JO\FLQHD 0DJQDSRUWKH JULVHD DQG 6WDJRQRVSRUD QRGRUXP $SSHO DQG *RUGRQ *RRGZLQ HW DO .RK HW DO .RKOL HW DO .RKOL HW DO /HRQDUG DQG /HDWK /HY\ HW DO /HY\ HW DO 0F'RQDOG DQG 0DUWLQH]

PAGE 23

0F'RQDOG HW DO 0F'RQDOG HW DO 6XMNRZVNL HW DO :KLVVRQ HW DO f 3RSXODWLRQ JHQHWLF VWXGLHV RI )XVDULXP R[\VSRUXP I VS PHORQLV KDYH DGGUHVVHG JHQHWLF VWUXFWXUH LQ ERWK DJULFXOWXUDO DQG QDWLYH VRLOV DQG DOVR KDYH H[DPLQHG WKH JHQHWLF UHODWLRQVKLSV DPRQJ SDWKRJHQLF DQG QRQSDWKRJHQLF LVRODWHV E\ FKDUDFWHUL]LQJ WKHLU 9&*V DQG PLWRFKRQGULDO KDSORW\SHV $SSHO DQG *RUGRQ *RUGRQ DQG 2NDPRWR *RUGRQ HW DO f %DVHG RQ SDUVLPRQ\ DQDO\VLV RI WKH PW'1$ GDWD QRQSDWKRJHQLF LVRODWHV IURP 0DU\ODQG ZHUH PRUH FORVHO\ UHODWHG WR ) R[\VSRUXP I VS PHORQLV WKDQ WR WKH &DOLIRUQLD QRQSDWKRJHQV $SSHO DQG *RUGRQ f %DVHG RQ PW'1$ ERWK SDWKRJHQ 9&*V ZHUH PRUH FORVHO\ UHODWHG WR D QRQSDWKRJHQ 9&* WKDQ WR HDFK RWKHU $SSHO DQG *RUGRQ f (VVHQWLDOO\ DOO GLYHUVLW\ LQ PW'1$ ZDV IRXQG DPRQJ WKH QRQSDWKRJHQLF VWUDLQV RI WKH VSHFLHV ZKHUHDV SDWKRJHQLF VWUDLQV ZHUH UHSUHVHQWHG E\ RQO\ D VLQJOH PW'1$ KDSORW\SH DW DQ\ RQH ORFDWLRQ *RUGRQ DQG 2NDPRWR -DFREVRQ DQG *RUGRQ Ef LQGLFDWLQJ WKDW SDWKRJHQLF VWUDLQV DUH OHVV JHQHWLFDOO\ GLYHUVH WKDQ QRQ SDWKRJHQLF VWUDLQV ,Q FRQWUDVW D VWXG\ RI WKH JHQHWLF UHODWHGQHVV DPRQJ VWUDLQV RI ) R[\VSRUXP LQ QDWLYH DQG FXOWLYDWHG VRLOV FRQVLVWHG RI GLIIHUHQW PW'1$ KDSORW\SHV RI ZKLFK WKH ILYH PRVW FRPPRQ KDSORW\SHV DFFRXQWHG IRU b RI WKH LVRODWHV ,VRODWHV UHSUHVHQWLQJ WKHVH KDSORW\SHV ZHUH IRXQG LQ ERWK QDWLYH DQG DJULFXOWXUDO VRLOV *RUGRQ HW DO f 6HYHQW\ WZR SHUFHQW RI WKH LVRODWHV IRXQG LQ FXOWLYDWHG VRLO ZHUH DVVRFLDWHG ZLWK WKH VDPH PW'1$ KDSORW\SH DV RQH RU PRUH RI WKH LVRODWHV LQ WKH QDWLYH VRLO *RUGRQ HW DO f +RVW SRSXODWLRQV FDQ DFW DV SRZHUIXO VHOHFWLYH IRUFHV RQ SDWKRJHQ SRSXODWLRQV DQG YLFH YHUVD 0F'RQDOG HW DO f )RU H[DPSOH WKH ULFH EODVW IXQJXV 0DJQDSRUWKH

PAGE 24

JUVHD FDQ EH GLYLGHG LQWR D ODUJH QXPEHU RI SDWKRW\SHV 3DWKRW\SHV DUH DQDORJRXV WR UDFHV LQ WKDW VSHFLILF SDWKRW\SHV RI D SDWKRJHQ DUH YLUXOHQW RQ FHUWDLQ JHQRW\SHV RI WKH KRVW 8VLQJ D UHSHWLWLYH ILQJHUSULQWLQJ SUREH WKDW VHUYHV DV D JHQHDORJLFDO LQGH[ DPRQJ ULFH EODVW LVRODWHV /HY\ HW DO f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f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

PAGE 25

&+$37(5 )86$5,802;<362580) 63 &8%(16( &216,676 2) $ 60$// 180%(5 2) ',9(5*(17 $1' */2%$//< ',675,%87(' &/21$/ /,1($*(6 ,QWURGXFWLRQ )XVDULXP ZLOW RI EDQDQD FRPPRQO\ UHIHUUHG WR DV 3DQDPD GLVHDVH LV FDXVHG E\ )XVDULXP R[\VSRPP 6FKOHFKWHQG )U I VS FEHQVH () 6Pf :& 6\QGHU t + 1 +DQV ,VRODWHV RI ) R[\VSRPP I VS FEHQVH SUHYLRXVO\ KDYH EHHQ FKDUDFWHUL]HG E\ PRUSKRORJ\ RU ELRFKHPLFDO DQG JHQHWLF PDUNHUV 2Q PRUSKRORJLFDO EDVHV ) R[\VSRPP ZDV LQFOXGHG LQ VHFWLRQ (OHJDQV E\ :ROOHQZHEHU DQG 5HLQNLQJ f 1R WHOHRPRUSK KDV EHHQ GHVFULEHG IRU WKH VSHFLHV 9HJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*Vf KDYH EHHQ XVHG WR FDWHJRUL]H LVRODWHV RI ) R[\VSRPP LQFOXGLQJ )XVDULXP R[\VSRPP I VS FEHQVH %DVHG RQ GDWD IURP RWKHU IXQJL LVRODWHV WKDW VKDUH LGHQWLFDO DOOHOHV DW WKH ORFL JRYHUQLQJ KHWHURNDU\RQ LQFRPSDWLELOLW\ FRPPRQO\ UHIHUUHG WR DV KHW RU YLH DUH YHJHWDWLYHO\ FRPSDWLEOH /HVOLH f &RQYHQWLRQDOO\ WKLV LV GHWHUPLQHG E\ WKH DELOLW\ RI QLWUDWHQRQXWLOL]LQJ QLWf DX[RWURSKLF PXWDQWV WR FRPSOHPHQW RQH DQRWKHU IRU QLWUDWH XWLOL]DWLRQ &RUUHOO HW DO f &XUUHQWO\ DW OHDVW 9&*V KDYH EHHQ GHVFULEHG IRU WKLV IRUPD VSHFLDOLV 3ORHW] DQG &RUUHOO 3ORHW] 3ORHW] HW DO f 7KH PDMRULW\ RI LVRODWHV EHORQJ WR WZR PDMRU 9&*V WKDW KDYH SDQWURSLFDO GLVWULEXWLRQ DQG FRQWDLQ PRUH WKDQ RQH UDFH RI WKH SDWKRJHQ 0LQRU 9&*V ZHUH IRXQG WR KDYH D PRUH OLPLWHG JHRJUDSKLFDO GLVWULEXWLRQ 3ORHW] DQG &RUUHOO 3ORHW] f )RU DQ DVH[XDOO\ UHSURGXFLQJ RUJDQLVP OLNH )

PAGE 26

R[\VSRUXP LW LV JHQHUDOO\ DVVXPHG WKDW LVRODWHV ZLWKLQ D 9&* DUH JHQHWLFDOO\ VLPLODU DQG UHSUHVHQW FORQDO SRSXODWLRQV $SSHO DQG *RUGRQ -DFREVRQ DQG *RUGRQ 7DQWDRXL HW DO f ,Q DGGLWLRQ WR FKDUDFWHUL]LQJ LVRODWHV E\ 9&* LVRODWHV RI ) R[\VSRUXP I VS FEHQVH KDYH EHHQ JURXSHG EDVHG RQ HOHFWURSKRUHWLF NDU\RW\SH UDQGRPO\ DPSOLILHG SRO\PRUSKLF '1$ 5$3'f DQDO\VLV DQG WKH DELOLW\ WR SURGXFH YRODWLOH RUJDQLF FRPSRXQGV LQ FXOWXUH %RHKP HW DO f SURSRVHG WZR JURXSLQJV RI LVRODWHV RI ) R[\VSRUXP I VS FEHQVH EDVHG RQ VLPLODULWLHV LQ FKURPRVRPH QXPEHU DQG JHQRPH VL]H *URXS FRPSULVHG LVRODWHV LQ 9&*V DQG *URXS ,, FRPSULVHG LVRODWHV LQ 9&*V DQG ,Q JHQHUDO WKHVH JURXSLQJV DJUHHG ZLWK 5$3' GDWD %HQWOH\ HW DO 6RUHQVHQ HW DO f :KHQ WKH SUHVHQFH RU DEVHQFH RI 5$3' EDQGV ZHUH WUHDWHG DV ELQDU\ GDWD DQG VXEMHFWHG WR SKHQHWLF DQDO\VLV EDVHG RQ WKH XQZHLJKWHG SDLU JURXS PHWKRG ZLWK DULWKPHWLF PHDQ 83*0$f LVRODWHV LQ 9&*V DQG IRUPHG RQH JURXS DQG LVRODWHV LQ 9&*V DQG IRUPHG D VHFRQG JURXS 6LPLODU PDMRU JURXSV RI LVRODWHV ZHUH HYLGHQW ZKHQ GLIIHUHQWLDWLRQ ZDV EDVHG RQ WKH IRUPDWLRQ RI DOGHK\GHV LQ FXOWXUH %UDQGHV f QRWHG WKDW FHUWDLQ LVRODWHV RI WKH SDWKRJHQ SURGXFHG WKHVH RGRURXV FRPSRXQGV ZKHQ JURZQ RQ VWHDPHG ULFH ZKHUHDV RWKHUV GLG QRW WKH ODWWHU LVRODWHV ZHUH FODVVLILHG DV YDULHW\ LQRGRUDWXP 6WRYHU f H[DPLQHG D ODUJHU VHW RI LVRODWHV IURP WURSLFDO $PHULFD DQG WKH &DULEEHDQ DQG QRWHG WKDW SURGXFWLRQ RI RGRURXV FRPSRXQGV ZDV D FRQVLVWHQW DQG UHSHDWDEOH WUDLW 7KRVH LVRODWHV ZKLFK SURGXFHG WKH DOGHK\GHV ZHUH UHIHUUHG WR DV FXOWLYDU 2GRUDWXP ZKHUHDV WKRVH ZKLFK GLG QRW ZHUH UHIHUUHG WR DV FXOWLYDU ,QRGRUDWXP 0RUH UHFHQWO\ 0RRUH HW DO

PAGE 27

f DQDO\]HG WKH SURGXFWLRQ RI WKHVH FRPSRXQGV ZLWK KLJK SUHVVXUH OLTXLG FKURPDWRJUDSK\ ,VRODWHV LQ 9&*V DQG SURGXFHG FKDUDFWHULVWLF YRODWLOH SURILOHV ZKHUHDV LVRODWHV LQ 9&*V DQG GLG QRW 5HVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV 5)/3Vf DOVR KDYH EHHQ HPSOR\HG WR GHWHUPLQH WKH JHQHWLF UHODWLRQVKLSV DPRQJ LVRODWHV RI ) R[\VSRUXP 7KHVH PDUNHUV DUH LGHDOO\ VXLWHG WR JHQHWLF GLYHUVLW\ VWXGLHV EHFDXVH RI WKH IROORZLQJ FKDUDFWHULVWLFV Lf PRVW DUH VHOHFWLYHO\ QHXWUDO LLf SRO\PRUSKLVPV WHQG WR EH PRUH QXPHURXV FRPSDUHG WR RWKHU W\SHV RI PDUNHUV VXFK DV LVR]\PHV LLLf WKH\ DUH UHSURGXFLEOH DQG LYf WKRVH LGHQWLI\LQJ UDQGRP VLQJOHFRS\ ORFL DYRLG SUREOHPV WKDW DUH DVVRFLDWHG ZLWK OLQNDJH RI PDUNHUV ,Q WKLV VWXG\ SUREHV IURP ) R[\VSRUXP I VS O\FRSHUVLFL WKDW FRUUHVSRQG WR VLQJOHFRS\ DQRQ\PRXV ORFL (OLDV HW DO f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f $GGLWLRQDOO\ WKUHH LVRODWHV RI ) R[\VSRUXP I VS O\FRSHUVLFL DQG D VLQJOH LVRODWH RI

PAGE 28

7DEOH /LVW RI LVRODWHV WKHLU YHJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*f FORQDO OLQHDJHV WKH FXOWLYDU DQG JHRJUDSKLFDO UHJLRQV IURP ZKLFK WKH\ ZHUH FROOHFWHG 9&*8 ,VRODWH &XOWLYDUY 2ULJLQ DQG FROOHFWRUf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

PAGE 29

7DEOH f§FRQWLQXHG 9&*8 ,VRODWH &XOWLYDUY 2ULVLQ DQG FROOHFWRU /LQHDJH 3: &DYHQGLVK 3KLOLSSLQHVP 9, 3: &DYHQGLVK 3KLOLSSLQHVP 9, '$9$2 6LON 3KLOLSSLQHVK 9,, 7 *URV 0LFKHO 7DL ZDQ I 9,, 3K/ /DWXQGDQ 3KLOLSSLQHVL 9,, 3KO /DWXQGDQ 3KLOLSSLQHV 9,, /DWXQGDQ 7DL ZDQ J 9,, -/7+ .OXH QDPZD 6PRHQJ +Z\ 7KDLODQGY ; -/7+ .OXH QDPZD 6PRHQJ +Z\ 7KDLODQGY ;r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f )ORULGD 86$ E % %XUUR )ORULGD86$E -&% %XUUR )ORULGD86$E O -/7+ .OXH QDPZD 6PRHQJ +Z\ 7KDLODQGY -/7+ .OXH QDPZD 6PRHQJ +Z\ 7KDLODQGY -/7+ .OXH QDPZD &KDL
PAGE 30

7DEOH f§FRQWLQXHG 9&*8 ,VRODWH &XOWLYDUY 2ULVLQ DQG FROOHFWRUf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
PAGE 31

7DEOH f§FRQWLQXHG 9&*8 ,VRODWH &XOWLYDUY 2ULJLQ DQG FROOHFWRU /LQHDJH $ $SSOH )ORULGD 86$E ,9 $ $SSOH )ORULGD 86$E ,9 &6% $SSOH )ORULGD 86$E ,9 -& $SSOH )ORULGD 86$E ,9 $ $SSOH )ORULGD 86$E O
PAGE 32

7DEOH f§FRQWLQXHG 9&*8 ,VRODWH &XOWLYDU 2ULDLQ DQG FROOHFWRU /LQHDDH ,1'2 0XVD VSS -DWHVDUL (DVW -DYD,QGRQHVLDQ ,,, ,1'= 0XVD VSS -DWHVDUL (DVW -DYD,QGRQHVLDQ ,,, ,1'= 0XVD VSS -DWHVDUL (DVW -DYD,QGRQHVLDQ ,,, )XVDULXP R[\VSRUXP I VS O\FRSHUVLFL 6& +RPHVWHDG %UDGHQWRQ )ORULGD 6& 2ULVWDQR ,WDO\ 6& 6XQQ\ %UDGHQWRQ )ORULGD )XVDULXP R[\VSRUXP I VS QLYHXP &6 )ORULGDZ f9HJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*Vf ZHUH DVVLJQHG XVLQJ QLWURJHQ PHWDEROLVP QLWf PXWDQWV DFFRUGLQJ WR WKH SURWRFROV RI &RYH f DV PRGLILHG E\ 3XKDOOD f Y&XOWLYDUV DUH LQWHU DQG LQWUDVSHFLILF GLSORLG RU WULSORLG K\EULGV RI 0XVD DFXPLQDWD $$f DQG 0XVD EDOELVLDQD %%f 7KH SORLG\ OHYHOV DQG FRQVWLWXWLRQV RI WKH FXOWLYDUV DUH DV IROORZV $$$ *URV 0LFKHO +LJKJDWH 0RQV 0RQV PDQf &DYHQGLVK 'ZDUI &DYHQGLVK *UDQGH 1DPH :LOOLDPV /DFDWDQ $$ 6+ DQG 6+ V\QWKHWLF FORQHVf $$% /DG\ )LQJHU 3DFRYDQ 3UDWD 6LON /DWXQGDQ 0DTXHR $%% 6DED %OXJJRH +DUDUH .KRORERZD 3LVDQJ DZDN .OXH QDPZD 'XFDVVH 0EXIX %XUUR =DPELD $% 1H\ SRRYDQ 6XNDOL $$$$ 7HWUDSORLG [ PLVVLQJ GDWD PDGH LW LPSRVVLEOH WR GHWHUPLQH FRHIILFLHQW RI VLPLODULW\ EXW EDVHG RQ DOO RWKHU GDWD LW LV SUHVXPHG WR EH LQ WKH OLQHDJH LQGLFDWHG \ XQLTXH LVRODWH WKDW KDG QR OLQHDJH DIILQLW\ EDVHG RQ GHILQHG FULWHULD WKH OLQHDJH DVVLJQHG LV EDVHG RQ FRHIILFLHQWV RI VLPLODULW\ WR WKH PRVW FORVHO\ UHODWHG LVRODWHV r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f $0 3HGURVD 3KLOLSSLQHV P 1 5RSHURV 3KLOLSSLQHV Q = 0 &RUGHLUR (0%5$3$ &UX] GDV $PDV %UD]LO YLD ( /RXGUHVf T )HUJXVRQ&RQLH %DQDQD %RDUG .LQJVWRQ -DPDLFD U % %UDXQZRUWK 2UHJRQ 6WDWH 8QLYHUVLW\ &RUYDOOLV X 7VDL\RXQJ &KXDQJ 1DWLRQDO 7DLZDQ 8QLYHUVLW\ 7DLSHL Y /HVOLH .DQVDV 6WDWH 8QLYHUVLW\ 0DQKDWWDQ Z ) 0DUWLQ 86'$ $56 6DOLQDV &$

PAGE 33

) R[\VSRUXP I VS QLYHXP ZHUH DQDO\]HG WR FRPSDUH UHVXOWV ZLWK JHQRW\SHV UHSUHVHQWLQJ GLIIHUHQW IRUPDH VSHFLDOHV '1$ ,VRODWLRQ $OO FXOWXUHV ZHUH GHULYHG IURP VLQJOH PLFURFRQLGLD DQG ZHUH VWRUHG DW r& RQ VWULSV RI :KDWPDQ ILOWHU SDSHU &RUUHOO HW DO f 3DSHU VWULSV ZHUH SODWHG RQWR SRWDWR GH[WURVH DJDU 3'$f 'LIFR 'HWURLW 0,f $IWHU DSSUR[LPDWHO\ GD\V RI JURZWK DW r& D PP EORFN ZDV H[FLVHG IURP WKH PDUJLQ RI FRORQLHV RQ 3'$ DQG WUDQVIHUUHG WR / (UOHQPH\HU IODVNV FRQWDLQLQJ DW OHDVW PO RI SRWDWR GH[WURVH EURWK 'LIFR J/f $IWHU WR GD\V RI JURZWK LQ VWLOO FXOWXUH WKH FRQWHQWV RI WKH IODVN ZHUH ILOWHUHG WKURXJK VWHULOL]HG FKHHVH FORWK WR FROOHFW P\FHOLXP 7KH P\FHOLXP ZDV SODFHG LQWR PO SODVWLF WXEHV IUR]HQ DW r& DQG O\RSKLOL]HG IRU DW OHDVW KRXUV %XIIHU IRU '1$ H[WUDFWLRQ ZDV FRPSULVHG RI D YROXPH RI WKH IROORZLQJ VROXWLRQV %XIIHU $ 0 6RUELWRO 0 7ULV P0 ('7$ DW S+ f %XIIHU % 0 7ULV DW S+ P0 ('7$ DQG P0 &7$%f DQG b 6DUNRV\O 7HQ PO RI WKH H[WUDFWLRQ EXIIHU ZHUH PL[HG ZLWK DSSUR[LPDWHO\ J RI JURXQG P\FHOLDO SRZGHU DQG WKH WXEHV ZHUH SODFHG LQ D r& ZDWHU EDWK IRU PLQXWHV 7KH FRQWHQWV RI WKH WXEHV ZHUH WKHQ VKDNHQ DQG PO RI VROXWLRQ ZDV WUDQVIHUUHG WR HDFK RI VWHULOH PO PLFURFHQWULIXJH WXEHV )LYH KXQGUHG PLFUROLWHUV RI FKORURIRUUPRFWDQRO f VROXWLRQ ZHUH DGGHG WR HDFK WXEH 7KH VROXWLRQ ZDV PL[HG XVLQJ D YRUWH[ VKDNHU IRU DSSUR[LPDWHO\ PLQXWHV EHIRUH FHQWULIXJDWLRQ IRU PLQXWHV DW J LQ D PLFURFHQWULIXJH DW URRP WHPSHUDWXUH 8QOHVV QRWHG DOO IXUWKHU FHQWULIXJH VWHSV ZHUH GRQH LQ D PLFURFHQWULIXJH DW URRP WHPSHUDWXUH 7KH VXSHUQDWDQW ZDV WUDQVIHUUHG WR VWHULOH PO WXEHV DQG WUHDWHG ZLWK SL RI PJ 51DVH $ 6LJPD &KHPLFDO &R 6W /RXLV 02fPO RI VROXWLRQ IRU

PAGE 34

PLQXWHV DW r& )ROORZLQJ 51DVH WUHDWPHQW SL RI PJ SURWHLQDVH 6LJPDfPO RI VROXWLRQ ZHUH DGGHG WR WKH WXEHV DQG WKH WXEHV ZHUH LQFXEDWHG IRU PLQXWHV DW r& $SSUR[LPDWHO\ RQH YROXPH LVRSURSDQRO ZDV DGGHG DQG WKH WXEHV ZHUH FHQWULIXJHG IRU PLQXWHV DW J 7KH LVRSURSDQRO WKHQ ZDV GLVFDUGHG DQG SL RI LFHFROG b HWKDQRO ZHUH DGGHG EHIRUH FHQWULIXJDWLRQ IRU ILYH PLQXWHV 7KH HWKDQRO ZDV GLVFDUGHG DQG WKH '1$ VDPSOH ZDV DLUGULHG IRU DW OHDVW PLQXWHV LQ D ODPLQDU IORZ KRRG $W OHDVW SL RI 7( EXIIHU P0 7ULV P0 ('7$ DW S+ f ZHUH DGGHG WR WKH WXEHV DQG WKH\ ZHUH SODFHG LQ D ZDWHU EDWK DW r& XQWLO WKH '1$ SHOOHW ZDV GLVVROYHG 6DPSOHV ZKLFK ZHUH GLIILFXOW WR EULQJ LQWR VROXWLRQ ZHUH VXEMHFWHG WR D /L&O WUHDWPHQW 7KUHH KXQGUHG PLFUROLWHUV RI LFHFROG 0 /L&O VROXWLRQ ZHUH DGGHG WR HDFK WXEH DQG WKH WXEHV ZHUH SODFHG RQ LFH IRU PLQXWHV EHIRUH FHQWULIXJDWLRQ DW J IRU PLQXWHV DW r& 7KH VXSHUQDWDQW ZDV WUDQVIHUUHG WR D VWHULOH PO WXEH FRQWDLQLQJ SL RI LVRSURSDQRO 7KLV VROXWLRQ ZDV PL[HG DQG WKH WXEHV ZHUH NHSW DW URRP WHPSHUDWXUH IRU PLQXWHV $IWHU FHQWULIXJDWLRQ DW J IRU PLQXWHV DW r& WKH VXSHUQDWDQW ZDV GLVFDUGHG DQG SL RI LFHFROG b HWKDQRO ZHUH DGGHG WR WKH WXEHV $IWHU FHQWULIXJDWLRQ DW J IRU PLQXWHV WKH HWKDQRO ZDV GLVFDUGHG DQG WKH '1$ ZDV DLU GULHG 7( EXIIHU SLf ZDV DGGHG WR WKH WXEHV DQG WKH\ ZHUH SODFHG LQ D ZDWHU EDWK DW r& XQWLO WKH SHOOHW ZDV GLVVROYHG 7KH FRQFHQWUDWLRQ RI '1$ LQ WKH VDPSOHV ZDV HVWLPDWHG E\ UXQQLQJ SL RI HDFK VDPSOH RQ DQ DJDURVH JHO DORQJ ZLWK '1$ IUDJPHQWV EDFWHULRSKDJH ODPEGD '1$ GLJHVWHG ZLWK 3VW,f RI D NQRZQ FRQFHQWUDWLRQ DQG PDNLQJ YLVXDO FRPSDULVRQV RI WKHLU UHODWLYH IOXRUHVFHQFH LQ WKH SUHVHQFH RI XOWUDYLROHW 89f OLJKW DQG SJ RI HWKLGLXP EURPLGHPO RI VROXWLRQ

PAGE 35

6RXWKHUQ %ORWWLQJ DQG +\EULGL]DWLRQ $SSUR[LPDWHO\ SJ RI '1$ ZHUH GLJHVWHG ZLWK DW OHDVW XQLWV RI HLWKHU 2UDO (FR5? RU +DHOOO UHVWULFWLRQ HQ]\PHV %HWKHVGD 5HVHDUFK /DERUDWRULHV *DLWKHUVEXUJ 0'f DQG LQFXEDWHG IRU DW OHDVW KRXUV 7KHVH UHVWULFWLRQ HQ]\PHV ZHUH FKRVHQ EDVHG RQ WKHLU LQVHQVLWLYLW\ WR '1$ PHWK\ODWLRQ DQG WKHLU DELOLW\ WR GLJHVW '1$ FRQVLVWHQWO\ 5HVWULFWLRQ IUDJPHQWV ZHUH VHSDUDWHG E\ HOHFWURSKRUHVLV LQ b DJDURVH IRU (FR5: DQG =fUDOGLJHVWHG '1$ RU b DJDURVH IRUmHOOOGLJHVWHG '1$ LQ 7%( EXIIHU DW S+ *HOV ZHUH UXQ DW HLWKHU YROWV IRU DSSUR[LPDWHO\ KRXUV RU YROWV IRU DSSUR[LPDWHO\ KRXUV %DFWHULRSKDJH ODPEGD '1$ GLJHVWHG ZLWK HLWKHU 3VWO RU +LQGOOO ZDV LQFOXGHG RQ HDFK JHO DQG XVHG WR FDOFXODWH WKH PROHFXODU PDVV RI UHVWULFWLRQ IUDJPHQWV REWDLQHG IURP ) R[\VSRPP I VS FEHQVH '1$ (WKLGLXP EURPLGH PJPOf ZDV GLVVROYHG LQ WKH DJDURVH JHO DW D FRQFHQWUDWLRQ RI SL PO DQG WKH GLJHVWHG '1$ ZDV LOOXPLQDWHG E\ 89 LUUDGLDWLRQ DQG SKRWRJUDSKHG 7KH '1$ ZDV WUDQVIHUUHG WR 1\WUDQ PHPEUDQHV 6FKOHLFKHU t 6FKXHOO ,QF .HHQH 1+f XVLQJ WKH FDSLOODU\ WUDQVIHU PHWKRG 6DPEURRN HW DO f 7KH '1$ WUDQVIHU SURFHHGHG IRU DW OHDVW KRXUV DQG WKH '1$ ZDV LPPRELOL]HG E\ 89 FURVV OLQNLQJ 89 QP FURVV OLQNHU PRGHO )% 89;/ )LVFKHU 6FLHQWLILF 3LWWVEXUJK 3$f 7R UHGXFH WKH LQFLGHQFH RI UHSHDWHGO\ VFRULQJ VLPLODU UHJLRQV RI WKH JHQRPH RU K\SHUYDULDEOH UHJLRQV FORQHV FRQWDLQLQJ VLQJOHFRS\ '1$ VHTXHQFHV REWDLQHG IURP 7DOPD .DWDQ 7KH 9ROFDQL &HQWHU %HW 'DJDQ ,VUDHOf ZHUH XWLOL]HG (OLDV HW DK f &ORQHV ZHUH FRQVLGHUHG WR EH VLQJOHFRS\ EDVHG RQ WKH FULWHULD GHILQHG E\ (OLDV HW DK f DV K\EULGL]LQJ WR RQO\ D VLQJOH '1$ IUDJPHQW LQ DQ\ RI WKH LVRODWHV WHVWHG XVLQJ DW OHDVW RQH RI WKUHH UHVWULFWLRQ HQ]\PHV )LJXUH f

PAGE 36

0ROHFXODU : HLJKW .E ,VRODWH 9&* 3RO\PRUSKLF 6WDWH *$/ $') 3DFRYDQ *0 6$%$ )&0: &9$ 6" -/7+ -/7+ $ 0: 6, )LJXUH 6LQJOHFRS\ VHTXHQFHV RI VHOHFWHG )XVDULXP R[\VSRUXP I VS FEHQVH LVRODWHV '1$ RI LVRODWHV ZDV GLJHVWHG ZLWK (FR 5) UHVWULFWLRQ HQ]\PH DQG SUREHG ZLWK FORQH 3RO\PRUSKLF VWDWHV LQGLFDWH WKH SUHVHQFH f RU DEVHQFH f UHVSHFWLYHO\ RI WKH RU .E EDQGV

PAGE 37

'1$ IRU HDFK FORQH ZDV ODEHOHG XVLQJ UDQGRP KH[DPHU SULPHUV WR LQFRUSRUDWH IOXRUHVFHLQG873 IROORZLQJ WKH SURFHGXUHV SURYLGHG E\ WKH PDQXIDFWXUHU 'XSRQW 1(1 5HQDLVVDQFH ( GX 3RQW GH 1HPRXUV t &R ,QF %RVWRQ 0$f '1$ ODEHOLQJ K\EULGL]DWLRQ DQG GHWHFWLRQ IROORZHG WKH SURFHGXUHV SURYLGHG E\ WKH PDQXIDFWXUHU 'XSRQW %RVWRQ 0$f 3UHK\EULGL]DWLRQ K\EULGL]DWLRQ DQG ZDVKLQJ ZHUH SHUIRUPHG DW r& XVLQJ D +\EDLG K\EULGL]DWLRQ RYHQ 'RW 6FLHQWLILF ,QF )OLQW 0,f 0HPEUDQHV ZHUH SODFHG EHWZHHQ DFHWDWH VKHHWV DQG H[SRVHG WR ;UD\ ILOP IRU DW OHDVW KRXUV 0LWRFKRQGULDO '1$ RI ) R[\VSRUXP I VS FEHQVH LVRODWH 6 9&* f ZDV LVRODWHG IROORZLQJ WKH SURFHGXUHV RI .LVWOHU DQG /HRQJ f DQG ODEHOHG DV GHVFULEHG DERYH 7KH PLWRFKRQGULDO '1$ SURILOHV RI D VXEVHW RI LVRODWHV ZHUH REWDLQHG E\ GLJHVWLQJ DSSUR[LPDWHO\ SJ RI WRWDO '1$ ZLWK DW OHDVW XQLWV RI WKH UHVWULFWLRQ HQ]\PH +DH:O DQG SURELQJ ZLWK WKH PLWRFKRQGULDO '1$ RI LVRODWH 6 5HVWULFWLRQ IUDJPHQWV ZHUH VHSDUDWHG E\ HOHFWURSKRUHVLV LQ b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

PAGE 38

SUREHHQ]\PH FRPELQDWLRQ 7KH GLIIHUHQW UHVWULFWLRQ VL]H IUDJPHQWV JHQHUDWHG IRU HDFK FRPELQDWLRQ RI SUREH DQG HQ]\PH ZHUH FRQVLGHUHG WR EH DOOHOHV DW D VLQJOH 5)/3 ORFXV DQG WKHLU SUHVHQFH RU DEVHQFH ZDV VFRUHG IRU HDFK LVRODWH 5)/3 SDWWHUQV IRU HDFK FRPELQDWLRQ RI SUREH DQG HQ]\PH ZHUH FRPELQHG WR DVVLJQ DQ 5)/3 KDSORW\SH WR HDFK LVRODWH 7KH GDWD ZHUH DQDO\]HG E\ D FODGLVWLF DSSURDFK EDVHG RQ SDUVLPRQ\ DQDO\VLV XVLQJ WKH FRPSXWHU SURJUDP 3$83 9HUVLRQ 6ZRIIRUG f DQG E\ D SKHQHWLF DSSURDFK XVLQJ GLVWDQFH PDWUL[ PHWKRGV 83*0$ FOXVWHULQJ 6QHDWK DQG 6RNDO f DQG WKH QHLJKERUMRLQLQJ DOJRULWKP RI3K\OLS 9HUVLRQ F )HOVHQVWHLQ 8QLYHUVLW\ RI :DVKLQJWRQf )RU SDUVLPRQ\ DQDO\VLV SK\ORJHQLHV ZHUH GHULYHG E\ XVLQJ WKH KHXULVWLF VHDUFK RSWLRQ DQG WKH GHJUHH RI VXSSRUW ZDV HYDOXDWHG XVLQJ ERRWVWUDS UHSOLFDWHV ,Q DGGLWLRQ FRHIILFLHQWV RI VLPLODULW\ EDVHG RQ VLPSOH PDWFKLQJ ZHUH FDOFXODWHG IRU WKRVH LVRODWHV LQ ZKLFK GDWD ZHUH DYDLODEOH IRU HYHU\ 5)/3 ORFL VFRUHG EDVHG RQ WKH IRUPXOD GHVFULEHG E\ 6QHDWK DQG 6RNDO f ,VRODWHV ZHUH DUELWUDULO\ FRQVLGHUHG WR EH ZLWKLQ WKH VDPH FORQDO OLQHDJH EDVHG RQ FRHIILFLHQWV RI VLPLODULW\ UDQJLQJ IURP WR 6LQFH PDQ\ RI WKH LVRODWHV KDG LGHQWLFDO PXOWLORFXV KDSORW\SHV RQO\ D VLQJOH LVRODWH ZDV XVHG WR UHSUHVHQW HDFK KDSORW\SH LQ GDWD DQDO\VHV 7R GHWHUPLQH ZKHWKHU WKLV FROOHFWLRQ RI LVRODWHV SURYLGHG HYLGHQFH IRU FORQDO UHSURGXFWLRQ WKH JDPHWLF GLVHTXLOLEULXP FRHIILFLHQW 'f ZDV FDOFXODWHG DPRQJ SDLUV RI DOOHOHV DW GLIIHUHQW ORFL E\ PHWKRGV GHVFULEHG E\ :HLU f &ORQHFRUUHFWHG DOOHOH IUHTXHQFLHV XVLQJ RQO\ D VLQJOH UHSUHVHQWDWLYH IRU HDFK KDSORW\SH ZHUH HPSOR\HG IRU WKH FDOFXODWLRQV $OVR WR DYRLG WKH SRWHQWLDO SUREOHP RI UHSHDWHGO\ VFRULQJ VLPLODU UHJLRQV

PAGE 39

RQO\ GDWD IURP D VLQJOH UHVWULFWLRQ HQ]\PH GLJHVWLRQ ZHUH XVHG IRU HDFK SUREH LQ WKH DQDO\VHV 1LQH KXQGUHG DQG HLJKW\HLJKW SDLUZLVH FRPSDULVRQV ZHUH SHUIRUPHG WR WHVW IRU GLVHTXLOLEULXP EHWZHHQ PXOWLSOH DOOHOHV DW QLQH ORFL $ WHVW IRU WKH VLJQLILFDQFH RI WKH GLVHTXLOLEULXP FRHIILFLHQW EHWZHHQ HDFK SDLU RI DOOHOHV DW WZR ORFL ZDV IRUPXODWHG ZLWK WKH FKLVTXDUH VWDWLVWLF 3X $f SY $f ZKHUH Q ZDV WKH QXPEHU RI LQGLYLGXDOV LQ WKH VDPSOH DQG 'XY ZDV WKH PD[LPXP OLNHOLKRRG HVWLPDWRU IRU WKH FRHIILFLHQW RI GLVHTXLOLEULXP EHWZHHQ DOOHOHV X DQG Y 7KH REVHUYHG DOOHOH IUHTXHQFLHV IRU WKH ORFL ZHUH 3X DQG 3Y UHVSHFWLYHO\ 0F'RQDOG HW DO :HLU f 7KH FKLVTXDUH VWDWLVWLF KDG RQH GHJUHH RI IUHHGRP DQG WKH SDLUV RI ORFL WKDW VKRZHG VLJQLILFDQW GHSDUWXUH IURP UDQGRP H[SHFWDWLRQV 3 f ZHUH FRQVLGHUHG WR EH LQ GLVHTXLOLEULXP $ WHVW IRU VLJQLILFDQFH RI WKH GLVHTXLOLEULXP FRHIILFLHQW DFURVV DOO DOOHOHV IRU HDFK SDLU RI ORFL ZDV IRUPXODWHG ZLWK WKH FKLVTXDUH WHVW VWDWLVWLF DV GHVFULEHG E\ 0F'RQDOG HW DO f 7KH LVRODWHV WHVWHG LQ WKH VWXG\ ZHUH DVVLJQHG SUHYLRXVO\ WR D 9&* DQG RQH WR VHYHUDO UHSUHVHQWDWLYH LVRODWHV ZLWKLQ HDFK RI WKH H[DPLQHG 9&*V ZHUH SDWKRJHQLF RQ DW OHDVW RQH DQG DV PDQ\ DV HLJKW GLIIHUHQW EDQDQD FXOWLYDUV 3ORHW] SHUVRQDO FRPPXQLFDWLRQf $OWKRXJK RQH VDPSOH RI '1$ ZDV LVRODWHG IURP HDFK LVRODWH PDQ\ LVRODWHV ZHUH WHVWHG DW OHDVW RQFH DQG DV PDQ\ DV WKUHH WLPHV IRU HDFK SUREH HQ]\PH FRPELQDWLRQ H[DPLQHG $GGLWLRQDOO\ D VXEVHW RI ) R[\VSRUXP I VS FEHQVH DQG ) R[\VSRUXP I

PAGE 40

VS O\FRSHUVLFL LVRODWHV WHVWHG KHUH ZHUH VHSDUDWHO\ DQDO\]HG XVLQJ '1$ VHTXHQFLQJ GDWD 2n'RQQHOO SHUVRQDO FRPPXQLFDWLRQf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f 7KH ILYH PRVW FRPPRQ KDSORW\SHV UHSUHVHQWHG b RI WKH LVRODWHV 7KH PHGLDQ QXPEHU RI DOOHOHV SHU ORFXV ZDV WKUHH DQG LI WKUHH DOOHOHV ZHUH SUHVHQW DW HDFK ORFXV WKHRUHWLFDOO\ [ f SRVVLEOH KDSORW\SHV FRXOG H[LVW IRU WKLV FROOHFWLRQ RI ) R[\VSRUXP I VS FEHQVH +RZHYHU WKH PDMRULW\ RI VLQJOHLVRODWH KDSORW\SHV IRXQG ZHUH WKH UHVXOW RI RQH WR D IHZ DOOHOLF GLIIHUHQFHV IURP D PRUH FRPPRQ KDSORW\SH 7R GHWHUPLQH UHODWHGQHVV DPRQJ LVRODWHV WKH 5)/3 KDSORW\SHV ZHUH VXEMHFWHG WR SKHQHWLF DQG FODGLVWLF DQDO\VHV %RWK W\SHV RI DQDO\VHV SURGXFHG WUHHV ZLWK VLPLODU EUDQFKLQJ SDWWHUQV 7KH b PDMRULW\ UXOH ERRWVWUDS FRQVHQVXV WUHH JHQHUDWHG E\ 3$83 LV SUHVHQWHG LQ )LJXUH $ GLFKRWRP\ ZLWK VWURQJ ERRWVWUDS VXSSRUW bf ZDV

PAGE 41

1XPEHU RI ,VRODWHV f§WW +DSORW\SH )LJXUH )UHTXHQF\ GLVWULEXWLRQ RI 5)/3 KDSORW\SHV DPRQJ WKH LVRODWHV UHSUHVHQWLQJ D ZRUOG FROOHFWLRQ RI )XVDULXP R[\VSRUXP I VS FEHQVH **

PAGE 42

PP B (81&66 )'/ 6& _f B )2/ 6& f§ (2/ 6& f ‹n&* & YFF RL RL c 9&* 3K 9&* /$3 3: 3 9&* 6$%$ @ ) R[\VSRUXP I VS O\FRSHUVLFL DQG ) R[\VSRUXP I VS QLYHDP LP 9&* 0: 9&* 0: Lf§ 9&* 0: f 0+ 9&* & f§ 9&* ) 9&* 3$9&* 0: 9&* 099&* 0: 9&* .: 9&* 0: // f 9&* ,& 9&* 3DFRYDQ ‘ ‘ 9&* & *0 -)2&; 9&* ) f‘ 9&* +, ‘f§ 9&* ‘/ 9&* 9&* 9&* 9&* ,1'2 ‘ TQO f // f t‘ 9&* 0*6$ 9&* 6$ 9&* 6$ 9&* 67$ 9&* 67% 9&* 6, 9&* '$9$2 9&* 3KL 9&* 7 9&* 3K/ f§ 9&* -/7+ 9&* -/7+ @n YFDM 9&* 1 )2& ,, 9&* $ 9&* ** 9&* 9&* $ n )2& ,9 f 9&* -& f§r )2& ,9 9&* 6+ f§" )2& &I f§ 9&* (6 9&* $ 9&* *0% 9&* 0DFD 9&* 679&* 0: 9&* 0: f§ 9&* -/7+ 9&* 0: M 9&* 61 ‘Lf§ 9&* 672 Lf§ 9&* 0: 9&* 0: 9&* 0: L 9&* 0: B 9&* 0: L\ 9&* 9&* 9&* f§ 9&* $ f 9&* 6713O 9&* 6713 9&* -/7+ 9&* -/7+ f§ 9&* -/7+ 9&* -/7+ )2& 9,,, )LJXUH 0LGSRLQW URRWHG b PDMRULW\ UXOH FRQVHQVXV WUHH UHSUHVHQWLQJ ERRWVWUDS UHSOLFDWHV 2QH LVRODWH UHSUHVHQWV HDFK RI WKH 5)/3 KDSORW\SHV RI )XVDULXP R[\VSRUXP I VS FEHQVH WKUHH LVRODWHV RI ) R[\VSRUXP I VS O\FRSHUVLFL DQG RQH LVRODWH ) R[\VSRUXP I VS QLYHXP %UDQFK OHQJWKV DUH LQGLFDWHG RQ HDFK EUDQFK DQG ERRWVWUDS YDOXHV DUH LQ SDUHQWKHVLV 7UHH OHQJWK &RQVLVWHQF\ ,QGH[ +RPRSODV\ ,QGH[ 5HWHQWLRQ ,QGH[

PAGE 43

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bf :LWKLQ RQH RI WKH EUDQFKHV LVRODWHV LQ 9&*V DQG IRUPHG RQH FODGH ,VRODWHV LQ 9&* ZHUH JHQHWLFDOO\ VLPLODU EXW GLVWLQFW b ERRWVWUDS VXSSRUWf VHH EHORZf IURP WKHVH LVRODWHV :LWKLQ WKH RWKHU EUDQFK VL[ FODGHV ZHUH LGHQWLILHG ,VRODWHV LQ 9&*V b ERRWVWUDS VXSSRUWf b ERRWVWUDS VXSSRUWf b ERRWVWUDS VXSSRUWf b ERRWVWUDS VXSSRUWf DQG LVRODWHV RI ) R[\VSRQLP I VS O\FRSHUVLFL b ERRWVWUDS VXSSRUWf HDFK IRUPHG WKHLU RZQ FODGH ,VRODWHV LQ 9&*V DQG WKUHH LVRODWHV LQ IRUPHG D FODGH RI ZHDN VXSSRUW bf DQG WKHUHIRUH FRXOG QRW EH FRQILGHQWO\ GLIIHUHQWLDWHG IURP LVRODWHV LQ 9&*V WZR LVRODWHVf DQG WKH VLQJOH LVRODWH RI ) R[\VSRQLP I VS QLYHXP $OO FODGHV ZLWK VWURQJ ERRWVWUDS VXSSRUW DUH FRPSULVHG RI LVRODWHV WKDW KDYH LGHQWLFDO RU QHDUO\ LGHQWLFDO PXOWLORFXV KDSORW\SHV DQG DUH UHIHUUHG WR KHUH DV FORQDO OLQHDJHV $GGLWLRQDOO\ PDQ\ RI WKH LVRODWHV ZKLFK FRXOG QRW EH UHVROYHG XVLQJ ERRWVWUDS

PAGE 44

DQDO\VLV VKDUHG QHDUO\ LGHQWLFDO PXOWLORFXV KDSORW\SHV ZLWK RWKHU LVRODWHV 7R IXUWKHU XQGHUVWDQG WKH JHQHWLF UHODWLRQVKLSV DPRQJ XQUHVROYHG LVRODWHV D VLPSOH PDWFKLQJ FRHIILFLHQW RI VLPLODULW\ IRU FRPSDULVRQ ZDV XVHG 7DEOH f ,VRODWHV ZLWK FRHIILFLHQWV RI VLPLODULW\ UDQJLQJ IURP WR ZHUH FRQVLGHUHG WR EH ZLWKLQ D FORQDO OLQHDJH 7KLV UDQJH UHIOHFWV DSSDUHQW QDWXUDO JURXSV )LJXUH f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f ,Q JHQHUDO 9&*V DOLJQHG ZLWK VLQJOH FORQDO OLQHDJHV H[FHSWLRQV WR WKLV ZHUH LVRODWHV LQ 9&*V 7DEOH OLVWV HDFK OLQHDJH ZLWK WKH SUHIL[ )2&f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

PAGE 45

7DEOH 6LPLODULW\ PDWUL[ RI VLPSOH PDWFKLQJ FRHIILFLHQWV EDVHG RQ UHVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV IRU VHOHFWHG LVRODWHV RI )XVDULLWP R[\VSRUXP I VS FEHQVH ) R[\VSRUXP I VS O\FRSHUVLFL DQG ) R[\VSRUXP I VS QLYHXP Ua &0 FU! 6 &' 2 6 FG &0 2 &' 2 R &2 &' 8B &1_ 2 &' 2 [ &0 &' &' 2 V F &1&1&' &' 2 9&* '$9$2 U &2 &1O &' 2 &1O [] &O &2 &1_ &' 2 &1_ B[] &/ &2 &1_ &' &' 2 17 ; Wf§ &2 &0 &' 2 &2 17 &1O &' &' 2 9&* 0: 9&* 0: &' &' &' FR L' &1_ &' &' 2 &1_ &2 &' &1_ &' &' 2 RR &' &1O &' &' R &' 0 ,' ; &' &1O &' &' 2 &0 &' &1O &' &' 2 &0 ‘r FR ; &' &1O &' 2 9&* 6713 &0 &' &0 &' &' 2 &0 &0 &' 2 &0 &' 2 9&* 0: &' &2 1M &0 &' 2 9&* ,QGR R &' &0 &2 R &2 R &2 e & 17 ,' R 9&* 9&*,& 9&* ) 9&* +, 9&* 6$%$ 9&* '$9$2 9&* 7 9&* 3KL 9&* 3K/ 9&* -/7+ 9&* 0: 9&* 0: 9&* 0: 9&* 9&* 67$ 9&* 6, 9&* 1 9&* $ 9&* 6+ 9&* 6713 9&* 9&* 0: 9&* 0:, 9&* 0: 9&* 0: 9&* ,QGR 6& O\FRSHUVLFL 6& O\FRSHUVLFL &6 QLYHXP

PAGE 46

7DEOH &ORQDO OLQHDJHV RI )XVDULXP R[\VSRUXP I VS FEHQVH LVRODWHV WKHLU JHRJUDSKLFDO GLVWULEXWLRQV DQG FRUUHVSRQGLQJ YHJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*f 1R RI 9&* /LQHDVH ,VRODWHV *HRVUDSKLF 'LVWULEXWLRQ 5HSUHVHQWHG )2& $XVWUDOLD %UD]LO %XUXQGL &RPRUHV ,VODQGV +RQGXUDV -DPDLFD 0DODZL 1LFDUDJXD 7DQ]DQLD 7KDLODQG 8JDQGD DQG WKH 8QLWHG 6WDWHV )ORULGDf DQG )2& ,, $XVWUDOLD %UD]LO &DQDU\ ,VODQGV &RVWD 5LFD +RQGXUDV -DPDLFD 0DOD\VLD 6RXWK $IULFD DQG 7DLZDQ DQG )2& ,,, ,QGRQHVLD DQG 7DLZDQ DQG )2& ,9 8QLWHG 6WDWHV )ORULGDf )2& 9 0DODZL )2& 9, 3KLOLSSLQHV )2& 9,, 3KLOLSSLQHV 7DLZDQ )2& 9,,, 7DQ]DQLD )2& ,; $XVWUDOLD )2& ; 7KDLODQG 7VRODWHV (6 LQ 9&* DQG *0 LQ 9&* GLG QRW DOLJQ ZLWK DQ\ OLQHDJH EDVHG RQ FRHIILFLHQW RI VLPLODULW\ GDWD

PAGE 47

WR FRPSDUHG WR LVRODWHV LQ )2& ,, 7KH VLQJOH LVRODWH RI ) R[\VSRUXP I VS QLYHXP KDG FRHIILFLHQWV RI VLPLODULW\ UDQJLQJ IURP WR FRPSDUHG WR LVRODWHV LQ )2& DQG IURP WR FRPSDUHG WR LVRODWHV LQ )2& ,, 7KXV WKH WZR ODUJHVW OLQHDJHV RI ) R[\VSRUXP I VS FEHQVH HDFK DUH PRUH JHQHWLFDOO\ VLPLODU WR WKH ) R[\VSRUXP I VS QLYHXP LVRODWH WKDQ WR HDFK RWKHU 6LPLODUO\ WKH\ DUH URXJKO\ DV JHQHWLFDOO\ GLVWLQFW IURP HDFK RWKHU DV HLWKHU LV WR WKH ) R[\VSRUXP I VS O\FRSHUVLFL LVRODWHV $OO LVRODWHV LQ 9&* FRPSULVH )2& 9 ,VRODWHV LQ WKLV OLQHDJH IRUPHG D FODGH H[KLELWLQJ WKH ORQJHVW EUDQFK OHQJWK FRPSDUHG WR DOO RWKHU FODGHV UHSUHVHQWLQJ ) R[\VSRUXP I VS FEHQVH LVRODWHV ,Q IDFW LWV OHQJWK ZDV FRPSDUDEOH WR WKH EUDQFK OHQJWKV RI WKH FODGHV FRQWDLQLQJ WKH LVRODWHV IURP WKH RWKHU IRUPDH VSHFLDOHV )LJXUH f $GGLWLRQDOO\ LVRODWHV FRPSULVLQJ WKLV OLQHDJH KDG FRHIILFLHQWV RI VLPLODULW\ ZKLFK GLG QRW DOLJQ FORVHO\ WR DQ\ RWKHU OLQHDJH )RU H[DPSOH LVRODWHV LQ )2& 9 KDG FRHIILFLHQWV RI VLPLODULW\ UDQJLQJ IURP WR FRPSDUHG WR LVRODWHV LQ )2& IURP WR FRPSDUHG WR LVRODWHV LQ )2& ,, DQG IURP WR FRPSDUHG WR WKH ) R[\VSRUXP I VS O\FRSHUVLFL LVRODWHV 7KLV JURXS KDG D ODUJH QXPEHU RI OLQHDJHVSHFLILF DOOHOHV DW VHYHUDO 5)/3 ORFL ZKLFK DFFRXQWHG IRU LWV UHODWLYH ODFN RI VLPLODULW\ WR RWKHU ) R[\VSRUXP I VS FEHQVH LVRODWHV $OO RI WKH LVRODWHV LQ 9&* KDYH D YHU\ OLPLWHG JHRJUDSKLFDO GLVWULEXWLRQ ,VRODWHV LQ WKLV JURXS DOVR GR QRW IRUP FKODP\GRVSRUHV 5& 3ORHW] XQSXEOLVKHGf DV RWKHU ) R[\VSRUXP LVRODWHV GR ZKLFK LV XVXDOO\ D GHILQLQJ WUDLW IRU WKH VSHFLHV )2& 9,, FRQVLVWHG RI WKH PDMRULW\ RI LVRODWHV LQ 9&* )2& ; FRQVLVWHG RI WZR DGGLWLRQDO LVRODWHV LQ 9&* 6LPLODU WR )2& 9 LVRODWHV FRPSULVLQJ WKHVH OLQHDJHV KDG FRHIILFLHQWV RI VLPLODULW\ ZKLFK GLG QRW DOLJQ FORVHO\ WR DQ\ RWKHU OLQHDJH

PAGE 48

6XUSULVLQJO\ WKH 5)/3 PXOWLORFXV KDSORW\SHV RI LVRODWHV ZLWKLQ WKHVH OLQHDJHV FRQVLVWHG RI DOOHOHV VLPLODU WR WKRVH LQ ERWK )2& DQG ,, LQ DGGLWLRQ WR VRPH OLQHDJHVSHFLILF DOOHOHV 7DEOH f $ UHSUHVHQWDWLYH )2& 9,, LVRODWH 3K/f VKDUHG b f RI SRO\PRUSKLF DOOHOHV ZLWK LVRODWHV LQ )2& ,, DQG b f RI WKHVH DOOHOHV ZLWK LVRODWHV LQ )2& %\ FRQWUDVW WKH LVRODWHV LQ )2& GLVSOD\HG RQO\ b f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f ,VRODWHV LQ 9&*V DQG IRUPHG RQH JURXS ,VRODWHV LQ 9&*V DQG IRUPHG D VHFRQG JURXS ,VRODWHV LQ 9&*V DQG IRUPHG D WKLUG JURXS ,Q JHQHUDO WKHVH JURXSLQJV DOLJQHG ZLWK WKRVH EDVHG RQ 5)/3 DQDO\VLV RI VLQJOHFRS\ ORFL DOWKRXJK PRUH JURXSV FRXOG EH UHVROYHG XVLQJ WKH ODWWHU PHWKRG 0HDVXUHV RI JDPHWLF GLVHTXLOLEULXP ZHUH SHUIRUPHG IRU DOOHOHV DW QLQH ORFL 7DEOH f (YHQ WKRXJK PDQ\ RI WKH LQGLYLGXDO FRPSDULVRQV ZHUH QRW VLJQLILFDQW RI WKH SDLUZLVH FRPSDULVRQV DPRQJ DOOHOHV DW GLIIHUHQW 5)/3 ORFL VKRZHG VLJQLILFDQW

PAGE 49

7DEOH $ FRPSDULVRQ RI DOOHOLF GDWD IRU LVRODWHV LQ )2& )2& ,, DQG )2& 9,, ,VRODWH 9&*f\ $OOHOHV $ 9&* f ,& 9&* f 3K/ 9&* f n,VRODWHV $ DQG ,& UHSUHVHQW WKH WZR ODUJHVW 5)/3 KDSORW\SHV LQ )2& DQG )2& ,, UHVSHFWLYHO\ DQG 3K/ UHSUHVHQWV DQ LVRODWH LQ )2& 9,, ]$OOHOHV IRXQG LQ WKH PRVW FRPPRQ KDSORW\SHV ZHUH JLYHQ WKH QXPEHU $OOHOHV LQ WKH VHFRQG PRVW FRPPRQ KDSORW\SH LI GLIIHUHQW IURP KDSORW\SH ZHUH JLYHQ WKH QXPEHU $OOHOHV LQ LVRODWH 3K/ LI GLIIHUHQW WKDQ WKRVH LQ WKH WZR PRVW FRPPRQ KDSORW\SHV ZHUH JLYHQ WKH QXPEHU

PAGE 50

0ROHFXODU :HLJKW .E ,VRODWH 9&* PW'1$ *URXS 257= 1: 3K '$9$2 6" 0: 6713 '$92 670 1 7%5 (6 -/7+ 6& ) R[\VSRUXP I VS O\FRSHUVLFL RWKHU )LJXUH 0LWRFKRQGULDO '1$ KDSORW\SHV RI VHOHFWHG )XVDULXP R[\VSRUXP I FEHQVH LVRODWHV

PAGE 51

7DEOH &ORQHFRUUHFWHG PHDVXUHPHQWV RI JDPHWLF GLVHTXLOLEULXP DPRQJ SDLUV RI DOOHOHV LQ D ZRUOGZLGH FROOHFWLRQ RI )XVDULXP R[\VSRUXP IVS FEHQVH 3UREH (Q]\PH &RPELQDWLRQV +DH +, 'P (FR59 (FR59 +DH LQ (FR59 (FR59 (FR59 'P f f f f f f f f +DH8O f f f f f f f 'P fr f fr f f f (FR59 f f f f f (FR59 f f f f +DH ', f f f (FR59 f f (FR59 f 1RWH $OO WHVWV KDG DW OHDVW PXOWLORFXV KDSORW\SHV LQ WKH FRPSDULVRQV DQG D PD[LPXP RI Q 7KH ILUVW URZ LQ HDFK ER[ UHSUHVHQWV WKH QXPEHU RI VLJQLILFDQW FKLVTXDUH WHVWV 3f EHWZHHQ LQGLYLGXDO DOOHOHV DW GLIIHUHQW 5)/3 ORFL RYHU WKH WRWDO QXPEHU RI WHVWV PDGH IRU HDFK SDLUZLVH FRPSDULVRQ WKH VHFRQG URZ LQ HDFK ER[ UHSUHVHQWV WKH UHVXOWV RI D FKL VTXDUH WHVW IRU WKH VLJQLILFDQFH RI DVVRFLDWLRQ EHWZHHQ DOO DOOHOHV DW WKH WZR ORFL 7KH QXPEHU LQ WKH SDUHQWKHVLV LV WKH GHJUHHV RI IUHHGRP IRU WKH WHVW $OO DUH VLJQLILFDQW DW 3 DQG 3 H[FHSW WKRVH QRWHG ZLWK r

PAGE 52

QRQUDQGRP DVVRFLDWLRQV DW WKH b OHYHO 0DQ\ LQGLYLGXDO DOOHOH FRPELQDWLRQV ZHUH QRW SUHVHQW LQ WKH SRSXODWLRQ EXW WKH FKLVTXDUH WHVW IRU WKHVH FRPELQDWLRQV ZDV QRW VLJQLILFDQW EDVHG RQ WKH RYHUDOO DOOHOLF IUHTXHQF\ WKH H[SHFWHG QXPEHU RI WKHVH FRPELQDWLRQV ZDV VPDOO ,Q FRQWUDVW IRU WKH PRVW FRPPRQ DOOHOHV WKH ODUJH QXPEHU RI QRQUDQGRP DVVRFLDWLRQV ZDV LQGLFDWHG E\ VLJQLILFDQWO\ ODUJHU RU VPDOOHU QXPEHUV RI REVHUYDWLRQV ZKHQ FRPSDUHG WR WKH H[SHFWHG IUHTXHQF\ RI WKHVH FRPELQDWLRQV 'LVFXVVLRQ 6HYHUDO OLQHV RI HYLGHQFH VXSSRUW WKH FRQFHSW WKDW ) R[\VSRQWP I VS FEHQVH KDV D FORQDO SRSXODWLRQ VWUXFWXUH LQ OLQH ZLWK FULWHULD HVWDEOLVKHG E\ 7LED\UHQF HW DO f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f DOVR DUH LQGLFDWLYH RI D FORQDOO\ UHSURGXFLQJ RUJDQLVP 7LED\UHQF HW DO f 7KLV VWXG\ FRQILUPV WKDW LQ SK\ORJHQHWLF DQDO\VLV RI ) R[\VSRUXP I VS FEHQVH XVLQJ SDUVLPRQ\ 9&* LV D VWURQJ SUHGLFWRU RI FODGLVWLF JURXSLQJV )XUWKHU GLIIHUHQWLDWLRQ LQWR OLQHDJHV PD\ EH GRQH EDVHG RQ FRHIILFLHQWV RI VLPLODULW\ RI 5)/3 KDSORW\SHV $W OHDVW 9&*V KDYH EHHQ GHVFULEHG IRU ) R[\VSRUXP I VS FEHQVH 3ORHW] DQG &RUUHOO

PAGE 53

3ORHW] 3ORHW] HW DO 3ORHW] 3ORHW] HW DO f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f $OWKRXJK 9&*V DUH JRRG LQGLFDWRUV RI JHQHWLF VLPLODULW\ DPRQJ WKH LQGLYLGXDOV FRPSULVLQJ WKHP WKH\ GR QRW SURYLGH LQIRUPDWLRQ UHJDUGLQJ WKH JHQHWLF VLPLODULW\ WR LQGLYLGXDOV LQ GLIIHUHQW 9&*V ,Q IDFW WKLV VWXG\ VKRZV WKDW LVRODWHV EHORQJLQJ WR GLIIHUHQW 9&*V FRXOG KDYH LGHQWLFDO RU QHDUO\ LGHQWLFDO 5)/3 KDSORW\SHV :LWK WZR H[FHSWLRQV WKH HQWLUH FROOHFWLRQ RI LVRODWHV FRQVLVWHG RI RQO\ WHQ GLVWLQFW FORQDO OLQHDJHV 7DEOH f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f DQG 5$3' SURILOH %HQWOH\ HW DO f %DVHG RQ WKHLU HOHFWURSKRUHWLF NDU\RW\SH %RHKP HW DO f GLYLGHG LVRODWHV LQWR

PAGE 54

WZR PDMRU JURXSV *URXS FRQWDLQHG LVRODWHV IURP 9&*V DQG DQG ZDV FKDUDFWHUL]HG E\ KLJK FKURPRVRPH QXPEHU DQG ODUJH UHODWLYH JHQRPH VL]H > PHJDEDVH SDLUV 0ESf@ *URXS ,, FRQWDLQHG LVRODWHV IURP 9&*V DQG ZKLFK KDG FRUUHVSRQGLQJO\ IHZHU FKURPRVRPHV DQG VPDOOHU JHQRPH VL]HV 0ESf 8VLQJ 5$3' DQDO\VLV %HQWOH\ HW DO f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b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

PAGE 55

VWXGLHV %HQWOH\ HW DO %RHKP HW DO f 8QOLNH SUHYLRXV LQYHVWLJDWLRQV LVRODWHV LQ 9&*V DQG HDFK IRUPHG LQGLYLGXDO FODGHV ZLWK ERRWVWUDS YDOXHV JUHDWHU WKDQ b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f 2QH LQWHUSUHWDWLRQ RI WKHVH UHVXOWV LV WKDW LVRODWHV EHORQJLQJ WR WKH GLVVLPLODU JURXSV DFTXLUHG WKHLU DELOLW\ WR EH SDWKRJHQLF RQ EDQDQDV LQGHSHQGHQWO\ )2& 9 FRQWDLQV LVRODWHV IURP WKH 0LVXNX +LOOV LQ 0DODZL D UHODWLYHO\ VPDOO DUHD DSSUR[LPDWHO\ VTXDUH NLORPHWHUVf RQ WKH FRXQWU\fV QRUWKHUQ ERUGHU ZLWK 7DQ]DQLD 3ORHW] HW DO f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f 'LYHUVH JHQRW\SHV RI EDQDQD DUH QRZ IRXQG LQ (DVW $IULFD PDQ\ RI ZKLFK DUH IRXQG QRZKHUH HOVH LW LV SRVVLEOH WKDW WKH SDWKRJHQ ZDV PRYHG IURP 6RXWKHDVW $VLD RQ WKH EDQDQDV LQWURGXFHG WR $IULFD DQG DV D UHVXOW RI

PAGE 56

PXWDWLRQ DQG VHOHFWLRQ RU WKURXJK DGDSWDWLRQ WR HQGHPLF EDQDQDV LVRODWHV LQ 9&* PD\ KDYH GLYHUJHG IURP WKHLU $VLDQ SURJHQLWRUV )2& 9,, DQG ; FRQWDLQ LVRODWHV LQ 9&* 7KH 5)/3 KDSORW\SH RI LVRODWHV EHORQJLQJ WR WKHVH JURXSV FDUU\ DQ DVVRUWPHQW RI DOOHOHV IURP WKH WZR PDMRU OLQHDJHV )2& DQG ,,f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f ,GHDOO\ EUHHGLQJ SURJUDPV FRXOG VFUHHQ QHZ K\EULGV DJDLQVW LVRODWHV IURP HDFK FORQDO OLQHDJH WR LQFUHDVH WKH SUREDELOLW\ RI GHYHORSLQJ FXOWLYDUV WKDW UHVLVW JHQHWLFDOO\ GLVWLQFW SRSXODWLRQV RI WKH SDWKRJHQ

PAGE 57

&+$37(5 *(1(7,& 9$5,$7,21 ,1 7:2 +21'85$1 ),(/' 3238/$7,216 2) )86$580 2;<632580 ) 63 &8%(16( ,QWURGXFWLRQ )XVDULXP R[\VSRUXP 6FKOHFKWHQG )U IRUPD VSHFLDOLV FEHQVH () 6Pf :& 6\QGHU t + 1 +DQV LV WKH FDXVDO DJHQW RI )XVDULXP ZLOW RI EDQDQD D GLVHDVH ZKLFK KDV EHHQ UHVSRQVLEOH IRU VLJQLILFDQW \LHOG ORVVHV LQ EDQDQD 7KLV GLVHDVH KDG D GHYDVWDWLQJ HIIHFW RQ WKH H[SRUW EDQDQD WUDGH LQ &HQWUDO $PHULFD GXULQJ WKH HDUO\ SDUW RI WKLV FHQWXU\ 6WRYHU f %DQDQDV ZHUH LQWURGXFHG WR +RQGXUDV ZHOO RYHU \HDUV DJR DQG FRPPHUFLDO WUDGH EHJDQ DERXW 8QWLO WKH ODWH V WKH UDFH VXVFHSWLEOH EDQDQD FXOWLYDU *URV 0LFKHO ZDV WKH PRVW ZLGHO\ SODQWHG FRPPHUFLDO FXOWLYDU LQ +RQGXUDV DQG HOVHZKHUH 6WRYHU f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f DQG WKHVH LVRODWHV EHORQJ WR RQO\ WKUHH YHJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*Vf DQG f 3ORHW] f 5DFH LVRODWHV RFFXU LQ DOO WKUHH

PAGE 58

RI WKHVH 9&*V ZKLOH UDFH LVRODWHV DSSHDU WR EH OLPLWHG WR 9&* ,Q D QXPEHU RI VXEWURSLFDO FRXQWULHV $XVWUDOLD 6RXWK $IULFD &DQDU\ ,VODQGV DQG 7DLZDQf PDQ\ FRPPHUFLDO SODQWDWLRQV RI &DYHQGLVK FXOWLYDUV VXFFXPEHG WR )XVDULXP ZLOW VKRUWO\ DIWHU WKH\ ZHUH SODQWHG 3ORHW] f 0RUH UHFHQWO\ LVRODWHV LQ 9&*V DQG KDYH FDXVHG GLVHDVH RQ &DYHQGLVK FXOWLYDUV LQ WKH WURSLFV ,VRODWHV FDSDEOH RI FDXVLQJ GLVHDVH RQ &DYHQGLVK FXOWLYDUV ZHUH LGHQWLILHG E\ 6X HW DO f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f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f

PAGE 59

$ VWXG\ RI D ZRUOG FROOHFWLRQ RI ) R[\VSRUXP I VS FEHQVH LVRODWHV UHSUHVHQWLQJ YHJHWDWLYH FRPSDWLELOW\ JURXSV 9&*Vf UHYHDOHG WKDW WKLV SDWKRJHQ LV FRPSULVHG RI DW OHDVW FORQDO OLQHDJHV &KDSWHU f 7KH JUHDWHVW GLYHUVLW\ RI 9&*V DQG OLQHDJHV DUH SUHVHQW LQ 0DOHVLD 3ORHW] &KDSWHU f ZKLFK LV DOVR WKH UHJLRQ RI RULJLQ RI WKH KRVW 6LPPRQGV f $OWKRXJK WKH OHYHO RI JHQRW\SLF GLYHUVLW\ WKDW ZDV UHSUHVHQWHG LQ D ZRUOGZLGH FROOHFWLRQ RI ) R[\VSRUXP I VS FEHQVH LVRODWHV KDV EHHQ GHVFULEHG SUHYLRXVO\ %HQWOH\ HW DO &KDSWHU f WKH DPRXQW RI GLYHUVLW\ IRXQG LQ LQGLYLGXDO ILHOG SRSXODWLRQV RI WKH SDWKRJHQ LV XQNQRZQ 6LPLODUO\ IDFWRUV WKDW PD\ DIIHFW WKH OHYHOV RI GLYHUVLW\ ZLWKLQ D ILHOG KDYH QRW EHHQ H[SORUHG IRU ) R[\VSRUXP I VS FEHQVH 0F'RQDOG HW DO f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f GHWHUPLQH WKH DPRXQW RI JHQRW\SLF GLYHUVLW\ LQ ILHOG SRSXODWLRQV RI WKH SDWKRJHQ LQ +RQGXUDV DQG LLf GHWHUPLQH ZKHWKHU VSHFLILF JHQRW\SHV RI WKH SDWKRJHQ SUHIHUHQWLDOO\ LQIHFW VSHFLILF KRVW JHQRW\SHV

PAGE 60

0DWHULDOV DQG 0HWKRGV 'HVFULSWLRQ RI 6DPSOLQJ 3URFHGXUH 6DPSOLQJ RI ) R[\VSRUXP I VS FEHQVH LVRODWHV ZDV FRQGXFWHG LQ WZR DGMDFHQW GLVHDVH VFUHHQLQJ ILHOGV )LHOGV DQG f DW WKH EDQDQD EUHHGLQJ VWDWLRQ RSHUDWHG E\ WKH )XQGDFLQ +RQGXUHD GH ,QYHVWLJDFLQ $JUFROD )+,$f LQ /D /LPD +RQGXUDV +LVWRULFDOO\ )LHOG KDG EHHQ SODQWHG WR FXOWLYDU +LJKJDWH ZKLFK LV VXVFHSWLEOH WR UDFH DQG KDG D KLJK LQFLGHQFH RI SODQWV ZLWK )XVDULXP ZLOW EDVHG RQ YLVXDO DVVHVVPHQW RI V\PSWRPV 7KHUHIRUH LW ZDV DQ RSWLPDO ORFDWLRQ WR HYDOXDWH WKH UHVLVWDQFH WR UDFH LQ VHYHQ RI )+,$fV WHWUDSORLG K\EULGV 5LYHUD SHUV FRPPf 3ULRU WR SODQWLQJ WKH WULDO LQRFXOXP OHYHOV ZHUH DUWLILFLDOO\ LQFUHDVHG LQ WKH ILHOG XVLQJ D QXPEHU RI SURFHGXUHV )LUVW IUHVKO\ LQIHFWHG SVHXGRVWHP DQG UKL]RPH WLVVXH IURP +LJKJDWH SODQWV LQIHFWHG ZLWK ) R[\VSRUXP I VS FEHQVH ZDV XQLIRUPO\ VSUHDG WKURXJKRXW WKH ILHOG 6XEVHTXHQWO\ WKH HQWLUH ILHOG ZDV SODQWHG WR WKH UDFH ,VXVFHSWLEOH FXOWLYDU 0DTXHR 3ULRU WR SODQWLQJ WKH 0DTXHR UKL]RPHV DSSUR[LPDWHO\ JUDP Jf FKXQNV RI V\PSWRPDWLF SVHXGRVWHP DQG UKL]RPH WLVVXH IURP +LJKJDWH ZHUH SODFHG LQ WKH ERWWRP RI HDFK KROH $IWHU D \HDU b RI WKH UHVXOWLQJ SODQWV H[KLELWHG LQWHUQDO V\PSWRPV RI )XVDULXP ZLOW YDVFXODU GLVFRORUDWLRQ RI UKL]RPHV DQG SVHXGRVWHPVf 3ULRU WR SODQWLQJ WKH GLVHDVH VFUHHQLQJ WULDO LQIHFWHG 0DTXHR SODQWV ZHUH LQFRUSRUDWHG LQWR WKH VRLO $GGLWLRQDOO\ SULRU WR SODQWLQJ HDFK H[SHULPHQWDO K\EULG J RI V\PSWRPDWLF 0DTXHR WLVVXHV ZHUH SODFHG LQ HDFK SODQWLQJ KROH 7KH H[SHULPHQW ZDV DUUDQJHG LQ D UDQGRPL]HG FRPSOHWH EORFN GHVLJQ ZLWK QLQH WUHDWPHQWV FXOWLYDUVf DQG VHYHQ UHSOLFDWLRQV 7DEOH f 7KH EDQDQD SODQWV XVHG LQ WKH H[SHULPHQWV ZHUH GHULYHG IURP WLVVXH FXOWXUH 7KH WUHDWPHQWV ZHUH SODQWHG LQ WKH VDPH

PAGE 61

7DEOH *HQRW\SHV RI EDQDQD SODQWHG LQ )LHOG +\EULG *HQRW\SH )+,$ $$% [ $$ $$$% 'ZDUI 3UDWD [ 3LVDQJ -DUL %XD\D GHULYDWLYHf )+,$ $$$ [ $ $ $$$$ :LOOLDPVr[ $$f [ $$f ERWK DUH 3LVDQJ -DUL %XD\D GHULYDWLYHVf )+,$ $%% [ %% $%%% $%%% [ $$ $$%% $$% RU $%% [ $$ $$$ $$% RU $%% &DUGDED [ %% $%%% [ $$f >VHOHFWHG WULSORLG [ $$f@ )+,$ $$% ; $$ $$$% 0DTXHR [ %ODFN VLJDWRND UHVLVWDQW 0XUD DFXPLQDWD VXEVS EXULPDQLFDf )+,$ $$% [ $$ $$$% 6DPH SDUHQWDJH DV )+,$ EXW WKLV VHOHFWLRQ LV IDVWHU WR UDWRRQ )+,$ $$$ [$$ $$$$ +LJKJDWH [ [ f )+,$ $$$ [ $$ $$$$ 6DPH SDUHQWDJH DV )+,$ EXW GLIIHUHQW VHOHFWLRQ +LJKJDWH $$$ UDFH VXVFHSWLEOH FRQWURO UDFH UHVLVWDQWff :LOOLDPV $$$ UDFH DQG UDFH UHVLVWDQW FRQWUROf %OXJJRH $%% UDFH VXVFHSWLEOH FRQWURO UDFH UHVLVWDQWff 1RWH :LWK WKH H[FHSWLRQ RI )+,$ DOO )+,$ K\EULGV DUH HLWKHU LQWHUVSHFLILF RU LQWUDVSHFLILF WHWUDSORLGV REWDLQHG E\ SHUIRUPLQJ FURVVHV RQ 0XVD DFXPLQDWD $ $f DQG 0 EDOELVLDQD %%f SDUHQWDO OLQHV +LJKJDWH :LOOLDPV DQG %OXJJRH DUH EDQDQD FXOWLYDUV XVHG DV KRVW GLIIHUHQWLDOV WR GLIIHUHQWLDWH DPRQJ WKH WKUHH UDFHV RI )XVDULXP R[\VSRUXP I VS FEHQVH r ,W LV TXHVWLRQDEOH DV WR ZKHWKHU RU QRW :LOOLDPV ZDV DFWXDOO\ WKH SDUHQW RI WKLV FURVV GXH WR WKH FXOWLYDU :LOOLDPV H[WUHPH VWHULOLW\

PAGE 62

URZV LQ ZKLFK WKH FXOWLYDU 0DTXHR KDG EHHQ SODQWHG (DFK SORW FRQWDLQHG WKUHH SODQWV VSDFHG DW P DQG WKH HQWLUH H[SHULPHQWDO DUHD HQFRPSDVVHG DSSUR[LPDWHO\ P 7KH ILHOG FRQVLVWHG RI VHYHQ URZV WKUHH RI WKHVH FRQWDLQHG WKH VHYHQ UHSOLFDWLRQV RI WKH H[SHULPHQW SODQWHG LQ D VHUSHQWLQH IDVKLRQ WKURXJKRXW WKH SORW ([SHULPHQWDO URZV ZHUH VHSDUDWHG IURP RQH DQRWKHU E\ D VLQJOH URZ RI WKH UDFH VXVFHSWLEOH FXOWLYDU %OXJJRH 7DEOH f 7KH ILUVW VDPSOHV ZHUH FROOHFWHG LQ )HEUXDU\ XVLQJ D QRQGHVWUXFWLYH PHWKRG ZKLFK FRQVLVWHG RI UHPRYLQJ WKH ILUVW IHZ RXWHU OD\HUV RI SVHXGRVWHP WLVVXH IURP V\PSWRPDWLF DQG DV\PSWRPDWLF SODQWV 3ORHW] f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f 6HYHQW\VL[ SHUFHQW RI WKH SODQWV VKRZHG YLVLEOH V\PSWRPV RI )XVDULXP ZLOW GXULQJ DQ LQLWLDO DVVHVVPHQW RI WKH ILHOG 6\PSWRPDWLF SODQWV ZHUH IDLUO\ HYHQO\ GLVWULEXWHG WKURXJKRXW WKH ILHOG DQG EDVHG RQ WKLV GLVWULEXWLRQ D V\VWHPDWLF VDPSOLQJ VWUDWHJ\ ZDV IROORZHG LQ ZKLFK RQH RI HYHU\ WKUHH SODQWV ZDV VDPSOHG IROORZLQJ D VHUSHQWLQH SDWWHUQ WKURXJK WKH SORW 6DPSOHV ZHUH

PAGE 63

7DEOH ([SHULPHQWDO GHVLJQ RI EDQDQD SODQWV LQ URZV RI )LHOG 5RZ 1XPEHUn % ) % ) % ) % % ) % ) % ) % % ) % ) % ) % % ) % ) % ), % % ) % ) % ), % % ) % ) % ), % % ) % : % ) % % ) % : % ) % % ) % : % ) % % ) % + % ) % % ) % + % ) % % ) % + % ) % % ), % ) % ) % % ), % ) % ) % % ), % ) % ) % % + % ) % ) % % + % ) % ) % % +] % ) % ) % % : % ) % ) % % : % ) % ) % % Z % ) % ) % % ) % +] % + % % ) % +] % + % % ) % + % + % % ) % ) % ) % % ) % ) % ) % % ) % ) % ) % % : % ) % ) % % : % ) % ) % % : % ) % ) % % ) % ) % : % % ) % ) % : % % ) % ) % : % % ) % ) % ), % % ) % ) % ), % % ) % ) % ), % % ) % : % ) % % ) % : % ) % % ) % Z % ) % % ) % ) % : % % ) % ) % : %

PAGE 64

7DEOH f§FRQWLQXHG 5RZ 1XPEHU\ % ) % ) % : % % ), % ), % ) % % ), % ), % ) % % ), % ), % ) % % ) % ) % + % % ) % ) % + % % ) % ) % + % % )7 % + % ) % % +] % + % ) % % + % + % ) % % ) % ), % ), % % ) %] ), % ), % % ) % ), % ), % % ) % ) % ) % % ) % )= % ) % % ) % ) % )] % % ) % ) % ) % % ) % ) % ) % % ) % ) % ) % % : % ) % ) % % : % ) % ) % % Z % ) % ) % \ (DFK OHWWHU RU OHWWHUQXPHUDO FRPELQDWLRQ UHSUHVHQWV D VLQJOH SODQW EROG W\SH LQGLFDWHV SODQWV WKDW ZHUH VDPSOHG IRU WKH SUHVHQFH RI )XVFWULXP R[\VSRPP I VS FEHQVH % FXOWLYDU %OXJJRH + FXOWLYDU +LJKJDWH : FXOWLYDU :LOOLDPV ) )+,$ ) ))+$ ) )+,$ ) )+,$ ) )+,$ ) )+,$ ) )+,$ ] LQGLFDWHV PXOWLSOH VDPSOHV ZHUH REWDLQHG IURP WKHVH SODQWV

PAGE 65

7DEOH )LHOG GHVLJQ RI EDQDQD SODQWV LQ URZV RI WKH UDFH GLVHDVH VFUHHQLQJ SORW 3ODQW 1XPEHU 5RZ 1XPEHU ' ' ' ' ' ' ' + + ' ' ' ' ' ' ' ' ' ' ' + ' ' ' + ' ' ' + ' ' ' ' ' ' + + ' ' ' + + + ' + + + ' + + ' + + + ' + + ' + + + + ' ' ' + ' ' + + + ' ' ' + ' + ' + + ' ' ' + ' + ' ' + ' ' ' + ' ' ' + ' + ' ' + ' ' ' + ' ' + ' + + ' ' + ' + ' + + + ' ' ' ' ' ' ' + ' ' ' + + ' ' ' + + + + + ' ' ' + ' ' + + + + + ' ' + + ' ' ' + + ' + + + ' ' ' ' + + + ' + + ' ]7KH HQWLUH ILHOG SODQWHG WR WKH UDFH VXVFHSWLEOH FXOWLYDU %OXJJRH (DFK OHWWHU UHSUHVHQWV DQ LQGLYLGXDO SODQW + DV\PSWRPDWLF SODQWV DQG V\PSWRPDWLF SODQWV EDVHG RQ YLVXDO DVVHVVPHQW VDPSOHV ZHUH REWDLQHG IURP SODQWV LQ EROG W\SH 7RWDO SODQWV 7RWDO bf 7RWDO + bf

PAGE 66

FROOHFWHG LQ 0DUFK DQG FRQVLVWHG RI GHVWUXFWLYH VDPSOLQJ RI LQQHU UKL]RPH DQG SVHXGRVWHP WLVVXHV 6DPSOLQJ LQFOXGHG ERWK V\PSWRPDWLF DQG QRQV\PSWRPDWLF WLVVXH )XQJDO ,VRODWLRQV 7LVVXH VDPSOHV ZHUH DOORZHG WR DLU GU\ IRU WR GD\V SULRU WR WUDQVSRUW WR )ORULGD ,Q )ORULGD HDFK VDPSOH ZDV FXW LQWR DSSUR[LPDWHO\ HLJKW FP SLHFHV ZKLFK HDFK FRQWDLQHG DW OHDVW RQH GLVFRORUHG DQG SUHVXPDEO\ LQIHFWHG YDVFXODU HOHPHQW 7KH WLVVXH SLHFHV ZHUH VXUIDFH GLVLQIHVWHG E\ SODFLQJ WKHP LQ b HWKDQRO IRU VHFRQGV IROORZHG E\ D PLQ VRDN LQ b EOHDFK DQG D ILQDO ULQVH IRU DW OHDVW PLQXWH LQ VWHULOH ZDWHU 3LHFHV ZHUH DLU GULHG LQ D ODPLQDU IORZ KRRG DQG WKHQ SODFHG LQWR WZR SHWUL GLVKHV WR SURYLGH IRXU SLHFHV RI WLVVXH LQ HDFK GLVK 2QH RI WZR PHGLD ZHUH SRXUHG RYHU WKH WLVVXH SLHFHV 0HGLXP FRQVLVWHG RI DPHQGHG ZDWHU DJDU VWUHSWRP\FLQ VXOIDWH PJ/ 6LJPD &KHPLFDO &RPSDQ\ 6W /RXLV 0Rf ULIDPSLFLQ PJ/ 6LJPDf DQG O RI GDQLWRO/ RI PHGLXP &KHYURQ &RUSRUDWLRQ 6DQ )UDQFLVFR &$f 0HGLXP FRQVLVWHG RI DPHQGHG SRWDWR GH[WURVH DJDU 3'$f FRQWDLQLQJ VWUHSWRP\FLQ PJ/f ULIDPSLFLQ PJ/f GDQLWRO c$// DQG WHULJLWRO 13 / 6LJPDf 0HGLXP DOORZHG IRU TXLFN JURZWK RI )XVDULXP VSS ZKLOH VXSSUHVVLQJ WKH JURZWK RI RWKHU PLFURRUJDQLVPV +RZHYHU LGHQWLILFDWLRQ RI ) R[\VSRUXP LVRODWHV ZDV GLIILFXOW RQ WKLV PHGLXP EHFDXVH RI SRRU P\FHOLDO JURZWK DQG VSRUH IRUPDWLRQ 0HGLXP DOORZHG IRU EHWWHU LGHQWLILFDWLRQ RI ) R[\VSRUXP FRORQLHV EXW WKH DGGLWLRQDO QXWULHQWV ZHUH DOVR PRUH FRQGXFLYH WR WKH JURZWK RI RWKHU PLFURRUJDQLVPV 3ODWHV ZHUH LQFXEDWHG DW URRP WHPSHUDWXUH DQG ZLWKLQ GD\V P\FHOLDO JURZWK ZDV REVHUYHG RQ WLVVXH SLHFHV 3XWDWLYH FRORQLHV RI )XVDULXP VSS ZHUH VHOHFWHG EDVHG RQ YLVXDO JURZWK FKDUDFWHULVWLFV DQG WKH IRUPDWLRQ RI PLFURFRQLGLD RU PDFURFRQLGLD $JDU VHFWLRQV PPf ZHUH H[FLVHG IURP WKHVH FRORQLHV DQG SODFHG RQ WKH

PAGE 67

UHFLSURFDO PHGLXP IURP PHGLXP WR DQG YLFH YHUVDf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f $SSUR[LPDWHO\ SJ RI '1$ ZHUH GLJHVWHG ZLWK DW OHDVW XQLWV RI HLWKHU +DH,,, RU +LQGOOO UHVWULFWLRQ HQ]\PHV %HWKHVGD 5HVHDUFK /DERUDWRULHV *DLWKHUVEXUJ 0'f DQG LQFXEDWHG IRU DW OHDVW KRXUV DW r& 5HVWULFWLRQ IUDJPHQWV ZHUH VHSDUDWHG E\ HOHFWURSKRUHVLV LQ b DJDURVH IRU !"L,,,GLJHVWHG '1$ RU b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t 6FKXHOO ,QF .HHQH 1+f XVLQJ WKH FDSLOODU\ WUDQVIHU PHWKRG 6DPEURRN HW DO f 7KH '1$ WUDQVIHU SURFHHGHG IRU DW OHDVW KRXUV DQG WKH '1$ ZDV LPPRELOL]HG E\ 89 FURVVOLQNLQJ 89 QP FURVVOLQNHU PRGHO )% 89;/ )LVKHU 6FLHQWLILF 3LWWVEXUJK 3$f

PAGE 68

0LWRFKRQGULDO '1$ PW'1$f ZDV LVRODWHG IURP +RQGXUDQ LVRODWH 6 9&* f IROORZLQJ D SURFHGXUH GHVFULEHG SUHYLRXVO\ .LVWOHU DQG /HRQJ f 7RWDO PW'1$ ZDV ODEHOHG XVLQJ UDQGRP KH[DPHU SULPHUV WR LQFRUSRUDWH IOXRUHVFHLQG873 '1$ ODEHOLQJ K\EULGL]DWLRQ DQG GHWHFWLRQ IROORZHG WKH SURFHGXUHV SURYLGHG E\ WKH PDQXIDFWXUHU 'XSRQW 1(1 5HQDLVVDQFH ( GX 3RQW GH 1HPRXUV t &R ,QF %RVWRQ 0$f 3UHK\EULGL]DWLRQ K\EULGL]DWLRQ DQG ZDVKLQJ ZHUH SHUIRUPHG DW r& XVLQJ D +\EDLG K\EULGL]DWLRQ RYHQ 'RW 6FLHQWLILF ,QF )OLQW 0,f 0HPEUDQHV ZHUH SODFHG EHWZHHQ DFHWDWH VKHHWV DQG H[SRVHG WR [UD\ ILOP IRU WR PLQXWHV 3UHOLPLQDU\ ZRUN RQ D ZRUOGZLGH FROOHFWLRQ RI ) R[\VSRUXP I VS FEHQVH LVRODWHV UHYHDOHG WKDW PLWRFKRQGULDO KDSORW\SHV ZHUH FRUUHODWHG ZLWK 9&* DQG LVRODWHV WHVWHG ZLWKLQ D 9&* KDG LGHQWLFDO PLWRFKRQGULDO KDSORW\SHV .LVWOHU DQG 0RPRO &KDSWHU f ,VRODWHV EHORQJLQJ WR WKH WKUHH 9&*V IRXQG LQ +RQGXUDV DQG f FRXOG EH GLIIHUHQWLDWHG EDVHG RQ WKHLU XQLTXH PLWRFKRQGULDO SURILOHV REWDLQHG E\ SURELQJ EORWV RI WRWDO DHOOOGLJHVWHG '1$ ZLWK PW'1$ LVRODWHG IURP +RQGXUDQ LVRODWH 6 8VLQJ WKLV PHWKRG D PLWRFKRQGULDO KDSORW\SH ZDV DVVLJQHG WR HDFK LVRODWH +RQGXUDQ LVRODWHV ZKLFK KDG PLWRFKRQGULDO SURILOHV LGHQWLFDO WR LVRODWHV LQ 9&* ZHUH GHVLJQDWHG PLWRFKRQGULDO KDSORW\SH PWKDSf WKRVH ZKLFK KDG PLWRFKRQGULDO SURILOHV LGHQWLFDO WR LVRODWHV LQ 9&* RU 9&* ZHUH GHVLJQDWHG PWKDS DQG PWKDS UHVSHFWLYHO\ 6LPLODUO\ ) R[\VSRUXP I VS FEHQVH LVRODWHV ZHUH GLIIHUHQWLDWHG EDVHG RQ WKHLU '1$ ILQJHUSULQW E\ GLJHVWLQJ WRWDO '1$ ZLWK WKH UHVWULFWLRQ HQ]\PH +LQGOOO DQG SURELQJ ZLWK FORQH S(
PAGE 69

DQ RSHQ UHDGLQJ IUDPH HQFRGLQJ D SRO\SHSWLGH ZLWK b DPLQR DFLG LGHQWLW\ WR WKH WUDQVSRVDVH RI WKH 0DJQDSRUWKH JUVHD WUDQVSRVRQ 3RW )DUPDQ HWDO 3HWWZD\ DQG .LVWOHU XQSXEOLVKHGf $OO SUHYLRXVO\ WHVWHG LVRODWHV IURP D ZRUOGZLGH FROOHFWLRQ RI ) R[\VSRUXP I VS FEHQVH KDG XQLTXH '1$ ILQJHUSULQWV XVLQJ S(
PAGE 70

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f 2I WKHVH LVRODWHV LVRODWHV ZHUH REWDLQHG IURP SODQWV WKDW ZHUH V\PSWRPDWLF GXULQJ ERWK VDPSOLQJ SHULRGV DQG UHSUHVHQW PXOWLSOH VDPSOHV WDNHQ IURP WKH VDPH SODQW $ PW'1$ KDSORW\SH ZDV DVVLJQHG WR HDFK LVRODWH E\ SURELQJ EORWV RI DHOOOGLJHVWHG '1$ IURP HDFK LVRODWH ZLWK ODEHOHG PW'1$ LVRODWHG IURP +RQGXUDQ LVRODWH 6 )LJXUH f ,VRODWHV FROOHFWHG IURP WKH ILHOG ZHUH DVVLJQHG D PW'1$ KDSORW\SH DV GHVFULEHG LQ WKH PDWHULDOV DQG PHWKRGV 7KH LQFLGHQFH DQG GLVWULEXWLRQ RI SDWKRJHQ PW'1$ KDSORW\SHV YDULHG ZLWK KRVW JHQRW\SH 7DEOH f 7KH +LJKJDWH DQG &DYHQGLVK FXOWLYDUV XVHG DV VXVFHSWLEOH DQG UHVLVWDQW H[SHULPHQWDO FRQWUROV KDG WKH KLJKHVW DQG ORZHVW LQFLGHQFH RI

PAGE 71

,VRODWH+RVW &XOWLYDU PW'1$ +DSORW\SH 0 %OXJJRH 8QNQRZQ W\SH %OXJJRH %OXJJRH %OXJJRH %OXJJRH \O %OXJJRH %OXJJRH %OXJJRH 8QNQRZQ W\SH %OXJJRH +LJKJDWH )+,$ +LJKJDWH )+,$ )+,$ )+,$ )+,$ )+,$ +LJKJDWH )+,$ )LJXUH PW'1$ KDSORW\SH RI VHOHFWHG +RQGXUDQ ILHOG LVRODWHV [ ,VRODWHV ZLWK GHVLJQDWLRQ ZHUH FROOHFWHG IURP WKH FXOWLYDU %OXJJRH LQ )LHOG \ ,VRODWHV ZLWK GHVLJQDWLRQ ZHUH FROOHFWHG IURP WKH FXOWLYDU %OXJJRH LQ )LHOG ] )+,$ K\EULGV GHVLJQDWHG DFFRUGLQJ WR WKHLU SORW QXPEHU LQ )LHOG

PAGE 72

LQIHFWLRQ UHVSHFWLYHO\ 1R LVRODWHV ZHUH REWDLQHG IURP WKH &DYHQGLVK FXOWLYDU :LOOLDPV ZKLFK LV UHVLVWDQW WR ERWK UDFH DQG UDFH RI WKH SDWKRJHQ ,Q FRQWUDVW LVRODWHV ZHUH FROOHFWHG IURP WKH +LJKJDWH SODQWV )RXUWHHQ RI WKH LVRODWHV KDG PW'1$ KDSORW\SHV ZKHUHDV RQH HDFK KDG DQG PW'1$ KDSORW\SHV 2QH LVRODWH FROOHFWHG IURP WKH FXOWLYDU +LJKJDWH KDG D SUHYLRXVO\ XQFKDUDFWHUL]HG PW'1$ KDSORW\SH IRU ) R[\VSRUXP I VS FEHQVH 3HUFHQW LQIHFWLRQ IRU WKH VHYHQ K\EULG EDQDQDV UDQJHG IURP OHVV WKDQ b IRU )+,$ WR b IRU )+,$ 7DEOH f 7KH REVHUYHG SHUFHQW LQIHFWLRQ RI HDFK FXOWLYDU ZDV REWDLQHG DQG FRPSDUHG DJDLQVW WKH QXOO H[SHFWDWLRQ RI QR GLIIHUHQFHV DPRQJ FXOWLYDUV ZLWK D VHULHV RI FKLVTXDUH WHVWV 7KH GLIIHUHQFHV LQ LQIHFWLRQ RQ WKH FXOWLYDUV ZHUH KLJKO\ VLJQLILFDQW S f LQGLFDWLQJ WKDW FXOWLYDUV DUH QRW HTXDOO\ OLNHO\ WR EH LQIHFWHG 7KLV VXJJHVWV GLIIHUHQFHV LQ KRVW UHVLVWDQFH JHQHV 7KH GLVWULEXWLRQ RI SDWKRJHQ JHQRW\SHV YDULHG RQ WKH GLIIHUHQW KRVW JHQRW\SHV 7KH RYHUZKHOPLQJ PDMRULW\ RI LVRODWHV FROOHFWHG IURP WKH K\EULGV KDG PW'1$ KDSORW\SHV ,VRODWHV ZLWK DQG PW'1$ KDSORW\SHV ZHUH UHFRYHUHG IURP VRPH RI WKH K\EULGV EXW DW D ORZHU IUHTXHQF\ FRPSDUHG WR LVRODWHV ZLWK PW'1$ KDSORW\SHV ,Q FRQWUDVW b RI WKH LVRODWHV FROOHFWHG LQ WKH VDPH ILHOG EXW REWDLQHG IURP WKH UDFH VXVFHSWLEOH FXOWLYDU %OXJJRH KDG WKH PW'1$ KDSORW\SH 7DEOH f 7KH FORQH S(
PAGE 73

7DEOH ,VRODWHV RI )XVDULXP R[\VSRUXP I VS FEHQVH FROOHFWHG LQ )LHOG DQG WKHLU LQFLGHQFH RQ GLIIHUHQW KRVW JHQRW\SHV 1XPEHU RI ,VRODWHV ZLWK PW'1$ +DSORW\SHV &XOWLYDU RU +\EULG 7RWDO ,VRODWHVn ,QIHFWLRQ\ b] +LJKJDWH UDFH VXVFHSWLEOHf :LOOLDPV UDFH t UHVLVWDQWf %OXJJRH UDFH VXVFHSWLEOHf )+,$ )+,$ )+,$ )+,$ )+,$ O )+,$ )+,$ [ )RU SODQWV ZKHUH PRUH WKDQ RQH LVRODWH ZDV FROOHFWHG RQO\ RQH LV UHSUHVHQWHG LI WKH PW'1$ KDSORW\SHV ZHUH LGHQWLFDO \ 7RWDO SHUFHQWDJH RI SODQWV LQ )LHOG IURP ZKLFK LVRODWHV ZHUH UHFRYHUHG ] &KLVTXDUH DQDO\VLV S GI

PAGE 74

,VRODWH+RVW &XOWLYDU PW'1$ +DSORW\SH [ %OXJJRH )+,$ %OXJJRH )+,$ +LJKJDWH %OXJJRH +LJKJDWH )+,$ )+,$ +LJKJDWH %OXJJRH )+,$ +LJKJDWH %OXJJRH %OXJJRH %OXJJRH )LJXUH +LQGOOO GLJHVWHG '1$ RI VHOHFWHG +RQGXUDQ ILHOG LVRODWHV WKDW ZHUH SUREHG ZLWK FORQH S(
PAGE 75

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f 6LPLODUO\ DOO WHVWHG LVRODWHV KDG D QXFOHDU '1$ ILQJHUSULQW LGHQWLFDO WR WKRVH LVRODWHV LQ )LHOG ZKLFK KDG D PW'1$ KDSORW\SH DQG ZHUH UHFRYHUHG IURP %OXJJRH DQG )+,$ K\EULGV DQG 7KH REVHUYHG IUHTXHQF\ RI HDFK SDWKRJHQ JHQRW\SH LQ WKH HQWLUH ILHOG DQG WKH KRVW JHQRW\SH IURP ZKLFK WKH\ ZHUH REWDLQHG ZHUH FRPSDUHG DJDLQVW WKH QXOO H[SHFWDWLRQ RI QR GLIIHUHQFHV QR KRVW VHOHFWLRQf ZLWK D VHULHV RI FKLVTXDUH WHVWV )LJXUH f 7KH GLIIHUHQFHV LQ FORQH IUHTXHQFLHV RQ WKH YDULRXV KRVW JHQRW\SHV ZHUH KLJKO\ VLJQLILFDQW &ORQHV RI WKH SDWKRJHQ GLG QRW UDQGRPO\ LQIHFW EDQDQD SODQWV UHSUHVHQWLQJ WKH YDULRXV KRVW JHQRW\SHV ,QVWHDG FORQHV ZHUH SUHIHUHQWLDOO\ GLVWULEXWHG RQ VSHFLILF KRVW JHQRW\SHV 'LVFXVVLRQ &ORQDO GLYHUVLW\ ZDV ORZ LQ WKH WZR ILHOG SRSXODWLRQV RI ) R[\VSRUXP I VS FEHQVH RQO\ ILYH FORQHV ZHUH UHFRYHUHG IURP WKH JHQRW\SHV RI EDQDQD $OWKRXJK WKH

PAGE 76

&KLVTXDUH DQDO\VLV S GI P &2 7 &0 FQ &' /2 U P ( 7f§ rf§ 2&7L Ua , ; ; FQ // // // // ; ; ; LI // // // pPW KDSORW\SH XPW KDSORW\SH ‘ PWKDSORW\SH +RVW FXOWLYDU 2EVHUYHG IUHTXHQF\ RI P WKDSORW\SH LQ )LHOG 2EVHUYHG IUHTXHQF\ RI PW KDSORW\SH LQ )LHOG 2EVHUYHG IUHTXHQF\ RI PW KDSORW\SH LQ )LHOG )LJXUH ,QFLGHQFH DQG GLVWULEXWLRQ RI PW'1$ KDSORW\SHV RI )XVDULXP R[\VSRUXP I VS FEHQVH RQ KRVW JHQRW\SHV LQ )LHOG

PAGE 77

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f ,Q FRQWUDVW ILHOG SRSXODWLRQV RI RWKHU DVH[XDOO\ UHSURGXFLQJ IXQJL LQFOXGLQJ 6WDJRQRVSRUD QRGRUXP DQG 0DJQDSRUWKH JUVHD DV ZHOO DV VH[XDOO\ UHSURGXFLQJ IXQJL LQFOXGLQJ 6FOHURWLQLD VFOHURWLRUXP 0\FRVSKDHUHOOD JUDPLQLFROD DQG VRPH SRSXODWLRQV RI 3K\WRSKWKRUD LQIHVWDQV FRQWDLQHG D ODUJHU QXPEHU RI JHQRW\SHV LQGLFDWLQJ KLJKHU OHYHOV RI FORQDO GLYHUVLW\ *RRGZLQ HW DO .RKOL HW DO .RKQ /HY\ HW DO 0F'RQDOG HW DO 0F'RQDOG HW DO f 7KH UHFRYHU\ RI GLIIHUHQW FORQHV RI ) R[\VSRUXP I VS FEHQVH LQ WKHVH WZR ILHOGV ZDV GHSHQGHQW RQ WKH JHQRW\SH RI WKH KRVW &KLVTXDUH DQDO\VHV ZHUH SHUIRUPHG WR WHVW

PAGE 78

ZKHWKHU LVRODWHV RI WKH SDWKRJHQ ZHUH UDQGRPO\ UHFRYHUHG IURP WKH GLIIHUHQW EDQDQD JHQRW\SHV WKHVH VKRZHG D VLJQLILFDQW DVVRFLDWLRQ EHWZHHQ D PW'1$ KDSORW\SH DQG LWV LQFLGHQFH RQ VSHFLILF KRVW JHQRW\SHV ,Q IDFW LVRODWHV ZLWK FHUWDLQ PW'1$ KDSORW\SHV ZHUH PRUH IUHTXHQWO\ DQG VRPHWLPHV DOPRVW H[FOXVLYHO\ UHFRYHUHG RQ FHUWDLQ KRVW JHQRW\SHV )LJXUH f )RU H[DPSOH b RI WKH LVRODWHV UHFRYHUHG IURP WKH UDFH VXVFHSWLEOH FXOWLYDU %OXJJRH LQ )LHOGV DQG KDG D PW'1$ KDSORW\SH DQG UHSUHVHQWHG D VLQJOH FORQH EDVHG RQ WKHLU LGHQWLFDO '1$ ILQJHUSULQWV ,Q FRQWUDVW b RI WKH LVRODWHV UHFRYHUHG IURP +LJKJDWH KDG D PW'1$ KDSORW\SH UHSUHVHQWLQJ WZR QHDUO\ LGHQWLFDO FORQHV 7KH RFFXUUHQFH RI D VSHFLILF JHQRW\SH RI WKH SDWKRJHQ DVVRFLDWHG ZLWK YLUXOHQFH RQ D VSHFLILF KRVW JHQRW\SH LV OLNHO\ WKH FRQVHTXHQFH RI OLQNDJH GLVHTXLOLEULXP IRXQG LQ FORQDOO\ UHSURGXFLQJ RUJDQLVPV 7LED\UHQF HW DO f 6LQFH VH[XDO UHSURGXFWLRQ GRHV QRW RFFXU LQ WKHVH RUJDQLVPV WKH HQWLUH JHQRPH LV HIIHFWLYHO\ OLQNHG $QGHUVRQ DQG .RKQ f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f 6XFK VHOHFWLRQ SUHVVXUH XQGRXEWDEO\ KDV DQ HYHQ PRUH VLJQLILFDQW HIIHFW RQ WKH SDWKRJHQ SRSXODWLRQV LQ SHUHQQLDO PRQRFXOWXUDO SURGXFWLRQ V\VWHPV VXFK DV EDQDQD SODQWDWLRQV VLQFH VXVFHSWLEOH

PAGE 79

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

PAGE 80

ILQJHUSULQW RI LVRODWHV UHSUHVHQWLQJ WKHVH WZR OLQHDJHV ZHUH GLVWLQFW IXUWKHU VXEVWDQWLDWLQJ WKHLU JHQHWLF GLVLPLODULW\ 5DFH LVRODWHV DUH NQRZQ WR RFFXU LQ DOO WKUHH RI WKH 9&*V IRXQG LQ +RQGXUDV 3ORHW] DQG &RUUHOO f DQG WKH SUHVHQW UHVXOWV VKRZLQJ PLWRFKRQGULDO KDSORW\SHV FRUUHVSRQGLQJ WR WKHVH WKUHH 9&*V UHFRYHUHG IURP GLVHDVHG +LJKJDWH SODQWV 7DEOH f VXSSRUW WKLV REVHUYDWLRQ 7KH GHILQLWLRQ RI UDFH DVVXPHV WKDW DOO UDFH LVRODWHV VKDUH LGHQWLFDO YLUXOHQFH FKDUDFWHULVWLFV FRUUHVSRQGLQJ WR VXVFHSWLELOLW\ LQ WKH FXOWLYDU *URV 0LFKHO DQG +LJKJDWHf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f IRXQG WKDW LVRODWHV RI D VLQJOH SDWKRW\SH KDG VLJQLILFDQWO\ GLIIHUHQW '1$ ILQJHUSULQWV DQG WKLV VXJJHVWHG WKDW WKH SDWKRW\SHV ZHUH GHULYHG IURP LQGHSHQGHQW FORQDO OLQHDJHV ,Q ) R[\VSRUXP I VS FEHQVH LVRODWHV LQ 9&* WKDW FDXVH GLVHDVH LQ +LJKJDWH PD\ SRVVHVV JHQHWLF IDFWRUV

PAGE 81

WKDW PDNH WKHP EHWWHU DGDSWHG WKDQ WKRVH LQ 9&* DQG WR WKH HFRORJLFDO DQG HQYLURQPHQW IDFWRUV LQ +RQGXUDV ,Q WKLV VWXG\ WKH PDMRULW\ RI LVRODWHV UHFRYHUHG IURP +LJKJDWH KDG PWKDS ZKHUHDV WKH PDMRULW\ RI LVRODWHV UHFRYHUHG IURP %OXJJRH KDG PWKDS 7DEOH f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

PAGE 82

QRW UHVXOW LQ UHVLVWDQFH WR SRSXODWLRQV RI WKH SDWKRJHQ LQ RWKHU ORFDWLRQV 3ORHW] 6WRYHU DQG %XGGHQKDJHQ f ,GHDOO\ QHZ JHQRW\SHV VKRXOG EH VFUHHQHG DJDLQVW LVRODWHV WKDW HQFRPSDVV WKH UDQJH RI JHQHWLF GLYHUVLW\ SUHVHQW LQ WKH HQWLUH IRUPDH VSHFLDOHV 7KHUHIRUH WKHUH LV D QHHG WR DVVHVV WKH UDQJH RI YDULDWLRQ LQ WKH SDWKRJHQ SRSXODWLRQV IRU YLUXOHQFH RQ SDUWLFXODU EUHHGLQJ OLQHV 0LOJURRP DQG )U\ LQ SUHVVf 8VLQJ ERWK WKH PW'1$ KDSORW\SH DQG D QXFOHDU '1$ ILQJHUSULQWLQJ SUREH WR HVWLPDWH JHQHWLF GLYHUVLW\ ZLWKLQ D ILHOG VLWH FDQ LPSURYH LQ VLWX VFUHHQLQJ IRU )XVDULXP ZLOW ,Q WKLV SDWKRV\VWHP WKHUH LV D UHODWLYHO\ ORQJWHUP LQWHUDFWLRQ EHWZHHQ KRVW DQG WKH SDWKRJHQ WKDW PD\ QRW EH DVVHVVHG LQ DUWLILFLDO LQRFXODWLRQ WHVWV %XGGHQKDJHQ f 8VLQJ PROHFXODU WHFKQLTXHV IRU DVVHVVLQJ WKH SDWKRJHQ SRSXODWLRQ DW D JLYHQ VLWH DQG GHWHUPLQLQJ WKH OLQHDJH RI LVRODWHV LQIHFWLQJ SDUWLFXODU KRVW JHQRW\SHV FDQ LQFUHDVH WKH HIIHFWLYHQHVV RI GLVHDVH VFUHHQLQJ DQG DVVLVW LQ WKH HYDOXDWLRQ RI QHZ EUHHGLQJ OLQHV

PAGE 83

&+$37(5 6800$5< $1' &21&/86,216 7KLV UHVHDUFK KDV SURYLGHG D JHQHWLF FKDUDFWHUL]DWLRQ DQG SK\ORJHQHWLF DQDO\VLV RI LVRODWHV RI ) R[\VSRUXP I VS FEHQVH UHSUHVHQWLQJ ERWK D ZRUOG FROOHFWLRQ DQG SRSXODWLRQ IURP WZR ILHOGV 3ULRU LQYHVWLJDWLRQV ZKLFK JHQHWLFDOO\ FKDUDFWHUL]HG LVRODWHV EDVHG RQ 9&*V 5$3'V DQG HOHFWURSKRUHWLF NDU\RW\SLQJ SURYLGHG RQO\ OLPLWHG LQIRUPDWLRQ RQ WKH JHQHWLF UHODWLRQVKLSV DPRQJ LVRODWHV %HQWOH\ HW DO %RHKP HW DO 3ORHW] DQG &RUUHOO 6RUHQVHQ HW DO f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

PAGE 84

WKH FROOHFWHG LVRODWHV DQG GHPRQVWUDWHG KRZ WKH JHQRW\SH RI WKH KRVW FDQ DIIHFW WKH SDWKRJHQ SRSXODWLRQV %RWK VWXGLHV GHPRQVWUDWHG WKH LPSRUWDQFH RI FKRRVLQJ DQ DSSURSULDWH '1$ PDUNHU LQ SK\ORJHQHWLF DQG SRSXODWLRQ JHQHWLF LQYHVWLJDWLRQV $Q LPSRUWDQW EHQHILW WR XVLQJ 5)/3V LQ SK\ORJHQHWLF DQDO\VLV LV WKDW WKH QXPEHU RI 5)/3 PDUNHUV DYDLODEOH IRU XVH LV ODUJH EHFDXVH DQ\ FORQHG ORZFRS\QXPEHU SLHFH RI '1$ FDQ EH XVHG DV D SUREH DQG VHYHUDO UHVWULFWLRQ HQ]\PHV FDQ EH DVVD\HG WR LGHQWLI\ 5)/3V ZLWK HDFK SUREH 0LFKHOPRUH DQG +XOEHUW f 2QH LPSRUWDQW FKDUDFWHULVWLF RI WKH 5)/3 PDUNHUV XVHG LQ WKLV SK\ORJHQHWLF DQDO\VLV LV WKDW WKH\ ZHUH OLNHO\ VHOHFWLYHO\ QHXWUDO 7KLV LV LQ FRQWUDVW WR VHYHUDO PDUNHUV WKDW KDYH EHHQ HPSOR\HG LQ WUDGLWLRQDO VWXGLHV VXFK DV SDWKRJHQ YLUXOHQFH JHQHV RU UHVLVWDQFH WR IXQJLFLGHV ZKLFK DUH XQGHU VWURQJ VHOHFWLRQ SUHVVXUH LQ DJULFXOWXUDO V\VWHPV DQG PD\ SURYLGH D ELDVHG HVWLPDWH RI JHQHWLF GLYHUVLW\ 0F'RQDOG DQG 0F'HUPRWW f 2WKHU PROHFXODU PDUNHUV WKDW DUH FRPPRQO\ XVHG LQ SK\ORJHQHWLF DQDO\VHV RI IXQJL LQFOXGH LVR]\PHV 5$3'V WKH QXFOHDU DQG PLWRFKRQGULDO U51$ JHQHV DQG LQWHUQDO WUDQVFULEHG VSDFHUV ,76f DQG PW'1$ PDSV ,VR]\PHV KDYH EHHQ XVHG WR FKDUDFWHUL]H IXQJDO LVRODWHV EXW WKHLU GLVDGYDQWDJH LV WKDW WKH SRVVLEOH QXPEHU RI HQ]\PHV WKDW DUH LQIRUPDWLYH IRU DQ RUJDQLVP LV OLPLWHG DQG PDQ\ ORFL DUH PRQRPRUSKLF DQG WKXV QRQLQIRUPDWLYH 0LFKHOPRUH DQG +XOEHUW f $QDO\VLV RI ) R[\VSRUXP I VS FEHQVH LVRODWHV XVLQJ 5$3'V ZDV DWWHPSWHG LQLWLDOO\ LQ WKLV UHVHDUFK +RZHYHU WKLV WHFKQLTXH SURYHG WR EH XQUHOLDEOH IRU SK\ORJHQHWLF VWXGLHV 7KH ODUJHVW REVWDFOH WR XVLQJ WKHP ZDV WKDW SRO\PRUSKLVPV FRXOG QRW EH VFRUHG XQDPELJXRXVO\ ZLWKRXW FORQLQJ WKH SXWDWLYH SRO\PRUSKLF EDQGV DQG WKHQ SHUIRUPLQJ DQ

PAGE 85

5)/3 DQDO\VLV (YHQ ZKHQ 5$3'V ZHUH XVHG WKDW FRXOG EH H[FLVHG ZLWK DQ DSSURSULDWH UHVWULFWLRQ HQ]\PH UHVXOWV ZHUH GLIILFXOW WR LQWHUSUHW VHH $SSHQGL[ ,f 7ZR SUREOHPV DVVRFLDWHG ZLWK PHWKRGV EDVHG RQ SRO\PHUDVH FKDLQ UHDFWLRQ 3&5f VXFK DV 5$3'V DUH WKDW f QRQDPSOLILFDWLRQ FDQ UHVXOW IURP EDVH SDLU VXEVWLWXWLRQV DW DQ\ RI WKH GLIIHUHQW QXFOHRWLGHV FRPSOHPHQWDU\ WR WKH SULPHUV DW WKH HQGV RI WKH DPSOLILHG IUDJPHQW DQG f UHVXOWV PD\ EH DIIHFWHG E\ FKDQJHV LQ HQYLURQPHQWDO FRQGLWLRQV DQG DW WLPHV DUH QRW UHSURGXFLEOH 5RVHZLFK t 0F'RQDOG f 0LWRFKRQGULDO '1$V DQG LQ FHUWDLQ RUJDQLVPV PLWRFKRQGULDO SODVPLGV KDYH EHHQ XVHG WR FKDUDFWHUL]H LVRODWHV RI ) R[\VSRUXP I VS PHORQLV I VS FRQJOXWLQDQV DQG I VS FEHQVH -DFREVRQ DQG *RUGRQ .LVWOHU HW DO .LVWOHU DQG 0RPRO f 7ZR GLVDGYDQWDJHV RI XVLQJ PW'1$ IRU SK\ORJHQHWLF DQDO\VHV LV WKDW WKH JHQRPH LV VPDOO VR FRQVHTXHQWO\ IHZHU SRO\PRUSKLVPV PD\ EH IRXQG DQG WKH UDWH RI HYROXWLRQ RI PW'1$ PD\ EH GLIIHUHQW WKDQ WKDW RI QXFOHDU '1$ 0LWRFKRQGULDO '1$ ZDV XVHG LQ ERWK WKH SK\ORJHQHWLF DQG SRSXODWLRQ VWXGLHV ,Q WKH SK\ORJHQHWLF VWXG\ JURXSLQJV RI LVRODWHV EDVHG RQ PW'1$ KDSORW\SHV ZHUH FRQJUXHQW ZLWK JURXSV EDVHG RQ WKH 5)/3 DQDO\VLV KRZHYHU IHZHU JURXSV FRXOG EH GLVWLQJXLVKHG EDVHG RQ PW'1$ DQDO\VLV ,Q WKH SRSXODWLRQ VWXGLHV D PW'1$ KDSORW\SH ZDV XVHG WR JURXS VLPLODU LVRODWHV ZKLFK ZHUH WKHQ VXEMHFWHG WR '1$ ILQJHUSULQW DQDO\VLV WR GHWHUPLQH ZKHWKHU LVRODWHV ZHUH FORQDOO\ GHULYHG 6HTXHQFLQJ RI ERWK WKH QXFOHDU DQG PLWRFKRQGULDO U51$ JHQHV DV ZHOO DV ,76 UHJLRQV KDYH EHHQ GHVFULEHG IRU IXQJL DQG XVHG LQ SK\ORJHQHWLF VWXGLHV %UXQV HW DO f $Q XQGHUO\LQJ DVVXPSWLRQ RI VHTXHQFH DQDO\VLV LV WKDW WKH SK\ORJHQ\ RI WKH UHJLRQ LV D JRRG LQGLFDWRU RI WKH SK\ORJHQ\ RI RUJDQLVPV %UXQV HW DO f $QRWKHU DSSURDFK

PAGE 86

WR LGHQWLI\LQJ VHTXHQFH GLIIHUHQFHV LQ WKH ,76 UHJLRQ LV WKURXJK DQ 5)/3 DQDO\VLV XVLQJ WKH 3&5DPSOLILHG UHJLRQ ,Q WKLV VWXG\ D QXPEHU RI SULPHUV ZHUH XVHG WR DPSOLI\ ,76 UHJLRQV RI) R[\VSRUXP I VS FEHQVH LVRODWHV 6HYHUDO IRXU DQG VL[EDVH UHVWULFWLRQ HQ]\PHV ZHUH XVHG WR GLJHVW WKH DPSOLILHG SURGXFWV $SSHQGL[ ,,f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f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f DOVR DUH LQGLFDWLYH RI D FORQDOO\ UHSURGXFLQJ RUJDQLVP 0LOJURRP 7LED\UHQF HW DO f 7KH +RQGXUDQ ILHOG SRSXODWLRQ VWXG\ DOVR LQGLFDWHG WKDW FORQDO UHSURGXFWLRQ DFFRXQWHG IRU WKH RFFXUUHQFH RI D VSHFLILF JHQRW\SH RI WKH SDWKRJHQ DVVRFLDWHG ZLWK

PAGE 87

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f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f REVHUYHG D FORVH UHODWLRQVKLS EHWZHHQ QRQSDWKRJHQLF VWUDLQV RI ) R[\VSRUXP DQG ) R[\VSRUXP I VS PHORQLV LVRODWHV LQ 9&* VXJJHVWLQJ DQ LQGHSHQGHQW RULJLQ DV D PHORQ SDWKRJHQ VHSDUDWH IURP ) R[\VSRUXP I VS PHORQLV LVRODWHV LQ 9&* +RZHYHU WKLV VWXG\ ZDV EDVHG RQ PW'1$ KDSORW\SH GDWD DQG LQWHUSUHWDWLRQ PD\ EH FRPSOLFDWHG E\ WKH SRVVLELOLW\ RI KRUL]RQWDO WUDQVIHU RI PW'1$

PAGE 88

,QWHUHVWLQJO\ WKH UHVXOWV RI WKH 5)/3 VWXG\ LQGLFDWH WKDW LVRODWHV LQ 9&* ZKLFK FRPSULVH OLQHDJH )2& 9 DUH JHQHWLFDOO\ GLVWDQW IURP RWKHU ) R[\VSRUXP I VS FEHQVH LVRODWHV 2QH H[SODQDWLRQ LV WKDW WKLV JURXS FRXOG KDYH RULJLQDWHG LQ (DVW $IULFD D UHJLRQ RXWVLGH WKH KRVWnV FHQWHU RI RULJLQ $QRWKHU H[SODQDWLRQ LV WKDW WKH JHQHWLF GLIIHUHQFHV LQ WKHVH LVRODWHV FRXOG KDYH DULVHQ E\ D IRXQGHU HIIHFW )RXQGHU HIIHFWV UHVXOW ZKHQ VPDOO JURXSV RI LQGLYLGXDOV DUH LVRODWHG IURP ODUJHU SRSXODWLRQV ,W LV SRVVLEOH WKDW RYHU D SHULRG RI WLPH LVRODWHV LQ 9&* ZHUH VXEMHFWHG WR GLIIHUHQW VHOHFWLRQ IRUFHV ZKLFK DOWHUHG JHQH IUHTXHQF\ LQ WKLV JURXS $GGLWLRQDOO\ PXWDWLRQV ZKLFK RFFXU LQ WKH IRXQGLQJ SRSXODWLRQ FRXOG EH IL[HG LQ WKH VXFFHVVLYH JHQHUDWLRQV GXH WR WKH FORQDO QDWXUH RI WKLV RUJDQLVP 7KH GDWD LQGLFDWH WKDW LVRODWHV LQ 9&* ZKLFK FRPSULVH ERWK OLQHDJHV )2& 9,, DQG ; DUH SK\ORJHQHWLFDOO\ GLVWLQFW 7KLV JURXS PD\ SURYLGH HYLGHQFH RI DQ DQFLHQW JHQHWLF H[FKDQJH EHWZHHQ LQGLYLGXDOV LQ OLQHDJHV )2& DQG ,, RU WKH\ PD\ UHSUHVHQW DQ DQFHVWUDO JURXS SRVVHVVLQJ SULPLWLYH FKDUDFWHU VWDWHV 7KHUH DUH QXPHURXV SUDFWLFDO DSSOLFDWLRQV RI WKLV UHVHDUFK KRZHYHU RQO\ WKH PRVW LPSRUWDQW DVSHFWV DUH KLJKOLJKWHG 7KH SK\ORJHQHWLF VWXG\ KHOSV FODULI\ WKH JHQHWLF UHODWLRQVKLSV DPRQJ WKH LVRODWHV ZKLFK UHSUHVHQW GLIIHUHQW UDFHV RI WKH SDWKRJHQ ,VRODWHV ZKLFK UHSUHVHQW UDFH RI WKH SDWKRJHQ RFFXU LQ WKH WZR ODUJHVW DQG GLYHUJHQW OLQHDJHV )2& DQG ,,f LQ WKH FROOHFWLRQ +RZHYHU GDWD IURP WKH +RQGXUDV ILHOG VWXG\ UHYHDOHG WKDW LVRODWHV ZKLFK KDYH EHHQ FRQVLGHUHG UDFH EDVHG RQ UHDFWLRQ WR WKH KRVW GLIIHUHQWLDO PD\ QRW VKDUH WKH VDPH KRVW LQIHFWLYLW\ ,Q ILHOG WHVWV LVRODWHV UHSUHVHQWLQJ WKHVH WZR OLQHDJHV ZHUH QRW UHFRYHUHG DW WKH VDPH IUHTXHQF\ IURP VXVFHSWLEOH KRVW WLVVXH HYHQ WKRXJK LVRODWHV UHSUHVHQWLQJ )2& DQG )2& ,, ZHUH DVVXPHG WR EH HYHQO\ GLVWULEXWHG

PAGE 89

WKURXJKRXW WKH ILHOG ,VRODWHV ZKLFK UHSUHVHQW UDFH DSSHDU WR EH OLPLWHG WR OLQHDJHV )2& DQG 9,,, ZKLFK DUH DVVRFLDWHG ZLWK RQH PDLQ EUDQFK RQ WKH PLGSRLQW URRWHG WUHH DQG )2& 9 ZKLFK LV RQO\ GLVWDQWO\ UHODWHG WR RWKHU ) R[\VSRUXP I VS FEHQVH LVRODWHV )LJXUH f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f SRLQWHG RXW WKDW UHVXOWV IURP GLVHDVH VFUHHQLQJ WHVWV ZKLFK UHOLHG RQ

PAGE 90

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

PAGE 91

$33(1',; $ (;3/25,1* 7+( 327(17,$/ 2) 86,1* 32/<0(5$6( &+$,1 5($&7,21 0(7+2'2/2*,(6 ,1 3+
PAGE 92

QRQFRGLQJ UHJLRQV EHWZHHQ WKHP :KLWH HW DO f 7KH SULPHUV DUH EDVHG RQ VHTXHQFHV IURP 6FLFFKFLURP\FHV FHUHYLVLDH 'LFW\RVWHOLXP GLVFRLGHXP DQG 6W\ORQLFKD SXVWXODWD 'DPV HW DO f $ GLDJUDP RI WKHVH SULPHUV DQG WKHLU ORFDWLRQ DORQJ WKH QXFOHDU U'1$ UHJLRQ RI 6DFFKDURP\FHV FHUHYLVLDH LV SURYLGHG E\ :KLWH HW DO f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f ZHUH XVHG LQ WKLV VWXG\ WR DPSOLI\ WKH WRWDO JHQRPLF '1$V IURP D VXEVHW RI LVRODWHV UHSUHVHQWLQJ D ODUJH QXPEHU RI 9&*V DQG D SDQWURSLFDO GLVWULEXWLRQ 7DEOH $Of 7KH DPSOLILFDWLRQ SURWRFRO IRU WKLV VWXG\ DV ZHOO DV WKH 5$3' DQDO\VLV DUH SURYLGHG LQ 7DEOH $ ,QWHUQDO WUDQVFULEHG VSDFHU SULPHUV DQG ZHUH XVHG WR DPSOLI\ WKH ,76 UHJLRQ RI WKH QXFOHDU U51$ JHQHV DQG SULPHUV ,76 DQG ZHUH XVHG WR DPSOLI\ WKH ,76 ,76 DQG WKH 6 U'1$ UHJLRQV LQ ) R[\VSRUXP I VS FEHQVH LVRODWHV IRU SULPHU VHTXHQFHV VHH 7DEOH $f 7KH

PAGE 93

UHVXOWDQW DPSOLILHG SURGXFWV ZHUH VLPLODU IRU DOO LVRODWHV 7KHVH SURGXFWV ZHUH WKHQ VXEMHFWHG WR UHVWULFWLRQ HQ]\PH GLJHVWLRQ XVLQJ HQ]\PHV +DHO,, 6DX0 DQG (FR5O $LO RI WKH WHVWHG LVRODWHV KDG H[DFWO\ WKH VDPH UHVWULFWLRQ SDWWHUQ IRU DOO RI WKH HQ]\PHV WHVWHG 5$33 $QDO\VLV 7KH LGHQWLILFDWLRQ RI SRWHQWLDO SRO\PRUSKLF UHJLRQV RI WKH JHQRPH ZDV DFFRPSOLVKHG XVLQJ 5$3' DQDO\VLV IROORZLQJ SURWRFROV GHVFULEHG SUHYLRXVO\ :HOVK DQG 0F&OHOODQG :LOOLDPV HW DO f $ QXPEHU RI ROLJRQXFOHRWLGH SULPHUV ZHUH VFUHHQHG IRU WKHLU DELOLW\ WR DPSOLI\ EDQGV LQ ) R[\VSRUXP I VS FEHQVH 7DEOH $f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f )UDJPHQWV ZHUH VXEMHFWHG WR DJDURVH JHO HOHFWURSKRUHVLV DQG SXWDWLYH SRO\PRUSKLF UHJLRQV ZHUH LGHQWLILHG 3RO\PRUSKLF DPSOLILHG SURGXFWV ZHUH H[FLVHG IURP ORZ PHOWLQJ SRLQW DJDURVH JHOV DQG ODEHOHG ZLWK D QRQUDGLRDFWLYH GLJR[LJHQLQ GLJfO G873 XVLQJ WKH UDQGRP SULPHG PHWKRG %RHKULQJHU 0DQQKHLP *HQLXV 6\VWHPf *HQRPLF '1$ LVRODWHG IURP D VXEVHW RI LVRODWHV UHSUHVHQWLQJ 9&*V DQG D SDQWURSLFDO GLVWULEXWLRQ 7DEOH $Of ZDV GLJHVWHG ZLWK UHVWULFWLRQ HQ]\PHV ZLWK

PAGE 94

UHFRJQLWLRQ VHTXHQFHV FRUUHVSRQGLQJ WR WKRVH IRXQG LQ WKH LQWHUQDO KH[DPHULF VHTXHQFHV RI WKH ROLJRQXFOHRWLGH SULPHU 6WDQGDUG SURWRFROV ZHUH XVHG IRU UHVWULFWLRQ HQ]\PH GLJHVWLRQV EORWWLQJ DQG K\EULGL]DWLRQV DV GHVFULEHG LQ FKDSWHUV DQG :KHQ JHQRPLF '1$V ZHUH DPSOLILHG ZLWK DQ ROLJRQXFOHRWLGH FRQWDLQLQJ D %DP+O UHFRJQLWLRQ VHTXHQFH *$**$7&&&7SULPHU )f D NLOREDVH NEf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f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

PAGE 95

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f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

PAGE 96

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f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

PAGE 97

7DEOH $O ,VRODWHV RI )XVDULXP R[\VSRUXP I VS FEHPH XVHG LQ 5$3' DQDO\VLV DQG SUHVHQFH RU DEVHQFH RI PDMRU 5$3' EDQGV ,VRODWH 9&* 3ULPHU ) NE EDQG 3ULPHU ) NE EDQG 3ULPHU ) NE EDQG 67*0 ) )&3DFRYDQ 257 ) 3K '$9$2 r 67' r %/8* 6" r % r 0: 671 r 0: r &9$ r 6713 r 67% r 1 $ $ 6+ 6713 r 0: r :HDN EDQG SUHVHQW

PAGE 98

7DEOH $ $PSOLILFDWLRQ SURWRFRO IRU 5$3' DQG ,76 DQDO\VLV QJ WHPSODWH '1$ QJ SULPHU '1$ 2SHURQ 7HFKQRORJLHV 6DQ -RV &$f L$ O2P0G$73 X? O2P0 G&73 c$ O2P0G773 c$ O2P0 G*73 8 7DT SRO\PHUDVH %RHKULQJHU 0DQQKHLP ,QGLDQDSROLV ,1f c$ O2P0 0J&O c$ O2[ 7DT SRO\PHUDVH EXIIHU VWHULOH ZDWHU WR YROXPH RI T? $PSOLILFDWLRQV ZHUH SHUIRUPHG LQ D *HQH 0DFKLQH ,, WKHUPDO F\FOHU 7KH PDFKLQH ZDV SURJUDPPHG IRU RQH F\FOH IRU VHFRQGV DW r& IROORZHG E\ F\FOHV RI PLQXWH DW r& PLQXWH DW r& DQG PLQXWHV DW r& IROORZHG E\ D ILQDO F\FOH RI PLQXWHV DW r&

PAGE 99

7DEOH $ 3ULPHU VHTXHQFH IRU WKH ,76 UHJLRQV ,76 3ULPHUn 6HTXHQFH *&7*&*77&77&$7&*$7*& 7&&7&&*&77$77*$7$7*& **$$*7 $$$$*7 & *7 $$& $$** n3ULPHUV REWDLQHG IURP 'U .HUU\ 2n'RQQHOO ZLWK WKH VHTXHQFHV EDVHG RQ :KLWH HW DO f

PAGE 100

7DEOH $ /LVW RI ROLJRQXFOHRWLGH SULPHUV WHVWHG RQ )XVDUKLP R[\VSRUXP I VS FEHQVH LVRODWHV 2SHURQ 3ULPHU $PSOLILFDWLRQ 5HFRJQLWLRQ 6HTXHQFH
PAGE 101

$33(1',; % $'',7,21$/ ,1)250$7,21 )520 3+
PAGE 102

3UREH (Q]\PH 6L]H RI $PSOLILHG %DQGV LQ .E 3UREH (Q]\PH 6L]H RI $PSOLILHG %DQGV LQ .E (FR59 [ 'UD ; ; (FR59 !\ ; +DH ,,, \ ; \ [ (FR59 (FR59 O[ ; ; r ; nV7 (FR59 [ (FR59 ; ; ; [ 5HSUHVHQWV WKH PRVW FRPPRQO\ RFFXULQJ DOOHOHVf \ 5HSUHVHQW PRQRPRUSKLF DOOHOHVf ] 6RXUFH RI 3UREHV (OLDV HW DO

PAGE 103

$33(1',; & $//(/,& '$7$ 6&25(' $6 35(6(1&( 25 $%6(1&( )25 ($&+ 352%( (1=<0( &20%,1$7,21 ,62/$7( 1$0( $1' 9(*(7$7,9( &203$7,%,/,7< *5283 ,1&/8'(' $6 $ 5(35(6(17$7,9( )25 ($&+ 81,48( 5(675,&7,21 )5$*0(17 /(1*7+ 32/<0253+,60 +$3/27<3(

PAGE 104

9&* & +DSORW\SH [ 9&* ) +DSORW\SH 9&* 3$+DSORW\SH 9&* ,& +DSORW\SH 9&* 3DFRYDQ +DSORW\SH 9&* 0*6$ +DSORW\SH 9&* 6$ +DSORW\SH """O """"" 9&* 6$ +DSORW\SH 9&* *0 +DSORW\SH 9&* ) +DSORW\SH 9&* +, +DSORW\SH 9&* +DSORW\SH """ 9&* 3K +DSORW\SH 9&* 3 +DSORW\SH

PAGE 105

9&* /$3 +DSORW\SH 9&* 6$%$ +DSORW\SH 9&* 3: +DSORW\SH 9&* '$9$2 +DSORW\SH 9&* -/7+ +DSORW\SH 9&* -/7+ +DSORW\SH """"""" 9&* 7 +DSORW\SH 9&* 3K/ +DSORW\SH 9&* 3K +DSORW\SH 9&* $ +DSORW\SH 9&* *0% +DSORW\SH 9&* 0DFD +DSORW\SH 9&* 67+DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH """""""

PAGE 106

9&* -/7+ +DSORW\SH 9&* 0: +DSORW\SH 9&* 671 +DSORW\SH 9&* 671 +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH """""" 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* -/7+ +DSORW\SH 9&* -/7+ +DSORW\SH """" 9&* -/7+ +DSORW\SH OO""""" 9&* -/7+ +DSORW\SH 9&* +DSORW\SH 9&* +DSORW\SH

PAGE 107

9&* +DSORW\SH 9&* 67% +DSORW\SH 9&* 67$ +DSORW\SH 9&* 6, +DSORW\SH 9&* $ +DSORW\SH """"""" 9&* 1 +DSORW\SH 9&* $ +DSORW\SH 9&* ** +DSORW\SH 9&* $ +DSORW\SH 9&* -& +DSORW\SH 9&* 6+ +DSORW\SH 9&* +DSORW\SH """ 9&* 6713 +DSORW\SH 9&* 6713 +DSORW\SH 9&* +DSORW\SH

PAGE 108

9&* +DSORW\SH 9&* +DSORW\SH 9&* (6 +DSORW\SH """"" 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* 0: +DSORW\SH 9&* ,1'2 +DSORW\SH 6&\ 6& """"""

PAGE 109

6& &6 [ 7KH RUGHU RI SUREHHQ]\PH LQSXW DUH 2UDO +DHOOO )FR59 7-P, +DHO?? (FR59 ?+DHOOO ?'UDO (FR59 ?+DHOOO (FR59 (FR59 DHO,, (FR59 (FR59 (FR59 =fUD, (FR59 DQG (FR59 \ ,VRODWHV LQ EROG VLJQLI\ WKH RXWJURXSV ] $OOHOHV LQ EROG DUH XQLTXH WR LVRODWHV UHSUHVHQWLQJ WKH )XVDULXP R[\VSRUXP I VS QLYHXP DQG )R[\VSRUXP I VS O\FRSHUVLFL

PAGE 110

9&*9 $33(1',; /,67 2) ,62/$7(6 2) ) 2;<632580) 63 &8%(16( 7+(,5 9(*(7$7,9( &203$7,%,/,7< *5283 +$3/27<3( $1' /,1($*( ,VRODWHr ,& 257 *$/ & $'&O $ $'67*0 3$257 *$/ %8( 1: 1+ 1% ) )&3DFRYDQ 0*6$ 6$ 6$ 6$ +DSORWYSH /LQHDJH ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,3 ,,

PAGE 111

9&* ,VRODWH +DSORWYSH /LQHDJH *0 \ ,,, ,,3 +, LOO 0/ ,,, 7%5 LOO 3K 9, 3K 9, 3 9, /$3 9, 6$%$ 9, 3: 9, 3: 9, 3: 9, '$9$2 9,, 7 9,, 3K/ 9,, 3KO 9,, 9,, -/7+ ; -/7+ ;r $ *0% 0DFD 673$ 67' %/8* 6" )&, )&, )&, )&, 67, 0: 0: 0: S 0: 0: 0: 0: 0: 0: 0: ,

PAGE 112

9&* FRQWf ,VRODWH 0: 671 671 671 671 673$ % % -&% -/7+ -/7+ -/7+ 0: 0: 0: 0: 0: 0: 0: 0: 0: 0: 0: -/7+ -/7+ -/7+ -/7+ -/7+ $ ,6" 673$ -/7+ 6, 67$ 670 67% +DSORWYSH } /LQHDJH , , , , , , , U L L L L L L L L L U U L L L L L S L L L L L L L LL LL LL LL

PAGE 113

9&* ,VRODWH +DRORWY'H /LQHDVH , $ ,r 1 ,, ,, 1 ,, ,, ,, ,, $ ,; $ ,; &6% ,; -& ,; $ ,;" $ ,; -& ,; ) ,; ) ,; -& ,; ** ,; ,;V 6+ ,; 6713 9,,, 6713 9,,, 6713 9,,, ,,, ,,, ,,, ,,, ,,, ,,, ,,, (6 \

PAGE 114

9&* ,VRODWH +DQORWYRH /LQHD 0: 9 0: 9 0: 9 0: 9 0: 9 0: 9 0: 9 0: 9 0: 9 &5 ,, &5 ,, &5 ,, &5 ,, ,1'2 ,,, ,1' ,,, ,1' ,,, Y 9HJHWDWLYH FRPSDWLELOLW\ JURXSV 9&*Vf ZHUH DVVLJQHG XVLQJ QLWURJHQ PHWDEROLVP QLWf PXWDQWV DFFRUGLQJ WR WKH SURWRFROV RI &RYH f DV PRGLILHG E\ 3XKDOOD f Z ,VRODWHV LQ EROG UHSUHVHQW WKRVH UHIOHFWHG LQ WKH 5)/3 WUHH DQG 5)/3 KDSORW\SH GDWD IRU WKRVH LVRODWHV DUH LQ $SSHQGL[ [ PLVVLQJ GDWD PDGH LW LPSRVVLEOH WR GHWHUPLQH FRHIILFLHQW RI VLPLODULW\ EXW EDVHG RQ DOO RWKHU GDWD LW LV SUHVXPHG WR EH LQ WKH OLQHDJH LQGLFDWHG \ XQLTXH LVRODWH WKDW KDG QR OLQHDJH DIILQLW\ EDVHG RQ GHILQHG FULWHULD WKH OLQHDJH DVVLJQHG LV EDVHG RQ FRHIILFLHQWV RI VLPLODULW\ WR WKH PRVW FORVHO\ UHODWHG LVRODWHV ] LVRODWHV DQDO\]HG ZLWK WKLV GHVLJQDWLRQ GR QRW FRUUHVSRQG WR DQ\ LVRODWHV FXUUHQWO\ KHOG LQ WKH +RPHVWHDG FROOHFWLRQ

PAGE 115

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

PAGE 116

%XUJUHVV /: *HQHUDO HFRORJ\ RI WKH )XVDULD SS ,Q )XVDULXP 'LVHDVHV ELRORJ\ DQG WD[RQRP\ 3( 1HOVRQ 7$ 7RXVVRXQ DQG 5&RRN HGV 7KH 3HQQ 6WDWH 8QLY 3UHVV 8QLYHUVLW\ 3DUN 3$ &ODUN $ DQG /XQLJDQ &06 3URVSHFWV IRU HVWLPDWLQJ QXFOHRWLGH GLYHUJHQFH ZLWK 5$3'V 0RO %LRO (YRO f &RUUHOO -& .OLWWLFK &-5 DQG /HVOLH -) 1LWUDWH QRQXWLOL]LQJ PXWDQWV RI )XVDULXP R[\VSRUXP DQG WKHLU XVH LQ YHJHWDWLYH FRPSDWLELOLW\ WHVWV 3K\WRSDWKRORJ\ (OLDV .6 DQG 6FKQHLGHU 5: 9HJHWDWLYH FRPSDWLELOLW\ JURXSV LQ )XVDULXP R[\VSRUXP I VS O\FRSHUVLFL 3K\WRSDWKRORJ\ (OLDV .6 =DPLU /LFKWPDQ3OHEDQ 7 DQG .DWDQ 7 3RSXODWLRQ VWUXFWXUH RI )XVDULXP R[\VSRUXP I VS O\FRSHUVLFLn 5HVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV SURYLGH JHQHWLF HYLGHQFH WKDW YHJHWDWLYH FRPSDWLELOLW\ JURXS LV DQ LQGLFDWRU RI HYROXWLRQDU\ RULJLQ 0RO 3ODQW0LFUREH ,QWHUDFW )$2 )$2 SURGXFWLRQ \HDUERRN )$2 5RPH )DUPDQ 0 / 7DXUD 6 DQG /HRQJ 6$ 7KH 0DJQDSRUWKH JULHVHD '1$ ILQJHUSULQWLQJ SUREH 0*5 FRQWDLQV WKH n HQG RI DQ LQYHUWHG UHSHDW WUDQVSRVRQ 0RO *HQ *HQHW )HUQDQGH] $VVLJEHWVH .% 'XERLV 03 DQG *HLJHU -3 0ROHFXODU FKDUDFWHUL]DWLRQ RI UDFHV DQG YHJHWDWLYH FRPSDWLELOLW\ JURXSV LQ )XVDULXP R[\VSRUXP I VS YDVLQIHFWXP $SSO (QYLURQ 0LFURELRO *RRGZLQ 6 % 6SLHOPDQ /0DWXV]DN -0 %HUJHURQ 61 DQG )U\ :( &ORQDO GLYHUVLW\ DQG JHQHWLF GLIIHUHQWLDWLRQ RI 3K\WRSKWKRUD LQIHVWDQV SRSXODWLRQV LQ QRUWKHUQ DQG FHQWUDO 0H[LFR 3K\WRSDWKRORJ\ *RUGRQ 75 DQG 2NDPRWR 3RSXODWLRQ VWUXFWXUH DQG WKH UHODWLRQVKLS EHWZHHQ SDWKRJHQLF DQG QRQSDWKRJHQLF VWUDLQV R )XVDULXP R[\VSRUXP 3K\WRSDWKRORJ\ *RUGRQ 75 2NDPRWR DQG 0LOJURRP 0* 7KH VWUXFWXUH DQG LQWHUUHODWLRQVKLS RI IXQJDO SRSXODWLRQV LQ QDWLYH DQG FXOWLYDWHG VRLOV 0ROHFXODU (FRORJ\ ,QWHUQDWLRQDO 1HWZRUN IRU WKH ,PSURYHPHQW RI %DQDQD DQG 3ODQWDLQ ,1,%$3f $QQXDO UHSRUW -DFREVRQ 'DQG *RUGRQ 75 D )XUWKHU LQYHVWLJDWLRQV RI YHJHWDWLYH FRPSDWLELOLW\ ZLWKLQ )XVDULXP R[\VSRUXP I VS PHORQLV &DQ %RW

PAGE 117

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
PAGE 118

.RKQ /0 7KH FORQDO G\QDPLF LQ ZLOG DQG DJULFXOWXUDO SODQWSDWKRJHQ SRSXODWLRQV &DQ %RW 6XSSO f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f ([S 0\FRO /HY\ 0 &RUUHD9LFWRULD )=HLJOHU 56 ;X 6 DQG +DPHU -( *HQHWLF GLYHUVLW\ RI WKH ULFH EODVW IXQJXV LQ D GLVHDVH QXUVHU\ LQ &RORPELD 3K\WRSDWKRORJ\ /HY\ 0 5RPDR 0DUFKHWWL 0$ DQG +DPPHU -( '1$ ILQJHUSULQWLQJ ZLWK D GLVSHUVHG UHSHDWHG VHTXHQFH UHVROYHV SDWKRW\SH GLYHUVLW\ LQ WKH ULFH EODVW IXQJXV 3ODQW &HOO 0DQLFRP %4 DQG %DD\HQ 53 5HVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV LQ )XVDULXP R[\VSRUXP I 6S GLDQWKL DQG RWKHU IXVDULD IURP 'LDQWKXV VSHFLHV 3ODQW 3DWKRO 0DQLFRP %4 %DU-RVHSK 0 .RW]H DQG %HFNHU 00 $ UHVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVP SUREH UHODWLQJ YHJHWDWLYH FRPSDWLELOLW\ JURXSV DQG SDWKRJHQLFLW\ LQ )XVDULXP R[\VSRUXP I VS GLDQWKL 3K\WRSDWKRORJ\ 0F'RQDOG %$ DQG 0DUWLQH] -3 '1$ UHVWULFWLRQ IUDJPHQW OHQJWK SRO\PRUSKLVPV DPRQJ 0\FRVSKDHUHOOD JUDPLQLFROD DQDPRUSK 6HSWRULD WULWLFLf LVRODWHV FROOHFWHG IURP D VLQJOH ZKHDW ILHOG 3K\WRSDWKRORJ\ 0F'RQDOG %$ DQG 0DUWLQH] -3 '1$ ILQJHUSULQWLQJ RI WKH SODQW SDWKRJHQLF IXQJXV 0\FRVSKDHUHOOD JUDPLQLFROD DQDPRUSK 6HSWRULD WULWLFLf ([S 0\FRO 0F'RQDOG %$ DQG 0F'HUPRWW -0 3RSXODWLRQ JHQHWLFV RI SODQW SDWKRJHQLF IXQJL %LR6FLHQFH 0F'RQDOG %$ 0F'HUPRWW -0 *RRGZLQ 6 % DQG $OODUG 5: 7KH SRSXODWLRQ ELRORJ\ RI KRVWSDWKRJHQ LQWHUDFWLRQV $QQX 5HY 3K\WRSDWKRO

PAGE 119

0F'RQDOG %$ 0LOHV 1HOVRQ /5 DQG 3HWWZD\ 5( *HQHWLF YDULDELOLW\ LQ QXFOHDU '1$ LQ ILHOG SRSXODWLRQV RI 6WDJRQRVSRUD QRGRUXP 3K\WRSDWKRORJ\ 0F'RQDOG %$ 3HWWZD\ 5( &KHQ 56 %RHJHU -0 DQG 0DUWLQH] -3 7KH SRSXODWLRQ JHQHWLFV RI 6HSWRULD WULWLFL WHOHRPRUSK 0\FRVSKDHUH OD JUDPLQLFRODf &DQ %RW 6XSSO $' 0HV -9DQ 'RRP 5RHEURHFN (-$ 9DQ (JPRQG ( 9DQ $DUWULMN DQG %RRQHNDPS 30 5HVWULFWLRQ OHQJWK SRO\PRUSKLVPV UDFHV DQG YHJHWDWLYH FRPSDWLELOLW\ JURXSV ZLWKLQ D ZRUOGZLGH FROOHFWLRQ RI )XVFLULXP R[\VSRUXP I VS JOFLGLRORL 3ODQW 3DWKRO 0LFKHOPRUH 5: DQG +XOEHUW 6+ 0ROHFXODU PDUNHUV IRU JHQHWLF DQDO\VLV RI SK\WRSDWKRJHQLF IXQJL $QQX 5HY 3K\WRSDWKRO 0LOJURRP 0* DQG )U\ :) &RQWULEXWLRQV RI SRSXODWLRQ JHQHWLFV WR SODQW GLVHDVH HSLGHPLRORJ\ DQG PDQDJHPHQW $GYDQFHV LQ %RW 5HV LQ SUHVVf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f $IULFD LWV SHRSOHV DQG WKHLU FXOWXUDO KLVWRU\ 0F*UDZ+LOO 1< 1DPLNL) 6KLRPL7 .D\DPXUD7 DQG 7VXJH7 &KDUDFWHUL]DWLRQ RI WKH IRUPDH VSHFLDOHV RI )XVDULXP R[\VSRUXP FDXVLQJ ZLOWV RI FXFXUELWV E\ '1$ ILQJHUSULQWLQJ ZLWK QXFOHDU UHSHWLWLYH '1$ VHTXHQFHV $SSO (QYLURQ 0LFURELRO 3ORHW] 5& 3RSXODWLRQ ELRORJ\ R )XVDULXP R[\VSRUXP I VS FEHQVH SS ,Q )XVDULXP ZLOW RI EDQDQD 5& 3ORHW]HG $36 3UHVV 6W 3DXO 01 3ORHW] 5& )XVDULXP ZLOW RI EDQDQD 3DQDPD GLVHDVHf SS ,Q 3ODQW GLVHDVHV RI LQWHUQDWLRQDO LPSRUWDQFH YROXPH ,,, 3UHQWLFH +DOO (QJOHZRRG &OLIIV 13ORHW] 5& 3DQDPD GLVHDVH 5HWXUQ RI WKH ILUVW EDQDQD PHQDFH ,QWHU -RXU RI 3HVW 0JPW 3ORHW] 5& )XVDULXP ZLOW DQG SKDVH ,, RI ,037 SS ,Q -RQHV '5 HGf 7KH ,PSURYHPHQW DQG WHVWLQJ RI 0XVD $ JOREDO SDUWQHUVKLS ,1,% $3 0RQWSHOOLHU )UDQFH

PAGE 120

3ORHW] 5& %UDXQZRUWK :6 -U +DVW\ 6 *DQWRWWL % &KL]DOD &7 %DQGD '/1 0DNLQD ': DQG &KDQQHU $* )XVDULXP ZLOW RI EDQDQD 3DQDPD GLVHDVHf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f RQ (DVW $IULFDQ KLJKODQG FXOWLYDUV RI EDQDQD )UXLWV 3ORHW] 5& 9£]TXH] $ 1DJHO %HQVFKHU 6LDQJOHZ 3 6ULNXO 6 .RRDUL\DNXO 6 :DWWDQDFKDL\LQJFKDURHQ : /HUWUDW 3 DQG :DWWDQDFKDL\LQJFKDURHQ &XUUHQW VWDWXV RI 3DQDPD GLVHDVH LQ 7KDLODQG )UXLWV LQ SUHVVf 3XKDOOD DQG 6SHLWK $ FRPSDULVLRQ RI KHWHURNDU\RVLV DQG YHJHWDWLYH LQFRPSDWLELOLW\ DPRQJ YDULHWLHV RI *LEEHUHOODIXMLNXURL )XVDULXP PRQLOLIRUPHf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f 6QHDWK 3+ DQG 6RNDO 55 1XPHULFDO WD[RQRP\ : + )UHHPDQ 6DQ )UDQFLVFR &$

PAGE 121

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f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

PAGE 122

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

PAGE 123

%,2*5$3+,&$/ 6.(7&+ 5RVDOLH /\QQ .RHQLJ ZDV ERUQ LQ )UHHKROG 1HZ -HUVH\ -DQXDU\ WR 0LOWRQ DQG /XFLOOH .RHQLJ 6KH ZDV UDLVHG RQ KHU IDPLO\f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f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

PAGE 124

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 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fA7f(GXDUGR 9DOOHMRV $VVRFLDWH 3URIHVVRU RI +RUWLFXOWXUDO 6FLHQFH 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI $JULFXOWXUH DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 0D\ 'HDQ *UDGXDWH 6FKRRO


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E47HIX5Y2_7P2GKX INGEST_TIME 2014-08-14T19:57:04Z PACKAGE AA00024916_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


PHYLOGENETIC AND POPULATION GENETIC ANALYSES
OF FUSAR1UM OXYSPORUMF. SP. CUBENSE,
THE CAUSAL AGENT OF FUSARIUM WILT ON BANANA
BY
ROSALIE LYNN KOENIG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1997

Copyright 1997
by
Rosalie Lynn Koenig

I dedicate this research to my father and mother who always stressed the
importance of education, and to my husband Tom and daughter Amaleah, who put
with me during this stressful period of my life.

ACKNOWLEDGMENTS
I would like to thank Dr. Corby Kistler for his support, patience, understanding
and encouragement through all stages of my graduate studies.
I would like to thank Dr. Dave Mitchell for his guidance, counsel, encouragement
and personal commitment to excellence and graduate students. Many thanks to my
committee members Drs. Randy Ploetz and Eduardo Vallejos for their input and
cooperation. Additional thanks to Ulla Benny, Gerald Benny and Lyndel Meinhardt who
always lended their technical expertise and support.
Special thanks to Dr. Brian Bowen for his expertise in genetic analysis; he always
made time for me and provided clear explanations of complex subject matters. I thank the
NSEP for funding the field work in Honduras. I thank Drs. Maricio Rivera and Phil Rowe
for their support in Honduras. Thanks to my many friends and classmates who shared
their dreams and aspirations.
IV

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iv
LIST OF TABLES vii
LIST OF FIGURES viii
ABSTRACT ix
CHAPTERS
1 INTRODUCTION 1
The Host 4
Genetic Characterization of Fusarium oxysporum f. sp. cúbense 8
Clonality and its Consequences in the Study of Population Genetics of
Asexually Reproducing Organisms 11
2 FUSARIUM OXYPSORUM F. SP. CUBENSE CONSISTS OF A
SMALL NUMBER OF DIVERGENT AND GLOBALLY
DISTRIBUTED CLONAL LINEAGES 15
Introduction 15
Materials And Methods 17
Results 30
Discussion 42
3 GENETIC VARIATION IN TWO HONDURAN FIELD
POPULATIONS OF FUSARIUM OXYSPORUM F. SP. CUBENSE 47
Introduction 47
Materials and Methods 50
Results 60
Discussion 65
4 SUMMARY AND CONCLUSIONS 73
v

APPENDICES
A EXPLORING THE POTENTIAL OF USING POLYMERASE CHAIN
REACTION METHODOLOGIES IN PHYLOGENETIC STUDIES
OF FUSA1UUM OXYSPORUM F. SP. CUBENSE 81
Introduction 81
Results and Discussion 82
B ADDITIONAL INFORMATION FROM PHYLOGENETIC STUDY 91
C ALLELIC DATA SCORED AS PRESENCE OR ABSENCE FOR EACH
PROBE ENZYME COMBINATION; ISOLATE NAME AND
VEGETATIVE COMPATIBILITY GROUP INCLUDED AS A
REPRESENTATIVE FOR EACH UNIQUE RESTRICTION
FRAGMENT LENGTH POLYMORPHISM HAPLOTYPE 94
D LIST OF ISOLATES OF F. OXYSPORUM Y. SP. CUBENSE, THEIR
VEGETATIVE COMPATIBILITY GROUP, HAPLOTYPE AND
LINEAGE 100
REFERENCES 105
BIOGRAPHICAL SKETCH 113
vi

LIST OF TABLES
Table page
2-1: List of isolates, their vegetative compatibility groups (VCG), clonal lineages,
the cultivar, and geographical regions from which they were collected 18
2-2: Similarity matrix of simple matching coefficients based on restriction fragment
length polymorphisms for selected isolates of Fusarium oxysporum f. sp.
cúbense, F. oxysporum f. sp. lycopersici and F. oxysporum f. sp. niveum .... 35
2-3: Clonal lineages oí Fusarium oxysporum f. sp. cúbense isolates, their
geographical distributions and corresponding vegetative compatibility
groups (VCG) 36
2-4: A comparison of allelic data for isolates in FOC I, FOC II and FOC VII 39
2-5: Clone-corrected measurements of gametic disequilibrium among pairs of
alleles in a world-wide collection of Fusarium oxysporum f sp. cúbense 41
3-1. Genotypes of banana planted in Field 1 51
3-3. Field design of banana plants in rows of the race 2 disease screening plot 55
3-4. Isolates of Fusarium oxysporum f. sp. cúbense collected in Field 1 and their
incidence on different host genotypes 63

LIST OF FIGURES
Figure page
2-1: DNA of isolates was digested with Eco RF restriction enzyme and probed
with clone 187 26
2-2: Frequency distribution of RFLP haplotypes among the 165 isolates
representing a world collection of Fusarium oxysporum f. sp. cúbense 31
2-3: Midpoint rooted 50% majority rule consensus tree representing 500
bootstrap replicates 32
2-4: Mitochondrial DNA haplotypes of selected Fusarium oxysporum f. sp.
cúbense isolates 40
3-1: mtDNA haplotype of selected Honduran field isolates 61
3-2: Hindlll digested DNA of selected Honduran field isolates that were probed
with clone pEYlO to generate DNA fingerprints 64
3-3: Incidence and distribution of mtDNA haplotypes of Fusarium oxysporum f.
sp. cúbense on host genotype 66
viii

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PHYLOGENETIC AND POPULATION GENETIC ANALYSES
OF FUSARJUM OXYSPORUMF. SP. CUBENSE,
THE CAUSAL AGENT OF FUSARIUM WILT ON BANANA
By
Rosalie Lynn Koenig
May 1997
Chairperson: Dr. H. C. Kistler
Major Department: Department of Plant Pathology
A worldwide collection of Fusarium oxysporum f. sp. cúbense was analyzed using
anonymous, single-copy, restriction fragment length polymorphism (RFLP) loci. Of the
165 isolates examined, 72 RFLP haplotypes were identified. Consistent with strict clonal
reproduction, individuals with identical haplotypes were geographically dispersed, and
clone-corrected tests of gametic disequilibrium indicated significant nonrandom
association among pairs of alleles. Parsimony analysis divided haplotypes into two major
branches. Ten clonal lineages were identified based on coefficients of similarity between
0.94-1.00. The two largest lineages had pantropical distribution. Isolates comprising one
lineage (FOC VII) may represent either an ancient genetic exchange between individuals in
the two largest lineages or an ancestral group. The two largest lineages (FOC I and II)
and a lineage from East Africa (FOC V) are genetically distinct; each may have acquired
the ability to be pathogenic on banana independently.
IX

Populations of the pathogen were sampled from two disease screening fields in
Honduras. Field 1 was planted to 10 genotypes of banana and 97 isolates were recovered
from host tissue. Isolates recovered in Field 1 comprised three mtDNA haplotypes, and
their incidences and distributions varied with the host genotype sampled. DNA
fingerprints indicated that the isolates represented five clones. Field 2 was planted to a
single genotype of banana. The 59 isolates collected from host tissue represented a single
mtDNA haplotype and clone. The frequencies with which the pathogen clones infected
the host were dependent upon host genotype. Two races of the pathogen (race 1 and 2)
have been reported in Honduras. Since the cultivar Highgate is susceptible to race 1 but
resistant to race 2, isolates recovered from the cultivar Highgate were assumed to
represent race 1. Isolates within a lineage associated with race 1 isolates were
preferentially recovered from Highgate and the Fundación Hondureña de Investigación
Agrícola (FHIA) hybrids. Isolates representing a second lineage were recovered at a low
frequency from Highgate, suggesting that these two lineages differ in their ability to infect
bananas susceptible to race 1. Assessing the pathogen populations and determining the
lineages of isolates infecting particular banana breeding lines could be useful for banana
breeding programs.
x

CHAPTER 1
INTRODUCTION
The genus Fusarium (Link) is comprised of fungi which have evolved from an
ancient ancestor whose geological age has been postulated to date back at least 200
million years (Snyder, 1981). Whether or not this is an accurate estimate of the age of this
important fungal group, the genetic diversity present in members of the genus Fusarium is
the result of evolutionary processes which have developed over a long time span.
Fusarium oxysporum (Schlecht.) Synd. & Hans, is an ubiquitous, soil borne fungus
that is comprised of strains that are ecologically adapted to diverse environments and
which have the ability to utilize an array of substrates in the soil (Burgress, 1981). The
reproductive success of strains within this species depends on the ability to produce a
large number of propagules, in the form of microconidia or macroconidia, as well as
resistant resting spores, referred to as chlamydospores, that can ensure long-term survival
during adverse environmental conditions. The morphological characteristics of these
propagules also have been used by Wollenweber and Reinking (1935) to include the
species in section Elegans. No teleomorph have been described for the species.
Plant pathogenic isolates of F. oxysporum cause diseases known as vascular wilts
and constitute a small portion of the strains within this species. However, because of their
importance to agriculture, Snyder and Hansen (1940) developed the concept oiformae
speciales to recognize plant pathogenic strains of F. oxysporum that were morphologically
1

2
indistinguishable from saprophytic strains of the species but had the ability to cause
disease on specific hosts. Over 70 formae speciales have been described (Kistler, 1997).
Although formae speciales are not recognized by the International Code of Botanical
Nomenclature, they are useful to plant pathologists because they provide informal
groupings of morphologically indistinguishable isolates. Isolates within a forma specialis
may be pathogenic on a specific host, such as F. oxysporutn f. sp. pi si, which causes a wilt
disease on peas, Pi sum sativum L., or may have a broader host range, such as F.
oxysporum f. sp. vasinfectum (Atk.) Snyd. & Hans., which is pathogenic on plants in the
families Malvaceae, Solanaceae and Fabaceae (Windels, 1991). Two assumptions that
have been made about isolates within a forma specialis are that 1) they are genetically
more similar to each other than to isolates of different formae speciales, and 2) that formae
speciales are monophyletic groupings (Kistler, 1997).
Fusarium wilt of banana, commonly referred to as Panama disease, is caused by
F. oxysporum Schlechtend.: Fr. f. sp. cúbense (E.F. Sm.) W.C. Synder & H. N. Hans.
This fungus systemically colonizes the xylem tissue of susceptible banana cultivars,
disrupting normal water uptake (Beckman, 1987). This disruption eventually results in wilt
and the subsequent death of the plant. Studies described by Beckman (1990) indicate that
conidia in aqueous solutions are readily taken up into the vascular elements and carried in
the transpiration stream to sites of entrapment at vessel endings. Subsequent secondary
colonization occurs in susceptible cultivars by the dispersal of conidia through the vascular
system. Resistant cultivars prevent spread by the formation of occluding gels and tyloses
that wall off the lumena of the infected vessels.

To date, three races of the pathogen, 1, 2 and 4, have been described which are
pathogenic on banana; these races are differentiated by the banana cultivars Gros Michel,
Bluggoe, and Cavendish, respectively. Historically, this pathogen has caused large scale
disease epidemics when susceptible banana cultivars have been grown in the presence of
the pathogen (Ploetz, 1994).
The first recording of Fusarium wilt of banana was made in southeast Queensland
in 1874, on the cultivar Silk, known locally as Sugar (Bancroft, 1976). The history of this
disease generally paralleled the dissemination of susceptible genotypes to new areas. The
edible bananas are asexually propagated, typically using rhizomes or sucker plants.
Occasionally, infested rhizomes were brought to new plantations and this is likely how the
pathogen was widely distributed. Interestingly, one region where the pathogen has been
found only recently is the vast area of Oceania where susceptible cultivars have been
grown for millennia (Shivas and Philemon, 1996).
Prior to the late 1950s, monoculture of the race 1-susceptible cultivar Gros Michel
resulted in severe and wide-spread epidemics in the export banana trade. The disease
became economically less important in export production once the trade converted to race
1-resistant Cavendish cultivars. However, approximately 20-40 years after Cavendish
cultivars were planted widely in a number of subtropical regions, they too began to
succumb to the disease. Outbreaks on Cavendish cultivars were reported in the Canary
Islands in 1926 (Ashby, 1926), the 1940s in South Africa, the 1950s in Australia, and the
1960s in Taiwan (Ploetz et al., 1990). In the late 1970s outbreaks on Cavendish were
observed in the Mindinao province of the Philippines. However, in contrast to outbreaks
in the subtropics, disease outbreaks in the Philippines occurred in localized patches that
m

4
tended not to spread (Ploetz et al., 1990). All isolates in VCG 0122 are from the
Philippines and they all have been recovered from Cavendish cultivars.
In the late 1970s Su et al. (1977) conclusively demonstrated that isolates capable
of causing disease on Cavendish were different from race 1 and 2 isolates. These isolates,
which also cause disease on race 1- and race 2- susceptible cultivars, were designated
race 4.
In addition to economic losses in the export trade, Panama disease is a serious
problem on locally consumed dessert bananas and cooking bananas (Ploetz, 1992; Ploetz
et al., 1992; 1994; 1995). Only 14% of the bananas that are produced in a given year are
exported. The remaining 86% (65 million metric tons in 1992) are consumed locally
(FAO, 1993) and are often staple foods for poor, subsistence farmers in the tropics
(INIBAP, 1993). In total, the impact of Fusarium wilt on these bananas is far greater than
on bananas produced for export.
The Host
The F. oxysporum f. sp. cúbense', banana pathosystem likely developed from a
process of co-evolution between the host and pathogen, resulting in complementary
evolutionary changes in both organisms. Bananas belong to the order Zingiberales, which
consists of eight families of tropical and subtropical rhizomatous perennial plants that
inhabit moist forested areas or invade disturbed sites (Kress, 1990). Bananas belong to
the family Musaceae, which contain three genera (Musa, Musella, and Ensete) whose
natural range extends from West Africa to the Pacific. Edible bananas are the most
important cultivated crop of this family. Two additional minor crops of economic

5
importance are abacá (M textilis), a fiber crop, and cultivated forms of Ensete
ventricosum, a food and fiber plant of upland Abyssinia.
Edible bananas have been domesticated from two species, M. acuminata and M
balbisiana. Simmonds and Shepherd (1955) used 15 phenotype characters and scored
each on the scale of 1 (accuminata expression) to 5 (balbisiana expression) to develop a
system to classify edible bananas. From this, the genetic constitution of any edible banana
can be specified using a genome formula indicating the contributions from M accuminata
(AA) and M. balbisiana (BB).
Edible M. acuminata cultivars are either diploid (AA) or triploid (AAA),
and their center of origin includes Southern Thailand, Sumatra, Java, Borneo, Burma, and
Malaysia, which is also the primary center of diversity for the species. This species is most
diverse in peninsular Malaysia and the equatorial islands where the wild species is native.
Six subspecies of M acuminata are recognized. Although the banana clone Gros Michel
has phenotypic characters that suggest descent from M acuminata subsp. malaccensis
(Stover, 1990), the parents of most clones are not known.
Simmonds (1962) argues that edibility in bananas arose in Malaysia through
parthenocarpy in M. acuminata diploids followed by triploidy (AAA). Through the
process of domestication, man has selected for both parthenocarpy and sterility, which are
genetically independent characteristics. Additionally, domestication has led to selection of
triploidy because it contributes to high plant vigor and fruit growth rate, and confers
gametic sterility (Simmonds, 1962).
In contrast, M. balbisiana has a center of origin in a region including the countries
of India, Burma, and Northern Thailand. It is less genetically diverse than M acuminata.

6
It is presumed that edibility did not arise in M ba/bisiatia because no naturally occurring
parthenocarpic clones have been identified. Similarly, no purely M. balbisiana triploids or
tetraploids are known. M balbisiana is naturally cross fertile with a number of Musa
species, including M acuminata, but it is reproductively quite isolated (Simmonds, 1962).
The hybrid groups (AB, AAB, ABB) are peripheral, being diversified in India and, to a
lesser extent, in Indochina and eastern Malaysian areas from which wild M. acuminata is
locally absent but in which M balbisiana is native (Simmonds, 1962). The transport by
humans of AA-type bananas into areas occupied by wild M. balbisiana is inferred to have
provided the opportunity for interspecific hybridization (Simmonds, 1962).
Characteristics such as plant hardiness, drought resistance, disease resistance to
nematodes and yellow Sigatoka, and starchiness of fruit that are found in interspecific
hybrid bananas are believed to be contributed fromM balbisiana (Simmonds, 1962).
Interspecific hybrids between M. balbisiana and M. acuminata occur in nature because
the species are sympatric over a large area of southeast Asia and introgression occurs
often.
A secondary center of diversity for bananas is East Africa. In East Africa,
mixtures of genotypes are grown for local use and particular dietary needs. The East
African baking and beer type bananas (AAA) are unique to the region and are often found
in mixed plantings with cultivars of exotic origin, such as ABBs and various dessert types
(Ploetz et al., 1990). Inland cultivation of banana in East Africa is based primarily on
triploid clones (AAA) but the coastal areas have much more diverse populations. Coastal
East Africa is the only location outside of Southeast Asia where there is an assemblage of
edible diploids cultivars and extensive local diversification. The most striking

7
accumulations of unique clones are in upland East Africa and the Pacific Islands, where it
appears that subsistence farmers actively preserved these clones (Simmonds, 1962).
Murdock’s (1959) discussion of plant introduction in relation to African history
suggests that Malaysian food plants entered Africa in the hands of Indonesian migrants
about 2000 B.P. Bananas, as well as taro, yams, breadfruit, coconut, some rices and
sugarcane, are adapted to moist climates and became staple foods of the wetter parts of
Central Africa (Simmonds, 1962). Madagascar appears to be the principal point of entry
of bananas into Africa. Simmonds' (1962) view is that the early African cultivars reached
Madagascar from Malaysia late in the first half of the first millennium A.D. From West
Africa, Portuguese travelers took plants to the Canary Islands early in the fifteenth
century, and a least one clone traveled to Hispaniola in 1516. Simmonds (1962) believed
that this represented the first of many introductions made into the New World. However,
Langdon (1993) suggests, based on linguistic, botanical, and archeological data, that
banana was present in the New World before the Spaniards arrived.
Presently, the only location with a high level of genetic diversity of M acuminata
is Papua New Guinea and surrounding islands. Here, the indigenous people have
preserved an intact population of primitive cultigens, a relic of an early stage in the
evolution of the group. Jarret (1992) used 98 phylogenetically informative restriction
fragment length polymorphisms (RFLPs) to characterize eight species representing three
sections of Musa from Papua New Guinea, as well as one species of Ensete. The study
found only a distant relationship between M balbisiana and other species of Musa; M.
balbisiana was not in the same clade as the other species within the section Musa (also

8
referred to as Eumusa). The diploid land races from this section clustered withM
acuminata. All accessions of M acuminata shared a large number of alleles.
Genetic Characterization of Fusarium oxvsponim f, sp. cúbense
Cladistic analyses of the phylogenetic relationships among isolates of a fungal
species are based on the concept that members of a species have evolved from a common
ancestor (Scotland, 1992). This ancestor possessed primitive genetic characteristics which
may have changed through the evolutionary history of the organism. Derived character
changes, those changes that are different from the ancestral type, are the result of
evolutionary processes which consist of both genetic (mutation, recombination, and
transposon rearrangements) and spatial (migration, divergence, and convergence) forces.
Evolutionary studies of organisms aim to determine phylogeny by examining the
genetic relatedness among members of a group. Studies may involve closely related
organisms, such as fungal isolates of a given species, or organisms more distantly related,
such as all fungal isolates within a location. However, many evolutionary studies are
conducted at the genus or species level. These types of studies rely on identifying genetic
markers that discern among individuals sampled in the study. Prior to the development of
molecular techniques, researchers primarily relied on phenotypic traits to develop
evolutionary theories. However, the development of molecular techniques, such as
isozyme analysis, restriction fragment length polymorphism analysis, random amplified
DNA analysis and gene sequencing, have made it possible to objectively assess traits at the
genotypic level.
Isolates of F. oxysporum have been characterized using morphological,
biochemical and genetic markers (Kistler, 1997). Additionally, vegetative compatibility

groups (VCGs) based on work on Fusarium moniliforme also have been used to
differentiate similar isolates of F. oxysporum (Puhalla and Spieth, 1985). Isolates of F.
oxysporum which share identical alleles at the loci governing heterokaryon incompatibility,
commonly referred to as het or vie, are vegetatively compatible. Conventionally, this is
determined via the ability of nitrate-nonutilizing (nit) auxotrophic mutants to complement
one another for nitrate utilization (Leslie, 1993). Since F. oxysporum has no known
sexual stage, it is not possible to determine genetically the number of loci involved in
vegetative incompatibility. However, genetic studies performed on F. moniliforme
(teleomorph- Gibberella fujikuroi) indicate that a minimum of 10 vie loci govern
vegetative compatibility in this closely related species (Puhalla and Spieth, 1985).
Unfortunately, the relationship among isolates comprising different VCGs is impossible to
discern based solely on this technique because a single change of alleles at any of the vie
loci will generate a new VCG.
Vegetative compatibility group diversity varies among the known formae speciales
of F. oxysporum. For example, among 115 isolates of F. oxysporum f. sp. lycopersici,
Elias and Schneider (1991) identified one major and two minor VCGs, as well as a number
of single-member VCGs. In contrast, a worldwide collection of 52 isolates of F.
oxysporum f. sp. vasinfectum isolates comprised 10 VCGs and isolates belonging to
distinct races of the pathogen were never in the same VCG (Fernandez et al., 1994).
Currently, at least 17 VCGs have been described for F. oxysporum f. sp. cúbense (Ploetz,
1994). The majority of VCGs appear to have limited geographical distribution.
However, the majority of isolates belong to a few major VCGs that are geographically
widespread and contain more than one race of the pathogen (Ploetz, 1994).

10
Until the advent of tissue culture, the rhizome of the banana plant was used to
propagate banana. It is likely that the worldwide movement of planting stock (rhizomes
and associated soil particles) from region to region is responsible for much of the
geographical distribution of the pathogen outside its region of origin. In terms of
evolution, the diversity of the pathogen in areas outside of its center of origin will depend
on the number of years the pathogen has been present in the new location and whether
single or multiple clonal lineages were introduced. For example, one might expect the
populations of F. oxyspomm f. sp. cúbense in East Africa to be more genetically variable
than populations in the USA because the crop was introduced into East Africa thousands
of years prior to the introductions to Florida.
Many earlier genetic characterizations of various formae speciales of F. oxysporum
focused on a small number of isolates with limited geographical distribution. These
studies gave an incomplete picture of genetic diversity when more extensive testing using
a larger number of isolates with a large geographical distribution was performed. For
example, when 170 isolates of the pathogen F. oxysporum f. sp. dianthi were collected
from 42 cultivars at 40 different sites in Israel, all of the isolates belonged to the same
VCG (Katan et al., 1989). However, when isolates of the same pathogen were sampled
from a number of geographical locations, including Italy, Spain, The Netherlands, the
United States and Germany, eight VCGs were identified (Aloi and Baayen, 1993). The
latter researchers found not only single races within single VCGs but also multiple races
within single VCGs (Aloi and Baayen, 1993).
Isolates of F. oxysporum f.sp. cúbense also have been grouped based on
electrophoretic karyotype, random amplified polymorphic DNA (RAPD) analysis and the

11
production of volatiles in culture. Boehm et al. (1994) proposed two groupings of isolates
of F. oxysporimr f.sp. cúbense based on similarities in chromosome number and genome
size. In general, these groupings agreed with RAPD data (Bentley et al., 1995, Sorensen
et al., 1993) and groups determined when the formation of aldehydes by different isolates
were examined (Brandes, 1919; Moore et al., 1991, 1993; and Stover, 1959).
Clonalitv and Its Consequences in the Study of Population Genetics of Asexually
Reproducing Organisms
Fusariiim oxysporum is considered to be an asexual fungus based on the fact that,
although many researchers have searched for a sexual stage in both the laboratory and
field, none has been reported. For an asexually reproducing organism like F. oxysporum,
it is generally assumed that isolates within a VCG are genetically similar and represent
clonal populations. Many asexually reproducing populations consist of a number of clonal
lineages with wide geographical distribution. Kohli et al. (1992), using a DNA
fingerprinting probe and mycelial compatibility, detected 39 clones of Sclerotinici
sclerotiorum on canola from three provinces in Canada. The most widely distributed clone
was found in all three provinces. Overall, they found that clones of S. sclerotiorum were
distributed over large geographic areas and field populations were composed of more than
one clone.
Patterns of genetic variability in natural populations of fungi may reflect strict
clonality, panmixia or a mixture of clonality and mating (Anderson and Kohn, 1995). In
asexually reproducing organisms, there is presumably no contribution to genotypic
variability through meiotic recombination or introgression. In general, genotypic
variability in asexual fungi may be generated through mutation, random genetic drift,

12
parasexuality, somatic recombination and transposable elements. Although mutation rates
are generally low per locus, mutations are persistent and over time they generate much
variability, summed over all loci (McDonald, et al., 1989). Tibrayrenc et al. (1990)
established criteria that characterize asexually reproducing organisms. They include
over-represented, ubiquitous multilocus genotypes in a population; the absence of
recombinant genotypes; correlation between two independent sets of genetic markers; and
the presence of linkage disequilibrium. Genetic analyses of a large collection of isolates
using neutral, single-copy, polymorphic loci ideally can test for the absence of
recombination.
Recently there has been an increased interest in genetically characterizing field
populations of plant pathogenic fungi. Many field studies have been conducted which
have characterized the genetic diversity of fungal populations within a defined
geographical location. The concept of a fungal population has been described by
McDonald and McDermott (1993) as a group of individuals sharing a common gene pool
and present in a limited geographical area. Population genetic analysis involves the
identification of genetic markers that are unambiguous and informative [polymorphic]
(McDonald and McDermott, 1993). These types of markers have been used to
characterize populations of a number of fungal species, including Cryphonectrict
parasitica, Fusarium oxysporum, Sclerotinia sclerotiomm, Phytophthora infestam,
Mycosphaerella graminicola, Cochliobolus carbonum, Phytophthora megasperma f. sp.
glycinea, Magnaporthe grísea and Stagonospora nodorum (Appel and Gordon, 1994;
Goodwin et al., 1992; Koh et al., 1994; Kohli et al., 1992; Kohli et al.,1995; Leonard and
Leath, 1990; Levy et al., 1991; Levy et al., 1993; McDonald and Martinez, 1990;

13
McDonald et al., 1994; McDonald et al., 1995, Sujkowski, et al., 1994; Whisson et al.,
1992).
Population genetic studies of Fusarium oxysporum f. sp. melonis have addressed
genetic structure in both agricultural and native soils and also have examined the genetic
relationships among pathogenic and nonpathogenic isolates by characterizing their VCGs
and mitochondrial haplotypes (Appel and Gordon, 1994, Gordon and Okamoto, 1992;
Gordon et al., 1992,). Based on parsimony analysis of the mtDNA data, nonpathogenic
isolates from Maryland were more closely related to F. oxysporum f. sp. melonis than to
the California nonpathogens (Appel and Gordon, 1994). Based on mtDNA, both
pathogen VCGs were more closely related to a nonpathogen VCG than to each other
(Appel and Gordon, 1994). Essentially all diversity in mtDNA was found among the
nonpathogenic strains of the species, whereas pathogenic strains were represented by only
a single mtDNA haplotype at any one location (Gordon and Okamoto, 1992; Jacobson
and Gordon, 1990b) indicating that pathogenic strains are less genetically diverse than non
pathogenic strains. In contrast, a study of the genetic relatedness among 120 strains of F.
oxysporum in native and cultivated soils consisted of 23 different mtDNA haplotypes, of
which the five most common haplotypes accounted for 78% of the isolates. Isolates
representing these haplotypes were found in both native and agricultural soils (Gordon et
al., 1992). Seventy two percent of the isolates found in cultivated soil were associated
with the same mtDNA haplotype as one or more of the isolates in the native soil (Gordon
et al., 1992).
Host populations can act as powerful selective forces on pathogen populations and
vice versa (McDonald et al., 1989). For example, the rice blast fungus, Magnaporthe

14
grísea, can be divided into a large number of pathotypes. Pathotypes are analogous to
races in that specific pathotypes of a pathogen are virulent on certain genotypes of the
host. Using a repetitive fingerprinting probe that serves as a genealogical index among
rice blast isolates, Levy et al. (1993) examined the population structure and associated
virulence properties among isolates in a resistance-breeding nursery in Colombia.
Sampling the nursery population enabled the recovery of 39 pathotypes among 151
isolates. Using the fingerprinting probe called MGR 586, they could distinguish six
distinct clonal lineages, each with a generally non-overlapping cultivar range and an
associated set of several related pathotypes. Eighty-five percent of the pathotypes were
lineage specific. Their results support the view that cultivar specificity and pathotype
evolution have developed within constraints imposed by the genetic background of each
lineage (Levy et al., 1993).
The overall objectives of this dissertation research were to determine the
phylogenetic relationships among isolates of F. oxyspontm f. sp. cúbense and to determine
the level of genetic variability and its distribution in field populations of the pathogen on
banana. An RFLP study of a world collection of isolates was undertaken to assess the
phylogenetic relationships among isolates. A study which characterized isolates by their
mitochondrial haplotype and distinguished between different clones by utilizing a DNA
fingerprinting probe was undertaken to assess the level and distribution of genetic
variability in two Honduran field populations. Both of these studies should contribute to
the overall understanding of the genetic structure and relationships among isolates ofF.
oxysporum f. sp. cúbense.

CHAPTER 2
FUSAR1UM OXYPSORUMY. SP, CUBENSE CONSISTS OF A SMALL NUMBER OF
DIVERGENT AND GLOBALLY DISTRIBUTED CLONAL LINEAGES
Introduction
Fusarium wilt of banana, commonly referred to as Panama disease, is caused by
Fusarium oxyspomm Schlechtend.: Fr. f. sp. cúbense (E.F. Sm.) W.C Synder & H. N.
Hans. Isolates of F. oxyspomm f sp. cúbense previously have been characterized by
morphology or biochemical and genetic markers. On morphological bases, F. oxyspomm
was included in section Elegans by Wollenweber and Reinking (1935). No teleomorph has
been described for the species.
Vegetative compatibility groups (VCGs) have been used to categorize isolates of
F. oxyspomm, including Fusarium oxyspomm f. sp. cúbense. Based on data from other
fungi, isolates that share identical alleles at the loci governing heterokaryon
incompatibility, commonly referred to as het or vie, are vegetatively compatible (Leslie,
1993). Conventionally, this is determined by the ability of nitrate-nonutilizing (nit)
auxotrophic mutants to complement one another for nitrate utilization (Correll et al.,
1987). Currently, at least 17 VCGs have been described for this forma specialis (Ploetz
and Correll, 1988; Ploetz, 1994; Ploetz et al., 1997). The majority of isolates belong to
two major VCGs that have pantropical distribution and contain more than one race of the
pathogen. Minor VCGs were found to have a more limited geographical distribution
(Ploetz and Correll, 1988; Ploetz, 1990). For an asexually reproducing organism like F.
15

16
oxysporum, it is generally assumed that isolates within a VCG are genetically similar and
represent clonal populations (Appel and Gordon, 1994; Jacobson and Gordon, 1991;
Tantaoui et al., 1996).
In addition to characterizing isolates by VCG, isolates of F. oxysporum f. sp.
cúbense have been grouped based on electrophoretic karyotype, randomly amplified
polymorphic DNA (RAPD) analysis and the ability to produce volatile organic compounds
in culture. Boehm et al. (1994) proposed two groupings of isolates of F. oxysporum f. sp.
cúbense based on similarities in chromosome number and genome size. Group I
comprised isolates in VCGs 0124, 0125, 0124/0125, 01210 and 01214. Group II
comprised isolates in VCGs 0120, 0121, 0122, 0123, 0129 and 01213. In general, these
groupings agreed with RAPD data (Bentley et al , 1995; Sorensen et al., 1993). When the
presence or absence of RAPD bands were treated as binary data and subjected to phenetic
analysis based on the unweighted pair group method with arithmetic mean (UPGMA),
isolates in VCGs 0120, 0121, 0122, 0126, 01210, 01211 and 01212 formed one group
and isolates in VCGs 0123, 0124, 0124/0125 and 0125 formed a second group.
Similar major groups of isolates were evident when differentiation was based on
the formation of aldehydes in culture. Brandes (1919) noted that certain isolates of the
pathogen produced these odorous compounds when grown on steamed rice, whereas
others did not; the latter isolates were classified as variety inodoratum. Stover (1959)
examined a larger set of isolates from tropical America and the Caribbean and noted that
production of odorous compounds was a consistent and repeatable trait. Those isolates
which produced the aldehydes were referred to as cultivar Odoratum whereas, those
which did not were referred to as cultivar Inodoratum. More recently, Moore et al. (1991;

17
1993) analyzed the production of these compounds with high pressure liquid
chromatography. Isolates in VCGs 0120, 0129 and 01211 produced characteristic volatile
profiles, whereas isolates in VCGs 0123, 0124, 0124/0125, 0125 and 0128 did not.
Restriction fragment length polymorphisms (RFLPs) also have been employed to
determine the genetic relationships among isolates of F. oxysporum. These markers are
ideally suited to genetic diversity studies because of the following characteristics: i) most
are selectively neutral; ii) polymorphisms tend to be more numerous compared to other
types of markers, such as isozymes; iii) they are reproducible; and iv) those identifying
random single-copy loci avoid problems that are associated with linkage of markers.
In this study, probes from F. oxysporum f. sp. lycopersici that correspond to
single-copy, anonymous loci (Elias et al., 1993) were used to identify polymorphic alleles
in F. oxysporum f. sp. cúbense. The objectives of the study were to determine whether
isolates within VCGs of F. oxysporum f. sp. cúbense were clonally derived and if clonal
lineages correlated with previously determined VCGs. Phylogenetic relationships among
the various VCGs of F. oxysporum f. sp. cúbense and F. oxysporum from other hosts
were also characterized to determine if host specificity is monophyletic.
Materials And Methods
Fungal Isolates
One hundred and sixty-five isolates representing worldwide distribution and 17
VCGs of F. oxysporum f. sp. cúbense from the collection located at the Tropical Research
and Education Center in Homestead, Florida, were selected for analysis (Table 2-1).
Additionally, three isolates of F. oxysporum f. sp. lycopersici and a single isolate of

Table 2-1: List of isolates, their vegetative compatibility groups (VCG), clonal lineages,
the cultivar, and geographical regions from which they were collected
18
VCGU
0120
0121
0122
Isolate
Cultivarv
Origin and collector”
Lineage
IC2
Cavendish
Icod de los Vinos, Canary Islands,d
II
22425
Cavendish
Wamuran, Queensland, Australian
II
ORT2
Cavendish
La Orotava, Canary Islands,d
II
0-1220
Mons
Queensland, Australian
II
GAL2
Cavendish
Las Galletas, Canary Islands,d
II
C2
Cavendish
Canary Islands/
II
ADJ2
Cavendish
Adeje, Canary Islands,d
II
Cl
Cavendish
Canary Islands, f
II
22424
Lady Finger
Moorina, Queensland, Australia,g
II
0-1222
Mons
Queensland, Australian
II
0-1219
Mons
Queensland, Australian
II
A2
Mons Mari
Australia/
II
ADJ1
Cavendish
Adeje, Canary Islands/
II
STGM1
Gros Michel
Costa Rica,i
II
3S1
Highgate
Honduras,i
IT
PAJ1
Cavendish
Pajalillos, Canary Islands/
II
ORT1
Cavendish
La Orotava, Canary Islands/
II
GAL1
Cavendish
Las Galletas, Canary Islands/
II
BUE1
Cavendish
Buenavista, Canary Islands/
II
NW
Williams
Natal, South Africa/
II
NH
Williams
Natal, South Africa/
II
NB
Cavendish
Natal, South Africa/
11
F9127
Grand Naine
South Africa,g
II
15638
X
Malaysian
II
FCJ7
Lacatan
Jamaica,q
II
Pacovan
Pacovan
Bahia, Brazil,n
II
MGSA1
SH3142
South African
II
SA6
Cavendish
Transvaal, South Africa/
II
SA4
SH3362
Natal, South Africa/
IT
SA3
Williams
Tansvaal, South Africa/
II
GM
Gros Michel
Taiwan/
9130
Cavendish
Taiwan,g
III
0-1124
X
Taiwan,c
IIIX
HI
Cavendish
Taiwan, e
III
ML
Cavendish
Taiwan/
III
TBR
Cavendish
Taiwan/
III
Ph3
Cavendish
Philippines,!
VI
Ph6
Cavendish
Philippines,!
VI
P79
Cavendish
Philippines/
VI
LAP
Cavendish
Philippines/
VI
SABA
Saba
Philippines/
VI
PW3
Cavendish
Philippines,m
VI

Table 2-1—continued
19
VCGU
Isolate
Cultivarv
Oriein and collector"
Lineage
PW6
Cavendish
Philippines,m
VI
PW7
Cavendish
Philippines,m
VI
0123
DAVAO
Silk
Philippines,h
VII
T1
Gros Michel
Tai wan, f
VII
PhL2
Latundan
Philippines,i
VII
Phl2
Latundan
Philippines,!
VII
9129
Latundan
Tai wan, g
VII
JLTH4
Klue namwa
Smoeng Hwy 1269, Thailand,v
X
JLTH5
Klue namwa
Smoeng Hwy 1269, Thailand,v
X*
0124
A3 6
X
Brazil,k
I
GMB
Gros Michel
Brazil,n
I
Maca
Maca
Brazil,n
I
STPA1
Pisang Awak
Burundi,i
I
STD2
Highgate
Honduras, i
I
BLUG
Bluggoe
Honduras, h
I
Sx
Tetraploid 1242
Bodies,Jamaica,i
I
FCJ2
Bluggoe
Jamaica,q
I
FCJ3x
Jamaica,q
I
FCJ8X
Jamaica,q
I
FCJ9
Tetraploid 1242
Jamaica,q
I
STJ2
Grande Naine
Jamaica, i
I
MW43
Harare
Chitipa, Karonga, Malawi,b
I
MW 4 5
Harare
Chitipa, Karonga, Malawi,b
I
MW47
Harare
Chesenga, Malawi,b
p
MW50
Harare
Chitipa,Karonga, Malawi,b
I
MW52
Sukali
Karonga South, Malawi,b
I
MW58
Harare
Karonga, Malawi, b
l
MW64
Harare
Kaporo North, Malawi,b
I
MW67
Kholobowa
Thyolo, Blantyre, Malawi, b
I
MW69
Kholobowa
Thyolo, Blantyre, Malawi,b
I
MW71
Kholobowa
Mulanje, Blantyre, Malawi,b
I
MW78
Harare
Vinthukutu, Karonga, Malawi,b
I
STN2
Bluggoe
Corinto, Nicaragua,i
I
STN5
Bluggoe
Corinto, Nicaragua,i
I
STN6
Bluggoe
Corinto, Nicaragua,i
I
STN7
Bluggoe
Corinto, Nicaragua,i
I
STPA2
Pisang Awak
Tanzania, i
I
B1
Burro(Bluggoe)
Florida,USA,b
I
0124
B2-1
Burro
Florida,USA,b
I
JCB1
Burro
Florida,USA,b
l
JLTH2
Klue namwa
Smoeng Hwy 1269, Thailand,v
I
JLTH7
Klue namwa
Smoeng Hwy 1269, Thailand,v
I
JLTH15
Klue namwa
Chai Yo Hwy, Thailand,v
I

Table 2-1—continued
20
VCGU
Isolate
Cultivarv
Orisin and collector” Lineage
0124/0125
MW9
Zambia
Kaporo, Malawi,r
I
MW11
Harare
Kaporo, Malawi,r
I
MW39
Harare
Chitipa, Karonga, Malawi,b
I
MW53
Sukali
Karonga, Malawi,b
r
MW56
Zambia
Karonga, Malawi,b
i
MW60
Zambia
Karonga, Malawi,b
i
MW61
Harare
Vinthukutu, Karonga, Malawi,b
i
MW63
Harare
Karonga, South, Malawi,b
i
MW66
Kholobowa
Thyolo, Blantyre, Malawi,b
i
MW70
Kholobowa
Thyolo, Blantyre, Malawi,b
i
MW86
Mbufu
Chitipa, Karonga, Malawi,b
i
JLTH1
Klue namwa
Ban Nok, Thailand,v
i
JLTH16
Klue namwa
Ban Nok, Thailand,v
i
TLTH17
Klue namwa
Ban Nok, Thailand,v
p
JLTH18
Klue namwa
Ban Nok, Thailand,v
p
JLTH19
Klue namwa
Ban Nok, Thailand,v
i
0125
A4
Lady Finger
Australia/
i
8606
Lady Finger
Currumbin, Queensland,g
i
8611
Lady Finger
Currumbin, Queensland,g
i
22468
Lady Finger
Tomewin, Queensland,g
i
22479
Ducasse
Bowen, Queensland, Australian
F
22600
Lady Finger
Murwillumbah, New S. Wales, Australian
I
22417
Lady Finger
Rocksberg, Queensland, Australian
I
22541
Lady Finger
Murwillumbah, New S. Wales, Australian
I
0-1223
Mons
Queensland, Australian
I
1SX
Williams
Bodies, Jamaica,i
I
STPA3
Pisang Awak
Uganda, i
I
JLTH20
Klue namwa
Ban Nok, Thailand,v
I
0126
SI
Highgate
Honduras,i
II
STA2
Highgate
Honduras, i
II
STM3
Maqueno
Honduras, i
II
STB2
Highgate
Honduras,i
II
0128
22993
Bluggoe
South Johnstone, Queensland,g
I
22994
Bluggoe
South Johnstone, Queensland, Australian
I
A47
Bluggoe
Comores Islands,j
P
0129
N5331
Cavendish
Yandina, Queensland, Australian
II
0-1221
Mons
Queensland, Australian
II
N5443
Cavendish
Doonan, Queensland, Australian
II
8627
Cavendish
North Arm, Queensland, Australian
II
22402
Cavendish
Wamuran, Queensland, Australian
II
8604
Cavendish
North Arm, Queensland, Australian
II

Table 2
-1—continued
21
VCGU
Isolate
Cultivarv
Origin and collector" Lineage
01210
A2-1
Apple
Florida, USA,b
IV
A4-1
Apple
Florida, USA,b
IV
CSB
Apple
Florida, USA,b
IV
JC14
Apple
Florida, USA,b
IV
A15
Apple
Florida, USA,b
IW
A3-1
Apple
Florida, USA,b
IV
JC8
Apple
Florida, USA,b
IV
F2
Apple
Florida, USA,b
IV
F3
Apple
Florida, USA,b
IV
JC1
Apple
Florida, USA,b
IV
GG1
Apple
Florida, USA,b
IV
01211
13721
X
x,a
DC
SH3142
SH3142
Queensland, Australia^
IX
01212
STNP1
Ney Poovan
Pemba Island, Zanzibar, Tanzania,i
VIII
STNP2
Ney Poovan
Tenguero Station, Tanzania, i
VIII
STNP4
Ney Poovan
Bukava Station, Tanzania,i
VIII
01213
1-1
Cavendish
Tai wan, u
III
2-1
Cavendish
Taiwan, u
III
6-2
Cavendish
Tai wan, u
III
5-1-1
Cavendish
Taiwan, u
III
4-2-1
Cavendish
Taiwan, u
III
4-1-1
Cavendish
Taiwan,u
III
2-2
Cavendish
Taiwan, u
III
ES2-1
Cavendish
Taiwan, u
01214
MW2
Harare
Misuku, Karonga, Malawi,r
V
MW40
Harare
Misuku Hills, Karonga, Malawi,b
V
MW41
Mbufu
Misuku Hills, Karonga, Malawi,b
V
MW42
Harare
Misuku Hills, Karonga, Malawi, b
V
MW44
Harare
Misuku Hills, Karonga, Malawi,b
V
MW46
Harare
Misuku Hills, Karonga, Malawi,b
V
MW48
Harare
Misuku Hills, Karonga, Malawi,b
V
MW51
Harare
Misuku Hills, Karonga, Malawi,b
V
MW 8 9
Harare
Misuku Hills, Karonga, Malawi,b
V
01215
CR1
Gros Michel
Isolona,Costa Rica,b
II
CR2
Gros Michel
Hamburgo,Rio Reventazón,Costa Rica,b
II
CR4
Gros Michel
Hamburgo,Rio Reventazón,Costa Rica,b
II
CR5
Gros Michel
Hamburgo,Rio Reventazón,Costa Rica,b
II

Table 2-1—continued
22
VCGU Isolate Cultivar"
Origin and collector"
Lineage
0120/01215
INDO20 Musa spp.
Jatesari, East Java,Indonesian
III
IND015Z Musa spp.
Jatesari, East Java,Indonesian
III
IND018Z Musa spp.
Jatesari, East Java,Indonesian
III
Fusarium oxysporum f. sp. lycopersici
SC548 Homestead
Bradenton, Florida
SC626 Oristano
Italy
SC761 Sunny
Bradenton, Florida
Fusarium oxysporum f. sp. niveum
0082 CS85-4
Florida,w
“Vegetative compatibility groups (VCGs) were assigned using nitrogen metabolism (nit) mutants
according to the protocols of Cove (1976) as modified by Puhalla (1985).
vCultivars are inter- and intraspecific diploid or triploid hybrids of Musa acuminata (AA) and
Musa balbisiana (BB). The ploidy levels and constitutions of the cultivars are as follows: AAA =
Gros Michel, Highgate, Mons (Mons man), Cavendish, Dwarf Cavendish, Grande Name,
Williams, Lacatan; AA = SH3142 and SH3362 (synthetic clones); AAB = Lady Finger, Pacovan,
Prata, Silk, Latundan, Maqueño; ABB - Saba, Bluggoe, Harare, Kholobowa, Pisang awak, Klue
namwa, Ducasse, Mbufu, Burro, Zambia; AB = Ney poovan, Sukali; AAAA = Tetraploid 1242
x=missing data made it impossible to determine coefficient of similarity but based on all other data
it is presumed to be in the lineage indicated.
y= unique isolate that had no lineage affinity based on defined criteria; the lineage assigned is based
on coefficients of similarity to the most closely related isolates.
*= isolates analyzed with this designation do not correspond to any isolates currently held in the
Homestead collection.
wCollector or original source: a, American Type Culture Collection; b, R.C. Ploetz, Homestead,
Florida; c, Paul E. Nelson, Fusarium Research Center, University Park, Pennsylvania; d, J.H.
Hernandez, Tenerife, Canary Islands; e, S.C. Hwang, Taiwan Banana Research Institute,
Pingtung; f, B. Manicom, Nelspruit, South Africa; g, K. Pegg, Brisbane, Australia; h, S. Nash
Smith, Alameda, California; i, R.H. Stover, La Lima, Honduras; j, IFRA, Montpellier, France (via
R.C. Stover); 1, A.M. Pedrosa, Philippines; m, N I. Roperos, Philippines; n, Z. J. M. Cordeiro,
EMBRAPA, Cruz das Amas, Brazil (via E D. Loudres); q, J. Ferguson-Conie, Banana Board,
Kingston, Jamaica; r, B. Braunworth, Oregon State University, Corvallis; u, Tsai-young Chuang,
National Taiwan University, Taipei; v, J. Leslie, Kansas State University, Manhattan; w, F.
Martin, USDA, ARS, Salinas, CA.

2
F. oxysporum f. sp. niveum were analyzed to compare results with genotypes representing
different formae speciales.
DNA Isolation
All cultures were derived from single microconidia and were stored at 4°C on
strips of Whatman filter paper (Correll et al., 1987). Paper strips were plated onto potato
dextrose agar (PDA) (Difco, Detroit, MI). After approximately 7 days of growth at 27°C,
a 5-mm3 block was excised from the margin of colonies on PDA and transferred to 4-L
Erlenmeyer flasks containing at least 100 ml of potato dextrose broth (Difco 24 g/L).
After 7 to 10 days of growth in still culture, the contents of the flask were filtered through
sterilized cheese cloth to collect mycelium. The mycelium was placed into 13-ml plastic
tubes, frozen at -80°C and lyophilized for at least 12 hours.
Buffer for DNA extraction was comprised of a 1:1:0 4 volume of the following
solutions: Buffer A (0.3 M Sorbitol, 0.1 M Tris, 20 mM EDTA at pH 7.5), Buffer B (0.2
M Tris at pH 7.5, 50 mM EDTA and 0.2 mM CTAB) and 5% Sarkosyl. Ten ml of the
extraction buffer were mixed with approximately 0.5 g of ground mycelial powder, and
the tubes were placed in a 65°C water bath for 30 minutes. The contents of the tubes
were then shaken and 1 ml of solution was transferred to each of 10, sterile, 1.5-ml
microcentrifuge tubes. Five hundred microliters of chloroforrmoctanol (24:1) solution
were added to each tube. The solution was mixed using a vortex shaker for approximately
2 minutes before centrifugation for 10 minutes at 12,000 g in a microcentrifuge at room
temperature. Unless noted, all further centrifuge steps were done in a microcentrifuge at
room temperature. The supernatant was transferred to sterile 1.5-ml tubes and treated
with 5 pi of 20 mg RNase A (Sigma Chemical Co., St. Louis, MO)/ml of solution for 30

24
minutes at 37°C. Following RNase treatment, 5 pi of 20 mg proteinase K (Sigma)/ml of
solution were added to the tubes, and the tubes were incubated for 20 minutes at 37°C.
Approximately one volume isopropanol was added, and the tubes were centrifuged for 15
minutes at 12,000 g. The isopropanol then was discarded, and 100 pi of ice-cold 70%
ethanol were added before centrifugation for five minutes. The ethanol was discarded,
and the DNA sample was air-dried for at least 30 minutes in a laminar flow hood. At least
100 pi of TE buffer (10 mM Tris, 1 mM EDTA at pH 7.4) were added to the tubes, and
they were placed in a water bath at 65°C until the DNA pellet was dissolved.
Samples which were difficult to bring into solution were subjected to a LiCl
treatment. Three hundred microliters of ice-cold 4M LiCl solution were added to each
tube, and the tubes were placed on ice for 30 minutes before centrifugation at 12,000 g for
10 minutes at 4°C. The supernatant was transferred to a sterile 1.5-ml tube containing 600
pi of isopropanol. This solution was mixed, and the tubes were kept at room temperature
for 30 minutes. After centrifugation at 12,000 g for 10 minutes at 4°C, the supernatant
was discarded and 100 pi of ice-cold 70% ethanol were added to the tubes. After
centrifugation at 12,000 g for 5 minutes, the ethanol was discarded and the DNA was air-
dried. TE buffer (100 pi) was added to the tubes and they were placed in a water bath at
65°C until the pellet was dissolved.
The concentration of DNA in the samples was estimated by running 3 pi of each
sample on an agarose gel along with DNA fragments (bacteriophage lambda DNA
digested with PstI) of a known concentration and making visual comparisons of their
relative fluorescence in the presence of ultraviolet (UV) light and 0.5 pg of ethidium
bromide/ml of solution.

25
Southern Blotting and Hybridization
Approximately 10 pg of DNA were digested with at least 10 units of either
Oral, EcoR\ or Haelll restriction enzymes (Bethesda Research Laboratories,
Gaithersburg, MD) and incubated for at least 3 hours. These restriction enzymes were
chosen based on their insensitivity to DNA methylation and their ability to digest DNA
consistently. Restriction fragments were separated by electrophoresis in 0.7% agarose for
EcoRW- and Z)ral-digested DNA or 1.5% agarose for//«elll-digested DNA in TBE
buffer at pH 7.0. Gels were run at either 30 volts for approximately 16 hours or 40 volts
for approximately 12 hours. Bacteriophage lambda DNA digested with either Pstl or
Hindlll was included on each gel and used to calculate the molecular mass of restriction
fragments obtained from F. oxyspomm f. sp. cúbense DNA. Ethidium bromide (10
mg/ml) was dissolved in the agarose gel at a concentration of 1 pi-100 ml, and the
digested DNA was illuminated by UV irradiation and photographed. The DNA was
transferred to Nytran membranes (Schleicher & Schuell, Inc., Keene, NH) using the
capillary transfer method (Sambrook et al., 1989). The DNA transfer proceeded for at
least 12 hours, and the DNA was immobilized by UV cross linking (UV 254-nm cross
linker, model FB UVXL 1000, Fischer Scientific, Pittsburgh, PA).
To reduce the incidence of repeatedly scoring similar regions of the genome or
hypervariable regions, clones containing single-copy DNA sequences, obtained from
Talma Katan (The Volcani Center, Bet Dagan, Israel), were utilized (Elias, et al., 1993).
Clones were considered to be single-copy based on the criteria defined by Elias et al.
(1993) as hybridizing to only a single DNA fragment in any of the isolates tested using at
least one of three restriction enzymes (Figure 2-1).

26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Molecular
W eight
Kb
6.5
5.1
4.2
Isolate
VCG
Polymorphic State
1
GAL2
0120
01000
2
ADJ2
0120
01000
3
01222
0120
01000
4
F9127
0120
01000
5
Pacovan
0120
01000
6
GM
0121
00100
7
SABA
0122
01000
8
FCJ9
0124
01000
9
MW60
0124/25
01000
10
CVA
0124/25
01000
11
S?
0125
01000
12
JLTH1
0124/25
01000
13
JLTH21
0125
01000
14
22994
0128
01000
15
A47
0128
01000
16
MW2
01214
10010
17
1-2
01213
10000
18
22507
0129
01000
19
SI
0126
01000
Figure 2-1: Single-copy sequences of selected Fusarium oxysporum f. sp. cúbense
isolates. DNA of isolates was digested with Eco RF restriction enzyme and probed with
clone 187. Polymorphic states indicate the presence (1) or absence (0), respectively of the
6.5, 5.1, 4.6, 4.2, or 3.4 Kb bands

27
DNA for each clone was labeled using random hexamer primers to incorporate
fluorescein-12-dUTP following the procedures provided by the manufacturer (Dupont
NEN "Renaissance," E. I. du Pont de Nemours & Co. Inc., Boston, MA). DNA labeling,
hybridization and detection followed the procedures provided by the manufacturer
(Dupont, Boston, MA). Prehybridization, hybridization and washing were performed at
65°C using a Hybaid hybridization oven (Dot Scientific Inc., Flint, MI). Membranes were
placed between acetate sheets and exposed to X-ray film for at least 5 hours.
Mitochondrial DNA of F. oxysporum f. sp. cúbense isolate 3S1 (VCG 0120) was
isolated following the procedures of Kistler and Leong (1986) and labeled as described
above. The mitochondrial DNA profiles of a subset of isolates were obtained by digesting
approximately 10 pg of total DNA with at least 10 units of the restriction enzyme Haelll
and probing with the mitochondrial DNA of isolate 3 S1. Restriction fragments were
separated by electrophoresis in 1.5% agarose in TBE buffer at pH 7.0. Gels were run at
30 volts for approximately 22 hours. Southern blotting and hybridization followed the
procedures described above.
Data Analyses
Initially, a subset of 38 geographically widespread isolates of F. oxysporum f. sp.
cúbense that represented all 17 VCGs was used to determine if a particular probe-enzyme
combination was polymorphic. Only polymorphic loci were considered informative for
phylogeny determinations. If polymorphisms were detected in the subset, then all 165
isolates were analyzed for that probe-enzyme combination. If all isolates within the subset
were monomorphic, it was assumed that the entire collection was monomorphic for that

28
probe-enzyme combination. The different restriction size fragments generated for each
combination of probe and enzyme were considered to be alleles at a single RFLP locus,
and their presence or absence was scored for each isolate. RFLP patterns for each
combination of probe and enzyme were combined to assign an RFLP haplotype to each
isolate.
The data were analyzed by a cladistic approach based on parsimony analysis using
the computer program PAUP Version 3.1.1 (Swofford, 1993), and by a phenetic approach
using distance matrix methods (UPGMA clustering, Sneath and Sokal, 1973) and the
neighbor-joining algorithm of Phylip Version 3.5c (J. Felsenstein, University of
Washington). For parsimony analysis, phylogenies were derived by using the heuristic
search option, and the degree of support was evaluated using 500 bootstrap replicates. In
addition, coefficients of similarity based on simple matching were calculated for those
isolates in which data were available for every RFLP loci scored, based on the formula
described by Sneath and Sokal (1973). Isolates were arbitrarily considered to be within
the same clonal lineage based on coefficients of similarity ranging from 0.94 to 1.00.
Since many of the isolates had identical multilocus haplotypes, only a single isolate was
used to represent each haplotype in data analyses.
To determine whether this collection of isolates provided evidence for clonal
reproduction, the gametic disequilibrium coefficient (D) was calculated among pairs of
alleles at different loci by methods described by Weir (1990). Clone-corrected allele
frequencies, using only a single representative for each haplotype, were employed for the
calculations. Also, to avoid the potential problem of repeatedly scoring similar regions,

29
only data from a single restriction enzyme digestion were used for each probe in the
analyses. Nine hundred and eighty-eight pairwise comparisons were performed to test for
disequilibrium between multiple alleles at nine loci. A test for the significance of the
disequilibrium coefficient between each pair of alleles at two loci was formulated with the
chi-square statistic
(Pu^-Pu) PA'-Pv)
where n was the number of individuals in the sample and Duv was the maximum likelihood
estimator for the coefficient of disequilibrium between alleles u and v.
The observed allele frequencies for the loci were Pu and Pv, respectively
(McDonald et al., 1994; Weir, 1990). The chi-square statistic had one degree of freedom
and the pairs of loci that showed significant departure from random expectations (P
<0.05) were considered to be in disequilibrium. A test for significance of the
disequilibrium coefficient across all alleles for each pair of loci was formulated with the
chi-square test statistic as described by McDonald et al. (1994).
The isolates tested in the study were assigned previously to a VCG, and one to
several representative isolates within each of the examined VCGs were pathogenic on at
least one and as many as eight different banana cultivars (Ploetz, personal
communication).
Although one sample of DNA was isolated from each isolate, many isolates were
tested at least once and as many as three times for each probe enzyme combination
examined. Additionally, a subset of 24 F. oxysporum f. sp. cúbense and 2 F. oxysporum f.

30
sp. lycopersici isolates tested here were separately analyzed using DNA sequencing data
(O'Donnell, personal communication). Similar phylogenetic relationships were observed
using this separate analysis, further validating the groups resolved in this analysis.
Results
In this study, 38 isolates of F. oxysporum f. sp. cúbense were screened for
polymorphisms using 19 probe-enzyme combinations. Only six of the 19 probe-enzyme
combinations were monomorphic among the 38 selected isolates, indicating a high degree
of genetic diversity among the isolates. The entire collection of 165 isolates was then
scored for polymorphisms using the 13 probe-enzyme combinations that were found to be
informative during the initial screening of isolates. A multilocus RFLP haplotype was
assigned to each isolate based on the allelic data for all probe-enzyme combinations. Only
72 distinct multilocus haplotypes were detected among the 165 isolates, 50 of which were
represented by a single isolate (Figure 2-2). The five most common haplotypes
represented 45% of the isolates. The median number of alleles per locus was three, and if
three alleles were present at each locus, theoretically 1.16 x 109 (319) possible haplotypes
could exist for this collection of F. oxysporum f. sp. cúbense. However, the majority of
single-isolate haplotypes found were the result of one to a few allelic differences from a
more common haplotype.
To determine relatedness among isolates, the 72 RFLP haplotypes were subjected
to phenetic and cladistic analyses. Both types of analyses produced trees with similar
branching patterns. The 50% majority rule bootstrap consensus tree generated by PAUP
3.1.1. is presented in Figure 2-3. A dichotomy with strong bootstrap support (99%) was

Number of Isolates
31
30 -r
25 -
Haplotype
Figure 2-2: Frequency distribution of RFLP haplotypes among the 165 isolates
representing a world collection of Fusarium oxysporum f. sp. cúbense
GG :

32
18
10
mm.
_ EUN-CS8S-4
J- FDL SC626
-|5190) _1
,FOL SC548
— EOL SC761
8(96)
É'CG C
vcc
; oi;
; oi;
¡122
,VCG 0122 Ph3
VCG 0122 LAP
0122 PW7
122 P79
VCG 0122 SABA
]
F, oxysporum
f. sp, lycopersici
and
F. oxysporum
f, sp. niveom
im
VCG 01214 MW2
VCG 01214 MW40
i— VCG 01214 MW41
4(69) 01214 MH42
VCG 0120 C2
— VCG 0120 F9127
VCG 0120 PAJ1
VCG 01214 MW48
- VCG 01214 MVJ44
VCG 01214 MW46
VCG 01214 KW51
- VCG 01214 MW89
Ji.
5(58)
VCG 0120 IC2
VCG 0120 Pacovan
â–  VCG C121 GM JFOCX
4 VCG 0121 F9130 “■
-1 VCG 0121 HI
VCG 01213 6-2
-L VCG 01213 1-2
VCG 01213 4-1-1
VCG 0121 0-1124
VCG 0120/01215 INDO20
3(90)
14(74)
LL
1 4
7(83)
-&â–  VCG 0120 MGSA1
VCG 0120 SA4
- VCG 0120 SA3
VCG 0126 STA2
VCG 0126 STB2
VCG 0126 SI
VCG 0123 DAVAO
VCG 0123 Phi2
VCG 0123 T1
VCG 0123 PhL2 —,
VCG 0123 JLTH4
- VCG 0123 JLTH5 I
]'
-Ij^vc
VCG 0129 N5443
FOC II
HE
VCG 01210 A2-1
VCG 01210 GG1
â–  VCG 01210 A15
VCG 01211 13721 —> FOC IV
VCG 01210 JC1 —* FOC IV
i VCG 01211 SH3142 1
i VCG 01213 ES2-1
13(99)
VCG 0124 A36
VCG 0124 GMB
J VCG 0124 Maca
3 i . VCG 0124 STJ2
VCG 0124 MW43
VCG 0124 MW47
— VCG 0124 JLTH15
VCG 0124 MW64
■{—— VCG 0124 S7N5
A— VCG 0124 STN7
i— VCG 0124/25 MW9
VCG 0124/25 MW11
J- VCG 0124/25 MW53
-i VCG 0124/25 MW63
— VCG 0124/25 MW86
-iy VCG 0125 8606
VCG 0125 22479
VCG 0125 22541
— VCG .0128 A47
(83) VCG 01212 STNPl
VCG 01212 STNP4
VCG 0124/25 JLTH16
VCG 0124/25 JLTH17
— VCG 0124/25 JLTH18
VCG 0124/25 JLTH19
FOC VIII
Figure 2-3: Midpoint rooted 50% majority rule consensus tree representing 500 bootstrap
replicates; One isolate represents each of the 72 RFLP haplotypes of Fusarium oxysporum
f. sp. cúbense, three isolates of F. oxysporum f. sp. lycopersici and one isolate F.
oxysporum f. sp. niveum. Branch lengths are indicated on each branch and bootstrap
values are in parenthesis. Tree length = 351; Consistency Index = 0.214; Homoplasy
Index = 0.786; Retention Index = 0.705.

33
observed on the midpoint-rooted tree among the 72 haplotypes representing the 165 F.
oxysponim f. sp. cúbense isolates and four isolates from other formae speciales. Isolates
of F. oxysponim f. sp. cúbense belonging to VCGs 0124, 0124/0125, 0125, 0128, and
01212 formed one main phylogenetic branch, while isolates belonging to VCGs 0120,
0121, 0122, 0123, 0126, 0129, 01210, 01211, 01213, 01214, 0120/01215 and 01215, as
well as the isolates of F. oxysponim f. sp. lycopersici and f. sp. niveum, were found on the
second main branch.
Isolates comprising these two branches could be further divided into eight major
clades which have moderate to strong bootstrap support (values greater than 70%).
Within one of the branches, isolates in VCGs 0124, 0124/0125, 0125, and 0128 formed
one clade. Isolates in VCG 01212 were genetically similar but distinct (83% bootstrap
support) (see below) from these isolates. Within the other branch, six clades were
identified. Isolates in VCGs 0122 (96% bootstrap support), 0123 (74% bootstrap
support), 0126 (83% bootstrap support), 01214 (97% bootstrap support), and isolates of
F. oxysponim f. sp. lycopersici (100% bootstrap support) each formed their own clade.
Isolates in VCGs 0121, 01213, and three isolates in 0120/01215 formed a clade of weak
support (58%), and therefore could not be confidently differentiated from isolates in
VCGs 0120, 0123 (two isolates) 0129, 01210, 01211, 0120/01215 and the single isolate
of F. oxysponim f. sp. niveum.
All clades with strong bootstrap support are comprised of isolates that have
identical or nearly identical multilocus haplotypes and are referred to here as clonal
lineages. Additionally, many of the isolates which could not be resolved using bootstrap

34
analysis shared nearly identical multilocus haplotypes with other isolates. To further
understand the genetic relationships among unresolved isolates, a simple matching
coefficient of similarity for comparison was used (Table 2-2). Isolates with coefficients of
similarity ranging from 0.94 to 1.00 were considered to be within a clonal lineage. This
range reflects apparent natural groups (Figure 2-3); isolates within a lineage possess either
small or no genetic differences. In cases where isolates did not fall within this range for all
pairwise comparisons, isolates were included in the lineage if they shared values near or
within the specified range with the majority of isolates comprising the lineage. These
isolates are marked with an asterisk in Table 2-1. Two isolates, ES2-1 and GM, could not
be assigned to lineage based on these criteria.
A similarity matrix, which includes coefficients of similarity for selected isolates
representing the major RFLP haplotypes and VCGs, is presented (Table 2-2). In general,
VCGs aligned with single clonal lineages; exceptions to this were isolates in VCGs 0123.
Table 2-3 lists each lineage (with the prefix FOC), the number of isolates represented, the
VCG of each lineage, and its geographical distribution. Seventy-four percent of the
isolates studied were represented by Lineages FOC I, FOC II and FOC III. Each of these
three lineages contain more than one VCG, with lineages FOC I and FOC II having a
pantropical distribution. Isolates in FOC IV through X each belong to a single VCG and
represent one to a few geographical regions.
Coefficients of similarity between isolates of FOC I and FOC II ranged from
0.66 to 0.74. By comparison, two isolates of F. oxysporum f. sp. lycopersici had
coefficients of similarity ranging from 0.55 to 0.71 compared to isolates in FOC I and 0.64

Table 2-2: Similarity matrix of simple matching coefficients based on restriction fragment length polymorphisms for selected isolates of
Fusarium oxysporum f. sp. cúbense, F. oxysporum f. sp. lycopersici and F. oxysporum f. sp. niveum.
r-~
CN|
cr>
S
o
CD
O
>
S
cd
CNl
O
CD
O
>
o
CO
CD
U_
CN|
O
CD
O
>
x
OJ
CD
CD
O
>
s
c/3
CNJ
CNJ
CD
CD
O
>
VCG 0123 DAVAO
CO
CNl
5
CD
O
>
CNl
xz
CL
oo
CN|
5
CD
O
>
CN|
_J
xz
CL
CO
CN|
CD
CD
O
>
■«r
X
t—
I
CO
CNl
5
CD
O
>
CO
NT
$
2
CNl
CD
CD
O
>
VCG 0124 MW64
VCG 0124/0125 MW11
CD
CD
CD
co
iD
CNJ
CD
CD
O
>
CN|
<
c/5
CD
CN|
CD
CD
O
>
oo
CD
CNl
CD
CD
o
>
CO
NT
LO
X
CD
CNl
CD
CD
o
>
CNl
<
CD
CNl
CD
CD
O
>
CNl
â– *3-
co
X
CD
CNl
5
CD
O
>
VCG 01212 STNP4
CNl
CO
CNj
CD
CD
O
>
CNl
$
CN
5
CD
O
>
CNJ
5
CD
O
>
VCG 01214 MW51
CD
CO
$
2
Nj-
CNJ
5
CD
O
>
VCG 0120/01215 Indo 20
o
£
8.
1
CD
CN|
8
CD
o
CO
8.
1
5
o
CD
£
D
2
c.
NT
UD
8
o
VCG 0120 9127
1
VCG0120IC2
0.98
1
VCG 0121 F9130
0.82
0.85
1
VCG 0121 HI
0.86
0.88
0.96
1
VCG 0122 SABA
0.91
088
0.82
0.86
1
VCG 0123 DAVAO
0.78
0.75
0.74
0.75
0.75
1
VCG 0123 T1
0 8
0.78
0.74
0.78
0.78
0.93
1
VCG 0123 Phi2
0.76
0.76
0.73
0.76
0.74
0.94
0.92
1
VCG 0123 PhL2
0.81
0.79
0.75
0.79
0.79
0.94
0.99
0.93
1
VCG 0123 JLTH4
0.78
0.75
0.72
0.75
0.8
0.88
0.93
0.87
0.94
1
VCG 0124MW43
0.73
0.71
0.62
0.66
0.75
0.74
0.79
0.73
0.8
0.81
1
VCG 0124 MW64
0.72
0.69
0.61
0.65
0.74
0.75
0.78
0.74
0.79
0.8
0.99
1
VCG 0124/0125 MW 11
0.74
0.72
0.64
0.67
0 76
0 73
0.78
0.72
0.79
0.85
0 96
0.95
1
VCG 0125 8606
0.71
0.68
0.6
0.64
0.73
0.74
0.76
0.73
0.78
0.79
0.98
0.99
0.94
1
VCG 0126 STA2
0.96
0.96
0.84
085
0.87
0 76
0.79
0.75
0.8
0.76
0.72
071
0.73
0.69
1
VCG 0126 SI
0.94
0.94
0.81
0.82
0.85
0.74
0.79
0.75
0.8
0.76
0.72
0.71
0.73
0.69
0.98
1
VCG 0129 N5443
0.96
0.99
0.86
0.87
0.87
0.74
0.76
0.75
0.78
0.74
0.69
0.68
0.71
0.67
0.95
0.93
1
VCG 01210 A2-1
0.89
092
0.91
0.92
0.87
081
0.84
0.82
0.85
0.81
0.72
071
0.73
0.69
0.93
0.91
0.91
1
VCG 01211 SH3142
0.91
0.93
0.85
0.86
0.86
0.75
0.75
0.74
0.76
0.78
0.71
0.69
0.69
0.68
0.89
0.87
0.92
0.89
1
VCG 01212 STNP4
0.73
0.71
0.6
0.64
0.75
0.72
0.76
0 71
0.78
0.79
0.95
0.94
0.92
0.93
0.72
0.72
0.69
0.69
0.71
1
VCG 01213 1-2
0.86
0.88
0.94
0.98
0.84
0.75
0.8
0.79
0.81
0.78
0.68
0.67
0.69
0.66
0.85
0.85
0.87
0.92
0.86
0.66
1
VCG 01214 MW2
0.76
0.74
0.75
0 74
0 79
0.75
0.8
0.74
0.79
0.8
0.68
0.67
0 69
0 66
0.78
0.75
0.75
0.8
0.76
0.68
0.74
1
VCG 01214 MW4I
0.76
0.74
0.78
0.74
0.79
0 78
0.8
0.74
0.79
0.8
0.68
0.67
0.69
0.66
0.78
0.75
0.75
0.8
0.79
0.71
0.74
0.95
1
VCG 01214MW51
0.79
0.76
0.75
0 74
0.79
0.75
0.8
0.74
0.79
0.8
0.68
0.67
0.69
0.66
0.8
0.78
0.78
0.8
0.79
0.68
0.74
0.98
0.95
1
VCG 01214 MW89
0.8
0.78
0.74
0.78
0.82
0.79
0.84
0.78
0.82
0.84
0.72
0.71
0.73
0.69
0.79
0.76
0.76
0.81
0.8
0.72
0.78
0.94
0.94
0.94
1
VCG 0120/01215 Indo20
0.87
0.89
0.95
0.99
0.85
0.76
0.79
0.78
0.8
0.76
0.67
0.66
0.68
0.65
0.86
0.84
0.88
0.93
0.87
0.65
0.99
0.75
0.75
0.75
0.79
1
SC626 lycopersici
0.71
0.73
0.76
0.75
0.75
0.69
0.72
068
0.73
0.74
0.69
0.68
0.71
0.67
0.72
0.72
0.74
0.76
0.75
0.67
0.75
0.66
0.68
0.66
0 67
0.74
1
SC 761 lycopersici
0.64
0.66
0.72
0 71
0.71
0.6
0.62
064
0.64
0.65
0.6
0.59
0.61
0.58
0.65
0.65
0.67
0.69
0.68
0.62
0.71
0.64
0.66
0.61
0.62
0.69
0.88
1
CS85-4 niveum
0.74
0.74
0.75
0 76
0.76
08
0.85
0.79
0.86
0.85
0.8
0.79
0.79
0 78
0.75
0.75
0.75
0.8
0.76
0.78
0.79
0.81
0.79
0.81
0.8
0.78
0.75
0.71
1

Table 2-3: Clonal lineages of Fusarium oxysporum f. sp. cúbense isolates, their
geographical distributions and corresponding vegetative compatibility groups (VCG).
36
No. of VCG
Linease
Isolates2
Geosraphic Distribution
Represented
FOC I
65
Australia, Brazil, Burundi,
Comores Islands Honduras
Jamaica, Malawi, Nicaragua,
Tanzania, Thailand, Uganda,
and the United States (Florida)
0124,0124/25
0125, and
0128
FOC II
43
Australia, Brazil, Canary Islands,
Costa Rica, Honduras,
Jamaica, Malaysia, South Africa,
and Taiwan
0120, 0126,
0129, and
01215
FOC III
15
Indonesia and Taiwan
0121, 01213,
and
0120/01215
FOC IV
11
United States (Florida)
01210
FOC V
9
Malawi
01214
FOC VI
8
Philippines
0122
FOC VII
5
Philippines, Taiwan
0123
FOC VIII
3
Tanzania
01212
FOC IX
2
Australia
01211
FOC X
2
Thailand
0123
Tsolates ES2-1 in VCG 01213 and GM in VCG 0121 did not align with any lineage based
on coefficient of similarity data

37
to 0.74 compared to isolates in FOC II. The single isolate of F. oxysporum f. sp. niveum
had coefficients of similarity ranging from 0.75 to 0.81 compared to isolates in FOC I and
from 0.73 to 0.76 compared to isolates in FOC II. Thus, the two largest lineages of F.
oxysporum f. sp. cúbense each are more genetically similar to the F. oxysporum f. sp.
niveum isolate than to each other. Similarly, they are roughly as genetically distinct from
each other as either is to the F. oxysporum f. sp. lycopersici isolates.
All isolates in VCG 01214 comprise FOC V. Isolates in this lineage formed a
clade exhibiting the longest branch length compared to all other clades representing F.
oxysporum f. sp. cúbense isolates. In fact, its length was comparable to the branch lengths
of the clades containing the isolates from the other formae speciales (Figure 2-3).
Additionally, isolates comprising this lineage had coefficients of similarity which did not
align closely to any other lineage. For example, isolates in FOC V had coefficients of
similarity ranging from 0.62 to 0.73 compared to isolates in FOC I, from 0.72 to 0.81
compared to isolates in FOC II and from 0.61 to 0.69 compared to the F. oxysporum f. sp.
lycopersici isolates. This group had a large number of lineage-specific alleles at several
RFLP loci which accounted for its relative lack of similarity to other F. oxysporum f. sp.
cúbense isolates. All of the isolates in VCG 01214 have a very limited geographical
distribution. Isolates in this group also do not form chlamydospores (R.C. Ploetz,
unpublished) as other F. oxysporum isolates do, which is usually a defining trait for the
species.
FOC VII consisted of the majority of isolates in VCG 0123. FOC X consisted of
two additional isolates in VCG 0123. Similar to FOC V, isolates comprising these
lineages had coefficients of similarity which did not align closely to any other lineage.

38
Surprisingly, the RFLP multilocus haplotypes of isolates within these lineages consisted of
alleles similar to those in both FOC I and II, in addition to some lineage-specific alleles
(Table 2-4). A representative FOC VII isolate (PhL2) shared 43% (6/14) of polymorphic
alleles with isolates in FOC II and 36% (5/14) of these alleles with isolates in FOC I. By
contrast, the isolates in FOC I displayed only 29% (4/14) allelic similarity with isolates in
FOC II. With the exception of FOC V, isolates in the other lineages more closely aligned
with isolates in either FOC I or II. Unlike isolates in other lineages, the multilocus
haplotype of isolates in FOC VII and X appear to represent a combination of alleles from
FOC I and II.
The mitochondrial DNA haplotypes of a subset of 55 isolates representing the
major VCGs were examined to determine their genetic relationships based on an additional
independent genetic marker. Isolates within the same VCG shared identical mitochondrial
DNA haplotypes. Isolates could be divided into three major groups based on visual
assessment of similar though not identical, mitochondrial RFLP patterns (Figure 2-4).
Isolates in VCGs 0123, 0124, 0124/0125, 0125, 0128, 01212, and 01214 formed one
group. Isolates in VCGs 0120, 0121, 0122, 0129, 01213, 0120/01215 and 01215 formed
a second group. Isolates in VCGs 0126 and 01210 formed a third group. In general,
these groupings aligned with those based on RFLP analysis of single-copy loci, although
more groups could be resolved using the latter method.
Measures of gametic disequilibrium were performed for alleles at nine loci (Table
2-5). Even though many of the individual comparisons were not significant, 34 of the 36
pairwise comparisons among alleles at different RFLP loci showed significant

39
Table 2-4: A comparison of allelic data for isolates in FOC I, FOC II and FOC VII.
Isolate (VCG)y Alleles2
A36 (VCG 0124)
IC2 (VCG 0120)
PhL2 (VCG 0123)
11111111111111
22122122221222
12132123331113
'Isolates A36 and IC2 represent the two largest RFLP haplotypes in FOC I and FOC II,
respectively, and PhL2 represents an isolate in FOC VII.
zAlleles found in the most common haplotypes were given the number 1. Alleles in the
second most common haplotype, if different from haplotype 1, were given the number 2.
Alleles in isolate PhL2, if different than those in the two most common haplotypes were
given the number 3.

Molecular Weight
Kb
23.1
Isolate
VCG
mtDNA Group
1
ORTZ
0120
2
2
NW
0120
2
3
Ph3
0122
2
4
DAVAO
0123
1
5
S?
0124
1
6
MW60
0124/25
1
7
8606
0125
1
8
STNP4
01212
1
9
22993
0128
- 1
10
DAVO
0123
1
11
STM3
0126
3
12
N5443
0129
2
13
TBR
0121
2
14
ES2-1
01213
2
15
JLTH7
0124
1
16
SC548
F. oxysporum
f. sp.
lycopersici
other
Figure 2-4: Mitochondrial DNA haplotypes of selected Fusarium oxysporum f.
cúbense isolates.

41
Table 2-5: Clone-corrected measurements of gametic disequilibrium among pairs of alleles
in a world-wide collection of Fusarium oxysporum f.sp. cúbense.
Probe / Enzyme Combinations
120
Hae IE
162
Dm I
177
EcoRV
187
EcoRV
204
Hae in
228
EcoRV
260
EcoRV
261
EcoRV
7
Dm I
13/35
77.7 (24)
4/10
25.1 (5)
4/18
56.8(10)
5/30
174.4
(20)
5/15
87.4(8)
7/30
50.0 (24)
5/20
183.1
(12)
22/40
86.6
(28)
120
HaeUl
8/14
31.33 (6)
17/21
68.3(12)
9/42
122.6
(30)
11/21
142.4
02)
9/42
121.1
(30)
13/28
87.5(18)
14/56
184.3
(42)
162
Dm I
4/6
4.91 (2)*
8/12
37.7
(5)
4/6
5.49(2)*
5/12
128.7(5)
5/8
29.3
(3)
6/16
23.3
(7)
177
EcoRV
5/18
33.0(10)
4/9
73.09
(4)
7/18
78.1 (10)
2/12
37.7(6)
9/24
54.6(14)
187
EcoRV
6/18
40.5
(10)
17/36
126.2
(25)
6/24
63.54
(15)
6/48
65.2 (35)
204
Hae DI
7/18
51.3 (10)
6/12
48.2 (6)
11/24
85.8(14)
228
EcoRV
8/24
53.7(15)
7/48
57.2 (35)
260
EcoRV
16/32
190.6
(21)
Note:
All tests had at least 68 multilocus haplotypes in the comparisons and a maximum of n=72.
The first row in each box represents the number of significant chi-square tests (P<0.05)
between individual alleles at different RFLP loci over the total number of tests made for
each pairwise comparison; the second row in each box represents the results of a chi-
square test for the significance of association between all alleles at the two loci. The
number in the parenthesis is the degrees of freedom for the test. All are significant at P<
0.05 and P<0.01 except those noted with *.

42
nonrandom associations at the 1% level. Many individual allele combinations were not
present in the population, but the chi-square test for these combinations was not
significant; based on the overall allelic frequency, the expected number of these
combinations was small. In contrast, for the most common alleles the large number of
nonrandom associations was indicated by significantly larger or smaller numbers of
observations when compared to the expected frequency of these combinations.
Discussion
Several lines of evidence support the concept that F. oxyspontm f. sp. cúbense has
a clonal population structure in line with criteria established by Tibayrenc et al. (1990). A
unifying feature of clonally reproducing organisms is widespread geographic distribution
of a few successful clones. Even though this study identified 72 multilocus haplotypes in a
worldwide collection of F. oxysporum f. sp. cúbense, the five most common haplotypes
accounted for nearly half of the isolates. Additionally, the two most common haplotypes
were found on all five continents sampled in this study, indicating the pantropical
distribution of a small number of genotypes.
Further evidence of clonal reproduction is the absence of recombinant genotypes.
Significant gametic disequilibrium for alleles at 34 of 36 loci tested supported nonrandom
association between alleles of different loci. In addition, the strong correlation between
independent genetic markers (VCG, mitochondrial and multilocus RFLP haplotype) also
are indicative of a clonally reproducing organism (Tibayrenc et al., 1990).
This study confirms that in phylogenetic analysis of F. oxysporum f. sp. cúbense
using parsimony, VCG is a strong predictor of cladistic groupings. Further differentiation
into lineages may be done based on coefficients of similarity of RFLP haplotypes. At
least 17 VCGs have been described for F. oxysporum f. sp. cúbense (Ploetz and Correll,

43
1988; Ploetz, 1990; Ploetz et al., 1992; Ploetz, 1994; Ploetz et al., 1997), and
representatives from each of these VCGs were included in this study. With the exception
of two isolates in VCG 0123, all isolates within a VCG were in the same clade and clonal
lineage. The correlation between VCG and RFLP patterns has been observed previously
in F. oxysporum formae speciales, including albedinis, conglutincins, dianthi, gladioli
lycopersici, melonis, pisi, raphani, and vasitfectum (Elias and Schneider, 1991; Elias et
al., 1993; Kim et al., 1992; Kistler et al., 1987; Manicom et al., 1990, Manicom and
Baayen, 1993; Mes et al., 1994; Namiki et al., 1994; Tantaoui et al., 1996; Whitehead et
al., 1992).
Although VCGs are good indicators of genetic similarity among the individuals
comprising them, they do not provide information regarding the genetic similarity to
individuals in different VCGs. In fact, this study shows that isolates belonging to different
VCGs could have identical or nearly identical RFLP haplotypes. With two exceptions, the
entire collection of isolates consisted of only ten distinct clonal lineages (Table 2-2).
Clonal lineages provide a conservative system for grouping similar isolates, and the
coefficient of similarity provides a numerical value to assess genetic relationships among
isolates representing different lineages. This is in contrast to VCG groupings where, for
an asexually reproducing organism such as F. oxysporum, it is impossible to determine the
quantitative differences among individuals in different VCGs. Thus, the use of clonal
lineages rather than VCG is proposed to genetically characterize similar isolates of F.
oxysporum f. sp. cúbense. In most instances, VCGs can be used to predict lineage.
Many of the isolates used in this study also have been classified based on their
electrophoretic karyotype (Boehm et al., 1994) and RAPD profile (Bentley et al., 1995).
Based on their electrophoretic karyotype, Boehm et al. (1994) divided 118 isolates into

44
two major groups. Group I contained isolates from VCGs 0124, 0124/0125, 0125, 01210
and 01214 and was characterized by high chromosome number and large relative genome
size [39.9-58.9 megabase pairs (Mbp)]. Group II contained isolates from VCGs 0120,
0121, 0122, 0123, 0129 and 01213, which had correspondingly fewer chromosomes and
smaller genome sizes (32.1-44.9 Mbp). Using RAPD analysis, Bentley et al. (1995)
similarly found that 54 isolates, representing 11 VCGs, could be divided into two major
groups. Group I contained isolates in VCGs 0120, 0121, 0122, 0126, 01210, 01211 and
01212 while group II contained isolates in VCGs 0123, 0124, 0124/0125 and 0125.
Cluster analysis indicated that VCG 01212 was distinct from the other VCGs in group I
and 0123 was distinct from group II.
Although the results presented here corroborate most of the broad conclusions
made previously, this study provides additional and sometimes disparate conclusions
regarding the affinities of some of these isolates. The bootstrap 50% majority rule
consensus tree showed strong support for more than two clades among isolates of F.
oxysporum f. sp. cúbeme. The midpoint rooted tree divides isolates into two major
groups. One group is comprised of isolates in five VCGs, which represent two significant
clades. Isolates in this major group are remarkably homogenous, and the branch lengths
that separate isolates are minimal. In contrast, the second branch encompasses isolates
representing eight lineages, 11 VCGs, a large number of significant clades, as well as
isolates belonging to other formae speciales of F. oxysporum. Isolates representing the
second group had more variable branch lengths compared to isolates in the first major
branch.
With the exception of VCGs 0122, 0123, 0126, 01210, 01212, and 01214, the
relationship of isolates in 10 of the 17 VCGs correspond to those defined by previous

45
studies (Bentley et al., 1995; Boehm et al., 1994). Unlike previous investigations, isolates
in VCGs 0122, 0126, 01212 and 01214 each formed individual clades with bootstrap
values greater than 70%. Additionally, because this study used numerous, independent
clones, had a large sample size, and provides bootstrap support for the clades, it gives a
greater resolution of the genetic relationships among isolates of F. oxysporum f. sp.
cúbense than do previous studies.
A number of the clonal lineages described here are phylogenetically distinct. Some
isolates of F. oxysporum f sp. cúbense are as genetically dissimilar to one another as they
are to other formae speciales of F. oxysporum (niveum and lycopersici). One
interpretation of these results is that isolates belonging to the dissimilar groups acquired
their ability to be pathogenic on bananas independently.
FOC V contains isolates from the Misuku Hills in Malawi, a relatively small area
(approximately 400 square kilometers) on the country’s northern border with Tanzania
(Ploetz et al., 1992). All isolates in FOC V are in VCG 01214, and this is one of the few
VCGs which has not been found in Southeast Asia, the center of origin of banana. Due to
numerous lineage-specific alleles, FOC V is distant from all other lineages. One
hypothesis is that this lineage of F, oxysporum f. sp. cúbense may have evolved
independently of other members of the taxon in East Africa.
Alternatively, isolates within FOC V could have arisen by a founder effect.
Bananas probably first arrived at the island Madagascar in the later half of the first
millennium A.D. and from there moved to the coastal and then interior regions of the
African continent (Ploetz et al., 1992). Diverse genotypes of banana are now found in
East Africa, many of which are found nowhere else, it is possible that the pathogen was
moved from Southeast Asia on the bananas introduced to Africa, and, as a result of

46
mutation and selection or through adaptation to endemic bananas, isolates in VCG 01214
may have diverged from their Asian progenitors.
FOC VII and X contain isolates in VCG 0123. The RFLP haplotype of isolates
belonging to these groups carry an assortment of alleles from the two major lineages (FOC
I and II) as well as a number of lineage-specific alleles. Additionally, isolates comprising
this clade are quite heterogenous; the five isolates comprising FOC VII belong to two
significant sister groups. Also, all of the isolates in VCG 0123 fall neither into the same
lineage nor in a single clade. Based on this information, this group may provide evidence
of an ancient genetic exchange between individuals in FOC I and II. Alternatively, it may
represent an ancestral group possessing primitive character states found in FOC I and
FOC II.
In conclusion, isolates of F. oxysporum f. sp. cúbense represent a genetically
diverse group of organisms, many of which are distantly related. Previous studies on
other formae speciales indicate that many are genetically diverse. However, this study is
the first based pm RFLP data of nuclear DNA to present evidence that a forma specialis of
F. oxysporum may be polyphyletic. The implied independent origin of pathogenicity to
banana in some of the lineages has practical implications for work on this disease. Much
effort is devoted towards developing cultivars of banana which are resistant to Fusarium
wilt. Clearly, new hybrids should be screened against isolates representing the two most
common lineages of the pathogen (FOC I and II). Ideally, breeding programs could
screen new hybrids against isolates from each clonal lineage to increase the probability of
developing cultivars that resist genetically distinct populations of the pathogen.

CHAPTER 3
GENETIC VARIATION IN TWO HONDURAN FIELD POPULATIONS OF
FUSARIUM OXYSPORUM F. SP. CUBENSE
Introduction
Fusarium oxysporum Schlechtend, Fr. forma specialis cúbense (E.F. Sm.) W.C.
Synder & H. N. Hans., is the causal agent of Fusarium wilt of banana, a disease which has
been responsible for significant yield losses in banana. This disease had a devastating
effect on the export banana trade in Central America during the early part of this century
(Stover, 1962).
Bananas were introduced to Honduras well over 200 years ago and commercial
trade began about 1876. Until the late 1950s, the race 1 susceptible banana cultivar Gros
Michel was the most widely planted commercial cultivar in Honduras and elsewhere
(Stover, 1962). However, disease epidemics in many commercial banana plantations
were exacerbated by an increased prevalence of race 1 isolates due to the perennial
monoculture production of this susceptible, clonally derived cultivar. Eventually,
Cavendish cultivars, which are resistant to race 1 and 2, were used to replace Gros Michel
for export production.
Three races of the pathogen that are pathogenic on banana have been reported in
prior work. Only two races of the pathogen have been identified in Honduras (Stover,
personal communication), and these isolates belong to only three vegetative compatibility
groups (VCGs) (0120, 0124 and 0126) (Ploetz, 1990). Race 1 isolates occur in all three
47

48
of these VCGs, while race 2 isolates appear to be limited to VCG 0124. In a number of
subtropical countries (Australia, South Africa, Canary Islands and Taiwan), many
commercial plantations of Cavendish cultivars succumbed to Fusarium wilt shortly after
they were planted (Ploetz, 1990). More recently, isolates in VCGs 01213 and 01216 have
caused disease on Cavendish cultivars in the tropics. Isolates capable of causing disease
on Cavendish cultivars were identified by Su et al. (1977) as race 4, a new race of the
pathogen. Race 4 isolates also are pathogenic on all cultivars susceptible to race 1 and
race 2 but presently have a more limited geographical distribution than races 1 and 2.
There have been no reports of race 4 in Central America.
Fusarium wilt has been a problem in most major banana-producing regions of the
world and the occurrence of the pathogen in areas where bananas are not indigenous is
most likely a consequence of human dissemination of infested rhizomes. Bananas are
rhizomatous, perennial plants that are grown in subtropical and tropical climates.
Cultivated bananas are commonly sterile diploid, triploid, or tetraploid clones derived
from Musa acuminata or interspecific hybridization between M. acuminata and M.
balbisiana (Simmonds, 1962). Since most edible bananas do not produce viable seed,
they must be asexually propagated, typically using rhizomes or sucker plants. As such,
they are usually also accompanied by soil and associated microorganisms, it is through the
movement of infested plant material that the pathogen has been introduced to nonendemic
regions. For example, isolates that have been introduced to the Americas on infested plant
material comprise only a limited number of VCGs and clonal lineages compared to those
found in southeast Asia, the center of origin of M. acuminata (Ploetz and Correll, 1988;
Chapter 2).

49
A study of a world collection of F. oxysporum f. sp. cúbense isolates representing
17 vegetative compatibilty groups (VCGs) revealed that this pathogen is comprised of at
least 10 clonal lineages (Chapter 2). The greatest diversity of VCGs and lineages are
present in Malesia (Ploetz, 1994; Chapter 2), which is also the region of origin of the host
(Simmonds, 1962).
Although the level of genotypic diversity that was represented in a worldwide
collection of F. oxysporum f. sp. cúbense isolates has been described previously (Bentley
et al, 1995; Chapter 2), the amount of diversity found in individual field populations of the
pathogen is unknown. Similarly, factors that may affect the levels of diversity within a
field have not been explored forF. oxysporum f. sp. cúbense. McDonald et al. (1995)
discussed several factors which may cause pathogen populations to undergo significant
changes in genetic structure over time. For a perennial, tropical crop such as banana,
cropping practices such as rotation, field burning, and fungicide application may affect the
pathogen population. However, the use of host disease resistant genotypes also likely
plays a major role. Similarly, the reproductive strategy and mutational rate of the
pathogen, as well as the frequency of immigration of new genotypes, may also have a
profound effect on the genetic composition of a field population.
The objectives of this study were to: i) determine the amount of genotypic
diversity in field populations of the pathogen in Honduras; and ii) determine whether
specific genotypes of the pathogen preferentially infect specific host genotypes.

50
Materials and Methods
Description of Sampling Procedure
Sampling of F. oxysporum f. sp. cúbense isolates was conducted in two adjacent
disease screening fields (Fields 1 and 2) at the banana breeding station operated by the
Fundación Hondureña de Investigación Agrícola (FHIA) in La Lima, Honduras.
Historically, Field 1 had been planted to cultivar Highgate, which is susceptible to race 1,
and had a high incidence of plants with Fusarium wilt based on visual assessment of
symptoms. Therefore, it was an optimal location to evaluate the resistance to race 1 in
seven of FHIA’s tetraploid hybrids (Rivera, pers. comm.). Prior to planting the trial,
inoculum levels were artificially increased in the field using a number of procedures. First,
freshly infected pseudostem and rhizome tissue from Highgate plants infected with
F. oxysporum f. sp. cúbense was uniformly spread throughout the field. Subsequently, the
entire field was planted to the race I-susceptible cultivar Maqueño. Prior to planting the
Maqueño rhizomes, approximately 250-gram (g) chunks of symptomatic pseudostem and
rhizome tissue from Highgate were placed in the bottom of each hole. After a year, 94%
of the resulting plants exhibited internal symptoms of Fusarium wilt (vascular discoloration
of rhizomes and pseudostems). Prior to planting the disease screening trial, infected
Maqueño plants were incorporated into the soil. Additionally, prior to planting each
experimental hybrid, 250 g of symptomatic Maqueño tissues were placed in each planting
hole.
The experiment was arranged in a randomized complete block design with nine
treatments (cultivars) and seven replications (Table 3-1). The banana plants used in the
experiments were derived from tissue culture. The treatments were planted in the same

Table 3-1. Genotypes of banana planted in Field 1
Hybrid Genotype
FHIA 1 AAB x AA = AAAB
Dwarf Prata x 3142 (Pisang Jari Buaya derivative)
FHIA 2: AAA x A A = AAAA
Williams*x 3397 (3142 (AA) x 3217 (AA); both are Pisang Jari Buaya derivatives)
FHIA 3 ABB x BB = ABBB; ABBB x AA = AABB; AAB or ABB x AA = AAA,
AAB, or ABB
Cardaba x BB = ABBB x 2741 (AA) = 3386 [selected triploid x 3320 (AA)]
FHIA 6: AAB X AA = AAAB
Maqueño x 3437 (Black sigatoka resistant Mura acuminata subsp. burimanica)
FHIA 15: AAB x AA = AAAB
Same parentage as FHIA 6, but this selection is faster to ratoon
FHIA 17: AAA xAA = AAAA
Highgate x 3362 (3142 x 3217)
FHIA 23: AAA x AA = AAAA
Same parentage as FHIA 17 but different selection
Highgate: AAA (race 1 susceptible control (race 2 resistant))
Williams: AAA ( race 1 and race 2 resistant control)
Bluggoe: ABB (race 2 susceptible control (race 1 resistant))
Note: With the exception of FHIA 3, all FHIA hybrids are either interspecific or
intraspecific tetraploids obtained by performing crosses on Musa acuminata (A A) and M.
balbisiana (BB) parental lines. Highgate, Williams and Bluggoe are banana cultivars used
as host differentials to differentiate among the three races of Fusarium oxysporum f. sp.
cúbense.
* It is questionable as to whether or not Williams was actually the parent of this cross, due
to the cultivar Williams extreme sterility.

52
rows in which the cultivar Maqueño had been planted Each plot contained three plants
spaced at 2.5 m, and the entire experimental area encompassed approximately 6500 m2.
The field consisted of seven rows; three of these contained the seven replications of the
experiment planted in a serpentine fashion throughout the plot. Experimental rows were
separated from one another by a single row of the race 2-susceptible cultivar Bluggoe
(Table 3-2).
The first samples were collected in February 1995 using a nondestructive method,
which consisted of removing the first few outer layers of pseudostem tissue from
symptomatic and asymptomatic plants (Ploetz, 1992). A second set of samples were
collected from the same planting in September 1995 using a destructive sampling method,
which consisted of excising inner tissue of both the rhizome and pseudostem. When
collecting symptomatic tissue, attempts were made to obtain samples displaying necrotic
vascular strands. Since sampling occurred during two periods, some plants were sampled
twice. Samples were obtained from every plant showing visible disease symptoms in all
treatments and also from the symptomatic race 2 susceptible Bluggoe plants in the
adjacent rows. Approximately one of every 12 asymptomatic plants in the entire field also
were sampled to determine incidence of infection on these plants.
Field 2 was planted to the cultivar Bluggoe in an effort to increase inoculum levels
for a future field trial to evaluate hybrids for resistance to race 2. The field was
approximately 3500 m2 and consisted of eight rows (Table 3-3). Seventy-six percent of
the plants showed visible symptoms of Fusarium wilt during an initial assessment of the
field. Symptomatic plants were fairly evenly distributed throughout the field, and, based
on this distribution, a systematic sampling strategy was followed in which one of every
three plants was sampled following a serpentine pattern through the plot. Samples were

53
Table 3-2. Experimental design of banana plants in rows of Field 1.
Row Number-'
1
2
3
4
5
6
7
B
F15
B
F15
B
F2
B
B
F15
B
F15
B
F2
B
B
F15
B
F15
B
F2
B
B
F6
B
F3
B
FI
B
B
F6
B
F3
B
FI
B
B
F6
B
F3
B
FI
B
B
F23
B
W
B
F6
B
B
F23
B
W
B
F6
B
B
F23
B
W
B
F6
B
B
F17
B
H
B
F6
B
B
F17
B
H
B
F6
B
B
F17
B
H
B
F6
B
B
FI
B
F17
B
F17
B
B
FI
B
F17
B
F17
B
B
FI
B
F17
B
F17
B
B
H
B
F23
B
F2
B
B
H
B
F23
B
F2
B
B
Hz
B
F23
B
F2
B
B
W
B
F6
B
F15
B
B
W
B
F6
B
F15
B
B
w
B
F6
B
F15
B
B
F2
B
Hz
B
H
B
B
F2
B
Hz
B
H
B
B
F2
B
H
B
H
B
B
F3
B
F3
B
F23
B
B
F3
B
F3
B
F23
B
B
F3
B
F3
B
F23
B
B
W
B
F17
B
F3
B
B
W
B
F17
B
F3
B
B
W
B
F17
B
F3
B
B
F17
B
F23
B
W
B
B
F17
B
F23
B
W
B
B
F17
B
F23
B
W
B
B
F15
B
F15
B
FI
B
B
F15
B
F15
B
FI
B
B
F15
B
F15
B
FI
B
B
F6
B
W
B
F2
B
B
F6
B
W
B
F2
B
B
F6
B
w
B
F2
B
B
F2
B
F2
B
W
B
B
F2
B
F2
B
W
B

54
Table 3-2—continued.
Row Numbery
1
2
3
4
5
6
7
B
F2
B
F2
B
W
B
B
FI
B
FI
B
F6
B
B
FI
B
FI
B
F6
B
B
FI
B
FI
B
F6
B
B
F3
B
F2
B
H
B
B
F3
B
F2
B
H
B
B
F3
B
F2
B
H
B
B
FT
B
H
B
F17
B
B
Hz
B
H
B
F17
B
B
H
B
H
B
F17
B
B
F23
B
FI
B
FI
B
B
F23
Bz
FI
B
FI
B
B
F23
B
FI
B
FI
B
B
F23
B
F6
B
F15
B
B
F23
B
F6Z
B
F15
B
B
F23
B
F6
B
F15z
B
B
F17
B
F3
B
F3
B
B
F17
B
F3
B
F3
B
B
F17
B
F3
B
F3
B
B
W
B
F15
B
F23
B
B
W
B
F15
B
F23
B
B
W
B
F15
B
F23
B
y Each letter or letter/numeral combination represents a single plant; bold type indicates
plants that were sampled for the presence of Fusarium oxyspomm f. sp. cúbense; B=
cultivar Bluggoe, H= cultivar Highgate, W= cultivar Williams, F1=FHIA 1, F2-FHIA 2
F3=FHIA 3, F6=FHIA 6, F15=FHIA 15, F17=FHIA 17, F23=FHIA 23
z indicates multiple samples were obtained from these plants

55
Table 3-3. Field design of banana plants in rows of the race 2 disease screening plot.
Plant Number
Row Number2
1
2
3
4
5
6
7
8
1
D
D
D
D
D
D
D
D
9
D
D
D
D
D
D
D
H
3
H
D
D
D
D
D
D
D
4
D
D
D
D
D
D
D
D
5
D
D
D
D
D
D
D
D
6
H
D
D
D
D
D
D
D
7
H
D
D
D
D
D
D
D
8
H
D
D
D
D
D
D
D
9
D
D
D
D
D
D
-
H
10
H
D
D
D
D
D
D
H
11
H
D
H
D
D
D
H
H
12
H
D
D
D
H
H
D
D
13
D
H
D
H
D
H
D
D
14
H
H
D
D
D
H
D
H
15
H
H
D
D
D
D
D
D
16
D
H
D
D
D
D
H
H
17
D
H
D
D
D
D
D
D
18
D
H
D
D
H
D
D
D
19
H
H
D
D
D
D
D
D
20
H
D
D
H
D
D
D
D
21
H
D
D
D
D
D
D
D
22
H
D
D
D
D
D
D
D
23
H
D
D
H
D
D
D
D
24
H
D
D
D
D
D
D
D
25
H
D
D
D
D
H
D
D
26
H
D
H
D
D
D
D
D
27
H
D
D
D
H
D
D
H
28
H
H
D
D
D
D
D
D
29
D
D
D
D
D
D
D
D
30
D
H
D
D
D
D
D
D
31
H
H
D
D
D
D
D
D
32
D
H
D
H
D
H
D
H
33
D
H
D
D
D
D
D
D
34
H
D
D
D
D
H
H
H
35
H
H
D
D
D
D
H
D
36
H
D
D
D
D
D
D
H
37
H
D
D
H
H
H
D
D
38
D
D
D
D
D
D
D
H
39
H
H
D
D
H
H
D
D
40
D
-
-
-
-
-
-
-
zThe entire field planted to the race 2 susceptible cultivar Bluggoe. Each letter represents an individual
plant. H = asymptomatic plants and D= symptomatic plants based on visual assessment; samples were
obtained from plants in bold type. Total plants = 312; Total D = 238 (76%); Total H = 74 (24%).

56
collected in March 1995 and consisted of destructive sampling of inner rhizome and
pseudostem tissues. Sampling included both symptomatic and nonsymptomatic tissue.
Fungal Isolations
Tissue samples were allowed to air dry for 3 to7 days prior to transport to Florida.
In Florida, each sample was cut into approximately eight, 1-cm2 pieces, which each
contained at least one discolored and presumably infected vascular element. The tissue
pieces were surface disinfested by placing them in 70% ethanol for 10 seconds, followed
by a 2-min soak in 10% bleach, and a final rinse for at least 1 minute in sterile water.
Pieces were air dried in a laminar flow hood and then placed into two petri dishes to
provide four pieces of tissue in each dish. One of two media were poured over the tissue
pieces. Medium 1 consisted of amended water agar, streptomycin sulfate (10 mg/L;
Sigma Chemical Company, St. Louis, Mo.), rifampicin (12 mg/L; Sigma) and 10 //l of
danitol/L of medium (Chevron Corporation, San Francisco, CA.). Medium 2 consisted of
amended potato dextrose agar (PDA) containing streptomycin (10 mg/L), rifampicin (12
mg/L), danitol 10 ¡AÍLÍL, and terigitol NP-10 (100 ¿¿1/L; Sigma). Medium 1 allowed for
quick growth of Fusarhim spp. while suppressing the growth of other microorganisms.
However, identification of F. oxysporum isolates was difficult on this medium because of
poor mycelial growth and spore formation. Medium 2 allowed for better identification of
F. oxysporum colonies but the additional nutrients were also more conducive to the
growth of other microorganisms. Plates were incubated at room temperature, and within
3 days mycelial growth was observed on tissue pieces. Putative colonies of Fusarium spp.
were selected based on visual growth characteristics and the formation of microconidia or
macroconidia. Agar sections (1 mm2) were excised from these colonies and placed on the

57
reciprocal medium (from medium 1 to 2 and vice versa). Colonies that were confirmed to
be F. oxysporum based on mycelium pigmentation and growth characteristics,
microconidia or macroconidia formation, phialide anatomy, or chlamydospore formation
were streaked onto plates containing PDA. Within 14 hours, four single, germinated
spores were picked off using a compound microscope and placed with a sterile needle on
plates containing PDA; at least one single spore isolate was collected for each sample.
DNA Extraction. Southern Blotting and Hybridization
Procedures for DNA extraction have been described previously (Chapter 2).
Approximately 10 pg of DNA were digested with at least 10 units of either HcieIII or
Hindlll restriction enzymes (Bethesda Research Laboratories, Gaithersburg, MD) and
incubated for at least 3 hours at 37°C. Restriction fragments were separated by
electrophoresis in 0.7% agarose for ///>? digested DNA in TBE buffer at pH 7.0. Gels were run at 30 volts for approximately 24
hours. Bacteriophage lambda DNA digested with either Pstl or Hindlll was included on
each gel and used to calculate the molecular mass of restriction fragments obtained from
F. oxysporum f. sp. cúbense DNA. The DNA of each isolate was visualized by
incorporating 100 ng of ethidium bromide/ml of solution into the agarose gels and then
illuminating the gels using UV irradiation. The DNA was transferred to Nytran
membranes (Schleicher & Schuell, Inc., Keene, NH) using the capillary transfer method
(Sambrook et al., 1989). The DNA transfer proceeded for at least 12 hours, and the DNA
was immobilized by UV crosslinking (UV 254-nm crosslinker, model FB UVXL 1000,
Fisher Scientific, Pittsburgh, PA).

58
Mitochondrial DNA (mtDNA) was isolated from Honduran isolate 3 S1 (VCG
0120) following a procedure described previously (Kistler and Leong, 1986). Total
mtDNA was labeled using random hexamer primers to incorporate fluorescein-12-dUTP.
DNA labeling; hybridization and detection followed the procedures provided by the
manufacturer (Dupont NEN "Renaissance," E. I. du Pont de Nemours & Co. Inc., Boston,
MA). Prehybridization, hybridization and washing were performed at 65°C using a
Hybaid hybridization oven (Dot Scientific Inc., Flint, MI). Membranes were placed
between acetate sheets and exposed to x-ray film for 15 to 30 minutes.
Preliminary work on a world-wide collection of F. oxysporum f. sp. cúbense
isolates revealed that mitochondrial haplotypes were correlated with VCG and isolates
tested within a VCG had identical mitochondrial haplotypes (Kistler and Momol, 1990,
Chapter 2). Isolates belonging to the three VCGs found in Honduras (0120, 0124 and
0126) could be differentiated based on their unique mitochondrial profiles, obtained by
probing blots of total //aelll-digested DNA with mtDNA isolated from Honduran isolate
3 S1. Using this method, a mitochondrial haplotype was assigned to each isolate.
Honduran isolates which had mitochondrial profiles identical to isolates in VCG 0120
were designated mitochondrial haplotype (mthap) 0120, those which had mitochondrial
profiles identical to isolates in VCG 0124 or VCG 0126 were designated mthap 0124 and
mthap 0126, respectively.
Similarly, F. oxysporum f. sp. cúbense isolates were differentiated based on their
DNA fingerprint by digesting total DNA with the restriction enzyme Hindlll and probing
with clone pEYlO (Kistler et al, 1991). The clone pEYlO hybridizes to a nuclear
dispersed, middle repetitive DNA sequence (Benny and Kistler, unpublished) and contains

59
an open reading frame encoding a polypeptide with 51% amino acid identity to the
transposase of thq Magnaporthe grísea transposon Pot3 (Farman et-al., 1996; Pettway
and Kistler, unpublished). All previously tested isolates from a world-wide collection of
F. oxysporum f. sp. cúbense had unique DNA fingerprints using pEYlO (Kistler et
al., 1991). The DNA labeling, hybridization and detection procedures followed those
described above.
Data Analysis
DNA fingerprints were scored by visual assessment. Only bands that were of
strong intensity and consistently present were scored. Similar patterns were assigned a
number based on their frequency of occurrence in the field; the most common pattern was
assigned the number 1. Two chi-square analyses were performed. In the first test,
observed percent infected and noninfected of each FHIA hybrid and the two controls,
Flighgate and Williams, were obtained and compared against the null expectation of no
differences among cultivars. The expected number of infected plants for this test was 6.1,
which was obtained by summing the number of isolates obtained from the FHIA hybrids
and the cultivars Highgate and Williams and dividing by 189 the total number of plants in
the experiment, excluding the cultivar Bluggoe. This number was then divided by 9,
which represents the total number of cultivars in the experiment. The expected number of
noninfected plants was 14.8, and this number was obtained as described above but based
on the total number on noninfected plants. The chi-square analysis had 8 degrees of
freedom. The second test determined whether the pathogen genotypes obtained from
infected plants were randomly distributed among the different host genotypes. In this test,
the observed frequency of each pathogen genotype in the field and the host genotype from

60
which they were obtained were compared against the null expectation of no differences
among the host genotypes. The expected frequency for each genotype was estimated by
dividing the total number of isolates obtained from each cultivar by three. For example, a
total of 16 F. oxysporum f. sp. cúbense isolates were obtained from the cultivar Highgate,
and the expected number of isolates obtained for each mtDNA haplotype was 5.3. The
chi-square analysis had 16 degrees of freedom.
Field 1 was sampled twice and isolates were collected from both symptomatic and
asymptomatic tissue during each period. Duplicate isolates were collected from
symptomatic tissue of treatments 104-2, 106-3, 208-2, 306-2, 308-2, 408-1, and 408.3.
Additionally, duplicate samples were obtained from one Bluggoe plant. Field 2 was
sampled once and duplicate isolates were analyzed for four of the Bluggoe plants.
Additionally, many of the isolates from both fields were duplicated on different gels to
validate the results.
Results
Ninety-seven isolates were collected from Field 1 (the race 1-screening field). Of
these 97 isolates, 12 isolates were obtained from plants that were symptomatic during both
sampling periods and represent multiple samples taken from the same plant. A mtDNA
haplotype was assigned to each isolate by probing blots of //aelll-digested DNA from
each isolate with labeled mtDNA isolated from Honduran isolate 3S1 (Figure 3-1).
Isolates collected from the field were assigned a mtDNA haplotype as described in the
materials and methods. The incidence and distribution of pathogen mtDNA haplotypes
varied with host genotype (Table 3-4). The Highgate and Cavendish cultivars used as
susceptible and resistant experimental controls had the highest and lowest incidence of

12 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
61
Isolate/Host Cultivar
mtDNA Haplotype
1
1M-20 Bluggoe
Unknown type
2
1-5-4 Bluggoe
0124
3
1-3-63 Bluggoe
0124
4
1-3-62 Bluggoe
0124
5
1-1-52 Bluggoe
0124
6
2y-5-ll Bluggoe
0124
7
2-8-26 Bluggoe
0124
8
1-3-51 Bluggoe
Unknown type
9
1-3-3 Bluggoe
0124
10
106-2 Highgate
0120
11
601-2 FHIA 6
0126
12
308-1 Highgate
0126
13
707-3 FHIA 15
0126
14
502-1 FHIA 17
0126
15
304-3 FHIA 15
0126
16
108-1 FHIA 2
0126
17
707-3 FHIA 15
0124
18
408-3 Highgate
0126
19
409-3 FHIA 6
0126
Figure 3-1: mtDNA haplotype of selected Honduran field isolates.
x Isolates with 1-designation were collected from the cultivar Bluggoe in Field 1.
y Isolates with 2-designation were collected from the cultivar Bluggoe in Field 1.
z FHIA hybrids designated according to their plot number in Field 1.

62
infection, respectively. No isolates were obtained from the Cavendish cultivar Williams,
which is resistant to both race 1 and race 2 of the pathogen. In contrast, 17 isolates were
collected from the Highgate plants. Fourteen of the 17 isolates had 0126 mtDNA
haplotypes, whereas one each had 0120 and 0124 mtDNA haplotypes. One isolate
collected from the cultivar Highgate had a previously uncharacterized mtDNA haplotype
for F. oxysporum f. sp. cúbense.
Percent infection for the seven hybrid bananas ranged from less than 1% for FHIA
1 to 57% for FHIA 15 (Table 3-4). The observed percent infection of each cultivar was
obtained and compared against the null expectation of no differences among cultivars with
a series of chi-square tests. The differences in infection on the cultivars were highly
significant (p < .001) indicating that cultivars are not equally likely to be infected. This
suggests differences in host resistance genes. The distribution of pathogen genotypes
varied on the different host genotypes. The overwhelming majority of isolates collected
from the hybrids had 0126 mtDNA haplotypes. Isolates with 0120 and 0124 mtDNA
haplotypes were recovered from some of the hybrids but at a lower frequency compared
to isolates with 0126 mtDNA haplotypes. In contrast, 96% of the isolates collected in the
same field but obtained from the race 2 susceptible cultivar Bluggoe, had the 0124
mtDNA haplotype (Table 3-4).
The clone pEYlO was used to generate DNA fingerprints of the isolates (Figure
3-2). Isolates which had identical DNA fingerprints were presumed to represent the same
clone. Clone frequency and distribution on host genotypes were examined in the two
field populations. The nuclear DNA fingerprints of the isolates with 0126 mtDNA
haplotypes showed little variation. Only two patterns, which were 99% similar, were

Table 3-4. Isolates of Fusarium oxysporum f. sp. cúbense collected in Field 1 and their
incidence on different host genotypes
63
Number of Isolates with mtDNA Haplotypes
Cultivar or Hybrid
0120
0124
0126
Total
Isolates'
Infectiony
%z
Highgate (race 1
susceptible)
1
1
14
16
76
Williams (race 1 &
2 resistant)
0
0
0
0
0
Bluggoe (race 2
susceptible)
0
26
1
27
11
FHIA 1
0
1
0
1
<1
FHIA2
1
1
7
9
43
FHIA 3
0
0
2
2
14
FHIA 6
0
0
8
8
38
FHIA 15
0
l
11
12
57
FHIA 17
0
0
3
3
14
FHIA 23
0
1
3
4
19
x For plants where more than one isolate was collected, only one is represented if the
mtDNA haplotypes were identical.
y Total percentage of plants in Field 1 from which isolates were recovered
z Chi-square analysis: p < .001; df = 8

i : 3
4 5
6 7 8 9 10
I! 12 13 14 15 16
64
Isolate/Host Cultivar
mtDNA Haplotype
1
2x-3-18 Bluggoe
0124
2
102-3 FHIA 6
0126
3
2-7-37 Bluggoe
0124
4
707-2 FHIA 15
0126
5
503-2 Highgate
0126
6
1-5-66 Bluggoe
0124
7
503-3 Highgate
0124
8
707-3 FHIA 15
0124
9
309-1 FHIA 2
0124
10
208-1 Highgate
0126
11
1-1-59 Bluggoe
0124
12
304-1 FHIA 15
0126
13
208-2 Highgate
0126
14
2-8-16 Bluggoe
0124
15
1-1-29 Bluggoe
0124
16
1-5-4 Bluggoe
0124
Figure 3-2: HindiII digested DNA of selected Honduran field isolates that were probed
with clone pEYlO to generate DNA fingerprints.
x Isolates with 2-designation were collected from the cultivar Bluggoe in Field 2.
y FHIA hybrids designated according to their plot number in Field 1.
2 Isolates with 1-designation were collected from the cultivar Bluggoe in Field 1.

65
observed, and they represented the same clonal lineage. The two isolates with 0120
mtDNA haplotypes, which infected the cultivars Highgate and FHIA 2, had identical
nuclear DNA fingerprints that were distinct from isolates representing other mtDNA
haplotypes. With the exception of a single isolate, all of the isolates with 0124 mtDNA
haplotypes, which infected the cultivar Bluggoe and FHIA hybrids 1,2,15 and 23, had
identical nuclear DNA fingerprints that were distinct from isolates representing other
mtDNA haplotypes. An isolate collected from the cultivar Highgate with a 0124 mtDNA
haplotype had a unique nuclear DNA fingerprint.
In contrast to Field 1, which contained ten different genotypes of banana, Field 2
was entirely planted to the cultivar Bluggoe. Fifty-nine isolates of F. oxysporum f. sp.
cúbense were collected in this field, and all of them shared the same 0124 mtDNA
haplotype (Table 3-3). Similarly, all tested isolates had a nuclear DNA fingerprint
identical to those isolates in Field 1 which had a 0124 mtDNA haplotype and were
recovered from Bluggoe and FHIA hybrids 1,2, 15 and 23.
The observed frequency of each pathogen genotype in the entire field and the host
genotype from which they were obtained were compared against the null expectation of
no differences (no host selection) with a series of chi-square tests (Figure 3-3). The
differences in clone frequencies on the various host genotypes were highly significant.
Clones of the pathogen did not randomly infect banana plants representing the various
host genotypes. Instead clones were preferentially distributed on specific host genotypes.
Discussion
Clonal diversity was low in the two field populations of F. oxysporum f. sp.
cúbense; only five clones were recovered from the 10 genotypes of banana. Although the

66
30
25 -
Chi-square analysis: p < .001; df = 16
m
CO
T
CM
n
CD
LO
r^-
m
16
E
<
<
<
<
*—
OJ
Dl
r-
d
X
I
X
X
<
<
<
cn
LL
LL
LL
LL
X
X
X
if
$
LL
LL
LL
®mt haplotype 0120
umt haplotype 0124
â–  mthaplotype 0126
Host cultivar
Observed frequency of m thaplotype 0120 = .02 in Field 1.
Observed frequency of mt haplotype 0124 = .38 in Field 1.
Observed frequency of mt haplotype 0126 = .60 in Field 1.
Figure 3-3: Incidence and distribution of mtDNA haplotypes of Fuscirium oxysporum f.
sp. cúbense on host genotypes in Field 1.

67
number of clonal genotypes in these two populations was low, there appeared to be a
relationship between the level of genotypic diversity in the pathogen and host populations
By increasing the number of genotypes of banana in a field, more genotypes of the
pathogen were detected. In fact, the number of different clones of the pathogen which
were recovered in each field was proportional to the number of host genotypes planted in
the field. For example, all of the isolates recovered from host tissues in Field 2, which was
planted to a single cultivar, belonged to a single mtDNA haplotype and represented a
single clone. In contrast, isolates recovered from host tissues in Field 1, which was
planted to 10 different genotypes of banana, comprised three mtDNA haplotypes and five
clones. Each mtDNA haplotype was represented by only one or two clones.
These findings were consistent with field studies which compared the genetic
diversity of nonpathogenic F. oxysporum in native soils to F. oxysporum f. sp. melonis
isolates. In contrast to populations of native, nonpathogenic Fusarium oxysporum,
pathogen populations of F. oxysporum f. sp. melonis were less diverse based on their
VCGs and mtDNA haplotypes (Gordon and Jacobson, 1990a and b; Gordon and
Okamoto, 1992). In contrast, field populations of other asexually reproducing fungi,
including Stagonospora nodorum and Magnaporthe grísea as well as sexually reproducing
fungi including Sclerotinia sclerotiorum, Mycosphaerella graminicola and some
populations of Phytophthora i ufes tans, contained a larger number of genotypes, indicating
higher levels of clonal diversity (Goodwin et al., 1992; Kohli et al.,1995; Kohn, 1995,
Levy et al., 1993; McDonald et al., 1994; McDonald et al., 1995).
The recovery of different clones of F. oxysporum f. sp. cúbense in these two fields
was dependent on the genotype of the host. Chi-square analyses were performed to test

68
whether isolates of the pathogen were randomly recovered from the different banana
genotypes; these showed a significant association between a mtDNA haplotype and its
incidence on specific host genotypes. In fact, isolates with certain mtDNA haplotypes
were more frequently and sometimes almost exclusively recovered on certain host
genotypes (Figure 3-3). For example, 99% of the isolates recovered from the race
2-susceptible cultivar Bluggoe in Fields 1 and 2 had a 0124 mtDNA haplotype and
represented a single clone based on their identical DNA fingerprints. In contrast, 87% of
the isolates recovered from Highgate had a 0126 mtDNA haplotype, representing two
nearly identical clones.
The occurrence of a specific genotype of the pathogen associated with virulence
on a specific host genotype is likely the consequence of linkage disequilibrium found in
clonally reproducing organisms (Tibayrenc et al., 1991). Since sexual reproduction does
not occur in these organisms, the entire genome is effectively linked (Anderson and Kohn,
1995). Therefore, alleles at different loci, such as those for mt haplotypes, VCG,
virulence, are repeatably associated. Additionally, in these field populations it appears that
the host genotype drives selection for particular clones of F. oxysporum f. sp. cúbense.
In fields planted to a susceptible cultivar, genotypes of the pathogen which are virulent on
the host genotype may have greater fitness. They may, for example, produce more
propagules. The pattern of introducing strains of F. oxysporum f. sp. melonis to
agricultural soils and the correspondingly intense selection pressure on the susceptible
crop, has been described previously (Gordon and Okamoto, 1992). Such selection
pressure undoubtably has an even more significant effect on the pathogen populations in
perennial, monocultural production systems, such as banana plantations, since susceptible

69
host plants are exposed to the pathogen for even longer times. In these systems, one
would expect a large increase in the number of propagules of virulent pathogen genotypes
when susceptible cultivars are grown.
In contrast to certain plant pathogens, such as rusts, where new pathotypes appear
annually and breeders are continually challenging the pathogens by introducing cultivars
with new combinations of resistant genes, resistance genes to Fusarium wilt in Cavendish
cultivars have provided stable, relatively long-term protection. In fact, Cavendish cultivars
have been planted in Central America for at least 40 years and no substantial reports of
Fusarium wilt on these cultivars have been made. However, in a limited number of
subtropical and tropical countries, race 4 isolates of the pathogen, which are capable of
causing disease on Cavendish cultivars, do occur. The asexual nature of this fungus may
preclude rapid change in the pathogen population. However, lack of genetic change due
to recombination in populations of the pathogen actually may be detrimental in
commercial banana plantations where monoculture production of clonally propagated
banana cultivars are typical. In these cropping systems, when a virulent pathotype
evolves, lack of recombination in the pathogen can serve to stably perpetuate the rapid
selection and growth of the most virulent pathotype.
This study, in combination with information from Chapter 2, indicates that race 1
isolates belong to more than one distinct lineage. The study of a world collection of F.
oxysporum f. sp. cúbense found that isolates in VCG 0124 belonged to one clonal lineage,
while isolates in VCG 0120 and 0126 belonged to a second lineage. Isolates in these two
lineages were as genetically dissimilar to one another as they were to isolates of F.
oxysporum f. sp. niveum and f. sp. lycopersici isolates. In the present study, the DNA

70
fingerprint of isolates representing these two lineages were distinct, further substantiating
their genetic disimilarity. Race 1 isolates are known to occur in all three of the VCGs
found in Honduras (Ploetz and Correll, 1988) and the present results showing
mitochondrial haplotypes corresponding to these three VCGs, recovered from diseased
Highgate plants (Table 3-4), support this observation. The definition of race assumes that
all race 1 isolates share identical virulence characteristics corresponding to susceptibility in
the cultivar Gros Michel (and Highgate) and resistance in the cultivar Cavendish.
However, in this study, the frequency with which genotypes representing these two
lineages were isolated differed in the race 1 -susceptible cultivar Highgate and the FHIA
hybrids. Isolates in the lineage represented by mthap 0126 preferentially infected the
cultivar Highgate and most of the FHIA hybrids. In contrast, isolates with mthap 0124,
which represent a second phylogenetic lineage, were recovered only at a low frequency
from these same host genotypes. Because isolates with mthap 0124 were abundantly
present in the field and were frequently recovered from Bluggoe plants, the basis for the
race 1 phenotype may be different in these two lineages. This may imply genetic
differences in the virulence determinants of these two nominally race 1 lineages. This of
course assumes that isolates with identical mthap and DNA fingerprints in field 1 have the
identical pathogenic phenotype.
Similar situations have been observed in other fungi. In a survey of Mcignciporthe
gi-isea pathotypes from the United States, Levy et al. (1991) found that isolates of a single
pathotype had significantly different DNA fingerprints, and this suggested that the
pathotypes were derived from independent clonal lineages. In F. oxysporum f. sp.
cúbense, isolates in VCG 0126 that cause disease in Highgate may possess genetic factors

71
that make them better adapted than those in VCG 0120 and 0124 to the ecological and
environment factors in Honduras.
In this study, the majority of isolates recovered from Highgate had mthap 0126,
whereas the majority of isolates recovered from Bluggoe had mthap 0124 (Table 3-2).
Isolates representing other haplotypes were recovered on these host genotypes, but at low
frequencies. Additionally, similar results were observed for some of the FHIA hybrids.
The presence of a low rate of mixed infections may provide one explanation for this
observation. During the initial selection of isolates from the infected tissue, occasionally
multiple samples were taken from a single plant when fungal colonies displayed different
growth characteristics. Of these samples, only one plant was infected with isolates having
both 0124 and 0126 mtDNA haplotypes. Additionally, five plants from Field 1, each
representing different genotypes of banana, were sampled multiple times by excising
different xylem elements within a single plant. A total of 45 samples were obtained and all
of the sampled plants were infected by isolates representing a single mtDNA haplotype.
Thus, it appears that mixed infections occur, albeit at a low frequency.
Banana breeders in Honduras as well as other regions are involved in developing
eultivars of banana with resistance to all races of this pathogen. Because of the polyploid
nature of bananas and the sterility of many of the important edible eultivars, it is
impossible to do traditional plant breeding and disease screening. Additionally, in situ
disease screening in regions where the pathogen is nonendemic is limited to isolates found
within the country. Since isolates belonging to only three VCGs and two clonal lineages
have been described in Honduras, yet over 17 VCGs and 10 clonal lineages have been
described worldwide, breeding for resistance using endemic populations in Honduras may

72
not result in resistance to populations of the pathogen in other locations (Ploetz, 1995;
Stover and Buddenhagen, 1986). Ideally, new genotypes should be screened against
isolates that encompass the range of genetic diversity present in the entire formae
speciales. Therefore, there is a need to assess the range of variation in the pathogen
populations for virulence on particular breeding lines (Milgroom and Fry, in press). Using
both the mtDNA haplotype and a nuclear DNA fingerprinting probe to estimate genetic
diversity within a field site can improve in situ screening for Fusarium wilt. In this
pathosystem there is a relatively long-term interaction between host and the pathogen that
may not be assessed in artificial inoculation tests (Buddenhagen, 1990). Using molecular
techniques for assessing the pathogen population at a given site and determining the
lineage of isolates infecting particular host genotypes can increase the effectiveness of
disease screening and assist in the evaluation of new breeding lines.

CHAPTER 4
SUMMARY AND CONCLUSIONS
This research has provided a genetic characterization and phylogenetic analysis of
isolates of F. oxysporum f sp. cúbense representing both a world collection and
population from two fields. Prior investigations, which genetically characterized isolates
based on VCGs, RAPDs and electrophoretic karyotyping, provided only limited
information on the genetic relationships among isolates (Bentley et al., 1995; Boehm et al.,
1994; Ploetz and Correll, 1988; Sorensen et al., 1993). However, the present research,
which utilized numerous, independent DNA clones to identify polymorphic loci among a
worldwide collection of isolates, has provided a comprehensive assessment of the genetic
relationships among isolates. Parsimony analysis of the data using bootstrapping also
provided a statistical evaluation of the RFLP data as well as an unbiased assessment of
significant groupings among the isolates.
In addition, this research represents the first attempt to determine the amount of
genotypic diversity present in field populations of F. oxysporum f sp. cúbense based on
DNA fingerprinting. The study of two Honduran field populations provides information
regarding the amount of genetic variability present in a defined location. By identifying
the mtDNA haplotypes of individual isolates and determining the extent of clonality in the
population using a DNA fingerprinting probe, this study assessed the genetic diversity of
73

74
the collected isolates and demonstrated how the genotype of the host can affect the
pathogen populations.
Both studies demonstrated the importance of choosing an appropriate DNA
marker in phylogenetic and population genetic investigations. An important benefit to
using RFLPs in phylogenetic analysis is that the number of RFLP markers available for use
is large because any cloned, low-copy-number piece of DNA can be used as a probe and
several restriction enzymes can be assayed to identify RFLPs with each probe
(Michelmore and Hulbert, 1987). One important characteristic of the RFLP markers used
in this phylogenetic analysis is that they were likely selectively neutral. This is in contrast
to several markers that have been employed in traditional studies, such as pathogen
virulence genes or resistance to fungicides, which are under strong selection pressure in
agricultural systems and may provide a biased estimate of genetic diversity (McDonald
and McDermott, 1993).
Other molecular markers that are commonly used in phylogenetic analyses of fungi
include isozymes, RAPDs, the nuclear and mitochondrial rRNA genes and internal
transcribed spacers (ITS), and mtDNA maps. Isozymes have been used to characterize
fungal isolates but their disadvantage is that the possible number of enzymes that are
informative for an organism is limited and many loci are monomorphic and, thus,
noninformative. (Michelmore and Hulbert, 1987).
Analysis of F. oxysporum f. sp. cúbense isolates using RAPDs was attempted
initially in this research. However, this technique proved to be unreliable for phylogenetic
studies. The largest obstacle to using them was that polymorphisms could not be scored
unambiguously without cloning the putative polymorphic bands and then performing an

75
RFLP analysis. Even when RAPDs were used that could be excised with an appropriate
restriction enzyme, results were difficult to interpret (see Appendix I). Two problems
associated with methods based on polymerase chain reaction (PCR), such as RAPDs, are
that 1) nonamplification can result from base pair substitutions at any of the different
nucleotides complementary to the primers at the ends of the amplified fragment, and 2)
results may be affected by changes in environmental conditions and at times are not
reproducible (Rosewich & McDonald, 1994).
Mitochondrial DNAs and, in certain organisms, mitochondrial plasmids have been
used to characterize isolates of F. oxysporum f. sp. melonis, f. sp. conglutinans, and f. sp.
cúbense (Jacobson and Gordon, 1990; Kistler et al., 1987; Kistler and Momol, 1990).
Two disadvantages of using mtDNA for phylogenetic analyses is that the genome is small,
so consequently fewer polymorphisms may be found, and the rate of evolution of mtDNA
may be different than that of nuclear DNA. Mitochondrial DNA was used in both the
phylogenetic and population studies. In the phylogenetic study, groupings of isolates
based on mtDNA haplotypes were congruent with groups based on the RFLP analysis;
however, fewer groups could be distinguished based on mtDNA analysis. In the
population studies, a mtDNA haplotype was used to group similar isolates, which were
then subjected to DNA fingerprint analysis to determine whether isolates were clonally
derived.
Sequencing of both the nuclear and mitochondrial rRNA genes as well as ITS
regions have been described for fungi and used in phylogenetic studies (Bruns et al.,
1991). An underlying assumption of sequence analysis is that the phylogeny of the region
is a good indicator of the phylogeny of organisms (Bruns et al., 1991). Another approach

76
to identifying sequence differences in the ITS region is through an RFLP analysis using the
PCR-amplified region. In this study, a number of primers were used to amplify ITS
regions ofF. oxysporum f. sp. cúbense isolates. Several four- and six-base restriction
enzymes were used to digest the amplified products (Appendix II). No polymorphisms
were observed using the restriction enzymes studied, and this methodology was
discontinued because it did not prove useful for a phylogenetic analysis.
A central point of discussion in both studies was the overwhelming evidence of
clonality in this organism. With the exception of isolates in VCG 0123, the genetic
evidence which resulted from this research is consistent with the hypothesis that F.
oxysporum f. sp. cúbense is a clonally reproducing organism. Based on criteria established
by Tibayrenc et al. (1990), a unifying feature of clonally reproducing organisms is
widespread geographic distribution of a few successful clones. The RFLP study identified
72 multilocus haplotypes in a worldwide collection of F. oxysporum f. sp. cúbense. The
five most common haplotypes accounted for nearly half of the isolates, while the two most
common haplotypes were found on all five continents sampled in this study, indicating the
pantropical distribution of a small number of genotypes. Additional evidence of clonal
reproduction is the absence of recombinant genotypes. Significant gametic disequilibrium
for alleles at 34 of 36 loci tested supported nonrandom association between alleles of
different loci. Finally, the strong correlation between independent genetic markers (VCG,
mitochondrial and multilocus RFLP haplotype) also are indicative of a clonally
reproducing organism (Milgroom, 1996; Tibayrenc et al., 1990).
The Honduran field population study also indicated that clonal reproduction
accounted for the occurrence of a specific genotype of the pathogen associated with

77
virulence on a specific host genotype. This observation is likely the consequence of linkage
disequilibrium found in clonally reproducing organisms where alleles at different loci, such
as mtDNA haplotypes, VCG, virulence, are observed to be repeatedly associated. The
clonal structure in agricultural populations of plant pathogenic fungi also has been found
in some populations of Sclerotinia sclerotiorum, Fusarium oxysponim, Phytophthora
infestans and Maguaporthe grísea (Appel and Gordon, 1994; Goodwin et al., 1992;
Gordon et al., 1992; Kohli et al., 1995; Kohn, 1995; Levy et al., 1991).
A significant finding of the phylogenetic analysis was that a number of the clonal
lineages are phylogenetically distinct. Host specialization, such as the ability to cause
disease on a particular host, appears to be polyphyletic in F. oxysponim f. sp. cúbense.
Based on coefficients of similarity, isolates in the two largest lineages of F. oxysponim f.
sp. cúbense are genetically more similar to the F. oxysponim f. sp. niveum isolate than to
each other. Similarly, they are roughly as genetically distinct from each other as either is to
the isolates of F. oxysponim f. sp. lycopersici. An interpretation of these results is that
isolates belonging to the dissimilar groups acquired their ability to be pathogenic on
bananas independently. This is the first reporting of a potential polyphyletic acquisition of
pathogenicity on a particular host for F. oxysponim f. sp. cúbense. Appel and Gordon
(1994) observed a close relationship between nonpathogenic strains of F. oxysponim and
F. oxysponim f. sp. melonis isolates in VCG 0131, suggesting an independent origin as a
melon pathogen separate from F. oxysponim f. sp. melonis isolates in VCG 0134.
However, this study was based on mtDNA haplotype data and interpretation may be
complicated by the possibility of horizontal transfer of mtDNA.

78
Interestingly, the results of the RFLP study indicate that isolates in VCG 01214,
which comprise lineage FOC V, are genetically distant from other F. oxysporum f. sp.
cúbense isolates. One explanation is that this group could have originated in East Africa,
a region outside the host's center of origin. Another explanation is that the genetic
differences in these isolates could have arisen by a founder effect. Founder effects result
when small groups of individuals are isolated from larger populations. It is possible that
over a period of time, isolates in VCG 01214 were subjected to different selection forces
which altered gene frequency in this group. Additionally, mutations which occur in the
founding population could be fixed in the successive generations due to the clonal nature
of this organism.
The data indicate that isolates in VCG 0123, which comprise both lineages FOC
VII and X, are phylogenetically distinct. This group may provide evidence of an ancient
genetic exchange between individuals in lineages FOC I and II, or they may represent an
ancestral group possessing primitive character states.
There are numerous practical applications of this research; however, only the most
important aspects are highlighted. The phylogenetic study helps clarify the genetic
relationships among the isolates which represent different races of the pathogen. Isolates
which represent race 1 of the pathogen occur in the two largest and divergent lineages
(FOC I and II) in the collection. However, data from the Honduras field study revealed
that isolates which have been considered race 1 based on reaction to the host differential
may not share the same host infectivity. In field tests, isolates representing these two
lineages were not recovered at the same frequency from susceptible host tissue, even
though isolates representing FOC I and FOC II were assumed to be evenly distributed

79
throughout the field. Isolates which represent race 2 appear to be limited to lineages FOC
I and VIII, which are associated with one main branch on the midpoint rooted tree, and
FOC V, which is only distantly related to other F. oxysporum f. sp. cúbense isolates
(Figure 2-2). In contrast, isolates that represent race 4 appear to be limited to lineages
FOC II, III, and VI, which form a second branch on the midpoint rooted tree. Therefore,
these two races appear to have evolved only in certain lineages of the pathogen. Race 4
isolates occur in many of the VCGs that are present in tropical areas, such as Central
America, the Caribbean, and South America, but where race 4 has not yet been reported.
If the ability to cause disease on Cavendish cultivars is controlled by a single gene it is
possible that race 4 isolates could evolve in regions where it has not yet been found.
However, it would most likely evolve in those isolates representing VCGs that presently
contain race 4 isolates. Additionally, because race 4 isolates have a wide host range, it is
important to identify and develop new sources of disease resistance. These new sources
of disease resistance must then be incorporated into commercially acceptable banana
cultivars.
Banana breeders worldwide are involved in developing cultivars of banana with
resistance to all races of F. oxysporum f.sp. cúbense but are faced with a number of
obstacles. Because of the polyploid nature of bananas and the low level of fertility of
important edible cultivars, it is impossible to use many of the desirable commercial clones
as parental lines for breeding work. Additionally, the results of container disease
screening experiments do not necessarily correlate in field studies. Stover and
Buddenhagen (1986) pointed out that results from disease screening tests, which relied on

80
seedlings, small rhizomes or in vitro meristem plants grown in containers and subjected to
artificial inoculation, did not always correlate with field tests.
Additionally, in situ disease screening in regions where the pathogen is not
endemic is limited to isolates found within the country. The Honduras field study
demonstrated that the mtDNA haplotype and a nuclear DNA fingerprinting probe can aid
in situ screening for resistance to Fusarium wilt. Using molecular techniques for
assessing the pathogen population at a given site and determining the lineage of isolates
infecting particular host genotypes can increase the effectiveness of screening programs
and assist in the evaluation of new breeding lines. To date, disease screening programs
primarily have relied on visual assessment to determine disease incidence on new breeding
material. Using molecular techniques to characterize isolates of F. oxysporum f. sp.
cúbense infecting banana breeding lines would enable breeders to determine host
resistance and susceptibility based on the lineage of the pathogen. This could aid in
directing breeding programs toward developing resistance to all lineages of the pathogen
rather than the three races that are known to infect bananas. This would provide a more
comprehensive approach to^breeding for resistance to this pathogen. Finally, the optimal
disease screening methodology requires that the isolates that are being used encompass the
range of genetic diversity present in the entire forma specialis. This could be
accomplished by screening breeding materials with representatives of each of the 10
lineages of F. oxysporum f. sp. cúbense described in this research. Ideally, this would
occur in a field in which all lineages of the pathogen could be introduced.

APPENDIX A
EXPLORING THE POTENTIAL OF USING POLYMERASE CHAIN
REACTION METHODOLOGIES IN PHYLOGENETIC STUDIES OF
FUSARIUM OXYSPORUM F. SP. CUBENSE
Introduction
Randomly amplified polymorphic DNAs (RAPDs) have been used as molecular
markers in a number of fungal species to distinguish among genetically different isolates,
to aid in genetic mapping and for DNA fingerprinting. Randomly amplified polymorphic
DNAs can be generated quickly using small quantities of DNA which do not need to be
highly purified. However, RAPD techniques are prone to artifactual variation, which can
result in erroneous data (Ellsworth et al., 1993). The possibility of artifactual variation
prevents the direct use of the technique in genetic diversity studies aimed at describing the
genetic relatedness among isolates. Similarly, the inability to determine directly if bands of
similar molecular size represent similar amplification products presents a problem in
scoring individuals in genetic diversity studies.
Sequencing of both the nuclear and mitochondrial rRNA genes, as well as the
internal transcribed spacer region (ITS) region, have been described for fungi and used in
phylogenetic studies (Bruns et al., 1991). The ITS region and intergenic spacer of the
nuclear rRNA repeat units evolve the fastest compared to other regions of the rRNA
genes and may vary among species and populations (White et al., 1990). The ITS primers
make use of the conserved regions of the 18S, 5.8S and 28S rRNA genes to amplify the
81

82
noncoding regions between them (White et al., 1990). The primers are based on
sequences from Scicchciromyces cerevisiae, Dictyostelium discoideum and Stylonicha
pustulata (Dams et al., 1988). A diagram of these primers and their location along the
nuclear rDNA region of Saccharomyces cerevisiae is provided by White et al., (1990).
Restriction fragment length polymorphisms and RAPD markers are advantageous
in phylogenetic studies because they are presumed to be random traits, not subject to
natural selection. In this study, the ITS regions were amplified and subjected to restriction
enzyme digestion to determine if polymorphisms could be detected. Additionally, RAPD
fragments were generated to provide quick identification of putative polymorphic loci.
Polymorphic amplification products were labeled and used as probes for Southern
hybridization, enabling the confirmation and description of each polymorphism. The
objective of this study was to identify polymorphic regions of the genome that could be
used in phylogenetic analysis of a large subset of a world collection of Fusarium
oxysporum f. sp. cúbense.
Results and Discussion
ITS Region Analysis
A number of ITS primers described by White et al. (1990), were used in this study
to amplify the total genomic DNAs from a subset of isolates representing a large number
of VCGs and a pantropical distribution (Table A-l). The amplification protocol for this
study, as well as the RAPD analysis, are provided in Table A-2. Internal transcribed
spacer primers 2 and 5 were used to amplify the ITS 1 region of the nuclear rRNA genes
and primers ITS4 and 5 were used to amplify the ITS 1, ITS2 and the 5.8S rDNA regions
in F. oxysporum f. sp. cúbense isolates (for primer sequences see Table A-3). The

83
resultant amplified products were similar for all isolates. These products were then
subjected to restriction enzyme digestion using enzymes HaelII, Sau3M, and EcoRl. All
of the tested isolates had exactly the same restriction pattern for all of the enzymes tested.
RAPP Analysis
The identification of potential polymorphic regions of the genome was
accomplished using RAPD analysis following protocols described previously (Welsh and
McClelland, 1990; Williams et al., 1990). A number of oligonucleotide primers were
screened for their ability to amplify bands in F. oxysponim f. sp. cúbense (Table A-4).
Many of the primers generated heterogenous products which could be used to differentiate
isolates. The majority of primers generated numerous bands of varying intensity. A
limited number of polymorphic bands were further tested for their potential in identifying
polymorphisms using Southern analysis. Southern analysis provided a method for
distinguishing whether a missing band observed in a RAPD profile was the result of the
presence of multiple alleles, deletions or a problem with the amplification procedure.
Randomly amplified polymorphic DNA fragments were generated using 10-base
oligonucleotide primers which contained internal hexameric palindromes that may be cut
by single restriction endonucleases specific to the sequence (Kit F, Operon Technologies,
Inc.). Fragments were subjected to agarose gel electrophoresis, and putative polymorphic
regions were identified. Polymorphic amplified products were excised from low melting
point agarose gels and labeled with a nonradioactive digoxigenin (dig)-l 1-dUTP using the
random primed method (Boehringer Mannheim, Genius System).
Genomic DNA isolated from a subset of 30 isolates representing 15 VCGs and a
pantropical distribution (Table A-l) was digested with restriction enzymes with

84
recognition sequences corresponding to those found in the internal hexameric sequences
of the oligonucleotide primer. Standard protocols were used for restriction enzyme
digestions, blotting, and hybridizations as described in chapters 1 and 2.
When genomic DNAs were amplified with an oligonucleotide containing a BamHl
recognition sequence (GAGGATCCCT-primer F2), a 0.68-kilobase (kb) amplification
product was present in isolates 0-1224, STD1, BLUG, S?, B2-1, MW65, STN3, MW11,
CVA, STNP5, 22994 and STNP1, but absent in the remaining isolates listed in Table A-l.
This fragment, designated BLUG-2-1, was excised from a gel containing the amplification
product from isolate Blug in VCG 0124 and labeled as described above. Genomic DNAs
from the same isolates were digested with BamHl and probed with BLUG-2-1. A 0.68-kb
fragment that hybridizes to BLUG-2-1 was present only in those isolates that amplified a
,68-kb fragment by PCR. The same probe has been used on a total of 80 isolates of F.
oxysporum f. sp. cúbense. The probe appears to be specific for DNAs from isolates in
VCGs 0124, 0124/25, 0125, 0128 and 01212.
When genomic DNAs were amplified with an oligonucleotide containing a EcoRV
recognition sequence (GGGATATCGG-primer F8), a 0.95-kb fragment was present in
isolates 0-1224, STD1, BLUG, S?, B2-1, MW65, STPN3, MW11, CVA, 22994 and
STNP1, but absent in isolate STNP5 in VCG0125 and the remaining isolates described in
Table A-l. This fragment, designated BLUG-8-1, was excised from a gel containing the
amplification product from isolate Blug in VCG 0124 and labeled as described above.
Genomic DNAs from the isolates were digested with EcoRV and probed with BLUG-8-1
The probe hybridized to a monomorphic 0.95-kb DNA fragment in all isolates.
Additionally, the probe hybridized to a 2.1-kb DNA fragment in isolates 0-1224, STD1,

85
BLUG, S?, B2-1, MW65, STPN3, MW11, CVA, 22994 and STNP1. The PCR
amplification using the 10-mer revealed a polymorphism not detected in Southern
hybridizations with restricted DNA, which suggests that the polymorphism is due to a
point mutation at the F8 primer sequence, but not in the EcoR\ recognition site. With the
exception of isolate STNP5, the 2.1-kb fragment appears to be specific for DNAs from
many isolates in VCGs 0124, 0124/25, 0125, 0128 and 01212.
When genomic DNAs were amplified with an oligonucleotide containing a EcoRl
recognition sequence (CCGAATTCCC-primer F5), a 1.6-kb amplification product was
present in some isolates, and either absent or less intensively amplified in other isolates.
This fragment, designated SH3142-5-1, was excised from a gel containing the
amplification product from isolate SH3142 and labeled as described above. Genomic
DNAs from the isolates were digested with EcoRl and probed with SH3142-5-1. A 1.6-
kb fragment was clearly present in isolates Ph6, MW40, ORT2, 1-1,0-1219, STGM1,
F9127, 15638, FCJ7, Pacovan, F9130, DAVAO, A2-1, STB-2, and SH3142. A 1.6-kb
PCR amplified fragment was also clearly visible in these isolates. Isolates 0-1224, STD1,
S?, STN3, B2-1, MW11, CVA, STNP5, 22994 and STNP1, which contained a very faint
1.6-kb amplified product, weakly hybridized to the 1.6-kb labeled probe. In the same
isolates, the probe hybridized more strongly to a 4.6-kb fragment. No hybridizations were
observed in isolates BLUG and MW65, which did not contain a 1.6 amplification product.
A monomorphic 3.0-kb fragment was present in isolates that contained either a 1.6-kb or
4.6-kb fragment.
Informative polymorphic single-copy and multi-copy loci can be identified quickly
by PCR using random 10-base oligonucleotide primers. However, a correlation between a

86
polymorphism observed through PCR amplification and a similar polymorphism observed
when the DNA of the same isolates is subjected to Southern analysis, does not always
occur. This poses a problem for directly interpreting the polymorphisms observed from
RAPD analysis in a phylogenetic study.
Polymorphic products generated from single primer amplification may be used as
probes for Southern hybridization. Restriction fragment length polymorphisms identified
through Southern hybridization can be used to discern genetic relationships among isolates
of the world collection. However, some RAPD bands may represent multiple-copy
sequences and therefore are not ideal in phylogenetic studies. However, probes that
appear to be lineage specific, such as BLUG-2-1, may provide a quick method for
determining the VCG of new isolates and for genetic studies of field populations.
In conclusion, using RAPD analysis for the determination of phylogenetic
relationships among isolates is difficult because in many cases interpretation of the data is
ambiguous. Clark and Lanigan (1993) have set forth criteria that should be met when
using RAPD analysis to estimate nucleotide divergence of closely related taxa. One
criterion identified is that polymorphic bands must be shown to behave as Mendelian
factors. Since F. oxyspomm f. sp. cúbense does not have a known sexual state,
Mendelian inheritance can not be proven. Another criterion is that allelism of bands must
be ascertained by Southern blotting or segregation analysis. For F. oxyspomm f. sp.
cúbense, segregation analysis is not an option and Southern blotting in two of the three
examples described above did not provide definitive answers to the allelic nature of the
polymorphisms present in the isolates.

87
Table A-l. Isolates of Fusarium oxysporum f. sp. cúbense used in RAPD analysis and
presence or absence of major RAPD bands.
Isolate
VCG
Primer F-2
0.68-kb band
Primer F-5
1.6-kb band
Primer F-8
0.95-kb band
0-1219
0120
+
STGM1
0120
-
+
-
F9127
0120
-
+
-
15638
0120
-
+
-
FCJ7
0120
-
+
-
Pacovan
0120
-
+
-
ORT2
0120
-
+
-
F9130
0121
-
+
-
Ph6
0122
-
+
-
DAVAO
0123
-
+
-
0-1224
0124
+
+*
+
STD1
0124
+
+*
+
BLUG
0124
+
-
+
S?
0124
+
+*
+
B2-1
0124
+
+*
+
MW65
0124
+
-
+
STN3
0124
+
+*
+
MW11
0124/25
+
+*
+
CVA
0124/25
+
+*
+
STNP5
0125
+
+*
-
STB-2
0126
-
+
-
22994
0128
+
+*
+
N5331
0129
-
+
-
A2-1
01210
-
+
-
A1
01210
-
+
-
SH3142
01211
-
+
-
STNP-1
01212
+
+*
+
1-1
01213
-
+
-
MW40
01214
+
-
* Weak band present

88
Table A-2. Amplification protocol for RAPD and ITS analysis
25 ng template DNA
15 ng primer DNA (Operon Technologies, San José, CA)
0.3 iA lOmMdATP
0.3 /u\ lOmM dCTP
0.3 ¡A lOmMdTTP
0.3 ¡A lOmM dGTP
0.2 U Taq polymerase (Boehringer Mannheim, Indianapolis, IN)
1 ¡A lOmM MgCl,
2.5 ¡A lOx Taq polymerase buffer
sterile water to volume of 25 [A
Amplifications were performed in a Gene Machine II thermal cycler. The machine was
programmed for one cycle for 30 seconds at 94°C followed by 35 cycles of 1 minute at 94°C,
1 minute at 35°C, and 2 minutes at 72°C, followed by a final cycle of 10 minutes at 72°C.

89
Table A-3. Primer sequence for the ITS regions
ITS Primer'
Sequence
2
4
5
GCTGCGTTCTTCATCGATGC
TCCTCCGCTTATTGATATGC
GGAAGT AAAAGT C GT AAC AAGG
'Primers obtained from Dr. Kerry O'Donnell with the sequences based on White et al. (1990).

90
Table A-4. List of oligonucleotide primers tested on Fusarium oxysporum f. sp. cúbense
isolates
Operon Primer
Amplification
Recognition Sequence (Yes/No)
B1
GTTTCGCTCC
Y
B2
TGATCCCTGG
Y
B3
CATCCCCCTG
Y
B4
GGACTGGAGT
Y
B5
TGCGCCCTTC
Y
B6
TGCTCTGCCC
Y
B7
GGTGACGCAG
Y
El
CCCAAGGTCC
Y
E2
GGTGCGGGAA
Y
E3
CCAGATGCAC
Y
E5
GTGACATGCC
Y
E6
AAGACCCCTC
Y
E7
AGATGCAGCC
Y
Ell
GAGTCTCAGG
Y
E12
TTATCTCCCC
Y
E13
CCCGATTCGG
Y
E14
TGCGGCTGAG
Y
E15
ACGCACAACC
Y
E16
GGTGACTGTG
Y
E17
CTACTGCCGT
Y
E18
GGACTGCAGA
Y
E19
ACGGCGTATG
Y
E20
AACGGTGACC
Y
FI
ACGGATCCTG
Y
F2
GAGGATCCCT
Y
F4
GGTGATCAGG
Y
F5
CCGAATTCCC
Y
F6
GGGAATTCGG
Y
F7
CCGATATCCC
N
F8
GGGATATCGG
Y
F10
GGAAGCTTGG
Y
FI 1
TTGGTACCCC
Y
F12
ACGGTACCAG
Y
F13
GGCTGCAGAA
Y
F14
TGCTGCAGGT
N
F15
CCAGTACTCC
N
F16
GGAGTACTGG
Y
F17
AACCCGGGAA
Y
F18
TTCCCGGGTT
Y
F19
CCTCTAGACC
N
F20
GGTCTAGAGG
Y

APPENDIX B
ADDITIONAL INFORMATION FROM PHYLOGENETIC STUDY
Probe and enzyme combinations used in the RFLP of F. oxysporum f. sp. cúbense study
with the molecular weights of each observed allele.
Probe2
Enzyme
Size of
Amplified
Bands in Kb
Probe2
Enzyme
Size of
Amplified Bands
in Kb
7
Dra I
6.4X
204
Hae III
1.0X
5.5
,92x
4.8X
.85
4.7
225
EcoRV
7.5
4.1
4.7X
7
Hae III
1.0X
4.5X
0.6X
3.8
0.4X
2.2X
7
EcoRV
2.8y
2.0
30
Dra I
5.6y
228
EcoRV
5.3
4.8y
4.2
2.8y
3.8
EcoRV
1.8y
3.2X
Hae III
1.6y
260
Dra I
>20x
120
Hae III
1.7
>10
1.5X
6.4X
1.2
4.8
00
LAi
4.2
,69x
6
91

92
Probe2
Enzyme
Size of
Amplified
Bands in Kb
Probe2
Enzyme
Size of
Amplified Bands
in Kb
.60
260
EcoRV
5.5
58x
4.6
162
Dra I
4.9X
4.5
2.8
3.8X
162
EcoRV
>14.07y
3.5X
162
Hae III
1.5y
1,4X
,63y
,94x
177
EcoRV
4.5
261
EcoRV
5. lx
2.2X
5.0X
1.2*
X
00
187
EcoRV
6.5
4.7
5.1x
4.6
4.6
4.4
4.2
204
EcoRV
6.4X
5.5X
4.5
4.2
4.1
4.0
2.2X
1.3
x Represents the most commonly occuring allele(s).
y Represent monomorphic allele(s).
z Source of Probes: Elias et al., 1993

APPENDIX C
ALLELIC DATA SCORED AS PRESENCE OR ABSENCE FOR EACH PROBE
ENZYME COMBINATION; ISOLATE NAME AND VEGETATIVE COMPATIBILITY
GROUP INCLUDED AS A REPRESENTATIVE FOR EACH UNIQUE RESTRICTION
FRAGMENT LENGTH POLYMORPHISM HAPLOTYPE

VCG 0120-C2 Haplotype 21
x 10000/100/1 /1 /1 /1 /0001001 /100/1 /1 /01000/01000/1000/010000100/0100000/001000/100000/000010100/1000000
VCG 0120-F9127 Haplotype 18
10000/100/1/1/1 /1/0001001/100/1/1/01000/01000/1000/010000100/0110000/001000/100000/000010100/1000000
VCG 0120-PAJ1 Haplotype 22
10000/100/1/1/1/1/0001001/100/1/1/01000/01000/1000/010000100/0101100/001000/100000/000010100/1000000
VCG 0120-IC2 Haplotype 1
10000/100/1/1/1/1 /0001001/100/1/1/01000/01000/1000/010000100/0100100/001000/100000/000010100/1000000
VCG 0120-Pacovan Haplotype 23
10000/100/1/1/1/1/0001001/100/1/1/01000/01000/1000/010100100/0110100/001000/100000/000010100/1000000
VCG 0120-MGSA1 Haplotype 24
10000/100/1/1/1/1 /0001001/100/1/1/01000/01000/1000/010000100/0101000/001000/100000/000010100/1000000
VCG 0120-SA4 Haplotype 25
10000/100/1 /1 /1/1/0001001 /???/l/1 /01000/?????/1000/010000100/0000100/001000/100000/000010100/1000000
VCG 0120-SA3 Haplotype 26
10000/100/1/1/1/1 /0001001/100/1/1/01000/01000/1000/010000100/0110100/001000/100000/000010100/1000000
VCG 0121 -GM Haplotype 27
00100/100/1/1/1/1/0001001/010/1/1/01000/00100/1000/011010100/0100100/010000/000110/000100000/1000000
VCG 0121-F9130 Haplotype 28
10000/100/1/1/1/1/0011001/010/1/1/01000/10100/1000/010100100/0100100/100000/100000/000100000/1000100
VCG 0121-HI Haplotype 19
10000/100/1/1/1/1/0011001/010/1/1/01000/10000/1000/010000100/0100100/100000/100000/000100000/1000000
VCG 0121 0-1124 Haplotype 29
10000/100/1/1/1/1/0001001 /???/1/1/01000/10000/1000/010000100/0100100/100000/100000/000100000/1000000
VCG 0122-Ph3 Haplotype 10
10000/100/1/1/1/1/0010010/100/1/1/01000/01000/1000/010000100/0011100/001000/100000/00100000/1000000
VCG 0122-P79 Haplotype 30
10000/100/1/1/1/1/0010010/100/1/1/01000/01000/1000/010100100/0010000/001000/100000/000100000/1000000

VCG 0122-LAP Haplotype 3 1
10000/100/1/1/1/1/0010010/100/1/1/01000/01000/1000/010100100/0011010/001000/100000/000100000/1000000
VCG 0122-SABA Haplotype 20
10000/100/1/1/1/1/0010010/100/1/1/01000/01000/1000/010000100/0010000/001000/100000/000100000/1000000
VCG 0122-PW7 Haplotype 32
10000/100/1/1/1/1/0010010/100/1/1/01000/01000/1000/010000100/0010010/001000/100000/000100000/1000000
VCG 0123-DAVAO Haplotype 33
01000/100/1/1/1/1/0001001/100/1/1/01000/10100/0010/011000010/0110000/001000/010000/001100000/0101000
VCG 0123-JLTH4 Haplotype 36
01000/100/1/1/1/1/0100000/100/1/1/01000/10000/0110/010000010/0110000/001000/010000/000100000/0010000
VCG 0123-JLTH5 Haplotype 37
01000/100/1/1/1/1 /???????/100/1/1/01000/10000/0010/100000000/0110000/001000/010000/000100000/0100000
VCG 0123-T1 Haplotype 34
01000/100/1/1/1/1/0101001/100/1/1/01000/10000/0010/010000010/0110000/011000/010000/000100000/0100000
VCG 0123-PhL2 Haplotype 13
01000/100/1/1/1/1/0101001/100/1/1/01000/10000/0010/010000010/0110000/001000/010000/000100000/0100000
VCG 0123-Ph 12 Haplotype 35
00010/100/1/1/1/1/0101001/100/1/1/01000/10000/0010/011000010/0111000/001000/010000/001100000/0101000
VCG 0124-A3 6 Haplotype 2
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000101000/0010000
VCG 0124-GMB Haplotype 38
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0111000/001000/010000/000101000/0010000
VCG 0124-Maca Haplotype 9
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0110000/001000/010000/000101000/0010000
VCG 0124-STJ2 Haplotype 15
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0111000/001000/010000/000101000/0010000
VCG 0124-MW43 Haplotype 3
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000101000/0100000
VCG 0124-MW47 Haplotype 39
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/???????/001000/010000/000101000/0100000

VCG 0124-JLTH15 Haplotype 11
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0111000/001000/010000/000101000/0100100
VCG 0124-MW64 Haplotype 6
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000101000/0101000
VCG 0124-STN5 Haplotype 40
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0100000/001000/010000/000101000/0010000
VCG 0124-STN7 Haplotype 41
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0101000/001000/010000/000101000/0100000
VCG 0124/25-MW9 Haplotype 42
01000/011/1/1/1/1/0100000/100/1/1/00100/01000/1100/100000000/0111000/001000/010000/000101000/0100000
VCG 0124/25-MW11 Haplotype 7
01000/011/1/1/1/1/0100000/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000101000/0100000
VCG 0124/25-MW53 Haplotype 43
01000/011/1/1/1/1/0100000/100/1/1 /00100/01000/1100/100000000/0110000/001000/??????/000101000/0100000
VCG 0124/25-MW63 Haplotype 44
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0100000/001000/010000/000101000/0100000
VCG 0124/25-MW86 Haplotype 45
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0111000/001000/010000/000101000/0100100
VCG 0124/25-JLTH16 Haplotype 46
00001/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000101000/0100000
VCG 0124/25-JLTH17 Haplotype 47
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/????/100000000/0110000/001000/010000/000101000/1000000
VCG 0124/25-JLTH18 Haplotype 48
01000/011/1/1/1/1/0100100/100/ l/l/?????/01000/1100/100100000/0111000/001000/000100/000101000/0100100
VCG 0124/25-JLTH19 Haplotype 14
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/000100/000101000/0100000
VCG 0125-8606 Haplotype 17
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0110000/001000/010000/000101000/0101000
VCG 0125-22479 Haplotype 49
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0110000/001000/010000/000001011/0101000

VCG 0125-22541 Haplotype 50
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0111000/001000/010000/000101000/0100000
VCG 0126-STB2 Haplotype 52
10000/100/1/1/1/1/0101001/100/1/1/01000/01000/1000/010000100/0100000/001000/100000/000010100/0000100
VCG 0126-STA2 Haplotype 12
10000/100/1/1/1/1/0001001/100/1/1/01000/01000/1000/010000100/0100000/001000/100000/000010100/0000100
VCG 0126-SI Haplotype 51
10000/100/1/1/1/1/0101001/100/1/1/01000/01000/1000/010000100/0100100/001000/100000/000010100/0000100
VCG 0128-A47 Haplotype 53
01000/011/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0010000/001000/010000/000101000/???????
VCG 0129-N5443 Haplotype 54
10000/100/1/1/1/1/0001001/100/1/1/01000/01000/1000/010100100/0100100/001000/100000/000010100/1000000
VCG 01210-A2-1 Haplotype 4
10000/100/1/1/1/1/0001001/100/1/1/01000/10000/1000/010000100/0100100/001000/100000/000100000/0000100
VCG 01210-GG1 Haplotype 57
10000/100/1/1/1/1/0001001/100/1/1/01000/10000/1000/010100100/0100100/001000/100000/000100000/0000100
VCG 01210-A15 Haplotype 55
10000/100/1/1/1/1 /0001001/100/1/1/11000/10000/1000/010100100/0111100/010100/100000/000100000/0000100
VCG 01210-JC1 Haplotype 56
10000/100/1/1/1/1/0001001/100/1/1/01000/10000/1000/010100100/0110100/001000/100000/000100000/0000100
VCG 01211-SH3142 Haplotype 59
10000/100/1/1/1/1 /0000100/100/1/1/01000/10100/1000/010000100/0100100/001000/100000/000010100/1000000
VCG 01211-13721 Haplotype 58
10000/???/1/1/1/1/000100/100/1/1/01000/10100/1000/010000100/0100100/001000/100000/000100000/1000000
VCG 01212-STNP1 Haplotype 60
01000/100/1/1/1/1/0100100/100/1/1/00100/01000/1100/100001000/0010000/001000/010000/010000000/0100000
VCG 01212-STNP4 Haplotype 16
01000/100/1/1/1/1/0100100/100/1/1/00100/01000/1100/100000000/0010000/001000/010000/010000000/0100000
VCG 01213-6-2 Haplotype 61
10000/100/1/1/1/1/0101001/010/1/1/01000/10010/1000/010000100/0100100/100000/100000/000100000/1000000

VCG 01213-1-2 Haplotype 5
10000/100/1/1/1/1/0101001/010/1/1/01000/10000/1000/010000100/0100100/100000/100000/000100000/1000000
VCG 01213-4-1-1 Haplotype 62
10000/100/1/1/1/1/0101001/010/1/1/01000/10000/1000/010100100/0100100/100000/100000/000100000/1000000
VCG 01213-ES2-1 Haplotype 63
01000/100/1/1/1/1/0101001/100/1/1/01000/10000/?????/010000110/0100100/00100/010000/000100000/1000000
VCG 01214-MW2 Haplotype 64
10000/100/1/1/1/1/1000010/100/1/1/01000/10010/1000/010100010/0111000/010000/010000/100000000/0000100
VCG 01214-MW40 Haplotype 65
10000/100/1/1/1/1/1000010/100/1/1/01000/10100/1000/010100010/0110000/010000/010000/100000000/0000100
VCG 01214-MW41 Haplotype 66
10000/100/1/1/1/1/1000010/100/1/1/01000/10100/1000/010100010/0110000/010000/010000/110000000/0000100
VCG 01214-MW42 Haplotype 67
10000/100/1/1/1/1/1000010/100/1/1/01000/10010/1000/010100010/0111000/010100/010000/100100000/0000100
VCG 01214-MW44 Haplotype 68
10000/100/1/1/1/1/1000010/100/1/1/01000/10100/1000/010100010/0110000/010100/010000/100100000/0000100
VCG 01214-MW46 Haplotype 69
10000/100/1/1/1/1/1000010/100/1/1/01000/10010/1000/010000010/0110000/010000/010000/100000000/0000100
VCG 01214-MW48 Haplotype 70
10000/100/1/1/1/1/1000010/100/1/1/11000/10010/1000/010100010/0111000/010100/010000/100100000/0000100
VCG 01214-MW51 Haplotype 71
10000/100/1/1/1/1/1000010/100/1/1/01000/10010/1000/010100010/0110000/010000/010000/100010000/0000100
VCG 01214-MW89 Haplotype 72
10000/100/1/1/1/1/1000010/100/1/1/01000/10000/1000/010000010/0110000/010000/010000/100000000/0000010
VCG 0120/01215-INDO20 Haplotype 8
10000/100/1/1/1/1 /0001001/010/1/1/01000/10000/1000/010000100/0100100/100000/100000/000100000/1000000
SC626y
01000/100/1/1/1/1/0110000/00171/1 /00010/00101/1001/010100000/1100100/001001/100000/000100000/0000001
SC548
01000/100/1/1/1/1/0110000/001/1/1/00010/00101/1001/010101000/1001100/??????/100000/000100000/0000001

SC761
00010/100/1/1/1/1/0110000/001/1/1/00010/00101/1001/010100001/1001100/000011/100000/010100000/0000001
CS85-4
01000/100/1/1/1/1/0100000/100/1/1/01000/10010/1000/000100000/0100001/000010/010000/000100000/0100000
x The order of probe/enzyme input are:
1-Dra\, 1-Haelll, 7-EcoRV, 30-TJmI, 30-/teIII, 30-EcoRV, \20-Haelll \62-Dral, 162-EcoRV \62-Haelll, 177-EcoRV, 187-A'co//V,
204-/7aelII, 204-EcoRV, 225-EcoRV, 228-EcoRV, 260-Z)raI, 260-EcoRV, and 261 -EcoRV
y Isolates in bold signify the outgroups.
z Alleles in bold are unique to isolates representing the Fusarium oxysporum f. sp. niveum and F.oxysporum f. sp. lycopersici

VCGV
0120
APPENDIX D
LIST OF ISOLATES OF F. OXYSPORUMF. SP. CUBENSE,
THEIR VEGETATIVE COMPATIBILITY GROUP,
HAPLOTYPE AND LINEAGE
Isolate*
IC2
22425
ORT2
0-1220
GAL2
C2
ADJ2
Cl
22424
0-1222
0-1219
A2
ADJ1
STGM1
PAJ1
ORT1
GAL1
BUE1
NW
NH
NB
F9127
15638
FCJ7
Pacovan
MGSA1
SA6
SA4
SA3
Haplotvpe
1
1
1
1
1
21
1
18
1
1
1
1
1
1
22
1
1
1
1
1
1
18
1
1
23
24
1
25
26
Lineage
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
IP
II
100

VCG
Isolate
Haplotvpe
Lineage
0121
GM
27
y
9130
28
III
0-1124
29
IIP
HI
19
ill
ML
19
III
TBR
5
ill
0122
Ph3
10
VI
Ph6
20
VI
P79
30
VI
LAP
31
VI
SABA
20
VI
PW3
10
VI
PW6
10
VI
PW7
32
VI
0123
DAVAO
33
VII
T1
34
VII
PhL2
13
VII
Phl2
35
VII
9129
13
VII
JLTH4
36
X
JLTH5
37
X*
0124
A36
2
I
GMB
38
I
Maca
9
I
STPA1
15
I
STD2
2
I
BLUG
2
I
S?
2
I
FCJ2
2
I
FCJ3
9
I
FCJ8
2
I
FCJ9
3
I
STJ2
15
I
MW43
3
I
MW45
3
I
MW47
39
p
MW50
3
I
MW52
11
I
MW58
11
I
MW64
6
I
MW67
6
I
MW69
6
I
MW71
6
I

VCG
0124 (cont.)
0124/0125
0125
0126
Isolate
MW78
STN2
STN5
STN6
STN7
STPA2
B1
B2-1
JCB1
JLTH2
JLTH7
JLTH15
MW9
MW11
MW39
MW53
MW56
MW60
MW61
MW63
MW66
MW70
MW86
JLTH1
JLTH16
JLTH17
JLTH18
JLTH19
A4
8606
8611
22468
22479
22600
22417
22541
0-1223
IS?
STPA3
JLTH20
SI
STA2
STM3
STB2
Haplotvpe
2
2
40
2
41
2
2
2
2
7
2
11
42
7
3
43
7
7
->
44
3
3
45
2
46
47
48
14
2
17
3
17
49
2
J
50
3
2
9
14
51
12
12
52
Lineage
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
r
i
i
i
i
i
i
i
i
i
p
r
i
i
i
i
i
p
i
i
i
i
i
i
i
ii
ii
ii
ii

103
VCG
Isolate
HaolotvDe
Linease
0128
22993
3
I
22994
3
I
A47
53
I*
0129
N5331
1
II
0-1221
1
II
N5443
54
II
8627
1
II
22402
1
II
8604
1
II
01210
A2-1
4
IX
A4-1
4
IX
CSB
4
IX
JC14
4
IX
A15
55
IX?
A3-1
4
IX
JC8
4
IX
F2
4
IX
F3
4
IX
JC1
56
IX
GG1
57
IX
01211
13721
58
IXs
SH3142
59
IX
01212
STNP1
60
VIII
STNP2
16
VIII
STNP4
16
VIII
01213
1-1
5
III
1-2
5
III
6-2
61
III
5-1-1
5
III
4-2-1
5
III
4-1-1
62
III
2-2
5
III
ES2-1
63
y

104
VCG
Isolate
Hanlotvoe
Linea;
01214
MW2
64
V
MW40
65
V
MW41
66
V
MW42
67
V
MW44
68
V
MW46
69
V
MW48
70
V
MW51
71
V
MW89
72
V
01215
CR1
1
II
CR2
1
II
CR4
1
II
CR5
1
II
0120/01215
INDO20
8
III
IND0152
8
III
IND018Z
8
III
v Vegetative compatibility groups (VCGs) were assigned using nitrogen metabolism (nit)
mutants according to the protocols of Cove (1976) as modified by Puhalla (1985).
w Isolates in bold represent those reflected in the RFLP tree and RFLP haplotype data for
those isolates are in Appendix D.
x missing data made it impossible to determine coefficient of similarity but based on all other
data it is presumed to be in the lineage indicated.
y unique isolate that had no lineage affinity based on defined criteria, the lineage assigned is
based on coefficients of similarity to the most closely related isolates.
z isolates analyzed with this designation do not correspond to any isolates currently held in
the Homestead collection.

REFERENCES
Aloi, C., and Baayen, R.P. 1993. Examination of the relationships between vegetative
compatibility groups and races in Fusarium oxysporum f. sp. dianthi. Plant Pathology
42:839-850.
Anderson, J.A., and Kohn, L.M. 1995. Clonality in soilborne, plant-pathogenic fungi.
Annu. Rev. Phytopathol. 33:369-391.
Appel, D.J. and Gordon, T.R. 1994. Local and region variation in populations of
Fusarium oxysporum from agricultural field soils. Phytopathology 84:786-791.
Ashby, S. F. 1926. Panama disease of bananas in the Canaries and West Africa. Trop.
Ag. 3:8.
Bancroft, J. 1876. Report of the board appointed to enquire into the cause of disease
affecting livestock and plants, Queensland, 1876. Votes and Proceedings 1877 3:1011-
1138.
Beckmann, C. H. 1987. The nature of wilt diseases of plants. APS Press, St. Paul, MN.
Beckmann, C. H. 1990. Host responses to the pathogen, pp. 93-105. In: Fusarium wilt of
banana, R. C. Ploetz ed., APS Press. St. Paul, MN.
Bentley, S., Pegg, K.G., and Dale, J.L. 1995. Genetic variation among a world-wide
collection of isolates of Fusarium oxysporum f. sp. cúbense analyzed by RAPD-PCR
fingerprinting. Mycol. Res. 99: 1378-1384.
Boehm, E. W. A., Ploetz, R.C., Kistler, H.C. 1994. Statistical analysis of electrophoretic
karyotype variation among vegetative compatibility groups of Fusarium oxysporum f. sp.
cúbense. Mol. Plant-Microbe Interact. 7:196-207.
Brandes, E.W. 1919. Banana wilt. Phytopathology 9:339-389.
Bruns, T.D., White, T.J., and Taylor, J.W. 1991. Fungal molecular systematics. Annu.
Rev. Ecol. Syst. 22:525-64.
Buddenhagen, I.W. 1990. Banana breeding and Fusarium wilt, pp 107-113. In Fusarium
wilt of banana, R.C. Ploetz ed., APS Press. St. Paul, MN.
105

Burgress, L.W. 1981. General ecology of the Fusaria. pp. 225-235. In: Fusarium:
Diseases, biology and taxonomy. P.E. Nelson, T.A. Toussoun, and R.J. Cook, eds. The
Penn. State Univ. Press, University Park, PA.
Clark, A., and Lunigan, C.M.S. 1993. Prospects for estimating nucleotide divergence with
RAPDs. Mol. Biol. Evol. 10(5): 1096-1111.
Correll, J.C., Klittich, C.J.R., and Leslie, J.F. 1987. Nitrate nonutilizing mutants of
Fusarium oxysporum and their use in vegetative compatibility tests. Phytopathology
77:1640-1646.
Elias, K.S. and Schneider, R.W. 1991. Vegetative compatibility groups in Fusarium
oxysporum f. sp. Iycopersici. Phytopathology 81:159-162.
Elias, K.S., Zamir, D , Lichtman-Pleban, T., and Katan, T. 1993. Population structure of
Fusarium oxysporum f. sp. Iycopersici'. Restriction fragment length polymorphisms
provide genetic evidence that vegetative compatibility group is an indicator of
evolutionary origin. Mol. Plant-Microbe Interact. 6:565-572.
FAO. 1993. 1992 FAO production yearbook. FAO, Rome.
Farman, M. L., Taura, S., and Leong, S.A. 1996. The Magnaporthe griesea DNA
fingerprinting probe MGR586 contains the 3' end of an inverted repeat transposon. Mol.
Gen. Genet. 251:675-681.
Fernandez, D., Assigbetse, K.B., Dubois, M.P. and Geiger J.P. 1994. Molecular
characterization of races and vegetative compatibility groups in Fusarium oxysporum f.
sp. vasinfectum. Appl. Environ. Microbiol. 60:4039-4046.
Goodwin, S. B., Spielman, L.J., Matuszak, J.M., Bergeron, S.N., and Fry, W.E. 1992.
Clonal diversity and genetic differentiation of Phytophthora infestans populations in
northern and central Mexico. Phytopathology 82:955-961
Gordon, T.R., and Okamoto, D. 1992. Population structure and the relationship between
pathogenic and nonpathogenic strains of Fusarium oxysporum. Phytopathology 82: 73-77.
Gordon, T.R., Okamoto, D., and Milgroom, M.G. 1992. The structure and
interrelationship of fungal populations in native and cultivated soils. Molecular Ecology
1:241-249.
International Network for the Improvement of Banana and Plantain (INIBAP). 1993.
Annual report.
Jacobson, D.J. and Gordon, T.R. 1990a. Further investigations of vegetative compatibility
within Fusarium oxysporum f. sp. me/onis. Can. J. Bot. 68:1245-1248.

107
Jacobson, D.J. and Gordon, T.R. 1990b. Variability of mitochondrial DNA as an indicator
of relationships between populations of Fusarium oxysporum f. sp. melonis. Mycol. Res.
94:734-744.
Jacobson D.J. and Gordon, T.R. 1991. Fusarium oxysporum f. sp. melonis: A case study
of diversity within a forma specialis. Phytopathology 81: 1064-1067.
Jarrett, R. L., Gawel, N., Whittemore, A., and Sharrock, S. 1992. RFLP-based phytogeny
of Musa species in Papua New Guinea. Theor. Appl. Genet. 84:579-584.
Katan, T., Hadar, E., and Katan, J. 1989. Vegetative compatibility of Fusarium
oxysporum f sp. dianthi from carnation in Israel. Plant Pathology 38:376-381.
Kim, D.H., Martyn, R.D., and Magill, C.W. 1992. Restriction fragment length
polymorphism group and physical map of mitochondrial DNA from Fusarium oxysporum
f. sp. niveiim. Phytopathology 82: 346-353.
Kistler, H.C. 1997. Genetic diversity in the plant pathogenic fungus, Fusarium oxysporum.
Phytopathology 81:474-479.
Kistler, H.C., Bosland, P.W., Benny, U., Leong, S., and Williams, P.H. 1987. Relatedness
of strains of Fusarium oxysporum from crucifers measured by examination of
mitochondrial and ribosomal DNA. Phytopathology 77:1289-1293.
Kistler, H.C. and Leong, S.A. 1986. Linear plasmidlike DNA in the plant pathogenic
fungus Fusarium oxysporum f. sp. conglutinaos. J. Bacteriol. 167:587-593.
Kistler, H.C , and Momol, E.A. 1990. Molecular genetics of plant pathogenic Fusarium
oxysporum. pp. 49-53. In: Fusarium wilt ofbanana. R. C. Ploetz, ed. APS Press, St.
Paul, MN.
Kistler, H.C., Momol, E.A., and Benny, U. 1991. Repetitive genomic sequences for
determining relatedness among strains of Fusarium oxysporum. Phytopathology 81: 331-
336.
Koh, Y.J., Goodwin, S.B., Dyer, A. T., Cohen, B.A., Ogoshi, A., Sato, N. and Fry, W.E.
1994. Migrations and displacements of Phytophthora infestaos populations in east Asian
countries. Phytopathology 84:922-927.
Kohli, Y., Brunner, L.J., Yoell, H., Milgroom, M.G., Anderson, J.B, Morrall, R.A.A. and
Kohn, L.M. 1995. Clonal dispersal and spacial mixing in populations of the plant
pathogenic fungus, Sc/erotinia sclerotiorum. Molecular Ecology 4:69-77.
Kohli,Y., Morrall, R.A.A., Anderson, J.B., and Kohn, L.M. 1992. Local and trans-
Canadian clonal distribution of Sclerotinia sclerotiorum on canola. Phytopathology
82:875-880.

108
Kohn, L.M. 1995. The clonal dynamic in wild and agricultural plant-pathogen populations.
Can. J. Bot. 73(Suppl 1 )S 1231-S1240.
Kress, W.J. 1990. The phylogeny and classification of the Zingiberales. Ann Missouri
Bot. Gard. 77:698-721
Langdon, R. 1993. The banana as a key to early American and Polynesian history. Jour,
of Pacific History 28:15-35.
Leonard,K.J., Leath, S. 1990. Genetic diversity in field populations of Cochliobolus
carbonum on corn in North Carolina. Phytopathology 80: 1154-59.
Leslie, J.F. 1993. Vegetative compatibility in fungi. Amur Rev. Phytopathol. 31:127-151.
Leslie, J.F., and Spieth, P.T. 1985. A comparison of heterokaryosis and vegetative
incompatibility among varieties of Gibberellafujikuroi (Fusarium moniliforme). Exp.
Mycol. 9:39-47.
Levy, M., Correa-Victoria, F.J., Zeigler, R.S., Xu, S., and Hamer, J.E. 1993. Genetic
diversity of the rice blast fungus in a disease nursery in Colombia. Phytopathology
83:1427-1433.
Levy, M., Romao, J., Marchetti, M.A. and Hammer, J.E. 1991. DNA fingerprinting with a
dispersed repeated sequence resolves pathotype diversity in the rice blast fungus. Plant
Cell 3:95-102.
Manicom, B.Q., and Baayen, R.P. 1993. Restriction fragment length polymorphisms in
Fusarium oxysporum f. Sp. dianthi and other fusaria from Dianthus species. Plant Pathol.
42:851-857.
Manicom, B.Q., Bar-Joseph, M., Kotze, J. and Becker, M.M. 1990. A restriction
fragment length polymorphism probe relating vegetative compatibility groups and
pathogenicity in Fusarium oxysporum f. sp. dianthi. Phytopathology 80:336-339.
McDonald B.A. and Martinez, J.P., 1990. DNA restriction fragment length
polymorphisms among Mycosphaerel'la graminicola (anamorph Septoria tritici) isolates
collected from a single wheat field. Phytopathology 80:1368-1373.
McDonald, B.A., and Martinez, J.P. 1991. DNA fingerprinting of the plant pathogenic
fungus Mycosphaerel la graminicola (anamorph Septoria tritici). Exp. Mycol. 15:146-
158.
McDonald, B.A. and McDermott, J.M. 1993. Population genetics of plant pathogenic
fungi. BioScience 43:311-319.
McDonald, B.A., McDermott, J.M, Goodwin, S B., and Allard, R.W. 1989. The
population biology of host-pathogen interactions. Annu. Rev. Phytopathol. 27:77-94.

109
McDonald, B.A., Miles, J., Nelson, L.R. and Pettway, R.E. 1994. Genetic variability in
nuclear DNA in field populations of Stagonospora nodorum. Phytopathology 84:250-255.
McDonald, B.A., Pettway, R.E., Chen, R.S., Boeger, J.M., and Martinez, J.P. 1995. The
population genetics of Septoria tritici (teleomorph Mycosphaere I la graminicola). Can. J.
Bot. 73 Suppl. 1A-D: 5292-5301.
Mes, J.J., Van Doom, J., Roebroeck, E.J.A., Van Egmond, E., Van Aartrijk, J., and
Boonekamp, P.M. 1994. Restriction length polymorphisms, races and vegetative
compatibility groups within a worldwide collection of Fusarium oxysporum f. sp.
g/adioloi. Plant Pathol. 43:362-370.
Michelmore, R.W., and Hulbert, S.H. 1987. Molecular markers for genetic analysis of
phytopathogenic fungi. Annu. Rev. Phytopathol. 25:383-404.
Milgroom , M.G., and Fry, W.F. Contributions of population genetics to plant disease
epidemiology and management. Advances in Bot. Res. (in press).
Moore, N.Y., Hargreaves, P.A., Pegg, K.G., and Irwin, JAG. 1991. Characterization of
strains of Fusarium oxysporum f. sp. cúbense by production of volatiles. Australian Jour.
Bot. 39:161-166.
Moore, N.Y., Pegg, K.G., Langdon, P.W., Smith, M.K., and Whiley, A. W. 1993. Current
research on Fusarium wilt of banana in Australia, pp 270-284. In: Proc. Int. Symp. Recent
Dev. Banana Cultivation Technol. R. V. Valmayor, S.C. Hwang. R. Ploetz, S.W. Lee, and
V. N. Roa, eds. International Network for the Improvement of Banana and Plantain, Asia
and Pacific Network, Los Banos, Laguna, Philippines.
Murdock, G.P. (1959). Africa, its peoples and their cultural history. McGraw-Hill, NY.
Namiki,F., Shiomi,T., Kayamura,T. and Tsuge,T. 1994. Characterization of the formae
speciales of Fusarium oxysporum causing wilts of cucurbits by DNA fingerprinting with
nuclear repetitive DNA sequences. Appl. Environ. Microbiol. 60:2684-2691.
Ploetz, R.C. 1990. Population biology oí Fusarium oxysporum f. sp. cúbense, pp. 63-76.
In: Fusarium wilt of banana, R.C. Ploetz,ed. APS Press, St. Paul, MN.
Ploetz, R.C. 1992. Fusarium wilt of banana (Panama disease), pp. 270-282 In: Plant
diseases of international importance, volume III. Prentice Hall, Englewood Cliffs, NJ.
Ploetz, R.C. 1994. Panama disease: Return of the first banana menace. Inter. Jour, of Pest
Mgmt. 40:326-336.
Ploetz, R.C. 1995. Fusarium wilt and phase II of IMPT. pp. 57-69 In: Jones, D.R. (ed.).
The Improvement and testing of Musa: A global partnership. INIB AP, Montpellier,
France.

110
Ploetz, R.C., Braunworth, W.S. Jr., Hasty, S., Gantotti, B., Chizala, C.T., Banda, D.L.N.,
Makina, D.W., and, Channer, A.G 1992. Fusarium wilt of banana (Panama disease) in
Malawi. Fruits 47:503-508.
Ploetz, R.C., Channer, A.G., Chizala, C.T., Banda, D.L.N., Makina, D.W., and
Braunworth, W.S., Jr. 1992. A current appraisal of banana and plantain diseases in
Malalwi. Trop. Pest Mgmt. 38: 36-42.
Ploetz, R.C. and Correll, J.C. 1988. Vegetative compatibility among races of Fusarium
oxysporum f. sp. cúbense. Plant Dis. 72:325-328.
Ploetz, R.C., Herbert, J., Sebasigari, K., Hernandez, J.H., Pegg, K.G., Ventura, J.A., and
Mayato, L.S. 1990. Importance of Fusarium wilt in different banana-growing regions. In:
Fusarium wilt of banana. R. C. Ploetz ed. APS Press, St. Paul, MN.
Ploetz, R.C., Jones, D.R., Sebasigari, K., and Tushemerirewe, W. 1995. Fusarium wilt
(Panama disease) on East African highland cultivars of banana. Fruits 49:253-260.
Ploetz, R.C., Vázquez, A., Nagel, J., Benscher, D , Sianglew, P., Srikul, S., Kooariyakul,
S., Wattanachaiyingcharoen, W., Lertrat, P., and Wattanachaiyingcharoen, D. 1997.
Current status of Panama disease in Thailand. Fruits (in press).
Puhalla and Speith, 1985. A comparision of heterokaryosis and vegetative incompatibility
among varieties of Gibberellafujikuroi (Fusarium moniliforme). Exp. Mycol. 9:39-47.
Rosewich, U.L. and McDonald, B.A. 1994. DNA fingerprinting in fungi. Meth. in Mol.
Cell. Biol. 5:41-48.
Sambrook, J., Fritsch, E.F. and Maniatis, T. 1989. Molecular cloning. A laboratory
manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Scotland, R.W. 1992. Cladistic theory. In: Cladistics a practical course in systematics.
Forey, P.L., Humphries, C.J., Kitching, I.L., Scotland, R.W., Siebert, D.J., and Williams,
D.M. eds. Clarendon press, Oxford, England.
Shivas, R.G., and Shilemon, E. 1996. First record oí Fusarium oxysporum f. sp. cúbense
on bananas in Papua New Guinea. Australasian Plant Pathol. 25:260.
Simmonds, N.W. 1962. The evolution of the bananas. Longmans, London
Simmonds, N.W., and Shepherd, K. 1995. The taxonomy and origins of the cultivated
bananas. J. Linn. Soc. (Bot.) 55:302-313.
Sneath, P.H. and Sokal, R.R. 1973. Numerical taxonomy. W. H. Freeman, San Francisco,
CA.

Snyder, W.C. 1981. Introduction, pp. 308 In: Fusarium: Diseases, biology and taxonomy.
PE. Nelson, T.A. Toussoun, and R.J. Cook, eds. The Penn. State Univ. Press, University
Park, PA.
Snyder, W.C. and Hansen, H. N. 1940. The species concept in Fusarium. Am. J. Bot.
27:64-67.
Sorensen, S., Pegg, K.G., and Dale, J.L. 1993. RAPD-PCR analysis of genetic variation
within Australian populations of Fusarium oxyspontm f. sp. cúbense, pp. 285-295 In:
Proc. Int. Symp. Recent Dev. Banana Cultivation Technol. R. V. Valmayor, S.C. Hwang.
R. Ploetz, S.W. Lee, and V. N. Roa, eds. International Network for the Improvement of
Banana and Plantain, Asia and Pacific Network, Los Banos, Laguna, Philippines.
Stover, R. H. 1959. Studies on Fusarium wilt of banana. IV. Clonal differentiation among
wild type isolates of Fusarium oxysporum f. sp. cúbense. Can. J. Bot. 37:245-255.
Stover, R.H. 1962. Fusarial wilt (Panama Disease) of bananas and other Musa species.
Phytopathol. Pap. No. 4. CMI.
Stover, R.H. and Buddenhagen, I.W. 1986. Banana breeding: Polyploidy, disease
resistance and productivity. Fruits 41:175-191.
Su, H.J., Chuang, T.Y. and Kong, W.S. 1977. Physiological race of fusarial wilt fungus
attacking Cavendish banana in Taiwan. Taiwan Banana Res. Inst. Spec. Publ. 2. 21 pp.
Sujkowski, L.S., Goodwin, S.B., Dyer, AT, and Fry, W.E. 1994. Increased genotypic
diversity via migration or possible occurrence of sexual reproduction of Phytophthora
infestans in Poland. Phytopathology 84:201-207.
Swofford, D.L. 1993. PAUP. Phylogenetic analysis using parsimony. Version 3.1. Illinois
Natural History Survey. Champaign, IL.
Tantaoui, A., Ouinten, M., Geiger, J. P. and Fernandez, D. 1996. Characterization of a
single clonal lineage causing Bayoud disease of date palm in Morocco. Phytopathology
86:787-792.
Tibayrenc, M., Kjellberg, F. and Ayala, F.J. 1990. A clonal theory of parasitic protozoa:
The population structures of Entamoeba, Giardia, Leishmania, Naegleria, Plasmodium,
Trichomonas and Trypanosoma and their medical and taxonomical consequences. Proc.
Natl. Acad. Sci. USA 87:2414-2418.
Weir, B.S. 1990. Genetic data analysis: Methods for discrete population genetic data.
Sinauer Associates, Sunderland, MA.
Whisson, S.C., Maclean, D.J., Manners, J.M., and Irwin, J.A.G. 1992. Genetic
relationships among Australian and North American isolates of Phytophthora megasperma
f. sp. glycinea assessed by multicopy DNA probes. Phytopathology 82:863-868.

112
White, T.J., Bruns, T., Lee, S., and Talyor, J. 1990. Amplification and direct sequencing
of fungal ribosomal RNA genes for phylogenetics, pp. 315-322. In: PCR protocols: A
guide to methods and applications. Academic Press, San Diego, CA.
Whitehead, D.S., Coddington, A. and Lewis, B.G. 1992. Classification of races by DNA
polymorphisms analysis and vegetative compatibility grouping in Fuscirium oxysporum f.
sp. pisi. Physiol. Mol. Plant Pathol. 41:295-305.
Windels, C. 1991. Current status of Fusariwv taxonomy. Phytopathology 81:1048-1051.
Wollenweber, H.W., and Reinking, O. A. 1935. Die Fusarien, ihre Beschreibung,
Schadwirkung und Bekampfung. Paul Parey, Berlin.

BIOGRAPHICAL SKETCH
Rosalie Lynn Koenig was born in Freehold, New Jersey, January 12, 1962, to
Milton and Lucille Koenig. She was raised on her family’s chicken and vegetable farm.
She attended Cook College of Rutgers University and received a Bachelor of Science
degree in agricultural science in May, 1984. After working for a year as an assistant plant
breeder for Northrup King seed company, she began a graduate program in international
agricultural development with an emphasis on plant breeding and genetics at the
University of California, Davis. She received a Master of Science degree in May 1989.
She then took a short-term position, which was partially funded by the United States
Agency for International Development (USAID), working as an agronomist intern for
Rodale International in Thies, Senegal. She entered the doctorate program in plant
pathology at the University of Florida in 1990. She married her husband Tom Mirti in
1991. In 1995, her daugther Amaleah was born. Upon completion of the Doctor of
Philosophy degree, she will continue her role as a mother and organic vegetable grower
while pursue writing grants for research studies in the areas of international and
sustainable agriculture. Rosalie maintains interests in people and in interesting places and
experiences.
113

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
rfC. /«¡t
Harold C. Kistler, Chair
Associate Professor of Plant Pathology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Randy C. Ploetz, Cochair Q
Professor of Plant Pathology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
David J. Mitdnell
Professor of Plant Pathology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
‘^T’Eduardo Vallejos
Associate Professor of Horticultural Science
This dissertation was submitted to the Graduate Faculty of the College of
Agriculture and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
May 1997
Dean, Graduate School