l: .

#333.

3 .w
.,

b .
.3 ..
(#1 . w . ,
.r. Mm .n .3”
Winn-.3...

an. s
w»
”it:

1: f; .

.x : «I
511......)

55..
a. cunt ‘

i .w. . . . ma. ,
.2 1.9: . . JéWJ—I
.

. 1,... n. .,
.flwwfim . 1. . u.

A ii
a.

a .

.t.
.

Wmumwwfi
n .,

now“

4
2:13).

v x.
1.7+ "y' 5
.cmfiu #4

,, éfiyw

5.:

u: .u
I‘d}

 

Iv

J

This is to certify that the
thesis entitled

DEVELOPING IMPROVED BUSH BEAN VARIETIES IN
ECUADORIAN MARKET CLASSES USING FARMER
PARTICIPATORY CROP IMPROVEMENT METHODS AND
MARKER ASSISTED SELECTION OF AN ANTHRACNOSE
RESISTANCE GENE

presented by

Emmalea Garver Ernest

has been accepted towards fulfillment
of the requirements for the

MS. degree in Plant Breeding and Genetics

 

 

,2, J M/Z/

 

 

/ Major Professor’ s Signature

02’/?' 06’

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

& pow,“- k”

6/01 c-JCIRC/DateDuest-p. 15

 

DEVELOPING IMPROVED BUSH BEAN VARIETIES IN ECUADORIAN MARKET
CLASSES USING FARMER PARTICIPATORY CROP IMPROVEMENT METHODS
AND MARKER ASSISTED SELECTION OF AN ANTHRACNOSE RESISTANCE
GENE

By

Emmalea Garver Ernest

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Plant Breeding and Genetics Program
Department of Crop and Soil Sciences

2004

ABSTRACT
DEVELOPING IMPROVED BUSH BEAN VARIETIES IN ECUADORIAN MARKET
CLASSES USING FARMER PARTICIPATORY CROP IMPROVEMENT METHODS
AND MARKER ASSISTED SELECTION OF AN ANTHRACNOSE RESISTANCE
GENE
By
Emmalea Garver Ernest

Bush beans are an important cash and subsistence crop in the Mira and Chota
River Valleys of northern Ecuador. This project was undertaken to introduce genetic
resistance to anthracnose into bean market classes grown in this region, and to determine
the bean production practices, problems and variety preferences of bean producers in the
Mira and Chota Valleys.

C0-42, a gene conferring broad-based resistance to anthracnose (Colletotrichum
lindemuthz’anum) was introduced into bush bean varieties from three Ecuadorian
commercial classes using marker assisted backcrossing. Anthracnose resistant lines were
developed in two of the market classes but the C0-42 gene failed to function in the genetic
background of two of the Ecuadorian varieties used as recurrent parents. Verification of
disease reaction through direct inoculation is essential to support the use of marker
assisted selection in resistance breeding.

Farmer surveys revealed that bean producers in northern Ecuador are switching to
bean types with a domestic market after losing the Colombian bean export market. Major
production constraints are bean rust, whitefly, bean weevil, anthracnose, and root
attacking pests and diseases. Few bean growers maintain their own seed but rather rely

on the local grain markets as a source of bean seed. This system of seed management

may be incompatible with the introduction of improved bean varieties in Ecuador.

 

ACKNOWLEDGEMENTS

I gratefully acknowledge the counsel and guidance of my advisor Dr. James D.
Kelly and the advice of my committee members Dr. Anne Ferguson and Dr. Russell
Freed. My thanks, as well, to Dr. Diane Ruonavaara, and Dr. Rick Bemsten for their help
as I developed the survey questionnaire. Dr. Mark Bassett graciously provided
germplasm and advice regarding the experiments with ‘Red Hawk’ and ‘Redcoat.’ The
Bean/Cowpea CRSP provided the funding for this project.

I am also indebted to the group from National Legume Program and many other
individuals at INIAP’s Santa Catalina research station, who provided assistance during
my time in Ecuador. To the bean producers of the Mira and Chota Valleys, my thanks
for your hospitality and candid answers. I also thank everyone from the dry bean lab at
MSU for their friendship and advice

Finally, my thanks to my family and to my husband, Jeremy, for their

encouragement.

iii

TABLE OF CONTENTS

 

 

 

 

 

List of Tables - .............. vi
List of Figures - viii
Key to Abbreviations ix
Chapter 1 Improving Anthracnose Resistance of Ecuadorian Beans 1
Introduction ..................................................................................................................... l
Beans in Ecuador ........................................................................................................ 1
Colletotrichum lindemuthianum ................................................................................. 4
Control of Bean Anthracnose ...................................................................................... 7
Genetic Resistance to Bean Anthracnose ................................................................... 8
Anthracnose in Ecuador ............................................................................................ 13
Materials and Methods .................................................................................................. 15
Use of Molecular Markers ........................................................................................ 15
Introduction of Co-42 into Ecuadorian Varieties Through Marker Assisted
Backcrossing ............................................................................................................. 16
Marker Assisted Selection in Ecuador ...................................................................... 20
Field Inoculation of Backcross Populations ............................................................. 20
Evaluation of Seed From Field Grown Populations ................................................. 24
Greenhouse Inoculations of Backcross Populations ................................................. 25
Results and Discussion ................................................................................................. 27
Flowering in Field Grown Populations ..................................................................... 27
Reaction to Field and Greenhouse Inoculations ....................................................... 27
Disease Reaction ....................................................................................................... 37
Evaluation of Seed Appearance ................................................................................ 42
Legend ....................................................................................................................... 52
Conclusions ................................................................................................................... 54
Literature Cited ....................................................................................................... - ...... 57
Chapter 2 Using Participatory Crop Improvement Methods in the Mira and Chota
Valleys of Ecuador 63
Introduction ................................................................................................................... 63
Participatory Crop Improvement .............................................................................. 63
Farmer vs Breeder Selection Criteria ........................................................................ 64
On—Farm vs. Research Station Selection Environment ............................................. 66
Implementation of PCI .............................................................................................. 69
Timing of Farmer Involvement ................................................................................. 70
Farmer Recruitment .................................................................................................. 71
Bean Variety Improvement for Northern Ecuador ................................................... 73
Geography of the Mira and Chota River Valleys ..................................................... 74
Study of Production, Post Harvest and Marketing Practices of Bean Producers in the
Chota Valley, 2000 ................................................................................................... 74
Participatory Rural Appraisal of Seven Communities in the Parish of La
Concepcion, 2002 ..................................................................................................... 77
A Brief Comparison of the Two Valleys .................................................................. 79
Materials and Methods .................................................................................................. 80

Farmer Survey ........................................................................................................... 80

 

Results ........................................................................................................................... 81
Age, Education and Family Size of the Bean Producers Interviewed ...................... 81
Land Ownership and Crops Cultivated by Farmers .................................................. 82
The Bean Classes Grown by F anners ....................................................................... 84
Farmers Seed Sources ............................................................................................... 98
Pest and Disease Problems Reported by Farmers ..................................................... 99
Pesticide Use ........................................................................................................... 108
Farmers’ Suggestions for Bean Variety Improvement ........................................... 109
Introduction of New Bean Varieties ....................................................................... 110

Discussion ................................................................................................................... 1 12
Farmer Demographics and Agricultural Systems ................................................... 112
Farmers’ Variety Choices in the Mira and Chota Valleys ...................................... 113
Prices and Marketing .............................................................................................. 114
Farmers’ Seed Sources ............................................................................................ 115
Farmers’ Production Problems in the Mira and Chota Valleys .............................. 118
Pesticide Use Among Bean Producers in the Mira and Chota Valleys .................. 124
Farmers’ Recommendations for Variety Improvement .......................................... 126

Conclusions ................................................................................................................. 128

Literature Cited ........................................................................................................... 13 1

Appendices - - -- 134

Appendix A: Results of Greenhouse Inoculations of Ecuadorian Recurrent Parents. 135

Appendix B: Questionnaire Used in Farmer Survey .................................................. 136

Appendix C. List of Pesticides Farmers Reported Using and their Ecuadorian Trade

Names ......................................................................................................................... 144

Appendix D. Characterization of Redcoat, a Virgarcus Seed Coat Pattern Mutant of

Red Hawk Dark Red Kidney Bean ............................................................................. 146

LIST OF TABLES

Table 1. Anthracnose differential cultivars, their gene pool, binary number and resistance
genes ........................................................................................................................... 6
Table 2. Anthracnose resistance genes and alleles, their previous names, differential
cultivars containing them, their gene pool of origin, other sources and linked
markers ...................................................................................................................... 1 0
Table 3. Parents used in crossing ...................................................................................... 17
Table 4: Planting dates for lines grown as the pollen parent for crosses, number of seeds
planted, number of plants selected for crossing, molecular marker used for selection,

and varieties with which the selected plants were crossed. ...................................... 18
Table 5: Foliage Symptom Ratings for Anthracnose* ..................................................... 23
Table 6: Pod Symptom Ratings for Anthracnosei' .......................................................... 23
Table 7. Samples used in farmer evaluations of seed quality .......................................... 25
Table 8. Days to flowering for the backcross populations an recurrent parents .............. 27
Table 9. Cocacho BC1F2, BCze and BC3F2 Marker Analysis and Reaction to Ecuadorian

Race 0 Isolate ............................................................................................................ 30
Table 10. Cocacho BC 1F 2, BCZFZ and BC3F2 Test for Dependence of Marker and

Disease Reaction ....................................................................................................... 31

Table 11. Determination of intermediate resistance or susceptibility in Paragachi
backcross and parent populations based on pod and leaf symptoms (frequency,

percent of population) ............................................................................................... 36
Table 12. Paragachi BCle, BCze and BC3F2 Marker Analysis and Reaction to
Ecuadorian Race 0 Isolate ......................................................................................... 37
Table 13. Paragachi BCle, BC2F2 and BC3F2 Test for Dependence of Marker and
Disease Reaction ....................................................................................................... 37
Table 14. ARME-2 BCze and BC3F2 Marker Analysis and Reaction to Race 2 ........... 39
Table 15. ARME-2 BCze and BC 3F 2 Test for Dependence of Marker and Disease
Reaction .................................................................................................................... 40
Table 16. Yunguilla BCze and BC3F2 Marker Analysis and Reaction to Race 2 .......... 40
Table 17. Yunguilla BC2F2 and BC3F2 Test for Dependence of Marker and Disease
Reaction ........... 41
Table 18. Seed appearance frequencies in the Cocacho populations ............................... 43
Table 19. Seed appearance frequencies in the ACE-1 populations ................................. 44
Table 20. Seed appearance frequencies in the Paragachi populations ............................ 45
Table 21. Seed appearance frequencies in the ARME-2 populations .............................. 46
Table 22. Seed appearance frequencies in the Yunguilla populations ............................. 47
Table 23. Age distribution of farmers interviewed .......................................................... 81
Table 24. Level of schooling completed by farmers interviewed .................................... 81
Table 25. Family size of farmers interviewed ................................................................. 82
Table 26. Crops grown by farmers interviewed ............................................................... 83
Table 27. Area farmers interviewed planted to beans ...................................................... 83
Table 28. Number of classes of beans farmers reported growing .................................... 84
Table 29. XZ-test for independence between landholding size and seed saving .............. 98
Table 30. Farmers’ names for production problems and number of farmers using each
name ........................................................................................................................ 100

vi

Table 31. Active ingredients (A.I.) in pesticide farmers used to control bean production

problems, an problems treated ................................................................................ 102
Table 32. Classification of farmers by most hazardous pesticide level used on bean 109
Table 33 . Desirable bean variety characteristics identified by farmers ........................ 110
Table A 1. Reaction of Ecuadorian recurrent parents to greenhouse inoculation with

various isolates of Colletotrichum lindemuthianum ............................................... 135

Table C 1. F ungicide products farmers reported using .................................................. 144

Table C 2. Insecticide products farmers reported using ................................................. 145
Table D 1. Crosses made between Redcoat and Red Hawk and the seed coat pattern tester

lines ......................................................................................................................... 148

Table D 2. Test of fit to a ratio of 3 self-colored : 1 partly-colored in the Redcoat/Red
Hawk, Red Hawk/Virgarcus, and Red Hawk/bipunctata F2 generations X2 Goodness
of Fit Test ............................................................................................... ' ................. 150

Table D 3. Flower Color in the Red Hawk and Redcoat by Seed Coat Pattern Tester F2
Populations and Inferred Genotypes for Red Hawk and Redcoat .......................... 151

Table D 4. Seed Coat Pattern Frequency in the Red Hawk and Redcoat/Seed Coat Pattern
Tester F2 Populations .............................................................................................. 152

vii

LIST OF FIGURES

Figure 1. Layout of field planted on the Santa Rosa Substation ....................................... 22
Figure 2. Seed samples farmers rated in survey .............................................................. 24
Figure 3. Average leaf symptom ratings for Cocacho backcross populations and parent 28
Figure 4. Determination of resistance or susceptibility in Cocacho backcross and parent
plants inoculated with race 0 isolate based on pod and leaf symptoms .................... 29
Figure 5. Average leaf symptom ratings for ACE-1 backcross populations and parent... 32
Figure 6. Determination of resistance or susceptibility in ACE-1 backross and parent

plants inoculated with race 0 based on pod and leaf symptoms ............................... 33
Figure 7. Average leaf symptom ratings for Paragachi backcross populations and parent

................................................................................................................................... 34
Figure 8. Seed of Cocacho parent (1) and BC2 and BC; plants ....................................... 48
Figure 9. Seed of ACE-1 parent ( 1) and BC2 and BC; plants .......................................... 48
Figure 10. Seed of Paragachi BC2 and BC; plants ............................................................ 49
Figure 11. Seed of ARME-Z parent (1) and BC2 and BC; plants ..................................... 49
Figure 12. Seed of Yunguilla parent (1) and BC2 and BC; plants .................................... 50
Figure 13. Seed of Genuine parent (1) and BC2 (2) and BC; plants (3) ........................... 50

Figure 14. Farmers’ ranking of seed samples from the Cocacho, ACE-1, and Yunguilla
backcross populations and their respective Ecuadorian recurrent parent using

Bradley-Terry modeling ............................................................................................ 52
Figure 15. Amount of each bean class sown by the farmers interviewed and number of

farmers growing each class ....................................................................................... 85
Figure 16. Maximum, minimum and mean prices for bean classes .................................. 86
Figure 17. Number of years farmers had grown bean class .............................................. 87
Figure 18. Farmer reported incidence and severity of bean pests and diseases .............. 101
Figure B 1. Spanish language questionnaire as it was used in farmer surveys ............... 136
Figure B 2. English translation of questionnaire used in farmer surveys ....................... 140
Figure D 1. Seed of Red Hawk and Redcoat .................................................................. 146
Figure D 2. Seed coat patterns and genotypes described by Bassett and McClean (2000)

................................................................................................................................. 148

Figure D 3. Seed coat patterns observed in the Red Hawk and Redcoat by seed coat
pattern tester F2 populations .................................................................................... 152

viii

KEY TO ABBREVIATIONS

CRSP Collaborative Research Support Program

CIAL Comité de Investigacion Agricola Local, Local Agricultural Research Committee
CIAT International Center for Tropical Agriculture

INIAP Instituto Nacional Autonomo de Investigaciones Agropecuarias

PCI Participatory Crop Improvement

PCR Polymerase Chain Reaction

PRONALEG-GA Programa Nacional de Leguminosas y Granos Andinos
PVS Participatory Varietal Selection

QTL Quantitative Trait Loci

RAPD Random Amplified Polymorphic DNA

RIL Recombinant Inbred Line

SCAR Sequence Characterized Amplified Region

WHO World Health Organization

ix

CHAPTER 1
IMPROVING ANTHRACNOSE RESISTANCE OF ECUADORIAN BEANS

Introduction

Beans in Ecuador

Production and Consumption

Common bean (Phaseolus vulgaris) is the most cultivated and most consumed
legume crop in Ecuador. Ecuadorians consume not only dry beans but also the immature
seed (tierno) and pods (snap bean). Although the Ecuador’s national average per capita
bean consumption rate of 6 kg/year is one of the lowest in Latin America (Broughton et
al, 2003), consumption rates are believed to be much higher in certain regions of the
country. According to Ecuador’s Third National Agricultural Census, about 122,000 ha
of beans for consumption dry or in tierno are planted every year (INEC-MAG-SICA,
2002). Climbing beans intercropped with maize constitute the majority of this area.
Monocropped beans, principally bush beans with determinant or upright indeterminate
growth habits, occupy 30% of the area planted to beans consumed in tierno and only 18%
of the area planted to dry bean. However, bush beans account for 63% of the tierno
production and 47% of the dry bean production. Additionally, the majority of the beans
sold in Ecuador are bush beans (INEC-MAG-SICA, 2002).

The bulk of Ecuador’s bean production area is in the highlands (80%), with
smaller areas on the coast (18%), and in the eastern lowlands (2%) (Peralta et al., 1997).
In the highlands, climbing beans are typically grown at 2400 to 2800 m.a.s.l., while the

bush types are grown at lower elevations between 1500 and 2400 m.a.s.l. (Voysest,

2000). The focus of this project is the highland bush bean production area in the

Northern Ecuadorian provinces of Imbabura and Carchi.

Bean Classes Grown in the Ecuadorian Highlands

Morphological (Singh et al., 1991) and genetic differences (Nodari et al., 1992)
between domesticated beans from Central America and Mexico and those from South
America indicate that there are two distinct bean gene pools and sites of domestication in
Phaseolus vulgaris. Andean beans are typically larger-seeded and some possess a
determinant growth habit, which is infrequent in beans of Middle American origin.
Despite possessing a wide variation in traits such as seed color and protein type, Andean
beans have been shown to be less genetically diverse than their Middle American
counterparts. This narrow genetic base may account for the difficulty breeders have
encountered in improving traits, such as yield, in Andean beans (Beebe et al., 2001).
Most of the bean classes cultivated in Ecuador are from the Andean bean gene pool.
However, beans of Middle American origin have been grown in South America since
Pre-Colombian times, and some landraces grown in Colombia and Northern Ecuador
Show evidence of gene introgression from Middle American beans (Beebe et al., 2000;
Beebe et al., 2001). Such landraces are an important genetic resource and potential
source of unique variability within the Andean gene pool (Beebe et al., 2001)

According to Voysest (2000) the predominant commercial classes in the highland
regions of Ecuador are large-seeded red mottle types (45-55 g/l 00 seeds), cranberry (45-
55 g/ 100 seeds), and the round-seeded bola types (60-70 g/ 100 seeds) in a range of colors
from yellow and red to white. Red mottle seed types have historically been grown for

export to Colombia, while the other classes are marketed domestically. Both bush (type I

 

and 11 growth habits) and climbing varieties (type IV growth habits) of these commercial
classes are cultivated (Singh, 1982). Other Ecuadorian bean classes reported by Voysest
(2000) include panamito, a small white bean, which may be Middle American in origin;
amarillo matahambre, a yellow seeded bean; and vaquita which includes a variety of

different kinds of Spotted beans that are grown primarily for home consumption.

Bean Production Problems in Ecuador

Anthracnose (C olletotrichum lindemuthianum) and rust (Uromyces
appendiculatus) are the two bean diseases which have been most researched and
characterized in Ecuador. Rust can result in severe yield losses if plants are infected
before or during flowering. The disease is transmitted by wind-blown spores but not
through seed. Rust can be controlled with fungicide applications made early in plant
development, or through the use of resistant varieties (Stavely and Pastor-Corrales,
1989). A survey of rust diversity in Ecuador detected 17 different populations of the
pathogen, and found that all of the Ecuadorian bean varieties tested were susceptible to at
least one of the collected rust populations (Ochoa et al., 2002). Work to develop rust
resistant bean varieties in Ecuadorian market classes is underway at INIAP (Murrillo,
2002)

Other bean diseases which have been reported in Ecuador include: ascochyta
blight (Ascochyta sp.), halo blight (Pseudomonas syringae pv phaseolicola), powdery
mildew (Erysiphe polygom'), angular leaf spot (Phaeoisariopsis griseola), root rot
(F usarium solani, Rhizoctonia solani) and bean common mosaic virus (BCMV) (Peralta

et al., 1997; van Schoonhoven and Voysest, 1989).

 

Insect pests are more problematic in bush beans than in climbing beans. Bean
pests that have been reported in Ecuador include: greenhouse whitefly (T rialeurodes
vaporariorum), Spider mite (T etranychus Sp. and Polyphagotasonemus latus), leaflropper
(Empoasca kraemeri), pod borer (Epinotia aporema and Laspeyresia leguminis), and
bean weevil (Acanthosalides obtectus) (Peralta et al, 1997). Whiteflies and Spider mites,
which also affect other crops, such as tomato, feed on the leaves of bean plants. Large
populations of the pests can cause yield losses by weakening the plant. However, the
whitefly Species reported, T raleurodes vaporariorum, does not transmit Bean Common
Mosaic Virus. Leafhoppers can be especially damaging and may cause severe yield
losses. Their feeding stunts the plant and causes leaf curling. Pod borers can have a
direct effect on yield by damaging developing seeds and allowing secondary rot
pathogens infect the pod. Bean weevil larvae feed on stored grain and can cause
substantial post-harvest losses (Schwartz et al., 1978).

The Bean/Cowpea Collaborative Research Support Program (CRSP) has funded
an initiative to improve bean yield and yield stability in Ecuador. The project involves
the bean breeding programs at Michigan State University and at Ecuador’s National
Agricultural Research Institution (INIAP). One goal of this project is to develop bean
varieties possessing improved genetic resistance to anthracnose as well as the agronomic
and seed quality traits that Ecuadorian farmers desire.

Colletotrichum lindemuthianum

Anthracnose of bean is caused by the fungus Colletotrichum lindemuthianum

(Sacc. & Magnus) Briosi & Cav. The disease has a world-wide distribution, although it

tends to be more problematic in temperate and subtropical climates since high humidity

and moderate temperatures of between 13 and 26 C favor disease development (Pastor-
Corrales and Tu, 1989).

Symptoms of the disease in bean include dark brown or black lesions on the
leaves, pods, stems and hypocotyl. Veins of infected leaves often become darkened and
necrotic. Conidia are produced in larger lesions and serve as a source of secondary
inoculum that may be spread by wind and rain (Tu, 1992). Mycelia of the fungus can
also infect seeds causing discoloration and cankers. Moreover, anthracnose is transmitted
efficiently by pathogen infected seed, which is a source of inoculum if planted the
following year (Pastor-Corrales and Tu, 1989).

The infection process of C. lindemuthianum begins with the germination of
conidia in free water on the plant surface. The fungus then penetrates the cuticle and
epidermal cells via appressoria. The infection peg enlarges between the plant cell wall
and the plasma membrane and primary hyphae from the infection peg ramify
intracellularly through the plant tissue. Within a few days, lateral branches of the
primary hyphae begin to grow into the host’s cells, marking the beginning of the
destructive, necrotrophic phase (O’Connell and Bailey, 1991). Eventually, acervulli
bearing conidia form in the water-soaked lesion infected by the fungus (Pastor-Corrales
and Tu, 1989).

C. lindemuthianum reproduces asexually through conidia -- the sexual phase of
the fungus has very rarely been observed (Pastor-Corrales and Tu, 1989). Despite its
infrequent sexual reproduction, C. lindemuthianum exhibits a wide range of pathogenic
variability. One means of assessing the variability of the pathogen is inoculation of a set

of twelve differential cultivars (Table l), which possess different anthracnose resistance

genes or alleles. The differential series includes representative cultivars from the Andean
and Middle American bean gene pools. Each differential cultivar is assigned a binary
number. An isolate of C. lindemuthianum is assigned a race number, which is the sum of
the binary numbers of the differential cultivars on which the isolate is pathogenic

(Drijfliout and Davis, 1989; Pastor-Corrales, 1991).

Table 1. Anthracnose differential cultivars, their gene pool, binary number and resistance
genes

 

 

Differential Cultivar Gene Pool* Binary Number Resistance Genes

A. Michelite MA 1
B. Michigan Dark Red Kidney A 2 Co-I
C. Perry Marrow A 4 C0-13
D. Cornell 49242 MA 8 Co-2

E. Widusa MA 16 Co-9?

F. Kaboon A 32 €0-12
G. Mexico 222 MA 64 Co-3
H. PI 207262 MA 128 Co-43, Co-9

I. TO MA 256 C 0-4

J. TU MA 512 Co-5
K. AB 136 MA 1024 Co-6, 00-8

L. G2333 MA 2048 Co-42, C0—5, Co—7

 

* MA- Middle American, A-Andean, ?- allele of Co-9, yet to be determined
However, the differential series does not detect all pathogenic variability, as there
have been reported differences in pathogenicity between isolates with the same race

number. In a study of C. lindemuthianum diversity in Ecuador, thirteen isolates which

were not pathogenic on any of the differential cultivars (race 0) had six distinct patterns
of pathenogenicity when inoculated on a set of nine “local differential cultivars” (Falconi
et al., 2003). Molecular genetic studies of C. lindemuthianum have detected additional
variability both within and between races of the pathogen (F abre, et al., 1995; Balardin, et
aL,l997)
Control of Bean Anthracnose

Cultural practices, fungicides and genetic resistance have all been used to control
anthracnose with varying degrees of success. Pathogen-free seed has been used
successfully in some countries, such as the United States, that have semiarid regions
where disease-free seed can be produced. In developing countries, however, farmers are
unlikely to have access to disease-free seed, and are unable to avoid possible introduction
of the disease into their fields through infected seed. Seed treatments and foliar fungicide
applications are not sufficiently effective to prevent crop losses under conditions
favorable for disease development (Pastor-Corrales and Tu, 1989). Overuse of chemical
controls could lead to the emergence of fungicide resistant biotypes of C.
lindemuthianum (Tu and McNaughton, 1980) resulting in loss of disease control.
Additionally, chemical controls can be prohibitively expensive to small-scale farmers in
developing countries (Pastor-Corrales and Tu, 1989).

Generally, genetic resistance is considered the best control for bean anthracnose.
Historically, the introduction of major anthracnose resistance genes into commercial
cultivars has reduced the economic importance of this disease in Europe, the United

States and Canada, but dependence on Single resistance genes has led to the breakdown

of cultivar resistance when new races of the pathogen are inadvertently introduced or
arise through mutation (Fouilloux 1976, Kelly et al., 1994).
Genetic Resistance to Bean Anthracnose

Both quantitatively and qualitatively inherited genetic resistance to C.
lindemuthianum are present in bean, but research on quantitative resistance has been
limited. Geffroy et al. (2000) located ten quantitative trait loci (QTL) involved in partial
resistance to C. lindemuthianum races 7 and 45 on the BAT93 x JaloEEP558 RIL map.
Two of the QTL identified were located near genes involved in plant defense and three
were adjacent to genes conferring specific disease resistance. This finding lends support
to the theory that defeated resistance genes are involved in conferring partial resistance
(Nass et al., 1981; Li et al., 1999) and may represent members of a cluster of resistance
genes as proposed by Michelmore and Meyers (1998).

One recessive and nine dominant anthracnose resistance genes are reported in the
literature. Some genes, namely Co-I, Co-3 and Co—4, have multiple resistance alleles
that have been characterized (Kelly and Vallejo, 2004). The names of the known
resistance genes, their gene pools of origin, sources, and linked molecular markers are
shown in Table 2. The Andean sources of resistance that have been characterized all
possess alleles at the same locus, C 0-1 . The Middle American sources of resistance are
more diverse and represent eight different loci.

A number of studies have demonstrated the gene pool specificity of C.
lindemuthianum (Sicard et al., 1997; Balardin and Kelly, 1998; Geffroy et al., 1999).
Generally, Andean races of C. lindemuthianum are more virulent on bean lines carrying

the Andean resistance alleles at the C 0-1 locus than on lines with resistance genes from

the Middle American gene pool. Yet Co-I alleles confer resistance to several highly
virulent races of Middle American origin (Balardin and Kelly, 1998). Similarly, the
Middle American resistance genes confer resistance to Andean races of the pathogen.
Consequently, the best approach to anthracnose resistance breeding is to employ genes
based on a knowledge of the C. lindemuthianum races present in the region. In regions
where Andean races predominate, the Middle American genes should be used and vice
versa. In the US, where both Andean and Middle American races of the pathogen are
present, resistance is best achieved by pyramiding both Andean and Middle American
anthracnose resistance genes into a cultivar (Balardin and Kelly, 1998). The most
complete and durable resistance to anthracnose will be achieved through gene
pyramiding, as evidenced by the outstanding resistance of a three gene pyramid in G

2333, one of the anthracnose differential cultivars (Table 1).

:moCoE< 2352 .<E m emove< .QZ< p

38389 E @8865 89:3: mem n

 

 

 

 

 

 

 

 

 

 

amoom
.._a e =52é§< o E 832 8:0 <2 2-8
32 .a a 3&8 2mm 8 e<m <2 EN: SNSN a see
58 .a e =E2a§< omw<mo <2 2.8: oz m< m8
2: amm <2 $32 380 58

2.3920

32 2:32 Ba was; $520 seesaw <2 $8: on m< 98
58 .232 as 20:; mm<m 82 #2 <2 :5 B. m 8382 new
<2 SN: SNBN E New
58 .23 Ba 28$ 2mm; .m :5. 82 2mm <2 $33 23.0 ergo

58 2:32 2a 2§<
$2 .a a was; 2m<m <2 SE OF N. 352 4-8
EN 8&2 <2 N36
<2 :3 NNN 852 N SSE2 new

$2

.a a 822m-§e< 85m .8238

82 .23 Ba was; 2.2400 <2 E 9.84 2.280 as. New
«88% e a52.222 ta oz< Dz< 1-8
88 .232 as 02282 072 E 3052 PE 1-8

.88 2:32 Ba 9.21.5
88 5.32 Be 0:202 82233 Qz< an 8882 aid
82 .232 Ea max; 2650 oz< E .2822 82 25 3222 w #8
39:283— 23.32 833m 2.8.— 73.5:— isfin— «Ea Z 0.27..
.850 250 1.33:0 33:98::— m=£>9£ \oaoU

 

.8832: @832 van 8858 .550 .Emto
mo Boa meow :05 .865 mew—5:8 megs—:0 Eccoeobme £25: 3255 :05 £223 can moeow 853%me 08535: .N 2an

10

CIAT accession G 2333 is a landrace known as Colorado de Teopisca that was
collected in Chiapas, Mexico. G 2333 belongs to a genetically unique group within the
Middle American gene pool, termed race Guatemala, which includes several highly
disease resistant landraces (Beebe et al., 2004; Beebe et. al., 2000). Schwartz et a1.
(1982) found that G 2333 was resistant to C. lindemuthianum isolates from Colombia,
Brazil, Guatemala and Europe. Subsequent testing revealed that G 2333 was resistant to
380 isolates of C. lindemuthianum (representing 15 different races) from 11 Central and
South American countries (Pastor-Corrales et al., 1994). Pastor-Corrales et a1. (1994)
and Young et a1. (1998) determined that G 2333 carries a pyramid of three anthracnose
resistance genes: Co-42, Co—5, and Co-7. The most effective gene in this pyramid is Co-
42 which conferred resistance to 33 out of 34 different races of C. lindemuthianum
collected from 9 different countries in the Americas (Balardin et al., 1997).

Pyramiding of resistance genes presents a challenge for breeders, as the effects of
two or more resistance genes may confound one another, making their detection through
disease screening difficult or impossible. Molecular markers, which allow breeders to
indirectly test for the presence or absence of a gene without disease inoculation, facilitate
gene pyramiding (Kelly, 1995). RAPD and SCAR markers have proven useful in such
applications because they are relatively inexpensive and efficient to run. Both RAPD and
SCAR markers make use of in vitro DNA synthesis, or PCR. For RAPDS short
sequences of DNA, called primers, anneal to particular sequences in the genomic DNA
and initiate DNA replication at the Site. Since RAPD primers are short, they initiate
replication at several sites in the genome and produce DNA fragments different lengths

which can be separated by agarose gel electrophoresis, and visualized as discrete bands.

11

Slight differences in the genomic DNA sequences of the individual plants being tested
affect primer annealing and thereby result in different patterns of DNA fragments
amplified. A RAPD marker for a disease resistance gene amplifies a fragment of the
genomic DNA in the resistant plant that is located near the gene of interest but is absent
in the susceptible plant. SCAR markers are developed by sequencing the fragment that is
polymorphic between the resistant and susceptible plant and designing longer primers (of
~24 base pairs) which are specific to the fragment of interest. SCAR markers then
amplify only a specific region of the genome, which is linked to the resistance gene,
resulting in only one band. The advantages of SCAR markers over RAPDS are that
SCARs are more reproducible and do require as long for electrophoresis (Staub et al.,
1996)

Molecular markers have been developed for many of the anthracnose resistance
genes that have been characterized, including Co—42, the most effective gene from G
2333. SCAR markers linked to Co-42 include: SASl3 at 0.39 cM from the Co-42 gene
(Young et al., 1998; Melotto and Kelly, 2001), SH18 at 4.27i2.37 cM from the gene and
SBBl4 at 5.87i1.93 cM from the gene (Awale and Kelly, 2001). SASl3 and SH18 are
dominant markers coupled to the Co-42 gene. SBB14 is codominant, which means it
amplifies two different sized bands with the larger band associated with the resistance
gene. SH18 and SBB14 are specific to the Co-42 allele while SASl3 is present in lines
carrying other alleles at the C 0—4 locus. Even though it is not as specific as the other two
markers SASl3 has the desirable characteristic of being tightly linked to the Co-4 locus.
In a population of 1,018 plants segregating for Co-42, only four recombinations between

the gene and SASl3 were observed. SASl3 has successfully been used to introduce Co-

12

42 into highly susceptible pinto bean through backcrossing without pathogen screening
(Miklas and Kelly, 2002).
Anthracnose in Ecuador

Anthracnose is one of the most important bean diseases in Ecuador, particularly in
cool, moist, highland regions. A 1994 survey of C. lindemuthianum diversity and
virulence in the country identified 14 races from 94 different isolates of the fungus
(INIAP, 1995). Twenty- one percent of the isolates were not pathogenic on any of the
differential cultivars and were designated race 0. The most frequently identified races,
133, 128, 129, 5, and 4, collectively overcome the resistance genes in the differentials
Perry Marrow (C o-1 3 ), and PI 207262 (C o-43 and Co-9). Other races identified in the
survey were, 2, 15, 65, 131, 1153, and 1409, which overcome resistance genes in
Michigan Dark Red Kidney (C o-I ), Cornell 49242 (C 0-2), Mexico 222 (C 0-3), TO (Co-
4), and AB136 (C 0-6 and co-8). Twenty-nine isolates collected from wild beans in
Ecuador were mostly race 0; only one isolate was pathogenic on the differentials and was
characterized as race 6 (Sicard, 1997). A more recent survey of 31 C. lindemuthianum
isolates identified races 0, 3, 4, 256, 260, and 1346. Again, most of the isolates were race
0, which suggests a low level of virulence among isolates from Ecuador. Collectively,
the six races identified in the most recent study overcame Co-I, Co-13, Co-3, Co-4, Co-6,
and co-8 (Falconi et al., 2003). None of the races identified thus far in the country are
pathogenic on G 2333 (C 0-42, C 0-5 and C o-7) or Kaboon (C 0-1 2). These two differential
cultivars are potential sources of genes which would confer resistance to all known races

of C. lindemuthianum in Ecuador.

13

This project was undertaken with the objective of developing Ecuadorian bean
varieties with enhanced anthracnose resistance. To reach this objective, Co-42 was
introduced into Six Ecuadorian bush bean varieties, which represent three different
commercial classes. Use of molecular markers allowed for rapid, indirect selection for
the resistance gene so that three backcrosses could be made in quick succession. A
backcross breeding approach was used in order to maintain the required adaptation and

seed quality traits important to Ecuadorian bean farmers.

14

 

 

Materials and Methods
Use of Molecular Markers

Indirect selection was conducted based on RAPD (A813) and SCAR (SASl3,
SBBl4, SH18) markers linked to the C 0-42 gene. DNA was extracted from bean leaf
tissue according to the mini-prep procedure of Afanador et al. (1993). The DNA was
quantified using a fluorometer (Hoefer DyNA Quant 200, Amersham Biosciences) or by
comparison to Low DNA Mass Ladder (Invitrogen) in a 0.8% agarose gel. DNA samples
were diluted to a standard concentration of 10 ng/ul. RAPD marker reactions were
prepared in a 25 pl reaction volume with the following component concentrations:
template DNA, 1.2 ng/ul; primer, 1.2 ng/ul; MgCl2, 5 mM; PCR buffer, 1x; dNTP, 0.2
mM each base; Taq polymerase, 0.05 u/ul. SCAR marker reactions were prepared in a
20 ul reaction volume with the following component concentrations: template DNA, 1
ng/pl; primer (Integrated DNA Technologies, Coralville, IA), 0.5 ng/ul each primer;
MgCl2, 3.75 mM; PCR buffer, 1x; dNTP, 0.2 mM each base; Taq polymerase, 0.05 u/pl.
Amplification by Polymerase Chain Reaction (PCR) in a MJ Research FTC-100 thermal
cycler (MJ Research, Inc., San Francisco) followed the thermal profiles recommended by
the markers’ developers (Young et al, 1998; Awale and Kelly, 2001). PCR products
were loaded into a 1.4% agarose gel containing 0.02 rig/ml ethidium bromide, 40 mM
Tris-acetate, and 1 mM EDTA. After electrophoresis the DNA bands were visualized

under ultraviolet light and photographed.

15

 

Introduction of Co-42 into Ecuadorian Varieties Through Marker Assisted
Backcrossing

Crossing between Ecuadorian varieties, J e.Ma. and Cocacho, and Red
Hawk*2/SEL 1308, a line carrying Co-42, began in November of 2001. A description
and pedigree for the parental materials appears in Table 3, and a summary of the crossing
and selection activities appears in Table 4. All crosses were made in the Michigan State
University Greenhouses, East Lansing, MI.

RH*2/SEL1308 is a Single backcross of Red Hawk, an improved dark red kidney
variety of Andean origin, to SEL 1308 a type II small black which carries C0—42 derived
from G2333 (Young et a1, 1998). Rather than using G2333 or SEL 1308 as the donor
parent, RH*2/SEL1308 was chosen as the source of C o-42 , since this line produced seed
nearer in size to the large-seeded Ecuadorian recurrent parents. The RH*2/SEL1308
plants were segregating for C 0—42 , so the RAPD marker A813 was used to select those
plants likely to be carrying the resistance gene. The seed for Je.Ma. and Cocacho was
supplied by Dr. James Beaver, University of Puerto Rico, Mayaquez. Je.Ma is a large
seeded Calima type of Colombian origin and Cocacho is a medium seeded yellow
Canario, which may have Peruvian origins (Voysest, 2000). The donor parent was also
crossed with Genuine, a McCaslan type climbing snap bean which carries two resistance
genes for bean rust, Ur-3 and Ur-4 and the Bean Golden Mosaic Virus resistance gene
bgm-I . Snap beans are grown in Ecuador and we speculated that this line could be useful

in Ecuador because of its multiple disease resistance genes.

16

$22 .a a maze: s
:32 :3 a 223:

outflow €883. Engagement. BBSon< .«o Eofitmaoa wBSm c225 - mm<-<DmD
ogsooaocwaq. 2858030 93:2: - <2

1338 .H «E_=owew< ow 3228585 3:60 .. BSD

. p
$2 .358

@2585 BmEEBoefi - >~

325a wEnfizo 080m mommommoa .8555 can 8on Ewta: 55 8.285885 1 9:
325a Manet—Q 83m— .mosoofin can 8on Ewan: £3, 8.35:3er i a:

 

 

e
SOON .amom>o>vb oEEEBoQ nu a
(CEO amoemEEmPKRmmmO match = E omv xoflm =~Em we: 15m
$2-3m: rem - . . - . 3 cc
dxm .w< emeoLE “OE—moueoEKiwoB—ELO N ob N ob seem N _ >269 com x80 kam Box
m%-<QmD
£0sz dxm .w< mecwomwogm... -EM . L . L 928 BQEESECE 2:25
COMM 0:03“ TNHW ma AW» U ~WMNNMN U N Q V D Mr D >H “OS CN—WNUUE . 0
BE .w< «ESE : E O\m\m*~v _ O 2
:5 20m ocaaaoahzo 05m 2 2 a Ca €26 _-mo<
penance: QEwCo Sitcom Co bflmmoa l _ Aw omv 0.5280 93800
<O=H<~O ems—NORNR—O -- 3: Am m3 «EEO 22.3 -- wzui<~zh
.32 <>m -- Am 3 282
CEO crammed EEC setup—om m bflmmom a: com 82on Eomwfimm
-- 3 we 0:82 -
.320 Eomwfimfimoe 92¢. a: com 8282 N m§<
ESQ EEORCNOENNQM Ev -- H R o3 «Ezmo BEES.» .. v3-2;
23.—sow 99.33.— 850 ooeaamfiom nasal 539.0 @6on 3:3 mod—U on:

 

wfimmoeo 5 com: 852$ .m 653.

17

”2882833 95 832?: m... 3 wok/BBL E33 E2508 ”$8833 25 330:2: N... 3 330:8 EBB Enhanced

 

game; Emmmfimfi. e : 82 2mm}.m§a2a§§a=aw§> 88
22282 Emmm .2m<m e 2 M52 2%...4322demeaning 83
05:50 Emmimmm e M: 82 ammarigraaaso 83
N-m2m< 2m<m a S we: 2mm\mtm§.a22\m\m*m-m2m< 85
vmo< flm<m : mm we: amemfiéafisooaa...Tmo< New

 

 

0:880 m _ m<m v N ~ we? AMmNaiygm *onomoou 8%
m mmoSo—omm
.323. m~m<m c a wofl amm\N*g\N*.mE.om N03
Eomwfiwm m fl m<m m 0 gm 2 AWmN...§\.m2.o_.\m\fiomwm8m N03
«EstCr m fl m<m m o womfi qmmN*§\.m2.o<m\£:=wcs> NoB
TmO< m ~ m<m N o womfi qmm\N*§\o:owooO\m\ fl -mO< Nob
Nam; m~m<m m o womfi dmfiNLfiS .mEdeNdé N03
0:850 m fl m<m o c mom fl qmm\N*§\N*oeScoO SR
05800 m _ m<m m c we? 4mm\N*E~<\N*onowooO No\m

 

N mmBoxomm
N-m§< .anmfimm .m_::w::> r322. mfim< we»: 4mm\N*g\.m2.om 8%
05:50 m_m< womfi 4mm\N*§\oE::oO No\N
TmO< .onomooO m~m< a N2 we: 4mm\N*E~<\o:omooU No\N

A mmoaoxomm

2
8550 £22 .2088 2m< 2 2 we: RENEE Q:

GOO
CO
v—Iv—1

 

cote—om 53.35
25>? .3396 flea—m vogue—om .5qu a a 25...?— :21:— 8a:

.3380 895 Sena @8028 05 £053 :23 mount? was £2828 com com: “BEE 3:622: .wemmmBo
com @8028 Sew—m No 638:: 6253 $53 No 398:: .8380 now “:83 5on 05 mm 56% 8:: 8m 8:3 mecca—m Av 2an

l8

In February of 2002, seed of four additional Ecuadorian varieties was obtained
from the National Legume and Andean Grains Program (PRONALEG-GA) at INIAP:
Yunguilla, an improved Calima type; ARME-2 and Paragachi, both medium seeded red
mottle types; and ACE-l, an improved Canario. These four varieties were introduced
into the crossing program as recurrent parents during the first backcross in February and
March of 2002. J e.Ma.// RH*2/SEL1308 plants carrying the A813 marker were crossed
with Yunguilla, ARME-Z, Paragachi and Je.Ma. J e.Ma., Paragachi, ARME-Z and
Yunguilla are sensitive to photoperiod and required the use of blackcloth to shorten the
daylight period to induce flowering in Michigan. Cocacho//RH*2/SEL1308 plants
carrying ASl3 were crossed with ACE-1 and Cocacho. Genuinel/ RH*2/SEL1308 plants
carrying A813 were crossed with Genuine. In all cases this was considered the first
backcross even though it was the first cross involving Yunguilla, ARME-Z, Paragachi
and ACE-l, since the first cross had been to plants of similar seed type.

Seed for the second backcross was sown in May and June of 2002. The J e.Ma.,
Yunguilla, ARME-2 and Paragachi crosses and parents were sown in a growth chamber
(22C, 10 hour daylength) and maintained there until they began flowering. At flowering,
they were moved to the greenhouse for crossing. The BC1F1 plants were screened with
the SCAR marker SASl3 and those carrying the marker were crossed with their
respective recurrent parent.

Seed of the BC2F1 generations for Cocacho, ACE-1, Yunguilla, ARME-Z,
Paragachi and Genuine was sown in August and September of 2002 in order to generate
the third backcross. J e.Ma. was not included since its long generation time did not allow

for harvest of seed in time to make a third backcross. SASl3 was used to select plants

19

from the Cocacho, AC E-l, and ARME-2 populations. Both SBB14 and SASlB were
used to select plants from the Yunguilla and Paragachi populations. SBB14 and SH18
were used to select plants from the Genuine population since it was discovered that that
Genuine tested positive for the SASl3 marker. The BCzFl plants which were positive for
these markers were crossed with their respective recurrent parent.

Marker Assisted Selection in Ecuador

In February of 2003 a sample of the BC2 and BC 3 generations was planted in pots
in a greenhouse on INIAP’s Santa Rosa substation near Quito, Ecuador. The greenhouse
is located at 2750 m.a.s.l. and consists of a fiberglass roof and screened walls. The
potting media was pasteurized soil and pumice. Fifteen seeds from the BCze
generations and ten from the BC 3F 1 generations of the Cocacho, ACE-1, Yunguilla,
Paragachi, ARME-Z and Genuine populations were planted in the greenhouse in addition
to 10 seeds from the J e.Ma BC2F1 generation. Two seeds of each recurrent parent were
also sown.

Leaf tissue was collected from each plant and DNA extracted as in previous
selections. SASl3 primers failed to amplify a band afier repeated attempts.
Consequently, the SBB14 and SH18 SCAR markers were used to select plants believed
to be carrying the C0-42 anthracnose resistance gene. Selfed seed of the selected plants
was harvested and was retained in Ecuador for further testing.

Field Inoculation of Backcross Populations

The majority of the Cocacho, ACE-1, Yunguilla, Paragachi, ARME-Z and

Genuine BCze and BC3F1 generation seed was planted in a research plot on the Santa

Rosa Substation. The plot was at an elevation of 2740 m.a.s.l. - higher than bush beans

20

are typically grown in Ecuador. However, the cool, wet conditions at this elevation were
ideal for anthracnose infection. The field was planted on February 11, 2003 according to
the layout in Figure l. Paragachi, which is highly susceptible to anthracnose, was planted
in the guard rows. Additionally, three seeds of Paragachi were planted at the end of each
experimental row to serve as a disease spreader.

The plot was cultivated at 28, 55, 88 and 118 days after planting in order to
control weeds. A foliar fertilizer was applied at 34 days after planting and a sidedressing
of urea (approximately 20 kg/ha) at 55 days. Dimetholate was applied at the
recommended rate at 88 days to control leafhoppers (Empoasca krameri).

The field was inoculated three times with a race 0 isolate of Colletotrichum
lindemuthianum collected from the Santa Rosa Substation. The fungus used for
inoculation was grown in Erlenmeyer flasks on chopped and sterilized snap beans pods.
When the fungus began to produce spores, the contents of the flask were pureed in a
blender and diluted with distilled water. The final spore concentration was calculated
using a hemicytometer. The spore solution was applied to the plants in late afiemoon
using a backpack sprayer. The plants were first inoculated at 59 days afier planting with
7 L of spore solution at a concentration of 1.2 x 105 spores/ml. The second inoculation
was at 74 days afier planting with 15 L of spore solution at a concentration of 4.2 x 105
spores/ml. The third inoculation was at 92 days after planting with 20 L of spore solution

at a concentration of 8.5 x104 spores/ml.

21

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

030.. E m 0.50.. E m 0.500 E n
0000w000m 0000w000m 0000w000m
000000003 000000000 0000w000m

2 ~00 0-0302 2 _-mo< 0 M00 200005

3 N0m NAG/92 NN N0m 000000 ON 0E=w00>

m0 N0m 0E=w00> m0 m0m 000.000 0N N0m. 0000000

n M0m N0; 00 ~0m N02”? 0 m0m 000000000

N 000000 2 ~0m N-m=>a< m 0 N0m 0:895.»

0 .00 080200 2 N00 700< 0 N00 080800

s 080000 2 N00 000... s 080000

..m 00 ~0m Tm0< 2 N0m Tm0< 5 m0m 0000000 Z
We 2 N0m Tm0< wfi N0m 000000000 00 m0m 0000000 m
m E N0m Tm0< E 0000000 0 M0m Nan"; m
um 00 M0m 0000000 0 00m 00000000 00 N0m Tm0< s
.m 2 ~00 20080 a 05:50 2 N00 000... m...
m N0 N0m 0000000 N E00w0000 2 N0m Tm0< W
m C M0m 0000000 w NAG/02 w Nd; aw
m C m0m 0000000 m0 N0m 0:8m03w E 0000000 w
3 NN N0m 0000000 0 M0m N0; M2 Tm0< m..

0N 0:590; 3 m0m Tm0< m _ 00m 005000

E 0000000 a m0m 0E=w00> 2 N0m 0000000

a mom 0:805; a 0:500; 2 ~00 £880

9 N0m 00000000 m _ N0m 0000000 “N 05.0000

3 Tm0< N0 N0m 0000000 3 m0m Tm0<

w Nd; C M0m 002000 n 0 N0m N030?

2 M0m Tm0< 00 M0m 0000000 2 ~0m N020?
0000w0a0m 0000m000m _000w000m
0000m000m 00003000 000M000

00000..— t 0030—000E005 00000.0 00 0050—0009005 00000.0 0 0050.00.50.03

 

 

 

00000005 000M 0000w 000 00 000003 200 .00 00904 ._ 0:.me

22

The number of plants that were flowering in each population was recorded at 55,
62, 69, 73, 83, and 87 days after planting. The anthracnose symptoms on each plant were
evaluated at 74, 88, 102 and 119 days after planting. Leaf symptoms were rated on the 1
to 9 scale described in Table 5. Plants having a rating of 1-3 were considered resistant,
those with a rating of 4 to 6 were considered moderately susceptible, and plants with a
rating of 7 to 9 were considered very susceptible. Pod symptoms were rated according to
the O to 3 scale shown in Table 6. A rating of O was considered resistant, a rating of 1

moderately susceptible, and a rating of 2 or 3 very susceptible.

Table 5: Foliage Symptom Ratings for Anthracnose*

 

 

Rating Symptoms
1 Plant has no visible symptoms on leaves.
2 Few small lesions in veins on back of leaf
3 Frequent small lesions in veins on back of leaf
4 Large lesions in the veins on back of leaf.
5 Large lesions visible on front and back of leaf.
6 Leaf lesions as in 5, some lesions on petioles, and stems
7 Many large lesions on leaves, petioles, and stems.
8 Many large lesions, accompanied by dead leaves, reduced growth
9 Plant is dead or nearly so.

 

*A rating of 1 to 3 is considered resistant; 4 to 6 is considered moderately susceptible, and 7 to 9 very
susceptible.

Table 6: Pod Symptom Ratings for Anthracnose’r

 

 

Rating Symptoms
0 Plant has no visible symptoms on pods.
1 Small superficial lesions on pods
2 Large, deep, sporulating lesions on pods
3 Plant has no pods due to severe anthracnose symptoms in foliage.

 

”(A rating of O is considered resistant, l is considered moderately susceptible, and 2 or 3 very susceptible.

23

Evaluation of Seed From Field Grown Populations

Seed from each of the plants grown in the field was harvested and maintained
separately. Plants were classified according to their seed coat pattern and color. Some
difference in seed size was also apparent. However, there was insufficient quantity of
seed to evaluate this trait.

Farmers recruited from among the participants in a Bean Seed Exhibition held by
PRONALEG-GA in the community of La Concepcion, June 29, 2003 were asked to rank
samples of the seed produced by the ACE-1, Cocacho, and Yunguilla BCze and BC3F1
populations according to their preference. Ties were permitted.

For presentation to the farmers the seeds were glued onto 2 x 2” pieces of card
stock. Figure 2 shows a photograph of the seed samples that were used. (Figure 2 in this

thesis is presented in color.) Table 7 gives a description of each sample.

Figure 2. Seed samples farmers rated in survey

 

24

Table 7. Samples used in farmer evaluations of seed quality

 

 

Cocacho Samples ACE-1 Samples Yunguilla Samples

A white F tan, yellow edges K red/pink/white mottle*
B white/yellow mottle G ACE-l, parent L Yunguilla, parent

C yellow* H yellow* M red/white mottle

D brown I white

E Cocacho, parent J white/yellow mottle

 

* sample which most resembled the recurrent parent
Greenhouse Inoculations of Backcross Populations

The following populations were screened with the A813 and SH18 markers for
C 0-42 and inoculated with C. lindemuthianum under greenhouse conditions at Michigan
State University: Paragachi BC 1F2, BCZFZ, and BC3F1; Cocacho BC le, BCze, and
BC3F1; ARME-2 BCze and BC3F1; and Yunguilla BCze and BC 3F1. Leaf tissue was
collected from the plants before inoculation and DNA extraction and marker analysis
were performed as previously described. According to the method described by Balardin
and Kelly (1998) plants were spray inoculated at ten days after planting with a spore
suspension containing 1.0x106 spores/ml and 0.01% Tween 80. Afier inoculation, plants
were maintained in a humid chamber for 48 hours.
The Paragachi and Cocacho populations were inoculated with the same Ecuadorian race 0
isolate that was used for field inoculations in Ecuador. The ARME-Z and Yunguilla
populations were inoculated with a Mexican race 2 isolate. The Ecuadorian parent of the
population and SEL 1308 were included in each inoculation as susceptible and resistant
controls respectively. Disease reactions were evaluated 6 to 10 days after inoculation.
Plants were rated as susceptible (a disease reaction like that of the Ecuadorian parent,
resulting in death of the plant), interrnediately resistant (more resistant than the

Ecuadorian parent but exhibiting susceptible symptoms -- these plants were often able to

25

survive inoculation), resistant (disease reaction like that of SEL 1308, no symptoms of

anthracnose).

26

Results and Discussion

Flowering in Field Grown Populations

Days to flower (DTF) of the twelve populations and six parents are given in Table
8. There was no difference in DTF between the backcross populations and their recurrent
parent. This is one indication that plants in the backcross populations resembled the
recurrent parent, and may be similarly adapted to the target Ecuadorian bean production
areas. DTF was longer than is typical for these varieties. The cool temperatures due to
the high elevation of the field site were probably the cause of the delay in flowering as
temperatures of 12 to 14 C have been shown to dramatically increase DTF in beans

(Wallace et al., 1991).

Table 8. Days to flowering for the backcross populations an recurrent parents

 

 

 

 

Population Days to Flower Population Days to Flower
Cocacho 62 ARME—Z 73
Cocacho BCze 62 ARME-2 BCze 73
Cocacho BC3F1 62 ARME-Z BC3F1 73
ACE-1 62 Yunguilla 69
ACE-1 BCze 62 Yunguilla BCze 69
ACE-l BC3F1 62 Yunguilla BC3F1 69
Paragachi 73 Genuine 73
Paragachi BC 2F 2 73 Genuine BCze 73
Paragachi BC 3F1 73 Genuine BC 3F 1 73

 

Reaction to Field and Greenhouse Inoculations

Cocacho Field Inoculation With Race 0

Figure 3 shows the average leaf symptom ratings for the two Cocacho backcross
populations and the parent in response to field inoculation with race 0. The average leaf
symptom rating for the parent ranged from 3.3 to 6.0. The BC2 and BC3 populations

exhibited a larger range of leaf symptom ratings than the parent: 1.67 to 6.67 and 1.33 to

27

4.67 respectively. 53% of the BC2 plants and 40% of the BC3 plants had an average leaf
symptom rating greater than three and were categorized as having susceptible leaves. All

of the control plants (parent) had an average leaf symptom rating that was greater than 3.

Figure 3. Average leaf symptom ratings for Cocacho backcross populations and parent

 

 

0.30 '
I
0.25 ~ , A ~ , , , [
g: i
a 0 0.20- — , .7 7 I
i 0 p _ __
1 g- ‘IParentii
' E 0.15 A 0 W 0 - 0 ~00 ~ . ~ , < {‘IBCZ
. a t,
l ,5 iCIBC3 ,
l 4': 0.10 --~ 0 ~— —
0
0.05 1% k — - 0

 

 

 

 

 

 

 

 

 

 

 

g?» 09 wt? .5. ,9 ,5» .3. .09
Average Leaf Syptom Rating

Three percent of the parent plants exhibited no pod symptoms and three percent
had small lesions. The remaining 94% of the plants had deep spore producing lesions on
their pods. All of the parent plants exhibited either susceptible pod or leaf symptoms and
were, overall considered susceptible. 58% of the BC2 plants and 60% of the BC3 plants
had susceptible pod symptoms. Figure 4 shows the assignment of the Cocacho
populations to one of two categories, resistant or susceptible, based on both pod and leaf
symptoms. 80% of the BC2 plants and 64% of the BC3 plants had either susceptible

leaves or susceptible pods and were overall considered susceptible.

28

Figure 4. Determination of resistance or susceptibility in Cocacho backcross and parent
plants inoculated with race 0 isolate based on pod and leaf symptoms

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00
l
1 0.800400777777404 ---_g
. >1
0
l G
1 “0’
.5. 0.60--- -03,
0
l i-
. a.
1 3
‘3 0.40.--m»---i-- 00-0
£3
1 0
1 D!
« 0.20 -0 .0 g 0 #1 .. ”—
l
0.00 ‘ ‘J—l ' . ;
; Rpod, ‘ Rpod, I Spod, ‘ Spod, : Resistanti Suscep- T
lRleafO‘Sleafl ,Rleaf2.Sleaf3y l I tible ‘
_ m _ ..._.l_---.___._-_.. l— ..._ __. TM. __ ___L» ___. _0 i._. ._. -1 0_ 0.“- .4iu__._._._ J, _ 0 _i
ilparentl 0.00 . .03 1 0.00 l 0.97 ' 1 0.00 I 1.00 .' 1
1.3g ; 0.29 ‘ 0.13 l 0.18 1 0.40 1 0.29 1 0.71 1 J
1— 0-0 .w 0* t~ ._.. v1.._-.. Ina-.. __1.__0_ __ .‘___—...—+ ..___._ 4.“,_, 0v?
10.893 1 0-36 g- 0-03 g 0:2: 133:, L L .1‘ £139 ._9-_6,4 : r - 1

Resistance Category

If the Co-42 resistance gene were segregating as expected in these populations,
25% of the BC; plants and 50% of the BC3 plants would be susceptible. The populations

did not exhibit these expected ratios.

Cocacho Marker Analysis and Greenhouse Inoculation with Race 0

The results of marker analysis and greenhouse inoculation for the Cocacho
populations are given in Table 9. In greenhouse inoculations the Cocacho parent was

confirmed to be susceptible to the race 0 isolate used in the field study. Three plants

29

 

from the BCle population exhibited complete resistance to the race 0 isolate of C.
lindemuthianum. Some additional plants in this population had intermediate resistance to
the pathogen. In the BC ze and BC3F1 populations, none of the plants were completely

resistant, although some had intermediate resistance.

Table 9. Cocacho BCle, BCze and BC3F2 Marker Analysis and Reaction to Ecuadorian
Race 0 Isolate

 

ASI3 and SH18 Markers

 

 

Disease Reaction Present Absent
BCle 8 4
. . BC2F2 2 1
Intermediate ReSIStance B C3F1 1 0
Total 11 5
BC1F2 8 2
. 1300 9 4
Susceptible BC3F1 5 7
Total 22 13

 

As shown in Table 10, the AS13 and SH18 markers do not appear to be associated
with any of the resistance, complete or partial, observed in the Cocacho populations.
Results from the x2 Test for Independence indicate that in each of the populations, the
marker results and disease reactions are independent of one another. Moreover, odds
analyses indicate that a positive marker result does not increase the chances of a plant
being anthracnose resistant, as would be expected were the marker linked to the gene(s)

conferring resistance.

30

Table 10. Cocacho BC 1F 2, BCze and BC3F2 Test for Dependence of Marker and
Disease Reaction

 

Population Xz-value Odds of Resistance Odds of Resistance Given Positive

 

Marker Result
BCle 0.49 1.20 1.20T
BCze 0.01 0.23 0.05T
BC3F1 1.26 0.08 0.021L
Total 0.17 0.46 0.23T

 

1' Odds of Resistance Given a Positive Marker Result 5 Overall Odds of Resistance indicates that a positive
marker result decreases or does not affect the chances of a plant being resistant

One possible explanation for this data is that a recombination between Co-42 and
the SASl3 occurred early in the breeding program. However this is unlikely given that
there is a tight linkage between SASl3 and Co-42, and several plants carrying the marker
were used as parents in each generation. Another possible explanation is that C 0-42 does
not confer resistance to anthracnose in the genetic background of Cocacho; the variety
may lack certain genes that are required for C 0-42 to be expressed or to function. The
anthracnose resistance genes in bean have generally been regarded as functioning
independently from other genes. However in the case of Bean Common Mosaic Virus
(BCMV) resistance, bc-22, a recessive resistance gene, requires another gene, bc-u, in
order to function (Kelly, 1997). Additionally, genetic background has been show to
affect the expression of resistance genes in wheat (van der Westhuizen et al., 1998) and
oat (Wilson and McMullen, 1997). Since the Co-42 gene is from an isolated bean race in
the Middle American gene pool, it is likely that the genetic background of G2333, in

which C o-42 functions, is quite different than that of Cocacho, an Andean landrace.

ACE-I Field Inoculation With Race 0

Figure 5 shows the average leaf symptom ratings for the two ACE-1 backcross

populations and the parent in response to field inoculation with race 0. The average leaf

31

symptom rating for the parent ranged from 1.3 to 2.3. The BC2 and BC3 populations
exhibited a larger range of leaf symptoms ratings than the parent: 1.0 to 4.3 and 1.3 to 3.0
respectively. Three percent of the BC; plants had an average leaf symptom rating
greater than three and were categorized as having susceptible leaves. All of the control
plants (parent), and the BC3F1 plants had an average leaf symptom rating that was less

than 3 and considered to have resistant leaves.

Figure 5. Average leaf symptom ratings for ACE-1 backcross populations and parent

 

0.40

0.30 ,

Relative Frequency

0.20 ~

0.10 ~

 

 

 

 

 

 

0.00- _
1 1.00 1.33 1.67 2.00 2.33 2.67 3.00 3.33 3.67 4.00 4.33

 

1
Average Leaf Rating '

None of the parent plants exhibited pod symptoms. The parent plants had no pod
or leaf symptoms and were, considered resistant, although the sample size was very small
due to poor germination. 17% of the BC2 plants and 7% of the BC3 plants had

susceptible pod symptoms. These plants were considered susceptible. (3% of the BC2

32

plants had susceptible leaves and susceptible pods). Figure 6 shows the determination of
resistance or susceptibility in the ACE-1 backcross and parent plants based on pod and

leaf symptoms.

Figure 6. Determination of resistance or susceptibility in ACE-1 backross and parent
plants inoculated with race 0 based on pod and leaf symptoms

 

 

 

 

 

 

 

 

 

 

1.20
1% ------
l
l
>. 0.80 i A 3 ~ , 0 .
U
0
0
0
g 3
‘ E 0.60 f 3 i i i w
0
l .2.
1 ‘5
l 0 .
at. 0.40 — ;
0.20- 3 i A ,
0.00- , . I .
. Rpod,R ‘ Spod,R 1 Spod,S : . .
leaf ! leaf I leaf : ‘ Resrstant lSuscepnble| [
,- - -_ .. .__ 2- . . .2 -. - -1- . ___ 0. m- 1 1
1.Parent‘ 1.00 . 0.00 1 0.00 ‘ 1.00 0.00 ' 1
IBC2 1 0.83 1 0.14 , 0.03 . 0.83 l 0.17
‘DBC3 0.93 0.07 g 0.00 0.93 : 0.07 ;
Resistance Category

Later greenhouse inoculations confirmed that ACE-l is resistant to the race 0
isolate that was used in the field inoculations (see appendix A). The few susceptible

individuals in the BCze generation may be explained by loss of the parents’ resistance

33

genes through independent assortment. The presence of susceptible plants in the BC 3F 1

population suggests that the resistance in the ACE-l parent is recessive, or that the ACE-

1 parental line is segregating for resistance to this race.

Paragachi Field Inoculation With Race 0

Figure 7 shows the average leaf symptom ratings for the Paragachi backcross

populations and the parent in response to field inoculation with race 0. The average leaf

symptom rating for the parent ranged from 7.5 to 8.8. The BC2 and BC3 populations

exhibited a larger range of leaf symptom ratings than the parent: 2.5 to 7.8 and 2.3 to 8.5

respectively.

Figure 7. Average leaf symptom ratings for Paragachi backcross populations and parent

,_, ,_ _.3,,_ . _. __ 2... ,__. _. ..____.____‘_______,,____,__ _ __.]

 

0.30

Relative Frequency

0.40 .

0.20 —- —.

0.101 -

 

0.00 _.

f».

 

nsmmaxv

   
   

I Parent
“— I BC2 "
V __ El BC3

 

 

 

 

(\‘0

no»
I
b. (\. /\. %. %.

«a ’0
<0- 0- 10’ 2’

Average Leaf Rating 1

 

Plants were categorized into three different leaf symptom categories. Those

plants with average leaf ratings between 0 and 3.0 were considered as having resistant

34

leaves. Plants with ratings between 3.1 and 5.5 were considered moderately susceptible
and those with average ratings greater than 5.51 were considered very susceptible. All of
the parental control plants had an average leaf symptom rating greater than 5.51 and were
considered very susceptible. In the BC; population 8% of the plants had resistant leaves,
33% had moderately susceptible leaves and 58% had very susceptible leaves. In the BC3
population 11% of the plants had resistant leaves, 22% of the plants had moderately
susceptible leaves and 67% of plants had very susceptible leaves.

Forty-three percent of the parental control plants had spore-producing lesions on
their pods. The remaining 57% of the control plants did not produce pods because of
stress or death resulting from severe leaf symptoms. In the BC2 populations 8% of the
plants had no pod symptoms, 31% had small lesions, 50% had spore-producing lesions
and 10% produced no pods due to stress or death resulting from inoculation. In the BC3
populations 11% of the plants had small pod lesions and 61% of the plants had spore-
producing lesions. The remaining 28% of the plants did not produce any pods. Table 11
shows the assignment of Paragachi backcross and parent plants to susceptibility
categories based on pod and leaf symptoms. Plants with resistant leaves and those with a
moderately susceptible leaf symptom rating and pods with small lesions or no lesions
were considered interrnediately resistant. 27% of the BC; plants and 22% of the BC3

plants fell into this category.

35

Table l 1. Determination of intermediate resistance or susceptibility in Paragachi
backcross and parent populations based on pod and leaf symptoms (frequency, percent of
population)

 

 

     

 

 

 

 

 

 

 

 

 

Pod Leaf Ratingb

Ratinga Population 0 1 2
Parent 0, 0%

0 BCze {27‘9“
BC3F1 0, 0%
Parent 0, 0%

l BC2F2 fi:’63“§
BC3F1 0, 0%
Parent 0, 0% 0, 0%

2 BCZFZ 0, 00/0 7. 0.15
110:. -
Parent 0, 0% 0, 0%

3 BCze 0, 0% 0, 0%
BC3F1 0, 0% 0, 0%

 

r0, no pod symptoms; 1, small lesions that did not produce spores; 2, large spore producing
lesions; 3, plant did not produce pods due to stress or death as a result of disease

b 0, resistant, average leaf symptom rating 0 to 3.0; 1, moderately susceptible, average leaf
symptom rating 3.1 to 5.5; 3, very susceptible, average leaf symptom rating 5.51 to 9.0

1 té'irfiiieqdiiitély’ré‘sisg ” y‘p

 

Paragachi Marker Analysis and Greenhouse Inoculation with Race 0

The results of marker analysis and greenhouse inoculation for the Paragachi

populations are given in Table 12. In each of the three populations some plants exhibited

partial resistance to the C. lindemuthianum isolate, but none of the plants were

completely resistant to the disease. This confirms the expression of partial or

intermediate resistance observed in the field inoculations of the Paragachi backcross

plants.

36

Table 12. Paragachi BCle, BCze and BC3F2 Marker Analysis and Reaction to
Ecuadorian Race 0 Isolate

 

 

 

Disease Reaction SASI3, SH18, and SBBI4 Markers
Present Absent
BC1F2 l4 0
. . BCze 12 0
Intermediate Re31stance B C3F1 4 1
Total 30 1
BC1F2 5 8
. BCZFZ 5 7
Susceptlble B C3F1 2 2
Total 12 17

 

As shown in Table 13, the SASl3, SBBl4 and SH18 markers are associated with
the partial resistance observed in Paragachi populations. Results from the x2 Test for
Independence indicate that in the BOP; and BCze populations, the marker results and
disease reactions are dependent. Failure of the test to detect dependence in the BC3F1
population is probably due to small sample size (n=9). The odds analyses indicate that in
all of the Paragachi populations, a positive marker result increases the chances of a plant
being anthracnose resistant, as is expected for a marker linked a segregating resistance

gene.

Table 13. Paragachi BC1F2, BCze and BC3F2 Test for Dependence of Marker and
Disease Reaction

 

Population Xz-value Odds of Resistance Odds of Resistance Given Positive

 

Marker Result
BC1F2 12.24**** 1.08 3021
BCze 9.88*** 1.00 2.401'
BC3F1 0.90 1.25 2.501
Total 21 .89**** 1.07 2.67T

 

**** marker and disease reaction are dependent at 01<.001, *** marker and disease reaction are dependent
at a=.005,**marker and disease reaction are dependent at a=.01 * marker and disease reaction are

dependent at 0t=.025
1’ Odds of Resistance Given a Positive Marker Result > Overall Odds of Resistance indicates that a positive

marker result increases the chances of a plant being resistant

37

As in the Cocacho populations, only partial resistance to anthracnose was
observed in the BC; and BC3 populations, but in the case of the Paragachi populations,
there is evidence that C 0-42 is the source of some of the partial resistance. In Paragachi’s
genetic background Co-42 is not fully functional. Instead Co-42 may act as a defeated
resistance gene which confers partial resistance similar to the anthracnose resistance QTL
reported by Geffroy et al. (2000). The plants that carried markers for Co—42, but did not
express any partial resistance may represent individuals that completely lack the other
genes needed for function of Co-42. This suggests that in a pure Paragachi genetic
background Co-42 may not function at all. In field inoculations, the most resistant plants
were from the BC2 population, perhaps because they retained a greater proportion of the

donor parent’s genes.

ARME—Z Field Inoculation With Race 0

The ARME-2 parent, the BC; and BC 3 populations were resistant to race 0 and
neither the pods nor the leaves had susceptible symptoms. Since ARME-Z apparently
carries resistance to race 0 it was not possible to distinguish plants in the field that were

carrying Co-42.

ARME—Z Marker Analysis and Greenhouse Inoculation with Race 2

The results of marker analysis and greenhouse inoculation with race 2 for the
ARME-2 populations are given in Table 14. ARME-2 was found to be susceptible to a
Mexican race 2 isolate of C. lindemuthianum. Plants from the BCze and BC 3F1

populations exhibited a clear resistant or susceptible reaction to the isolate, with some

38

plants from each population showing a completely resistant reaction identical to that of

SEL 1308 and the remainder exhibiting susceptibility identical to the Ecuadorian parent.

Table 14. ARME-Z BCze and BC3F2 Marker Analysis and Reaction to Race 2

 

 

 

. . AS13 and SH18 Markers
Disease Reaction
Present Absent
BCze 11 O
Resistant BC 3F1 6 0
Total 17 0
BCze 0 5
Susceptible BC3F 1 0 6
Total 0 ll

 

As shown in Table 15, positive results for the AS13 and SH18 markers are
associated with resistance in the ARME-2 populations. Results from the x2 Test for
Independence indicate that in each of the populations, the marker results and disease
reactions are dependent. All resistant plants carried the two markers for C042 and all of
the susceptible plants tested negative for the markers, indicating that, as expected, C0-42
confers complete resistance to race 2 in the ARME-Z genetic background. Since ARME-
2 is an advanced line from the INIAP bean breeding program it may carry additional

. . 2 .
Mesoamencan genes Wthh are necessary for Co-4 to fiinctlon.

39

Table 15. ARME-2 BCze and BC3F2 Test for Dependence of Marker and Disease
Reaction

 

 

Population X‘-value
BCze 16.00****
BC3F1 12.00****
Total 28.00****

 

**** marker and disease reaction are dependent at 01<.001, *** marker and disease reaction are dependent
at 01=.005,**marker and disease reaction are dependent at 01:01 * marker and disease reaction are
dependent at 0t=.025

Yunguilla Field Inoculation With Race 0

The Yunguilla parent, the BC2 and BC 3 populations were resistant to race 0 and
neither the pods nor the leaves had susceptible symptoms. Since Yunguilla apparently
carries resistance to race 0 it was not possible to distinguish plants that were carrying Co-

42

Yunguilla Marker Analysis and Greenhouse Inoculation with Race 2

The results of marker analysis and greenhouse inoculation with race 2 for the
Yunguilla populations are given in Table 16. Yunguilla was found to be susceptible to a
Mexican race 2 isolate of C. lindemuthianum. Plants from the BCze and BC3F1
populations exhibited a clear resistant or susceptible reaction to the isolate, with some
plants from each population showing a completely resistant reaction identical to that of

SEL 1308 and the remainder exhibiting susceptibility identical to the Ecuadorian parent.

Table 16. Yunguilla BCze and BC3F2 Marker Analysis and Reaction to Race 2

 

 

 

. . ASI3 and SH18 Markers
Disease Reaction
Present Absent
BC ze 12 2
Resistant BC 3F1 8 1
Total 20 3
BCze 0 2
Susceptible BC 3F1 0 2
Total 0 4

 

40

As shown in Table 17, positive results for the AS13 and SH18 markers are
strongly associated with resistance in the Yunguilla populations. Results from the x2 Test
for Independence indicate that in each of the populations, the marker results and disease
reactions are dependent. All plants carrying the two markers for Co-42 were resistant in
the greenhouse inoculations, indicating that C 0-42 functions as expected in the Yunguilla
genetic background. Some of the plants from which the marker was absent were still
resistant. Other anthracnose resistance genes from the Red Hawk or Je.Ma. parents,
which may be segregating in the populations, could confer this additional resistance that

is unlinked to C 0-42 .

Table 17. Yunguilla BCng and BC3F2 Test for Dependence of Marker and Disease
Reaction

 

 

Population XI-value
BCze 6.85””
BC 3F1 652*

Total l3.42****

 

**** marker and disease reaction are dependent at 01<.001, *** marker and disease reaction are dependent
at a=.005,**marker and disease reaction are dependent at a=.01 * marker and disease reaction are
dependent at 0t=.025

Genuine Field Inoculation with Race 0

The Genuine parent, the BC; and BC3 populations were resistant to race 0 and
neither the pods nor the leaves had susceptible symptoms. Since Genuine apparently
carries resistance to race 0 it was not possible to distinguish plants that were carrying Co-
42. The plants did suffer from bronzing of the pods and leaves, possibly due to sunscald.
Genuine, which was bred for a sea level production area (Florida), may be poorly adapted
to the high light intensity conditions at 2700 m.a.s.l. Consequently, Genuine may not be

as usefiil in the Ecuadorian breeding program as was hoped.

41

Evaluation of Seed Appearance

C ocacho Seed Appearance

The Cocacho backcross populations produced seed in eight different seed coat
pattern and color classes, which are shown in Figure 8: (1) solid yellow with darkened tan
veins and an orange/brown corona, this was considered the parental class (2) yellow and
white mottle with darkened tan veins and an orange/brown corona (3) solid brown with
dark brown veins (4) brown and cream mottle with an orange/brown corona (5) red and
dark red mottle (6) solid red (7) solid cream with tan veins and an orange/brown corona
(8) gray purple with dark veins lightening to cream at edges with an orange/brown
corona.

The frequency of each seed type in the two backcross populations is shown in
Table 18. Certain seed types, 5, 6, 7 and 8, were present only in the BCze population,
possibly because they result from homozygous alleles from the donor parent. The
yellow/white mottle pattern may segregate 1:2:1 solid yellow : yellow/white mottle :
white when selfed. The yellow/white mottle seed type was the most frequent class in the
BCze population, 26%, while the parental seed type was rather infrequent, 8%. The
parental type was much more frequent in the BC3F1 population at 41%. The second most
frequent class in the BC3F 1 population, yellow/white mottle at 39%, may segregate and

produce parental type plants in future generations after further selfing.

42

Table 18. Seed appearance frequencies in the Cocacho populations

 

 

 

Frequency % of Population
Seed Appearance B Cze B C3F1 B Cze B C3F1
2 solid yellow; tan ve1ns; orange/brown 5 37 8.06 40.66
corona
3 yellow/whlte mottle; tan 16 3 5 25.81 38.46
ve1ns;orange/brown corona
4 solid brown; dark brown veins 8 10 12.90 10.99
5 brown/cream mottle; orange/brown 13 6 20.97 6.59
corona
6 red/dark red mottle 3 0 4.84 0.00
7 solid red 2 0 3.23 0.00
8 solld cream; tan ve1ns; orange/brown 11 0 17.74 0.00
corona
9 gray purple; darkened ve1ns; cream 1 0 1.61 0.00
edges; orange/brown corona.
Missing (plant did not produce seed) 3 3 4.84 3.30

 

ACE-1 Seed Appearance

The ACE-1 backcross populations produced seed in seven different seed coat
pattern and color classes, which are shown in Figure 9: (2) solid yellow with tan veins
and a tan/yellow corona, which was considered the parental class (3) solid cream with tan
veins and a tan/yellow corona (4) yellow and white mottle with tan veins and a
tan/yellow corona (5) brown and cream mottle with a tan/yellow corona (6) tan/yellow
with yellow green edges and tan/yellow corona (7) solid brown with dark brown veins (8)
sulfur yellow with tan veins and tan/yellow corona.

The frequency of each seed type in the two backcross populations is shown in
Table 19. Certain seed types, 2 and 6, were present only in the BCze population,

possibly because they result from homozygous alleles from the donor parent. The

43

yellow/white mottle pattern may segregate 1:2:1 solid yellow : yellow/white mottle :
white. The solid yellow, yellow/white mottle and solid cream seed types were the most
frequent class in the BCze population, at 25%, 28% and 27% respectively. Although
fairly frequent in the BC 2F; population the parental type was more frequent in the BC 3F1
population at 48%. The second most frequent class in the BC3F1 population,
yellow/white mottle at 21%, may segregate and should produce parental type plants in
future generations as it is selfed. The cream-colored seed class may also be useful to
breeders at INIAP, as this seed resembles Bola Blanca, another Ecuadorian commercial

class.

Table 19. Seed appearance frequencies in the ACE-1 populations

 

 

 

Frequency % of Population
Seed Appearance B Cze B C3F1 B Cze B GE
2 solid yellow; tan ve1ns; tan/yellow 28 14 23.73 50.00
corona
3 SOlld cream; tan ve1ns; tan/yellow 32 0 27.12 0.00
corona
4 yellow/white mottle; tan ve1ns; 33 6 27.97 21.43
tan/yellow corona
5 brown/cream mottle; tan/yellow corona 13 2 11.02 7.14
6 tan/yellow; yellow green edges; 1 0 0.85 0.00
tan/yellow corona
7 solid brown; dark brown veins 2 0 1.69 0.00
8 sulfur yellow; tan ve1ns; tan/yellow 4 5 3.39 17.86
corona
Missing (plant did not produce seed) 5 1 4.24 3.57

 

44

Paragachi Seed Appearance

The Paragachi backcross populations produced seed in three different seed coat
pattern and color classes, which are shown in Figure 10: (1) dark purplish red, red and
white mottle (2) dark purplish red and white mottle (3) white with a white corona.

Table 20 shows the frequency of each seed class in the two backcross populations.
The white seed type was observed only in the BC; population. The red—purple and white
mottle was the most frequent class in the BC; population at 42%. Only 10% of the plants
in this population produced seed in the parental class. As expected, in the BC 3 generation

the parental seed type was much more frequent at 44%.

Table 20. Seed appearance frequencies in the Paragachi populations

 

 

 

1.51:1“121, 3311;011:311
l purplish red/red/white mottle 5 8 10.42 44.44
2 purplish red/white mottle 20 3 41.67 16.67
3 white 14 0 29.17 0.00

Missing (plant did not produce seed) 9 7 18.75 38.89

 

ARME-Z Seed Appearance

The ARME-2 backcross populations produced seed in three different seed coat
pattern and color classes, which are shown in Figure 11: (2) dark purplish red, light red
and white mottle (3) dark purplish red and white mottle (4) white with a white corona.

Table 21 shows the frequency of each seed class in the two backcross populations.

The white seed type was observed only in the BC; population. The red-purple and white

45

mottle was the most frequent class in the BC; population at 52%. Twenty-two percent of
the plants in this population produced seed in the parental class. In the BC3 generation
the parental seed type was much more frequent at 56%. The two colored mottle seeds
(class two) may segregate 1:2:1 for three-colored mottle : two-color mottle : white, upon

further selfing.

Table 21. Seed appearance frequencies in the ARME-2 populations

 

 

 

Seed Appearance B cfpffquenlgap‘, g/Efézpop “33:1?!
2 purplish red/red/white mottle l6 9 21.92 56.25
3 purplish red/white mottle 38 7 52.05 43.75
4 white 17 0 23.29 0.00
Missing (plant did not produce seed) 2 0 2.74 0.00

 

Yunguilla Seed Appearance

The Yunguilla backcross populations produced seed in five different seed coat
pattern and color classes, which are shown in Figure 12: (2) red, pink and white mottle
(3) pink and white mottle (4) white with a yellow/brown corona (5) dark purple and white
mottle (6) dark purple, purple and white mottle.

Table 22 shows the frequency of each class in the two backcross populations.

The white and two color purple mottle seed types were observed only in the BC;
population. The white seed type was the most frequent class in the BC; population at
29%. Only 16% of the plants in this population produced seed equivalent to the parental
class. In the BC 3 generation the parental seed type was much more frequent at 43%. The

two colored mottle seeds (classes 2 and 5) may segregate 1:2:1 for three-colored mottle to

46

two-color mottle to white upon further selfing. The three-color dark purple mottle seed
may also be useful in INIAP’s bean breeding program, as this seed resembles another

Ecuadorian commercial class, Calima Negro.

Table 22. Seed appearance frequencies in the Yunguilla populations

 

 

 

Seed Appearance B cfgfquengé3pl géfézpopuggfgl
2 red/pink/white mottle 5 9 16.13 42.86
3 pink/white mottle 6 10 19.35 47.62
4 white; yellow/brown corona 9 0 29.03 0.00
5 dark purple/white mottle 6 0 19.35 0.00
6 dark purple/purple/white mottle 2 2 6.45 9.52
Missing (plant did not produce seed) 3 0 9.68 0.00

 

Genuine Seed Appearance

Both of the Genuine backcross populations produced only white seed that

resembled seed of the parent as shown in Figure 13.

Figures 8 through 13 in this thesis are presented in color.

47

Figure 8. Seed of Cocacho parent (1) and BC2 and BC3 plants

  

Figure 9. Seed of ACE-1 parent (1) and BC2 and BC3 plants

 

48

Figure 10. Seed of Paragachi BC2 and BC3 plants

  

Figure 1 1. Seed of ARME-2 parent (1) and BC2 and BC3 plants

 

49

 

Figure 13. Seed of Genuine parent (1) and BC2 (2) and BC3 plants (3)

 

50

Farmer Evaluation of Seed from Cocacho, ACE-1 and Yunguilla Backcross Populations

Figure 14 shows the farmers’ ranking of bean samples as determined using the
Bradley-Terry model for pairwise comparisons modified to account for ties (Dittrich et
al., 1998; Coe, 2002). From among the Cocacho derived samples, farmers preferred the
brown seed type. Some farmers noted that brown beans are traditionally grown in the
area, which may explain farmers’ interest in this seed type. The sample resembling the
Cocacho parent and the parent itself were the second and third ranked samples
respectively. The majority of the farmers, 81%, tied these two samples in their ranking
and some indicated that they did
see a difference between the two samples. This would suggest that the sample classified
as the parental type sufficiently resembles the Ecuadorian recurrent parent for it to be
considered the same commercial class.

Among the samples derived from the ACE-1 recurrent parent the parent itself
received the highest ranking from farmers. The white, tan and yellow samples (I, F and
H) are essentially tied for the second place ranking. Fifty-seven percent of the farmers
ranked the seed of ACE-1 higher than the yellow—seeded sample and 36% tied the two
samples. In this case the farmers were able to differentiate between the seed of the AC E-
1 parent and the seed classified as the parental type and they preferred the seed type of
the parent. This would suggest that the parental type sample used in the survey lacked
some seed quality traits that are important to the farmers. The seed size of the samples
was slightly different (Figure 2), and the larger size of the ACE-1 seed may be one reason
that farmers preferred it. Seed which more closely resembled the ACE-l parent (in the

author’s opinion) was harvested after the survey was completed (Figure9).

51

Figure 14. F armers’ ranking of seed samples from the Cocacho, ACE-1, and Yunguilla
backcross populations and their respective Ecuadorian recurrent parent using Bradley-

 

Terry modeling
of 1.30
D 0.89
C 0.73
I 0.62
E 0.59 F0 060
H 0.60
A 0.24
B 0 J I 0
Legend C ocacho ACE-1

K a 0.55

 

M 0.17

.I.

Yunguilla

 

Cocacho Samples

ACE-1 Samples

Yunguilla Samples

 

 

B white/yellow mottle

 

J white/yellow mottle

 

D brown G ACE-1, parent K red/pink/white mottle*
C yellow* I white M red/white mottle

E Cocacho, parent F tan, yellow edges L Yunguilla, parent

A white H yellow*

 

*sample which most resembled the recurrent parent

Among the samples derived from Yunguilla, the farmers preferred the seed which

was classified as the parental type and ranked the seed of the parent last (Figure 14).

F ifty-three percent of the farmers preferred the parental type sample (K) and 33% tied the

two samples. Some farmers commented that the three Yunguilla class seed samples were

very similar, but others differentiated between them. Based on their comments, farmers

differentiated between the seed samples on the basis of color. Bright colors and less

52

 

white were two particular traits that farmers reported using to differentiate the samples.
These results suggest that some seed from the Yunguilla derived backcross plants is

acceptable to farmers and has reached commercial quality for the Calima Rojo class.

Number of Backcrosses

With the exception of the Genuine populations, the percent of plants producing seed
resembling the recurrent parent ranged from 8 to 24 % in the BC; populations and from
41% to 56% in the BC 3 populations. In each case there was at least a two-fold increase in
the percent of plants in the parental class. Two backcrosses (and one generation of
selfing) were sufficient to obtain some plants with seed like the recurrent parent.
However, three backcrosses resulted in a substantial increase in the frequency of the
parental type seed. Three backcrosses more effectively recovered the desired seed
quality of the Ecuadorian varieties than did two backcrosses. Additionally the adaptation
of the backcross populations to the target environment must be considered, as BC 3
populations may be more adapted than BC2 populations. Adaptation was not tested in
this study since the field experiment location was at a higher elevation than bush beans

are typically grown.

53

Conclusions

The objective of this study was to introduce a Mesoamerican anthracnose
resistance gene, Co-42, into Ecuadorian bean market classes using molecular markers to
select for the gene and multiple backcrosses to the Ecuadorian recurrent parent in order to
obtain lines with commercial seed characteristics and local adaptation.

Six different Ecuadorian varieties were used as recurrent parents. In backcross
populations generated from the varieties Yunguilla (a red Calima type) and ARME-2 (a
medium red mottle type) plants carrying the markers for Co-42 were resistant to
anthracnose in greenhouse inoculations. However, the marker assisted backcrossing did
not work as expected in the Cocacho and Paragachi derived populations. In the Cocacho
backcross populations, plants carrying the markers for Co-42 were no more likely to be
anthracnose resistant, while in the Paragachi populations the markers for Co-42 were
associated with partial resistance but none of the plants in the populations were
completely resistant to anthracnose.

One possible explanation for this observation is that in these populations a
recombination between the Co-42 gene and the markers occurred early in population
development. However this is unlikely since the SASl3 is tightly linked to Co-42.
Additionally, multiple plants carrying the markers for Co-42 were used as parents in each
generation (Table 4), which means that recombination would have had to occur multiple
times.

Another explanation is that Cocacho and Paragachi lack certain genes necessary
for the function of Co-42, possibly because of their genetic distance from G 2333, the

source of C 0-42 . In support of this theory, evidence exists for the presence of

54

independent complimentary genes for anthracnose resistance in the Andean cultivars
Kaboon (Muhalet et al., 1981; Melotto and Kelly, 2000), and Algarrobo (Cardenas et al.,
1964). Inoculation of resistant by susceptible crosses using these cultivars as sources for
anthracnose resistance yielded resistant : susceptible segregation ratios of 9:7 or 57:7,
which suggest that two independent genes are required for resistance. It is proposed that
Co-42 requires the presence of another complimentary gene(s) which is present in Middle
American gerrnplasm but absent in certain Andean genotypes. Using the molecular
markers for Co-42 resulted in selection for only this gene, so in resulting populations the
complimentary gene(s) was not retained and Co-42 was not functional (Cocacho) or only
partially functional (Paragachi). ARME-2 and Yunguilla, which were recently selected
from CIAT breeding lines are more likely to possess some Mesoamerican genes.

If Co-42 requires other genes, which are absent from unimproved Andean bean
germplasm, this particular gene may not be as useful a source of anthracnose resistance in
certain Andean genotypes. The situation, at least, limits the usefulness of marker-assisted
selection for introgression of this gene into genotypes which lack the gene(s) that
compliments Co—4". In the case of the Paragachi and Cocacho populations described
here, direct selection through disease inoculation would likely have produced useful
anthracnose resistant backcross plants; whereas marker assisted selection did not. This
highlights the necessity of incorporating direct selection, through disease screening, into
breeding projects utilizing molecular markers to select for disease resistance genes. The
apparent failure of C 0-42 in certain genetic backgrounds also highlights the importance of

identifying new and diverse sources of anthracnose resistance.

55

For all of the Ecuadorian parents used in the study, two backcrosses and one
generation of selfing was sufficient to obtain some plants which produced seed of
commercial quality, that was acceptable to farmers. However, the proportion of plants
producing parental type seed increased dramatically with the third backcross.

The plants carrying C 0-42 from the ARME-2 and Yunguilla backcross
populations will be useful in the INIAP bean breeding program as potential breeding
lines or new varieties with superior anthracnose resistance. Both Yunguilla and ARME-
2 carry uncharacterized anthracnose resistance genes which, when pyramided with Co-42,
should confer resistance to all reported races of the pathogen in Ecuador. The Cocacho
and Paragachi backcross plants with promising agronomic qualities, parental type seed,
and the markers for Co-42 should be crossed with resistant ARME-2 or Yunguilla (and
possibly ACE-l) backcross plants carrying Co-42. In this way the gene or genes
complimentary to C 0-42 can be introduced from the ARME-2 and Yunguilla backcross
plants without reintroducing genes from unadapted gennplasm (i.e. SEL1308 and

RH*2/SEL 1308).

56

Literature Cited

Adam-Blondon AF, Sevignac M, Bannerot H, Dron M (1994) SCAR, RAPD and RFLP
markers linked to a dominant gene (Are) conferring resistance to anthracnose.

Theor Appl Genet 88:865-870

Afanador LK, Haley SD, Kelly JD (1993) Adoption of mini-prep DNA extraction
method for RAPD marker analysis in common bean (Phaseolus vulgaris). Ann

Rep Bean Improv Coop 36:10-11

Alzate-Marin AL, de Almeida KS, de Barros EG, Moreira MA (2001) Identification of a
recessive gene confening resistance to anthracnose in common bean lines derived
from the differential cultivar AB 136. Annu Rep Bean Improv C00p 442117-118

Alzate—Marin AL, Arruda KM, de Barros EG, Moreira MA (2003a) Allelism studies for
anthracnose resistance genes of common bean cultivar AND 277. Annu Rep

Bean Improv Coop 46:173-174

Alzate-Marin AL, Costa MR, Arruda KM, de Barros EG, Moreira MA (2003b)
Characterization of the anthracnose resistance gene present in Ouro Negro

(Honduras 35) common bean cultivar. Euphytica 1331165-169

Awale HE, Kelly JD (2001) Development of SCAR markers linked to Co-42 gene in
common bean. Ann Rep Bean Improv Coop 442119-120

Balardin RS, J arosz AM, Kelly JD (1997) Virulence and molecular diversity in
Colletotrichum lindemuthianum South, Central, and North America.

Phytopathology 87:1184-1191.

Balardin RS, Kelly JD (1998) Interaction between Colletotrichum lindemuthianum races
and gene pool diversity in Phaseolus vulgaris. J Amer Soc Hort Sci 123:1038-

1047.

Beebe S, Skroch PW, Tohme J, Duque MC, Pedraza F, Nienhuis J (2000) Structure of
genetic diversity among common bean landraces of Middle American origin
based on correspondence analysis of RAPD. Crop Sci 40:264-273

Beebe S, Rengifo J, Gaitan E, Duque MC, Tohme J (2001) Diversity and origin of
Andean landraces of common bean. Crop Sci 41: 854-862

Beebe S, Tohme J, Nienhuis J, Pedraza F, Rengifo J, Tovar E, Islam A (2004) Studies in
Phaseolus germplasm diversity: A review of work at CIAT. Ann Rep Bean

Improv Coop vol. 47

Bl‘oughton WJ, Hernandez G, Blair M, Beebe S, Gepts P, Vanderleyden J (2003) Beans
(Phaseolus spp.) — model food legumes. Plant and Soil 252:55-128

57

Cardenas F, Adams MW, Andersen A (1964) The genetic system for reaction of field
beans (Phaseolus vulgaris L.) to infection by three physiologic races of
Colletotrichum lindemuthianum. Euphytica 132178-186

Coe R (2002) Analyzing ranking and rating data from participatory on-farm trials. In
Bellon MR, Reeves J (eds) Quantitative analysis of data from participatory
methods in plant breeding. International Maize and Wheat Improvement Center,
Mexico pgs 44-65

Dittrich R, Hatzinger R, Katzenbeisser W (1998) Modeling the effect of subject-specific
covariates in paired comparison studies with an application to university rankings.
Appl Statist 47:511-525

Drij flrout E, Davis JHC (1989) Selection of a new set of homogeneously reacting bean
(Phaseolus vulgaris) differentials to differentiate races of Colletotrichum
lindemuthianum. Plant Path 38:391-396

Fabre JV, Julien J, Parisot D, Dron M (1995) Analysis of diverse isolates of
Colletotrichum lindemuthianum infecting common bean using molecular markers.
Mycol Res 992429-435

F alconi E, Ochoa J, Peralta E, Danial D (2003) Virulence pattern of Colletotrichum
lindemuthianum in common bean in Ecuador. Ann Rep Bean Improv Coop vol.
46

Fouilloux G (1976) Bean anthracnose: New genes of resistance. Ann Rep
Bean Improv Coop 19:36-37

Geffroy V, Sicard D, de Oliveira J CF , Sévignac M, Cohen S, Gepts P, Neema C, Langin
T, Dron M (1999) Identification of an ancestral resistance gene cluster involved
in the coevolution process between Phaseolus vulgaris and its fiingal pathogen
Colletotrichum lindemuthianum. Mol Plant-Microbe Interact 12:774-784

Geffroy V, Sévignac M, de Oliveira JCF, Fouilloux G, Skroch P, Thoquet P, Gepts P,
Langin T, Dron M (2000) Inheritance of partial resistance against
Colletotrichum lindemuthianum in Phaseolus vulgaris and co-localization of
quantitative trait loci with genes involved in specific resistance. Mol Plant-
Microbe Interact 13 :287-296

INEC-MAG-SICA (2002) III Censo Nacional Agropecuario.
(www.sica.gov.ec/censo/index.htm)

INIAP (1995) Identificacion de razas fisiologicas del hongo Colletotrichum

lindemuthianum en el Ecuador. In: Estacion Experimental Santa Catalina
Programa de Leguminosas Infonne Annual 1994. INIAP, Quito, Ecuador, pp 1-4

58

Kelly JD (1995) Use of random amplified polymorphic DNA markers in breeding for
major gene resistance to plant pathogens. HortScience 30:461-465

Kelly JD (1997) A review of varietal response to bean common potyvirus in Phaseolus
vulgaris. Plant Varieties and Seeds 10: 1-6

Kelly JD, Afanador L, Cameron LS (1994) New races of Colletotrichum
lindemuthianum in Michigan and implications in dry bean resistance breeding.
Plant Disease 782892-894

Kelly JD, Hosfield GL, Vamer GV, Uebersax MA, Long RA, Taylor J (1998)
Registration of ‘Red Hawk’ dark red kidney bean. Crop Sci 38:280-281

Kelly JD, Vallejo VA (2004) A comprehensive review of the major genes conditioning
resistance to anthracnose in common bean. HortScience (in press)

Li ZK, Luo L], Mei HW, Paterson AH, Zhao XH, Zhong DB, Wang YP, Yu XQ, Zhu L,
Tabien R, Stansel JW, Ying CS (1999) A “defeated” rice resistance gene acts as a
QTL against a virulent strain of Xanthomonas oryzae pv.oryzae. Mol Gen Genet
261 :58-63

Melotto M, Kelly JD (2000) Allelic series at the Co-I locus conditioning resistance to
anthracnose in common bean of Andean origin. Euphytica 116: 143-149

Melotto M, Kelly JD (2001) Fine mapping of the Co-4 locus of common bean reveals a
resistance gene candidate, C 0K-4, that encodes for a protein kinase. Theor Appl
Genet 103:508-517

Michelmore R, Meyers BC (1998) Clusters of resistance genes in plants evolve by
divergent selection and a birth-and-death process. 8:1113-1130

Miklas PN, Kelly JD (2002) The use of MAS to develop pinto bean germplasm
possessing Co-42 gene for anthracnose resistance. Ann Rep Bean Improv Coop
45:68-69

Muhalet CS, Adams MW, Saettler AW, Ghaderi A (1981) Genetic system for the
reaction of field beans to beta, gamma, and delta races of Colletotrichum
lindemuthianum. J Amer Soc Hort Sci 106:601-604

Murillo A (2002) Uso de germoplasma exotico para ampliar la base genética para
resistencia a roya y otras enfermedades de fréjol comun (Phaseolus vulgaris L.)
(16 Ecuador. Thesis: Universidad de Puerto Rico, Recinto Universitario de
Mayaguez.

59

Nass HA, Pedersen WL, MacKenzie DR, Nelson RR (1981) The residual effects of some
“defeated” powdery mildew resistance genes in isolines of winter wheat.
Phytopathology 71:1315-1318

Nodari RO, Tsai RM, Gilbertson RL, Gepts P (1993) Towards an integrated linkage
map of common bean. 2. Development of an RF LP-based linkage map. Theor
Appl Genet 85:513-520

O’Connell RJ, Bailey JA (1991) Hemibiotrophy in Colletotrichum lindemuthianum.
Pages 211-222 In: Electron Microscopy of Plant Pathogens. Mudgen and
Risemann eds. Springer-Verlag, New York

Ochoa J, Cruz E, Danial D (2002) Physiological variation of bean rust in Ecuador. Ann
Rep Bean Improv Coop 45:138-139

Pastor-Corrales MA, Tu J C (1989) Anthracnose. In: Schwartz HF, Pastor-Corrales MA
(eds) Bean production problems in the tropics. Centro Intemacional de
Agricultura Tropical, Cali, Colombia, pp 77-104

Pastor-Corrales MA (1991) Estandarizacion de variedades diferenciales y de
designacion de razas de Colletotrichum lindemuthianum. Phytopathology 81 :694
(abstract)

Pastor-Corrales MA, Erazo OA, Estrada EI, Singh SP (1994) Inheritance of anthracnose
resistance in common bean accession G2333. Plant Dis 78:959-962

Peralta E, Murillo A, Pinzon J (1997) Fréjol, su importancia economica, ecologica y
social y la oferta tecnologica del programa nacional de leguminosas. In: Danial D
(ed) Primer Taller de Preduza en Resistencia Duradera en Cultivos Altos e1 la
Zona Andina. Quito, Ecuador.

van Schoonhoven A, Voysest O (1989) Common beans in Latin America and their
constraints. . In: Schwartz HF, Pastor-Corrales MA (eds) Bean production
problems in the tropics. Centro Intemacional de Agricultura Tropical, Cali,
Colombia, pp33-57

Schwartz HF, Galvez GE, van Schoonhoven A, Howeler RH, Graham PH, Flor C (1978)
Field problems of beans in Latin America. International Center for Tropical
Agriculture: Cali, Colombia

Schwartz HF, Pastor-Corrales MA, Singh SP (1982) New sources of resistance to
anthracnose and angular leaf spot. Euphytica 31:741-754

Sicard D, Michalakis Y, Dron M, Neema C (1997) Genetic diversity and pathogenic

variation of Colletotrichum lindemuthianum in the three centers of diversity of its
host, Phaseolus vulgaris. Phytopathology 87:807-813

60

Singh SP (1982) A key for identification of different growth habits of Phaseolus vulgaris
L. Ann Rep Bean Improv Coop 25:92-95

Singh SP, Gutierrez JA, Molina A, Urrea C, Gepts P (1991) Genetic diversity in
cultivated common bean: II. Marker-based analysis of morphological and
agronomic traits. Crop Sci. 31:23-29

Staub J E, Serquen F C (1996) Genetic markers, map construction, and their application in
plant breeding. HortScience 31(5):729-741

Stavely JR, Pastor-Corrales MA (1989) Rust. In: Schwartz HF, Pastor-Corrales MA
(eds) Bean production problems in the tropics. Centro Intemacional de
Agricultura Tropical, Cali, Colombia, pp159-194

Tu J C (1992) Colletotrichum lindemuthianum on bean: Population dynamics of the
pathogen and breeding for resistance. In: Bailey J A, J eger MJ (Eds)
Colletotrichum: Biology, pathology and control. C .A.B. International:
Wallingford, UK

Tu J C, McNaughton ME (1980) Isolation and characterization of benomyl-resistant
biotypes of the delta race of Colletotrichum lindemuthianum. Can J Plant Sci
60:585-5 89

Vallejo V, Kelly JD (2001) Development of a SCAR marker linked to the Co-5 locus in
common bean. Ann Rep Bean Improv Coop 44:121-122

Vallejo V, Kelly JD (2002) The use of AFLP analysis to tag the Co-12 gene
conditioning resistance to bean anthracnose. (http://www.intl-
pag.org/pag/ 1 0/abstracts/PAGX_P233.html) Plant and Animal Genome X
Conference 2002, San Diego, CA

Voysest CV (2000) Mejoramiento genético del frijol. International Center for Tropical
Agriculture, Cali, Colombia.

Wallace DH, Gniffl(e PA, Masaya PN, Zobel RW (1991) Photoperiod, temperature, and
genotype interaction effects on days and nodes required for flowering of bean. J
Amer Soc Hort Sci 116:534-543

van der Westhuizen AJ, Qian XM, Botha AM (1998) Differential induction of apoplastic
peroxidase and chitinase activities in susceptible and resistant wheat cultivars by

Russian wheat aphid infestation. Plant Cell Reports 18: 132-137

Wilson WA, McMullen MS (1997) Dosage dependent genetic suppression of oat crown
rust resistance gene Pc-62. Crop Sci 37:1699-1705

61

Young RA, Kelly JD (1996) RAPD markers flanking the Are gene for anthracnose
resistance in common bean. J Amer Soc Hort Sci 121 :37-41

Young RA, Kelly JD (1997) RAPD markers linked to three major anthracnose
resistance genes in common bean. Crop Sci 37:940-946

Young RA, Melotto M, Nodari RO, Kelly JD (1998) Marker-assisted dissection of the
oligogenic anthracnose resistance in the common bean cultivar, ‘G2333.’ Theor
Appl Genet 96:87-94

62

CHAPTER 2
USING PARTICIPATORY CROP IMPROVEMENT METHODS IN THE MIRA AND
CHOTA VALLEYS OF ECUADOR

Introduction

Participatory Crop Improvement

Improved crop varieties are an important part of agricultural technology change,
and one that can particularly benefit small-scale farmers with limited resources. For the
cost of the improved seed, new varieties with higher yields, pest or disease resistance, or
tolerance to stresses such as drought or cold can dramatically increase a farmer’s
productivity. No additional investment is required to use the new variety -- a farmer can
simply plant the crop as in the past. If farmers maintain their own seed, a one time
expense will provide benefit for years. In order for the potential impact of improved
varieties to be realized, however, farmers must have access to varieties that will perform
competitively under their environmental conditions and meet quality requirements set by
the farmer or target markets. Participatory Crop Improvement (PCI), the practice of
involving farmers in the development or testing of crop varieties, is one methodology
which is being used to give farmers access to better crop varieties.

Farmer involvement in plant breeding can increase the effectiveness with which
improved varieties are selected by 1) identifying and taking into consideration farmers’
variety selection criteria and 2) providing opportunity for selection under farmers’

growing conditions.

63

Farmer vs Breeder Selection Criteria

Plant breeders who have implemented PCI often find that their variety selection
criteria differ from farmers’ selection criteria. Generally, plant breeders are concerned
with improving a crop’s yield or stress tolerance (such as drought tolerance or disease
resistance), whereas farmers tend to be concerned with more obvious traits (such as seed
quality). Dissimilarity between farmer and breeder selection criteria can occur in the
following circumstances:

Crop has alternative uses. Small-scale farmers sometimes have multiple uses for
a crop, not all of which are obvious to plant breeders. Traits related to the crop’s
alternative uses may not be considered by breeders. In Morocco, barley grain is used for
human consumption and the leaves, grain and straw are used as forage for sheep. When
making selections from a segregating barley population, Moroccan farmers mentioned
grain yield as a selection criterion 40 times and straw yield 45 times. Moroccan farmers
were most interested in barley varieties with both good forage and good grain production
(Ceccarelli et al., 2001).

Stringent quality requirements exist for market or home consumption. In some
regions, a crop’s users (be they the farm family or urban consumers) have particular
quality preferences. These quality requirements are not always apparent to plant breeders
but are important to farmers. In Colombia, farmers grow dry beans as a cash crOp.
Consumers prefer solid red or red mottle seed types, and the seed appearance determines
the price a farmer gets for his crop. In a comparison of advanced lines selected by
farmers and breeders (from the same initial crosses), farrner-selected lines produced

larger seeds with more attractive colors and patterns. In a trial of the farmer and breeder

64

selected lines farmers preferred varieties that produced attractive seed, which would bring
a higher market price, despite the varieties’ lower yields (Komegay et al., 1996).

Certain traits are useful in view of the farmers ' production practices or
environment. Plant breeders cannot be fully aware of all farmers’ agricultural practices
and problems. Therefore, they cannot always know what traits are necessary or could be
useful from the perspective of a farmer’s agroecosystem. During a participatory trial of
rice varieties in eastern India, farmers were interested in a particular variety which had a
purple sheath. This trait allowed the farmers to differentiate between the cultivated
variety and wild rice plants during weeding (Curtois et al., 2001).

Breeders and farmers differ in their understanding of traits and their inheritance.
Plant breeders often focus their efforts on developing varieties with stress tolerance (such
as drought tolerance or disease resistance) which are not immediately apparent to
farmers. Barley breeders working with Moroccan farmers found that farmers did not
realize that varieties could differ in their degree of disease resistance. Instead the farmers
thought that disease occurrence depends only on weather conditions. Consequently, only
a few farmers considered disease incidence when making selections from a segregating
population, despite the presence of two barley diseases in the trial plots (Ceccarelli et al.,
2001). Similarly, farmers in the Great Lakes Region of Africa believe the complex of
diseases that affects their bean crOp to result from “different types of rain” or soil fertility
problems (Trutmann, 1996). Plant breeders’ understanding of disease resistance genes
and their inheritance allow for the development of varieties with specific disease
resistance. However, farmers may not recognize the benefit of such traits unless the

resistant variety’s disease tolerance is proven in a field trial.

65

In the aforementioned examples researchers used farmer interviews to better
understand farmers selection criteria and variety preferences. With such information
breeders can (a) incorporate farmer selection criteria into a formal breeding program, (b)
recommend varieties that are better suited to farmers’ needs, (c) choose parental material
that will produce useful variability for future breeding projects.

On-Farm vs. Research Station Selection Environment

PCI often involves testing varieties or making selections under farmers’ field
conditions. The environment in which selection occurs has a substantial impact on the
resulting varieties due to genotype by environment interaction. A variety selected in one
environment may not perform equally well in a location with different conditions. In
mountainous areas with many different microclimates, the diversity of environmental
conditions can make breeding adapted varieties more difficult. Through on-farrn testing
of varieties, farmers and plant breeders can select those varieties which perform best
under farmers’ agricultural conditions. Australian wheat breeders found that they were
better able to increase yields under low, input rainfed conditions when they moved their
trials from research stations to farmers fields (Cooper et al., 1997; Banziger and Cooper,
2001)

On-fann testing and a PCI approach allow farmers and breeders to better take
advantage of repeatable genotype by environment interaction than is possible with formal
plant breeding. Formal plant breeding programs tend to select varieties with broad
adaptation, which will perform well in many environments. Meanwhile, varieties with
specific adaptation, which perform very well in some environments but not others, are

often overlooked.

66

Despite the opportunity for very specific adaptation through PCI, varieties
selected by this method are often more broadly adapted than might be expected.
Examples of extremely narrowly adapted varieties are rare. However, such varieties are
adopted when environmental constraints severely restrict the number of compatible
genotypes (Atlin et al., 2001). Varieties that have tolerance to particular stresses but
maintain a broader adaptation are more commonly chosen through PCI. Additionally,
there are instances of very broadly adapted varieties that were developed through PCI.
The three following examples illustrate the range of adaptation in varieties developed
through PCI.

A narrowly adapted Nepalese cold water rice variety. In a PCI program aimed at
developing rice varieties for high altitude areas of Nepal, one participating farmer chose a
plot with cold water and high nutrient conditions to test the segregating bulk variety,
Nilgiri-l, that he received from researchers. Rice tends to produce sterile spikelets under
such conditions and, in fact, the farmer had previously tried to grow rice in this field, but
it never produced any grain. Unexpectedly, the variety Nilgiri-l, yielded 3.3 tons of
grain per hectare in the problem field. The farmer continued to grow his selections from
the Nilgiri-l bulk, but only in the problem field because the grain was prone to shattering
and had a poor taste (Sthapit et al., 1996). Nilgiri-l is an example of a very narrowly
adapted variety. Its desirable ability to withstand unfavorable environmental conditions
but undesirable taste and shattering make it useful in very few instances.

More broadly adapted Nepalese rice varieties bred for cold tolerance. In the
same Nepalese rice improvement program, two of the farmer selected varieties were

more broadly adapted than Nilgiri-l. The variety M-3 was selected by farmers from

67

segregating bulk varieties that they planted in their own fields. The variety M-9 was
selected by farmers from on station and on-fann variety trials. These two varieties were
adopted by farmers from villages at a range of altitudes, although the varieties were most
widely adopted in mid-altitude villages where they yielded more than the local landraces.
This is particularly interesting since M-3 was originally selected by farmers in high
altitude villages where it did not yield better than the local red grained landraces but did
have a preferred white grain color. The two varieties were not widely adopted in low
altitude villages where they did not yield as well (Joshi et al., 2001). M-3 and M-9 are
examples of more broadly adapted varieties. They were bred for a high altitude
environment, but also performed well at mid-range altitudes. However, M-3 and M-9’s
lower yields and adoption rates at low altitudes show that the varieties are not adapted to
these conditions.

A very broadly adapted Brazilian maize variety bred for low nitrogen tolerance.
Brazilian farmers and plant breeders developed a maize variety, Sol da Manha, which
yields well under different nitrogen conditions. The farmers and breeders’ goal was to
develop a variety that would tolerate low nitrogen soils. However, by making selections
from populations grown in fields with varying nitrogen levels, the resulting variety is an
efficient nitrogen user and yields well in both high and low nitrogen conditions. In
addition, Sol da Manha can use both ammonium and nitrate forms of nitrogen, an ability
that some other modern maize varieties lack (Machado and F emandes, 2001). S01 da
Manha is an example of a very broadly adapted variety. Although the objective of the

farmers and breeders was to develop a variety that was tolerant of low nitrogen,

68

selections were made in a variety of nitrogen conditions to ensure that the resulting
variety would not be adapted only to low nitrogen soils.

PCI has been promoted as a means to develop or identify locally adapted varieties
for farmers in marginal environments (Atlin et al., 2001), but one possible drawback of
this approach is that the products of PCI may be so narrowly adapted that they will be
useful in only a few locations. This would make PCI rather inefficient at meeting the
needs of the world’s many poor farmers. However, as the three examples illustrate, the
varieties developed through PCI are not exclusively narrowly adapted, nor are farmers
specifically looking for varieties with narrow adaptation. The same Nepalese rice
breeding program produced varieties with both broad and narrow adaptation, and farmers
found both types of varieties useful. Either broad or narrow adaptation may be desirable,
depending on the situation.

Implementation of PCI

PCI has been used effectively to develop or identify new varieties that benefit
small-scale farmers. Additionally, the new varieties have, in many cases, been available
to farmers much faster than they would have if developed though formal plant breeding,
since formal release of varieties takes years in some countries. However, the use of PCI
does not guarantee success. Successful PCI projects have some common traits that may
be considered guidelines when deciding if PCI is an appropriate solution to farmers
problems: (a) farmers’ needs are not met by their current crop varieties, (b) an improved
crop variety is an appropriate solution to the problem, (c) the crop being improved is
important to the farmers, (d) farmers have a tradition or prior knowledge of seed

selection, (e) farmers maintain their own seed over the long term.

69

One example of a PCI project which has each of these five traits is the effort by
the Institut des Sciences Agronomiques du Rwanda (ISAR) and the International Center
for Tropical Agriculture (CIAT) to improve Rwandan bean varieties. Rwandan farmers
were in need of higher yielding varieties as Rwanda’s population was growing, the
country’s soil fertility was declining, and landholding size was decreasing (Sperling,
1992). Yield and stress tolerance, two problems that can be addressed through breeding,
were the main concerns of farmers (Sperling, 1992). Beans are an important subsistence
crop in Rwanda where they are grown by 97% of farmers (Sperling, 1992) and provide
32% of the caloric intake (Sperling and Scheidegger, 1996). Women farmers are skilled
in maintaining and using bean variety mixtures, which are altered depending on the agro-
ecological conditions (Sperling and Scheidegger, 1996). The bean breeders asked
Rwandan women who were considered “bean experts” to make selections from bean
populations grown on the research station, and then test these selections in their own
fields (Sperling, 1992). During the first two years of this project farmers selected and
adopted 21 new varieties, which was equal to the total number of varieties released by the
national bean breeding program in the previous 25 years (Sperling and Scheidegger,
1996)

Timing of Farmer Involvement

As a part of PCI projects, farmers have been involved in various stages of the
plant breeding process (modified from Sperling et al., 2001):

1. Planning and goal setting

2. Generating variation through crossing

3. Early generation selection (from large segregating populations)

4. Late generation selection (from smaller segregating populations)

5. Variety testing
6. Seed multiplication and distribution

70

The majority of PCI experiences previously reported have involved farmers in the
last two stages of the plant breeding process, which is sometimes referred to as
Participatory Varietal Selection or PVS. Some have suggested that since PVS is simpler
and quicker than involving farmers in earlier stages of plant breeding, this approach
should by tried first. Then if no useful varieties are identified through this method it may
be necessary to involve farmers in the earlier steps of plant breeding (Almekinders and
Elings, 2001; Witcombe and Virk, 2001).

In projects where farmer involvement began earlier, they are most often involved
in planning, late generation selection, variety testing and seed multiplication. This
approach allows farmers to have input into the direction of the breeding program from the
very beginning. Also, it allows the breeders to make some selections in the early
generations before the farmers make selections later. This way, breeders can eliminate
obviously unacceptable plants and those which may have undesirable traits that will not
be as apparent to farmers (such as disease susceptibility) (Sperling and Scheidegger,
1996). In only a few cases have farmers made selections in early generations as a part of
PCI (Komegay et al., 1996; Ceccarelli et al., 2001; Humphries, 2001). This may be
because it is difficult to produce enough seed of early generations to allow for selection
in very many sites. Farmers have rarely been involved in crossing to generate genetic
variation. This step has mostly remained the responsibility of plant breeders.

Farmer Recruitment

For the Rwandan PCI project the plant breeders chose and recruited expert

farmers to participate in the crop improvement process. This approach has been used in a

number of PCI projects (e. g. Komegay et al., 1996; Sthapit et al., 1996; Smith et al.,

71

2001), and has some advantages. Researchers can choose farmers who are
knowledgeable about the crop, communicate well, and are socially connected (to
facilitate the spread of information and varieties). However, choosing only expert
farmers to participate can result in a bias toward wealthier or more educated farmers.
Meanwhile, the poorer farmers’ crop improvement needs may remain unmet.
Additionally, this approach does not give the farmers much power in the decision making
process.

One way in which farmers have been given more power in the PCI process is the
establishment of local farmer research committees (known by their Spanish acronym,
CIAL). CIAT developed this approach for farmer participatory research in general and it
has been used in several Latin American countries. The CIAL consists of fanner-
researchers, who are elected by community members. The CIAL oversees a community
meeting where farmers decide which of their agricultural problems the CIAL should
address. With the help of the supporting researchers the CIAL then develops, and carries
out an experiment to address top priority problems identified by the community. At the
end of the experiment the CIAL reports back to the community members (Ashby et al.,
1996).

In terms of PCI, CIALs have mostly been involved in variety testing (PVS).
Because the CIAL experiments last one season before the community reevaluates the
research direction, it would be difficult for CIALs to address breeding problems, which
require more long-term research. At least one group, however, has applied the CIAL

method to more long-term breeding work. After five years of PVS trials several CIALs

72

in Honduras are involved in the earlier stages of plant breeding in collaboration with the
Panamerican Agricultural School, Zamorano (Humphries, 2001; Rosas, 2001).

One potential problem with CIALs is that they may still be biased toward the
more powerful or wealthy community members, since communities may tend to elect the
more wealthy and powerful individuals to the CIAL while poorer farmers problems are
still neglected (Ashby et al., 1996). When the Rwandan bean researchers later turned the
responsibility of choosing representative farmer experts over to the communities, they
found that the representatives were often chosen based on their political and social
connections rather than on their bean expertise. Consequently, they were not as skilled at
selecting new bean varieties (Sperling and Scheidegger, 1996). Breeders involved in the
Honduran project have attempted to make the CIALs more representative of the
communities by not limiting CIAL membership to “research oriented farmers.”
According to socioeconomic surveys conducted in the communities, the goal of including
farmers from all social strata has largely been met (Humphries, 2001).

Bean Variety Improvement for Northern Ecuador

The Mira and Chota river valleys in the northern Ecuadorian provinces of
Imbabura and Carchi are among the most important bean growing regions in the country,
and constitute an estimated 40% of Ecuador’s total bush bean production area. In 2002,
INIAP founded a CIAL in La Concepcién, a community in this region, with the goal of
improving bean production, in part through PCI. Bean variety improvement decisions for
the region should also be guided by survey data recently collected from bean growers in

the Mira and Chota valleys.

73

Three surveys of bean producers have recently been conducted in the area: a study
of bean production, post-harvest and marketing practices in the Chota Valley, concluded
in 2000; a participatory rural assessment of bean growers in the Mira Valley, concluded
in 2002; and a study of bean production practices and problems, marketing and variety
preferences in the Mira and Chota valleys in 2003. Findings from the 2000 and 2002
surveys have been published as project reports by PRONALEG and are summarized here.
The findings from the 2003 survey are presented in this thesis.

Geography of the Mira and Chota River Valleys

The Mira and Chota Valleys are located in the highland region of northern
Ecuador. The median annual temperature is 19.5 C and varies little throughout the year.
Rainfall varies substantially depending on location, however the climate is generally dry
and agriculture in some areas is reliant upon irrigation. Most of the rainfall occurs
between October and May. A large percentage of the population is of African descent,
although in the country as a whole, blacks are an ethnic minority and account for less
than 10% of the population.

Study of Production, Post Harvest and Marketing Practices of Bean Producers in
the Chota Valley, 2000'

For this study, researchers selected, from bean growing communities in the Chota
Valley, 105 farmers who owned more than 0.5 ha of land. The farmers were divided into
three categories depending on their use of five recommended bean production practices:
use of improved varieties, use of quality seed, application of recommended levels of

fertilizer, proper timing of fertilizer application, and proper use of pesticides.

 

l A summary of IN IAP Project Report (2001) Estudio de la Produccién, Poscosecha, Mercadeo y Consumo
de Fréjol Arbustivo en el Valle del Chota, Ecuador. Quito, Ecuador

74

Those farmers who were classified as Type 1 used, on average, two of the five
recommended practices. Sixty percent of the Type 1 farmers owned their land. The
average land holding size was 3.5 ha and an average of 2.2 ha were planted to bean.
Farmers in this category cultivated both improved varieties and landrace varieties in the
Calima Negro, Calima Rojo, Medium Red Mottle, and Panamito commercial bean
classes. None of the farmers used quality seed. They tended to use less than the
recommended fertilizer levels and applied all of the fertilizer at planting. (The
recommended practice is to fertilize 15 to 30 days after planting.) All of the farmers in
this group used pesticides and made an average of three applications per growing cycle.
However, they tended to apply less than the recommended dose of the products they
used. The Type 1 farmers all marketed their beans through intermediaries who then sold
to local vendors or exported to Colombia.

The farmers who were classified as Type 2 used an average of three of the five
recommended practices. Eighty—five percent of the Type 2 farmers owned their land.
The average land holding size was 4.9 ha and an average of 2.2 ha were planted to bean.
Like the Typel farmers, those in the Type 2 category cultivated both improved varieties
and landraces varieties in the Calima Negro, Calima Rojo, Medium Red Mottle, and
Panamito commercial classes. Thirty percent of the farmers used quality seed. Some
farmers in this group used adequate fertilizer and others used less than the recommended
fertilizer levels. All of the farmers applied half of the fertilizer at planting and half about
20 days after planting. All of the farmers in this group used pesticides and made an
average of two applications per growing cycle. Pesticide use among the Type 2 farmers

was considered acceptable. Twenty percent of the farmers sold their beans in the

75

Colombian market directly and the remaining 80% marketed their beans through
intermediaries who then sold to local vendors or exported to Colombia.

The farmers in the Type 3 group used an average of four of the five recommended
practices. Sixty-seven percent of the Type 3 farmers owned their land. The average land
holding size was 30.3 ha and an average of 4.8 ha were planted to bean. Farmers in this
group cultivate almost exclusively improved varieties in the commercial classes for
Colombian export: Calima Negro, Calima Rojo, and Medium Red Mottle. All of the
farmers used quality seed when it was available. Some farmers in this group used
adequate fertilizer and others used less than the recommended fertilizer levels. All of the
farmers applied fertilizer at the recommended 15 to 30 days after planting. The farmers
in this group all used pesticides and made an average of 3 applications per growing cycle.
Pesticide practices among the Type 3 farmers were considered appropriate. All of the
farmers in this group sold their beans in the Colombian market. Across the farmer groups
about one third of the producers grew only bean. Others grew additional crops, including
maize, sweet pepper, hot pepper, cassava, onion, tomato, pea and sugar cane.

In all of the farmer groups, the majority considered pests and disease to be their
biggest bean production problem. The most problematic diseases of bean that farmers
reported were anthracnose (Colletotrichum lindemuthianum), rust (Uromyces
appendiculatus), and mildew (Erysiphe polygoni). Farmers reported many pest
problems, including whitefly (T rialeurodes vaporariorum), thrips, spider mite
(T etranychus sp. and Polyphagotasonemus latus), leaf miner (Liriomyza sp.), crickets,

and cut worms. For disease control the most common active ingredients used were

76

mancozeb, cyproconazole, sulfur, and maneb. The most common active ingredients for
pest control were methamidophos, profenofos, and cyperrnethrin.

Other production problems that farmers reported included lack of water,
commercialization, and poor quality seed. The researchers also regarded the farmers’
post harvest practices as problematic. They noted that the farmers in this region thresh
their beans by piling them in the road so that they will be run over by large trucks. This
method damages the seed and can result in a 20% loss depending on seed moisture.
Participatory Rural Appraisal of Seven Communities in the Parish of La
Concepcién, 20022

La Concepcién is located in the Mira River Valley. In conjunction with the
founding of a CIAL in the community of La Concepcién, INIAP researchers interviewed
a total of 80 households from La Concepcién and the nearby communities of El
Empedradillo, La Convalencia, El Rosal, El Chamanal, Santa Lucia, and Santa Ana.
Seventy-five percent of those interviewed were of African descent and the remaining
quarter were Mestizo.

Agriculture is the primary activity of the men in the communities while women
work mostly in the home. A few women (15%) devoted the majority of their time to
agriculture. Most families have small land holdings --the majority (56%) owned 1 ha of
land or less, another 24% owned between 1 and 2 ha, and 20% owned more than 2 ha.
Beans are the most important crop in this area and are grown for home consumption and
for sale. Of the farmers interviewed, 75% were growing beans at the time and 50% had

grown beans during the previous cropping cycle. The second most important crop is

 

2 A summary of INIAP Project Report (2003) Ge'nero e Investigacién Participativa en el Mejorameinto de
Variedades de Fréjol y Sistemas de ProducciOn de Semillas en la Sierra de Ecuador. Quito Ecuador

77

maize which was being grown by 5% of the farmers at the time of the interview but had

been grown by 39% in the previous crop cycle. Other crops include tomato (9% present
cycle, 5% previous), cassava (5%, 4%), sugar cane (3%, 0%), sweet pepper, banana, and
hot pepper.

The beans grown the parish of La Concepcién belong to the Calima Negro,
Calima Rojo, Cranberry, Small Black, Panamito, Medium Red Mottle, and Canario
market classes. Some farmers also reported growing bean variety mixtures. Beans are
sown twice during the year, typically from February to March and from August to
September. However, for farmers without irrigation, planting depends on the timing of
rainfall. The majority (80%) of the farmers save their own seed, the rest either purchase
seed (18%) or obtain it through exchange (2%). Due to poor seed quality, farmers plant
from 2 to 5 seeds per hill in order to get an acceptable plant stand. About half of the
farmers fertilized their bean crop.

The bean pest and disease problems are rust, whitefly and bean weevil
(Acanthoscelides obtectus). Farmers used products containing the following active
ingredients to control bean pests and disease: carbofuran, chlorothalonil, cypermethrin,
lambda cyhalothrin, malathion, mancozeb, maneb, metalaxyl, methamidophos,
profenofos, propineb, and sulfur. Pesticide use in La Concepcién is minimal, with
farmers making at most two pesticide applications and some using no pesticides at all.
Seventy-four percent of the farmers use some method for post-harvest control of the bean
weevil. Chemical controls included aluminum phosphide fumigation, or malathion

applied to the seed. Other farmers used garlic, eucalyptus, or ash to control weevils.

78

Some farmers reported treating beans for consumption with malathion and then washing
them before cooking.

Bean producers in La Concepcién thresh beans by beating the harvested plants
with sticks. This method does not damage the seed as threshing in the road does. One
third of the bean growers in the region grow only for home consumption. The majority
(60%) sell to intermediaries who come to the community and a small number (7%) take
their beans to Ibarra or Mira to sell to vendors directly.

A Brief Comparison of the Two Valleys

Although farmers from both valleys sell beans, production in the Chota Valley is
more commercialized. Farmers use more inputs, such as fertilizer and pesticides, and
their proximity to the Pan-American Highway makes markets more accessible. Some of
the farmers in the Mira Valley grow beans only for home consumption and they grow a
larger variety of bean classes. The predominant classes in the Chota Valley were those
grown for export to Colombia. Soon after the study in the Chota Valley was completed,
however, Plan Colombia effectively ended all export of beans from this region of
Ecuador to Colombia. Some results of this change in the market are reflected in the most

recent survey findings, which are presented in this thesis.

79

Materials and Methods

Farmer Survey

A questionnaire was developed to gather information that would be useful in
setting breeding goals. The questionnaire that was used appears in Appendix B, Figure
Bl. An English translation of the questionnaire is in Appendix B, Figure B2. For a part
of the interview farmers were presented with photographs of the following bean pests and
diseases mounted on 5 x 8” cards: whitefly (T rialeurodes vaporariorum), leaf miner
(Liriomyza sp.), spider mite (T etranychus sp.), bean weevil (Acanthoscelides obtectus),
anthracnose (C olletotrichum lindemuthianum), rust (Uromyces appendiculatus), common
bacterial blight (Xanthomonas axonopodis pv. phaseoli), angular leaf spot
(Phaseoisariopsis griseola), powdery mildew (Erysiphe polygoni), and bean common
mosaic virus (BCMV). Farmers were asked to identify those bean pests and diseases that
were problematic in their crOp, and farmers were questioned further about the particular
pests or diseases that they identified as problematic.

Fifty-six bean growers from 10 different communities in the Mira and Chota
Valleys of Ecuador were interviewed in May of 2003. Recruitment for the interviews
was accomplished by approaching people in the communities, especially those who were
working in bean fields or threshing beans. Individuals were only interviewed if they
were currently growing beans.

Data fi'om the interviews was compiled in SPSS.

80

Results

Age, Education and Family Size of the Bean Producers Interviewed

A total of 56 bean producers were interviewed, 10 women and 46 men. They
ranged in age from 23 to 83 years old. As shown in Table 23, the largest group (41% of
the farmers) was 36 to 50 years old. Approximately equal numbers of farmers were

between 20 to 35 years old and 51 to 65 years old.

Table 23. Age distribution of farmers interviewed

 

 

Age Category # of Farmers Percent of Farmers Interviewed
20 to 35 years old 12 21.4
36 to 50 years old 23 41.1
51 to 65 years old 13 23.2
>65 years old 8 14.3

 

The level of schooling completed by the farmers interviewed is shown in Table
24. Most of the farmers interviewed (63%) had attended only primary school, and 23%
had attended secondary school. Only 5% had had no schooling and another 5% had

attended university.

Table 24. Level of schooling completed by farmers interviewed

 

 

Level of Instruction # of Farmers Percent of Farmers Interviewed
No Schooling 3 5.4
Primary School 35 62.5
Secondary School 13 23.2
University 3 5.4
Not Responding 2 3.6

 

Table 25 shows the distribution of family sizes of the farmers interviewed. Sixty-
two percent of the farmers interviewed had five or more family members living at home.

Only 9% were members of 1 or 2 person households.

81

Table 25. Family size of farmers interviewed

 

 

Number of Family Members # of Farmers Percent of Farmers Interviewed
l to 2 5 8.9
3 to 4 15 26.8
5 to 6 18 32.1
7 to 8 14 25.0
>8 3 5.4
Not Responding l 1.8

 

Land Ownership and Crops Cultivated by Farmers

Almost all of the farmers interviewed (93%) owned the land they cultivated. The
remaining 7% of the farmers rented land. The size of farmers’ land holdings ranged from
0.5 to 30 ha. The median land holding size was 2 ha. However, 39% of the farmers
interviewed owned 1 ha or less and 64% owned less than 2 ha. Only 14 % of the farmers
owned more than 5 ha. Those farmers who rented land rented between 1 and 2 ha.

Although the interviews were conducted in a mountainous region, excessive slope
was not a problem for most farmers. The majority of the growers (59%) farmed only flat
fields (usually located in the river valley), and 68% had at least one flat field. A few, 7%,
owned only very sloped fields.

Within the two years preceding the interview most of the farmers had grown two
or three different crops, including bean. Twenty-nine percent had grown four or more
different crops. As shown in Table 26, farmers grew a variety of crops in addition to
bean. Beside bean, tomato was the crop farmers most frequently reported growing.

Peppers, maize, cassava, sugar cane and onion were also frequently grown.

82

Table 26. Crops grown by farmers interviewed

 

 

Crop Percent of Farmers Growing Crop
Bean 100%

Tomato 39%

Sweet Pepper, Maize 20 to 30%

Cassava, Sugar Cane, Hot Pepper, Onion 10 to 20%

Anise, Camote, Pea 5 to 10 %

Alfalfa, Avocado, Carrot, Papaya, Pigeon

0
Pea, Prickly Pear, Potato, Blackberry >5 /0

 

Table 27 shows the distribution of land area that farmers planted to bean. Most
frequently farmers planted between 0.5 and 1 ha of bean. About 20% planted less than
0.5 ha and another 20% planted between 1 and 2 ha. A few of the farmers, with larger
land holdings, planted more than 3 {ha of beans. Ten hectares was the largest area that a

farmer reported planting to beans.

Table 27. Area farmers interviewed planted to beans

 

 

Hectares Planted to Bean # of Farmers Percent of Farmers Interviewed
50.5 11 19.6
0.51 to 1 26 46.4
1.1 to 2 11 19.6
2.1 to 3 4 7.1
>3 4 7.1

 

Data for bean consumption by the farmers interviewed suggest that the
consumption rate in this region is much higher than the national average of 6 kg/year.
Farmers reported eating beans an average of 5 times per week, and the consumption rate

was calculated to be 38.106.68 kg/year.

83

 

The Bean Classes Grown by Farmers
Overall, the farmers interviewed reported growing thirteen different classes of
beans. However, even with this diversity, most farmers grew only one or two classes of

beans (Table 28). The maximum number of classes that any farmer grew was four.

Table 28. Number of classes of beans farmers reported growing

 

 

Number of Classes Grown # of Farmers Percent of Farmers Interviewed
1 31 55.411
2 16 28.6
3 6 10.7
4 3 5.4

 

Figure 15 shows the number of farmers growing each commercial class and the

 

total quantity sown in kilograms reported by farmers and in millions of seeds. The
calculation of millions of seeds sown was made to adjust for differences in seed weight
among the classes. Medium seeded red mottles were the most frequently grown class.
Other classes that were frequently grown were small black, Blanco de Leche, Calima
Negro and Calima Rojo. A few farmers grew Panamito and yellow seeded Canario and
Matahambre classes. The most rarely reported classes were Bayo Blanco, Vaquita, solid
red, Musgo and snap beans. A description of each class, its distribution, uses, prices

(Figure 16), and farmers’ comments follow.

Medium Red Mottle

Farmers call medium seeded red mottle (MRM) beans Injerto, Rojo Injerto, or by
the names of several released varieties in this class -- Cargabello, Paragachi, and Selva.
Seed of this class is medium in size and has a mottled pattern of dark and light red and
white. Farmers reported growing MRM varieties with both determinant (type I) and type

II indeterminate growth habits.

84

uMOS spans jo suogmw

000.0 000m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

31 9.39.10 #1801. 113.01, 0N. 31m o1~|8| _11. Na N173 3 ND 13. NNN NW». N 1231 m1. mafia 8000. No aeezl
N11 l. 1.1 1 11-1 1. -1 11 l . , . . N111 11 ltl . 11.11 .
u 0 U3 : 3 _1_NN|81.3 mm 1: _ 8N1 N N..0N 0180 fl 0% . $10 _ NNN N1..- 88. 0000200058.!
341' is i i 0i i oll ill 14 '11.» 11 ..14| i. ,.i|..| i 111 hi .lii..1 1|..l 11.1 14 ..... l .1. Ii
,3 _ n _ 3| N 3 N 33 N 1m _ N u 0 3 21 1f 1: . N_.1_N11 2 N N _ 0538000800001
.111 l... ._.|. 11 Ii 1 i 1.1 .1....11 11-1.11, 11.11111 1 1N1. .1111, . 111.1% ....| .| . 11 -L 1| 11 l 11 l. .11 i .
3 , M 3 4
3 0 . 0w . N 0000 00005 000 0C0: 008 0.3% 0, 00m 02 00000 0005 0:002
_ . 0.8m _ a 1.000 . _ 00 _ 00m
00$. -82 -N> . o «0 -002 00 -050 6:00 _ 80.00 #8020“ :20. .002 _
8.0 1. I]: 3.... . , o
N3
W .. NW N N .,
00 N N .N a.“ 3M. 1 com
ooN - N... 9..
N 99.: m
000 1 .N am
,. . 1 002 m
. 1 3|. S
00 v S
O
. SON m
006 1
cod - comm
00.x. ooom

000—0 0000 NEBONN 00000.00 .00 00800 000 0030300000 0008000 000 N0 0300 000—0 0000 0000 .00 N000N0< .2 0003.0

85

Figure 16. Maximum, minimum and mean prices for bean classes

 

 

 

 

 

20
15
10
5
O

(nun/s) aaud

86

Bean Class (unit)

 

.. . If...

mud—U :00m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.2? 00.9
4‘ a .9 0v 0
0.4 2N. no 0 00 / %
.9on 9.104 ore nwmo 00¢? 00%. 02%.
T11 11 1 1 .1 1 111 1 .d
9
11 111 1 11 11 1N T1 1-1+ i—1 M
m
w -1 11 - 11 1r 1111 0..
00003157 1- 1 1 | 1 | 11 1 111 m
L L J
0000»??? m
0000>oT©IH 1 11 1 1 1 1 1 1 11 1 | 1 u
,mN00onAIh, 1 1 1 1 1 1 1 11 1 1 mg
- - 1-- M
111 T 11 11 11111 11111111 1 11111 105 m.
L 3
1 1 1 1 1 1 1 11 1 1 1 1 1 1 1 1 1 .3 m
11 1111 11 111111 1 11111 1 1 1 1100
oo—

 

 

 

 

 

 

 

 

 

0030 E09 :05on N00: E0~E£ $00» No 005832 .: 003wE

87

MRMs rank first among the classes in number of farmers cultivating the class,
millions of seeds sown and in amount of seed sown. Twenty-five farmers, 45% of those
interviewed, reported growing this class within the past year. As shown in Figure 17
12% of the farmers growing the class had grown it for more than 10 years and 40% had
grown it for 6 to 10 years. Twenty-four percent had grown the variety for 3 to 5 years
and another 24% were very recent adopters, who had grown the variety for 1 to 2 years.

The majority (81%) of the farmers who had grown MRM in the past had sold
some of their harvest. A few farmers (12%) grew the class only for sale and did not
consume it themselves. Farmers reported sowing a total of 2,693 kg of seed of this class
within the last year. This class was more frequently grown by farmers from higher
elevation communities between 1800 and 2460 m.

MRMS are sold and consumed both dry and in tiemo. Of the farmers who had
sold this class in the past, 76% had sold the beans dry and 59% had sold them as shell
beans, with some farmers selling both dry and in tierno. Of the farmers who had
consumed MRMs in the past year, the majority, 72%, used them both dry and in tierno.
Twenty-two percent consumed the class only as dry beans and 5% only in tiemo.

The average price farmers reported receiving for dry MRMs was $30/quintal
(about 45 kg or 100 lbs), with a minimum of $20/quintal and a maximum of $40/quintal.
The mean price for MRM in tierno was $14/bulto, with a minimum of $lO/bulto and a
maximum of $18/bulto. (A bulto, the typical unit for unshelled beans in tierno, is a
measure of volume. One bulto of beans in tierno weighs between 30 and 35 kgs). Of the
farmers selling MRMs, 47% sold their crop in their own community, 35% sold MRM in

the town of Pimanpiro and 18% sold MRM in the city of Ibarra.

88

 

Most farmers growing this variety do not save their own seed. The majority,
84%, of farmers growing MRMs had obtained their seed from an outside source -- either
purchased (72%), exchanged (4%) or obtained from their landlord (8%). Only 16% used
seed they had saved themselves.

Farmers had much praise for MRMs and few complaints. They regard the class
as resistant to diseases, especially rust, and high yielding. They also complimented its
flavor, and high quality seed for tierno, which brings a good price. Most farmers did not
mention any negative qualities for this class, although three farmers thought MRMs took

too long to mature.

Small Black

Farmers call small seeded black beans Negro Pequefio or simply Fréjol Negro.
One farmer used the Colombian name for the class, Caraota. The black bean varieties
farmers were growing had a type II indeterminate grth habit. This class was more
frequently grown by farmers from lower elevation communities between 1340 and 1700
m.a.s.l.

Small blacks are second among the classes for number of farmers cultivating the
class and for millions of seeds sown. Sixteen farmers, 29% of those interviewed,
reported growing this class within the past year. Nineteen percent of the farmers growing
the class had grown it for more than 10 years and 13% had grown it for 6 to 10 years.
Thirteen percent had grown the variety for 3 to 5 years and another 56% were very recent

adopters, who had grown the variety for 1 to 2 years.

89

Of the farmers who had grown small blacks in the past, 82% had sold some of
their harvest. None of the farmers interviewed grew this class only for sale -- all those
who grew small blacks also consumed them.

The farmers interviewed sold small blacks exclusively as dry beans. The majority
also consumed small blacks only as dry beans, although 25% consumed small blacks dry
and in tierno.

The average price farmers reported receiving for small blacks was $28/quintal(45
kg), with a minimum of $20/quintal and a maximum of $36/quintal. Of the farmers
selling small blacks, one thirds sold their crop in their own community, one third sold
them in the city of Ibarra, and one third sold them to a cooperative.

The majority, 88%, of the farmers growing small blacks had obtained their seed
from an outside source. However, this figure may be a bit misleading since a number of
farmers were growing the class for the first time. Fifty-six percent of the farmers
purchased seed, and 31% got seed from a cooperative or NGO. (Several groups were
promoting black beans, which have a domestic market, as an alternative to the Calima
classes.) Only 12% of black bean growers planted seed they had grown and saved
themselves.

Of all the classes, small blacks garnered the most positive comments on their
flavor. Farmers also commented that this class is high yielding, disease resistant and has
a local market. Only two farmers had negative comments for the class. One thought it
took too long to mature and the other found it difficult to thresh since the small seeds fall

into cracks in the road and cannot be picked up.

90

Blanca de Leche

The class known to farmers as Blanco de Leche produces large, white, kidney
shaped seeds. The plants are determinant.

Blanco de Leche is third among the classes in both number of farmers cultivating
the class but only sixth in total kilograms of seed sown of the class, indicating that the
farmers who grow Blanco de Leche plant less of it relative to other varieties. Twelve
farmers, 21% of those interviewed, reported growing this class within the past year. Fifiy
percent of the farmers growing the class had grown it for more than 10 years and 17%
had grown it for 6 to 10 years. Seventeen percent had grown the variety for 3 to 5 years
and another 17% were very recent adopters, who had grown the variety for l to 2 years.
None were growing it for the first time.

The majority (92%) of the farmers who grow Blanco de Leche sold some of their
harvest. Twenty-five percent of the farmers grew the class only for sale and did not
consume it themselves. Farmers reported sowing a total of 653 kg of seed of this class
within the last year. This class was grown at lower elevations almost exclusively by
farmers from two neighboring communities: Tumbatu and Pusir Grande.

Blanco de Leche is nearly always sold and consumed in tierno. All of the farmers
who had sold this class, sold it in tierno. One farmer reported also selling Blanco de
Leche dry, for seed. Of the farmers who consumed this class, all ate it in the tierno stage,
and one farmer also consumed it as a dry bean.

The mean price for Blanco de Leche in tierno was $16/bulto, with a minimum of

$10/bulto and a maximum of $20/bulto. Of the farmers selling Blanco de Leche, 18%

91

sold their crop in their own community, 18% sold it in the town of Pimanpiro and 64%
sold it in the city of Ibarra.

The majority, 75%, of the farmers growing Blanco de Leche saves their own seed.
The 25% of farmers who got their seed from an outside source purchased it or obtained it
through exchange.

Blanco de Leche is grown for sale in tierno during Holy Week prior to Easter, and
is a part of the traditional dishes for this celebration. Farmers reported that they like
Blanco de Leche because it has a short growing cycle, it brings a very good price during
Holy Week, and it yields well. However, the said it is also very susceptible to disease
and that prices drop substantially afier Holy Week. Although grown by a relatively large
number of farmers, Blanco de Leche can be a risky investment, and this probably

explains why farmers plant less of it, relative to the other classes.

Calima Negro

Calima Negro, sometimes called Mil Uno, the name of a released variety in this
class, produces large seed with a mottled pattern of dark and light purple and white. The
Calima Negro varieties grown by the interviewed farmers were all determinant.

Calima Negro ranks fourth among the classes in number of farmers cultivating the
class. Eleven farmers, 20% of those interviewed, reported growing this class within the
past year. Nine percent of the farmers growing the class had grown it for more than 10
years and 36% had grown it for 6 to 10 years. Nine percent had grown the variety for 3
to 5 years and 45% were very recent adopters, who had grown the variety for l to 2 years.

All of the farmers who had grown Calima Negro in the past had sold some of their

harvest. Twenty-two percent had grown the class only for sale and did not consume it

92

"'3

 

themselves. Farmers reported sowing a total of 890 kg of seed of this class within the last
year.

Calima Negro is mostly sold dry, but one farmer reported selling it in tierno. Of
the farmers who had consumed Calima Negro in the past year about half (5 7%)
consumed the class only as dry beans and the rest (43%) consumed them dry and in
tiemo.

The average price farmers reported receiving for dry Calima Negro was
$32/quintal(45 kg), with a minimum of $25/quintal and a maximum of $38/quintal. Of

the farmers selling Calima Negro, 78% sold their crop in their own community, 11% sold

 

them in the town of Pimanpiro and 11% sold them in the city of Ibarra.

Most farmers growing Calima Negro do not plant their own saved seed. The
majority, 64%, had obtained their seed from an outside source -- either purchased (54%)
or through exchange (9%). Of the 36% that planted seed they had saved themselves, all
had purchased Calima Negro seed within the past 3 years.

Farmers commented that they liked the seed and market quality of this class.
They also liked Calima Negro’s fast maturation and good yields. Some of the farmers’
criticism of this class was that it required too much water to grow, its flavor and large

size made it less than desirable, its price was too variable and it sometimes yields poorly.

C alima Rojo
Calima Rojo, also known as Margarita and Catio produces large seed with a
mottled pattern of red, pink and white. Farmers reported growing Calima Rojo varieties

with both determinant (type I) and type II indeterminate grth habits.

93

Calima Rojo ranks fifth among the classes in number of farmers cultivating the
class. Ten farmers, 18% of those interviewed, reported growing this class within the past
year. Thirty percent of the farmers growing the class had grown it for more than 10 years
and 50% had grown it for 6 to 10 years. Ten percent had grown the variety for 3 to 5
years and another 10% were very recent adopters, who had grown the variety for 2 years.
None of the farmers were growing the class for the first time.

All of the farmers who were growing Calima Rojo had sold some of their harvest.
Ten percent grew the class only for sale and did not consume it themselves. Farmers
reported sowing a total of 803 kg of seed of this class within the last year.

Calima Rojo is mostly sold dry, but two farmers reported also selling it in tierno.
Of the farmers who had consumed Calima Rojo in the past year, 66% consumed the class
only as dry beans and the rest (33%) consumed them dry and in tierno.

The average price farmers reported receiving for dry Calima Rojo was
$3 l/quintal(45 kg), with a minimum of $8/quintal and a maximum of $40/quintal. Of the
farmers selling Calima Rojo, 56% sold their crop in their own community, 22% sold
them in the town of Pimanpiro and 22% sold them in the city of Ibarra.

The majority, 60%, of the farmers growing Calima Rojo planted their own saved
seed. Thirty percent purchased seed and 10% obtained it through exchange. Of the 60%
that planted seed they had saved themselves, half had purchased Calima Rojo seed within
the past 3 years.

The farmers’ comments about Calima Rojo were much like those for Calima
Negro. They liked the seed quality, marketability, and fast maturation. Some farmers

also liked its flavor. There were mixed opinions on the yields of this class, as some

94

farmers thought it yielded well and others thought it yielded poorly. They also

considered it susceptible to rust and drought.

Panamito

Six farmers, 11% of those interviewed, were growing a small seeded white bean
they called Panamito. This class can have either a determinant (type I) or type II
indeterminate growth habit. The Panamito class was sixth among the classes in terms of
the number of farmers growing it.

Half of the farmers growing this class were growing it for the first time. The
other half had been growing the class for more than 10 years. All of those who had
grown it before the current cycle had sold some of the harvest, and consumed the class
themselves. None of the farmers reported consuming or selling Panamito in any form
other then as a dry bean.

The average price farmers reported receiving for Panamito was $26/quintal(45
kg), with a minimum of $25/quintal and a maximum of $27/quintal. Of the farmers
selling Panamito, 33% sold their crop in their own community, and 67% sold it in the city
of Ibarra.

Sixty-six percent of the farmers growing Panamito obtained seed from an outside
source. However, as previously mentioned, half of them were growing the class for the
first time. Overall, 33% of the farmers saved their seed, 33% purchased seed and 33%
obtained seed from a cooperative or NGO.

Farmers like Panamito beans because they are high yielding and have a local
market. However, the farmers also commented that they did not like the flavor of this

class, that it is susceptible to disease, particularly rust, and takes a long time to mature.

95

C anario

Four farmers were growing Canarios, which have a round, medium sized, light
yellow seed, and a type I or 11 growth habit. Canarios are seventh in terms of number of
farmers growing the class.

Of the four farmers growing this class, three were growing it for the first time, the
other had grown Canarios for more than 10 years. The only farmer who had grown it
before the present cycle had sold it in Pimanpiro, as a dry bean and consumed it both dry
and in tierno.

All of the farmers had obtained the seed they planted from and outside source --
two from a cooperative or NGO, one from a landlord, and one purchased.

The farmer who had grown Canarios before grew it for its flavor, others were
growing the class because they believed it would yield well and bring a good price.

However, two farmers thought Canarios were susceptible to disease.

Matahambre

Three farmers were growing a type 11, medium, yellow seeded bean which is
slightly longer and darker in color than the Canario. They called this class Amarillo or,
more picturesquely, Matahambre. One of the farmers was growing the class for the first
time, the other two farmers had grown the class for more than 10 years. The two farmers
with more experience with Matahambre sold the class and consumed it themselves. One
had sold Matahambre only as a dry bean and the other had sold it dry and in tierno. Both
had consumed the class dry and in tierno. The average price farmers received for dry

Matahambre beans was $25/quintal, and both had sold the beans in Pimanpiro.

96

All three farmers obtained seed of Matahambre from an outside source -- two
purchased it and one made an exchange with another farmer. According to farmers, the

outstanding quality of this class is its flavor.

Bayo Blanca

Two farmers were growing Bayo Blanco, a cream colored bean with a medium

sized round seed. Both farmers were growing the class for the first time and had obtained

‘30}? ',' . ‘q

the seed from a cooperative. The seed was a mixture, producing plants with both type I

‘
—r w.

and type II growth habits. The farmers were growing the class because they believed it

would bring a good price. However, one thought it susceptible to disease.

 

Vaquita

One farmer was growing a type 11, medium seeded, cream and pink bean called
Vaquita. The farmer had, within the past year, purchased seed of the variety and was
growing it exclusively for sale in tierno. She had sold her first harvest for $15/buleto in
Ibarra. The farmer thought the variety yielded well but complained that it did not mature

at the same time.

Solid Red

One farmer was growing a type 11, medium seeded, solid red bean simply called
Rojo. She had cultivated the variety for more than 10 years. The farmer planted her own
seed of the variety and grew it for home consumption and sale both dry and in tierno.

She had received a price of $10/quintal of dry seed in Ibarra.

97

 

Musgo

One farmer was growing, Musgo, a determinant bean with a thin, medium-sized
brown seed. The farmer planted his own seed of this variety, which he had grown for
more than 10 years, and used it solely dry for home consumption. He thought the variety

yielded well and he liked the flavor and seed size and shape.

Vaifiz'ta

One farmer grew determinant snap beans that had a large white seed. She had
begun growing the variety within the past year from seed she purchased. The farmer
grew the variety for sale as a snap bean in Ibarra, and used it for home consumption dry,
in tierno and as a snap bean for green pod consumption.
Farmers Seed Sources

Overall, only 36% of the farmers reported planting their own saved seed. Nearly
all of the farmers interviewed, 91%, had obtained some seed from an off-farm source in
the past three years. As shown in Table 29, farrners’ decision to save or to purchase seed
was not related to the amount of land they owned (a proxy for poverty). A Xz-test for
independence shows that landholding size and seed saving are independent of one

another.

Table 29. XZ-test for independence between landholding size and seed saving

 

 

 

 

 

 

 

 

 

Landholding Size Did Not Save Seed Saved Some Seed
0-1 .0 ha 13 9
1.1-2.0 ha 9 5
2.1-3.0 ha 6 1
3.1-6.0 ha 5 3
>6 ha 3 2

 

XZ-value=l .71 p-value=0.79; Fail to reject H0, conclude variables are independent

98

 

Pest and Disease Problems Reported by Farmers

When farmers were shown photographs of the bean pests whitefly, weevil, spider
mite and leaf miner, 96% identified whitefly as a problem in their bean crop, 75%
identified weevils as a problem, 50% identified spider mite as a problem, and 46%
identified leaf miner as a problem. When shown photographs of bean diseases, 86%
identified rust as a problem in their bean crop, 59% identified root rot as a problem, 52%
identified anthracnose as a problem, 48% identified mildew as a problem, 48% identified
common bacterial blight as a problem, and 23% identified angular leaf spot as a problem.
Additional data on the severity of each production problem, farmers’ names for the pest
or disease, and farmers’ control methods follows. A complete list of farmers’ names for
the production problems is given in Table 30. A comparison of the incidence and
severity of the production problems is shown in Figure 18. Table 31 lists all of the active
ingredients in the products that farmers reported using, the problems the products were
used to treat, and the World Health Organization’s acute toxicity rating for the each
active ingredient (IPCS-IOMC, 2000). A list of all the active ingredients of pesticides

and their Ecuadorian trade names is found in Appendix C.

99

 

Table 30. Farmers’ names for production problems and number of farmers using each

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

name
Whitefly # Farmers Leaf Miner # Farmers Spider Mite # Farmers
Mosca blanca 25 Minador 9 Arafia roja l6
Palomilla 12 Toston 5 Arafiita roja 4
Mariposita 10 Enrollador 2 Arafiita 3
Mariposa blanca 3 Lancha 2 Acaro 1
Plumilla 1 Rollador 1 No Name 4
Barreno 1 Arafia negra l
Arafia 1 Hongos 1
No Name 3 No Name 8
Weevil # Farmers Rust # Farmers ALSa # Farmers
Gorgojo 27 Roya 42 Laucha 2
Redondilla(o) 9 Eumollador 1 Lancha 2
Polilla 5 Royal 1 Antracnosis 1
Pulgon 1 No Name 5 Mancha Negra 1
No Name 1 No Name 7
Anthracnose # Farmers Root Rot # Farmers CBBb # Farmers
La Gota 8 Pudricion de raiz 8 Lada (Helada) 4
Antracnosis 6 Pudricion 2 Lancha 4
Lancha 2 Hongos de tallo 2 Bacteria 2
Laucha 2 F usario 1 Lancha amarilla 1
Manchas 2 Hongos 1 Baterosis 1
Pudricion mata 1 Cancer de raiz 1 Laucha 1
Sancochado l Tizon l Antracnosis l
Mancha negra 1 Trozador gusano l Minador 1
Septolia l Enrollador 1 Enrollador 1
Helada l Barrenador 1 Picudo 1
Enrolleador 1 No Name 3 Toston 1
No Name 6 Mosco 1
No Name 7
Mildew # Farmers BCMVc # Farmers
Ceniza 20 Deficiencias 1
Pudricion l Toston 1
Cenicilla 1 Desconocido 1
Acaros l Enrollador l
Mosca I No Name 7
No Name 4

 

a Angular Leaf Spot; b Common Bacterial Blight; c Bean Common Mosaic Virus

100

.i—‘fl'v‘qni h It?"
,.

‘—
.-|_-—

M

 

 

 

wEmemQ 30> I
wEmemQ
3280on I

wEmemD Ho Z D
3:2
8: :5 venomom D

 

.1 ‘3-

8339 .8 «8m
00

 

 

 

 

 

saauueg JO .laqumN

 

 

o0

 

$336 was mwmoa Swen— mo bto>om can 85203 Cotonou $8.8m .f oSwE

101

 

visculr I

:NsoEmNNNN

383358: .23.. .mmzoEmNm: bfiwgm: i. ...m:owNNNmNN on 8 38:25: * ”3%:on 838 Now meNmN NNONNNNNEwwNO 5:8: 25>? 4N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N N o o o o o m o o o .2. 3 885
N o o o N. o N o o o N .. e :58
N o o N o N N. o o o o .. e 8288
N o o o N N o o o o o 3.... N 288888
N o o o o NN o o N o o .. 2 58888
N o o o o o o o o N o ...... N N288
N o o o N o o o o o o ...... N N88323:.
N o o o o o o o o N o .. N 8885
N o o o o o N o o o o r. N 33.88
N o o o c o o o o o N .. N 888
N o N N N N e o c N N .. NN 8888.
N N o o o o o o o c o _.. N 8&8
N o o o o N N o o o o .2. N 28888ee
N o o o N N: N o N N o ...... 2 28888?
N o N N o o N o o o o t. N. 28858
N o o c o N o o o o o ...... N 88288 888
N. o N N o c N o o c N .. N Esmeoszo
N N o o o o o m o o o .. E 5898
N. N. o N o o N o o N o .. o 88888
N N: o o o o o e o o o .. 2 888
N o o o N o o o o o o .. N Santana
m o N N N N N o o o o .. N 1888
:5 v—QHNEN Nam 80:. 22 NA: 552 358% .N.< mam: 3203:?”—
NEuEEN— u 80% mmU mq< 360:2 NNNNM owNNNNNN< N383 NoENNm -233 N34 0E3 225$ a unum—NQNME— o>mau<

 

 

 

 

 

 

 

 

 

@885 25303 :a .mEoEoNNN :oNNoNNNNoNNN :3: 33:8 2 Neema 308.2% 02263 NNN 33$ mNNNoNNoonE e>NNo< .Nm 2an

102

0:3 2: .3 ENE No: ..Nd ...NNNoNNNNNNNNNN b08055: .........._.... ...msoNNNmNNNNN NEE? ...._....... ...mNNoNNNNNNNNNN
322358.. ....I. ...NNNoNNNmNNE 3an5: ...... ...mzovaNNNNN on 8 385:2: ... r3698 83% N8 meNmN :oNNmNNcmwNO 5:82 283 N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N. N. e a N N N: m N: 9N N .539:— NPNNNNS 3 com: N.N.< n =20>O

N o o o o c o o N e N ...... N 2288 8863 628.8%
m o N o o N o N N o N ... .. ... o NONoNNoNoNNN
N o N o o o o o o N o ... ... .. m ENNNNoENoNN
N o o o o N o o o v o 3...... m 38052:
n o o o o o o o N N N 1...... m 83028265
N o o o o o 0 MN o o N ... .. 3 :oNNNNNNNE
N o o o o o o o o m N .NNN m 85:82
v o o o N o o o N N N .. .. .. N 53835.8 «Nonnum—
v o o o o N o o m v N ... ... .. oN cargoes
N o o o o o o N o o o ... .. .. N cocNNNNNN
N o N c c o o o o N N ...... N 5.52888
N o o o o o o o o o N ...: N 5.5.63
N N o o o o o o o N N ...... N 8:8828
N O O O o O O o N o O ... ... ... N 93.80
m N. o o o o o N o N o ... ... ... ... 0 5230280
N o o o o c o o o N o ... N 58883
N o o o o o o o N o N .N.: N 532502..
N o o o o o c o v o o ... .. v NmbNEN
N o o o o o o w o o o .NNN w uNcENNNoNNNN ENNNNNENNNNN
N N o o o O o O o o O 33...; N 9306?.
\B have: Nod 08:. 0:2 .5 SEE .NwNNNNNMN .N.< wEmNN 3303003.—
NEoEPNNN u Noom mmU mN< Bow—:2 NNNNMN omNNNNNN< N333 Seam -853 .284 03>? NBENNN a NNNQNNNoNMNNN o>NNu<

 

NooNNNNNNcoo ON 2an

103

Whitefly

Whitefly was the most commonly reported pest of beans. Ninety-six percent of
farmers reported this pest as problematic in their bean crop, and 91% of the farmers
considered it very damaging to bean yield. The names farmers used for Whitefly are
given in Table 30. It is most commonly referred to as mosca blanca, palomilla, or
mariposita. Only three farmers were unfamiliar enough with Whitefly that they did not
have a name for it and one apparently confused it with spider mite, calling it araiia. The
vast majority of producers (91%) used some sort of chemical control for Whitefly. The
most frequently reported active ingredient for chemical control of Whitefly was
buprofezin. This chemical was used by 45% of the farmers interviewed, and unlike other

frequently used products, buprofezin was used to treat only one production problem.

Weevil

Weevil was the second most commonly reported pest of beans. Seventy-five
percent of farmers reported this pest as problematic in their bean crop, and 55% of the
farmers considered it very damaging to bean yield. The names farmers used for weevil
are given in Table 30. It is most frequently called gorgojo. Other names include
redondilla or polilla. One farmer had no name for the weevil and one called it pulgon, a
name typically used for aphids. The majority of producers (57%) used some sort of
chemical control for weevil. The most frequently reported active ingredient for control of
weevil, malathion, was used by 23% of the farmers interviewed to control the pest.
Farmers added the chemical to harvested seed, and a few farmers reported using

malathion on seed that they later consumed. Fourteen percent of the farmers interviewed

104

used aluminum phosphide fumigation to control weevils in harvested seed and about 9%
reported using a fungicidal seed treatment containing captan, carboxin and thiram to
control weevils. Weevils were the only production problem of beans for which farmers
reported using some sort of alternative method for control. Such controls were usually
used for seed set aside for consumption, and included mixing ash, oil, or eucalyptus
leaves with the seed, sunning harvested seed, and sorting seed. Thirteen percent of

farmers used at least one of the alternative methods to control weevils.

Spider Mite

Spider mite was a problem in the bean crops of half of the farmers interviewed
and 36% of the farmers considered mites very damaging to bean yield. The names
farmers used for spider mites are given in Table 30. It is most commonly referred to as
arafia or arafiita roja. Four farmers did not have a name for it. Thirty-eight percent of
producers applied some sort of chemical control for spider mite. Farmers reported using
ten different active ingredients to control spider mite, the most common of which were

endosulfan, amitraz, methamidophos, and profenofos.

Leaniner

Leaf miners were identified as problematic by 46% of the farmers, but only 25%
considered it very damaging to their bean crop. The names farmers used for leaf miner
are given in Table 30. The most commonly used names were minador and toston.
Thirty-one percent of the farmers who identified leaf miner as a problem did not have a
name for it. A few farmers, who used the names lancha or hongos, apparently confused

the damage caused by leaf miner with fungal pathogens. Forty percent of farmers

105

reported using some sort of chemical control for leaf miner. Farmers reported using
fifteen different active ingredients to control leaf miner. The most commonly used
chemicals were cypermethrin, lufenuron, methamidophos, profenofos, and thiocyclam

hydrogen oxalate.

Rust

Rust was the most commonly reported disease of beans. Eighty-five percent of
farmers reported this disease as problematic in their bean crop, and 80% of the farmers
considered it very damaging to bean yield. The names farmers used for rust are given in
Table 30. It is almost universally known as roya, although the names eumollador and
royal are also used. Five farmers did not have a name for rust. The vast majority of
producers (84%) used some sort of chemical control for the disease. The most frequently
reported active ingredients for control of rust were oxycarboxin and cyproconazole,
although farmers reported using nine other chemicals as well, three of which were

insecticides.

Root Rot

Fifty-nine percent of the farmers interviewed identified the photographs of root
rot as a problem they had had in their bean crop although only 23% considered it very
damaging. Some farmers recognized the disease as being caused by a fimgal pathogen
while others reported seeing root damage, similar to that in the photograph, which they
believed was caused by an insect larva. The names farmers used for root rot are given in
Table 30. The most common name for root rot was pudricion de raiz. Thirty-eight

percent of farmers used some sort of chemical control for the disease. The most common

106

control reported was use of a seed treatment containing captan, carboxin, and thiram.
Farmers who correctly or incorrectly believed the root damage to their beans to be caused

by an insect used carbofuran, aldicarb and chlorpyrifos to treat the problem.

Anthracnose

Fifty-two percent of the farmers interviewed identified the photographs of
anthracnose as a problem in their bean crop, and 21% considered it very damaging. The
many names farmers used for anthracnose are given in Table 30. The most commonly
used names were la gota and antracnosis. F orty-one percent of the farmers interviewed
used some sort of chemical control to treat anthracnose. The most frequently used active

ingredients were mancozeb, propineb, and cymoxanil.

Mildew

F orty-ei ght percent of the farmers reported problems with mildew in their bean
crop and 21% considered it very damaging. As shown in Table 30 ceniza was the most
common name used for mildew. Some farmers apparently confused the photograph of
mildew with damage caused by spider mites or Whitefly, since they called the disease
mosca or acaros. Thirty-six percent of the farmers used some sort of chemical control

for mildew. The most commonly used active ingredient was sulfur.

Common Bacterial Blight (CBB)

F orty-ei ght percent of the farmers interviewed reported that they had seen damage
similar to that caused by CBB in their bean fields, and 20% believed the problem to be

very damaging. The twelve different names farmers used for the damage caused by CBB

107

are given in Table 30. Twenty-nine percent of the farmers used pesticides to treat CBB,

the most common of which were chlorothalonil, cymoxanil, and mancozeb.

Angular Leaf Spot (ALS)

Only 23% of the farmers interviewed identified ALS as a problem in their bean
crop and only 2% considered it very damaging. Most farmers who reported seeing
damage like that caused by ALS did not have a name for the disease. Those who did
report a name called it by one of the generic terms for foliar diseases, mancha, laucha or
lancha. One farmer called the disease antracnosis. Only 14% of the farmers used some

sort of chemical control for ALS and each farmer used a different active ingredient.

Bean Common Mosaic Virus(BCMV)

Damage like that caused by BCMV was reported by 20% of the farmers
interviewed and 5% considered it very damaging. Most farmers did not have a name for
the problem. Those names which farmers did report are given in Table 30. Farmers did
not report using any chemical control for the problem.

Pesticide Use

Overall, 98% of the farmers interviewed applying pesticides to their bean crOp.
As shown in Table 32, 59% of the farmers applied chemicals classified as “moderately
hazardous” or “highly hazardous” by the WHO (IPCS-IOMC, 2000). Twenty-seven

percent applied only pesticides rated “unlikely to be hazardous” or “slightly hazardous.”

108

 

Table 32. Classification of farmers by most hazardous pesticide level used on bean

 

 

Pesticide Use Category # of Farmers Percent of Farmers Interviewed

Most dangerous pesticide used is 1 1 8
classified “Extremely Hazardous" '

Most dangerous pesticide used is 13 23 2
classified “Highly Hazardous” '

Most dangerous pesticide used is 20 35 7
classified “Moderately Hazardous” °

Most dangerous pesticide used is 12 21 4
classified “Slightly Hazardous” '

Most dangerous pesticide used is 3 5 4 l-

classified “unlikely to be hazardous” '

Did not recall names of pestrcrdes 6 10.7

used
Did not apply pesticides to bean 1 1.8

 

4:.

 

Farmers’ Suggestions for Bean Variety Improvement

Farmers were asked what characteristics of their current bean varieties should be
improved and they were asked to list the characteristics of a good bean variety. Farmers’
responses to this question are summarized in Table 33.

Yield and disease resistance were the traits that farmers most often thought
needed to be improved in their current varieties. Few farmers mentioned seed
characteristics in their lists of traits to improve. However, 61% of the farmers mentioned
traits related to seed quality as important in a good bean variety. This suggests that most
farmers do not think the seed quality of their current varieties needs to be improved.
However, since seed quality is important to farmers, new bean varieties must posses a

seed quality similar to that of the varieties currently grown

109

 

Table 33 . Desirable bean variety characteristics identified by farmers

 

Desirable Characteristic Cited for Improvement Listed as Important for a

 

 

Categorized in Current Varieties Good Bean Variety
Specific # Farmers % Farmers # Farmers % Farmers
Yield Characteristics 30 53.6 25 44.6
high yield 30 53.6 24 42.9
long pods l 1.8 l 1.8
Seed Characteristics 9 16.1 34 60.7
heavy seed 4 7.1 23 41.1
seed color 0 0.0 15 26.8
shiny seed 1 1.8 7 12.5
"quality " seed 4 7.1 O 0.0
Plant Characteristics 24 42.9 29 51.8 I
disease resistant/healthy 20 35.7 11 19.6
large plant 2 3.6 7 12.5
vigorous/abundant foliage O 0.0 15 26.8
abundant flowers 2 3.6 2 3.6
locally adapted 1 1.8 2 3.6
drought resistant 1 1.8 0 0.0
Harvest Characteristics 8 14.3 8 14.3
early maturity 7 12.5 7 12.5
uniformity at harvest 1 1.8 1 1.8
Flavor 1 1.8 3 5.4
Marketability 6 10.7 7 12.5

 

Introduction of New Bean Varieties

When asked how they hear about new bean varieties, 23% of farmers responded
that they got such information from an institution or training program. The remainder of
the farmers heard about new varieties from friends or neighbors (43%) and at markets
from intermediaries or vendors (46%).

Farmers were also asked where they would go if they wanted to obtain seed of a
new variety. Over half of the farmers (52%) said they would go to the markets to get
seed of a new variety, another 11% said they would go to other farmers in the
community. One quarter said they would go to an institution or non-governmental

organization (N GO). Sixteen percent mentioned INIAP in particular, and 9% named

110

other institutions (such as the Ecuadorian Ministry of Agriculture) or NGOs. Five

percent of the fanners said they did not know where to get seed of new bean varieties.

111

Discussion

Farmer Demographics and Agricultural Systems

Most of the bean growers interviewed were 35 or older, had only a primary school
education, and were members of households of five or more people. Although almost all
of the farmers were land owners, most owned one hectare of land or less. Such data
support the observation that most of the bean growers in the Mira and Chota valleys are
practicing small scale, labor intensive agriculture, as farm sizes are small and household
size is large.

The majority of the farmers reported growing only two or three different crops
within the past two years. However, the diversity of crops grown in the region was quite
large (Table 26) with 19 different crops reported in addition to bean. Some of this
diversity is probably due to the different microclimates present in the Mira and Chota
Valleys. Sugar cane and cassava, for example, were grown in lower elevation
communities, while carrot, pea, and potato were grown at somewhat higher elevations.

Almost all of the crops that farmers reported growing are unquestionably cash
crops (i.e. tomato, sugar cane, pepper, anise) and not subsistence crops. Only a few crops
could serve as either cash or subsistence crops, specifically bean, maize, cassava, potato,
pigeon pea, and camote. It seems that farmers use their limited land to grow only a few
labor intensive cash crops. The fact that these farmers are not subsistence farmers, but
are instead small-scale producers of labor intensive fruit and vegetable crops means that
prices and markets of the crops they grow have a substantial impact on their livelihoods.
Beans are uniquely suited to this type of system since they bring a good market price, yet

could sustain the farmers and their families nutritionally if prices dropped so low as to

112

 

make selling the crop unprofitable. Some of the other cash crops that the farmers
frequently grow, such as tomato, peppers, sugar cane, and onion, do not afford this
potential. The fact that beans are such an important part of the diets of the producers in
the Mira and Chota valleys reinforces their dual value to farmers in this type of
agricultural system.
Farmers’ Variety Choices in the Mira and Chota Valleys

Previous surveys of bean growers in the Chota Valley reported that farmers were
primarily growing the Calima classes for export to Colombia, with the exception of some
smaller farmers who grew Panamito type beans. In the survey reported here, however,
farmers most frequently reported growing medium red mottle and black beans. The data
presented in Figure 17 reveal that Calima Rojo has very low recent adoption rates
compared to the other important classes. In comparison, a high percent of the farmers
growing the Black, Panamito, Matahambre, Canario and Bayo Blanco classes have
adopted them recently. Although these are classes that have historically been grown in
the region, they have not been the farmers’ choice for commercialization. The farmers’
comments and the survey results suggest that since exports to Colombia have been halted
some bean producers are dropping the Calima Rojo class in favor of classes that have a
domestic market. Classes such as MRM, Blanco de Leche, and Calima Negro, which
have been grown as cash crops in the past, continue to be grown because they have a
domestic market. Finally, the trend toward growing some of the previously minor classes
for sale is probably not exclusive to small-scale producers in the region. Although
primarily small farmers were interviewed, the producer with the largest area planted to

bean (10 ha) was growing Canario, Bayo Blanco, Matahambre, and MRM beans.

113

Prices and Marketing

The prices that farmers reported receiving for their most recent bean harvest
validate their move away from Calima Rojo. The highest prices that farmers reported for
dry beans, $40/quintal, were for MRM and Calima Rojo (Figure 16). However, Calima
Rojo also brought the lowest price reported for dry beans -- $8/quintal. Calima Rojo’s
variable price is probably an important part of farmers’ decision to grow other classes.
The prices reported for Calima Negro and MRM remained more consistently high, and
this may explain why these two classes have more recent adopters than Calima Rojo.

On average Black beans brought lower prices than the Calima and MRM classes,
but the prices were more reliable than that of Calima Rojo. Additionally, farmers
reported that black beans are more disease resistant and yield better than the Calima
classes, and this may make them equally, if not more, profitable than Calima. Like black
beans, Panamito and Matahambre classes brought, on average, lower but stable prices.
The solid red landrace grown by one farmer brought a very low price; probably because
there is little market for this class.

Among the varieties sold in tierno Blanco de Leche, which is grown specifically
for sale during Holy Week, brought the highest price. This explains why farmers risk
growing the class when the window of time for marketing it is so short. However, to
minimize risk they plant limited amounts of the variety. In the case of the MRM class,
farmers reported that they decided whether to sell their crop in tierno or dry depending on
current market prices. If tierno prices are good it is desirable to harvest the crop so that
the next cycle of cropping can begin sooner. However, if tierno prices are low farmers

can wait and hope to make more money on the dry seed. The option of selling MRM at

114

two different stages may make the class particularly appealing to farmers since it offers
some flexibility and protection against low prices.

As expected among producers growing a crop for sale, farmers’ decision of which
bean classes to grow seem to be based in part on the price that the class will bring and the
stability of the price. Calima Rojo, a class with few recent adopters, is also the class with
the most variable price. The most frequently cultivated dry bean classes, MRM, Blacks,
and Calima Negro, are those with stable, acceptable prices.

INIAP bean breeders should consider that farmers’ bean variety choices are based
at least in part on the marketability of the class, and that farmers in the Mira and Chota
valleys are changing the classes they grow as a result of changes in the market. It is
important to include in the PRONALEG breeding program some of the previously minor
classes, which are gaining importance. Some of the classes are basically landraces and
have received little attention from breeders.

The interviewed farmers marketed beans to vendors in the city of Ibarra and the
town of Pimanpiro, or through intermediaries who came to the farmers’ communities.
Since none of the farmers reported selling beans directly to the consumer, farmers may
not have a good idea of what class and quality of bean is most marketable. Consequently,
participatory bean breeding efforts would benefit from involvement of bean
intermediaries or vendors, who will be knowledgeable about the local domestic market
and its quality requirements.

Farmers’ Seed Sources
The majority of the farmers interviewed did not plant seed that they had grown

and saved. This finding is surprising considering that small-scale farmers are usually

115

 

characterized as saving their own seed. Although farmers were not specifically asked
why they decided to save their own seed or to buy it, there are several reasons why
farmers might choose to buy seed. Some farmers commented that they thought it was
good to plant seed that was grown in a different climate (i.e. a farmer living in a warmer,
low elevation community should plant seed grown at a higher elevation and vice versa).
Farmers thought that this exchange of seed resulted in healthier plants. Another possible
reason is that farmers’ stored seed may be damaged by weevils. (Some farmers reported
that their weevil control method is to sell the whole harvest as soon as possible.) If a
farmer does not grow beans each season, it may be desirable to purchase fresher, higher
quality seed that has just been harvested. Although the high incidence of farmers
purchasing seed could be a product of so many farmers switching bean classes, this is
probably not the only explanation. Classes with a lower percent of recent adopters,
MRM and Calima Negro, still had a majority of farmers using purchased seed.

The exceptions to the trend of purchased seed were the solid red, Musgo, and
Blanco de Leche classes. The solid red and Musgo beans can be characterized as
noncommercial, farmer-maintained landraces, which explains why farmers saved their
own seed of these classes. Blanco de Leche, in contrast, is grown as a cash crop but
unlike the other commercial classes, most farmers growing it maintained their own seed
over the long term. This may be because production of the class is localized in two
neighboring communities, and seed of the variety is not readily available in the markets --
especially since it is always sold in tierno. One of the farmers growing Blanco de Leche

commented that seed of the class is very expensive.

116

 

 

The long-term success of PCI can hinge on whether or not farmers save their own
seed. In a participatory maize breeding project in Honduras, half of the participating
farmers switched to different varieties and discontinued their work with the project over
four cycles of selection. The breeders concluded that “For areas where farmers are
accustomed to switching varieties or ‘refreshing’ their seed from outside sources, even if
infrequently, PPB [participatory plant breeding] may not be appropriate due to its long
term nature (Smith et al., 2001).” Whether farmers select a variety through trials of
advanced lines or participate in the development of a variety through selection in earlier '
generations, the impact of the improved variety and its usefulness to farmers depends on

their access to seed of the variety year after year.

 

Farmers in the Mira and Chota valleys are without a formal seed system and for
the most part rely on the grain markets for seed. Consequently the purchased seed that
they plant is often a mixture of varieties that produce the same seed type. A study of
pigeonpea in Kenya demonstrated that a variety’s integrity can be maintained in such a
system if it is distinctive. The particular pigeonpea variety tracked in the study was
easily distinguished from local varieties by its white seeds and determinant growth habit
(Jones et al., 2001). Since a difference in seed appearance is part of what allowed the
pigeonpea variety to be maintained in the informal seed system, pure seed of a new
variety would not likely be maintained if the seed closely resembled existing varieties. In
the case of beans in Ecuador seed appearance determines marketability. Consequently
the characteristic shape, size, pattern, and color of the existing commercial classes must

be maintained in new varieties if they are to be marketable. Without a major change in

117

the current seed system, the best way for farmers to assure themselves access to seed of a
preferred improved variety is to plant seed they have grown themselves.

Farmers’ frequent seed purchases and the current bean seed system in this region
of Ecuador are, for the most part, not compatible with the PCI approach. The exception
is the farmers growing and maintaining the Blanco de Leche class. This class should be
considered as a candidate for improvement through PPB. Furthermore, the reasons for
farmers’ seed saving choices should be investigated, and any problems that lead farmers
to purchase seed (i.e. weevil infestation) should be addressed.

Farmers’ Production Problems in the Mira and Chota Valleys

Whitefly

Whitefly was the most frequently reported pest problem of bean in the Mira and
Chota valleys. The vast majority of farmers are familiar with the Whitefly, as they called
the pest by one of four distinctive common names: mosca blanca, palomilla, marz’posita,
or mariposa blanca. Its prevalence and negative impact are evidenced by the fact that
almost all farmers were familiar with the pest and considered it very damaging to bean
yield (Figure 18). Although almost all the farmers applied some sort of chemical control
for whitefly, many expressed frustration in trying to control the pest. There is evidence
that Whitefly in the region has already developed resistance to some pesticides, as a CIAT
affiliated group of researchers working in the Chota valley reported finding whiteflys that
were resistant to cypermethrin, methamidophos, and methomyl. Almost half of the
farmers interviewed reported using buprofezin to control whitefly. This finding is
encouraging since buprofezin is very safe (Table 30). For effective control of Whitefly,

however, it must be applied at a precise stage in the insect’s life cycle. The

118

 

aforementioned group was working to teach farmers to scout their fields for Whitefly
larvae and apply buprofezin at the critical time. Such efforts are imperative if farmers are
to learn to safely manage this pest. Any Bean/Cowpea CRSP supported work regarding
whitefly control should be planned with consideration to the research that has already

been done in this area by CIAT in order to avoid a duplication of previous work.

Weevil

Weevils were the second most frequently reported pest of bean. Farmers were
familiar with weevils as evidenced by the fact that most called them by one of three
commonly used names for the pest: gorgojo, redondilla, or polilla. Farmers primarily
used malathion or the fumigant aluminum phosphide to control weevil in stored seed.
Disturbingly, a few farmers reported applying malathion to seed they later consumed.
Although malathion has a low oral toxicity, this is probably not the safest way to control
weevils in beans for consumption. Weevil was the only pest or disease for which farmers
reported using some sort of non-pesticide control. Some of the non-pesticide controls
farmers use, such as mixing stored seed with vegetable oil or ash, have been reported
elsewhere as effective controls for the pest (Cardona, 1989).

Although practical controls for weevils exist, over half the farmers still considered
the pest very damaging to bean yield, suggesting that it is not being adequately
controlled. This pest problem should be addressed as a part of the Bean and Cowpea
CRSP sponsored PCI initiative, since the pest may prevent farmers from saving their own
seed and farmer seed saving is vital to the success of PCI. A first step would be to inform
farmers of the available control options and possibly test them as a part of the CIAL

activities. A suggested second step is to consider implementation of genetic resistance.

119

Seven different Arc gene alleles that encode the arcelin protein, which confers resistance
to the Mexican bean weevil Zabrotes subfasciatus, have been characterized. It has been
reported that the weevil pest in the Mira and Chota Valleys is Acanthoscelides obtectus.
The Arc genes are not as effective against this species. However, effective control of A.
obtectus has been obtained using a combination of arcelin-enriched bean varieties and the

parasitoid Dinarmus basilis (Schmale et al., 2003).

Spider Mite and Leaf Miner

Spider mites and leaf miners are not as common as weevil and Whitefly, and not
considered as damaging to bean yield by farmers. Farmers were more familiar with
spider mites than they were with leaf miners. Most of the farmers who reported mite
problems in beans called them by one of three commonly used names: araiia roja,
arafiita roja, or arafiita. In contrast, a number of farmers did not have a name for leaf
miner or confused it with fungal pathogens. Although only a quarter of farmers
considered leaf miners very damaging to bean yield, forty percent reported using a
chemical control for the pest. Leaf miners are generally considered to have little
economic impact on bean yield making chemical control unwarranted (Schwartz et al.,
1978). The farmers who apply pesticides to control leaf miner probably do so

unnecessarily.

Rust

Rust was the most commonly reported disease of bean. Most of the farmers are
familiar with rust and call it roya. Beside whitefly, rust is the other major production

problem that farmers applied pesticides to control. Most farmers used products

120

 

specifically recommended for rust: oxycarboxin and cyproconazole. Although a few
farmers reported choosing to grow rust resistant varieties or classes, most farmers used
fungicides to control the disease. PRONALEG’s current CRSP sponsored work to
develop rust resistant bean varieties for the Mira and Chota valleys could be very

beneficial to farmers by reducing production costs and pesticide exposure.

Root Rot

Although over half of the farmers identified the photograph of root rot as a picture
of something that is a problem in their bean fields it is uncertain that all such farmers
were actually experiencing a problem caused by fungal root rot pathogens; particularly
since some of the farmers identified the picture as damage caused by an insect larva that
bores into the plant’s stern. Despite this potential confusion, it is likely that some of the
farmers are experiencing root rot infestation in their bean fields, since over half of those
reporting the problem identified it as a fungal problem.

Root rot is controlled most effectively by cultural practices, particularly crop
rotation, and potentially by use of resistant varieties. The seed treatment containing
captan, carboxin, and thiram that many farmers reported using protects bean plants
against death (damping off) from the root rot pathogens as seedlings, and would allow the
plant to establish itself. However, this treatment would not continue to protect the mature
plant’s roots against root rot pathogens throughout the growing season.

A survey of fungal pathogens and root attacking pests in bean fields in the Mira
and Chota valleys is necessary to determine the true prevalence of the problem and to
identify the actual pests or pathogens present. Root rot resistant varieties may be the best

long-term solution to root pathogens in this region. Crop rotation may also be a viable

121

 

solution to both pest and fungal problems but depends on the practicality of the rotations
on such small, intensively cultivated farms. If the root problems are primarily caused by
insect larvae, farmers would benefit from recommendation of chemical control options
that are less toxic than carbofuran and aldicarb, which some reported using to treat root

problems (Table 30).

Anthracnose

About half of the farmers interviewed identified anthracnose as a problem in their
bean crop. Farmers had many names for anthracnose, some of which were specific to the
disease and others which were generic names for foliar diseases. The symptoms of
angular leaf spot (ALS) are similar to those of anthracnose. However, the difference in
frequency with which ALS and anthracnose were reported suggests that farmers
differentiate between the two. With the exception of two cases, farmers used the most
common specific names for anthracnose, la gota and antracnosis, only when referring to
this disease.

Although many farmers reported anthracnose in their fields, less than a quarter
considered it very damaging. Farmers probably consider anthracnose less damaging
because its occurrence is intermittent and its severity depends on the presence of cool,
wet environmental conditions, which are somewhat rare in this region. Nevertheless,
many of the farmers interviewed used fungicides to treat the disease. If anthracnose is
truly not very damaging, this may be another example of farmers applying pesticides
unnecessarily.

Farmers in the Mira and Chota Valleys, particularly those growing at higher

elevations where anthracnose is more common, could benefit from anthracnose resistant

122

 

varieties. Such varieties would protect farmers from any losses that do result from the
disease and would prevent unnecessary fungicide applications prompted by the

appearance of diseased plants.

Mildew

Nearly half the farmers interviewed reported mildew as a problem in their bean
crop but less than one quarter considered it very damaging. Most of the farmers who had
experienced problems with the disease identified it by its common name, ceniza. The
most commonly reported control for mildew, sulfur, is fairly safe and considered
effective against the disease. Mildew is not a problem for most bean growers, and those

who do see the disease in their fields control it effectively.

Common Bacterial Blight (CBB)

Almost half of the farmers reported seeing damage like that caused by CBB in
their bean fields. However, as evidenced by the names given for the disease, there exists
some confusion over what causes this damage. A few farmers identified CBB as a
bacterial disease (bacteria, bacterosis) or gave it a unique name different than other foliar
diseases (lancha amarilla). It is assumed that these few farmers were familiar with the
disease. Other farmers called CBB laucha or lancha, which are unspecific names used
for foliar diseases. Some farmers thought that the damage had been caused by freezing
(helada) and others thought the damage was from insects (minador, toston, and mosca).
It is unclear if farmers did actually have CBB in their fields or not.

Since very few farmers were familiar with CBB and only a few applied any

pesticide to treat the problem, it is unlikely that CBB is currently a major bean production

123

problem in this region. If the disease is present, however, it could be easily spread
through infected seed. Farmers are probably unprepared to recognize or treat this disease
since so few identified it correctly and none of the fungicides or insecticides that they

reported applying for CBB are likely to control it.

Angular Leaf Spot(ALS)

Few farmers reported ALS as a problem in their bean fields and very few
considered it damaging. As mentioned earlier, it seems that farmers are able to
distinguish between ALS and anthracnose. None of the farmers gave the disease a
distinctive name and few reported applying a chemical control for it. These data suggest

that ALS is, at most, a minor bean production problem in this region of Ecuador.

Bean Common Mosaic Virus (BCM V)

A few farmers reported damage like that caused by BCMV and very few
considered it damaging to yield. None of the farmers who reported BCMV-like
symptoms in their bean plants identified the problem as a virus, which may indicate that
the disease is not a problem in the region. Some thought the damage was caused by
insects (toston, enrollador) and one identified the problem as nutrient deficiencies. Some
of the reported instances could be cases of a physiological problem that is common in the
variety Paragachi. BCMV is probably not a problem in this part of Ecuador, and the
Whitefly species present in the region reportedly does not vector the disease.

Pesticide Use Among Bean Producers in the Mira and Chota Valleys
Previous studies have found pesticide use among small-scale Ecuadorian farmers

to be quite high. A study in the northern Ecuadorian province of Carchi (the Chota

124

Valley is in this same province) found that small-scale potato producers made frequent
applications of a variety of pesticides. They also found that some pesticides that are quite
dangerous, notably carbofuran and methamidophos, are in common use (Crissman et al.,
1998). Pesticide use among bean growers in the Chota Valley is characterized as high in
the report of the 2000 PRONALEG survey. According to this study, the most commonly
used fungicides were mancozeb, cyproconazole, sulfur and maneb and the most common
insecticides were methamidophos, profenofos, and cypermethrin. In the recent
participatory rural appraisal done in the Mira Valley the most commonly reported
pesticides were: mancozeb, maneb, sulfur, propineb, chlorothalonil, malathion,

methamidophos, profenofos, cypermethrin, carbofuran and lambda cyhalothrin. The

 

If

most dangerous among these are the insecticides, methamidophos, cypermethrin, and
carbofuran.

Concurrent with previous findings, practically all of the farmers interviewed for
this study applied pesticides to their bean crop. However, the product which farmers
most frequently reported using, buprofezin, did not appear in any of the earlier surveys.
As noted previously, farmers used buprofezin only to treat Whitefly. It is much safer than
the other insecticides used to control the pest, but requires precise timing of application.
The widespread use of buprofezin is surprising, but encouraging, since it suggests that
farmers are choosing less toxic alternatives to some of the products they had been using.
Similarly, oxycarboxin, a fungicide used to treat rust, does not appear in the reports from
previous surveys but was used by one third of the farmers interviewed in this study.
According to the World Health Organization (WHO) pesticide ratings for acute toxicity

oxycarboxin is safer than the other commonly used rust fungicide, cyproconazole.

125

Despite some indication that farmers are choosing to use less toxic pesticides,
many are still using the insecticides carbofuran, methamidophos and methomyl, which
are classified as “highly hazardous” by the WHO (Table 30). The observation that
methomyl, and methamidophos were commonly reported for the control of leaf miner
and Whitefly, suggests that application of these pesticides is, in most cases, unjustified.

As noted previously, leaf miner is not an economically important pest and Whitefly has

 

already developed resistance to these chemicals in some areas of the Chota Valley.
Likewise, most use of carbofuran is probably unnecessary. Farmers most commonly
reported using this insecticide for the control of root rot. If the problem that these

farmers were experiencing was indeed fungal, carbofuran will be of little use. Although

 

nearly all of the small-scale bean growers in the Mira and Chota Valleys continue to use
pesticides some are using them more safely (and possibly more effectively) than others.
As indicated by the number of farmers using pesticides, bean producers in the
Mira and Chota Valleys are willing and able to spend money on chemical control,
probably because most of them sell a part of their bean harvest depend it as a source of
income. Consequently, chemical inputs should not be excluded as an option for
recommended bean management practices in this region. Farmers would benefit from
straight-forward instruction in safe pesticide use, identification of pests and diseases in
bean, and recommended control methods.
Farmers’ Recommendations for Variety Improvement
According to farmers, yield and disease resistance are the two traits that most
need improvement in their current varieties (Table 32). Therefore, the goals of the,

Bean/Cowpea CRSP supported project, to improve yield and disease resistance fit well

126

 

with farmers’ needs. Although not many farmers thought that seed characteristics needed
improvement, most farmers mentioned at least one seed characteristic when describing a
good bean variety. This indicates that most farmers are satisfied with the seed quality of
their current varieties, but new varieties must possess similar seed quality in order to be
acceptable.

The farmers’ preference for earliness (ready for harvest in 3 months) is
contradictory to the recommendations to increase yield and plant size. It is necessary to
clarify farmers’ priorities regarding these traits. An economic analysis of the potential
benefits and drawbacks of varieties with shorter cycles and lower yields versus those with
longer cycles and higher yields in this agricultural system could help to elucidate the best

combination of traits.

127

 

 

Conclusions

The objective of this study was to learn about the production practices, problems
and variety preferences of bean growers in the Mira and Chota Valleys in northern
Ecuador, with the purpose of making recommendations for developing improved
varieties for the region. In the Mira and Chota valleys beans are important part of the diet
and an important source of income. Most of the bean producers are intensively
cultivating a few cash crops on small landholdings of less than 1 ha.

Although farmers in the Mira and Chota Valleys traditionally grew Calima type
beans for export to Colombia, it appears that bean growers are responding to the loss of
this market by growing different bean classes which are consumed domestically. In this
survey the most widely cultivated bean class in this region was medium red mottle, with
black beans as the second most common class. Other classes that farmers are adopting
include Calima Negro, Panamito, Canario and Bayo Blanco. If export to Colombia
remains an impossibility the aforementioned classes, in addition to MRM and small black
beans, merit more attention from plant breeders than Calima Rojo. Improved Central
American black bean germplasm from CIAT and the Michigan State University bean
breeding program should be tested in Northern Ecuador and integrated into the
PRONALEG-GA bean breeding program.

Unexpectedly growers tended not to save their own seed of the beans they grew,
but rather purchased it. This practice may be incompatible with the participatory crop
improvement (PCI) methodology being implemented in the area, since farmers will not

maintain the selected lines or varieties themselves over the long term.

128

 

The problem using PCI and releasing new varieties in an area where farmers do
not maintain their own seed should be addressed from several different angles. The
question of why farmers purchase seed warrants further investigation and any problems
which prevent them from saving their own seed (i.e. weevil infestation) should be
addressed. PRONALEG-GA could try restricting their PCI efforts to farmers who
maintain their own seed. This approach may be difficult, since so few farmers saved seed
over the long term. Perhaps the most effective solution would be for PRONALEG-GA to
collaborate with informal seed vendors and cooperatives that are operating in this region.
At least one cooperative sold bean seed to farmers and then purchased the ensuing
harvest. If the cooperative provided farmers with seed of an improved variety this would
be an ideal way for PRONALEG-GA to release a new variety in this region. The
cooperative would be able to maintain the new vareity reliably, by reselling some of the
seed purchased from farmers to whom they had provided the improved variety.

Farmers’ biggest bean production problems were Whitefly and rust. Other pests
and diseases of concern are anthracnose, root rot, and bean weevil. The problem of bean
weevil should be addressed by characterization of the weevil species present in the two
valleys followed by intro gression of Arc genes for weevil resistance in to important
commercial classes. The problem that farmers identified as root rot may be exacerbated
or caused by insect pests. Further investigation is necessary to determine the cause and
an appropriate control for the root problems that farmers are experiencing.

Although some farmers did report spider mites, leaf miner, and mildew these
problems are probably not economically damaging. However, they do prompt farmers to

apply pesticides unnecessarily. It has been noted in past surveys that farmers in this

129

 

11"

region tend to misuse pesticides. Concurrent with previous findings some of the farmers
surveyed used highly dangerous pesticides while others used them unnecessarily or
applied the wrong product. There are some indications, however, that farmer are
choosing to use less hazardous pesticides and that, for the most common problems, they
are applying an effective chemical. Since they get most of their pesticide information
from chemical vendors, farmers would certainly benefit from current, straight-forward
information or publications concerning pest and disease control practices. PRONALEG-
GA should make it a priority to provide such information, particularly publications, to
farmers.

Yield and disease resistance are the two traits that farmers think need to be
improved in the varieties that they currently grow. This farmer recommendation fits well
with the goals of the Bean/Cowpea CRSP project. Although they rarely cited seed traits
for improvement, these characteristics are important to farmers. Therefore, any new
varieties must maintain the seed quality of farmers’ current varieties. In the absence of a
formal seed certification and multiplication industry in Ecuador, PRONALEG-GA could
take the approach of introducing improved varieties that deviate slightly from traditional
varieties in seed shape, size or color. This would allow farmers to distinguish between
seed of the improved variety and that of the traditional varieties. If such an approach is
taken, bean breeders should consult with bean market intermediaries and vendors to make

sure that the new variety maintains the marketability of the traditional varieties.

130

 

"....

 

 

Literature Cited

Almekinders CJM, Elings A (2001) Collaboration of farmers and breeders: Participatory
crop improvement in perspective. Euphytica 122:425-238

Ashby JA, Gracia T, Guerrero MP, Quiros CA, Roa JI, Beltran JA (1996) Innovation in
the organization of participatory plant breeding. In: Eyzaguirre P and Iwanaga M
(Eds) Participatory plant breeding, pp. 77-97 Proceedings of a workshop on
participatory plant breeding, 26-29 July 1995, Wageningen, The Netherlands.
IPGRI, Rome, Italy.

Atlin GN, Cooper M, Bjornstad A (2001) A comparison of formal and participatory
breeding approaches using selection theory. Euphytica 122:463-475

Banziger M, Cooper M (2001) Breeding for low input conditions and consequences for
participatory plant breeding: Examples from tropical maize and wheat. Euphytica
122:503-519

Cardona C (1989) Insects and other invertebrate bean pests in Latin America. In:
Schwartz HF, Pastor-Corrales MA (eds) Bean production problems in the tropics.
Centro Intemacional de Agricultura Tropical, Cali, Colombia, pp 505-570

Ceccarelli S, Grando S, Bailey E, Amri A, El-Gelah M, Nassif F, Rezgui S, Yahyaoui A
(2001) Farmer participation in barley breeding in Syria, Morocco and Tunisia.
Euphytica 122:521-536

Crissman CC, Espinosa P, Ducrot CEH, Cole DC, Carpio F (1998) The case study site:
Physical, health, and potato farming systems in Carchi province. In: Crissman
CC, Antle JM, Capalbo SM (Eds) Economic, Environmental, and Health
Tradeoffs in Agriculture: Pesticides and the Sustainability of Andean Potato
Production. Kluwer Academic Publishers, Boston

Cooper M, Stucker RE, DeLacy IH, Harch BD (1997) Wheat breeding nurseries, target
environments, and indirect selection for grain yield. Crop Sci 37:1168-1176

Courtois B, Bartholome B, Chaudhary D, McLaren G, Misra CH, Manda] NP, Pandey S,
Paris T, Piggin C, Prasad K, Roy AT, Sahu RK, Sahu VN, Sarkarung S,
Sharma SK, Singh A, Singh HN, Singh ON, Singh NK, Singh RK, Singh RK,
Singh S, Sinha PK, Sisodia BVS, Takhur R (2001) Comparing farmers and
breeders rankings in varietal selection for low-input environments: A case study
of rainfed rice in eastern India. Euphytica 122:537-550

Humphries S, Jimenez J, Gallardo O, Gonzales J (2001) Metodologias participativas

para el mejoramiento genético del fiijol comr'rn. Report of a PRGA Small Grant
for Dec 2000 — May 2001. Part 2

131

 

 

!" __.,...< . , , . 4
. .

International Programme on Chemical Safety IPCS, Inter-Organizational Programme for
the Sound Management of Chemicals IOMC (2000) The WHO Recommended
Classification of Pesticides by Hazard and Guidelines to Classification 2000-
2002. (www.who.int/pcs/jmpr/jmpr.htm)

Jones RB, Audi PA, Tripp R (2001) The role of informal seed systems in disseminating
modern varieties. The example of pigeonpea from a semi-arid area of Kenya.
Expl Agric 372539-548

Joshi KD, Sthapit BR, Witcombe JR (2001) How narrowly adapted are the products of
decentralized breeding? The spread of rice varieties from a participatory plant
breeding programme in Nepal. Euphytica 1222589-597

“q
‘1
I

Komegay J, Beltran J A, Ashby J (1996) Farmer selections within segregating populations
of common bean in Colombia. In: Eyzaguirre P and Iwanaga M, (Eds)
Participatory plant breeding, pp. 151-159 Proceedings of a workshop on
participatory plant breeding, 26-29 July 1995, Wageningen, The Netherlands.
IPGRI, Rome, Italy.

_-N-"'&‘_’o"d“"fll m "J L- '

 

Machado AT, Femandes MS (2001) Participatory maize breeding for low nitrogen :— !
tolerance. Euphytica 122:567-573

Rosas J C (2001) Metodologias participativas para el mejoramiento genético del fiijol
comr’rn. Report of a PRGA Small Grant for Dec 2000 - May 2001. Part 1

Schmale I, Wéickers FL, Cardona C, Dorn S (2003) Combining parasitoids and plant
resistance for the control of the bruchid Acanthoscelides obtectus in stored beans.
J Stored Products Research 392401-411

Smith ME, Castillo F, Gomez F (2001) Participatory plant breeding with maize in
Mexico and Honduras. Euphytica 1222551-565

Sperling L (1992) Farmer participation and the development of bean varieties in Rwanda.
In: JL Moock & RE Rhoades, (Eds) Diversity, farmer knowledge, and
sustainability, pp. 96-112, Cornell University Press, Ithaca, NY.

Sperling L, Ashby JA, Smith ME, Weltzien E, McGuire S. (2001) A framework for
analyzing participatory plant breeding approaches and results. Euphytica
122:439-450

Sperling L, Scheidegger U (1996) Results, methods and institutional issues: The case of
beans in Rwanda. In: Eyzaguirre P and Iwanaga M, (Eds) Participatory plant
breeding, pp. 44—56 Proceedings of a workshop on participatory plant breeding,
26-29 July 1995, Wageningen, The Netherlands. IPGRI, Rome, Italy.

132

 

 

Sthapit BR, J oshi KD, Witcombe JR (1996) Farmer participatory crop improvement. III.

Participatory plant breeding, a case study for rice in Nepal. Expl Agric 32:479-
496.

Trutrnann P (1996) Participatory diagnosis as an essential part of participatory breeding:
A plant protection perspective. In: Eyzaguirre P and Iwanaga M, (Eds)
Participatory plant breeding, pp. 37-43 Proceedings of a workshop on
participatory plant breeding, 26-29 July 1995, Wageningen, The Netherlands.
IPGRI, Rome, Italy.

Witcombe JR, Virk DS (2001) Number of crosses and population size for participatory
and classical plant breeding. Euphytica 122:451-462

133

 

‘1" .L

 

APPENDICES

134

Appendix A: Results of Greenhouse Inoculations of Ecuadorian

Recurrent Parents

Table A 1. Reaction of Ecuadorian recurrent parents to greenhouse inoculation with
various isolates of C olletotrichum lindemuthianum

 

 

Parental Line Race 0" Race 2 Race 73
ARME-Z Rb s S/R
Paragachi S S/R R

ACE-1 R R R
Cocacho S S --
J e.Ma. -- R --
Yunguilla R S R
SEL 1308 R R R

 

a - the same Ecuadorian race 0 isolate that was used in field inoculations
b — R denotes resistant; S, susceptible; S/R segregating with some resistant and some susceptible plants

135

 

Appendix B: Questionnaire Used in Farmer Survey

Figure B 1. Spanish language questionnaire as it was used in farmer surveys

Preguntas Acerca de T enencia de T ierra

gQué superficie de tierra posee?

LCuanta superficie esta cultivada?

 

 

Dibuje un mapa de la propiedad que cultiva.

 

Lote #

{Que superficie
uene?

gCuél de estas
cuatro lineas
parece a la ladera
de este lote?

{,Qué cultivos ha sembrado en este lote en los dos

ultimos afios?

 

I . F re'jol

2. Tomate
3. Yuca

4. Pimiento
5. Maiz

6. A ji

 

Platano
Camote
Hortalizas
Cafia

. Pastos
l 2.

Otros

 

 

 

 

 

 

 

 

 

 

 

 

 

136

 

 

 

 

 

 

 

 

NMVIOQF

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

moummaa
2sz magma
.NoonNNNoNoE $568 No vapor? 2.8 No
up New“??? sugar? 88 NNNNNNENNSN 83 up 36:? 20on 36:? 83 new: Negotg .2
SN 9:090 up 8.88 82% 2 on .254 888 83% 2 30¢ ovconNN “.30 9:30... 29:54 25:880..v
m
N.
o
m
N.
m
N
N
«No @3888 S385.» «ennui?
«NEE; Newton? 5 2.8: Name 83 up NNNNEuN «N @588 2206 NNNNNBNNEN moment?
88 o 88 883 on 28.5 o NewsNoNNNNwNNo831%.N EDGE 8a. mumov Name 83 USN:
ENNNNNoo up 2:88 NNNNNEoN NNNoNENooNo newsman @553 Noam: 9588 2:323
3 «N oNo N28 N83 oNN SEE can No 055% 26 nuvocg 88 «553 $035 mesa NNN Noam BEE;
0260..N No No NEON ow NEON NEON oNN mNNNEoN «N oNNNwNNNNOQ GENO... NNNNNNNNUN NNov oNNNNNNNNUN N..oNNNN.NN..U..N up 05:52

 

 

 

 

Nabmofi «Nova 2:. wovacotg 33 2d «588 «saw? N56: 25th

-6me 0:53 No 83583 No New on Newswog>
NonNnN on wowwcotm>

137

 

 

 

 

 

 

 

 

 

wfiocuimvh

 

 

c8 33:? 83 «Sn: 8 oEoU.v 66c ow «:8on E «as 02?; ““3238 oEoU.~

E mnzomzmg .5 awavoficaco 63c ow 2088 E a ovflovoE 8% «£528 .a unremmmk .5

98 a «65382 ammo“ 2.5 38¢ at 2880 2 m ocmv 0:02: 332 C mbgoctomqo mwmvuztomco

30¢ on 33:? SE a 80:qu “Umvuctflao 98 52:50 am cm 83 28:89 BED... 2% «Eu: 8 9:004 “x "~th

 

$88 360 m6 «:2 x 0828mm cm 25 33¢ aw @3056“ch «saw? or» 3334 .mowa moEE: mow m2 cm

 

 

 

138

 

 

 

3.6: av £038 3 «mg oz .m

33c ow £038 3 a onEocoE 05% «RED .~
EEC on 2088 S m 0mg 05:8 320 ._ aflobcg 2:004 .a mum??? .5 2”me
awn—a £3 52an mm :m mama—Q Swo 305:8 BHme Emu mEm: mm 089% a SEQ

 

 

 

 

 

 

$98 mfiwu By an: a 0.3808 8 25 36¢ ow mwfla «saw? a 03 @3me .womm 8:53 mow m2 cm

 

wcwcoaofi am so mofimfiomg mi: 3.33% on somoosuoa at £80305 8— :8 moflwsow

afiaususk 3: ngEEE mu 923‘ mu~==umb~

 

~82: Emu :o 03:30 33¢ ow wows—Big “.5an .a 325%? .5
oZ \ Wm mofizso etc :8 35% Exam com: 323qu .883 2.8 E

 

$5318 moumcotg “.504 .2 3:83.? a.
02 \ .5 ~22 08%.: _o :0 36¢ ow mocmcotg 2U £82: a: o>£=0w .833 918 E :m

Preguntas Acerca de las Preferen cias para las Nuevas Variedades de F re'jol

LCuales son las caracteristicas mas importantes de una variedad de fréjol?

 

 

 

 

LQué caracteristicas quisiera que se mejore en las variedades que cultiva?

 

 

 

 

LCo'mo se entera de las nuevas variedades de fréjol? 2

 

 

 

 

 

11..

(JAdénde iria usted para obtener una nueva variedad de fréjol?

 

 

 

 

Si sembra una variedad de fréjol por la primera vez, gouanto semilla quisiera sembrar?

 

 

 

 

(JCuantos afios tiene?

' ué nive] dc ensefianza ha tenido usted? 1. Nin mm 2. AI aben'zacién 3. Pre rimario
6 g P
4. Primario 5. Secundario 6. Superior

Género (por observacién): Hombre Mujer
Cuantas personas conforman su familia?

Con qué frecuencia consume fréjo]?
Diariamente (cuantas veces)? Semanalmente (cuantas veces)?

Qué cantidad de fréjol prepara para cada comida?

139

 

 

Figure B 2. English translation of questionnaire used in farmer surveys

Questions About the Farm

How much land do you own?
How much of the land is cultivated?

Draw a map of the land that you cultivate.

 

 

Field # What is the area Which of these What crops have you planted in this field in the

 

of this field? four lines looks past two years?
like the slope 0f [Bean 7. Bannana
this field? 2. Tomato 8. Camote
3. Yuca 9. Vegetables
4. Sweet pepper [0. Sugar cane
5. Corn 1 I. Pasture
6. Hot pepper 12.0ther

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

140

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m
h
e
m
w
m
N
_
«88> «2: $8.» $2
$338 2:; but? 88; 25
.8895 8 mon— 8oc @383 $8 «but? 88:8 so» so» Eu so» v.6 £8 :8 08:88
3289» 05 8am Scan 85 so» race :83 8: «zoom 85 so» 06 an; Ev 82;? 88 28>? 80:8 3oz so» ED :9» BO
w
h
o
m
w
.m
N
~
A“ ...oNV. 8 $58 when? 88
38:88 >898; £5 538w .8 808 Emacs 30> 88m 83% NAM—88:9 v.88? m5“ abut? 28
bot?» 3102:; 83888 fiber—g >58 Rom 398% 83 «$3 5 some Ev 33. 38 so; ow Esaum :9»
£5 £5 80 Bow v.38 035 m8“ «Burma—wififla 8m but? 85 so» ow 88m 3808 :83 028 88% but;
8 Be! 08 288383 .8 8255 me :83 80 Bow 2: How so» EU 82$, 80:8 Bow :5 5:3 >58 >63 05 80 088A

 

 

141

$8 308w 8.8 :9» 35 8:08; 085 .8 38 088 028 so» on

-89» “man 05 8 8853 Swan .8 mocutm>
8:285» 83:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

what? 22% amen GEE go: moon .m
E832 mEH mo 08m: 2: M. :25) 3% 22» :mon 05 2 umemc 2336:. 8930 .N
b 5830:“ flfi 9 023362 was :2: 23% :83 2: 9 uwmfinv :02: momsau ._ 36$ bib $336 a
:39 go @659» m .3 Sci :3 cm ”83va mi“ doEEo Box E $386 $5 Hub so» 0Q £5 :8 so» ow :23 2.0

 

$993 82: mo 0:0 BE $62 35 0836 :89 a :03 so» PE; 98% 02: “man 2: E

 

 

 

 

 

 

 

22% :82 ES; 5: moon .o

Eu; anon 2: 8 umenw 28258 momsmu .m
EB» :82 2t 2 owmfimw 50:8 mumsmu .v 36: $15 @685 u
“685 £5 doan So» 5 $82,. 25 he Sub so» on 25 :8 so» oc 3:3 Emu

 

 

 

 

 

$993 32: mo 28 8E $62 85 283 So» E 835 cm 58 so» 96: mum?» 95 “man on. E
“HEB.“ So» no 25305 couoscoa Swen EatonEm $08 2: 8m :23

MEmBEE =885~9£ “23‘ 23.5.25

 

933 $5 38w so» Eu 8:2: :39 fink/83%
oZ \ 8.» macho 5598 5:3 Bfiowoa mason 38w so» Ev EPA as 05 E

 

mohdfiE m E 39m 30% Ev 3:29 3:3 aux \N
02 \ 8% ““22.“ 25m 2: E mocotg anon mo 2:35: a 38m :9» Ev so» am. 2: E

142

Questions About Preferences for New Bean Vareities

What are the most important characteristics for a bean vareity?

 

 

 

 

What characteristics do you want in a bean variety which your current varieties don’t

have?

 

 

 

 

How do you hear about new bean varieties?

 

 

 

 

where would you go to obtain a new variety of bean?

 

 

 

 

If you grew 3 new bean vareity for the fist time, how much seed would you want to sow?

 

 

 

 

How old are you?

What level Of education have you had? I . None 2. A lfabetz‘zacz’o’n 3. Preprimario
4. Primario 5. Secundario 6. Superior

Gender (observed): Male Female
How many people are in your family?

How often do you eat beans?
Daily (How many times?) Weekly (How many times?)

What quantity of beans do you prepare for each meal?

143

 

 

 

Appendix C. List of Pesticides Farmers Reported Using and their

Ecuadorian Trade Names

Table C 1. Fungicide products farmers reported using

 

 

Common Name Common Name Commercial Name Family
(English) (Spanish)
benomyl benomil Benomyl, Benlate, Benzimidazole
Pillarben
bupirimate buprimato Nimrod Pyrimidine
captan captan Captan, VitaVax* Thiophthalimide
carbendazim carbendazin Bavistin, Carbendazim Benzimidazole
carboxin carboxin Vitavax* Carboxamide
chlorothalonil clorotalonil Daconil, Fungil Substituted
benzene
copper oxychloride oxicloruro de cobre Sulcox Inorganic copper
cymoxanil cymoxanil Curzate*, Fitoraz* Unclassified
cyproconazole cyproconazole Alto 100 Azole
difenoconazole difenoconazol Score Azole
folpet folpet Folpan Thiophthalimide
mancozeb mancozeb Dithane, Mancozeb, Dithiocarbamate,
Manzate, Triziman, inorganic zinc
Curzate*, Ridomil*,
Sandofan*
maneb maneb Maneb Dithiocarbamate
metalaxyl metalaxil Ridomil* Xylylalanine
metiram metiram Polyram Dithiocarbamate,
inorganic zinc
myclobutanil myclobutanil Rally Azole
oxadixyl oxadixil Sandofan* Anilide
oxycarboxin oxycarboxin Plant Vax Carboxamide
propiconazole propiconazole Tomeo Azole
propineb propineb Antracol, Fitoraz* Dithiocarbamate,
inorganic zinc
sulfur azufre Azufre, Cosan Inorganic
thiram thioram Vitavax* Dithiocarbamate

 

* product contains more than one active ingredient

144

 

 

Table C 2. Insecticide products farmers reported using

 

Common Name Common Name Commercial Name Family

 

(English) (Spanish)
aldicarb aldicard Temik N—methyl carbamate
aluminum fosfuro de alumino Gastoxin Inorganic
phosphide
amitraz amitraz Mitac Formadine
avermectin abermectina Averrnectina Botanical
buprofezin buprofezim Applaud Unclassified
carbofuran carbofuran Furadan N-methyl carbamate
cartap cartap Padan Unclassified
chlorpyrifos clorpirifos Lorsban Organophosphorus
cyfluthrin ciflutrin Bulldock Pyrethroid
cyperrnethrin cipermetrina Cipermetrina, Master Pyrethroid
diazinon diazinon Diazinon Organophosphorus
endosulfan endosulfan Endosulfan, Pahnarol, Organochlorine
Thiodan, Thionex,
Methofan*
lambda cyhalothrin lambda cihalotrina Karate Pyrethroid
lufenuron lufenuron Match Benzoylurea
malathion malathion Malathion Organophosphorus
methamidophos metamidofos Pillaron, Rector Organophosphorus
methomyl metomil Methomex, Metomil, N-methyl carbamate
Methofan*
permethrin permetrina Permasec Pyrethroid
profenofos profenofos Curacron Organophosphorus
sulfur azufre Azufre, Cosan Inorganic
thiocyclam thiocyclam- Evisect Unclassified

hydrogen oxalate hydrogenoxalato

 

* product contains more than one active ingredient

145

 

 

Appendix D. Characterization of Redcoat, a Virgarcus Seed Coat
Pattern Mutant of Red Hawk Dark Red Kidney Bean

Introduction

The soldier bean variety released with the name ‘Redcoat’ originated from a few
off-type seeds found in 1999 in a Foundation Seed lot of ‘Red Hawk,’ a dark red kidney
bean variety (Kelly et al., 1998). Unlike Red Hawk, which has totally colored seed,
Redcoat possess white seed with a red Virgarcus or soldier bean pattern on its ventral side
(Figure D1). The only other observed phenotypic difference between the varieties is in
flower color. Redcoat’s flowers are pure white, while Red Hawk’s flowers are white
with pale red veins in the wing petals. Red veins in the flower are conferred by the
recessive rk‘i allele for red seed coat color.

Figure D 1. Seed of Red Hawk and Redcoat

 

 

 

 

 

The circumstances of Redcoat’s discovery aroused suspicions that it could be a
seed coat pattern mutant of Red Hawk, and preliminary testing supported this possibility.
When grown in the greenhouse in Winter 1999/2000, the few original Redcoat seeds
produced progeny and seed that were true to type and did not segregate, indicating that
the off type seeds were not the result of cross pollination. Additionally, Virgarcus

patterned beans are not grown commercially in Presque Isle County in Northeast

146

Michigan where Redcoat was first detected, making contamination of the Red Hawk seed
lot with a commercial soldier bean variety unlikely. Finally, the reaction of Redcoat to
inoculation with Bean Common Mosaic Necrosis Virus strain NL3 and C olletotrichum
lindemuthianum races 7 and 73 was identical to the reaction of Red Hawk to these
pathogens and different than the reactions of three other commercial soldier bean
varieties. Red Hawk carries resistance genes (Co-1, C 0-2, and I) that confer a known
reaction to these pathogen isolates. Redcoat appeared to carry the same genes, which
supported the theory that Redcoat is a seed coat pattern mutant of Red Hawk.

According to the work of Bassett (Bassett and McClean, 2000), the inheritance of
partly colored seed coat patterns, like that of Redcoat, is controlled by at least five
interacting loci: T, Z, Bip, J, and F ib. A dominant T allele results in totally colored (also
called self—colored) seed, while the recessive genotype t/t allows the other seed coat
pattern genes to be expressed. This t/t genotype also has a pleiotropic effect resulting in
white flower color. The other seed coat pattern loci determine the shape and the extent of
the colored area. With a dominant J allele, the genotype t Z Bip produces self-colored or
expansa patterned seed; t Z/z Bip produces ambigua; tz Bip produces Virgarcus; and t 2
hip produces bipunctata (Bassett and McClean, 2000). These patterns are depicted in

Figure D2

147

 

 

il'l‘uuxt; \. s
,.

 

Figure D 2. Seed coat patterns and genotypes described by Bassett and McClean (2000)

Self Expansa Ambigua Virgarcus Bipunctata

    

T/-’or.t ZBip r tZBiP fizz BPII V ' i ‘ ’ ' ~ I tz bip
Since two different genotypes could confer Red Hawk’s self-colored seed (T /- or t

Z Bip), a mutation at either the T or Z locus in Red Hawk could result in the Virgarcus
pattern of Redcoat. This study was undertaken to elucidate Red Hawk’s genotype and
determine which gene in Red Hawk had mutated in order to produce the soldier pattern of
Redcoat.

Materials and Methods

The seven crosses shown in Table D1 were made.

Table D 1. Crosses made between Redcoat and Red Hawk and the seed coat pattern tester
lines

 

Female Parent Pollen Parent
Redcoat x Red Hawk

 

 

 

t self-colored BC3 5-593
Redcoat x t cl 2 g b v Virgarcus BC3 5-593
tz bip bipunctata BC3 5—593

 

 

t self-colored BC3 5-593
Red Hawk x 1 cl 2 g b v Virgarcus BC3 5-593
tz bip bipunctata BC3 5-593

 

The lines t self-colored BC3 5-593; 1 cl 2 g b v Virgarcus BC3 5-593; and t z bip

bipunctata BC 3 5-593 are seed coat pattern tester stocks that were developed and

148

provided by M. J. Bassett. The Virgarcus and bipunctata patterned testers were
developed by backcrossing PI 527712, a line derived from the bipunctata patterned
variety Incomparable, to Florida dry bean breeding line 5-593, which has solid black
seeds (Bassett, 1996). The self-colored tester line carrying t was developed by back
crossing the variety Early Wax, to 5-593 (Bassett and Blom, 1991). For all crosses, three
to four pods were harvested from each F 2 plant, and the seed coat pattern was recorded.
The flower color of a sample of the F2 plants was also noted.
Results and Discussion
Redcoat/Red Hawk Population

Of the 273 Redcoat/Red Hawk F 2 plants, 211 produced self-colored seed and 62
produced Virgarcus patterned seed. This fits a segregation ratio of 3:1, self-colored to
Virgarcus (p=0.382) (Table D2). The 28 plants which were recorded as having white
flowers with pink veins all produced self-colored seed, and the 9 plants which were
recorded as having pure white flowers all produced Virgarcus patterned seed. This
suggests that Red Hawk carries a dominant gene conferring a self-colored seed coat,
while Redcoat carries a recessive allele at this locus. When homozygous, the recessive
allele in Redcoat, results in a Virgarcus seed coat pattern, and has a pleiotropic effect,
producing white flowers. These data support the genetic hypothesis that Red Hawk
carries the dominant T allele, which, in this variety, masks other seed coat pattern genes
that confer a Virgarcus pattern (Figure D1). The dominant T allele in Red Hawk mutated

to a recessive t, producing the Virgarcus pattern and white flowers of Redcoat.

149

Table D 2. Test of fit to a ratio of 3 self-colored : 1 partly-colored in the Redcoat/Red
Hawk, Red Hawk/Virgarcus, and Red Hawk/bipunctata F2 generations X2 Goodness of Fit
Test

 

 

 

Seed Coat RH/RC* RH/virgarcus RH/bipunctata
Pattern Observed Expected Observed Expected Observed Expected
Self colored 211 204.75 110 122.25 63 63.75

Partly colored 62 68.25 53 40.75 22 21.25

 

x2 (p-value) 0.76 (0.382) 4.91 (0.027) 0.04 (0.851)

* RH denotes Red Hawk; RC, Redcoat; Virgarcus, t cl 2 g b v Virgarcus BC3 5-593; and bipunctata, tz bip
bipunctata BC3 5-593

 

Flower Color in Redcoat and Red Hawk by Seed Coat Pattern Tester F2 Populations
The flower colors in the F2 generation of the Red Hawk and Redcoat by seed coat
pattern tester lines are given in Table D3. The tester lines are all white flowered since
they carry the recessive t allele. However, the Red Hawk/self-colored and Red Hawk/
bipunctata F2 populations included plants with violet, white, and red-veined white
flowers. The Red Hawk/Virgarcus F2 plants had either white flowers or red-veined white
flowers. This suggests that Red Hawk carries the dominant T allele, since violet flowers
(conferred by V) would not be expressed if Red Hawk possessed t. Violet flowers were
not present in the Red Hawk/Virgarcus F2 plants since the tester lines cam'es v and Red
Hawk apparently does as well. There was no segregation for flower color in the Redcoat

by seed coat pattern tester crosses, which supports the hypothesis that Redcoat carries t.

150

Table D 3. Flower Color in the Red Hawk and Redcoat by Seed Coat Pattern Tester F2
Populations and Inferred Genotypes for Red Hawk and Redcoat

 

 

 

 

Cross*

flavvézlffteyngp“ RH T v rkd/ RC t v rkd/

" self vir bip self vir bip

tVRk thk tVRk tVRk thk tVRk

Violet
T/- V/- X X
White/Red Veins
:r/. v/v rkd/rk" X X X
:pre White X X X X X X

 

* RH denotes Red Hawk; RC, Redcoat; self, I self BC3 5-593; vir, tel 2 g b v Virgarcus BC3 5-593; and bip,
t2 bip bipunctata BC3 5593

Red Hawk/self, Red Hawk/Virgarcus, and Red Hawk/bipunctata Populations

The seed coat pattern frequencies in the Red Hawk and Redcoat by seed coat
pattern tester line F2 populations are given in Table D4 and Figure D3 depicts the seven
seed coat patterns that were observed in the Red Hawk and Redcoat by seed coat pattern
tester F2 populations. Three distinct Virgarcus patterns were observed in these
populations: tester-like Virgarcus, Redcoat—like Virgarcus, and weak Virgarcus.

Seed coat pattern frequencies in the Red Hawk/tester F2 populations support the
theory that Red Hawk carries T. If Red Hawk’s self-colored seed coat were conferred by
the genotype t Z Bip, no partly colored patterns other than expansa would be expected in
the Red Hawk/self F2 population, since the self-colored tester has this same genotype.
However, a few individuals in the population expressed the ambigua and Virgarcus
patterns.

If Red Hawk cam'es T, there is an expected segregation ratio of 3 self—colored : 1
partly colored in the Red Hawk/Virgarcus and Red Hawk/bipunctata F2 populations. As

shown in Table D2, the observed values for the RH/bipunctata population fit a 3:1 ratio

151

 

 

F;.:

 

 

but the data for the RH/virgarcus population, although close to this expected ratio, do not

fit the expected ratio according to the X2 goodness of fit test.

Table D 4. Seed Coat Pattern Frequency in the Red Hawk and Redcoat/Seed Coat Pattern
Tester F2 Populations

 

Cross*

Seed coat Pattern RH/self RH/vir RH/bip RC/self RC/vir RC/bip

 

 

 

Self 78 1 10 63 42 0 0
Expansa 4 0 O 36 O O
Ambigua O 17 5 26 27 O
Redcoat Virgarcus 3 1 l 9 13 31 58
Tester Virgarcus 2 20 O 6 24 0
Weak Virgarcus 1 5 5 5 13 4O
Bipunctata O 0 3 0 7 22
TOTAL 88 163 85 128 102 120

 

* RH denotes Red Hawk; RC, Redcoat; self, I self BC3 5-593; vir, t cl 2 g b v Virgarcus BC3 5-593; and bip,
tz bip bipunctata 5-593

Figure D 3. Seed coat patterns observed in the Red Hawk and Redcoat by seed coat
pattern tester F2 populations

 

 

 

t
r
,f‘x ‘ .I L
. v . .

  

6 coat is eak‘“ 'TOBlpunctata

Virgarcus Virgarcus Virgarcus

 

 

 

 

Finally, the data in Table D4 show that Red Hawk carries the genes for Redcoat’s
Virgarcus pattern cryptically since all of the Red Hawk/tester F2 populations contained

some plants expressing Redcoat-like Virgarcus patterned seed.

152

Redcoat/self Redcoat/Virgarcus, and Redcoat/bipunctata Populations

The frequencies of the different seed coats patterns in the Redcoat/self,
Redcoat/Virgarcus and Redcoat/bipunctata F2 populations are given in Table D4. The
data from the Redcoat/Virgarcus F2 population suggest the alleles conferring Redcoat’s
Virgarcus pattern are not the same as those carried by the Virgarcus tester line, since
plants in this population segregated for seed coat pattern. Unexpectedly, 27 plants
expressed the ambigua pattern and seven plants expressed the bipunctata pattern from the
RC/virgarcus cross. The seven bipunctata plants represent approximately 1/16 of the
population, which suggests that two seed coat pattern genes are segregating.

Redcoat’s seed coat pattern is different than that of the Virgarcus tester. The
colored area on seeds of the Virgarcus tester completely encircles the hilum, while in seed
of Redcoat the arcs of color, which extend from the caruncular end of the seed, do not
join with the micropyle stripe (Figure D3). Since the Virgarcus tester was derived from a
cross between 5—593 and a plant with bipunctata patterned seed, rather than 5-593 and a
plant expressing a Virgarcus pattern, it is possible that this tester does not carry the same
genes as a plant with the “classic” soldier or Virgarcus pattern displayed by Redcoat.
Plants expressing a tester-Virgarcus pattern occurred in the Red Hawk/self, Red
Hawk/Virgarcus, Redcoat/self and Redcoat/Virgarcus F2 populations. This suggests that
the Hip allele from 5-593 is required for the expression of tester-Virgarcus, and that
Redcoat and the tester differ at this locus.

My results from the Virgarcus (Redcoat)/bipunctata cross are similar to those
reported by Lamprecht (1940) for a Virgarcus/bipunctata cross, and different than those

obtained by Bassett (1996). Bassett (1996) observed a 3:1 segregation ratio for

153

 

Virgarcuszbipunctata in his Virgarcus/bipunctata F2 population. Lamprecht (1940),
however, observed three different seed coat patterns segregating in the
Virgarcus/bipunctata F2 population. The three patterns were Virgarcus, bipunctata and an
intermediate pattern which resembles my weak Virgarcus class, in a 9:4:3 ratio. In order
to explain his results Lamprecht proposed that the Bip locus interacts with alleles at an
independent locus, which he called Arc. Redcoat (and Red Hawk) may carry different
alleles than the tester lines at the Arc locus theorized by Lamprecht (1940). F
Conclusions -

Red Hawk carries the dominant T allele, which confers its self colored seed coat

 

and masks the expression of other seed coat pattern genes that the variety carries. The T
allele mutated to recessive t in Redcoat, resulting in a Virgarcus patterned seed coat and
suppressing the expression of red veins in the flowers. The genes conferring Redcoat’s
Virgarcus pattern are different than those of the Virgarcus tester. It is proposed that
Redcoat (Red Hawk) and the testers differ at the Bip locus and possibly the Arc locus as
well.

Literature Cited

Bassett MJ (1996) Inheritance of the partly colored seedcoat pattern, bipunctata, in
common bean. J Amer Soc Hort Sci 121:1032-1034

Bassett MJ, Blom A (1991) A new genotype for white seed coat discovered in ‘Early
Wax’ snap bean. J Amer Soc Hort Sci 116:131-136

Bassett MJ, McClean P (2000) A brief review of the genetics of partly colored seed coats
in common bean. Annu Rep Bean Improve Coop 43299-101

Kelly JD, Hosfield GL, Vamer GV, Uebersax MA, Long RA, Taylor J (1998)
Registration of ‘Red Hawk’ dark red kidney bean. Crop Sci 38:280-281

Lamprecht H (1940) Regarding the genetics of Phaseolus vulgaris XVI. Further
contributions regarding the heredity of partial color. Hereditas 262277-291

154

   

u((101)01ng(11(1))