ENVIRONMENTAL RISK ASSESSMENT FOR INTRODUCTION OF GENETICALLY
ENGINEERED WEEVIL RESISTANT SWEETPOTATO IN UGANDA
By
Barbara Mugwanya Zawedde

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Plant Breeding, Genetics and Biotechnology - Horticulture - Doctor of Philosophy
2013

ABSTRACT
ENVIRONMENTAL RISK ASSESSMENT FOR INTRODUCTION OF GENETICALLY
ENGINEERED WEEVIL RESISTANT SWEETPOTATO IN UGANDA
By
Barbara Mugwanya Zawedde
Genetic engineering (GE) tools have been used for crop improvement for over two decades,
however, widespread acceptance and cultivation of GE crops has been constrained in part by
associated environmental concerns. Bt sweetpotato is one of the GE products being
developed to address weevil infestation, which causes high yield losses in some parts of
Africa where sweetpotato is considered a food security crop. To address the frequently raised
concern of possible impact of GE crops on land race and varietal diversity, microsatellite
marker analysis was performed to assess the level of genetic diversity in sweetpotato in
Uganda and other parts of East Africa and compare to diversity present in samples from the
proposed center of origin in South America A total of 260 sweetpotato accessions, collected
from Uganda, Kenya, Tanzania, Ghana, Brazil and Peru, were characterized using 13 to 16
microsatellite markers. The Ugandan collection has a large number of distinct landrace
genotypes, but there was low (6%) genetic variability between the Ugandan germplasm and
accessions obtained from other East African countries. Within Uganda higher genetic
variability (76%) was recorded within regions than among regions. To gain a better
understanding of how diversity is maintained in farmers’ fields, ethno-botanical surveys were
performed within three sweetpotato producing regions of Uganda to analyze the
considerations used by farmers when making decisions to maintain, incorporate and or
discard varieties, and to assess the changes in the crop management practices implemented
when adopting new cultivars that are, or are not, introduced in concert with promotional
campaigns. In most districts, farmers were growing both landraces and released cultivars.

Maintenance of traditional varieties was influenced by ecological conditions, socio-economic
factors, and crop management practices such as drought, market and extension service access.
To contribute to an environmental risk assessment for Bt sweetpotato, which is required in
Ugandan prior to open release of any new GE crops, problem formulation was used to
identify valued ecological entities in the receiving environment and possible harms associated
with introducing a new technology. A participatory approach was taken using consultations
with local Ugandan scientists and regulators with diverse expertise to identify agronomic
practices and possible harms that may be associated with three GE crop-trait scenarios. Based
on these consultations, the potential harms viewed to be most relevant for each scenario were
identified. For GE weevil resistant sweetpotato, development of resistance by the pest,
sweetpotato varietal diversity loss, diversity loss for other crops, and secondary pests were
identified as the concerns that warrant greatest regulatory consideration. Information obtained
from the sweetpotato biodiversity and farmer practices studies combined with existing
literature was used to address the prioritized concerns. These analyses indicated that
diversity loss for other crops and emergence of secondary pests were unlikely; development
of resistance by the pests and loss of sweetpotato varietal diversity appear to be more likely,
although not unique to deployment of GE crops. Possible risk management strategies to
mitigate these concerns were suggested for consideration prior to open cultivation of weevil
resistant sweetpotato varieties.

DEDICATION
I dedicate this dissertation to my family: my husband, Peter, and our children,
Christopher, Gabriella, Isabella and Michaella, for their constant support and
unconditional love. I love you all dearly.

iv

ACKNOWLEDGEMENTS

Foremost, I would like to express my sincere gratitude to Professor Jim Hancock,
who recognized my potential and offered to sponsor my PhD program under his
Blueberry Project. I am very grateful to my advisors Professors Rebecca Grumet and Jim
Hancock for the continuous support of my PhD study and research, their immense
knowledge, and tireless motivation and enthusiasm for the past four years. Their guidance
helped to ensure that I conduct the research and write of the dissertation in a timely
manner.
I thank other members of my thesis committee: Professors Harris Craig, Russell
Freed, and Hector Quemada, for their encouragement, technical advice and knowledge,
and their helpful career advice. I am also very grateful to Dr. Marc Ghislain of the
International Potato Center (CIP)-Nairobi office who was my CGIAR mentor for the
Norman E. Borlaug Leadership Enhancement in Agriculture Program (LEAP) fellowship,
for sharing his knowledge and the useful comments provided during the conduct of the
research and writing of the chapter funded by this fellowship.
My sincere thanks also go to Dr. Karim Maredia and Dr. Cholani Webadde, who
together with my advisors invited me to Michigan State University to work with them to
develop a web-based resource for the African Biosafety Network of Experts (ABNE), an
African Union/NEPAD initiative. This interaction opened the doors to my PhD training.
I thank my collaborators during the conduct of my research especially Dr Geovani
Amaro of EMBRAPA-Brazil, Eric Magembe of CIP-Nairobi office, Bramwel Wanjala of
the Biosciences Eastern and Central Africa (BecA)-Nairobi, Agnes Alajo and Francis
Osingada of National Crop Resources Research Institute (NaCRRI).

v

I will forever be very grateful to Dr. Theresa Sengooba, my former boss under the
Program for Biosafety Systems. Dr. Sengooba has been helpful in providing advice and
support both for professional and social advancement. She is my role model as a scientist,
mentor, leader, wife and mother. Professor Samuel Kyamanywa of Makerere University
and Dr. Yona Baguma (NaCRRI) have also been very instrumental in guiding my career
advancement, for which I am grateful.
A good support system is important to surviving and staying sane in grad school. I
am very grateful for the support I received from my labmates both in Rebecca Grumet
and Jim Hancock’s laboratory as well as from the members of the Plant Breeding,
Genetics and Biotechnology Program. I am grateful for all the friendships that developed
both at MSU and BecA. I thank all my friends (old and new) for providing support and
encouragement that I needed.
I will forever be thankful to my family for allowing me to take this journey. It has
been very challenging being a student in a foreign land with the family far away, back
home in Uganda. I thank my parents who have worked hard to care for me and my
children, and for the unconditional love they have given us. I thank my siblings and inlaws for caring for the children, and for supporting us during this journey. Most
especially, I thank my best friend and the love of my life, Peter. Peter has been a great
supporter and has unconditionally loved me during the good and bad times. He has been
non-judgmental and has instilled confidence in me because he has faith in me. Together,
this journey has taught us a lot about life and has strengthened our commitment to each
other.

vi

TABLE OF CONTENTS

list of tables ......................................................................................................................... ix
LIST OF FIGURES ...........................................................................................................xii
LITERATURE REVIEW .................................................................................................... 1
Introduction .......................................................................................................................... 1
Part 1. Sweetpotato .............................................................................................................. 2
Taxonomy and biology of sweetpotato ............................................................................ 2
Origin and importance of sweetpotato ............................................................................. 3
Production constraints ...................................................................................................... 6
Genetic Diversity of the Sweetpotato .............................................................................. 8
Sweetpotato breeding ....................................................................................................... 9
Genetic engineering ....................................................................................................... 11
Developing weevil resistant GE sweetpotato ................................................................ 12
Part 2. Environmental risk assessment............................................................................... 15
Problem formulation ...................................................................................................... 17
Objectives of dissertation ................................................................................................... 19
LITERATURE CITED ...................................................................................................... 22
CHAPTER 1 CHARACTERIZATION OF THE GENETIC DIVERSITY OF
UGANDA’S SWEETPOTATO GERMPLASM USING MICROSATELLITES .... 31
Introduction ........................................................................................................................ 31
Material and Methods ........................................................................................................ 35
Plant material ................................................................................................................. 35
DNA extraction and Simple Sequence Repeat (SSR) Amplification ............................ 35
Allele scoring and data analysis ..................................................................................... 37
Results ................................................................................................................................ 42
Genotype description ..................................................................................................... 42
SSR markers amplification ............................................................................................ 42
Genetic relationship between varieties with varying response to weevil infestation .... 42
Determining relatedness between farmers’ varieties and cultivars in Uganda .............. 45
Genetic relationship between genotypes from Uganda’s agro-ecological zones and
accessions collected from other East African countries ................................................ 45
Discussion .......................................................................................................................... 61
Conclusions ........................................................................................................................ 64
APPENDIX ....................................................................................................................... 66
LITERATURE CITED ...................................................................................................... 73
CHAPTER 2 ASSESSMENT OF ON-FARM SWEETPOTATO DIVERSITY, AND
FACTORS INFLUENCING THE LEVEL OF FARMERS’ CROP DIVERSITY:
IMPLICATIONS FOR CONSERVATION .............................................................. 80
Introduction ........................................................................................................................ 80
Materials and Methods ....................................................................................................... 87
Description of the study area ......................................................................................... 87
Site selection .................................................................................................................. 87
vii

Survey methodology ...................................................................................................... 90
Results and Discussion ...................................................................................................... 92
Respondents’ information .............................................................................................. 92
Reasons for sweetpotato production .............................................................................. 92
Land size and ownership cultivated with sweetpotato ................................................... 95
Varietal diversity on-farm and at regional level ............................................................ 97
Impact of farmers’ variety exchange ........................................................................... 102
Observation of volunteers or potential hybrids in farmers’ fields ............................... 105
Criteria for adopting new variety or improved cultivars selection .............................. 106
Frequency of adoption of new varieties or newly bred cultivars ................................. 109
Challenges to adoption of new varieties ...................................................................... 111
Criteria for maintaining a variety in the farmers’ fields .............................................. 115
Challenges to maintaining varieties in farmers’ field .................................................. 116
Crop management practices and how they affect diversity ......................................... 118
Conclusion ....................................................................................................................... 123
APPENDIX ...................................................................................................................... 128
LITERATURE CITED .................................................................................................... 132
CHAPTER 3 INTEGRATING DIVERSE SCIENTIFIC EXPERTISE AND
PRACTITIONER KNOWLEDGE IN PROBLEM FORMULATION FOR
ENVIRONMENTAL RISK ASSESSMENT: A CASE STUDY OF
INTRODUCTION OF GE CROPS IN UGANDA .................................................. 138
Introduction ...................................................................................................................... 138
Materials and Methods ..................................................................................................... 143
Designing the framework and defining the boundaries and scope of the ERA ........... 143
Identifying the participants .......................................................................................... 143
Identifying the potential harm to biodiversity ............................................................. 143
Results and Discussion .................................................................................................... 156
Defining the boundaries and scope of the analysis ...................................................... 156
Identifying the participants .......................................................................................... 157
Identifying potential harm to biodiversity ................................................................... 159
Weevil resistant sweetpotato........................................................................................ 160
Herbicide tolerant cotton.............................................................................................. 166
Drought tolerant maize ................................................................................................ 168
Conclusion ....................................................................................................................... 168
LITERATURE CITED .................................................................................................... 172
CONCLUSIONS AND FUTURE DIRECTIONS: PROBLEM FORMULATION FOR
THE ENVIRONMENTAL RISK ASSESSMENT OF GE WEEVIL RESISTANT
SWEETPOTATO .................................................................................................... 179
Introduction ...................................................................................................................... 179
Development of resistance to Bt toxins by sweetpotato weevils ..................................... 180
Loss of sweetpotato varietal diversity.............................................................................. 182
Risk hypothesis 1 ......................................................................................................... 184
Risk hypothesis 2 ......................................................................................................... 184
Risk hypothesis 3 ......................................................................................................... 188
Changes in crop management systems ............................................................................ 189
Risk hypothesis 1 ......................................................................................................... 184
Conclusion ....................................................................................................................... 192
LITERATURE CITED .................................................................................................... 195
viii

LIST OF TABLES

Table 1-1 Crop Area and Production by region for selected food crops, 2008. .................. 5
Table 1-2 A general scheme for step-by-step consideration of the risk hypotheses used to
evaluation likelihood of a given harm........................................................................ 20
Table 2-1 Characteristics of SSR markers used to evaluate East African sweetpotato
accessions: name, labeled dye, motifs, annealing temperature and reference ........... 38
Table 2-2 Observed base pair range, number of alleles and polymorphic information
content for the SSR markers used to characterize sweetpotato genotypes from
Uganda, Kenya and Tanzania. ................................................................................... 43
Table 2-3 AMOVA for genetic variability within and among the different weevil response
categories of sweetpotato genotypes obtained from Uganda. .................................... 44
Table 2-4 Number of individuals, number of rare alleles, and genetic diversity
parameters1 for the different weevil response categories of sweetpotato genotypes
obtained from Uganda................................................................................................ 44
Table 2-5 AMOVA for genetic variability within and among the different varietal
categories . ................................................................................................................. 46
Table 2-6 Number of individuals, number of rare alleles, and genetic diversity
parameters1 for the different varietal categories of sweetpotato genotypes obtained
from Uganda. ............................................................................................................. 46
Table 2-7 AMOVA for genetic variability within and among the populations of
sweetpotato from different Uganda’s agro-ecological zones and other East African
countries as well as Uganda’s improved cultivars. .................................................... 49
Table 2-8 Number of individuals, number of unique alleles, and genetic diversity
parameters1 for sweetpotato populations from various Uganda’s agro-ecological
zones and other East African countries, as well as Uganda’s improved cultivars. .... 49
Table 2-9 A pair-wise FST for populations from various Uganda’s agro-ecological zones
and other East African countries, as well as Uganda’s improved cultivars ............... 51
Table 2-10 AMOVA for genetic variability within and among sweetpotato populations
from Uganda, other African countries, and American countries. .............................. 53
Table 2-11 Number of individuals, number of rare alleles, and genetic diversity
parameters1 for populations from Uganda, other African countries and American
countries (Brazil, Peru, USA). ................................................................................... 53
Table 2-12 Pair-wise FST1 for populations from Uganda, other African countries (Ghana,
Kenya, Tanzania), and American countries (Brazil, Peru, USA)2. ........................... 55

ix

Table 2-13 Number of individual obtained from Uganda, other African countries and
American countries and the proportion of members of region that is found in
clusters K1 and K2.1 .................................................................................................. 59
Table A-1 Description of the sweetpotato accessions studied ........................................... 71
Table 3-1 Crop Area and Production by region for selected food crops 2008 ................. 82
Table 3-2 Detailed descriptions of the agro-ecological zone: altitude, weather patterns,
soil characteristics, and farming systems. .................................................................. 88
Table 3-3 Number of respondents, and percentage of respondents based on gender and
membership in farmers’ groups, in each district. ....................................................... 93
Table 3-4 Number of respondents cultivating sweetpotato on the different land sizes in
each district ................................................................................................................ 96
Table 3-5 Land ownership (percentage of rented of respondents in each district) ............ 96
Table 3-6 Level of sweetpotato diversity at district level: number of farmers’ varieties and
released cultivars at district level, total number of varieties and cultivars, and the
average number and range of varieties grown per household in each district. .......... 98
Table 3-7 Varieties discarded or lost in the various regions and the reasons for discarding
or for the loss. .......................................................................................................... 101
Table 3-8 Number of farmers practicing varietal exchanges in the different regions. .... 103
Table 3-9 Measure of association between different variables that influence crop
diversity.................................................................................................................... 107
Table 3-10 Number of farmers who adopted new cultivars/varieties and the frequency of
adoption.................................................................................................................... 110
Table 3-11 Number of subsistence and commercial farmers, and the percentage of total
respondents that are conducting the different crop management practices ............ 119
Table B-1 Characteristics of sweetpotato varieties/cultivars grown in the central region of
Uganda ..................................................................................................................... 129
Table B-2 Characteristics of sweetpotato varieties/cultivars grown in the northern region
of Uganda ................................................................................................................. 131
Table B-3 Characteristics of sweetpotato varieties/cultivars grown in the eastern region
of Uganda ................................................................................................................. 132
Table 4-1 Summary of the profiles of the participants engaged in the ERA including their
professional affiliation, educational background and skills acquired through their
work experience and their regulatory roles .............................................................. 145
Table 4-2A Characteristics of the crop and the engineered trait, which are relevant for an
environmental risk assessment for weevil resistant sweetpotato. ............................ 148
x

Table 4-2B Characteristics of the crop and the engineered trait, which are relevant for an
environmental risk assessment for herbicide tolerant cotton. .................................. 148
Table 4-2C Characteristics of the crop and the engineered trait, which are relevant for an
environmental risk assessment for drought tolerant maize. ..................................... 148
Table 4-3 Potential adverse effects associated with interaction of GE crops with
biodiversity, and the number of respondents who listed the effects. ...................... 154
Table 4-4 Section of the questionnaire that was used to identify the environmental
concerns that the participants thought were likely warrant great consideration prior to
cultivation of weevil resistant sweetpotato in Uganda. ........................................... 155
Table 4-5 Participant – identified potential risks that should receive greatest attention in
performing environmental risk assessments for the three crop-trait scenarios in
Uganda. .................................................................................................................... 170
Table 5-1 Exposure scenarios, risk hypotheses and action plan for assessing the harm
through reduced abundance or displacement of landrace varieties resulting from
increased planting of GE weevil resistant sweetpotato. ........................................... 183
Table 5-2 Exposure scenarios, risk hypotheses and actions plan for assessing the harm
caused by changes in rates of pesticide application in GE weevil resistant
sweetpotato plots. ..................................................................................................... 191

xi

LIST OF FIGURES

Figure 1-1 Damage caused by sweetpotato weevil on the storage roots. ............................ 7
Figure 2-1 Map of Uganda showing the different agro-ecological zones in Uganda.
Samples used in this study were originally collected from five agro-ecological zones
.................................................................................................................................... 36
Figure 2-2 A) Cluster analysis of 171 sweetpotato genotypes generated by unweighted
neighbor joining algorithm.. ...................................................................................... 47
Figure 2-2 B) Non-metric multidimensional scaling representation of the principal
coordinates analysis (PCoA) by regions. ................................................................... 47
Figure 2-3 Non-metric multidimensional scaling representation of the principal
coordinates’ analysis (PCoA) comparing genotypes from different Uganda’s agroecological zones and other East African countries .................................................... 52
Figure 2-4 Non-metri multidimensional scaling representation of the PCoA by regions.
Others stand for other African countries.. .................................................................. 56
Figure 2-5 Frequencies of pair-wise similarity coefficients for 260 sweetpotato accessions
from Uganda, other African countries, and American countries ............................... 57
Figure 2-6 A dendrogram of the unweighted pair group method analysis (UPGMA)
cluster analysis on the basis of Jaccard’s simple sequence repeat (SSR) showing
genetic similarities among 102 genotypes. ................................................................ 58
Figure 2-7 Determining the best K value using Structure Harvester. K=2 had the highest
ad hoc Quantity Delta K thus selected as the most appropriate population structure.
.................................................................................................................................... 60
Figure 2-8 Probabilities of membership for each individual in clusters K1 and K2 as
determined by the Bayesian clustering method. ........................................................ 60
Figure 3-1 Map of Uganda showing: A) the different agro-ecological zones (Roman
numerals). The study areas are marked with a black circle1. B) The districts visited
in the study. ................................................................................................................ 89
Figure 3-2 Uses of sweetpotato: A) provided by the key respondents; B) deduced from
data collected during the Uganda Agricultural Consensus (2010). ........................... 94
Figure 3-3 Key challenges faced by farmers engaged in material exchange practices. .. 104
Figure 3-4 Criteria for selection of A) new varieties/cultivars; and B) maintaining current
varieties. ................................................................................................................... 108
Figure 3-5 Key challenges associated with adoption of new diversity. ........................... 112

xii

Figure 3-6 Key challenges associated with maintaining current varieties when adopting
new varieties/cultivars. ............................................................................................ 117
Figure 3-7 Major challenges affecting crop management practices necessary to realize
optimal sweetpotato yields. ...................................................................................... 122
Figure 4-1 A risk assessment framework adapted from those of Wolt et al., (2010) and
Dana et al., (2012) for the ERA for introducing GE crops in Uganda. ................... 144
Figure 4-2 An interim biosafety regulatory approval process for introduction of
genetically engineered organisms in Uganda. ......................................................... 158
Figure 4-3 Scoring of concerns associated with introduction of weevil resistant
sweetpotato in Uganda by the different regulatory categories. ................................ 161
Figure 4-4 Scoring of concerns associated by introduction of herbicide tolerant cotton in
Uganda by the different regulatory categories. ........................................................ 162
Figure 4-5 Scoring of concerns associated by introduction of drought tolerant maize in
Uganda by the different regulatory categories. ........................................................ 163

xiii

LITERATURE REVIEW

Introduction
The work described in this thesis aims to provide baseline information that may be
included as a component of the biosafety package for open release of genetically
engineered (GE) sweetpotato in Uganda, and will provide relevant information for other
East African countries like Kenya and Tanzania. One of the potential environmental
concerns associated with introduction of GE crops is loss of biodiversity. This study uses a
common conceptual framework for environmental risk assessment (ERA) to determine the
likelihood of biodiversity loss being a risk resulting from an introduction of GE weevil
resistant sweetpotato.

Specifically, I use existing literature to carry out problem

formulation for ERA for cultivation of GE weevil resistant sweetpotato in Uganda. As part
of the problem formulation phase, consultations with local scientific experts, and
practitioners involved in the regulatory process were used to identify environmental
concerns associated with GE weevil resistant sweetpotato. Using molecular markers, I
estimated the level of genetic diversity in sweetpotato in Uganda and how it relates to
diversity from elsewhere. To gain a better understanding of how this diversity is maintained
in farmers’ fields, I used an ethno-botanical survey to analyze 1) considerations by farmers
when making a decision to maintain, incorporate and/or discard varieties in their fields; 2)
their conservation practices when adopting new cultivars, and 3) how these decisions and
practices have influenced the level of sweetpotato diversity within selected regions of
Uganda. Using the information obtained from the above studies, I evaluated the likelihood
of the identified risks. The work is reported in three chapters of this thesis. This literature
review chapter consists of two main parts: Part 1 describes sweetpotato including its origin,
importance, production constraints, and crop improvement research efforts such as
1

development of genetically engineered (GE) varieties. Part 2 discusses the concept of
environmental risk assessment and its role during introduction of GE crops into the
environment.
Part 1. Sweetpotato
Taxonomy and biology of sweetpotato
Sweetpotato (Ipomoea batatas L. Lam) belongs to series Batatas, genus Ipomoea
and family Convolvulaceae (Morning glory) (Mwanga 2001a; Loebenstein and
Thottappilly, 2009). Cultivated sweetpotato is a hexaploid (2n=6x=90), however wild
tetraploid (2n=4x) forms were reported in Ecuador (Bohac et al., 1993). There have been
several theories regarding the genetic origin of hexaploid sweetpotato (Huamán, 1999).
Some theories suggest that its origin is an allopolyploid involving I. leucantha, I. triloba
and I. trifida (Austin, 1987; Nishiyama, 1971) while others suggest an autopolyploid
origin involving only I. trifida (Kobayashi, 1983; Shiotani and Kawase, 1987). However,
the most recent literature supports the allopolyploid origin theory (Srisuwan et al, 2006;
Gao et al., 2011).
Sweetpotato is a perennial dicotyledonous herb (Austin and Huamán, 1996) that is treated
as an annual crop under cultivation (Lebot, 2010). Sweetpotato is vegetatively propagated
and is mainly cultivated in the tropics. However, it can tolerate a wide range of edaphic
and climatic conditions, including cold and high altitudes (as high as 2,500 m) (Lebot,
2010). Its growth habit is characterized by an aerial vine system that spreads rapidly
horizontally on the ground, and an underground adventitious root system that forms
storage roots. Most sweetpotato genotypes flower naturally in short-day in the tropics
(Mwanga, 2001a; Lebot, 2010). Sweetpotato is an out-crossing species because it is selfincompatibile. The main pollinators of sweetpotato are insects, particularly honey bees
(Lebot, 2010). Not much is known about the survival of sweetpotato pollen, except that
2

pollen vaibility may continue for 3 to 4 hours after pollination (Martin and Cabanillas,
1966; Murata and Matsuda, 2003). Sweetpotato is only sexually compatible with relatives
in the series Batatas, but successful fertilization is hindered by sporophytic self-and
cross-incompatibility (Jones 1967; Mwanga et al., 2007), on average each pollinated
flower produces only two to three seeds, often less (Lebot, 2010). Sweetpotato seed has a
very hard seed coat and thus has a dormancy tendency: it germinates very slowly and
irregularly. The fruit is a dehiscent capsule. Sweetpotato is not listed on the US Federal
noxious weed list (USDA-APHIS, 2010). The crop has been grown in various parts of the
world but there is no report that it is a weed. Volunteers, mostly from vegetative
propagules are often observed in the following growing season but these are usually
controlled using manual and chemical measures.

Origin and importance of sweetpotato
The center of origin for sweetpotato is believed to be in the Meso- and South
American lowlands and its cultivation started around 2000-2500 B.C (Zhang et al., 2004;
Loebenstein and Thottappilly, 2009). It was introduced to East Africa by Portuguese
explorers around the 16th century. East Africa is considered a secondary center of
diversity for sweetpotato because of the presence of high level of crop diversity in that
region (Huamán and Zhang, 1997; Mwanga et al., 2001b; Gichuki et al., 2003).
Sweetpotato is grown in over 100 countries on about 7.9 million hectares, yielding
~104 million tons (FAO, 2011). China is the largest global producer of sweetpotato
accounting for about 44% of the global area, and 72% of the global output. Uganda is the
third largest producer of sweetpotato in Africa (contributing 6.6% and 2.4% of the total
global area and output, respectively), after Tanzania (8.7% and 3.4%) and Nigeria (11.8%
and 2.6%). It is the fourth most important food crop in terms of fresh weight in Eastern

3

Africa (FAO, 2011), and in Uganda (Table 1.1). It is grown for its enlarged starchy roots
and immature leaves, which are for human consumption and animal feed. Uganda is
among the top five highest global consumers with per capita consumption of about
88kg/person/annum (FAO, 2003). It produces more edible energy (194 MJ ha-1day-1) than
any other major food crop (Woolfe, 1992), and with the on-going promotion to consume
more orange-fleshed sweetpotato, it has become a major source of provitamin A
especially in rural areas (Low et al., 2001). It is an important crop for small-scale farmers
with limited land, labor and capital (Bashaasha et al., 1995). It is also a source of income
for some households. About 44% of Ugandan farmers are involved in sweetpotato
production (MAAIF, 2011).
The importance of sweetpotato as a food security crop is increasing rapidly in Uganda
because of the virus epidemic in cassava, and the rapid spread of a parasitic weed (Striga
hermonthica) that attacks maize and other cereal crops (Abidin, 2004; MAAIF, 2011).
The crop plays an important role in providing household food security because it stores
well in the soil as a famine reserve, and it can withstand extreme weather conditions and
marginal soils (Bashaasha et al., 1995). One of its greatest assets is its ability to be
harvested piecemeal as needed for home consumption or income generation. Price
fluctuation for cash crops, population increase, civil wars and changes in weather patterns
are also influencing sweetpotato production in different regions (Abidin, 2004). For
example, many farmers in Kumi district are growing sweetpotato as a cash crop because
of the decline in market prices for cotton. In most regions, the crop is mostly grown by
resource-poor small-scale farmers on plots less than one acre (Bashaasha et al., 1995). Its
production is mostly managed by women farmers who are able to distinguish the varieties
(Hakiza et al., 2000).

4

1

Table 1-1 Major crops grown in Uganda, their production area and total output in 2008 .
Crop
Maize
Banana
Cassava
Sweetpotato
Beans
Sorghum
Ground nuts
Finger millet
Rice
Irish potato
Soya bean

1

Area (Ha)
1,014,260
915,877
871,389
440,256
509,523
399,252
345,232
249,987
75,086
32,759
36,444

Production (MT)
2,361,956
4,297,345
2,894,311
1,818,769
929,276
375,795
244,683
276,928
190,736
154,354
59,393

Uganda Bureau of Statistics (UBOS), 2010. Uganda Census for Agriculture, 2008/09.
Volume IV: Crop Area and Production Report.

5

Production constraints
Uganda’s average yield of 4.8 ton/ha is very low compared to the global and
China’s average yield of 13 and 23 tons/ha, respectively (FAO, 2011). The major
constraints to sweetpotato production in Uganda include sweetpotato weevils (Cylas
puncticollis and C. brunneus), Alternaria stem blight, and viral diseases, particularly
sweetpotato virus disease (SPVD), which is caused by combined infection of sweetpotato
feathery mottle virus (SPFMV) and sweetpotato chlorotic stunt virus (SPCSV)
(Loebenstein and Thottappilly, 2009). Other constraints include limited access to high
quality planting materials, high usage of low yielding and stress-susceptible varieties, as
well as limited processing and marketing avenues.
Sweetpotato weevils may account for more than 60% of yield loss, and this can
reach 100% depending on the climatic conditions and susceptibility of the variety (Zhang
et al., 1996; CIP, 2008). The primary damage involves tunneling through the storage root
both pre-harvest and post-harvest (Figure 1.1) rendering them unfit for human
consumption (Smit, 1997). Since sweetpotato is mostly grown by resource-poor small-scale
farmers, chemical control is rarely an option.

Biological control agents such as entomo-pathogenic fungi have been reported to be
effective in an integrated pest management sweetpotato weevil program in Cuba (Stathers
et al., 1999); however, they were not effective in Uganda (Smit 1997; Stathers et al.,
1999). Sex pheromonea of Cylas formicarius were found to be effective in checking
weevil populations in many parts of Asia (Hwang and Hung, 1991; Pillai et al., 1993), but
sex pheromones of C. puncticollis and C. brunneus were ineffective in Uganda
(Downham et al., 1998). Thus, the most feasible option for addressing this production
constraint is through host resistance. This can be achieved through identification and

6

Figure 1-1 Damage caused by sweetpotato weevil the storage roots. Source: CIP. Credit
to N. Smit.
“For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation.”

7

promotion of resistant varieties in the existing gene pool, and through breeding of
improved cultivars.

Genetic Diversity of the Sweetpotato
Understanding of the level of crop genetic variation is crucial for selection of
genotypes with desirable traits, for establishment of core collections for diversity
conservation and designing optimal gene pools for plant breeding programs (Zhang et al.,
2000; McCouch et al., 2012). Maintaining genetic diversity is essential for sustainable
improvement of agricultural productivity given the changing weather patterns, pests and
diseases, and consumer preferences. There are more than 8000 accessions of sweetpotato
maintained at 83 gene banks in the world (Elameen et al. 2008). In Uganda, a national
collection at the National Crops Resources Research Institute (NaCRRI), Namulonge,
holds 1303 accessions collected from different agro-ecological zones of Uganda (Yada et
al., 2010a). All the Ugandan accessions were characterized using morphological and
agronomic traits and only 946 were identified as distinct accessions (Yada et al., 2010a).
Molecular characterization using microsatellites (simple sequence repeats, SSR markers)
has been used to assess the genetic relationship between accessions showing superiority
for certain traits like resistance to common diseases, high dry matter content and high
yields (Yada et al., 2010b). SSR markers were also used to determine the genetic
relationship between the orange fleshed varieties and the white or cream-fleshed
accessions within the collection (Tumwegamire et al., 2011). None of these studies
focused on understanding the level of genetic variability within the different agroecological zones and their genetic relatedness.
Besides understanding the level of genetic variability within Uganda, it is also
important to know the genetic relationship between Ugandan sweetpotato germplasm and
8

germplasm from elsewhere. This can help to determine whether the Ugandan genotypes
carry unique alleles that may be of importance to present or future breeding programs,
and whether replacement of the traditional Ugandan varieties can result in loss of allelic
diversity. One study has compared East African varieties (including four Ugandan
accessions) with genotypes from different centers of diversity (South America,
Caribbean, USA, Asia, Oceania) using RAPD markers (Gichuki et al., 2003). Results of
this study indicated that genetic variability between these broad geographic regions was
very low (6.6% contribution to total variance) as compared to within regions (93.4%
contribution to total variance). These results suggest that over time genotypes have
evolved within regions to enhance the adaptability of the populations to their habitats.
This study also reported unique characteristics in some East African sweetpotato varieties
including higher storage root dry matter content with low ß-carotene content, later
maturity and high virus resistance. However, because all genotypes from East African
countries were grouped together, it is not clear whether the unique alleles for these traits
were observed in Ugandan accessions. A study involving more Ugandan accessions
would provide a better understanding of the genetic relationship between Uganda’s
sweetpotato germplasm and germplasm from elsewhere. This is necessary for effective
conservation and sustainable utilization of the germplasm for the crop improvement
through breeding.

Sweetpotato breeding
Sweetpotato breeding started in the US in the 1920s, and involved transfer of
desirable traits from varieties introduced from the tropics and favorable mutations (Yen
1974). A major break-through in sweetpotato breeding came with development of
sweetpotato types that flowered under long-day conditions (Miller 1939; Wang 1964). In
9

1965, Jones proposed a breeding procedure for sweetpotato using polycross and recurrent
selection to enhance the effectiveness of combining desired genes and alleles in parental
genotypes. This procedure has been used successfully in sweetpotato to improve various
desirable traits, and is particularly useful for traits with low heritability (Jones, 1986;
Grüneberg et al., 2009; Jackson et al., 2010).
In Uganda, sweetpotato breeding started in 1982 with technical assistance from
the International Institute for Tropical Agriculture [IITA] (Hakiza et al., 2000). Since
then, significant progress has been made including evaluation of traditional varieties and
introduced germplasm, generation and screening of breeding populations, and multilocational testing of varieties and lines to identify superior genotypes for the different
agro-ecological zones. Selection of superior traditional varieties varies depending on the
target agro-ecological zone including for northern and eastern regions where weevils and
drought are most critical; the central and western region where viruses are most severe;
and the southwestern highlands where Alternalia blight and low soil fertility are the major
constraints (Hakiza et al., 2000). In 1995, the national programme released six superior
varieties (Tanzania, Tororo3, New Kawogo, Bwanjule, Wagabolige and Sowola)
(Mwanga et al., 2001b). Only Sowola was an improved cultivar selected from bulked
seed from a polycross of 18 parents. The rest were superior farmers’ varieties selected
after evaluation of 380 landraces. Release of six improved cultivars (NASPOT 1-6)
followed in 1999 (Mwanga et al., 2003), then five (NASPOT 7-10 and Dimbuka
Bukulula) in 2007 (Mwanga et al., 2009) and one (NASPOT 11) in 2010 (Mwanga et al.,
2011). Dimbuka Bukulula is also a superior farmers’ variety. The selection criteria for the
improved cultivars includes traits such as high root yield of more than 21tha-1, moderate
to high resistance to sweetpotato virus disease (SPVD) and Alternaria bataticola blight ,
high dry matter content (more than 29%) and high consumer acceptance, depending on
10

the growth conditions (Mwanga et al., 2011). However, breeders are faced with a
challenge when it comes to improving weevil resistance through conventional breeding
(Grüneberg et al., 2009). This is because of limited genetic variation for this trait in
sexually compatible relatives. For this trait, genetic engineering has been adopted as a
breeding tool to incorporate new variation into the gene pool (Grüneberg et al., 2009;
Loebenstein and Thottappilly, 2009).

Genetic engineering
Genetic engineering (GE, also referred to as genetic modification) is an advanced
method of biotechnology used to improve organisms for human benefits. A number of GE
crops with desirable traits such as herbicide-tolerance, insect pest-resistance, and virus
resistance are now cultivated in different parts of the world (James, 2012). Cultivation of
GE crops has grown from 1.7 million hectares in 1996 to 170 million hectares in 2012.
The GE crops grown are maize, cotton, soyabean, canola, sugarbeet, alfalfa, papaya,
squash, tomato and poplar. However, research is underway involving more crop species
and traits (Grumet et al., 2011).
GE also holds great promise for sweetpotato crop improvement because it allows
the introduction of new desired genes that may not be known in sweetpotato or in its close
relatives, and enables manipulation of traits within sweetpotato that are difficult to handle
using conventional breeding. Conventional breeding in sweetpotato is limited by the fact
that it is vegetatively propagated, highly polymorphic, and exhibits self- and crossincompatibility (Loebenstein and Thottappilly, 2009). Efficient and reliable plant
transformation protocols for sweetpotato have been developed (Song et al. 2004; Lim et
al., 2007; Okada and Saito, 2009). However, some protocols are highly genotype

11

dependent; this together with the low rate of regeneration of sweetpotato from cultured
tissue or cells has resulted in limited application of GE in sweetpotato breeding.
Sweetpotato transformation has been carried out using Agrobacterium rhizogenes
and organogenic plant regeneration from hairy roots (Dodds et al., 1991); A. tumefaciens,
particle bombardment, and protoplast electroporation coupled to either somatic
embryogenesis (Cipriani et al., 1999, 2001) or organogenesis-based (Luo et al., 2006)
plant regeneration. So far genetic engineering is being used to develop sweetpotato with
resistance to weevils or viral diseases, and alteresd starch composition (Loebenstein and
Thottappilly, 2009).

Developing weevil resistant GE sweetpotato
Susceptibility to sweetpotato weevil was observed to vary from mildly susceptible
to highly susceptible genotypes (Stathers et al., 1999; Thompson et al., 1999). However,
there has been slow progress in conventional breeding of weevil resistant varieties
because of limited native resistance and heritability of the trait is extremely low.
Therefore, weevil resistance was considered a suitable trait for application of genetic
engineering in sweetpotato. The first weevil-resistant transgenic sweetpotato was
transformed with Vigna unguiculata trypsin inhibitor from cowpea and a mannosebinding snowdrop lectin (Newell et al., 1995). Subsequent transformations have involved
introduction of a soybean (Glycine max) Kunitz-type trypsin inhibitor and rice (Oryza
sativa) cystein proteinase inhibitor (Cipriani et al., 2001). However, a very low level of
weevil resistance was observed from field testing of these products (Loebenstein and
Thottappilly, 2009). Recent studies have involved introduction of Bacillus thuringiensis
(Bt) toxins against Cylas puncticollis, C. brunneus, and C. formicarius. In Africa,
research has involved introduction of Bacillus thuringiensis (Bt) toxins (Mwanga et al.,
12

2011b). Bt proteins (Cry7Aa1, ET33/34, and Cry3Ca1) with high toxicity levels to the
two African weevils (C. puncticollis, C. Brunneus) (Ekobu et al., 2010) have been used
for Agrobacterium-mediated transformation of two African varieties in Kenya and
Uganda (Kreuze et al., 2009; Mwanga et al., 2011b).
Bacillus thuringiensis (Bt) is a naturally occurring soil bacterium that produces
crystals consisting of proteins that exhibit highly specific insecticidal activity (Hofte and
Whiteley, 1989). The insecticidal activity of Bt toxins is switched on when they are
ingested by the insect, solubilized in the insect gut, and are activated by specific midgut
enzymes. Active toxins bind to specific receptors on the surface of the insect midgut, and
cause cell lysis, which results in midgut damage and insect death (Bravo et al., 2007).
This mode of action makes Bt toxins specific in activity against a limited range of
susceptible insect species.
In 1995, The US EPA registered the first Bt plant-incorporated protectants for use
in the United States (US EPA, 2001). Currently there are over 49 Bt events in cotton,
maize, potato and tomato that have been released for food/feed and marketing (Annex 1)
(http://cera-gmc.org; http://www.isaaa.org/gmapprovaldatabase/gmtrait/). Bt maize and
cotton are examples of GE crops that are under cultivation, and whose acreage continues
to grow annually (James, 2012). However, use of Bt genes for development of insect
resistance in crops has been associated with potential environmental concerns including
development of resistance by the pest to the Bt toxins, horizontal transfer of transgenes to
other organisms, non-target effects, and transgene escape to sexually compatible relatives
that may result in loss of biodiversity (Lu, 2008; Pimentel and Paoletti, 2009; Webadde
and Maredia, 2011).
Widespread cultivation of Bt crops over long durations has been reported to result
in development of resistance in the targeted pests (Tabashnik and Carriere 2009;
13

Webadde and Maredia, 2011; Gray, 2012). For example, resistance of bollworm
(Helicoverpa zea) to Cry1Ac cotton was observed in USA in 2003; resistance of stem
borer (Busseola fusca) to Cry1Ab corn was reported in South Africa in 2006, and
resistance of fall armyworm (Spodoptera frugiperda) to Cry1F corn was recorded in
Puerto Rico in 2006 (Tabashnik and Carriere 2009). Development of Bt resistance by the
target pest is a very likely risk, therefore mechanisms to delay evolution of this resistance
are recommended to farmers growing Bt crops such as use of refugia.
Horizontal gene flow refers to transfer of the transgene to other organisms by
processes that are independent of reproduction. Keese (2008) provides a comprehensive
review of this concern. Non-target effects of GE crops have been discussed in many
studies (Conner et al., 2003; Craig et al., 2008; Romeis et al., 2008; 2011). Non-target
effects involve risks associated with interaction of the GE crop with various animals,
particularly invertebrates, in the receiving environments. In the case of Bt crops, such
non-target organisms maybe directly affected by the Bt toxin, or indirectly by change in
the plant or pest quality and behavior (Gray, 2012). Studies for non-target effects of
weevil resistant GE sweetpotato are currently underway (Runyararo Rukarwa, personal
communication).
Biodiversity loss is very broad concern as it involves assessing impacts of the GE
crop on the genetic variability of all flora and fauna (Convention of Biological Diversity,
1992). Most discussions linking GE crops to biodiversity refer to terms like species
reduction or loss of genetic diversity (Gray et al., 2003). This includes considerations of
loss of interspecific (within wild/weedy relatives) and intraspecific (within crop)
diversity. Hybridizations between GE crops and its wild/weedy relatives may cause
reduction in genetic variability if the trangene confers a fitness or selective advantage that
could result in replacement the natives (Ellstrand et al., 1999). Many reviews have raised
14

concerns regarding introduction of GE crops in centers of diversity for specific crops
because they could result in losses of genetic variability (Messeguer et al., 2001; Quist
and Chapela, 2001; Gepts and Papa, 2003; Rong et al., 2007; Lu, 2008) due to
hybridization between GE crops and traditional varieties especially within the centers of
diversity, and extensive cultivation of the GE crop that could result in displacement of
traditional varieties.
These potential environmental concerns have been major obstacles to widespread
acceptance and cultivation of GM crops (Gray, 2012). Science-based environmental risk
assessment on a case-by-case basis remains a crucial step in evaluation of these concerns.

Part 2. Environmental risk assessment
Despite the history of safe use of crops currently on the market, every new GE
crop has to undergo some form of environmental risk assessment (ERA) prior to
unrestricted introduction into the environment. The purpose of an ERA of a GE crop is to
determine the potential adverse effects on human and animal health and the environment,
relative to a non-GE plant comparator (EFSA, 2010). The principle of comparative safety
assessment (Kok and Kuiper, 2003) helps to identify intended and unintended effects of
the GE trait(s) by using the ‘concept of familiarity’, which was developed by the OECD
(1993). The assumption is that the biology of the crop and the impacts of the non-GE
counter-part on the environment are known and accepted. Plants developed by
conventional breeding are a suitable comparator (Conner et al., 2003). Selection of the
most suited comparator may be based on the European Food Safety Authority assessment
guidelines (EFSA, 2011). The ERA involves using available scientific information, and
where necessary applying common methodologies to identify and gauge how serious the

15

potential risks are. To ensure transparency, assumptions made during the ERA are clearly
defined as well as the nature and level of uncertainties and variability.
Assessment methodologies include molecular characterization, agronomic
performance and phenotypic evaluation, and compositional characterization (OECD,
1993; FAO/WHO, 1996). Most of this information can be accessed from the developer
because it is part of good product development and testing practices. What remains is to
establish whether the difference between the GE plant and its non-GE counterparthas a
potential to cause harm. Crop biology documents produced by various international
bodies and regulatory agencies [such as Organisation for Economic Cooperation and
Development (OECD), Office of Gene Technology Regulation, Australia (OGTR) or
Uganda National Council for Science and Technology (UNCST)] describe the expected
range of variation for different crops. Currently, there is no crop biology document for
sweetpotato produced by regulatory agencies. In addition to the methodologies indicated
above, determining the expected interaction of the GE crop with the receiving
environment is crucial (ESFA, 2010). There are different ERA frameworks developed by
different countries, however, they all share a common conceptual framework comprising
of: i) problem formulation; ii) risk characterization (exposure and consequence
assessment); and iii) risk evaluation (Wolt et al., 2010).
The first step of problem formulation involves assessing the problem using
available information about the plant and the new/modified trait, to identify hazards,
define assessment endpoints, and determine risk hypotheses to guide the next risk
assessment steps (Nickson, 2008; ESFA, 2010). The hazard characterization step involves
qualitative and/or quantitative evaluation of the magnitude of environmental harm
associated with the hazards described in the risk hypotheses. Exposure characterization

16

involves evaluation of the possible routes by which exposure to the hazard occurs and
estimating the likelihood that the exposure will occur. Based on the information gathered,
qualitative and/or quantitative evaluation of the magnitude of environmental harm
associated with the hazards described in the risk hypotheses. Exposure characterization
involves evaluation of the possible routes by which exposure to the hazard occurs and
estimating the likelihood that the exposure will occur. Based on the information gathered,
the risk is characterized by combining the magnitude of the consequences of a hazard and
the likelihood of exposure occurring, as well as the level of uncertainties. The final step
of the risk assessment is risk evaluation where the acceptability or unacceptability of the
risk is determined. If a risk is likely to occur, and is determined to be unacceptable, then
the next step is to identify effective risk management strategies and determining the
capacity to implement them. Risk assessment continues beyond open release of the GE
product through post-release monitoring and evaluation.

Problem formulation
Problem formulation involves assessing what valuable aspects of the environment
are most at risk of harm (Wolt et al., 2010; Gray, 2012). Over the years, many studies
have been conducted that can provide information about crop biology, agronomic
performance and impacts on the environment, as well as the effects of introducing or
modifying particular trait (s) for various crops (Raybould and Cooper, 2005; Lu, 2008;
Romeis et al., 2008; Hokanson et al., 2010; Raybould, 2011; Gray 2012; Raybould et al.,
2012). This information can serve as a basis for identifying the most important questions
that merit detailed risk characterization for new GM crop varieties (EFSA, 2010).

17

Gray (2012) suggests that during problem formulation the following questions
need to be addressed: (1) what are the ecological entities that need to be protected? (2)
Are these entities likely to be harmed by introduction of the GE crop? (3) How do we
determine the likelihood and level of harm? (4) If the harm happens, does it matter? The
first and last questions are policy and societal concerns, and therefore may be determined
through stakeholders’ consultations. The second and third questions require evidencebased, scientific analysis. The first step of problem formulation is to frame the problem
context, where the broad concerns based on international and national legislations and
policies, plus societal concerns are used to define more specifically the valued
environmental entities and their assessment endpoints. The assessment endpoints define
the valued entities in terms of attributes that can be evaluated to determine occurrence of
harm (Gray, 2012). The search for assessment endpoints will be guided on a case-by-case
basis by the crop biology, the modified trait and the conditions under which the crop will
be cultivated.
The next step is problem definition where the concerns are critically evaluated,
using available information about the characteristics of the plant, the trait and the
receiving environment plus the likely interactions among these factors, to identify those
that pose a risk and demand more attention. This can be achieved by clearly articulating
the interaction between the cultivation of the GE crop and the valued entities. This
activity can greatly benefit from ‘brainstorming’ sessions involving scientists with
biology and agriculture backgrounds because it results in developing relevant and realistic
exposure scenarios (Gray, 2012). It is better to start by generating all the possible
scenarios in which harm can arise, and then narrowing the list to the scenarios that are
worth devoting resources (Gray, 2012). For each possible harm, a conceptual model
tracing the ‘pathways to harm’ is constructed consisting of various exposure scenarios.
The next step is to formulate risk hypotheses of no harm addressing the likelihood of
18

exposure scenario (Raybould et al., 2007; Nickson, 2008; Gray 2012). A general scheme
for step-by-step considerations of the risk hypotheses is provided by Raybould (2010)
[Table 1.2]. Further analysis is only necessary if the risk hypothesis is rejected. For
example if the hypothesis is “insecticidal protein is not expressed in the pollen”, then
further analysis should proceed only if there is evidence to show that the pollen contains
the insecticidal proteins. Where necessary, an analysis plan suggesting the experimental
designs for testing the risk hypotheses, the type of data to be collected, and data synthesis
will be developed (Nickson, 2008; Gray, 2012). A tiered approach for conducting risk
assessment experiments, where conservative assumptions are made in the first tier
experiment so as to test the worst-case scenario, and the level of realism is increased in
making assumptions for every higher tier experiment, is now considered the most
effective method for testing risk hypotheses (Raybould and Cooper, 2005; Nickson,
2008; EFSA, 2010). The final step of problem formulation is selecting the harms on
which the next steps of the risk assessment should focus.

Objectives of dissertation
Prior to release of the GE weevil resistant sweetpotato varieties, an environmental
risk assessment is required to assess the potential impact of introducing transgenic
varieties to the receiving environment. This dissertation provides a baseline study of the
crop diversity existing in the receiving environment, which is an important component of
the biosafety package.
The first objective of this study was to measure the level of genetic diversity in
Uganda’s sweetpotato germplasm. Molecular characterization using SSR markers was
used to determine the genetic variability between Ugandan farmers’ varieties and
improved cultivars, as well as to establish the genetic relationship between Uganda’s
germplasm, and collections from other East African countries, Meso- and South America.

19

Table 1-2 A general scheme for step-by-step consideration of the risk hypotheses used to
evaluation likelihood of a given harm for a GE crop. (Raybould, 2010)
Exposure scenario

Risk hypothesis

Cultivation of the GE crop
Event A will not occur
Event A
Event B will not occur
Event B
Event C will not occur
Event C
Event D will not occur
Event D (Harm)

20

The second objective was to analyze the factors that influence farmers’ decisions
to maintain or discard traditional varieties during the adoption of new and or improved
cultivars. This involved conducting a survey in different agro-ecological zones to find out
farmers’ considerations and challenges during variety selection for adoption and
conservation, as well as their crop management practices and how they impact crop
diversity.
Objective three was to identify environmental concerns associated with the
introduction of genetically engineered weevil resistant sweetpotato, and to determine the
country policy objectives relevant for environmental risk assessment. Consultations with
scientific experts and practitioners were used to find out whether Ugandan regulators
considered crop diversity loss as a likely environmental concern to result from cultivation
of GE weevil resistant sweetpotato.
The final objective was to evaluate the probability that GE weevil resistant
sweetpotato will have the potential harms identified in objective 3. Where necessary,
exposure scenarios were developed, risk hypotheses formulated and analysis plans for
testing the risk hypotheses were designed. Risk management strategies were also
suggested for risks that were determined to be likely significant concerns during the
decision-making process.

21

LITERATURE CITED

22

LITERATURE CITED

Abidin, P.E., 2004. Sweetpotato breeding for northeastern Uganda: Farmer varieties,
farmer-participatory selection, and stability of performance. PhD thesis Wageningen
University, Wageningen, 152 pp.
Austin D.F. and Huamán Z. 1996. A synopsis of Ipomoea (Convolvulaceae) in the
Americas. Taxon 45(1): 3-38.
Austin DF (1987) The taxonomy, evolution and genetic diversity of sweetpotato and
related wild species. In: Exploration, Maintenance, Utilization of Sweetpotato Genetic
Resources. Proceedings of the 1st Planning Conference, Lima, Peru, International Potato
Center (CIP), pp. 27–59.
Bashaasha B, Mwanga ROM, Ocitti p’Obwaya C, Ewell PT (1995). Sweetpotato in the
farming and food system in Uganda: A farm survey report. CIP, sub –Saharan African
region, Nairobi, Kenya /National Agricultural Research organization, Kampala, Uganda.
63 pp.
Bohac, J.R., Duke, P.D. and Austin, D.F. 1995. Sweetpotato. In: Smart, J, Simmonds NW
(eds). Evolution of crop plants. Longman Scientific and Technical, Harlow, UK, pp 5762.
Bravo, A., Gill, S.S. and Soberón, M., 2007. Mode of action of Bacillus thuringiensis Cry
and Cyt toxins and their potential for insect control. Toxicon 49: 423-435.
CIP, 2008. Weevil resistant sweetpotato through biotechnology. Sweetpotato Action for
Security and Health in Africa. International Potato Center.
Cipriani, G., Fuentes, S., Bello, V., Salazar, L.F., Ghislain, M. and Zhang, D.P. (2001)
Transgene expression of rice cysteine proteinase inhibitors for the development of
resistance against sweetpotato feathery mottle virus. In: CIP program report 1999–2000.
International Potato Center, Lima, pp. 267–271
Cipriani, G., Michaud, D., Brunelle, F., Golmirzaie, A. and Zhang, D.P. (1999)
Expression of soybean proteinase inhibitor in sweetpotato. In: CIP program report 1997–
1998. International Potato Center, Lima, pp. 271–277
Conner AJ, Glare TR and Nap J. 2003. The release of genetically modified crops into the
environment. Part II. Overview of ecological risk assessment. Plant J. 33:19-46.
Convention on Biological Diversity (CBD). 1992. United Nations Environment
Programme. 30pp. http://www.cbd.int/
Craig W, Tepfer M, Degrassi G & Ripandelli D 2008. An overview of general features of
risk assessments of genetically modified crops. Euphytica 164: 853-880.
Dodds, J.H., C. Merzdorf, V. Zambrano, C. Sigueñas, and J. Jaynes. 1991. Potential use
of Agrobacterium-mediated transfer to confer insect resistance in sweetpotato, pp. 203–
23

220. In: R. Jansson and K.V. Raman (eds.). Sweetpotato pest management, a global
perspective. Westview Press, Boulder, Colo.
Downham, M.C.A., N.E.J.M. Smit, D.R. Hall, D.I. Farman, B. Odongo, and P.O. Laboke.
1998. Development of pheromones for monitoring and control of sweetpotato weevils,
Cylas breunneus and C. punctollis. DFID, RNRRS Final Technical Report R6115.
EFSA 2010. Guidance on the environmental risk assessment of genetically modified
plants. European Food Safety Authority Panel on Genetically Modified Organisms
(GMO), 100pp.
EFSA, 2011. Guidance document on Selection of Comparators for the Risk Assessment
of GM Plants. European Food Safety Authority, EFSA Journal 9(5):2149 21p.
Ekobu, M., Solera, M., Kyamanywa, S., Mwanga, R.O.M., Odongo, B., Ghislain, M.,
Moar, W.J., 2010. Toxicity of seven Bacillus thuringiensis Cry proteins against Cylas
puncticollis and Cylas brunneus (Coleoptera: Brentidae) using a novel artificial diet.
Journal of Economic Entomology 103: 1493-1502.
Elameen A., Fjellheim S., Larsen A., Rognli A.O., Sundheim L., Msolla S., Masumba E.,
Mtunda K. and Klemsdal S.S., 2008. Analysis of genetic diversity in a sweetpotato
(Ipomoea batatas L.) germplasm collection from Tanzania as revealed by AFLP. Genet
Resour Crop Evol 55:397–408
Ellstrand N. C., H. C. Prentice, and J. F. Hancock, 1999. “Gene flow and introgression
from domesticated plants into their wild relatives.” Annual Review of Ecology and
Systematics, 30: 539–563.
FAO, 2003. The state of food insecurity in the world 2003. Food and Agriculture
Organization, Rome.
FAO, 2011. FAO crop production statistics. Food and Agriculture Organization.
http://faostat.fao.org/ Retrieved March 3 2012.
Gao M, Ashu GM, Lakeisha Stewart, Akwe WA, Victor Njiti, Sandra Barnes (2011). Wx
intron variations support an allohexaploid origin of the sweetpotato [Ipomoea batatas (L.)
Lam]. Euphytica, 177: 111–133.
Gepts P, Papa R. 2003. Possible effects of (trans)gene flow from crops on the genetic
diversity from landraces and wild relatives. Environmental Biosafety Research 2: 89–103.
Gichuki, S.T., Berenyi, M., Zhang, D., Herman, M., Schmidt, J., Glossl, J and Burgh, K.
2003. Genetic diversity in sweetpotato [Ipomoea batatas (L.) Lam.] in relationship to
geographic sources as assessed with RAPD markers. Genet. Resources and Crop Evol.
50: 429-437.
Gray A., 2012. Problem Formulation in Environmental Risk Assessment for Genetically
Modified Crops: A Practitioner’s Approach International Centre for Genetic Engineering
and Biotechnology (ICGEB) Collection of Biosafety Reviews, 6: 10-65
http://www.icgeb.org/biosafety/publications/collections.html

24

Gray AJ, Daniels RE, Raybould AF, Cooper JI, Maskell LC, Pallet D, Edwards M-L,
Thurston M & Alexander M 2003. The conservation of genetic diversity: Gene flow from
agriculture. In: Methods for risk assessment of transgenic plants. IV. Biodiversity and
Biotechnology. Ammann K, Jacot Y and Braun R (eds), Birkhauser Verlag, Basel,
Switzerland. pp 105-110.
Grumet R, Wolfenbarger L, Ferenczi A. 2011. Future possible genetically engineered
crops and traits and potential environmental concerns. In: Environmental safety of
genetically engineered crops. Grumet R, Hancock J, Maredia K, Weebadde C (eds).
Michigan State Univ. Press, East Lansing, pp. 47-57.
Grüneberg W., R. Mwanga, M. Andrade, and H. Dapaah 2009. Unleashing the potential
of sweetpotato in Sub-Saharan Africa: Current challenges and way forward. CIP Social
Sciences Working Paper.
Hakiza, J.J, Turyamureeba, R.M., Kakuhenzire, B., Odongo, R.M., Mwanga, R.,
Kanzikwera, R. and Adipala, E. 2000. Potato and Sweetpotato improvement in Uganda:
A Historical Perspective. African Potato Association Conference Proceedings 5: 47-58.
Hofte H, Whiteley HR. 1989. Insecticidal crystal proteins of Bacillus thuringiensis.
Microbiol Rev: 53: 242–55.
Hokanson KE, Ellstrand NC, Ouedraogo JT, Olweny PA, Schaal BA and Raybould AF
2010. Biofortified sorghum in Africa: using problem formulation to inform risk
assessment. Nature Biotechnology 28(9): 900-903.
Huamán, Z. & D.P. Zhang, 1997. Sweetpotato. In: D. Fuccillo et al. (Eds), Biodiversity in
Trust-Conservation and Use of Plant Genetic Resources in CGIAR Centers, pp 29-38.
Cambridge University Press, Cambridge, UK.
Hwang, J.S. and C.C. Hung. 1991. Evaluation of the effect of integrated control of
sweetpotato weevil, Cylas formicarius Fabricius with sex pheromone and insecticide.
Chinese Journal of Agricultural Science 85:477-493.
Jackson D.M., J. R. Bohac, J.A. Thies, and H.F. Harrison 2010. 'Charleston Scarlet'
Sweetpotato. HortScience 45(2):306-309
James C, 2012. Global Status of Commercialized Biotech/GM Crops: 2010. ISAAA Brief
44. Available at http://www.isaaa.org/resources/publications/briefs/44/executivesummary/
Jones A, 1986. Sweetpotato heritability estimates and their use in breeding. HortScience
21(1): 14–17.
Keese P 2008. Risks from GMOs due to horizontal gene transfer. Environmental
Biosafety Research 7: 123-149.
Kobayashi M., 1983. The Ipomoea trifida complex closely related to sweet potato, in:
S.F. Shideler, H. Rincon, (Eds.), Proceedings of the 6th Symposium of the International
Society of Tropical Root Crops, Lima, Peru, 21–26 February, 1983, International Potato
Center, Lima, Peru, pp.561–568
25

Kok, E.J. and Kuiper, H.A., 2003. Comparative safety assessment for biotech
crops. Trends Biotechnology, 21: 439-444.
Kreuze, J.F., Valkonen, J.P.T., Ghislain, M., 2009. Genetic Engineering. In The
Sweetpotato. G. Loebenstein and G. Thottappilly (Eds.), Springer, Berlin.
Lebot V, 2010. Root and Tuber Crops: Handbook of Plant Breeding, 7: 97-125
Lim, S., Kim, Y.H., Kim, S.H., Kwon, S.Y., Lee, H.S., Kim, J.S., Cho, K.Y., Paek, K.Y.
and Kwak, S.S., 2007. Enhanced tolerance of transgenic sweetpotato plants that express
both CuZnSOD and APX in chloroplasts to methyl viologen-mediated oxidative stress
and chilling. Mol. Breed. 19 :227–239
Loebenstein G. and Thottappilly G. (eds.), 2009. The Sweetpotato. DOI 10.1007/978-14020-9475-0 2, C _ Springer Science+Business Media B.V. 2009, 290pp.
Low J, Walker T and Hijmans R, 2001. The potential impact of orange-fleshed
sweetpotatoes on vitamin A intake in Sub-Saharan Africa. Paper presented at a regional
workshop on food-based approaches to human nutritional deficiencies. The VITAA
Project, vitamin A and orange-fleshed sweetpotatoes in Sub-Saharan Africa 9–11 May
2001, Nairobi, Kenya
Lu B.L, 2008. Transgene escape from GM crops and potential biosafety consequences:
An environmental perspective. Collection of Biosafety Reviews 4: 66-141.
Luo, H.R., Santa Maria, M., Benavides, J., Zhang, D.P., Zhang, Y.Z. and Ghislain, M.
2006 Rapid genetic transformation of sweetpotato (Ipomoea batatas (L.) Lam) via
organogenesis. Afr. J. Biotech. 19 :1851–1857
MAAIF, 2011. Statistical Abstract. Agriculrtural Planning Department, Ministry of
Agriculture, Animal Industry and Fisheries (MAAIF). 64 pages
Martin, F.W. and Cabanillas, E. 1966. Post-pollen-germination barriers to seed set in
sweetpotato. Euphytica 15:404-411.
McCouch S.R., K.L. McNally, W. Wang and R.S. Hamilton 2012. Genomics of gene
banks: A case study in rice. American Journal of Botany 99(2): 407–423. 2012
Messeguer J, Fogher C, Guiderdoni E, Marfa V, Catala MM, Baldi G, Mele E (2001)
Field assessment of gene flow from transgenic to cultivated rice (Oryza sativa L.) using a
herbicide resistance gene as tracer marker. Theor Appl Genet 103: 1151–1159
Miller, R.E., Rausher, M.D., Manos, P.S., 1999. Phylogenetic systematics of Ipomoea
(Convolvulaceae) based on ITS and waxysequences. Syst. Bot. 24, 209–227.
Murata T., Y. Matsuda, 2003. Histological Studies on the Relationship between the
Process from Fertilization to Embryogenesis and the Low Seed Set of
Sweetpotato, Ipomoea batatas (L.) Lam. Breeding Science 53 (1): 41-49

26

Mwanga R.O.M., B. Odongo, C. Ocitti p’obwoya, R.W. Gibson, N. Smith, E. Carey,
2003. Release of six (“NASPOT 1” to “NASPOT 6”) sweetpotato cultivars in Uganda.
HortScience 38: 475-476.
Mwanga R.O.M., B. Odongo, C.N. Niringiye, A. Alajo, B. Kigozi, R. Makumbi, et al.,
2009.“NASPOT 7”, “NASPOT 8”, “NASPOT 9 O”, “NASPOT 10 O”, and “DimbukaBukulula” sweetpotato. HortScience 44: 828-832.
Mwanga R.O.M., M. Ghislain, J. Kreuze, G. N. Ssemakula and C. Yencho, 2011a.
Exploiting the use of biotechnology in sweetpotato for improved nutrition and food
security: Progress and future outlook. In: Proceedings of the International Conference on
Agro-Biotechnology, Biosafety and Seed Systems in Developing Countries, pp 25-31
Mwanga R.O.M., C. Niringiye, A. Alajo, B. Kigozi, J. Namakula, I. Mpembe, et al.,
2011b. “NAPOT 11”, a sweetpotato cultivar bred by a participatory plant-breeding
approach in Uganda. HortScience 46: 317-321.
Mwanga, R.O.M., Odongo, B., Niringiye, C. and Alajo, A., 2007. Release of Two Orange
-fleshed Sweetpotato Cultivars, ‘SPK004’ (‘Kakamega’) and ‘Ejumula’, in Uganda.
HortScience 42(7):1728–1730.
Mwanga, R.O.M., Odongo, B., Ocitti p’Obwoya, C.N., Gibson, R.W., Smith, N.E.J.M.
and Carey, E.E. 2001b. Release of five Sweetpotato cultivars in Uganda. HortScience 36
(2): 385-386.
Mwanga, R.O.M, 2001a. Nature of Resistance and Response of Sweetpotato to
Sweetpotato Virus Disease. PhD Thesis. http://www.lib.ncsu.edu/resolver/1840.16/5395
Newell, C.A., Lowe, J.M., Merryweather, A., Rooke, L.M. and Hamilton, W.D.O. 1995.
Transformation of sweetpotato (Ipomoea batatas (L.) Lam.) with Agrobacterium
tumefaciens and regeneration of plants expressing cowpea trypsin inhibitor and snowdrop
lectin. Plant Sci. 107: 215–227
Nickson T.E. 2008. Planning Environmental Risk Assessment for Genetically Modified
Crops: Problem Formulation for Stress-Tolerant Crops. Plant Physiology, 147: 494–502
Nishiyama, I. 1971. The origin of the sweetpotato plant. In: Barrau J. (ed.) Plants and the
migrations of Pacific peoples. Bihop Mus. Press Honolulu. Pp. 119-128.
OECD 1993. Safety considerations for biotechnology: Scale-up of crop Plants.
Organisation for Economic Cooperation and Development (OECD), Paris,France.
Available at http://www.oecd.org/dataoecd/26/261958527.pdf.
Okada Y, Saito A, 2009. Evaluation of resistance to complex infection of SPFMVs in
transgenic sweetpotato (Ipomoea batatas). Breed Sci (Jpn) 58(3): 243–250.
Pillai, K.S. and S.G. Nair. 1981. Field performance of some pre-released sweetpotato
hybrids to weevil incidence. Journal of Root Crops, 7:37-39.

27

Pimentel D. and Paoletti, M.G., 2009. Developing a 21st century view of agriculture and
the environment. Environmental Impact of Genetically Modified Crops, Chapter 5 (eds N.
Ferry and A.M.R. Gatehouse). CAB International, 42-58
Quist D, Chapela IH, 2001. Transgenic DNA introgressed into traditional maize landraces
in Oaxaca, Mexico. Nature 414: 541–543
Raybould A and Cooper I, 2005. Tiered tests to assess the environmental risk of fitness
changes in hybrids between transgenic crops and wild relatives: the example of virus
resistant Brassica napus. Environmental Biosafety Research 4, 127-140.
Raybould A, Stacey D, Vlachos D, Graser G, Li X, Joseph R, 2007. Non-target organisms
risk assessment of MIR604 maize expressing mCry3A for control of corn rootworms. J
Appl Ent., 131:391–399
Raybould A, Tuttle A, Shore S, Stone T. 2010. Environmental risk assessments for
transgenic crops producing output trait enzymes. Transgenic Research 19:595–609.
Raybould A.F, L.S. Higgins, M. J. Horak, R.J. Layton, N.P. Storer, J. M. DeLaFuente and
R.A. Herman 2012. Assessing the ecological risks from the persistence and spread of
feral populations of insect-resistant transgenic maize. Transgenic Research 21:655–664
Romeis J, Bartsch D, Bigler F, Candolfi MP, Gielkens MMC, Hartley SE, Hellmich RL,
Huesing JE, Jepson PC, Layton R, Quemada H, Raybould AF, Rose RI, Schiemann J,
Sears MK, Shelton AM, Sweet J, Vaituzis Z & Wolt JD, 2008. Assessment of risk of
insect-resistant transgenic crops to nontarget arthropods. Nature Biotechnology 26: 203208.
Romeis J, Hellmich RL, Candolfi MP, Carstens K, De Schrijver A, Gatehouse AMR,
Herman RA, Huesing JE, McLean MA, Raybould AF, Shelton AM and Waggoner A,
2011. Recommendations for the design of laboratory studieson non-target arthropods for
risk assessment of genetically engineered plants. Transgenic Research 20: 1-22.
Rong J, Lu B-R, Song ZP et al., 2007. Dramatic reduction of crop-to-crop gene flow
within a short distance from transgenic rice fields. New Phytol 173: 346-353
Runyararo Rukarwa. PhD candidate in Plant Breeding & Biotechnology, Makerere
University.
Shiotani I, Kawase T, 1987. Synthetic hexaploids derived from wild species related in
Ipomoea trifida. Jpn J Breed 39: 57–66.
Smit N.E.J.M., 1997. Integrated pest management for sweetpotato in Eastern Africa.
Wageningen, (Netherlands): Grafish Service Centrum Van Gils b.v. pp. 2–117
Song, G., H. Honda and K. Yamaguchi, 2004. Efficient Agrobacterium tumefaciensmediated transformation of sweetpotato (Ipomoea batatas (L.) Lam.) from stem explants
using a two-step kanamycin hygromycin. In Vitro Cellular and Developmental BiologyPlant article, 40(4): 359-365.

28

Srisuwan S., Sihachakr D, Siljak-Yakovlev S 2006. The origin and evolution of
sweetpotato (Ipomoea batatas Lam.) and its wild relatives through the cytogenetic
approaches. Plant Sci 171: 424–433.
Stathers, T.E., D. Rees, and D. Jeffries. 1999. Investigating the potential of cultivar for
differences in susceptibility to sweetpotato weevil as a means of control. Department
International Development (DFID), Natural Resources Institute (NRS), Final Technical
Report R6769.
Tabashnik, B.E., and Carrière, Y., 2009. Insect resistance to genetically modified crops.
Environmental Impact of Genetically Modified Crops, Chapter 5 (eds N. Ferry and
A.M.R. Gatehouse). CAB International, 74-100
Thompson, P.G., Schneider, J.C., Graves, B. and Sloan, R.C., 1999. Insect resistance in
sweetpotato plant introductions. HortScience 34 :711–714
Tumwegamire S., R. Kapinga, P.R. Rubaihayo, Do.R. LaBonte, W.J. Grüneberg,
G.Burgos, T. zum Felde, R. Carpio, E. Pawelzik, and R.O.M. Mwanga 2011. Evaluation
of Dry Matter, Protein, Starch, Sucrose, β-carotene, Iron, Zinc, Calcium, and Magnesium
in East African Sweetpotato [Ipomoea batatas (L.) Lam] Germplasm. HortScience 46:
348-357
USDA-APHIS, 2010a. "Federal Noxious Weed List (as of Dec. 10, 2010)."
http://www.aphis.usda.gov/plant_health/plant_pest_info/weeds/downloads/weedlist.pdf
US-EPA, 2001. "Biopesticides Registration Action Document - Bacillus thuringiensis
Plant-Incorporated Protectants " http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm
Wang, H. 1964. A study of the self- and cross-incompatibilities in the sweetpotato in
Taiwan (Formosa). Proc. Amer. Soc. Hort. Sc. 84:424-430.
Webadde C, and Maredia M.K, 2011. Environmental biosafety issues associated with
genetically engineered crop. In: Environmental safety of genetically engineered crops.
Grumet R, Hancock J, Maredia K, Weebadde C (eds). Michigan State Univ. Press, East
Lansing, pp. 47-57.
Wolt JD, Keese P, Raybould AF, Fitzpatrick JW, Burachik M, Gray AJ, Olin SS,
Schiemann J, Sears M & Wu F 2010. Problem formulation in the environmental risk
assessment for genetically modified plants. Transgenic Research 19: 425-436.
Woolfe, J.A. 1992. Sweetpotato, an untapped food resource. Cambridge University Press,
Cambridge, U.K.
Yada B., P. Tukamuhabwa, Villordon A., A. Alajo, and R.O.M. Mwanga 2010a. An
online database of sweetpotato germplasm collection in Uganda. Hortscience 45(1):153–
153.
Yada B., P. Tukamuhabwa, Wanjala B, Kim D-J, Skilton R.A., A. Alajo, and R.O.M.
Mwanga 2010b. Characterization of Ugandan sweetpotato germplasm using fluorescent
labeled simple sequence repeat markers. Hortscience 45(2):225–230. 2010.
29

Yen, D.E. 1982. Sweetpotato in historical perspective. pp. 17-30. In: Sweetpotato.
Villareal, R.L. and T.D. Griggs (eds.). Proceedings of the First International Symposium,
AVRDC, Publ. No. 82-172.
Zhang D.P., Golmirzaie A., Cipriani G., Panta A., Ghislain M., Smit N., Rety I., and
Michaud D. 1996. Developing weevil resistance in sweetpotato with genetic
transformation. International Potato Center (CIP) Program Report 95-96.
Zhang D.P., Rossel G., Kriegner A. and Hijmans R. 2004. AFLP assessment of diversity
in sweetpotato from Latin America and the Pacific region: Its implications on the
dispersal of the crop. Genet Resour Crop Evol 51:115-120.
Zhang, D.P., D. Carbajulca, G. Rossel, S. Milla, C. Herrera, and M. Ghislain. 2000.
Microsatellite Analysis of Genetic Diversity in Sweetpotato Cultivars from Latin
America. CIP Program Report 1999-2000.

30

CHAPTER 1
CHARACTERIZATION OF THE GENETIC DIVERSITY OF UGANDA’S
SWEETPOTATO GERMPLASM USING MICROSATELLITES

Introduction
Crop diversity loss is increasingly becoming a concern because of agricultural
intensification where a few improved cultivars are grown in homogenous farming systems
(FAO 2007a). Conserving a wide range of crop varieties with varying environmental
adaptive capacity can be both a source of genes for future crop improvement, as well as a
critical resource for the small-scale farmers who cannot afford the cost of chemical inputs
usually needed in intensified production systems. The highest levels of genetic diversity
for the majority of the important global food crops is in the South (FAO, 2007b), where
crop centers of origins are commonly found, and centers of diversity emerged due to
prolonged periods of farmer selection.
Sweetpotato, Ipomoea batatas L. (Lam.), is the sixth most important food crop in
the World after rice, wheat, potato, maize, and cassava (CIP, 2008). It is a new world crop
which was domesticated most likely somewhere in Central America but disseminated
rapidly to South America. Archeological data reveal that the oldest remains have been
found in the caves of the Chilca Canyon in Peru (Woolfe, 1992). However, based on
morphological relationships among related species, the center of origin appears to be
between the Yucatan Peninsula in Mexico and the Orinoco River in Venezuela (Austin,
1977). Evaluations of genetic variability patterns among germplasm from different parts
of the world have resulted in the suggestion that China, Southeast Asia, New Guinea and
East Africa are secondary centers of diversity. Sweetpotato was introduced to the East
African borders by Portuguese explorers during the 16th century (Zhang et al., 2000,
2004), however no study has assessed the relationship between East Africa’s and Brazil’s
germplasm.
31

A known fact is that Uganda has high level of sweetpotato diversity. In 2005, the
national sweetpotato program collected over 1300 farmers’ varieties (Yada et al., 2010a).
The high level of diversity can be attributed to the method of propagation (vegetative
propagation which maintain cultivar diversity when propagated in bulk), the allogamous
and hexaploidy nature of sweetpotato (Lebot, 2010), as well as widespread farmer
exchanges of varieties (Doyle et al., 2001; Zawedde, Chapter 2). Over 1300 African
varieties were characterized using morphological methodologies to determine the level of
genetic diversity (Yada et al., 2010a), and 946 of these were found to be morphologically
distinct accessions.
Knowledge about the genetic diversity and structure of existing crop varieties can
aid in making better conservation decisions, and help direct breeding programs towards
under-utilized gene pool. Characterization of crop diversity can be achieved through
morphological and molecular tools. Morphological characterization is an important first
step in assessment of diversity; however there are major limitations in relying only on
morphological characterization including low levels of polymorphism, low repeatability,
late expression for certain traits; phenotypic plasticity and parallel evolution (Smith and
Smith, 1992; Prakash & He, 1996; Karuri et al., 2010). A number of molecular markers
including random amplified polymorphic DNAs (RAPDs), restriction fragment length
polymorphism (RFLPs), amplified fragment length polymorphism (AFLP),
microsatellites or simple sequence repeats (SSRs), single nucleotide polymorphisms
(SNPs) have been developed and used to complement morphological characterization.
Selection of any particular DNA marker in a crop depends largely on the objective of the
research, available resources and technical skills (Otoo et al., 2009).
In sweetpotato research, a number of molecular markers have been used to study
the genetic diversity including RAPDs (Connolly et al., 1994; Gichuki et al., 2003; He et
al., 2006), DNA amplification fingerprinting (DAF) (He et al., 1995), AFLPs (Zhang et

32

al., 2000; Elameen et al., 2008), inter simple sequence repeats (ISSRs) (Hu et al., 2003),
selective amplification of microsatellite polymorphic loci (SAMPL) (Tseng et al., 2002),
and SSRs markers (Zhang et al., 2000; Gichuru et al., 2006; Veasey et al., 2008).
Microsatellites or SSRs have also been developed for genotyping sweetpotato (Jarret and
Bowen, 1994; Buteler et al., 1999; Hu et al., 2004; Yada et al., 2010b; Tumwegamire et
al., 2011).
Sweetpotato is a hexaploid (2n=6x=90) crop believed to have originated from
natural hybridization between several wild species (Lebot, 2010). There are three
hypotheses for generation of hexaploid sweetpotato: autopolyploidy (Kobayashi, 1983;
Shiotani and Kawase, 1989); allo-autopolyploidy (Zhang et al., 2001) and allopolyploidy
(Austin, 1987; Nishiyama, 1971; Srisuwan et al., 2006; Gao et al., 2011). The hexaploid
nature of sweetpotato and the complexity of its genomic makeup create a challenge when
using molecular markers for germplasm characterization (Lebot, 2010). However, SSR
markers have proved to be useful in unveiling genetic relationships that are closely
similar to morphological characterization in hexaploid Guinea yam (Dioscorea rotundata
Poir.) (Mignouna et al., 2003) and detecting polymorphism in sweetpotato (Veasey et al.,
2008). Small sets of SSR markers, as low as 10 primer pairs, have been found to
efficiently trace the movement of specific alleles between populations (Lebot, 2010), and
distinguish distinct individuals within a population (Yada et al., 2010b).
In this work, we used SSR markers to study the genetic diversity patterns of
Uganda’s sweetpotato germplasm. Two studies have previously used SSR markers to
analyze the genetic diversity of the sweetpotato germplasm in Uganda. One study focused
on assessed the genetic relationship among 192 selected superior farmers’ varieties (high
yielding, sweetpotato virus disease or Alternaria blight disease resistance, or high dry
matter content) and concluded that there were 190 distinct farmers’ varieties [Yada et al.,
2010b]. This study did not consider the weevil resistance trait yet weevils are one of the

33

major constraints to sweetpotato production in Uganda. They are responsible for more
than 60% yield losses especially in the semi-arid areas (Zhang et al., 1996; CIP, 2008).
The second study used 75 accessions from the same collection to assess the genetic
relationship between East African orange-fleshed and the white-fleshed farmers’
varieties, and how they related to accessions from China, USA, Papua New Guinea and
Peru (Tumwegamire et al., 2011). They found that the orange-fleshed types from East
Africa were distinct from those of non-African origin, and that orange-fleshed and and
white-fleshed farmers’ varieties from East Africa are closely related.
Since 1995, the national breeding programme has released twenty cultivars
(Mwanga et al., 2011). Most of these cultivars (Sowola and NASPOT 1 to 11) were
selected from bulked seed from a polycross between 18-24 parents. However, to our
knowledge, no study has used molecular markers to assess the genetic variability between
the farmers’ varieties and the released cultivars in Uganda. In addition, there has been no
study to assess the genetic relationship between East African germplasm and varieties
from Brazil and Mexico, and no studies to date have assessed the genetic relationship
between Ugandan accessions in relation to weevil resistance and the geographic source
(agro-ecological zones).
The purpose of this study was to determine the structure of genetic diversity in
Ugandan’s sweetpotato, and use this information to make recommendations that will
enhance efficient low-cost conservation of sweetpotato diversity. Specifically, we were
interested in characterizing the genetic relationship between farmers’ varieties of varying
weevil infestation response, and different agro-ecological zones; to measure the genetic
variability among and within farmers’ varieties and released varieties in Uganda; and to
compare the East African germplasm with accessions from Brazil and Peru.

34

Material and Methods
Plant material
Using the collection database maintained by the Sweetpotato Program in the
National Crop Resources Research Institute (NaCRRI) at Namulonge, a total of 220
Ugandan genotypes were selected for characterization. Varieties were selected based on
three categories; response to weevil infestation (Muyinza et al., 2012); agro-ecological
zone (Yada et al., 2010a); and varietal classification (Mwanga et al., 2011). Farmers’
varieties were selected from the five major sweetpotato growing agro-ecological zones of
Uganda: northern region or Zone III; eastern region or Zone V; central region or Zone VI;
western region or Zone VII; and southern region or Zone X (Figure 2.1). To determine
their genetic relationship with accessions from other parts of the world, the Ugandan
genotypes were compared with genotypes from other sweetpotato growing African
countries, Brazil and Peru. Planting material of selected varieties was grown in a screen
house for two months and young leafs were put immediately on ice and stored at -80oC
until DNA extraction.

DNA extraction and Simple Sequence Repeat (SSR) Amplification
Genomic DNA was isolated from 200mg of frozen leaf tissue using a modified
CTAB method (Doyle and Doyle, 1990), according to (Yada et al., 2010b). A total of 308
DNA samples were quantified, diluted to 5ng and prepared for polymerase chain reaction
(PCR) amplification. We tested 31 labeled SSR primer pairs used in previous sweetpotato
studies (Karuri et al., 2010; Yada et al., 2010b; Tumwegamire et al., 2011) for DNA
amplification, but only 19 SSR markers (Table 1) were confirmed for use in this work.
Other SSR markers were rejected due to non-specific amplification or for being
monomorphic. A total reaction volume of 10μl containing 1μlof the 10X Fermentas PCR
35

(50)

(50)

(50)
(50)

(20)

Legend
X- Highland Range
V- Kyoga plains
VI- Lake Victoria Cresent
I- Northern Eastern Drylands
II- NE Savannah Grasslands
III- NW Savannah Grasslands
IV- Para Savannah
VIII-Pastoral Rangelands
VII- W Savannah Grasslands

Figure 2-1 Map of Uganda showing the different agro-ecological zones (Roman
numerals) as described by the Ministry of responsible for agriculture, in Uganda.
Samples used in this study were originally collected from five agro-ecological zones:(III,
V, VI, VII and X, which are referred to in this study as northern, eastern, central, western
and southern region, respectively. The numbers of samples collected from each region is
provided in parenthesis.

36

buffer (10 mM Tris-HCl (pH 9.0), 30 mM KCl, 1.5 mM MgCl2), 0.1μl of
lyophilized5U/μl of Fermentas Taq DNA polymerase, 0.2μl of 10mM dNTPs, 0.5μl of
10µM of the primers, 2.5μl of 20ng/ul DNA template and 5.7μl of double distilled water,
was used for PCR. The PCR program was set as follows: 94oC for 3 minutes for initial
denaturation, followed by 35 cycles each consisting of denaturation at 94oC for 30 sec,
annealing at the indicated temperature for each primer pair (Table 2.1) and
polymerization at 72oC for 30 sec, and then final extension at 72oC for 20 minutes.
Amplified PCR products were prepared for each DNA sample and used for fragment
analysis using ABI 3730 capillary sequencer (Applied Biosystems).

Allele scoring and data analysis
Genemapper v3.7 software (Applied Biosystems) was used for peak detection and
fragment size estimation. AlleloBin software (Prasanth et al., 1997) was used to correct
any errors in the scored alleles due to slippage of DNA polymerase during PCR
(Schlotterer and Tautz, 1992). Analyses were performed using two types of data coding
previously employed to describe diversity in polyploids (Esselink et al., 2004; JØrgensen
et al., 2008; Kloda et al., 2008; Garcia-Verdugo et al., 2009; Sampson and Byrne, 2012).
Firstly, the multilocus data were transformed into binary arrays of the presence/absence
of an allele for each individual, using ALS-Binary software (Prasanth and Chandra,
1997). Secondly, the MAC-PR (microsatellite DNA allele counting-peak ratios) method
(Esselink et al., 2004) was employed to obtain allele-dosage of 16 markers for each of the
polyploidy individuals. The MAC-PR method makes use of the quantitative values for
peak areas provided by the GeneMapper software. For each locus, all alleles were
analysed in pairwise combinations in order to determine their copy numbers in the
individual samples. This was accomplished by calculating ratios between
37

Table 2-1 Characteristics of SSR markers used to evaluate East African sweetpotato
accessions: name, labeled dye, motifs, annealing temperature and reference.
1

Name
IB-S07

Dye
6-FAM

Motif
(tgtc)7

Ta
60

IB-R03
IB-R16
IB-R13
J10A

PET
VIC
NED
PET

(gcg)5
(gata)4
(ttc)6
(aag)6

58
59
58
TD (57-62)

J175
VIC
(aatc)4
IBCIP-9
6-FAM (cca)2ac(acc)6
IBCIP-13 NED
(acc)3+(cgg)2+(tgc)3+(gtc)2
IB-R19
PET
(cag)5b
BU691984 6-FAM (tgg)6
JB1809
VIC
(cct)6(ccg)6
IBSSR09
NED
(gaa)5(gag)3
IB-R08
PET
(t3a)4
BU690524 VIC
(cag)5
BU690708 6-FAM (ccg)5
1B-S18
6-FAM (tagc)4
IBCIP-1
6-FAM (acc)7a
IBSSR07
PET
(ct)7a(tc)4
IB-R12
NED
(cag)5a
1
Ta- annealing temperature;

TD (57-62)
TD (50-60)
TD (57-62)
TD (57-62)
TD (57-62)
TD (50-60)
TD (57-62)
TD (50-60)
TD (57-62)
TD (57-62)
TD (57-62)
TD (57-62)
TD (57-62)
TD (57-62)

2

unpublished data developed from 2002 to 2003 at CIP

3

unpublished data developed from 2005 to 2006 at CIP

38

Reference

2

Benavides (unp.)
Benavides (unp.)
Benavides (unp.)
Benavides (unp.)
3
Solis et al. (unp.)
Solis et al. (unp.)
Yañez 2002
Yañez 2002
Benavides (unp.)
Hu et al., 2004
Solis et al. (unp.)
Hu et al., 2004
Benavides (unp.)
Hu et al., 2004
Hu et al., 2004
Benavides (unp.)
Yañez 2002
Hu et al., 2004
Benavides (unp.)

peak areas for two alleles in all samples where these two alleles occurred together. Access
to individuals with six different alleles or three different alleles provided a baseline for
calculating the single- and double copy dosage, respectively.
The binary data was used to determine the polymorphic information content
(PIC), which is the measure of the usefulness of each marker in distinguishing between
individuals that was formulated by Weir (1996) as:
n
2
PICi= 1-∑ Pij
j=1
Where PICi is the polymorphic information content of a marker i; Pij is the frequency of
the jth pattern for marker i and the summation extends over n patterns.
The standard measures of genetic diversity for each population including number of
polymorphic loci (P), percentage of polymorphic loci (%P) and Nei’s (1973) gene
diversity (D) were estimated from binary data using GenAlex 6.4 (Peakall & Smouse,
2006). The Euclidean distance was calculated; this distance was used because it does not
categorize the absence of an allele in two individuals as a shared characteristic (Legendre
and Legendre 1998; Kloda et al., 2008; Sampson and Byrne, 2012).
Analysis of molecular variance (AMOVA) was also performed using GenAlEx
version 6.4 to estimate the total variance and distribution of diversity within and between
populations. This method was included because it has been used to determine genetic
diversity in sweetpotato (Yada et al., 2010b). Wright’s F-Statistic (FST, fixation index)
was also computed, using GenAlEx software, to estimate the amount of genetic variance
that can be explained by population structure (Holsinger and Bruce, 2009).
Fixation index, FST = (HT - HI) / HT,

39

where HI is the mean observed heterozygosity per individual within subpopulations and
HT is the expected heterozygosity in a random mating total population. FST can range
from 0.0 (no differentiation) to 1.0 (complete differentiation, that is, subpopulations fixed
for different alleles).
The phylogenetic relationship among populations was assessed using DARwin
version 5 (Perrier and Jacquemoud-Collet, 2006). Similarity matrices were constructed
from the binary data with Jaccard’s coefficients (Jaccard, 1908). Jaccard’s coefficient
=Nab/(Na + Nb), where Nab is the number of alleles shared by two individuals a and b, Na
is total number of alleles in sample a, and Nb is total number of alleles in sample b.
Genetic distances between populations were obtained by computing the usual Euclidian
distance matrix based on haplotype frequencies. From this matrix, a dendrogram was
constructed using the neighbor joining method (NJ) from Saitou and Nei (1987). The
significance of each node was evaluated by bootstrapping data over a locus for 5000
replications of the original matrix. We examined hierarchical genetic variation between
individuals using the unweighted pair group method analysis (UPGMA), as suggested by
Sneath and Sokal (1973).
Clustering patterns of individuals and populations were examined using
STRUCTURE version 2.3.3 (Pritchard, Stephens & Donnelly, 2000), which is reported to
have the capability to generate more accurate population structuring (Pritchard, Wen &
Falush, 2009). Using the allele dosage (MAC-PR) data for each individual, individuals
were assigned probabilistically to genetic clusters (K). The STRUCTURE program was
run using no prior assumptions of population structure with an admixture ancestry model
and the recommended methods for recessive alleles, and allele frequencies correlated.
The analysis was used to determine whether biologically relevant clusters could be
40

determined among the plants sampled, and establish the proportion of an individual’s
genome (Q) that originated from each cluster. For all analyses, the Markov chain Monte
Carlo (MCMC) parameters were set to a burn-in period of 50000 with 50000 iterations.
The optimum K, indicating the number of true clusters in the data, was determined from
20 replicate runs for each value of K (K set to 10) using the method described by Evanno
et al. (2005) and the ad hoc Quantity Delta K, based on the rate of change in the log
probability of the data between successive K values. Parameters of the method of Evanno
et al. (2005) were calculated using the program Structure Harvester version 0.6.92(Earl
and von Holdt, 2011). Similarity among different runs was calculated by the method of
Jakobsson and Rosenberg (2007) as used in their computer program CLUMPP 1.1.2. This
method calculates a similarity coefficient h′, which allows the assessment of the similarity
of individual runs of the program STRUCTURE. The optimal alignment of 20 replicates
of K values was determined using the computer program CLUMPP 1.1.2 (Jakobsson
&Rosenberg, 2007) and clusters were visualized using the program DISTRUCT 1.1
(Rosenberg, 2004).The optimum number of clusters found in the total sample was K = 2,
corresponding to the two subspecies (see Results). Each subspecies was further analysed
separately under the same conditions.

41

Results
Genotype description
A total of 105 out of the 220 Ugandan farmers’ varieties selected from the original
collection database were available in the national sweetpotato ‘gene bank’. Only 62% of
the 136 varieties used by Muyinza et al. (2012) for weevil resistance morphological
characterization could be accessed for this work. Based on this observation, it is assumed
that at least 40% of the crop diversity maintained at the national sweetpotato ‘gene bank’
has been lost since its collection in 2005. This loss was attributed to caterpillar and viral
incidences at the collection site (Agnes Alajo, personal communication). To increase the
sample size for Ugandan accessions, 34 farmers’ varieties were collected from the
farmers’ fields in different zones. A list of the varieties included in this work is available
in Annex 1.

SSR markers amplification
A total of 106 polymorphic alleles were scored for the 19 SSR markers (Table
2.2). The number of alleles per locus ranged from 3 to 9. Three markers had very low
PIC; IB-S07 (0.22), JB1809 (0.19) and IBSSR09 (0.20) thus were excluded from further
analyses.

Genetic relationship between farmers’ varieties with varying response to weevil
infestation
There was a significant (P<0.018) difference in levels of genetic variability among

accessions with different weevil responses (Table 2.3). Analysis of molecular variance
(AMOVA) indicated that only 4% of the genetic variation was partitioned among the
categories. On average, 61 alleles out of the 92 alleles used were polymorphic (Table 2.4).

42

Table 2-2 Observed base pair (bp) range, number of alleles and polymorphic information
content (PIC) for the SSR markers used to characterize sweetpotato genotypes from
Uganda, Kenya and Tanzania.
Name
IB-S07
IB-R03
IB-R16
IB-R13
J10A
J175
IBCIP-9
IBCIP-13
IB-R19
BU691984
JB1809
IBSSR09
IB-R08
BU690524
BU690708
1B-S18
IBCIP-1
IBSSR07
IB-R12

bp range
173-177
244-258
194-212
226-296
175-229
112-144
176-194
196-374
191-208
252-267
196-264
196-208
205-216
272-293
235-259
211-235
135-189
158-178
303-341

No. of alleles
5
5
5
9
8
6
4
7
4
3
4
5
4
8
7
4
8
4
7

43

PIC
0.31
0.60
0.66
0.77
0.82
0.81
0.80
0.52
0.63
0.67
0.19
0.31
0.64
0.80
0.81
0.76
0.89
0.70
0.79

Table 2-3 AMOVA for genetic variability within and among the different weevil response
categories of sweetpotato genotypes obtained from Uganda.
Source

df

SS

MS

Est. Var.

%

FST

P value

Among categories 2

39.273

19.636

0.629

4%

0.045

0.018

Within categories

27

364.394

13.496

13.496

96%

Total

29

403.667

14.125

100%

Table 2-4 Number of individuals, number of rare alleles (RA), and genetic diversity
1
parameters for the different weevil response categories of sweetpotato genotypes
obtained from Uganda.
Weevil response
Individuals RA
P
%P
D
Moderate
12
1
60
64.6
0.332
Resistant
11
0
61
62.2
0.328
Susceptible
7
2
61
62.2
0.340
Mean
60.5
63.0
0.333
Total
30
3
1
Where P is total number of polymorphic loci per category, %P is percentage of
polymorphic loci, and D is Nei’s (1973) gene diversity estimated with computer program
GenAlex 6.4 (Peakall & Smouse, 2006).

44

Determining relatedness between farmers’ varieties and released cultivars in
Uganda
A total of 10 newly bred cultivars released by the national program were
compared with 158 Ugandan farmers’ varieties. AMOVA showed that there was
significant genetic variation (P<0.014) between the farmers’ varieties and the released
cultivars, but only 3% of the genetic variation was explained by differences among the
cultivars and farmers varieties (Table 2.5). An average of 73 polymorphic alleles was
observed in each category (Table 2.6). Farmers’ varieties had 32 unique alleles, while
improved cultivars had none. The heterozygosity value for both categories was very low.
The unweighted neighbor joining (NJ) algorithm cluster analysis generated three
mega-clusters each with 2-3 sub-clusters (Figure 2.2A). Bred cultivars were grouped into
two of the three clusters, and there were only three sub-clusters only contained by
farmers’ varieties. Many (5/11) bred cultivars were grouped together with a Kenyan
variety Kakamega, which was purposely included in this analysis because it is a known
maternal parent for many of these cultivars. There was improved cultivars appear to have
higher genetic similarity to the northern and eastern varieties. (Figure 2.2B).

Genetic relationship between genotypes from Uganda’s agro-ecological zones and
accessions collected from other East African countries
Analysis of Molecular Variance (AMOVA) indicated that only 6% of the genetic
variation was explained by differences among the regions (Table 2.7). Analysis of the
sixteen microsatellites yielded a total of 92 presumptive loci in the 206 sweetpotato
genotypes from the eight predefined populations (Table 2.8). An average of 70
polymorphic loci was observed in each population. The level of genetic diversity varied

45

Table 2-5 AMOVA for genetic variability within and among the different varietal
categories (farmers’ varieties vs. released cultivars from Uganda).
Source
Between variety type
Within variety type
Total

df
1
166
167

SS
MS
24.854
24.854
2185.110 13.163
2209.964

Est. Var.
0.622
13.163
13.785

%
FST P value
3
0.045 0.005
97
100

Table 2-6 Number of individuals, number of rare alleles (RA), and genetic diversity
1
parameters for the different varietal categories of sweetpotato genotypes obtained from
Uganda.
Varietal category
Individuals Rare alleles P
%P
D
Farmers varieties
158
32
89
91.6
0.340
Improved cultivars
10
0
57
50.5
0.306
Mean
73
75.0
0.323
Total
168
32
1
Where P is total number of polymorphic loci per category, %P is percentage of
polymorphic loci, and D is Nei’s (1973) gene diversity estimated with computer program
GenAlex 6.4 (Peakall & Smouse, 2006).

46

A

0

0.2

Figure 2-2 A) Cluster analysis of 171 sweetpotato genotypes generated by unweighted
neighbor joining algorithm using DARwin software. The Ugandan improved cultivars are
indicated by the blue circles; orange represents accessions from the northern region; black
represents accessions from the eastern region; cyan for accessions from the central region;
magenta for accessions from the western region; and green for accessions from the
Southern region. There was no clustering by regions or by varietal type.

47

B

0.6

Southern Ug

0.4
Eastern Ug
0.2

Western Ug

0

Improved Ug
-0.2

Northern Ug

Central Ug

0.2

-0.4
-0.4

0.1
-0.2

0

0
0.2
0.4

-0.1
0.6 -0.2

Figure 2-2 B) Non-metric multidimensional scaling representation of the principal
coordinates analysis (PCoA) by regions in Uganda. The percentage of variation explained
by the three axes is 100%. Pairwise comparison of the different Ugandan populations
shows that improved cultivars appear to have higher genetic similarity to the northern and
eastern varieties.

48

Table 2-7 AMOVA for genetic variability within and among the populations of
sweetpotato from different Uganda’s agro-ecological zones and other East African
countries (Kenya, and Tanzania) as well as Uganda’s improved cultivars.
Source
Among regions
Within regions
Total

df
7
220
227

SS
MS
275.238 34.405
3206.696 13.646
3481.934

Est. Var.
0.805
13.646
14.451

%
6
94
100

FST
0.056

P value
0.010

Table 2-8 Number of individuals, number of unique alleles, and genetic diversity
1
parameters for sweetpotato populations from various Uganda’s agro-ecological zones
and other East African countries (Kenya, Tanzania), as well as Uganda’s improved
cultivars.
Regions
Northern Uganda
Eastern Uganda
Central Uganda
Western Uganda
South-western Uganda
Improved cultivars
Kenya
Tanzania
Mean
Total

1

Individuals
23
66
34
43
14
11
15
22
228

RA
2
1
1
4
0
0
2
4
13

P
69
78
73
79
62
57
62
78
70
92

%P
75.3
84.4
79.2
85.7
67.5
62.3
67.5
84.4
75.8
-

D
0.433
0.433
0.442
0.449
0.431
0.347
0.457
0.514
0.438
-

Where P is total number of polymorphic loci per region, %P is percentage of
polymorphic loci, and D is Nei’s (1973) gene diversity estimated with computer program
GenAlex 6.4 (Peakall & Smouse, 2006).

49

among the different populations. Most regions in Uganda had populations with unique
alleles ranging from 1 to 4, except the south-western region, which had none. Tanzanian
accessions also had a high number of unique alleles (4), and they had the highest level of
heterozygosity (D). Overall the level of heterozygosity (D) for the collected samples was
low; only 44% of the genotypes are expected to be heterozygous at a given locus under
random mating.
Pairwise comparisons of genetic differentiation among regions indicated that
accessions from Tanzania were significantly different (P<0.001) from all the Ugandan
populations and the Kenyan one (Table 2.9). The Kenyan accessions were significantly
different from only the central Ugandan populations. The significant difference between
the accessions from Tanzania and the populations from Uganda and Kenya is clearly
shown in the genetic distance matrix (Figure 2.3).

Genetic relationship between Ugandan genotypes and accessions collected from
elsewhere
Analysis of Molecular Variance (AMOVA) indicated that only 24% of the genetic
variation was explained by differences among the regions (Table 2.10). A total of 84 loci
were recorded after screening 260 individuals with thirteen markers (Table 2.11). On
average, 51 loci were polymorphic for each region. There were unique alleles recorded
for accessions obtained from Ghana, Peru, Tanzania and Uganda. The accessions from
Ghana had the highest heterozygosity value (D).

50

1

Table 2-9 A pair-wise FST for populations from various Uganda’s agro-ecological zones and other East African countries (Kenya, Tanzania),
as well as Uganda’s improved cultivars

2

Regions
Improved Northern
Eastern
Central
Western Southern
Kenya
Tanzania
Improved Uganda
0.000
Northern Uganda
0.045
0.000
Eastern Uganda
0.027
0.008
0.000
Central Uganda
0.059
0.004
0.015
0.000
Western Uganda
0.034
0.016
0.009
0.002
0.000
Southern Uganda
0.055
0.032
0.008
0.026
0.011
0.000
Kenya
0.032
0.041
0.026
0.056
0.035
0.039
0.000
Tanzania
0.134
0.123
0.145
0.152
0.156
0.130
0.120
0.000
1
FST is the mean reduction in observed heterozygosity of a subpopulation (relative to the total population) due to genetic drift among
subpopulations. 0.0 means no differentiation while 1.0 means complete differentiation.
2
Shaded values indicate significant difference between the populations (P<0.001).

51

Improved Ug

0.6

Kenya
Northern Ug

0.4

Central Ug
0

Southern Ug

Eastern Ug

Western Ug

Tanzania

-0.4

0.4

-0.6

0.2
-0.4
0

0
-0.2

0.4
-0.4

0.8
1.0 -0.6

Figure 2-3 Non-metric multidimensional scaling representation of the principal
coordinates’ analysis (PCoA) comparing genotypes from different Uganda’s agroecological zones and other East African countries. The percentage of variation explained
by the three axes is 89.1%.

52

Table 2-10 AMOVA for genetic variability within and among sweetpotato populations
from Uganda, other African countries (Ghana, Kenya, Tanzania), and American countries
(Brazil, Peru, USA).
Source
Among Regions
Within Regions
Total

df
7
252
259

SS
MS
328.088 46.87
1513.793 6.031
1841.88

FST
%
24 0.244
76
100

Est. Var.
1.95
6.031
7.981

P value
0.001

Table 2-11 Number of individuals, number of rare alleles (RA), and genetic diversity
1
parameters for populations from Uganda, other African countries (Ghana, Kenya,
Tanzania), and American countries (Brazil, Peru, USA).
Region
Brazil
Peru
Ghana
Kenya
Tanzania
Uganda
USA
Other African
countries
Mean
Total

Individuals
22
20
19
13
10
166
5
5

RA
0
2
2
0
1
3
0
0

P
54
57
61
45
59
64
30
35

%P
63.8
68.2
73
53.4
70
75.9
36.2
41.4

D
0.464
0.469
0.514
0.454
0.467
0.361
0.313
0.365

260

9

51
84

60.2
-

0.421
-

1

Where P is total number of polymorphic loci per region, %P is percentage of polymorphic
loci, D is Nei’s (1973) gene diversity estimated with computer program GenAlex 6.4
(Peakall & Smouse, 2006).

53

Pairwise comparisons of genetic differentiation among regions indicated that
Uganda’s germplasm was significantly different (P<0.001) from accessions from Brazil,
Peru and Ghana (Table 2.12 and Figure 2.4). The Jaccard’s similarity coefficients
generated using the DARwin software ranged from 0.0 to 0.95 with a mean of 0.56
(Figure 2.5). More than 70% of the pair-wise similarity coefficients were between 0.50
and 0.63. Some accessions had very low genetic values (< 0.05) indicating little genetic
differentiation.
A dendrogram generated by DARwin software revealed three clusters (Figure
2.6). The dendrogram was pruned from the complete tree to show clustering between
genotypes that had bootstrap values greater than 60. This pruned tree shows similar broad
clustering as observed for the complete tree (data not shown). East African germplasm
and accessions from USA were found in Cluster A, majority of accessions from Brazil
and Peru were in cluster B while most Ghanaian accessions were found in Cluster C.
The Bayesian model of STRUCTURE (Pritchard, Stephens & Donnelly, 2000)
assigned the individuals to two major genetic clusters, since the highest Delta K was
observed at K=2 (Figure 2.7). Each region had a proportion of its members allocated to
cluster K1 and K2 based on size of genetic constitution that originates from each cluster
(Table 2.13). All individuals appeared to have a component of both clusters in their
genome; however, the Ugandan and Kenyan accessions had a very high proportion of
their genome originating from cluster K1, Brazil, Ghana and Peru had a very high
proportion of their genome originating from cluster K2, while Tanzanian accessions were
composed of a mixture of the two clusters (Figure 2.8).

54

1

Table 2-12 Pair-wise FST for populations from Uganda, other African countries (Ghana,
Kenya, Tanzania), and American countries (Brazil, Peru, USA)

2

.

Regions
Others3 Brazil Ghana Kenya Peru
Tanzania Uganda USA
Other African
Countries
0.000
Brazil
0.187 0.000
Ghana
0.258 0.347 0.000
Kenya
0.006 0.294 0.314 0.000
Peru
0.181 0.127 0.221 0.308 0.000
Tanzania
0.010 0.197 0.246 0.190 0.166
0.000
Uganda
0.000 0.285 0.387 0.068 0.299
0.100
0.000
USA
0.029 0.225 0.257 0.210 0.116
0.033
0.133
0.000
1
FST is the mean reduction in observed heterozygosity of a subpopulation (relative to the
total population) due to genetic drift among subpopulations. 0.0 means no differentiation
while 1.0 means complete differentiation.
2
Shaded values indicate significant difference between the populations (P<0.001)
3Others stand for accessions obtained from other African countries including Rwanda and
Mozambique.

55

0.6
Ghana
Kenya
0.4

USA
Others
Tanzania

Uganda
0

Peru

-0.4
Brazil

0.6

-0.6

0.4
-0.6

0.2
-0.4

0

0
0.4

-0.2
0.6

-0.4
0.8 -0.6

Figure 2-4 Non-metri multidimensional scaling representation of the PCoA by regions.
Others stand for other African countries. The percentage of variation explained by the
three axes is 88.8%.

56

12000
10000
8000
6000
4000
2000
0

0

0.18

0.24

0.36

0.48

0.60

0.72

0.84

0.96

Figure 2-5 Frequencies of pair-wise similarity coefficients for 260 sweetpotato accessions
from Uganda, other African countries (Ghana, Kenya, Tanzania), and American countries
(Brazil, Peru, USA) provided by DARwin software. The minimum value was 0,
maximum 0.95 and the mean was 0.56. Zero genetic value means 100% genetic
similarity, while 1.0 means completely dissimilarity.

57

A

B

C

0

Bootstrap - 5000

0.1

Figure 2-6 A dendrogram of the unweighted pair group method analysis (UPGMA)
cluster analysis on the basis of Jaccard’s simple sequence repeat (SSR) showing genetic
similarities among 102 accessions. Colors brown, magenta, red, blue, cyan, green, black
and pink represents accessions obtained from Brazil, Ghana , Kenya, Peru, Rwanda,
Tanzania, Uganda and USA, respectively.
58

Table 2-13 Number of individuals obtained from Uganda, other African countries
(Ghana, Kenya, Tanzania), and American countries (Brazil, Peru, USA) and the
1
proportion of members of region that is found in clusters K1 and K2.
Region
No. of individuals
Cluster K1
Cluster K2
Brazil
22
0.17
0.83*
Peru
20
0.14
0.86*
Ghana
19
0.02
0.98*
Tanzania
19
0.33
0.67*
Kenya
13
0.81
0.19
Uganda
166
0.91
0.09
Other African countries 5
0.87
0.13
USA
5
0.69
0.31
*Regions with majority of its members found in cluster K2.
1
The Bayesian clustering method was implemented in STRUCTURE version 2.3.3
(Pritchard, Stephens & Donnelly, 2000).

59

Delta K = mean (L(K)) / sd(L(K))
40

Delta K

30

20

10

0
2

3

4

5

K

6

7

8

9

Figure 2-7 Determining the best K value using Structure Harvester (Earl and von Holdt
2012). K=2 had the highest ad hoc Quantity Delta K thus selected as the most appropriate
population structure.

1.00
0.80
0.60
0.40
0.20
0.00
BR

PE GH TZ KE AF

UG

USA

Figure 2-8 Probabilities of membership for each individual in clusters K1 and K2 as
determined by the Bayesian clustering method implemented in STRUCTURE version
2.3.3 (Pritchard, Stephens & Donnelly, 2000). BR, PE, GH, TZ, KE, AF, and UG stands
for Brazil, Peru, Ghana, Tanzania, Kenya, other African countries (Mozambique and
Rwanda) and Uganda, respectively. Other accessions were obtained from USA. Each
individual is represented as a vertical bar and the colors correspond to its membership
probabilities in clusters K1 (green) and K2 (red).
60

Discussion
The number of alleles per primer pair observed in this work is close to that
obtained by Yada et al. (2010) using the same SSR markers. However, our number of
alleles varies somewhat from those reported by Tumwegamire et al. (2011) on similar
sweetpotato germplasm using the same markers. Higher number of alleles was observed
for some markers in this work and this may be attributed to the accession sample size and
or method of DNA fragment analysis used. Yada et al. (2010b) assessed 192 samples
using the ABI system similar to what was used in this work, while Tumwegamire et al.
(2011) screened 75 samples with the LiCOR system.
A total of 92 out of 106 markers were highly polymorphic, which confirms the
high discriminating power of the SSR markers (Hwang et al., 2002; Gichuru et al., 2006,
Veasey et al., 2008, Yada et al., 2010; Tumwegamire et al., 2011). Hwang et al (2002)
suggested the high level of polymorphism of molecular markers in sweetpotato was due
to the large genome size and high heterozygosity. If the sweetpotato is an allopolyploid as
some suggest (Austin, 1987; Nishiyama, 1971; Srisuwan et al., 2006; Gao et al., 2011), it
would be expected to have high levels of heterozygosity, which would be maintained by
the common practice of vegetative propagation of sweetpotato. In addition, the mating
system of sweetpotato, which is outcrossing combined with self incompatibility, would
likely support high levels of heterozygosity. However, genetic diversity within our
populations was low, along with levels of heterozygosity. The high levels of
polymorphism we observed in the sweetpotato germplasm were apparently partitioned
primarily across regions, not among samples within a region. Heterozygote deficiency
may be due to presence of null alleles; inbreeding as a result of reduced population size;
or unrecognized genetic structure within populations (Pemberton et al., 1995).

61

The low levels of genetic variation observed among accessions from the different
weevil response categories may indicates that the weevil resistance source is highly
dispersed and does not originate from a narrow source. Genetic linkage mapping is
underway to dissect this trait from one variety with partial resistance to weevils.
Preliminary analysis confirms that this trait is highly polygenic (Yada pers. comm.).
There was a significant difference in the levels of genetic variation between the
farmers’ varieties and the improved cultivars. Majority of the improved cultivars were
found in one cluster probably because of by the source of parents used in breeding the
improved cultivars. These improved cultivars were generated from a polycross of a single
cultivar crossed with 18-24 parents including introductions from other parts of the world
(Mwanga et al., 2003; 2009; 2011). For example NASPOT 7-11 were generated from a
polycross of 24 parents with ‘Kakamega’ (a Kenyan farmers’ variety) as the maternal
parent. The cross included ‘Zapallo’ (from Peru), ‘Beauregard’ and ‘Jewel’ (from USA),
and three introductions from the International Institute of Tropical Agriculture (IITA)
based in Nigeria.
In the cluster analysis, three sub-clusters were identified that were composed of
only farmers’ varieties, which suggests that there is still a large genetic diversity in
Uganda that has not been tapped by breeding programs. The fourteen unique alleles
observed in farmers’ varieties compared to released cultivars suggests a need to conserve
this diversity for future breeding programs or farmers needs.
There was a low mean genetic similarity coefficient of 0.56 between the Ugandan
varieties and varieties from other African countries, similar to that observed by
Tumwegamire et al. (2011). This suggests that there is a large amount of sweetpotato
genetic diversity in Uganda that needs to be preserved. Very few Ugandan farmers’
varieties had high genetic similarity coefficients (<0.05) with other varieties, suggesting
62

high levels of genetic diversity are being maintained by the farmers. This observation was
supported by the AMOVA results, which showed over that 94% of the variation was
found within populations. Similar results were observed in previous studies assessing East
African germplasm (Gichuru et al., 2006; Tumwegamire et al., 2011). The high level of
genetic variability within a given gene pool is influenced by a number of factors such
gene flow through intra-specific introgression, and movement of plant materials between
localities during farmer plant material exchange and levels of farmer selection (Gichuru et
al., 2006; Veasey et al., 2008). Farmers exchange can explain our observation that a
number of Ugandan farmers’ varieties had high genetic similarity with varieties from
Kenya, Rwanda and Tanzania (Zawedde, Chapter 2).
While the majority of genic diversity was spread evenly across Uganda and East
Africa, there were distinct regional clusters of genotypes. Pairwise comparisons of
genetic differentiation among regions indicated that accessions from Tanzania were
significantly different (P<0.001) from all the Ugandan populations and the Kenyan one;
Uganda’s germplasm was significantly different (P<0.001) from accessions from Brazil,
Peru and Ghana. This suggests that the same genes were being assorted in different arrays
in the various regions. Even though sweetpotato was likely introduced into Africa from
Brazil, we observed substantial differentiation between the Brazilian accessions and all of
the African populations. This could imply that only a sample of the Brazilian germplasm
arrived in Africa, but we found no alleles in the Brazilian accessions that were not in the
African accessions. A more likely explanation for the higher degree of differentiation
among populations is that the sweetpotato gene pool evolved to local conditions and
human preferences as it was spread from America throughout Africa. This would explain
why the majority of the American accessions were found in one cluster, and most of the
East African germplasm were found in another cluster. However, some of the accessions
63

from Tanzania clustered more closely with the American group. This may be because the
Tanzanian collection contained ancient genotypes that have been cultivated in isolation
on some of Tanzania’s islands. The Ghanaian germplasm was more closely related to the
American accessions than with the East African genotypes, although they still formed a
unique cluster. The accessions from the USA had a higher genetic similarity with the East
African germplasm compared with other American genotypes, probably because these
accessions were introduced in Uganda about 15 years ago and have been used as paternal
parents in the polycrosses (Mwanga et al. 2003).It is likely that hybridization and
introgression has occurred between these accessions and Ugandan germplasm.

Conclusions
The presence of unique alleles in populations from various Uganda’s agroecological zones and other global regions, as well as the regional diversity patterns,
indicates a need to collect and characterize in more depth the germplasm that is at a threat
of being abandoned in favor of more productive varieties. This will allow making better
choices of what need to be preserved in order to secure wide array of the farmers’
varieties across Africa. These genotypes need to be incorporated in the collections at the
national gene bank and managed to ensure their long-term conservation. The focus should
be in regions where cultivars are actually replacing traditional varieties. Zawedde
(Chapter 2) found that only a fraction of sweetpotato growers have access to improved
varieties, and many of those that do, maintain some traditional varieties.
Ex situ conservation can be expensive and in situ conservation strategies should be
employed whenever possible (Rao and Campilan, 2002; Lebot, 2010). To do this most
effectively, periodic monitoring of regional patterns of in situ diversity need to be
undertaken to identify genetic diversity at risk and minimize ex situ collection size. In a
64

direct survey of Ugandan sweetpotato farmers, Zawadde (Chapter 2) found that most
farmers maintained multiple varieties in spite of many constraints, although they
experimented freely with other varieties and the turnover of their predominant variety was
high.
The losses of sweetpotato germplasm in the clonal gene bank also suggest a need
to improve current ex situ conservation methods. We suggest applying conservation
methods such as in vitro (slow grow) storage of different plant parts and or cryostorage
(Keller et al., 2010) in Uganda. These methods have been applied in a number of other
vegetatively propagated species including pears (Reeds et al., 1998), a number of
medicinal plants (Krishnan et al., 2011), strawberry (Höfer and Reeds 2011) and potato
(Keller et al., 2010; Yamamoto et al., 2012). However, ex situ conservation is very
expensive and there are losses that are inevitable in ex situ conservation, thus there is a
need to complement these approaches with in situ conservation. Since most small-scale
farmers in Uganda tend to grow many varieties together, this presents an opportunity to
develop in situ conservation strategies (Rao and Campilan, 2002; Lebot, 2010). To
develop effective in situ conservation methods, there is need for a clear understanding of
how farmers make decisions to maintain or drop varieties, which is discussed in Chapter 2.

65

APPENDIX

66

Table A-1 Description of the sweetpotato accessions studied including their local code,
clone name of the varieties, category of the varieties, their original source and response to
weevil infestation.
Local
Code
BR01
BR02
BR03
BR04
BR05
BR06
BR07
BR08
BR09
BR10
BR11
BR12
BR13
BR14
BR15
BR16
BR17
BR18
BR19
BR20
BR21
BR22
BR23
BR24
BR25
BR26
BR27
GH01
GH02
GH03
GH04
GH05
GH06
GH07
GH08
GH09
GH10
GH11
GH12
GH13
GH14
GH15
GH16
GH17
GH18
GH19

1

2

Clone name

Type

Zone

Country

Santo Amaro
B.D105
B.D130
B.D152
B.D188
B.D219
B.D247
B.D261
B.D286
B.D317
B.D351
B.D384
B.D41
B.D461
B.D529
B.D627
B.D686
B.D717
B.D761
B.D786
B.D862
B.D87
B.D948
B.RX
BDB.BR
BDCOQ
BDPRI
P1
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
P2
P20
P3
P4
P5
P6
P7
P8

I
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS

NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
BR
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH
GH

67

Weevil
3
response
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

Table A-1 (cont’d)
Local
Clone name
Code
GH20
P9
KE01
Kakamega
KE02
Oguroiwe
KE03
Kemb10
KE04
Musinyamu
KE05
KSP20
KE06
Ex-Mukumweivi
KE07
Bungoma
KE08
Nyathiodiavo
KE09
Sinia 2
KE10
Nyawo
KE11
Wera
KE12
Mwavuli
KE13
SPK013
MZ01
Rainha
MZ02
Zambezi
IC11
TIS 9101
IC12
Breeding line
PE01
Zapallo
PE02
Jonathan
PE03
Paramonguino NPPE04
CIP401055
47
PE05
CIP401104
PE06
CIP401148
PE07
CIP401205
PE08
CIP401208
PE09
CIP401215
PE10
CIP401217
PE11
CIP401243
PE12
CIP401248
PE13
CIP401255
PE14
CIP420697
PE15
CIP420720
PE16
CIP420746
PE17
CIP420819
PE18
CIP421111
PE19
CIP422527
PE20
CIP422649
PE21
CIP441724
RW01
Mugande
RW02
Caceaperdo
TZ01
Mayai
TZ02
Carrot Dar
TZ03
Jeshi
TZ04
Kilimanjaro
TZ05
Isaka la basihani
TZ06
Damu ya Mzee
TZ07
Budagara
TZ08
Simama
TZ09
Chigambilenyoko

1

2

Type

Zone

Country

DS
I
I
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
I
I
I
DS
I
I
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
I
DS
I
I
DS
DS
DS
DS
DS
DS
DS

NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

GH
KE
KE
KE
KE
KE
KE
KE
KE
KE
KE
KE
KE
KE
MZ
MZ
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
PE
RW
RW
TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ

68

Weevil
3
response
NA
Mild
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA

Table A-1 (cont’d)
Local
Clone name
Code
TZ11
Umeme
TZ12
Sinia
TZ13
Kituli
TZ14
Polista
TZ15
Changarawe local
TZ16
Shangazi
TZ17
Manjano
TZ18
Athuman
TZ19
Kata Mkononi
TZ20
Mbilimbi
TZ21
Carrot C
IC01
NASPOT 1
IC02
NASPOT 2
IC03
NASPOT 3
IC04
NASPOT 4
IC05
NASPOT 5
IC06
NASPOT 6
IC07
NASPOT 8
IC08
NASPOT 9
IC09
NASPOT 10
IC10
NASPOT 11
UG01
ARA228
UG02
Andinyaku
UG03
Tedolo Kereni
UG04
APA379
UG05
ARA222
UG06
Imatojony
UG07
Tutwech
UG08
Twongwengo
UG09
APA323
UG10
Oleke
UG100
Kibogo
UG101
Kitambi
UG102
BSH762
UG103
Kwezi kumwe
UG104
Mugiga
UG105
MSD480
UG106
BSH777
UG107
Kikuyu
UG108
Rashid
UG109
Old Kawogo
UG11
Tanzania
UG110
Kibanda
UG111
Kahungezi
UG112
HMA496
UG113
Karebe
UG114
Suwedi
UG115F
Kakobe
UG116F
SPE
UG117F
Dimbuka Bukulula

1

2

Type

Zone

Country

DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
DS
RC
RC
RC
RC
RC
RC
RC
RC
RC
RC
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
RC
FV
FV
FV
FV
FV
FV
RC
RC

NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Improved
Northern
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Northern
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Western
Ug
Central Ug
Ug
Central Ug
Central Ug

TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ
TZ
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG

69

Weevil
3
response
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Susceptible
NA
NA
NA
NA
NA
Susceptible
NA
NA
NA
Moderate
Moderate
Mild
NA
NA
NA
NA
NA
NA
Moderate
Moderate
NA
NA
NA
NA
NA
NA
NA
NA
Moderate
Susceptible
NA
Mild
NA
NA
NA
NA
NA
Susceptible

Table A-1 (cont’d)
Local
Clone name
Code
UG119F
Busuku
UG12
LIR288
UG120F
Baganyobekere
UG122F
Sukali
UG123
KBL625
UG124
KBL620
UG125
Makara
UG126
Kamamanzi
UG127
Africare
UG128
Mulelabana
UG129
Burundi
UG130
Mukazi
UG131
KBL639
UG132
KBL642
UG133
Kijambire
UG134
Kimotoka
UG135
Kashusha
UG137
Mpologoma
UG139
Nasoga
UG13F
Koromojo Red
UG14
Koromojo
UG140
KBL631
UG15
Ana Moito
UG16F
Lira Lira Red
UG17F
Otada
UG18F
Pam dero
UG19
Wagaborige
UG20
Mary
UG21
Shock
UG22
Tengerere
UG23
KML941
UG24
KML885
UG25
KML919
UG26F
Kampala Red
UG27
IGA1003
UG28
Abuket 1
UG29
Opira
UG30
Araka White
UG31
Ngujja Uganda
UG32
Uganda mali
UG33
Gunyombekere
UG34
Muyambi
UG35
Bunduguza empyaka
UG36
SRT32
UG37F
Silimu
UG38
Muiruki
UG39
Duduma
UG40
Mukoma
UG41
SRT37
UG42
Malagalya

1

2

Type

Zone

Country

FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
Fv
FV
FV
FV
FV
RC
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV

Central Ug
Northern
Central Ug
Ug
Central Ug
Southern
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Southern
Ug
Central Ug
Ug
Southern
Western
Ug
Eastern Ug
Ug
Northern
Northern
Ug
Southern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Northern
Ug
Eastern Ug
Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug

UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG

70

Weevil
3
response
NA
NA
NA
NA
Mild
Mild
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Susceptible
Susceptible
NA
NA
Mild
Susceptible
NA
NA
Mild
Moderate
NA
NA
NA
NA
Susceptible
NA
NA
NA
NA
NA
NA
NA
NA
Mild
NA
NA
NA
NA
NA
NA
NA

Table A-1 (cont’d)
Local
Clone name
Code
UG44
Araka Red
UG45
KML955
UG46
Anyumel
UG47
Osapat
UG48
Tororo 3
UG49
Kigaire
UG50
Ejumula
UG51
PAL148
UG52
KMI61
UG53F
Kassim
UG54
PAL98
UG55
Igang Uganda
UG56
Sudan
UG57
PAL158
UG58
Opade
UG59
Esamiat
UG60
Ikala 1
UG61
Kala 2
UG62F
Night
UG63F
Ateseke
UG64F
Mbale yellow
UG65F
Mbale white
UG66
Bwanjule
UG66F1
Bwanjule white
UG66F2
Bwanjule red
UG67
RAK835
UG68
RAK786
UG69
Nantongo
UG70
Kyebandula-omupya
UG71
MPG1128
UG72
Bitambi
UG73
MKN1169
UG74
Sowola
UG75
Dimbuka
UG76
New Kawogo
UG77
Dduka enzala
UG78
MPG1122
UG79
MSK1094
UG80
Bikiramaria
UG81
Mpambire
UG82F
Entebbe
UG83F
Sula oluti
UG84F
Mbizimbu family
UG85
Nylon
UG86F
Ndikiryanomwami
UG87F
Kifuta
UG88F
Orange Musoke
UG89
MBR536 Karebe
UG90
Old Silk
UG91
Malagalia

1

2

Type

Zone

Country

FV
FV
FV
FV
RC
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
RC
FV
RC
FV
FV
FV
FV
FV
FV
FV
RC
FV
RC
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV

Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Eastern Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Improved
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Central Ug
Western
Western
Ug
Western
Ug
Ug

UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG
UG

71

Weevil
3
response
NA
NA
NA
Mild
NA
Mild
Susceptible
NA
NA
Mild
NA
NA
NA
NA
Moderate
Susceptible
Mild
Susceptible
Moderate
NA
NA
NA
NA
NA
NA
NA
NA
NA
Mild
NA
NA
NA
Moderate
Susceptible
Moderate
NA
NA
NA
NA
NA
NA
Moderate
Moderate
NA
NA
NA
NA
NA
NA
Susceptible

Table A-1 (cont’d)
Weevil
Local
1
2
Clone name
Zone
3
Type
Country
Code
response
UG93
Rwabuganda
FV
Western
UG
NA
UG94
Dimbuka Omupya
RC
Western
UG
Susceptible
Ug
UG95
KRE724
FV
Western
UG
NA
Ug
UG96
Kasoga
FV
Western
UG
NA
Ug
UG97
LUW1254 Silk
FV
Western
UG
Mild
Ug
UG98
Kwezi kumu
FV
Western
UG
NA
Ug
UG99
LUW1269
FV
Western
UG
NA
Ug
USA01
Jewel
I
NA
US
NA
Ug
USA02
Beauregard
I
NA
US
NA
USA03
Resisto
I
NA
US
NA
USA04
Tenian USA
DS
NA
US
NA
1
Clone types: FV- Uganda’s farmers’ variety; I- introductions to Uganda; RC- Uganda’s
released cultivar; DS- different source
2
Original source: BR- Brazil; GH- Ghana; KE- Kenya; MZ- Mozambique; PE- Peru;
RW- Rwanda; TZ- Tanzania; UG- Uganda; and US – United States of America
3
Source of weevil response data: Yada 2009; Muyinza et al. 2012. NA means that the
response is not known to this research team.

72

LITERATURE CITED

73

LITERATURE CITED

Austin D.F. and Huamán Z. 1996.A synopsis of Ipomoea (Convolvulaceae) in the
Americas. Taxon 45(1):3-38.
Austin, D.F. 1977. Hybrid polyploids in Ipomoea section of Batatas. Heredity 68: 259260.
Buteler, M.I., Jarret, R.L. and LaBonte, D.R. 1999. Sequence characterization of
microsatellites in diploid and polyploid Ipomoea. Theoretic and Applied Genetics 99:
123-132.
CIP, 2008.Weevil resistant sweetpotato through biotechnology.Sweetpotato Action for
Security and Health in Africa.International Potato Center.
Connolly, A.G., Godwin, I.D., Cooper, M. and Delacy, I.H. 1994. Interpretation of
randomamplified polymorphic DNA marker data for fingerprinting sweetpotato [Ipomoea
batatas (L.) Lam] genotypes. Theoretic and Applied Genetics 88: 332–336.
Doyle, J. J. and Doyle, J. L. 1990. Isolation of plant DNA from fresh tissue. Focus 12:
13-15.
Earl, Dent A. and vonHoldt, Bridgett M., 2012. STRUCTURE HARVESTER: a website
and program for visualizing STRUCTURE output and implementing the Evanno method.
Conservation Genetics Resources 4 (2): 359-361
Elameen A., Fjellheim S., Larsen A., Rognli A.O., Sundheim L., Msolla S., Masumba E.,
Mtunda K. and Klemsdal S.S., 2008. Analysis of genetic diversity in a sweetpotato
(Ipomoea batatas L.) germplasm collection from Tanzania as revealed by AFLP. Genet
Resour Crop Evol, 55:397–408
Esselink GD, Nybom H, Vosman B (2004) Assignment of allelic configuration in
polyploids using the MAC-PR (microsatellite DNA allele counting-peak ratios) method.
Theoretical and Applied Genetics, 109, 402–408.
Evanno et al., 2005.Detecting the number of clusters of individuals using the software
STRUCTURE: a simulation study. Molecular Ecology 14: 2611 - 2620
FAO, 2007a. Report of Panel of Eminent Experts on Ethics in Food and Agriculture.
Food and Agriculture Organization of United Nations, Rome. November 2007.
FAO, 2007b.The state of food insecurity in the world 2003. Food and Agriculture
Organization, Rome.
Gao M, Ashu GM, Lakeisha Stewart, Akwe WA, Victor Njiti, Sandra Barnes (2011). Wx
intron variations support an allohexaploid origin of the sweetpotato [Ipomoea batatas (L.)
Lam]. Euphytica, 177: 111–133.

74

García-Verdugo C., M. F. Fay, C. Granado-Yela, R . Rubio De Casas, L. Balaguer, G.
Besnard And P. Vargas, 2009. Genetic diversity and differentiation processes in the
ploidy series of Olea europaea L.: a multiscale approach from subspecies to insular
populations. Molecular Ecology 18: 454–467
Gichuki, S.T., Berenyi, M., Zhang, D., Herman, M., Schmidt, J., Glossl, J and Burgh, K.
2003.Genetic diversity in sweetpotato [Ipomoea batatas (L.) Lam.] in relationship to
geographic sources as assessed with RAPD markers. Genet.Resources and Crop Evol. 50:
429-437.
Gichuru, V., V. Aritua, G.W. Lubega, R. Edema, E. Adipala, and P.R. Rubaihayo,
2006.A preliminary analysis of diversity among East African sweetpotato landraces using
morphological and simple sequence repeats (SSR) markers. Acta Hort. (ISHS) 703:159164.
He, G., C.S. Prakash and R.L. Jarret. 1995. Analysis of genetic diversity in a sweetpotato
(Ipomoea batatas) germplasm collection using DNA amplification fingerprinting.
Genome 56:25-29.
He, X., Liu, Q., Ishiki, K., Zhai, H. and Wang, Y. 2006. Genetic diversity and genetic
relationships among Chinese sweetpotato landraces revealed by RAPD and AFLP
markers. Breeding Science 56: 201-207.
Höfer M. and B.M. Reed 2012. Cryopreservation of strawberry genetic resources in
Germany. ISHS Acta Horticulturae 918: XXVIII International Horticultural Congress on
Science and Horticulture for People (IHC2010): III International Symposium on Plant
Genetic Resources
Holsinger K.E and Bruce S.W, 2009. "Genetics in geographically structured populations:
defining, estimating and interpreting FST". Nat Rev Genet 10 (9): 639–650.
Hu J, Nakatani M, MizunoKand Fujimura T, 2004. Development and characterization of
microsatellite markers in sweetpotato. Breeding Science 54:177-188.
Hu, J., Nakatani, M., Lalusin, A.G., Kuranouch, T and Fujimura, T. 2003. Genetic
analysis of sweetpotato and its wild relatives using inter-simple sequence repeat (ISSRs).
Breeding Science 53: 297-304.
Hwang, S.Y., Tseng, Y.T. and Lo, H.F. 2002. Application of simple sequence repeats in
determining the genetic relationships of cultivars used in sweetpotato polycross breeding
in Taiwan. Sci. Hort. 93: 215-224.
Jaccard, P. 1908. Nouvelles recherches sur la distribution florale. (In French.) Bull. Soc.
Vaud. Sci. Nat. 44:223–270.
Jakobsson and Rosenberg 2007. CLUMPP: a cluster matching and permutation program
for dealing with label switching and multimodality in analysis of population structure.
Bioinformatics 23(14): 1801-1806.
Jarret, R.L. and Bowen, N. 1994. Simple sequence repeats (SSRs) for sweetpotato
characterization. Plant Genetic Resources Newsletter 100: 9-11.
75

Jørgensen MH, Carlsen T, Skrede I, Elven R. 2008. Microsatellites resolve the taxonomy
of the polyploidy Cardamine digitata aggregate (Brassicaceae). Taxon 57: 882–892.
Karuri, H.W., E.M. Ateka, R. Amata, A.B. Nyende, A.W.T. Muigai, E. Mwasame and
S.T. Gichuki, 2010. Evaluating diversity among Kenyan sweetpotato genotypes using
morphological and SSR markers. Int. J. Agric. Biol., 12: 33–38
Keller E.R.J., A. Senula, C. Zanke, M. Grübe , A. Kaczmarczyk 2010. Cryopreservation
and in vitro culture - state of the art as conservation strategy for genebanks. ISHS Acta
Horticulturae 918: XXVIII International Horticultural Congress on Science and
Horticulture for People (IHC2010): III International Symposium on Plant Genetic
Resources
Kloda JM, Dean PDG, Maddren C, MacDonald DW, Mayes S. 2008. Using principal
component analysis to compare genetic diversity across polyploidy levels within plant
complexes: an example from British Restharrows (Ononis spinosa and Ononis repens).
Heredity 100: 253–260.
Kobayashi M., 1983. The Ipomoea trifida complex closely related to sweet potato, in:
S.F. Shideler, H. Rincon, (Eds.), Proceedings of the 6th Symposium of the International
Society of Tropical Root Crops, Lima, Peru, 21–26 February, 1983, International Potato
Center, Lima, Peru, pp.561–568
Krishnan P.N., S. W. Decruse and R. K. Radha 2011. Conservation of medicinal plants
of Western Ghats, India and its sustainable utilization through in vitro technology. In
Vitro Cell.Dev.Biol.—Plant 47:110–122
Lebot V 2010. Root and Tuber Crops: Handbook of Plant Breeding, 7: 97-125
Legendre P, Legendre L. Amsterdam: Elsevier Science; 1998. Numerical Ecology, 2nd
edn.
Mignouna HD, Abang MM, Fagbemi SA, 2003. A comparative assessment of molecular
marker assays (AFLP, RAPD and SSR) for white yam (Dioscorea rotundata Poir.)
germplasm characterization. Annals of Applied Biology 142: 269-276.
Muyinza H., H.L. Talwana, R.O.M. Mwanga and P.C. Stevenson, 2012. Sweetpotato
weevil (Cylas spp.) resistance in African sweetpotato germplasm. International Journal
of Pest Management 58(1), 78-81
Mwanga R.O.M., B. Odongo, C. Ocitti p’obwoya, R.W. Gibson, N. Smith, E. Carey,
2003. Release of six (“NASPOT 1” to “NASPOT 6”) sweetpotato cultivars in Uganda.
HortScience 38: 475-476.
Mwanga R.O.M., B. Odongo, C.N. Niringiye, A. Alajo, B. Kigozi, R. Makumbi, et al.,
2009.“NASPOT 7”, “NASPOT 8”, “NASPOT 9 O”, “NASPOT 10 O”, and “DimbukaBukulula” sweetpotato. HortScience 44: 828-832.
Mwanga R.O.M., C. Niringiye, A. Alajo, B. Kigozi, J. Namakula, I. Mpembe, et al.,
2011. “NAPOT 11”, a sweetpotato cultivar bred by a participatory plant-breeding
approach in Uganda. HortScience 46: 317-321.
76

Mwanga R.O.M., C. Niringiye, A. Alajo, B. Kigozi, J. Namakula, I. Mpembe, et al.,
2011. “NAPOT 11”, a sweetpotato cultivar bred by a participatory plant-breeding
approach in Uganda. HortScience 46: 317-321.
Mwanga, R.O.M., Odongo, B., Niringiye, C. and Alajo, A., 2007. Release of Two
Orange-fleshed Sweetpotato Cultivars, ‘SPK004’ (‘Kakamega’) and ‘Ejumula’, in
Uganda. HortScience 42(7): 1728–1730.
Nei, M. 1973. Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci.
USA70: 3321-3323.
Nishiyama, I. 1971. The origin of the sweetpotato plant. In: Barrau J. (ed.) Plants and the
migrations of Pacific peoples. Bihop Mus. Press Honolulu. Pp. 119-128.
Otoo E, Akromah R, Kolesnikova-Allen M, Asiedu R, 2009. Delinearation of pona
complex of yam in Ghana using SSR markers. International Journal of Genetics and
Molecular Biology 1: 006-016.
Peakall R, Smouse PE. 2006. GENALEX 6: genetic analysis in Excel. Population genetic
software for teaching and research. Molecular Ecology Notes 6: 288–295.
Pemberton J.M, Slate J, Bancroft D.R, and Barrett J.A, 1995. Nonamplifying alleles at
microsatellite loci: a caution for parentage and population studies. Mol Ecol 4:249–252.
Perrie LR, Shepherd LD, De Lange PJ, Brownsey PJ. 2010. Parallel polyploid speciation:
distinct sympatric genepools of recurrently derived allo-octoploid Asplenium ferns.
Molecular Ecology 19: 2916–2932.
Prakash, C.S. and G. He, 1996. DNA marker-based genetic relatedness in United States
sweetpotato cultivars. J. American Hortic. Sci., 121: 1059–1062
Prasanth, V.P. and Chandra, S. 1997. ALS-Binary. A program for converting allele size in
microsatellite markers into 0-1 (binary) data. ICRISAT_Hyderabad, India.
Prasanth, V.P., Chandra, S., Jayashree, B. and Hoisington, D. 1997. AlleloBin. A
software for allele binning of microsatellite markers based on the algorithms of Idury and
Cardon. ICRISAT_Hyderabad, India.
Pritchard JK, Stephens M, Donnelly P. 2000. Inference of population structure from
multilocus genotype data. Genetics 155: 945–959.
Pritchard JK, Wen X, Falush D. 2009. Documentation for structure software. Version 2.3.
Available at: http://pritch.bsd.uchicago.edu/structure.html
Rao R. and Campilan D (eds.) 2002. Exploring the complementarities of ex situ and in
situ conservation strategies for Asian sweetpotato genetic resources. Proceedings of the
3rd International Workshop of the Asian Network of Sweetpotato Genetic Resources
(ANSWER). IPGRI-APO, Serdang, Malaysia, 191p.
Reeds M.B, Dumet D., Denoma J.M., Benson E.E, 2001. Validation of cryopreservation
protocols for plant germplasm conservation: a pilot study using Ribes. Biodiversity and
Conservation 10: 939–949
77

Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing
phylogenetic trees. Mol Biol Evol 4: 406–425
Sampson JF, Byrne M (2012). Genetic diversity and multiple origins of polyploid
Atriplex nummularia Lindl. (Chenopodiaceae). Biol J Linn Soc 105: 218–230.
Schafleitner R, Tincopa LR, Palomino O, Rossel G, Robles RF, Alagon R, Rivera C,
Quispe C, Rojas L, Pacheco JA, 2010. A sweetpotato gene index established by de novo
assembly of pyrosequencing and Sanger sequences and mining for gene-based
microsatellite markers. BMC Genomics, 11:604.
Schlotterer, C., and Tautz, D. 1992. Slippage synthesis of simple sequence DNA. Nucleic
Acid Research 20: 211-215.
Shiotani I, Kawase T, 1987. Synthetic hexaploids derived from wild species related in
Ipomoea trifida. Jpn J Breed 39: 57–66.
Smith JSC, Smith OS, 1992. Fingerprinting crop varieties. Advanced Agronomy 47: 85140.
Sneath, P.H.A. and Sokal, R.R. 1973. Numerical Taxonomy. Freeman, San Francisco, pp.
573.
Srisuwan S., Sihachakr D, Siljak-Yakovlev S 2006. The origin and evolution of
sweetpotato (Ipomoea batatas Lam.) and its wild relatives through the cytogenetic
approaches. Plant Sci 171: 424–433.
Tseng, Y.T., Lo, H.F. and Hwang, S.Y. 2002. Genotyping and assessment of genetic
relationships in elite polycross breeding cultivars of sweetpotato in Taiwan based on
SAMPL polymorphisms. Botanical Bulletin of Academia Sinica 43: 99-105.
Tumwegamire S., R. Kapinga, P.R. Rubaihayo, Do.R. LaBonte, W.J.
Grüneberg, G.Burgos, T. zum Felde, R. Carpio, E. Pawelzik, and R.O.M. Mwanga 2011.
Evaluation of Dry Matter, Protein, Starch, Sucrose, β-carotene, Iron, Zinc, Calcium, and
Magnesium in East African Sweetpotato [Ipomoea batatas (L.) Lam] Germplasm.
HortScience, 46: 348-357
Weir, B.S. 1996. Genetic data analysis II. Methods for discrete population genetic data.
Sinauer Associates, Inc. Sunderland, Massachusetts. 445pp.
Woolfe, J.A. 1992. Sweetpotato, an untapped food resource.Cambridge University
Press,Cambridge, U.K.
Yada B., P. Tukamuhabwa, Villordon A., A. Alajo, and R.O.M. Mwanga 2010a.An
online database of sweetpotato germplasm collection in Uganda. Hortscience 45(1): 153–
153. 2010.
Yada B., P. Tukamuhabwa, Wanjala B, Kim D-J, Skilton R.A., A. Alajo, and R.O.M.
Mwanga 2010b. Characterization of Ugandan sweetpotato germplasm using fluorescent
labeled simple sequence repeat markers.Hortscience 45(2):225–230. 2010.

78

Yamamoto S., Fukui K, Hirah D, Niino T., 2012. 74. Cryopreservation of in vitro shoot
tips of potato varieties by V- Cryo-plate method. Cryobiology, 65(3), 363
Zhang D.P., Golmirzaie A., Cipriani G., Panta A., Ghislain M., Smit N., Rety I., and
Michaud D. 1996. Developing weevil resistance in sweetpotato with genetic
transformation. International Potato Center (CIP) Program Report 95-96.
Zhang D.P., Rossel G., Kriegner A. and Hijmans R. 2004. AFLP assessment of diversity
in sweetpotato from Latin America and the Pacific region: Its implications on the
dispersal of the crop. Genet Resour Crop Evol 51:115-120. Zhang DP, Rossel G,
Kriegner A and Hijmans R (2004) AFLP assessment of diversity in sweetpotato from
Latin America and the Pacific region: Its implications on the dispersal of the crop. Genet
Resour Crop Evol 51:115-120.
Zhang, D.P., D. Carbajulca, G. Rossel, S. Milla, C. Herrera, and M. Ghislain. 2000.
Microsatellite Analysis of Genetic Diversity in Sweetpotato Cultivars from Latin
America. CIP Program Report 1999-2000.

79

CHAPTER 2
ASSESSMENT OF ON-FARM SWEETPOTATO DIVERSITY, AND FACTORS
INFLUENCING THE LEVEL OF FARMERS’ CROP DIVERSITY:
IMPLICATIONS FOR CONSERVATION

Introduction
In situ conservation of crop diversity is acknowledged as a complementary
method to circumvent “inevitable losses” likely to be experience by ex situ conservation,
particularly for vegetatively propagated crops like sweetpotato (Rao and Campilan, 2002;
Lebot, 2010). Crop diversity is the basis for adaptation to the heterogenous and ever
changing environment, and generation of resistance to evolving pests and diseases. It is
also widely accepted that having high crop diversity enables agricultural systems to
maintain productivity over a wide range of environmental conditions (Carpenter, 2011;
Thomas et al., 2011).
There is increasing concern about the loss of crop biodiversity due to agricultural
intensification that is causing displacement of landraces by a few improved cultivars
(Brush 1991, Harlan, 1992; Ammann, 2005; Carpenter, 2011). Loss of landrace diversity
may be a valid concern in centers of origin and centers of diversity where varieties have
developed high adaptability to the local environment (Bellon, 1996), as well as in
developing countries where farmers rely on crop diversity for food security (FAO, 2007).
However, there is evidence to show that while varietal substitution is a common practice
in such areas, complete displacement of landraces is not likely because farmers tend to
maintain a diverse pool of varieties in their field even when they adopt new varieties or
cultivars (Brush, 1991; Perales et al., 2003; Gepts and Papa, 2003; Bellon and Berthaud,
2004). In this discussion, we will use the term ‘variety’ to mean a product of natural and
human selection that occurs in the farmers’ field, and the term ‘cultivar’ to mean a
80

product of formal crop improvement or an advanced variety released by the national
variety release committee.
Sweetpotato is a highly important food crop in Eastern and Southern Africa. It
ranks third in terms of weight harvested in seven Eastern and Central African countries
including Uganda; and fourth in six Southern African countries (CIP, 2008). Uganda is
the third largest producer of sweetpotato in Africa (contributing 6.6% and 2.4% of the
total global area and output, respectively), after Tanzania (8.7% and 3.4%) and Nigeria
(11.8% and 2.6%) [FAO, 2011]. The importance of sweetpotato as a staple food in
Uganda varies by region: the largest area allocated to sweetpotato production, and the
highest production occurs in the eastern region, although substantial production is also
reported in the northern and central regions (Table 3.1).

The importance of sweetpotato as a food security crop is increasing rapidly in
regions where there is virus epidemic of cassava, and rapid spread of the parasitic weed
(Striga hermonthica) that attacks maize and other cereal crops (Ndolo et al., 2007).
Sweetpotato also plays an important role in providing household food security because it
stores well in the soil as a famine reserve where there is no weevil infestation, and it can
withstand extreme weather conditions and marginal soils (Bashaasha et al., 1995). More
recently it has been promoted as a source of pro-Vitamin A for malnourished children
(Osiru et al., 2007). Price fluctuation for cash crops, population increase, civil wars and
changes in weather patterns are also influencing sweetpotato production in different
regions (Abidin, 2004). For example, many farmers in Kumi district are growing
sweetpotato as a cash crop

81

Table 3-1 Crop area and production by region for selected food crops in Uganda for 2008/09 (Source: UBOS, 2010)
Crop

Central
Area
(Ha)

Maize
189,135
Banana
326,082
Cassava
127,788
Sweetpotato 98,054
Sorghum
Finger millet
Rice
Irish potato

2,261
5,832
2,637
4,798

Production
(MT)

Eastern
Area
(Ha)

Northern
Western
Total
Production Area
Production Area
Production Area
(MT)
(Ha)
(MT)
(Ha)
(MT)
(Ha)

449,859
1,039,837
409,812
312,402

388,762
69,504
342,387
159,948

1,108,554 247,780
342,234
9,195
1,061,186 269,886
847,140 121,681

2,678
13,734
2,173
13,209

101,645
86,911
36,033
1,271

133,313 249,330
106,838 105,656
128,195 25,912
4,624
594

82

305,798 188,583
31,626 511,096
983,124 131,328
486,295 60,573
177,088
78,572
43,719
1,311

46,016
51,588
10,504
26,096

Production
(MT)

497,745
2,883,648
440,189
172,932

1,014,260
915,877
871,389
440,256

2,361,956
4,297,345
2,894,311
1,818,769

62,716
77,784
16,649
135,210

399,252
249,987
75,086
32,759

375,795
276,928
190,736
154,354

because of the decline in market prices for cotton (Abidin, 2004). In most regions, the
crop is mostly grown by resource-poor small-scale farmers on plots less than 0.5 hectare
(Bashaasha et al., 1995).
Sweetpotato originated from Central and South America and it was spread to East
Africa by Portuguese explorers in the 16th century (Zhang et al., 2004; Loebenstein and
Thottappilly, 2009). Over time, the region has accumulated a large number of landraces
that are adapted to the local growing conditions and native pests and diseases. Based on
this large number of adapted landraces, East Africa is considered a secondary centre of
diversity for sweetpotato (Huamán and Zhang, 1997; Mwanga et al., 2001; Gichuki et al.,
2003). Within Uganda, there is a large landrace pool (Yada et al., 2010) and most
sweetpotato production in the country relies on this landrace diversity. As part of the
global effort to preserve crop diversity, the national sweetpotato program collected 433
sweetpotato accessions in 1986, which were maintained at two institutions: the National
Semi-Arid Resources Research Institute (NaSARRI) in Serere, and National Crops
Resources Research Institute (NaCRRI) in Namulonge. However, since the collections
were maintained in open fields, substantial amounts were lost due to viral infection
(Mwanga and Otim-Nape, 1992). In 2005, the Uganda national programme made another
collection of over 1300 accessions, which were maintained at NaCRRI (Yada et al.,
2010). However, during sample selection for molecular characterization of the Uganda’s
germplasm it was observed that over 40% of this collection had been lost (Chapter 1)
likely due to caterpillar and viral incidences at Namulonge (Agnes Alajo, personal
communication). The challenges associated with maintaining ex situ collections indicate
the importance of in situ conservation of crop diversity for sweetpotato germplasm
(Khoury et al., 2010).

83

Individual farmers and farming communities play an important role in conserving
sweetpotato diversity (Rao and Campilan, 2002; Lebot, 2010). Farmers are not only
custodians of varietal diversity, but through their decisions and crop management systems
they influence the level of diversity at the local and country level. Farmers make
decisions based on (1) goals, (2) resources, and (3) constraints. Their decisions are also
influenced by local knowledge or information provided with information currently
provided by technical institutions and extension services. Some knowledge is handed
from previous generations, from shared experiences by members within the community or
from other communities, and from information provided by technical institutions or
extension services. This knowledge influences various aspects of farmers’ decisions
including: which morphological characteristics to consider during variety selection; the
location and size of the area allocated for cultivation; the source of the planting material;
and the crop management practices (Jarvis and Hodgkin, 2000). Gaining a good
understanding of farmers’ local knowledge and how it is used in the decision making
process is a necessary step for prioritization of crop improvement to assure that new
varieties meet farmers needs and to facilitate preservation of the landraces.
Over two decades ago (between 1989 and 1992), a farmer’s survey study was
conducted in four of Uganda’s agro-ecological zones (AEZs) to establish the role of
sweetpotato in the farming and food system of Uganda (Bashasha et al., 1995). The study
showed that sweetpotato was one of the top three food crops in northern region/north
western savannah grasslands AEZ (zone III), the eastern region/Kyoga plains AEZ (zone
V) and the central region/Lake Victoria Crescent AEZ (zone VI) in Uganda (Figure 1).
The average number of varieties grown by each farmer was five. The eastern region has
the highest production per unit area compared to other regions (Table 1). The high
average yield per unit area may be attributed to the region having large sections of sandy
84

loam soil type (Uganda Government, 1967). Sweetpotato requires loose soil, which
allows the storage roots to enlarge with less hindrance (Dantata et al., 2010).
Farmers depended on traditional varieties for sweetpotato production because they
did not have access to improved cultivars. The first lot of improved cultivars was released
by the national sweetpotato program in 1995 (Mwanga et al., 1995). Other studies
conducted in 1999 (Abidin et al., 2005) and 2005 (Yada, 2009) assessed farmers’
knowledge of the crop diversity and farming system during the process of germplasm
collection. It was observed that farmers were able to describe the key morphological
characteristics of each variety such as time to maturity, resistance to pests and diseases,
tolerance to drought, soil type suitability and culinary quality. However, none of these
studies has looked at how adoption of new varieties/cultivars has influenced farmers’
decisions to maintain or discard current varieties/cultivars.
Exchange of planting material by farmers is reported to also influence on-farm
crop diversity (Thomas et al., 2011). The quantity and frequency of planting material
exchanges plus the effects of genetic drift and gene flow, influence the genetic structure
and adaptation of the populations within a given location (Hodgkin et al., 2007; Whitlock
2003). The prior studies (Bashasha et al., 1995; Abidin, 2004; Yada 2009) did not assess
the planting material exchange practices by farmers and its likely impact on crop
diversity.
Gene flow is an evolutionary force that may cause an increase or decrease in
genetic diversity of a species depending on its interaction with other forces like genetic
drift and selection. Impact of gene flow also depends on the crop’s mating system and
level of hybridization and introgression. Sweetpotato has a mating system that encourages
gene flow: cross pollination combined with self-incompatibilty. However, Mwanga et al.
(2007) indicated that development of hybrids from crossing between cultivars is
85

extremely difficult in sweetpotato due to multiple compatibility barriers, low seed set and
poor germination. Other literature indicates that volunteer plants from seed resulting from
cross pollination are common, and may explain part of the high level of crop diversity
observed in farmers’ fields in Uganda (Abidin, 2004). Sweetpotato farmers tend to grow
many varieties in their fields and they take volunteer plants from previous seasons as
sources of material for the next season (Smit, 1997; Abidin, 2004). This suggests that new
farmers’ varieties may arise, which may persist in the gene pool due to intentional or
unintentional selection. No study has explored farmers’ knowledge of potential hybrids,
and how such hybrids are managed in sweetpotato production.
This study assessed the varietal diversity in farmers’ fields in Uganda,
documented the local knowledge related to this germplasm, and examined factors
influencing decisions about whether to maintain, incorporate and or discard varieties.
Specifically, we assessed the level of crop diversity at farm, district and regional level.
We also identified farmers’ criteria for adoption and maintenance of sweetpotato
varieties; explored the crop management system; assessed farmers’ knowledge and
management of volunteer plants and potential hybrids; and investigated how various
sociological factors including planting material exchange practices, source of
information, land ownership and size, and market access affect diversity of sweetpotato
varieties grown by farmers in Uganda.

86

Materials and Methods
Description of the study area
Ethno-botanical surveys were conducted in three agro-ecological zones (AEZs)
selected based on area allocated for sweetpotato production (Table 3.1, UBOS, 2010);
and whether it was ranked among the top three most important food crops (Yada, 2009).
The AEZs were demarcated by the Ministry of Agriculture, Animal Industry and
Fisheries based on vegetation and farming systems (Figure 3.1A). The study areas
include: zone III, the north western savannah grasslands, referred to as the northern
region in this work; zone V, the Kyoga plains, referred to as the eastern region; and zone
VI, the Lake Victoria Crescent, referred to as the central region. Table 3.2 provides
detailed description of the AEZs including the altitude, weather patterns, soil type, and
farming systems.

Site selection
Districts within each agro-ecological zone were selected based on their ranking of
sweetpotato as a staple crop (Yada, 2009) and proximity from the national research
institutes (Figure 3.1B). In the central region, Luwero and Mukono are close to the
National Crop Resources Research Institute while Rakai is distant. In the eastern region,
Soroti is close to the National Semi-Arid Resources Research Institute while Kumi is
distant. Bukedea (also found in the eastern region) was added later to the study after a
request by the regulators to assess the impact of introducing orange fleshed sweetpotato
(OFSP) on on-farm diversity; it is distant from the research institute. In the northern
region, Lira is close to Ngetta Zonal Agricultural Research and Development Institute
while Apac is distant.

87

Table 3-2 Detailed descriptions of the agro-ecological zone (AEZs)/regions in Uganda: altitude, weather patterns, soil characteristics, and
farming systems.
AEZ/ region

Altitude
1

Climate

m.a.s.l
Lake Victoria
1194-1384 Average rainfall of 1,200 to 1,450 mm
with two rainy seasons: March to May and
Crescent/
August to November. Temperature ranges
central region
from 15 – 30 °C.
Kyoga plains/
1133-1255 Average rainfall range of 1215 mm -1328
mm with two rainy seasons: March to
eastern region
May and August to November.
Temperature ranges from 15 – 32.5 °C.
Northwestern
1142-1215 Average rainfall range of 1340 mm –
1371mm.One rainy season, which lasts
savannah
over 7 months from April to about mid
grasslands/
November. One long dry season of about 4
northern region
months from mid-November to late March.
Temperature ranges from 15 - 25 °C.
1
Altitude for the districts surveyed.
2

Source: Uganda Government, 1967.

88

2

Vegetation

Cropping system

Sandy loam/
sandy clay
loams

Tall grasslands
and forests

Banana-coffee

Sandy
sediments

Hyparrhenia and
short grasslands

Annual crops and
livestock

Alluvial
deposits/sandy
loams

Short grasslands

Drought tolerant
annuals, and
livestock with
communal grazing

Soil types

Legend
X- Highland Range
V- Kyoga plains
VI- Lake Victoria Cresent
I- Northern Eastern Drylands
II- NE Savannah Grasslands
III- NW Savannah Grasslands
IV- Para Savannah
VIII-Pastoral Rangelands
VII- W Savannah Grasslands
Figure 3-1 Map of Uganda showing: A) the different agro-ecological zones (Roman
numerals) *. The study areas are marked with a black circle1. The districts visited were
Apac and Lira districts in the northern region; Bukedea, Kumi, and Soroti districts in
eastern region; and Luwero, Mukono, and Rakai in central region.
*Source: Ministry of Agriculture, Animal Industry and Fisheries, Uganda

89

Survey methodology
A survey was conducted to assess the varietal diversity in farmers’ fields, and to
establish how farmers make decisions whether to maintain, incorporate or discard
varieties. Structured questionnaires were used to obtain information. A pre-test of the
questionnaire was performed with 20 respondents from Mukono and Luwero districts.
Based on the feedback from the pre-test, a final questionnaire was prepared (included as
Annex). Topics covered by the questionnaire included: personal information of the
respondent (age, gender), size and ownership of land cultivated with sweetpotato,
characteristics of varieties currently grown, varieties dropped and reasons for dropping,
criteria for adopting new varieties, proportion of land allocated to new verses old
varieties, criteria for maintaining old varieties, challenges with adopting/ maintaining
varieties, farmers’ variety exchange practices, common crop management practices and
challenges associated with crop management, and role of extension service in variety
selection.
Face to face interviews were conducted in seven districts between January to June
2012, and in Bukedea in December 2012. Conduct of the survey was done in
collaboration with the national sweetpotato program, which is part of the National Crop
Resources Research Institute located in Namulonge, Uganda. Households were the
sampling unit with a minimum of ten per district (Table 3-3). The survey targeted
households growing sweetpotato at the time so that we were able to evaluate the key
morphological traits on-site. A sweetpotato genetic diversity specialist from the national
program, and an extension worker and or farmers group leader were involved in the
survey to ensure that details of farmers’ descriptions of varieties/cultivars were captured
accurately, and act as interpreters wherever necessary.

90

Data were analyzed using the Statistical Package for Social Sciences (SPSS). Chisquared test was used to determine whether there was relationship between categorical
variables such as relationship between frequency of adoption of new varieties and
challenges associated with adopting new varieties. One-way analysis of variance
(ANOVA) was performed for continuous variables.

91

Results and Discussion
Respondents’ information
A total of 102 respondents participated in the study (Table 3.3). A higher
percentage of sweetpotato farmers in most districts, except Bukedea, were women. This
provides supporting evidence for prior reports indicating that sweetpotato production in
Uganda is mainly done by women (Mutuura et al., 1992; Bashasha et al., 1995). Some
farmers in Luwero, Rakai, Bukedea and Soroti are members of farmers’ groups that were
formed separately in each district.

Reasons for sweetpotato production
The main reasons given for sweetpotato production were food security or home
consumption, although a majority of the respondents also indicated that they sold surplus,
thus it was also a source of income (Figure 3.2A). In all regions, storage roots were
consumed fresh. Farmers used a piecemeal harvesting method where they harvest just
enough roots for immediate consumption and they “store” the rest of the potential harvest
in-ground. In addition, in eastern and northern Uganda, some roots were harvested and
converted to dried chips which are stored for longer periods (over 6 months) to ensure
food security, or pounded to form flour or meal, which is used for traditional brewing. In
northern and some parts of eastern Uganda, young leaves are consumed as leafy
vegetables while in all regions leaves and damaged roots are used as animal feed. Similar
observations were reported by the national agricultural census (Figure 3.2B, UBOS,
2010).

92

Table 3-3 Number of respondents, and percentage of respondents based on gender and
membership in farmers’ groups, in each district in Uganda.
Region

District

Central

Eastern

Northern
Mean
Total

Luwero
Mukono
Rakai

No. of farmers
interviewed
15
10
15

Women
(%)
87
80
71

Members of farmers’
groups (%)
26.7
0.0
33.3

Bukedea
Kumi

15
10

33
60

60.0
0.0

Soroti
Apac
Lira

15
10
12

74
50
75

13.3
0.0
0.0

13
102

66.3
67

19.6
20

93

80

A
Consumed

Sold

Both

Central
Northern

B
80

70

70

60

60

50

50

40

40

30

30

20

20

10

Eastern
Western

10

0
Consumed

Sold

Both

0
%

Sold

Consumed

Stored

Other uses

Figure 3-2 Uses of sweetpotato: A) provided by the key respondents; B) deduced from
data collected during the Uganda Agricultural Consensus (2010).

94

Land size and ownership cultivated with sweetpotato
The area under sweetpotato production in a given farm ranged from 0.1to 5
hectares. Bukedea had the largest average land size while Mukono had the smallest
(Table 3.4). The overall average land size for study area (0.43 Ha) is higher than the
national average of 0.20 Ha (UBOS, 2010). The average allocated land size for Luwero
district, and that for the northern region, observed in this study is the same as that
reported by Bashaha et al. (1995)., which is much lower than the average recorded in all
the other districts studied.
Brush (1995) and Alvarez et al. (2005) reported that land size allocated for crop
production increases with farmers’ age because as farmers become older, they acquire
more experience, resources, and prestige. However, this trend was not observed for
sweetpotato production in this study most likely due to the land ownership conditions,
which varied by district (Table 3.5). On average, 20% of the land used for sweetpotato
production was rented. Half of the respondents in Mukono were cultivating sweetpotato
on rented land because Mukono is very close to Kampala the capital city of Uganda. This
has resulted in high level of land fragmentation due to urbanization of a large segment of
the district. As a result farmers have to rent land for crop production in the rural parts of
the district. Lira had the second highest percentage of rented land used for sweetpotato
cultivation, and this too can be attributed to Lira municiplity, which is the second largest
business center in the northern region. Young farmers within these districts rented
relatively large landholding (average of one hectare) for commercial production of
sweetpotato because they have not yet acquired their own land.

95

Table 3-4 Number of respondents cultivating sweetpotato on the different land sizes in
each district in Uganda.
Area (Ha)

Northern
region
Luwero Mukono Rakai Bukedea Kumi Soroti Apac Lira Total
Central region

Less than 0.25
7
5
0.25 to 0.49
3
4
0.50 to 0.99
3
1
1.00 to 2.99
2
3.00 to 5.00
Average area
0.29
0.20
*** denotes significance at 0.1%.

Eastern region

2
3
1
7
0
6
6
6
3
6
2
0.37 1.09*** 0.31

3
2
6
3
1
0.50

4
5
1
0.27

3
2
4
1
0.45

28
29
30
12
3
0.43

Table 3-5 Land ownership (percentage of respondents renting land in each district in
Uganda)
Region

District

Central

Luwero
Mukono

No. of farmers
interviewed
15
10

Rakai
Bukedea
Kumi

15
15
10

11
13
20

Soroti
Apac

15
10

5
0

Lira

12
13
102

40
20.1
-

Eastern

Northern
Mean
Total

Land rented (%)
27
50

96

Varietal diversity on-farm and at regional level
Respondents indicated that most of the varieties they grow have been given
vernacular names based on source (the origin or person who introduced the variety), vigor
or yield performance, duration to maturity, and plant organ shapes and colors (particularly
roots or leaves). Some varieties/cultivars have names given at the source. Farmers grow
varieties that are presumably chance seedlings selected and retained by farmers (Bashasha
et al.,1995; Abidin et al., 2005), and released cultivars including superior farmers’
varieties. Some superior varieties were selected by farmers, but were later scientifically
evaluated and released by the Ugandan Plant Variety Release Committee for countrywide cultivation. These include Bwanjule, New Kawogo, Tanzania, Wagabolige, and
Dimbuka Bukulula (Mwanga et al., 2011) plus two orange fleshed varieties Kakamega
and Ejumula (Tumwegamire et al., 2011). Other released cultivars are newly bred
cultivars such as Sowola and NASPOT 1 to 11 (Mwanga et al., 2011). The released
cultivars were selected by breeders based on criteria of consumer acceptance, consistently
high yields and disease resistance across a wide variety of agro-ecological zones in
Uganda (Mwanga et al., 2011).
The number of varieties per household ranged from 1 to 12 (Table 3.6) with an
overall average of 6 varieties per household. There was a significant difference (P<0.05)
in number of varieties grown per household in Soroti and Apac compared to Bukedea and
Mukono. Although Soroti and Apac both had an average of seven varieties per
household, there were more varieties cultivated in Soroti (28) than in Apac (15). These
results suggest that Soroti district had the highest level of crop diversity. In some districts,
we recorded additional farmers’ varieties that were not reported in the previous studies
(Bashasha et al., 1995, Abidin et al., 2004; Yada 2009). These additional varieties could
be introductions from other areas, or a result of somatic mutation which are then later
selected by the farmers (Aldrich 1963, cited in Hakiza et al., 2000).

97

Table 3-6 Level of sweetpotato diversity at district level: number of farmers’ varieties and
released cultivars at district level, total number of varieties and cultivars, and the average
number and range of varieties grown per household in each district.

No. of Farmers'
Region
District farmers varieties
Central Luwero
15
13
Mukono
10
8
Rakai
15
15
Eastern Bukedea
15
6
Kumi
10
14
Soroti
15
23
Northern Apac
10
15
Lira
12
10

Total Average no. Range of
Released varieties/ of varieties varieties per
1
2
HH
cultivars cultivars per HH
7
7
1
4
2
5
0
3

1

20
15
16
10
16
28
15
13

5
3
6
4
6
7
7
5

1-8
1-6
2-8
2-6
3-8
1-12
5-8
2-8

Released cultivars include newly bred cultivars and farmers’ varieties with superior
desirable traits, which are promoted by the national program.
2
HH stands for Household

98

In most districts, farmers were cultivating traditional varieties together with the
released cultivars. The majority (74%) of the respondents planted traditional varieties
with or without released cultivars; and overall 78% of the diversity found in farmers’
fields was farmers’ varieties. No newly bred cultivars were reported by respondents in
Apac, while the majority of the farmers interviewed in Bukedea were growing
predominantly orange-fleshed cultivars (NASPOT 9 and 10, Kakamega and Ejumula)
[data not shown]. Farmers in Luwero, Mukono, Soroti and Lira districts, which are close
to the research institutes, tended to grow released cultivars more than respondents in
distant districts. This may suggest uneven distribution of released cultivars across the
various districts. This is supported by information provided by various farmers. For
example, in Bukedea, which is a district distant to the research institution found in eastern
Uganda, a substantial number of the respondents indicated that they wanted to grow a
newly bred cultivar NASPOT 11 but they did not have access to the planting material.
They indicated that they know that farmers in Soroti are growing the cultivar. Similarly,
in Rakai, which is a district distant to the research institution found in central Uganda,
few respondents (29%) were growing released cultivars, and the cultivars grown were
those released over a decade ago. The majority of the respondents in Rakai district were
not aware of the released cultivars.
A survey was done of the varieties/cultivars grown in central, eastern and northern
regions, and their relevant characteristics. These data are summarized in Annex 2, 3,and
4, respectively. These findings are important because the decision to maintain or discard
a variety is greatly influenced by observations of morphological characteristics made by
the farmer (Jarvis and Hodgkin, 2000; Smale et al., 2001). Only 13 out of the 19 released
cultivars are still cultivated by the respondents. Besides the released cultivars, very few
varieties appeared in more than one region. Only three farmers’ varieties (Araka Red,
99

Araka White, and Kampala Red) were recorded in both the northern and eastern region.
The central region did not have any farmers’ varieties in common with northern or eastern
Uganda. This could be a result of variation in climate and soil types thus favoring
different varieties, variety renaming after introduction or the sample size used in the
study. Further studies with a larger sample size may help to elucidate the reasons for the
varietal distribution among the regions.
We determined whether farmers were still growing varieties that had been
reported to be commonly grown in a recent study (Yada, 2009). Overall 82% of the
varieties recorded in this study were reported in the same region by Yada (2009).
Eighteen varieties, recorded by Yada (2009) as commonly grown in certain regions, were
no longer grown by any of the respondents. With the help of the extension workers and
farmers’ leaders within each region, it was suggested that some of these varieties were
most likely lost from the region overall (Table 3.7). Respondents indicated that some
varieties have been lost due to prolonged droughts that have resulted in loss of the
planting materials. Other varieties were lost or dropped due to deterioration of
performance resulting from increasing susceptibility to pest infestation or viral infection,
or proven superiority of a new variety to local conditions. The respondents no longer
cultivated Sowola, the first newly bred cultivar released in 1995 because of increased
susceptibility to the common virus diseases. Some farmers’ varieties have been displaced
by varieties or cultivars demanded by the markets; for example increased cultivation of
the orange-fleshed cultivars, Osapat and Kampala in the eastern region contributed to loss
of certain varieties (Table 3.7). On the other hand, three respondents in Luwero district
indicated that they had replaced two released cultivars i.e., NASPOT 5 and NASPOT 7
with a farmers’ variety demanded by the markets (Sukali). Veasey et al. (2008) reported
that sometimes the varieties of sweetpotato, which were lost in certain communities, may
be found in other communities
100

Table 3-7 Names of varieties that were discarded or lost in the various regions of Uganda,
and the reasons for discarding or for the loss.
Region
Central

Varieties dropped
Silk

Kyebandula
Rashid
Sula oluti
Nylon
Bubakali
Nyendo yamulalo
Eastern

Ebor
Epura Amojong
Mukoma
Opila
Original

Northern

Imatojony
Ocaaka amani
Odeo Cani
Olul
Tutwech
Anam Anam

Reason for dropping/ loss
Susceptible to caterpillar, storage root
size reduced tremendously, and it is
difficult to access vines which were lost
during drought.
Susceptible to viruses, and late maturity
(over 6 months)
Susceptible to viruses
Difficult to access vines lost due to
drought
Susceptible to viruses
Susceptible to weevils
Susceptible to viruses
Difficult to access vines, increasing
thready flesh, or poor yield
Low yields
Replaced by new market preferred
varieties
Low yields
Rough and cracked skin, which allows
pest infestation
Low yields
Low yields
Susceptible to viruses
Poor yields, increasing thready flesh
Replaced by new market preferred
varieties
Replaced by new market preferred
varieties

101

where the production constraints were less prevalent as a result of material exchange.
However, in other cases, such displacements have resulted in genetic erosion as observed
in Ethiopian sorghum by Tsegaye and Berg (2007).

Impact of farmers’ variety exchange
Plant exchange by farmers is a known source of genetic diversity (Subedi et al.,
2003; Veasey et al., 2008; Thomas et al. 2012). Of the respondents in this study, 92%
engaged in this practice (Table 3.8). The flow of materials between farmers occurred
mainly through free exchange as gifts or barter exchange (either for a different variety or
different crop planting materials), or purchasing the materials. About 36% of the
respondents used both channels to access planting materials. Farmers indicated that in
many cases buying material resulted in accessing the desired variety but it requires
purchasing power.
Farmers who did not engage in material exchange practices were found in Luwero
(27%), Lira district (17%), Bukedea and Soroti districts (7% each). These farmers
indicated that they avoided this practice because they think it is a key channel for plant
disease spread. They also indicated that whenever they want to adopt new varieties or if
they need planting material for an older variety, they access material from the national
research institutes. As a result, they are able to protect their fields from common pests and
diseases. Respondents engaged in farmers’ variety exchange also indicated a number of
challenges (Figure 3.3). Lack of accurate information about the variety is a critical
challenge especially if the farmer purchased the planting materials and it performed
poorly in the local environment.Financial constraints, cooperation amongst farmers,
access to materials to ensure timely planting, and limited options were the other key
challenges.
102

Table 3-8 Number of farmers practicing sweetpotato varietal exchanges in the different
regions of Uganda.
Region
Central

District
Luwero
Mukono
Rakai
Eastern
Bukedea
Kumi
Soroti
Northern Apac
Lira
Total

Free/barter
exchange Buying
5
2
5
12
1
7
11
8
6
55

2

Both
4
5
3
13
3
3
2
4
37

103

No
Total
exchange farmers
4
15
10
15
1
15
10
1
15
10
2
12
8

102

Free/barter

80
%age of the respondents

70
60

Buying

*
*

50
40

*

30
20
10

0
Pest and
Lack of
disease variety info
transmission

Financial Cooperation
constraint
amongst
farmers

Timely
access

Limited
options

Figure 3-3 Key challenges faced by farmers engaged in material exchange practices. Yaxis shows the percentage of farmers engaged in one form of material exchange (for
example buying) who selected the variable as a key challenge. Farmers were asked to
provide up to three key challenges. Six farmers carrying out the exchange practice
indicated that they had no challenges. * denotes significance at 5%.

104

Plant material exchange is usually most challenging following a long dry spell.
Only farmers who own or can rent portions of the wetland to preserve their future
planting materials or can afford irrigation have ready access to planting material in the
next season. In many cases, this meant that better-off farmers were the suppliers of the
planting material. Better-off farmers being key suppliers of planting material in informal
exchange have been reported in other literature (Richards 1990; Cromwell 1996).
However, these better-off farmers indicated that they conserve planting material for their
own use in the next season, thus have little or nothing to share. This may be the reason
why other farmers indicated lack of cooperation by fellow farmers as a key constraint for
material exchange. A tendency for farmers in need of planting material labeling others as
stingy or uncooperative is a common phenomenon observed in seed exchange (Scott,
1985; McGuire 2008). However, lack of cooperation from fellow farmers has also been
reported in other agricultural studies in Uganda (Byamugisha et al., 2008). Gaining
access to seed or planting materials has also been identified as the major challenge in
other crops (Jones et al., 2002; Sperling et al., 2003; McGuire 2008).
Weakening social ties among farmers has been identified as a major factor
leading to some of the challenges faced during material exchange (Bellon, 2001; McGuire
2005; Winter et al. 2006). Without the traditional social ties such as community
gatherings, and working in groups, farmers miss the opportunity to learn about and obtain
access to the best new diversity available.

Observation of volunteers or potential hybrids in farmers’ fields
Spontaneous seedlings can also increase varietal diversity (MacDonald, 1965;
Carey et al., 1998; Mwanga et al., 2001). When we assessed the occurrence of volunteers
or potential hybrids in farmers’ fields, we found that only 38% of the farmers interviewed
105

indicated that they had observed volunteers or potential hybrids in their fields. These were
farmers who have been growing sweetpotato for more than 15 years and in each case had
observed potential hybrids only 1-3 times during that period. There was a significant
(P<0.001) positive association between observing potential hybrids and the farmer’s age
(Phi=0.57) (Table 3.9). The majority of farmers who observed potential hybrids were
over 40 years old. This evidence may support the hypothesis that occurrence of
sweetpotato hybrids is a rare event due to multiple compatibility barriers, low seed set
and poor germination (Mwanga et al., 2007; Lebot, 2010).
Only 17% of the respondents replanted the volunteer plants to evaluate their
agronomic and quality characteristics. The majority of these were farmers who had at
least three encounters with volunteer plants. In most cases, first and second encounters
were ignored or the seedlings were removed and destroyed. It is possible that if the
ignored seedlings survived, they probably were planted the next season, and based on
their fitness and agronomic/quality characteristics they may have become part of the
propagation material for future generations.

Criteria for adopting new variety or improved cultivars selection
The three criteria leading most respondents to adopt new varieties or cultivars
were yield, followed by taste and duration to maturity (Figure 3.4A). Other factors
indicated to be important included market demand, and tolerance to biotic and abiotic
stresses. A similar ranking was recorded by Abidin et al. (2004) for a study conducted in
the northeastern region of Uganda. Bashasha et al. (1995) had a slightly different ranking
with the three key factors including yield followed by resistance to pests and diseases,
and good piecemeal harvesting (i.e., ability to maintain high quality under in-ground
storage). Consideration of market demand as a key constraint was ranked

106

Table 3-9 Measure of association between different variables that influence crop
diversity.
Variable 1
Membership in farmers' groups
Observing potential hybrids
Frequency of adoption of new
varieties

Variable 2
Interaction with extension
service
Age of farmer
Gender of farmer
Age of farmer
Gender of farmer
Land ownership
Land size

Measure of
1
association
0.703***
0.574***
0.128
0.580***
0.285
0.420*
2

0.155
0.488***
0.419**

Membership in farmers' groups
Interaction with extension
service
Market access
0.425*
Weather conditions
0.504*
Maintenance of current
Age of farmer
0.615***
varieties
Gender of farmer
0.304*
Land ownership
0.370*
Land size
0.2472
Membership in farmers' groups
0.229
Interaction with extension
0.248
service
Market access
0.400*
Weather conditions
0.309*
Criteria for adoption of new
Considerations for maintaining
0.587***
varieties
current varieties
***, ** and * denote significance at 0.1%, 1% and 5%, respectively.
1
Phi was used as the symmetric measure of association, based on chi-square, between
nominal variables.
2
Eta was used as the measure of association between one nominal and one interval
variables.

107

87%

A

56%

48%
37%

Total respondents- 102

29%

25%
17%

17%
10%

Higher
yields

63%

Taste

Duration Market Access to ResistanceDry matter Beta Piecemeal
to
demands planting to stresses content carotene harvesting
maturity
materials
content

B
57%

42%

Total respondents- 102
32%

15%
10%

Yield
stability

Adaptability
to abiotic and
biotic stresses

Taste

Piecemeal
harvesting

Duration to
maturity

10%

Market
demands

Availability
of planting
material

Figure 3-4 Criteria for selection of A) new varieties/cultivars; and B) maintaining current
varieties. Numbers above the peaks indicate percentage of the total number of
respondents (102) that selected the variable as a key factor in their decision-making.

108

by a lower number of respondents. There was a significant difference (P<0.05) in
consideration of market demand between male and female farmers. Close to 70% of the
male respondents, who typically produce sweetpotato for commercial production, ranked
market demand as a key consideration for adopting new varieties/cultivars. In contrast,
only 22% of the female farmers who primarily grow sweetpotato for home consumption
and food security, considered market demands as a key factor. This may explain the lower
ranking of market demand since the majority of the respondents in the different regions
were women.
Consideration of market demand is most relevant for farmers with access to
market outlets and those expecting to sell a significant portion of the sweetpotato
produced. Subsistence-oriented farmers tend to focus on consumption attributes such as
good taste, and high dry matter or beta carotene content depending on the community,
plus piecemeal harvesting ability.
Respondents also indicated that in some cases the varieties they plant do not
necessarily indicate their preference or what they consider suitable for their environment,
but rather their selection is driven by accessibility to disease-free planting materials at the
time of planting. These findings were also observed by Hall and Clark (1995).

Frequency of adoption of new varieties or newly bred cultivars
The respondents ranged from farmers growing sweetpotato for the first time to
farmers who have not adopted new varieties/cultivars in more than five years and
sometimes even more than ten years (Table 3.10). More than half of the respondents
cultivated the same varieties for over three years before adopting new varieties/cultivars.
A substantial number of respondents (22) adopted new varieties/cultivars whenever they
identified one that performed better than their current varieties. Gender did not influence
frequency of adoption of new varieties/cultivars, while a majority of the farmers above
109

Table 3-10 Number of farmers who adopted new cultivars/varieties and the frequency of
adoption (Total number of respondents = 102).
New
Region
District farmers
Central
Luwero
1
Mukono
2
Rakai
1
Eastern
Bukedea
0
Kumi
0
Soroti
1
Northern Apac
0
Lira
1
Total

6

When I find
something
better
12
3
2
0
1
1
1
2

Every
season
0
0
0
1
1
2
0
0

Every
2 years
0
0
2
2
3
4
2
4

After
3-5
years
0
3
6
1
2
3
2
3

Over 5
years
2
2
4
11
3
4
5
2

22

4

17

20

33

110

the age of 50 had reduced frequency of adoption (Table 3.9). Membership in farmers’
groups and frequency of interaction with extension agents also affected the rate of
adoption of new varieties/cultivars. Farmers that interacted more frequently with an
extension service (at least once a year) tended to change their varieties more frequently
than those who had less (between 2-3 years) or rare (over 3 years) interactions with an
extension service (data not shown). In this case, the extension service is provided by the
agricultural officers assigned by the Ministry responsible for agriculture or from NGOs,
as well as trained farmer-groups’ leaders. Asrat et al. (2010) also observed that, in
Ethiopia, teff and sorghum farmers with frequent interaction with extension services
adopt new varieties more frequently.
Size of land allocated to sweetpotato production did not significantly influence
frequency of adoption of new varieties. However, the low measure of association
(Eta=0.155) is largely attributed to respondents from Bukedea districts where the majority
have cultivated the same OFSP varieties for at least five years regardless of the land size.
With the exception of Bukedea district, farmers allocating over 0.5 hectare to sweetpotato
production tended to adopt new varieties or cultivars more frequently. This is probably
because they are more market-oriented and they tended to interact more frequently with
extension service than subsistence-oriented farmers.

Challenges to adoption of new varieties
Having only information about the varieties that are being considered for adoption
was the main challenge to adoption of new varieties; it was indicated by close to 70% of
the respondents (Figure 3.5). Access to planting materials for known better varieties was

111

Limited interactions with extension service

% number of respondents

60
50

Frequent interactions
*

40
30
20

10
0
Limited info about Access to better
variety
varieties
performance

Market issues

Variation in soil
type

Figure 3-5 Key challenges associated with adoption of new diversity. Respondents were
asked to identify two key challenges to adoption of new diversity. Total number of
respondents was 102. * denotes significance at 5%.

112

also identified as a key challenge by 65% of the respondents. Market issues and variation
in variety performance due to differences in soil type between farmers’ fields were also
identified as key challenges.
Interaction with extension services resulted in a significant difference in what
were considered to be key challenges. The proportion of respondents who considered
limited information about varieties and lack of access to better varieties as key constraints
was significantly less (P<0.05) for farmers that frequently interacted with extension
services compared with farmers that rarely met with extension providers. Farmers that
rarely met with extension services relied on the information provided by the source of
planting material, in many cases a fellow farmer or merchant, however in some cases
such information is biased to performance in one or a few location(s) or season (s)
(McGuire, 2005). This may result in unreliable information about the variety. There was a
strong association between membership in farmers’ groups and frequency of interactions
with extension providers (Table 3.9). This may explain why respondents in farmers’
groups did not consider limited information on varieties and access to better varieties as
the major challenges. These findings support the hypothesis that weakening social ties
among farmers causes farmers to miss opportunities to obtain information about better
existing or new diversity, and to access this diversity (Bellon, 2001; Winter et al., 2006).
Market issues such as access to market outlets and price fluctuations were also
key challenges for 46% of the respondents who are involved in commercial production,
regardless of the frequency of interaction with extension providers (Figure 3.5). This is
because in many cases farmers are not provided with sufficient information regarding the
supply, demand and price of a produce, and yet this is a prerequisite for effective
marketing (Brush 2000; Van Dusen and Taylor, 2005). Since sweetpotato production is
largely dependent on rainy seasons, if many farmers grow the same varieties then there is
113

a glut at the harvest time. In this circumstance, the farmers receive low prices or
sometimes are not able to sell their produce. This challenge is worst for farmers where
distances between production areas and main markets are greatest. Production of
sweetpotato by many households in the communities makes marketing of the produce to
the surrounding communities a big challenge, thus farmers have to rely on distant traders
(Abidin, 2004). These observations suggest that crop diversity maintenance may be
enhanced by providing farmers with solutions to the market issues, including innovations
that facilitate non-rain-fed production, and increase ability to store long term and develop
new sweetpotato products and processed formss.
The challenge of variation in soil type can only be solved by case-by-case
evaluation of the varieties or new cultivars by the farmers in their fields. The majority of
the respondents (73%) indicated that first they evaluate performance of the new varieties
by planting a few mounds. Some of these farmers (28%) conduct further evaluation by
testing the varieties in multiple locations in their field. Locations were selected based on
slope, soil structure or amount of organic matter. For example, a farmer who owns land
on a hill slope indicated that he will plant a few mounds at the top, middle and lower part
of the slope to evaluate the performance of the variety in varied soil types. On-farm
evaluation of variety performance by individual farmers is a common practice that has
been reported in other crops (Collinson, 1981; Smale et al., 2001; Mikkelsen and
Langohr, 2004). Only the new varieties or cultivars that perform well in the receiving
environment are adopted to increase the farmers’ crop diversity or replace an older
variety.

114

Impact of adoption of new cultivars/varieties on maintenance of existing varieties in
the farmers’ field
The majority of the respondents (73%) indicated that they kept some of the
existing varietal diversity in their field when they adopted new varieties or cultivars.
Other respondents (27%) completely replaced their current varieties because of limited
land resources, and/or poor performance of the current varieties. Poor performance most
frequently resulted from reduced tolerance to the environmental stresses especially pests
and diseases. Almost equal numbers of respondents allocated less than quarter, between
quarter and a half, and between half and three quarters (27%, 23% and 23%, respectively)
of their sweetpotato cultivation area to maintaining previously cultivated varieties when
they adopted new ones.

Criteria for maintaining a variety in the farmers’ fields
The criteria used for maintaining older varieties when adopting new varieties or
cultivars were different from considerations for adopting new varieties (Figure 3.4B). The
largest number of respondents considered yield stability as critical, but this had to be
supplemented by tolerance to native biotic and abiotic stresses, and/or good taste. Farmers
indicated that in most cases the new varieties or cultivars are not adapted to the local
environment thus their higher performance tends to diminish over time. Asrat et al (2010)
observed that the majority (over 60%) of the teff and sorghum farmers interviewed
considered environmental adaptability more critical than yield stability (less than 25%).
Increased desire for environmental adaptability was attributed to the increasing frequency
of environmental stresses such as prolonged drought events and soil degradation, plus
lack of resources and technologies to help farmers to overcome these stresses.
Piecemeal harvesting capability was also considered a key factor by a substantial
number of the respondents (32%). Respondents indicated that most of the new
115

varieties/cultivars are early maturing varieties, but do not have good piecemeal harvesting
characteristics. Other factors considered when selecting varieties to conserve included
early maturity, market demand, and availability of planting materials. These results are
important for both in situ conservation of genetic diversity and crop varietal technology
adoption because they demonstrate that yield stability complemented by environmental
adaptability need to be breeding priorities if the goal is to motivate farmers to maintain
crop diversity for food security and climate change resilience. Asrat et al. (2010) also
reported that farmers considered not only the productivity attributes but also the
environmental adaptability traits during variety selection for teff and sorghum in Ethiopia.

Challenges to maintaining varieties in farmers’ field
The challenges identified as key for maintaining varietal diversity were different from
those indicated for adopting new diversity. Respondents indicated four major challenges
they face when maintaining varieties in their fields (Figure 3.6). These are: land limitation
(identified as a key challenge by 70% of the respondents that maintain older varieties);
loss of planting material due to prolonged drought events (63%); lack of training in better
conservation techniques (47%) and loss of planting materials due to feeding by animals
and pest infestation (19%). Two respondents had no problems with maintaining all
desired varieties in their fields. Respondents who are rented the land and female farmers
whose husbands allocate land use indicated land limitation as a key constraint. The male
owner tended to have different priorities for land use compared to the woman farmers’
priorities, which affect the size of land allocated for sweetpotato production, thus
reducing the probability for diversity preservation on-farm. This may explain the
significant association observed between land ownership issues and maintenance of
varieties by respondents (Table 3.9).

116

Men

90%
80%
70%

Women

Total

*

60%
50%
40%
30%
20%
10%
0%
Limitation of land

Losses due to
drought

Lack of info about
better conservation
methods

Losses due to
pests/animals

Figure 3-6 Key challenges associated with maintaining current varieties when adopting
new varieties/cultivars. Respondents were asked to identify two key challenges to
adoption of new diversity. Total number of respondents was 96. * denotes significance at
5%.

117

Prolonged drought events have resulted in loss of planting material in farmers’
fields. To cope with this constraint, respondents indicated that they use various strategies
including planting the vines near wetlands, on river banks or within banana plantations
especially during the dry season. However, very few of the respondents had access to
wetlands or had the resources to carryout irrigation (Table 3.11). Drought was also
reported as an important constraint that resulted in substantial reduction of faba bean
diversity in Morocco by influencing seed composition and reserves over time (Sadiki et
al., 2004).
Inadequate training of farmers in conservation practices has also been identified as
a key constraint to in situ conservation of other crops such as teff and wheat in Ethiopia
(Feyissa, 2000), coffee in El Salvador (Méndez, 2008), and potato in the Andes (De
Haan, 2009). There is need to incorporate training on conservation of planting material as
part of any new sweetpotato variety release package.

Crop management practices and how they affect diversity
Common crop management practices carried out by the respondents in their fields
are shown in Table 3.11. Sweetpotato performs better in sandy soils with lower moisture
content and organic matter (Dantata et al. 2010), however, in the study areas it was grown
in all types of soils ranging from sandy to clay loams (Table 3.2). The majority of the
farmers in all the study regions grow sweetpotato on mounds, which are conical heaps of
soil with variable sizes depending on the soil type, and availability and capacity of labor.
A typical mound does not exceed 1m in diameter and height. Ridges were used instead of
mounds by respondents (11%) that cultivated sweetpotato in wetlands. Ridges help to
improve drainage in wetlands. Bashasha et al. (1995) also reported use of both mounds
and ridges for sweetpotato production in different agro-ecological zones of Uganda
118

Table 3-11 Number of subsistence and commercial farmers, and the percentage of total
respondents that are conducting the different crop management practices (Total number
of respondents = 102).

Management practice

No. of
subsistence
farmers

No. of
Percentage
commercial of total
farmers
respondents

Weeding

32

66

96

Early field preparation

25

58

81

Timely planting

22

46

67

Using clean plant material 22

41

62

Rogue out diseased plants

18

27

44

Hilling up the mounds

6

36

41

Intercropping
Conserve vines in
wetlands

22

11

32

5

23

27

Irrigation

0

26

25

Pesticide application

0

17

17

Digging ridges

4

7

11

Crop rotation

5

5

10

Soil fertility management

2

8

10

Herbicide application

0

7

7

119

because they control the soil moisture content and provide a loose soil environment. The
number of vines and varieties planted per mound is heterogeneous depending on the
household. Planting can occur any time throughout the year depending on the weather
patterns, access to planting materials, and labor availability. In Uganda, sweetpotato is a
rain-fed crop (Bashasha et al., 1995, Abidin, 2004, Yada 2009). Farmers select planting
material depending on the conditions of the leaves (they reject diseased and wilted
leaves). Some respondents (62%) indicated that starting with clean planting material is
critical to realize high yields, and to avoid spreading of diseases. Monocropping was
predominant (68%) in all regions. This finding is lower than the proportion reported in
the national report, which shows that 83.4% of sweetpotato grown in pure plots (UBOS,
2010).
A substantial number of farmers (32%) intercropped sweetpotato with cereals or
legumes. The most common cultural practice carried out by the respondents (94%) was
weeding. Most farmers weed their fields only once every season because over time the
vegetation covers the ground and controls the weeds. Manpower is also a constraint
since weeding is done by hand hoe. Other common practices included early preparation of
the field (81%), and planting at the on-set of the rains (67%). Hilling involved adding soil
to the mounds to protect the storage roots from pests like weevils and rodents as well as
sunlight. Exposure to sunlight causes the storage roots to turn green due to production of
chlorophyll. At harvesting, most farmers left part of the crop in the field so that they have
reserves for planting materials for next season (piecemeal harvesting). Other farmers who
had access to wetlands would conserve the planting materials in the dry season, and in
cases of prolonged drought events these were considered the sources of material for the
next season.
Pesticide and herbicide application as well as soil fertility management were
carried out by very few respondents (Table 3.11). These practices were carried out by

120

farmers engaged in commercial production of sweetpotato. Abidin et al. (2004) also
reported limited application of pesticides and inorganic fertilizers in sweetpotato fields in
northeastern Uganda. The negligible application of these inputs was a result of limited of
access, and where they were available, farmers indicated low cost/benefit ratio especially
due to unreliable market factors. Farmers are less likely to adopt technologies or practices
if they are not convinced that they advance their goals sufficiently to outweigh its costs
(Pannell et al., 2006). The majority of the respondents preferred to use fertilizers and
manure in fields of more profitable crops like cotton/maize in the eastern region,
maize/millet in the northern region, and the banana/maize in central region. As a result,
sweetpotato production normally occurs under low soil fertility levels, which may
explain, to a large extent, the low average yields realized in the country.
The key challenges with sweetpotato crop management as reported by the
respondents include financial constraints to incorporate some of the desirable inputs
(57%), increased frequency and duration of drought (51%), labor constraints to
implement the necessary practices (47%), limited access to information on how to
improve current cultural practices so as to realize better yields (44%), and pests especially
moles, monkeys and weevils (42%). (Figure 3.7). Age and gender of the farmer influence
some of the key constraints reported by the respondents. For example, the majority of the
respondents (75%) who reported labor as a key constraint were farmers over 41 years old,
and its importance increased with age. Women reported the labor constraint more than the
men (Figure 3.7). Some respondents indicated that they deal with the labor constraints by
hiring labor. However, since financial resources are scarce, many farmers opt to reduce
the operations in the field. Financial constraints to incorporate some of the desirable
inputs like irrigation and pest and disease control may also explain the loss of
varieties/cultivars that are susceptible to these environmental stresses.

121

Male

%age of the respondents

60
*

50

Female

Total

*

40
30
20
10
0
Financial
constraints

Unpredictable
dry season

Labour
constraints

Limited advice Pests attacks
from experts

Figure 3-5 Major challenges affecting crop management practices necessary to realize
optimal sweetpotato yields. The challenges varied depending on the gender of the farmer.
Respondents were asked to give at most three major challenges. Total number of
respondents is 102. * denotes significance at 5%.

122

Conclusion
Success of in situ conservation is judged by establishing the impact of adopting
modern cultivars on the level of local diversity; by assessing the number of genotypes
maintained in a target area; by evaluating the number of farmers maintaining this
diversity; and by assessing the level of variety exchange within and between different
communities (Brush, 1999).
The results of this study show that in most districts, farmers were growing both
landraces and released cultivars. However, maintenance of traditional varieties was
influenced by ecological conditions, socio-economic factors, and crop management
practices. Drought was identified as a major constraint to sweetpotato production. It was
also observed that sweetpotato production is largely carried out on small plots under lowinput conditions. Marginal conditions with high land fragmentation tend to favor
maintenance of high landrace diversity especially in centers of diversity (Brush 1995;
Bellon, 2004). Lack of access to information about performance of varieties, and best
agronomic and conservation practices is another key constraint to maintenance of
farmers’ varieties. Economic conditions such as financial constraints to adopt modern
farming technologies/inputs, and unreliable market environment were found to also favor
cultivation of traditional varieties.
Brush (1995) observed that the ecological and socio-economic conditions favoring
cultivation of landraces are not fixed because they will be forced to change in the near
future due to rapid population growth and adverse effects of climate change, which
necessitates increasing agricultural productivity per unit area. Uganda has a rapidly
increasing population (CIA, 2008), and is located in Sub-Saharan Africa, a region
identified as the most vulnerable to climate change (IPCC, 2007). This may necessitate
adopting modern agricultural technologies and practices to increase agricultural
123

productivity. Such interventions will have great impact on traditional varietal diversity,
and since there are no appropriate incentive mechanisms to encourage farmers to
implement in situ conservation strategies, the crop diversity will have to rely on selection
factors.
Diversity is a product of natural and human selection (Boster, 1985; Brush 1999).
Some varieties will be lost due to competition between genotypes within the farmers’
fields, while others will be dropped by the farmers based on human preferences. Farmers
only maintain varieties that fulfill their private demands since it is not their responsibility
to maintain diversity (Asrat et al., 2010). Criteria for variety selection varies based on
household characteristics and their uses of sweetpotato, micro-climatic conditions in the
farmers’ fields, and level of access to agricultural extension. Because of this
heterogeneity, breeding programs are less likely to provide cultivars with most of the
preferred characteristics. The observed heterogeneity in selection criteria has important
implications for setting breeding priorities, adoption of modern cultivars, in-situ
conservation, and preservation of allelic diversity.
Preference of varieties with high environmental adaptability and yield stability
may explain why there is so far low adoption rate of the high-yielding modern cultivars
over the past two decades. This information points to the need for breeding programs to
prioritize traits that will enhance a cultivar’s adaptability to the common stresses of the
receiving environment under minimal input conditions. Interventions aimed at diffusion
of modern cultivars to increase crop diversity need to integrate mechanisms to address the
current unreliable market environment including introducing non-rain-fed production and
agro-processing of sweetpotato. Farmers will only adopt technologies if they are
convinced that the benefits outweigh the costs (Pannell et al., 2006). Therefore a key
responsibility of extension service should be to ascertain that new technologies have net
124

benefits to the farmers under their jurisdiction before they embark on education and
promotional activities.
Results also show that depending on household characteristics certain farmers are
likely to maintain traditional varieties de facto. In addition, depending on household
preferences some varieties are likely to be maintained even with adoption of new
cultivars. However, to facilitate continuous and improved in situ conservation, external
incentives such as market access and conservation trainings should target these farmers.
Weitzman (1993) suggests assigning in situ conservation priorities to diversity not yet at
risk so that it never becomes threatened. Using the information provided in Tables 7, 8,
and 9, certain varieties adapted to specific local environments or that meet specific local
preferences may be selected for conservation. Further studies are necessary to determine
the proportion of crop diversity that needs to be maintained on-farm, and identify other
incentives that can be adopted to encourage continued in situ conservation of sweetpotato
diversity for many decades to come.
There is a positive relationship between genetic diversity and strong social ties
(Bellon, 2001). Networking among farmers enables sharing of experiences and
information about available diversity and its management, strengthens planting material
exchange, and enables collective action to access superior varieties and markets beyond
the surrounding community. At a national level, the results suggest that as the country's
agriculture develops to meet the needs of the growing population and ever changing
environment, there will be need for a national policy to ensure effective ex situ
conservation of varieties that possess unique desirable traits/alleles but have been dropped
or lost by the farmers. There is also need for national policies to address key constraints
to crop diversity conservation such as unreliable weather conditions with prolonged
drought events and market access and price fluctuation.
125

APPENDIX

126

Table B-1 Characteristics of sweetpotato varieties/cultivars grown in the central region of Uganda

Farmers
Variety name
growing
Baganyombekele
3
Bamutola
1
Bitambi
5
Bubakali
2
Busuku
1
Dimbuka
4
Entebbe
1
Kakobe
2
Kalebe
3
Kavunza
7
Kawogo
8
Kawungezi
5
Kifuta
1
Kimotoka
2
Matungakibe
3
Mbale White
5
Mbale Yellow
3
Mbizibu family
2
Mulelabana
2
Munafu adimbuka
2
Mwezigumu
2
Nambi
1
Nantongo
2
Somba Buseno
2
Sukali
5

2

Type
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV

Months
to
maturity
3
3
4
6
3
4
4
6
8
4
6
6
4
4
4
3
4
4
3
3
2.5
7
5
4
4

Yield
2
1
2
2
3
1
2
2
3
2
2
2
2
3
3
2
2
3
2
2
2
1
2
3
2

3

Skin
4
color
1
2
2
3
1
3
2
5
2
2
2
3
2
2
5
2
2
2
1
2
3
2
2
3
2

Flesh
5
color
1
1
1
2
1
3
1
1
1
1
1
1
1
1
1
1
3
1
3
1
1
1
1
1
1
127

1

Piecemeal
Weevil
Virus
harvest
6
3
3
resistance resistance suitability
1
2
2
1
3
2
2
1
2
2
1
3
1
2
2
1
2
1
2
1
2
1
1
2
3
2
3
2
2
2
2
1
2
1
1
1
2
2
2
2
2
2
2
2
3
3
2
2
2
2
2
3
1
2
1
1
1
2
2
2
1
1
1
1
2
2
2
1
3
3
2
3
2
2
2

Sweetness
2
2
2
2
3
2
2
3
2
2
2
2
2
2
3
2
2
2
3
2
2
2
2
2
3

3

Dry
3
matter
2
2
2
2
2
2
2
3
3
2
2
2
2
2
2
2
2
3
2
3
2
2
2
2
2

Table B-1 (cont’d)

Farmers
growing
5

Variety name
Bwanjule
Dimbuka
Bukulula
8
Naspot 1
22
Naspot 10
6
Naspot 11
8
Naspot 2
5
Naspot 8
4
Naspot 9
3
New Kawogo
8
1
Number of respondents – 40

Type
RC

Months
to
maturity
4

Yield
3

RC
RC
RC
RC
RC
RC
RC
RC

3
3
3
3
3
3
3
4

3
3
2
3
3
2
2
3

2

3

Skin
4
color
5

Flesh
5
color
1

3
3
5
5
5
5
5
2

2
3
4
2
2
4
4
1

2

FV- Farmers’ varieties, RC- Released cultivars

3

1= low, 2= moderate, 3= high

4

1= white, 2= red, 3= cream, 4= brown, 5= purple or purple red

5

1= white, 2= cream, 3= yellow, 4= orange

6

1= very low, 2=mild, 3= moderate

128

Piecemeal
Weevil
Virus
harvest
6
3
3
resistance resistance suitability
2
2
1
2
1
1
1
1
1
1
2

2
2
2
2
3
2
2
3

1
1
1
1
1
1
1
2

Sweetness
2
2
3
2
2
3
2
2
2

3

Dry
3
matter
3
3
3
2
3
2
2
2
2

Table B-2 Characteristics of sweetpotato varieties/cultivars grown in the northern region of Uganda

Variety
name

Farmers
growing

2

Type

Months
to
maturity

Yield

3

Skin
4
color

Aber
3
FV
4
2
1
Acil Acil
4
FV
5
2
1
Araka Red
8
FV
4
3
5
Araka White
7
FV
3
2
3
Kampala
2
FV
5
2
2
Koromojo
4
FV
3
3
2
Lira Lira
12
FV
4
2
2
Museba
1
FV
4
2
3
Okonyineto
4
FV
3
3
3
Okonyonedo
4
FV
4
2
1
Oleke
8
FV
4
3
2
Onamyito
1
FV
4
2
2
Otada
8
FV
4
3
2
Pamdero
4
FV
4
2
3
Twongweno
2
FV
4
2
2
Ejumula
4
RC
3
2
5
Kakamega
2
RC
4
3
5
3
Tanzania
3
RC
3
3
Other varieties/cultivars grown include NASPOT 8 (3 farmers)
1
Number of respondents – 22
2

FV- Farmers’ varieties, RC- Released cultivars

Flesh
5
color

1

Piecemeal
Weevil
Virus
harvest
Dry
6
3
3
3
3
resistance resistance suitability Sweetness matter

1
3
1
3
3
2
1
1
2
2
2
2
1
2
2
3
2
2
2
2
2
3
1
2
4
3
1
3
3
2
1
3
3
1
2
1
1
2
1
2
2
1
2
2
1
4
1
2
4
2
2
3
1
2
3
1= low, 2= moderate, 3= high
4
5
6

2
2
1
2
2
2
2
2
3
3
3
2
2
2
2
1
2
1

2
2
3
2
2
3
2
2
3
2
2
2
3
3
2
3
2
2

1= white, 2= red, 3= cream, 4= brown, 5= purple or purple red
1= white, 2= cream, 3= yellow, 4= orange
1= very low, 2=mild, 3= moderate
129

2
2
3
3
2
2
3
2
2
2
2
2
3
1
2
2
2
3

Table B-3 Characteristics of sweetpotato varieties/cultivars grown in the Eastern region of Uganda

Farmers
Variety name
growing
Amongin
1
Anyara
2
Ateseke
2
Biscuit
1
Edipalait
1
Elaleit
1
Esamait
4
Eweta
1
Igangamalayan
1
Ikala
1
Inego
1
Iwela
1
Kala
1
Kassim
9
Kigaire
4
Latest
3
Mary
6
Mukoma
1
Night
3
Opade
1
Original
2
Osapat/Boy
9
Socaddido
2
Bunduguza
2

2

Type
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
FV
RC

Months
to
maturity
4
5
4
3
3
3
4
5
4
4
5
3
4
5
3
3
4
4
5
4
3
4
4
4

Yield
2
2
2
2
2
2
2
2
2
2
1
2
3
2
1
2
3
2
3
2
3
3
2
2

3

Skin
4
color
1
5
3
3
5
5
2
3
3
3
3
3
3
3
3
2
2
3
3
2
1
3
2
3

Flesh
5
color
2
2
2
3
2
1
2
3
3
4
3
2
3
3
1
1
1
3
4
1
2
2
1
2
130

1

Piecemeal
Weevil
Virus
harvest
Dry
6
3
3
3
3
resistance resistance suitability Sweetness matter
1
2
1
2
2
2
3
3
2
1
1
2
2
2
3
1
2
1
3
2
2
2
1
3
1
3
1
2
3
1
2
1
2
3
1
2
2
2
2
3
2
2
2
3
3
2
2
2
2
3
1
3
2
1
2
1
2
1
2
2
3
3
3
2
1
2
2
1
2
3
2
1
1
2
1
3
3
3
2
2
1
2
1
3
2
3
3
3
2
3
1
2
1
3
2
1
1
2
2
1
3
2
2
3
2
3
3
2
3
2
2

Table B-3 (cont’d)
Other varieties/cultivars grown: Araka Red (12 farmers); Araka White (6), Dimbuka Bukulula (1), Ejumula (15), Kakamega (17), Kampala Red
(13), Naspot 2 (2), Naspot 8 (3), Naspot 9 (13), Naspot 10 (11), Naspot 11 (3), Tanzania/ Osukutu (6)
1
Number of respondents – 40
2

FV- Farmers’ varieties, RC- Released cultivars

3

1= low, 2= moderate, 3= high

4

1= white, 2= red, 3= cream, 4= brown, 5= purple or purple red

5

1= white, 2= cream, 3= yellow, 4= orange

6

1= very low, 2=mild, 3= moderate

131

LITERATURE CITED

132

LITERATURE CITED

Abidin, P.E., 2004. Sweetpotato breeding for northeastern Uganda: Farmer varieties, farmerparticipatory selection, and stability of performance. PhD thesis Wageningen University,
Wageningen, 152 pp.
Abidin, P.E., van Eeuwijk, F.A., Stam, P., Sturik, P.C., Malosetti, M., Mwanga R.O.M.,
Odongo, B., Hermann, M., Carey, E.E. (2005) Adaptability and stability of sweetpotato
varieties for low-input systems in Uganda. Plant Breed. 124:491–497
Altieri M.A., L.C. Merrick, 1987. In situ conservation of crop genetic resources through
maintenance of traditional farming systems. Economic Botany, 41 (1) (1987), pp. 86–96
Alvarez, N., E. Garine, C. Khasah, E. Dounias, M. Hossaert-McKey, and D. McKey. 2005.
Farmers’ practices, metapopulation dynamics, and conservation of agricultural biodiversity
on-farm: a case study of sorghum among the Duupa in sub-sahelian Cameroon. Biological
Conservation 121:533–543.
Asrat S., M. Yesuf, F. Carlssonc, E.Wale, 2010. Farmers' preferences for crop variety traits:
Lessons for on-farm conservation and technology adoption. Ecological Economics 69:2394–
2401
Bashaasha B, Mwanga ROM, Ocitti p’Obwaya C, Ewell PT (1995). Sweetpotato in the
farming and food system in Uganda: A farm survey report. CIP, sub –Saharan African region,
Nairobi, Kenya /National Agricultural Research organization, Kampala, Uganda. pp.63.
Bellon M.R. 1997. Understanding farmer seed and variety systems On-farm conservation as a
process: an analysis of its components. In: Using Diversity: Enhancing and Maintaining
Genetic Resources On-farm. Available from http://www.idrc.ca/en/ev-85109-201-1DO_TOPIC.html. Accessed May 2013
Bellon M.R., 1996. The dynamics of crop infraspecific diversity: a conceptual framework at
the farmer level. Economic Botany, 50 (1) (1996), pp. 26–39
Bellon, M.R., 2001. Demand and Supply of Crop Infraspecific Diversity on Farms: Towards a
Policy Framework for On-Farm Conservation. CIMMYT Economics Working Paper 0101.Mexico, D.F.: CIMMYT
Bellon MR (2004) Conceptualizing interventions to support on-farm genetic resource
conservation. World Dev 32: 159-172
Bellon, M.R., Berthaud, J., 2004. Transgenic maize and the evolution of landrace diversity in
Mexico. The importance of farmers’ behavior. Plant Physiol. 134, 883–888.
Bowman D.T., May O.L., Creech J.B. 2003.Genetic uniformity of the US upland cotton crop
since the introduction of transgenic cottons. Crop Sciences 43:515-8.
Brush S., 1991. A farmer-based approach to conserving crop germplasm. Econ Bot 45:153–
165
133

Brush S., 1999. Genes in the Field: On-Farm Conservation of Crop Diversity. IDRC, 2000
Technology and Engineering - 288 pages
Brush S.B., 1995. In situ conservation of landraces in centers of crop diversity. Crop Science,
35 (2) (1995), pp. 346–354
Byamugisha H.M., R. Ikoja-Odongo, G.W. Nasinyama, S. Lwasa, 2008. Information seeking
and use among urban farmers in Kampala District, Uganda. In: Agricultural Information
Worldwide: An International Journal for Information Specialists in Agriculture, Natural
Resources, and the Environment 1(3): 94-101
Camacho Villa, T.C., Maxted, N., Scholten, M., Ford-Lloyd, B., 2005. Defining and
identifying crop landraces. Plant Gen. Resour. 3: 373–384.
Carpenter JE. 2011. Impact of GM crops on biodiversity. GM Crops 2:7-23.
CIA 2008. The World Factbook: Uganda. Central Intelligence Agency
https://www.cia.gov/library/publications/the-world-factbook/geos/ug.html
Collinson M. 1981.A low cost approach to understanding small farmers. Agricultural
Administration 8(6): 433–450
Dantata, IJ, Babatunde,FE, Mustapha, S, Fagam, AS (2010). Influence of variety and plant
spacing on tuber size, tuber shape and fresh marketable yield of sweetpotato in Bauchi
Nigeria. Biol. Environ. Sci.J. Tropics. 7(3):140-144.
De Haan, S. 2009. Potato Diversity at Height: multiple dimensions of farmer-driven in-situ
conservation in the Andes. PhD thesis. Wageningen University, Wageningen.
FAO, 2007. Report of Panel of Eminent Experts on Ethics in Food and Agriculture. Food and
Agriculture Organization of United Nations, Rome. November 2007.
Feyissa R. (2000) Community seed banks and seed exchange in Ethiopia: a farmer-led
approach. In Participatory Approaches to the Conservation of and Use of Plant Genetic
Resources, pp. 142–148. Eds E. Friis-Hansen and B. Sthapit. Rome, Italy: IPGRI
Gepts P, Papa R. 2003. Possible effects of (trans)gene flow from crops on the genetic
diversity from landraces and wild relatives. Environmental Biosafety Research 2: 89–103.
Gichuki, S.T., Berenyi, M., Zhang, D., Herman, M., Schmidt, J., Glossl, J and Burgh, K.
2003. Genetic diversity in sweetpotato [Ipomoea batatas (L.) Lam.] in relationship to
geographic sources as assessed with RAPD markers. Genet. Resources and Crop Evol. 50:
429-437.
Hakiza, J.J, Turyamureeba, R.M., Kakuhenzire, B., Odongo, R.M., Mwanga, R., Kanzikwera,
R. and Adipala, E. 2000. Potato and Sweetpotato improvement in Uganda: a Historical
Perspective: African Potato Association Conference Proceedings vol 5. pp 47-58.
Hall, A. and Clark, N. (1995) “Coping with Change, Complexity and Diversity in Agriculture:
The case of Rhizobium Innoculants in Thailand”, World Development. 23(9): 1601 -1614.

134

Harlan J.R., Crops and man, 1992. (2nd ed.)American Society of Agronomy, Madison, WI
(1992)
Hodgkin, T., J. Rana, J. Tuxill, D. Balma, I. Mar, D. Karamura, R. Valdivia et al. 2007. Seed
systems and crop genetic diversity in agroecosystems. In D. Jarvis, C. Padoch, and H. D.
Cooper, eds. Managing Biodiversity in Agricultural Ecosystems, pp. 77–116. CABI
Publishing, Rome, Italy.
Huamán, Z. and Zhang, D.P. 1997. Sweetpotato. In: Fuccillo, D., Sears, L. and Stapleton, P
(eds.). Biodiversity in trust. Cambridge Univ Press, Cambridge, UK. Pp. 29-38.
IPCC, 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability. Intergovernmental
Panel on Climate Change.
Jarvis D and T. Hodgkin. 2000. Farmer decision making and genetic diversity: linking
multidisciplinary research to implementation on-farm. In: Genes in the field. On-farm
conservation of crop diversity. Edited by Stephen B. Brush. International Development
Research Centre and International Plant Genetic Resources Institute
Lebot V 2010. Root and Tuber Crops: Handbook of Plant Breeding, 7: 97-125
Loebenstein G. and Thottappilly G. (eds.), 2009. The Sweetpotato. DOI 10.1007/978-1-40209475-0 2, C _ Springer Science+Business Media B.V. 2009, 290pp.
Khoury C, Laliberte B, Guarino L., 2010. Trends in ex situ conservation of plant genetic
resources: a review of global crop and regional conservation strategies. Genetic Resources
and Crop Evolution 57: 625–639
Krishna V., Zilberman D., Qaim M. 2003.Transgenic technology adoption and on-farm
varietal diversity. International Association of Agricultural Economists Conference. Beijing,
China 2009.
MAAIF, 2004. Increasing Incomes through Exports: A Plan for Zonal Agricultural
Production, Agro-Processing and Marketing, 2004. Ministry of Agriculture, Animal Industry
and Fisheries, Uganda.
UBOS, 2010. Uganda Census for Agriculture, 2008/09. Volume IV: Crop Area and
Production Report. Uganda Bureau of Statistics (UBOS), 178 pages.
McGuire, S. J. 2008. Securing access to seed: social relations and sorghum seed exchange in
Eastern Ethiopia. Human Ecology 36:217–229
Méndez, V. E. (2008) Farmers' livelihoods and biodiversity conservation in a coffee
landscape of El Salvador. pp. 207-236. In C.M. Bacon, V. E. Méndez, S. R. Gliessman, D.
Goodman & J. A. Fox (eds.) Confronting the coffee crisis: Fair Trade, sustainable livelihoods
and ecosystems in Mexico and Central America. MIT Press: Cambridge, U.S.A.
Mikkelsen J.H. and Langohr R., 2004. Indigenous knowledge about soils and a sustainable
crop production, a case study from the Guinea Woodland Savannah (Northern Region,
Ghana). Geografisk Tidsskrift, Danish Journal of Geography 104(2):13-26, 2004

135

Mutuura, J.N., P.T. Ewell, A. Abubaker, T. Munga, S. Ajanga, J. Irungu, F. Omari, and S.
Maobe. 1992. Sweetpotato in the food systems of Kenya: Results of a socio-economic survey,
p. 51-66. In: Kabira J.N. and P.T. Ewell, (Eds.). Current Research for the Improvement of
Potato and Sweetpotato in Kenya. Proceedings of KARI/CIP Technical Workshop on
Collaborative Research held from 20th February to 1st March 1991 in Nairobi, Kenya.
Mwanga, R.O.M., Odongo, B., p’Obwoya, C.O. and Turyamureba, G.M. 2001. Sweetpotato
(Ipomoea batatas (L.) Lam.). In: J.K. Mukiibi (ed.). Agriculture in Uganda. Vol II: Crops, pp.
233-251. National Agricultural Research Organisation (NARO)-CTA. Fountain Pub.
Kampala, Uganda.
Mwanga, R.O.M., Odongo, B., Niringiye, C. and Alajo, A., 2007. Release of Two Orangefleshed Sweetpotato Cultivars, ‘SPK004’ (‘Kakamega’) and ‘Ejumula’, in Uganda.
HortScience 42(7):1728–1730.
Mwanga R.O.M., C. Niringiye, A. Alajo, B. Kigozi, J. Namakula, I. Mpembe, et al., 2011
“NAPOT 11”, a sweetpotato cultivar bred by a participatory plant-breeding approach in
Uganda. HortScience 46: 317-321.
Ndolo P.J., Nungo R.A., Kapinga R.E and Agili S., 2007. Development and promotion of
orange-fleshed sweetpotato varieties in Western Kenya Proceedings of the 13th ISTRC
Symposium, 689 - 695
Osiru M, Adipala E, Olanya OM, Lemaga B and Kapinga R (2007). Occurrence and
distribution of Alternaria leaf petiole and stem blight on sweetpotato in Uganda. Plant Path.
Journal 6:112-119.
Pannell, D., Marshall, G., Barr, N., Curtis, A., Vanclay, F., Wilkinson, R., 2006.
Understanding and promoting adoption of conservation practices by rural landholders.
Australian Journal of Experimental Agriculture 46, 1407–1424
Perales, H., Brush, S., Qualset, C., 2003.Dynamic management of maize landraces in Central
Mexico. Economic Botany 57, 21–34.
Rao R, Campilan D (eds.) 2002. Exploring the complementarities of ex situ and in situ
conservation strategies for Asian sweetpotato genetic resources. Proceedings of the 3rd
International Workshop of the Asian Network of Sweetpotato Genetic Resources (ANSWER).
IPGRI-APO, Serdang, Malaysia, 191p.
Sadiki, M., M. Arbaoui, L. Ghaouti and D. Jarvis. 2004. Seed exchange and supply systems
and on-farm maintenance of crop genetic diversity. A case study: faba bean in Morocco.: In
Seed Systems and Crop Genetic Diversity OnFarm (D.I. Jarvis, R. Sevilla-Panizo, J.-L.
Chavez-Servia and T. Hodgkin, eds.). IPGRI, Rome.
Smale M., Bellon M.R., Alfonso J., Gómez A., 2001. Maize diversity, variety attributes, and
farmers’ choices in south-eastern Guanajuato, Mexico. Economic Development and Cultural
Change, 50(1): 201-222
Smale M., M.R. Bellon, 1999. A conceptual framework for valuing on-farm genetic
resources. In Agrobiodiversity: characterization, utilization and management, CABI
Publishing, Wallingford, UK (1999), pp. 387–408. D. Wood, J.M. Lenné (Eds.),
136

Smit NEJM. 1997. Integrated pest management for sweetpotatoin Eastern Africa.
Wageningen, (Netherlands): Grafish Service Centrum Van Gils b.v. p. 2–117
Sneller C.H. 2003. Impact of transgenic genotypes and subdivision on diversity within elite
North American soybean germplasm. Crop Sciences 43:409-14
Suszkiw J., 2007. Uncorking genetic bottlenecks in soy. USDA/ARS Agricultural Research
Magazine 55(10): 18-19
Thomas M., J.C. Dawson, I. Goldringer, C. Bonneuil, 2011, Seed exchanges, a key to analyze
crop diversity dynamics in farmer-led on-farm conservation. Genetic Resources Crop
Evolution Journal 58:321–338
Tsegaye, B. and Berg, T. 2007. Genetic erosion studies of Ethiopian tetraploid wheat
landraces in Eastern Shewa, Central Ethiopia. Genetic Resources and Crop Evolution 54:
715-726.
Tumwegamire S., R. Kapinga, P.R. Rubaihayo, Do.R. LaBonte, W.J. Grüneberg, G.Burgos,
T. zum Felde, R. Carpio, E. Pawelzik, and R.O.M. Mwanga 2011. Evaluation of Dry Matter,
Protein, Starch, Sucrose, β-carotene, Iron, Zinc, Calcium, and Magnesium in East African
Sweetpotato [Ipomoea batatas (L.) Lam] Germplasm. HortScience, 46:348-357
Van Dusen, M.E. and J.E. Taylor. 2005. Missing markets and crop diversity: Evidence from
Mexico. Environment and Development Economics 10(4): 513-531.
Veasey EA, Borges A, Rosa MS, Queiroz-Silva JR, Bressan EDA, Peroni N. 2008. Genetic
diversity in Brazilian sweetpotato (Ipomoea batatas (L.) Lam., Solanales, Convolvulaceae)
landraces assessed with microsatellite markers. Genet. Mol. Biol. 31:725–33
Whitlock, M. C., and D. E. McCauley. 1990. Some population genetic consequences of
colony formation and extinction: genetic correlations within founding groups. Evolution
44:1717–1724.
Winters, P., Cavatassi, R., Lipper, L., 2006. Sowing the seeds of social relations. ESA
Working Paper No. 06-16, FAO, Rome.
Yada B. 2009. Characterization and evaluation of sweetpotato genetic potential in Uganda
using agro-morphological and molecular approaches. MSc. Thesis Makerere University,
Kampala, Uganda.
Yada B., P. Tukamuhabwa, Villordon A., A. Alajo, and R.O.M. Mwanga 2010. An online
database of sweetpotato germplasm collection in Uganda. Hortscience 45(1):153–153. 2010.
Zhang D.P., Rossel G., Kriegner A. and Hijmans R. 2004. AFLP assessment of diversity in
sweetpotato from Latin America and the Pacific region: Its implications on the dispersal of
the crop. Genet Resour Crop Evol 51:115-120.

137

CHAPTER 3
INTEGRATING DIVERSE SCIENTIFIC EXPERTISE AND PRACTITIONER
KNOWLEDGE IN PROBLEM FORMULATION FOR ENVIRONMENTAL RISK
ASSESSMENT: A CASE STUDY OF INTRODUCTION OF GE CROPS IN UGANDA

Introduction
Genetic engineering (GE, also referred to as genetic modification) is one of the tools
that plant breeders can use for crop improvement. In light of climate change, deteriorating
environmental health, and increasing demand for land-use due to growing populations,
changing consumption patterns and declining agricultural productivity in Africa, innovative
approaches such as GE, that may contribute to increasing agricultural productivity per unit
area and to improving the quality of agricultural products, need to be considered (Fedoroff et
al., 2010; Godfray et al., 2010; Tester and Langridge, 2010). To-date, a number of
genetically engineered (GE) crops with desirable traits like herbicide-tolerance, insect pestresistance, and virus resistance are cultivated in different parts of the world (James, 2012).
Cultivation of GE crops has grown from 1.7 million hectares in 1996 to 170 million hectares
in 2012. The GE crops grown are maize, cotton, soybean, canola, sugarbeet, alfalfa, papaya,
squash, tomato and poplar. In Africa, as of 2013, four countries are cultivating GE crops:
Burkina Faso (one event of Bt cotton), Egypt (one event of Bt maize), South Africa ( 49
events of herbicide tolerant canola, herbicide tolerant and insect resistant cotton, herbicide
tolerant and insect resistant maize, herbicide tolerant rice, and herbicide tolerant soybean) and
Sudan (one event of Bt maize).
Uganda also has a vision of safely utilizing genetic engineering for agricultural
development as indicated by the National Biotechnology and Biosafety Policy (Republic of
Uganda, 2008). In addition, the National Development Plan (NDP) (Republic of Uganda,
2010) provides for application of biotechnology as one of the science and technology tools

138

that will contribute to national development (NDP, Para 353). The country has begun to make
progress in agricultural biotechnology research and development with a number of activities
initiated by Ugandan scientists and institutions. Uganda is testing a number of GE crops for
possible release including weevil resistant (Bt) sweetpotato, herbicide tolerant cotton, virus
resistant cassava and drought tolerant maize. The problem formulation for the environmental
risk assessment that was conducted for this study covered these three crops.
Sweetpotato is an important food crop in many parts of Africa but its productivity is
constrained by factors such as weevils, which are responsible for more than 60% yield losses
especially in the semi-arid areas (Zhang et al., 1996; CIP, 2008). Engineered weevil resistant
sweetpotato is under consideration because of the scarcity of significant naturally occurring
weevil resistance in sexually compatible relatives (Grüneberg et al., 2009; Stevenson et al.,
2009). Weevil resistant sweetpotato was developed by a partnership between the National
Agricultural Research Organization of Uganda, Auburn University, USA and the International
Potato Center using synthetic Bacillus thurengensis genes, Cry3C and Cry7A, conferring
resistance to Coleopteran insects (Ekobu et al., 2010). Bt toxins, which have a very narrow
range of toxicity, have been widely used to confer resistance in transgenic crops (Zhao et al.,
2003). Laboratory and greenhouse research has been conducted to test the efficacy of the
transgene expression on fitness of two African weevil species (Cylas puncticollis and C.
brunneus). It has not yet been released for cultivation in any country. However, other Bt crops
have been cultivated for more than 15 years, and acreage of Bt maize and cotton cultivation
continues to grow annually (James, 2012).
Glyphosate herbicide tolerant cotton is considered because current practices of weed
control in Uganda are very labor intensive and farm labor is a major constraint (CDO, 2006,
Baffe, 2009) that is increasing due to urbanization. Glyphosate is a broad spectrum contact
synthetic herbicides that is used to control many annual and perennial weeds (Duke and
139

Powels, 2008). Glyphosate is perhaps the least toxic pesticide used in agriculture (Williams
et al., 2000), with a lower acute toxicity than aspirin or many other commonly ingested
compounds. Glyphosate tolerant cotton is cultivated in different parts of the world, including
the USA (James, 2009). Herbicide tolerant cotton accounted for 88% of the US cotton
production in 2009 (National Agricultural Statistics Service, 2009).
Drought tolerant maize is under development because drought is a major constraint to
maize production in Uganda and Africa (Langyintuo et al., 2008), making farming risky for
millions of smallholder farmers who rely on rainfall to irrigate their crops. Water efficient
maize (referred to here as drought tolerant maize) was developed by a partnership between the
International Maize and Wheat Improvement Center (CIMMYT) and Monsanto. The maize is
under field-testing in Uganda, Kenya, Mozambique, South Africa and Tanzania (Edmeades,
2008). However, it has yet not been released for cultivation in any country.
Acceptance and cultivation of GE crops in many countries has been constrained, in
part, by environmental concerns (Gray, 2012). The Cartagena Protocol on Biosafety (CPB)
(UN, 2000), whose mandate is protection of biodiversity, requires that an environment risk
assessment (ERA) should be conducted prior to unconfined release of any new GE organism
into the environment. Similarly, regional legislation such as the Directive 2001/18/EC (EU,
2001), and national legislation/regulations such as those provided by Australia’s Gene
Technology Act (2000), and the US Department of Agriculture’s Biotechnology Permits 7
Code for Federal Regulations part 340 (USDA, 7CFR340) require ERA. An ERA should be
used to identify and evaluate the potential adverse effects from utilizing GE organisms in the
receiving environment (both natural and agricultural environment). The key question to
address in the ERA is whether the changes, intended or unintended, make the GE organism
significantly different from the non-GE counterpart, and whether these changes have the
potential to cause harm (Wolt et al., 2010; Gray, 2012). The findings of the ERA should be
140

science-based, and should inform decisions on handling, transportation and use of GE
organisms (CPB Articles 15 and 16, Annex III), plus support risk management strategies
(ESFA, 2010).
Although ERA has become a standard requirement for GE crops, there are no
universally agreed methods for conducting such ERAs, and in some cases the guidelines are
frequently revised, for example the case of the European Food Safety Authority (EFSA,
2010). However, most ERAs for GE crops have adopted the concept of ‘familiarity’ (OECD,
1993), which requires assessing whether changes, measured as a difference between the GE
crop and its non-GE comparator, fall within the range of variability observed in the crop
species as a whole (Gray, 2012). In addition, most methodologies of ERA for GE crops have
been modified from a common framework derived from assessing the risks associated with
agrochemicals (Hills, 2005). A version of the ERA for GE crops is provided by Wolt et al.
(2010). The major steps this framework include: 1) problem formulation, 2) risk
characterization (exposure and consequence assessment), 3) risk evaluation, and 4) risk
mitigation or management.
Problem formulation involves framing questions relevant to assessing which valuable
aspects of the environment are most at risk of harm, and then developing a plan to answer the
questions (Wolt et al., 2010; Gray, 2012). The two aspects of problem formulation are
determination of problem context and problem definition. Problem context uses
environmental policies and goals to define the boundaries and scope of the ERA, and to
derive assessment endpoints. The assessment endpoints should be attributes of the valuable
entity that can provide evidence of the harm in case it happens, for example insect pollinators
such as honeybees are a valuable ecological entity, and their abundance is a broad assessment
endpoint (Nickson, 2008). Problem definition involves translating the broad assessment
endpoints into specific measurable and verifiable variables that can be used to assess the risk,
141

for example measuring the toxicity of the stressor on honeybees. Using available information
about the plant and the new or modified trait, as well as the receiving environment, the most
important questions that merit detailed risk characterization for new GM crop varieties can be
identified (EFSA, 2010).
The Cartegena Protocol of Biosafety has recommended that risk assessment should be
science-based and should address issues relevant to society (UN, 2000). Engaging a wide
range of scientific expertise helps to generate relevant information during assessment of the
potential risks for introducing a GE crop (NRC, 2009; Dana et al., 2012; Gray 2012).
Specifically, involving experiential sources of knowledge during the problem context phase of
the problem formulation allows for a more comprehensive understanding of local ecologically
relevant sets of hazards.
In this study, we used this approach to identify the most relevant potential
environmental impacts of introducing GE crops in Uganda. Participants were identified based
on consultation with the national competent authority for regulating GE organisms in Uganda,
the National Council for Science and Technology (UNCST). A group of 35 individuals, who
provided a range of ecological, agronomic, biotechnological and biosafety expertise were
engaged to consider three examples representing different crops and traits currently under
development: weevil resistant sweetpotato, herbicide resistant cotton, and drought resistant
maize. This approach allowed a science-based approach to the identification and
prioritization of concerns most important for consideration during risk assessment for each
scenario.

142

Materials and Methods
Designing the framework and defining the boundaries and scope of the ERA
The ERA framework guiding this study was based on that of Wolt et al. (2010) with
modifications as suggested by Dana et al. (2012) (Figure 4.1). The boundaries and scope of
the ERA were defined using the guidance of international and national obligations that
Uganda has committed to implement to protect the environment.

Identifying the participants
Identification of stakeholders to participate in definition of problem context for the
ERA was done in consultation with the Uganda National Council for Science and Technology
(UNCST), which is the national competent authority for regulating GE organisms (Figure
4.2). A set of 39 participants was selected including representatives of the National Biosafety
Committee, institutional biosafety committees, Ministry of the Environment, the National
Environmental Management Authority and other scientists with relevant expertise. Of those,
35 agreed to participate in the survey. (Table 4.1)

Identifying the potential harm to biodiversity
Participants were engaged in a survey, which was conducted in Uganda in June-July
2010 using face-to-face interviews (16 respondents) or online responses (9). In the first phase
of the survey, the participants were presented with three examples of GE crop-trait scenarios
(weevil resistant sweetpotato, herbicide resistant cotton, and drought resistant maize) and
were provided with background information regarding the crop biology and relevant
engineered trait (Table 4.2A, B, and C, respectively). The participants were asked to describe
the management practices likely to be implemented for the three crops and to list potential
ways in which release of the GE varieties could harm biodiversity. The purpose of this

143

Identify boundaries and scope
Identify
participant
s

Conceptualize potential interactions
between stressors and environment

Problem context

Incorporate human
practices dimension
Identify hazards
Prioritize
hazards
Define assessment
endpoints

Problem definition

Generate risk
hypotheses
experimental designs
Risk estimation by
technical team

Reconvene diverse expertise
to deliberate on risk estimates
and risk acceptance criteria

Risk characterization

Risk evaluation

Figure 4-1 A risk assessment framework adapted from those of Wolt et al., (2010) and Dana
et al., (2012) for the ERA for introducing GE crops in Uganda. The gray triangle
schematically represents the level of diverse participation that should be done at each step. A
step for generating risk hypotheses and designing test experiments was added as suggested
by Raybould (2011).

144

Table 4-1 Summary of the profiles of the participants engaged in the ERA including their
professional affiliation, educational background and skills acquired through their work
experience and their regulatory roles
Professional
affiliation

Education
4
training

Work experience

Regulatory
role

Academia

Molecular
a
biology

Biotechnology, agriculture and
environmental regulation, policy
making

NBC

Academic

Molecular
a
biology

Biotechnology, plant genetics

TSC

Academic

Biochemistry

a

Biotechnology, environmental
science, molecular biology, plant
genetics, socio-economics research

NBC

Academic

Plant genetics

a

IPM, plant breeding

TSC

Academic

Entomology

Ecology, IPM

NBC

Academic

Molecular
b
biology

Ecology, plant genetic

TSC

Academic

Plant pathology

Molecular biology, plant diseases,
plant genetics

NBC

Academic

Agronomy

Plant agronomy

TSC

Agriculture
ministry

Plant
b
biotechnlogy

Biotechnology, agriculture and
environmental regulation, policy
making

Inspectorate

Agriculture
ministry

IPM

Agriculture and environmental
regulation, agriculture extension
services

Inspectorate

Agriculture
ministry

Plant pathology

Agriculture and environmental
regulation

NBC

a

Biotechnology, plant genetics,
plant conservation

TSC

Biotechnology, molecular biology

TSC

Biotechnology, agro-biodiversity
sustainable use and conservation,
environmental regulation, policy
making

NBC

CGIAR

1

b

a

a

b5

a

Plant breeding
a

8

CGIAR

Biochemistry

CGIAR

Entomology

CGIAR

Plant pathology

Biotechnology plant diseases,
policy making, science and
environmental regulation

TSC

Consumers'
protection NGO

Agricultural
b
extension

Agriculture extension services,
biotechnology, community
development consumer protection,
policy making

NBC

a

a

145

7

Table 4-1 (cont’d)
Professional
affiliation

Education
4
training

Work experience

Regulatory
role

Environment
ministry

Weed Science

Environmental science and
regulation

Inspectorate

Environment
ministry

Agronomy

Agriculture and environmental
regulation

Inspectorate

Environmental
NGO

Botany

Agriculture and environmental
regulation

TSC

Environmental
NGO

Agricultural
b
extension

Agriculture extension services,
community development,
environment conservation

TSC

Food safety
Bureau

Food
b
microbiology

Food safety regulation, policy
making

NBC

Food safety
Bureau

Food
b
microbiology

Food safety regulation

Inpectorate

Human health
ministry

Human
b
medicine

Biotechnology, environmental and
human health regulation, policy
making

NBC

NARO

Plant pathology

Biotechnology, agriculture and
environmental regulation, policy
making

IBC

NARO

Molecular
a
biology

Biotechnology, plant physiology

IBC

NARO

Plant genetics

Biotechnology, molecular biology

IBC

NARO

IPM

Entomology, IPM

IBC

NARO

Soil science

Soil fertility management,
biochemistry

IBC

NARO

Plant breeding

Plant breeding and genetics, IPM,
disease management

IBC

b

b

b

a

b

a
a
a

2

b

9

NARO

Food science

Food safety assessment

IBC

NEMA

Plant
b
conservation

Biotechnology and environmental
regulation, policy making

NBC

NEMA

Botany

Plant taxonomy, environmental
regulation

Inspectorate

NEMA

Zoology

Insect physiology, IPM

Inspectorate

NEMA

Botany

Environment conservation

Inspectorate

Plant
a
biotechnology

Biotechnology, microbiology,
agriculture and environmental
regulation

NBC

UNCST
1

b
b

c

3

Consultative Group on International Agricultural Research
146

Table 4-1 (cont’d)
2

National Agricultural Research Organization

3

Uganda National Council for Science and Technology

4

a

b

c

Highest degree completed: PhD, Masters, and Bachelors

5
6

Integrated pest management

7

8

National biosafety committee
Technical sub-committee
Institutional biosafety committee

147

Table 4-2A Characteristics of the crop and the engineered trait, which are relevant for an environmental risk assessment for weevil resistant
sweetpotato.
1
Weevil resistance
Sweetpotato, Ipomoea batatas (L.) Lam











The common method of propagation is vegetative propagation.
It is mainly an outcrossing species with self-incompatibility.
Most sweetpotato genotypes flower naturally in short-day in the
tropics.
Pollen can remain viable for up to 4 hours after pollination. Dispersal
is mainly by insects. No reports on pollen dispersal distances, and
level of out-crossing.
Seeds have a hard seed coat, which causes mechanical dormancy. As
a result germination is poor and irregular. Seedlings are thus rare.
Sweetpotato is a perennial that grows vigorously but it is easy to
control using cultural, mechanical or chemical methods.
Meso and South America are the primary centers of diversity. There
are wild relatives (Ipomoea spp.) are found in Uganda and Africa but
these are not sexually compatible with sweetpotato.

2

has large landrace diversity.


Bt crops have been cultivated in different parts of the
4
world for more than ten years . More than 21.7 million
Ha of Bt crops were grown in 2009. Before
commercialization all these crops were tested for
environmental and food safety.



Development of resistance by the pest to the Bt
5
proteins is known to occur .
Adoption of the refugia system can affect the
5
cropping system.
Bt toxins are highly specific but adverse effects may
6
occur to non-target insects in the same Order .
However, the impacts are considered minor relative
to those associated with the alternative of broad
5
spectrum insecticide applications .
There are no known adverse human health effects
3
associated with Bt proteins .
The product was developed by National Agricultural
Research Organization of Uganda, Auburn
3
University, USA and CIP .
At least two Bt proteins are inserted in two Ugandan
3
landrace varieties (Wagabolige and Tanzania) .






East Africa is considered a secondary center of diversity because it
2

Sweetpotato weevils may account for 60-100% yield loss depending
on the severity of attack, climatic conditions and susceptibility of the
variety.
148

Bacillus thuringiensis (Bt) is a naturally occurring
3
soil bacterium that produces insecticidal toxins .



Crossing between cultivated forms and wild relatives depends on
ploidy level. Intra-specific outcrossing is possible but success rates
are unpredictable.









Table 4-2A (cont’d)
1

CIP, 2009. Confined Field Trial application submitted to Uganda NBC. National Agricultural Research Organization in partnership with
International Potato Center.
2
Lebot V., 2010. Chapter 3: Sweetpotato. In: Bradshaw JE, editor. Root and Tuber Crops, Handbook of Plant Breeding. pp. 97–125.
3

CIP, 2009. Sweetpotato Action for Security and Health in Africa: Weevil resistant sweetpotato through biotechnology.

4

James C., 2009. Global Status of Commercialized Biotech/GM Crops: The first fourteen years, 1996 to 2009. ISAAA Brief 41-2009.

5

Zhao et al., 2003. Transgenic plants expressing two Bacillus thuringiensis toxins delay insect resistance evolution. Nature Biotechnology 12:
1493-1497
6
Bravo et al., 2007. Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon 49: 423-435

149

Table 4-2B Characteristics of the plant and the trait, which are relevant for an environmental risk assessment for herbicide tolerant cotton.
Cotton, Gossypium hirsutum













1

Round-up Ready herbicide tolerance

The common method of propagation is seed.
It is mainly a self-fertilizing species with less than 10% out-crossing
(insect pollination).
Pollen grains are large, heavy and sticky therefore wind pollination is
not possible. Pollen can stay viable for up to 8 hours. Level of insect
out-crossing observed after 7 meters was below 1%.
Seed can stay viable for 5 years. However, hard-seed coat trait
increasingly has been eliminated from commercial varieties because
it causes irregular and poor germination.
Some species in genus Gossypium have weedy tendencies.
The primary centers of genetic diversity for the genus Gossypium are
central and southern Mexico, north east Africa and Arabia, and
Australia.
The close wild relatives of G. hirsutum are found in Meso-America,
islands of the West Indies and islands in the Pacific.
Seed or pollen dispersal to feral cotton populations and some wild
relatives are the possible modes of gene escape.
Expansion of cotton cultivation in Uganda is constrained by
2

challenges of weed control .
1
2



3

Roundup Ready Flex® (RRF®) cotton is tolerant to

the glyphosate herbicides such as Roundup®.
 Most glyphosate chemicals are broad-spectrum, nonselective systemic herbicides that kill all plants
species.
 Glyphosate chemicals are the most commonly used
herbicides in Uganda.
RRF® trait reduces use of other chemicals which may
have high toxicity and residues left in the environment.
 Roundup herbicide has very low toxicity to humans.
 It is slightly toxic to birds and fish.
 It easily breakdown in soils and waterways to non-toxic
forms.


RRF® cotton has been released for commercial
production in different parts of the world. More than 1.1
4
million ha of HT cotton were grown in 2009 . Before
commercialization it was tested for environmental and
food safety.
 Product was developed by Monsanto.


Australian Office of Gene Technology Regulator, 2002. The Biology and Ecology of Cotton (Gossypium hirsutum) in Australia.

Owang, J.; Sekamatte, B.; Tindyebwa, A. 2005. Aspect on the organic cotton sub-sector in Uganda. Report on the ground situation of the
organic cotton production in selected areas of the Lango Sub-Region. 29p
150

Table 4-2B (cont’d)
3

Australia's Commonwealth Scientific and Industrial Research Organization (CSIRO), 2009. Bollgard II: the new generation of GM cotton.
Pesticide Information Project of Cooperative Extension Offices of Cornell University, Michigan State University, Oregon State University, and
University of California at Davis.
4
James, 2009. Global Status of Commercialized Biotech/GM Crops: The first fourteen years, 1996 to 2009. ISAAA Brief 41-2009.

151

Table 4-2C Characteristics of the plant and the trait, which are relevant for an environmental risk assessment for drought tolerant maize.
Maize, Zea mays











1

Drought tolerance

The common method of propagation is seed.
It is mainly an outcrossing species.
It produces large amounts of small pollen that is dispersed by wind.
Pollen can be carried for over long distances.
Maize produces large seeds that are dispersed by humans. Seed can
stay dormant in the soil and germinate in the next rainy season.
Maize does not have any weedy tendencies.
Mexico is the primary center of diversity.
It can cross with one of its closest wild relative, Tripsacum but it is
not found in Africa.
Pollen dispersal is the major mode of gene escape since there is great
sexual compatibility between maize varieties. However, exchange of
seed between farmers is also common.
Drought is a major constraint to maize production in Uganda and
2

Africa .
1










3

Introduced in maize from a bacterium to resist drought.
The gene enhances the ability of the crop to use water
more efficiently, at low soil moisture levels.
Some drought tolerant maize varieties have been
developed using conventional methods for example
ZM521 hybrid, which yields up to 50% more than
traditional varieties under drought conditions.
The impacts of the drought tolerance gene on maize
yields under sufficient rain conditions are still unclear.
Drought tolerance trait can increase fitness of a plant
species.
Out-crossing with weedy and wild invasive crops could
result in persistence of the trait in the environment.
GM maize was developed by a partnership between
International Maize and Wheat Improvement Center
(CIMMYT) and Monsanto.

Organization for Economic Co-operation and Development (OECD), 2003. Consensus Document on the Biology of Zea mays subsp. mays
(Maize).
2
Langyintuo, A.S., W. Mwangi, A.O. Diallo, J. MacRobert, J. Dixon, and M. Bänziger. 2008. An Analysis of the Bottlenecks Affecting the
Production and Deployment of Maize Seed in Eastern and Southern Africa. Harare, Zimbabwe: CIMMYT.
3
Edmeades G. O. 2008. Drought Tolerance in Maize: An Emerging Reality; Companion Document to Executive Summary ISAAA Briefs 39 –
2008.

152

activity was to provide the agronomic, biological, and environmental context in which to
conceptualize the interactions between the stressors and the different biodiversity components
that have potential to cause harm to ecologically valuable entities.
Responses from the first phase (Table 4.3) were used to identify the most frequently
cited potential harms. Based on these results, a structured questionnaire was developed to
obtain the participants’ views about the concerns likely to warrant greatest consideration
during the regulatory approval process for open release for the three crop-trait scenarios
(Table 4.4). The rating of likelihood of potential harm used a 5 point scale, with 1= very
likely and 5= very unlikely. In each case, participants were also asked to provide reasons for
their rating. In most cases (69%), the questionnaire was delivered by face to face interview.
In the remaining cases, participants responded via email. In the structured questionnaire, the
participants were also asked to provide information about professional affiliation, educational
disciplines, degrees received, areas of professional specialization, and regulatory roles.

153

Table 4-3 Potential adverse effects associated with interaction of GE crops with biodiversity,
and the number of respondents who listed the effects. Total number of respondents is 35.
Adverse effects

No. of respondents

Change in rate of chemical use

30

Development of resistance by pests (including weeds)

29

Loss of crop diversity

26

Effect on the beneficial organisms that interact with the modified
plants or the pests (including tri-tropic effects)

22

Changes in the crop management systems

20

Gene flow to wild relatives

13

Increased invasiveness of wild relatives causes loss of important
native plant species

10

Reduced plant species abundance and diversity affects the
beneficial organisms that depend on them for survival

10

Contamination of soil and water bodies, which affects the flora
and fauna in those habitats

9

Horizontal gene flow to soil microorganisms

7

Contamination of neighboring fields

6

Impact on soil types/structure, which affects soil microorganisms

4

154

Table 4-4 Section of the questionnaire1 that was used to identify the environmental concerns
that the participants thought were likely warrant great consideration prior to cultivation of
weevil resistant sweetpotato in Uganda. The survey was conducted June-July 2010.
Part B. Environmental concerns

1

An application for confined field testing of weevil resistance GE sweetpotato (please see
attached information about the crop and the modified trait) has been submitted for
regulatory review in Uganda. In your opinion, which of the following possible
environmental concerns warrant the greatest regulatory consideration before open release
of GM sweet-potato?
(Scale 1 to 5: 1-very unlikely, 2-Unlikely, 3-Uncertain, 4-Likely and 5-Very likely)
1

2

3

4

5

Reasons for the score

Change in rate of chemical use
Development of resistance by
pests to the trait
Loss of diversity of landraces or
other cultivated varieties
Non-target effects
Change in crop management
systems
Gene flow to weedy or wild plants
Other likely concerns
1

A similar question was asked for herbicide tolerant cotton and for drought tolerant maize.

155

Results and Discussion
This study utilized a participatory approach to involve scientific experts and
practitioners in an environmental risk assessment process for release of genetically engineered
crops in Uganda. The framework guiding this study was adapted from Wolt et al., (2010)
with modifications as suggested by Dana et al. (2012) (Figure 4.1). Diverse participation is
considered to be most important during the problem context stage of the risk assessment
process to obtain a more comprehensive understanding of the local ecologically relevant sets
of hazards (Gray, 2012; Dana et al., 2012). Participants were therefore engaged to assist in
definition of the problem context.
Defining the boundaries and scope of the analysis
Uganda has signed and ratified two international commitments to biodiversity
protection: the United Nations’ Convention on Biological Diversity (CBD) (UN, 1992) and
the Cartagena Protocol on Biosafety (UN, 2000). The CBD is a legally binding treaty that
requires parties to ensure conservation and sustainable use of biodiversity, as well as equitable
distribution of benefits from genetic resources. In 1992, Uganda signed the CBD and ratified
it in 1993 (http://www.cbd.int/convention/parties/list/). The Cartagena Protocol, which was
signed in 2000, includes obligations for safe transboundary movement of living modified
organisms while minimizing adverse effects on conservation and sustainable use of
biodiversity.
Objective 13 of the 1995 Constitution of the Republic of Uganda provides for
protection of valuable natural resources including flora and fauna biodiversity. At the same
time, in Objective 11 (ii) the Constitution obligates the State to stimulate agricultural,
industrial, technological and scientific development for sustainable use and management of
these natural resources so as to meet the development and environmental needs of present and
future generations of Ugandans. The National Environment Act of 1995 mandates the
156

National Environmental Management Authority to conduct environmental impact assessment
before any event likely to have an impact on the environment is undertaken, and to monitor
the impacts after the event.
The amended Plant Protection Act (1962) mandates the Department of Crop
Protection, under the Ministry of Agriculture, Animal Industry and Fisheries (MAAIF), to
prevent the introduction and spread of plant pests. The Act regulates the introduction of exotic
plants and microorganisms. The Uganda National Council for Science and Technology
(UNCST) Act (1990) mandates UNCST to oversee effective utilization of science and
technology for national development. Using its mandate, the UNCST was designated as the
competent authority for coordinating GE activities in the different sectors; it formulated the
national biosafety committee and spearheaded the process for enactment of the National
Biotechnology and Biosafety law to provide for an enabling environment for safe use of GE
products to address relevant national priorities.

Identifying the participants
Participants were invited to take part in a survey used to identify biodiversity
components that may be adversely affected by cultivation of GE crops in Uganda. We
identified practitioners and experts who have professional experience and knowledge of the
relevant subjects for environmental risk assessment, and have been involved in the regulatory
approval process for GE organisms in Uganda. (Figure 4.2).This activity was done in
consultation with the UNCST. Targeted stakeholder groups included: (1) members of the
national biosafety committee (NBC) who provide technical advice to Government on
biosafety issues, particularly with respect to the continued assessment of risks and benefits
associated with the production and/or application of biological materials; (2) members of
institutional biosafety committees (IBCs) from institutions that are conducting GE crop
157

Ministry of Agriculture
(involved in ERA review)

National Competent
Authority, UNCST
(appoints NBC)

Decision document

Import permit

National Biosafety Committee
(NBC)
(responsible to conduct/review
the ERA and make the decision)

Application
Product developer or supplier
(submits application)

Ministry of Environment
National focal Point
(involved in ERA review)
Institutional biosafety
committees (IBCs)
(may be recruited for ERA)

Research scientists
(recruited for ERA)

Authorization/rejection

Figure 4-2 An interim biosafety regulatory approval process for introduction of genetically
engineered organisms in Uganda. The gray rectangles highlight the groups invited to
participate in the ERA.
Source: Uganda National Council for Science and Technology; accessed in 2007.

158

research and development who are sometimes recruited by UNCST to participate in technical
subcommittees to make risk assessment and risk management recommendations; (3)
representatives of the Ministry of Water, Land and Environment, which is the national focal
point for the Biosafety Clearing House; (4) representatives of National Environmental
Management Authority, which is mandated to conduct EIAs for every event with potential to
harm the environment; (5) representatives from the Department of Crop Protection under the
Ministry responsible for agriculture, which is responsible for conducting pest risk analysis and
for agricultural import and export regulation; and (6) scientists who are recruited by the NBC
to provide technical assistance in evaluating the ERA findings.
The majority (90%) of the targeted participants agreed to participate in the survey
(Table 4.1). This group of 35 individuals represented a broad range of academic and
professional affiliations. Collectively, they provided diverse expertise including agronomy,
biotechnology, ecology, entomology, environmental science, food safety, human health, plant
genetics, and plant pathology, plus agricultural extension, regulatory and policy expertise.

Identifying potential harm to biodiversity
The participants were asked to consider possible ecological harm that may result from
interactions between GE crops and entities of ecological value. Three GE crops representing
different species and traits under consideration for future release in Uganda were included to
allow for contextualization of concerns on a case-by-case basis. Background information
about the characteristics of the crop and the engineered trait was provided for each of the
three GE crop-trait combinations (Table 4.2A, B and C). The participants were also asked to
consider the kind of crop management practices that the farmer would implement in each crop
case to provide the agronomic context for the environmental risk considerations. Practices
identified included adoption of modern varieties or hybrids, farmers seed exchange, soil
159

tillage practices, soil fertility management, crop rotation, intercropping, irrigation, pesticide
application, weed control and harvesting practices.
Based on the participants’ responses a list of potential adverse effects was compiled
(Table 4.3). Concerns listed by at least 1/3 of the respondents were: change in rate of
chemical use, development of resistance by the pests, loss of crop biodiversity, non target
effects, changes in the crop management systems and transgene flow to wild relatives.
In the second phase of the study, a questionnaire was designed to explore which of
these top six environmental concerns were considered likely for each crop-trait scenario,
(Table 4.4). Using a rating scale of 1-5, the participants were asked to indicate which of the
postulated adverse effects warrant the greatest regulatory consideration before permitting
open cultivation of the GE crop and to indicate reasons for their rating. Comparison of ratings
showed variation in relative scoring of the different environmental concerns for the three
different scenarios as described below. (Figure 4.3, 4.4 and 4.5 respectively).

Weevil resistant sweetpotato
The item receiving the greatest concern for weevil resistant sweetpotato (rated likely
or very likely by 63% of the respondents) was development of resistance by the pests to the Bt
toxins. (Figure 4.3). The concern that pests are likely to develop resistance to Bt toxins has
received a good deal of attention in recent literature (Bravo and Sobero´n 2008, Onstad 2008;
Tabashnik et al. 2008). Development of resistance by pests is a common phenomenon
connected with application of pesticides in general (Newcomb et al. 1997; Van Leeuwen et
al., 2008; 2012; Heckel, 2012; Guillette and Iguchi, 2012). Development of resistance by
diamondback moth to Bt foliar spray product was reported in earlier studies (Bates et al.,
2005; Sarfraz et al., 2006). Recent literature reports evidence of development of resistance

160

25

Development of resistance
by the pests
TSC
Inspectorate
IBC
NBC

25

Loss of crop diversity

25

20

20

15

15

10

10

10

5

5

5

0

0

Change in crop
management system

20
15

Likely
25

Change in rate of pesticide
use

0
Likely

Uncertain UnLikely
25

Uncertain UnLikely

Non-target effects

Likely
25

20

20
15

15

10

10

10

5

5

5

0

0

Gene flow to wild relatives

20

15

Uncertain UnLikely

Likely

Uncertain UnLikely

0
Likely

Uncertain UnLikely

Likely

Uncertain UnLikely

Figure 4-3 Scoring of concerns associated with introduction of weevil resistant sweetpotato in
Uganda by the different regulatory categories. TSC stands for technical subcommittees, NBC,
national biosafety committee, IBC, institutional biosafety committee. Total number of
respondents was 35. Scores of very likely and likely were combined to represent “likely”, and
scores of very unlikely and unlikely were combined to represent “unlikely”.

161

25
20

Development of
resistance by the pests
TSC
Inspectorate
IBC
NBC

25

Loss of crop diversity

25

20

20

15

15

10

10

10

5

5

Change in crop
management system

5

15

0

0
25

Change in rate of
pesticide use

0
Likely Uncertain UnLikely

Likely Uncertain UnLikely

Likely Uncertain UnLikely
25

Non-target effects

25

20

20

20

15

15

15

10

10

10

5

5

Gene flow to wild
relatives

5

0

0

0
Likely Uncertain UnLikely

Likely Uncertain UnLikely

Likely Uncertain UnLikely

Figure 4-4 Scoring of concerns associated by introduction of herbicide tolerant cotton in
Uganda by the different regulatory categories. TSC stands for technical subcommittees, NBC,
national biosafety committee, IBC, institutional biosafety committee. Total number of
respondents was 35. Scores of very likely and likely were combined to represent “likely”, and
scores of very unlikely and unlikely were combined to represent “unlikely”.

162

35

Development of
resistance

35

Loss of crop diversity

35

30

30

25

25

20

20

15

15

15

10

10

10

5

5

Change in crop
management system

5

30

TSC
Inspectorate
IBC
NBC

25
20

0

0
Likely
35

Change in rate of
pesticide use

0
Likely

Uncertain UnLikely
35

Uncertain UnLikely

Non-target effects

Likely
35

30

30

25

25

20

20

20

15

15

15

10

10

10

5

5

Gene flow to wild
relatives

30

25

Uncertain UnLikely

5

0

0
Likely

Uncertain UnLikely

0
Likely

Uncertain UnLikely

Likely

Uncertain UnLikely

Figure 4-5 Scoring of concerns associated by introduction of drought tolerant maize in
Uganda by the different regulatory categories. TSC stands for technical subcommittees, NBC,
national biosafety committee, IBC, institutional biosafety committee. Total number of
respondents was 35. Scores of very likely and likely were combined to represent “likely”, and
scores of very unlikely and unlikely were combined to represent “unlikely”.

163

by pests to Bt toxins expressed in GE crops (Gray, 2011; Hodgson and Gassmann, 2011;
Downes and Mahon, 2012; Mahon et al., 2012). Although not introducing a new
environmental harm, development of resistance is considered an environmental concern if
planting of the GE crop allows for reduced pesticide application; loss of the efficacy of the
GE crop would mean returning to the lesser environmentally benevolent options for pest
control (Quemada, 2011).
The majority of the respondents indicated that loss of crop diversity is a likely
environmental concern (25%) or that they need more information before they can decide
(55%). The uncertain group indicated that the manner in which the GE crop is introduced (i.e.,
whether it is extensively promoted) will influence the rate of adoption and displacement of
other varieties. The respondents who thought that this was a likely concern thought that
weevil resistant sweetpotato is likely to displace the diverse landrace varieties because it
addresses one of the major production constraints in Uganda. The country possesses
extensively diverse landraces, some of which have adapted to specific local environments
(Yada 2009; Zawedde, Chapter 2). Maintaining such varietal diversity is considered important
especially in developing countries where farmers rely on it for food security (FAO, 2007).
Varietal diversity also is critical for further crop improvement (Lu, 2008). Significant
displacement of local varieties by GE cultivars has been observed in China where Bt cotton
now accounts for approximately 95% of the crop production in the northern region (Lu et al.,
2012). Similarly, Bt cotton was reported initially to cause a significant reduction in the
varietal diversity in India due to incorporation of the new traits in a small number of the local
varieties, however in this case, the concern was substantially offset over time by introducing
the trait into other popular varieties (Carpenter, 2011).
Some respondents (43%) indicated that loss of crop diversity was a likely concern
because cultivation of weevil resistant sweetpotato may result in changes in crop management
164

practices that may not be favorable for cultivation of other varieties or other crops. Possible
changes indicated by the respondents included implementation of a refuge requirement to
minimize evolution of resistance to the Bt toxin, and potential increase in mono-cropping. It
was considered that use of refugia could increase land devoted to sweetpotato, and so may
decrease diversity of plots planted to other crops. If mono-cropping were to occur, it would
reduce crop diversity and can also result in increased incidence of diseases and other pests
(Evans, 2003).
Only 29% of the respondents considered changes in rate of chemical use a likely
concern; however 35% of the group were uncertain of the impacts of these changes. Both
groups indicated that this would be a concern if reduced use of pesticides results in emergence
of secondary pests. There is evidence that the reduced application of broad spectrum
pesticides facilitated by cultivation of Bt crops in China has led to increased abundance of
secondary pests (Naranjo et al., 2008; Wang et al., 2010). Two respondents indicated that
there may be an increase in chemical use due to development of resistance by the weevils if
the farmers had not been controlling the weevils by chemical means prior to adoption of the
Bt crop, but after the use of the Bt crop began to use chemical controls.
More than half of the respondents (54%) thought that non-target effects were not a likely
concern, although 12% thought that it was likely and 34% were uncertain of the impact. Most
respondents indicated that is not a likely concern because the negative impacts on non-target
organisms due to the weevil resistant sweetpotato are lower than the impacts expected from
the use of broad-spectrum pesticides. However, only a small number of farmers (17%) are
using pesticides for sweetpotato pests’ control, majority of farmers rely on host resistance.
The respondents, who thought that non-target effects were likely, indicated that there are
beneficial Coleoptera insects in Uganda. The uncertain group indicated that it will depend on
Bt toxin toxicity levels in the plants and their impacts on the beneficial organisms that interact
165

with the modified plants or the affected insects, i.e.., do the beneficial insects come in direct
contact with the Bt crop or do they eat sweetpotato weevil larvae. While some negative nontarget effects of Bt toxins have been detected in the laboratory (Hilbeck et al., 1998, 1999;
Losey et al., 1999; Ponsard et al., 2002), a large number of field studies used to assess
arthropod non-target effects of Bt crops have demonstrated no effect (Marvier et al., 2007;
Wolfenbarger et al., 2008; Naranjo, 2009; Duan et al., 2010). Other field studies have
recorded increase in biological diversity due to the reduction in pesticide application in Bt
crops’ fields (Dively, 2005; Obrist et al., 2006; Marvier et al., 2007).
Finally, the majority of the participants (62%) indicated that transgene flow to wild relatives
is unlikely. This is consistent with the information provided that there are not sexually
compatible wild relatives of sweetpotato in Uganda (CIP, 2009).

Herbicide tolerant cotton
Most respondents (60%) thought that planting of herbicide resistant cotton will result in
development of weeds resistant to the herbicide, possibly accelerated by chemical misuse
(Figure 4.4). This could result in adoption of application of more hazardous herbicides.
Development of resistance by weeds to herbicides is a common incident in agriculture with
more than 300 herbicide resistant weed biotypes reported in over 190 plant species (Heap
2010; Owen et al., 2011). Over 21 glyphosate-resistant weed species in 15 countries have
been recorded where glyphosate resistant crops are grown (Carpenter, 2011). Development of
resistance to herbicides by weeds is “an inevitable incidence” that can at best be delayed in
systems where herbicides are applied (Ervin et al., 2010; Owen et al., 2011).
The majority of the respondents (66%) considered loss of crop diversity an unlikely
environmental concern. Cotton landraces are not grown in Uganda, and only one variety is
grown in the country (Horna et al., 2009). It is this variety that this will be transformed to
166

incorporate the herbicide tolerance transgene (NARO application submitted for CFT
approval) therefore there is not an anticipated change in level of diversity of the cotton crop.
Most respondents (60%) indicated that changes in crop management practices was an unlikely
concern because most farmers are already growing cotton as a monocrop and they apply
herbicides for weed control. However, a small group of the respondents (20%) indicated that
cultivation of the herbicide tolerant cotton, which is accaompanied by increased herbicide
application, will affect farming systems where cotton is grown in a mixed cropping system.
This is because if you put one glyphosate tolerant crop in a mixed cropping system and spray
the glyphosate you kill all the other crops. The increased usage of glyphosate herbicide
application anticipated with cultivation of the GE crop was considered a likely concern by
52% of the respondents, especially if it is not accompanied with proper education to minimize
chemical misuse. However, another segment of the respondents (32%) indicated that this is an
unlikely concern because herbicides are already applied by cotton farmers, and adoption of
herbicide resistant cotton may be accompanied by an educational requirement, which will
reduce chemical misuse.
There was no consensus on the possibility of non-target effects for the glyphosate
tolerant cotton. Respondents who thought that non-target effects were likely (32%) indicated
that the anticipated increase in herbicide use will adversely affect other flora species in the
agricultural fields and surrounding natural habitats. The uncertain group (32%) indicated that
it will depend on the current herbicide application regimes, and how they change due to
adoption of herbicide tolerant cotton. Non-target effects have been reported for arthropods
that depend on the weeds in herbicide tolerant crop field (Chassy et al., 2003; Freckleton et
al., 2003). However, this effect was observed for any weed control measure adopted because
the abundance of most arthropods will depend on abundance of other host plant species
(Freckleton et al., 2003).
167

Finally, the majority of the respondents (66%) indicated that transgene flow to wild
relatives is unlikely to be a concern because close relatives of cotton are not known to be
weedy, and there are no reports of presence of wild/weedy wild relatives of cotton in Uganda
(NARO application submitted for CFT approval). The close wild relatives of Gossypium
hirsutum, which is the species cultivated in Uganda, are found in Meso-America, islands of
the West Indies and islands in the Pacific (OGTR, 2008). Another Gossypium species
(Gossypium herbaceum) has its center of origin in north east Africa and southern Arabia but
there are no reports of its wild/weedy relative in Uganda (Leipzig, 1996).

Drought tolerant maize
Most environmental concerns were not considered likely by the majority of the
participants except for loss of crop diversity and changes in cropping systems (Figure 4.5).
For both concerns, participants indicated that the impacts will depend on farmers’ decisions
on whether to maintain other varieties/crops or change the crop management practices after
they adopt the drought tolerant maize. The majority of respondents (69%) indicated that they
anticipated widespread adoption of these cultivars because drought causes the highest maize
yield losses and there is a growing increase in commercial maize production in the country.

Conclusion
Using a participatory framework for problem formulation, we were able to identify the
biodiversity components of the receiving environment with the greatest potential to be
adversely affected by cultivation of weevil resistant sweetpotato, herbicide tolerant cotton and
drought tolerant maize. Using the structured questionnaire, and tapping into the broad
scientific and agronomic expertise of the stakeholders, we were able to identify the most
important concerns associated with each crop-trait scenario. From these findings, we were
168

able to prioritize the risks that should receive greatest attention in performing environmental
risk assessments for these crops (Table 4.5).
Evaluation of the potential impacts of GE crops requires a good understanding of the
current level of biodiversity and the crop management systems. It must also take into account
anticipated rate of adoption and changes in the management system based on knowledge of
the relevant farming and agricultural extension systems. The ratings of perceived risks of the
different concerns were generally consistent among representatives of the different regulatory
groups. There were a few exceptions, however especially with the weevil resistant
sweetpotato case. Inspectors were more inclined to rate non-target effects and gene flow to
wild relatives as “likely” concerns. IBC members and research scientists were less concerned
about the changes in crop management systems

Scientists were also less likely to be

concerned about development of resistance by the pests. This may reflect the relative
knowledge-base and hands-on experiences of the different groups, indicating the importance
of bringing together multiple kinds of expertise.
The variation in the scoring of the environmental concerns for each of the crop-trait
scenarios indicates that the regulators evaluated the GE crops on a case-by-case basis. The
number of uncertain respondents observed for the various parameters suggests areas in which
additional information is needed. In the case where such information exists, measures should
be put in place to ensure timely and effective provision of relevant information. In cases
where such information does not exist, this analysis allows for identification of aspects that
may require further data collection or analysis. Further studies should identify the knowledge
gaps and most appropriate information delivery channels for each group.

169

Table 4-5 Participant – identified potential risks that should receive greatest attention in
performing environmental risk assessments for the three crop-trait scenarios in Uganda.
---------------------------------------------------------------------------------------------------------------Weevil resistant sweetpotato:
Development of resistance of the weevils to Bt toxins
Loss of varietal diversity
Changes in the crop management system leading to loss of sweetpotato varietal
diversity, loss of diversity of other crops and increase in secondary pests
Herbicide tolerant cotton:
Development of resistance of the weeds to glyphosate
Changes in chemical application rates
Non-target effects
Drought tolerant maize
Loss of varietal diversity
Change in crop management systems leading to loss of maize varietal diversity and
diversity of other crops in the maize fields.
-----------------------------------------------------------------------------------------------------------------

170

In summary, this approach to problem formulation helps to focus risk assessment
efforts on the most important factors. The inclusion of the practitioners’ expertise helped to
translate the broad environmental policy goals embedded in national and international
legislation into relevant concerns appropriate to the receiving environment. Utilizing a wide
range of expertise at the problem context was invaluable to identify and prioritize relevant
hazards and potential harms. This process began by developing a comprehensive list of
possible concerns and then identifying the most important potential harms appropriate to the
individual crop, trait and receiving environment. Furthermore, by including representatives
of multiple responsible agencies and regulatory bodies (i.e., the Ministry of Agriculture,
Ministry of the Environment, Biosafety Committees), this approach also improves
transparency of the ERA process.

171

LITERATURE CITED

172

LITERATURE CITED

Baffes J. 2009. The Cotton Sector Of Uganda. Africa Region Working Paper Series No. 123.
Paper prepared for the World Bank Group. p.30
Bates, S. L., J. Z. Zhao, R. T. Roush, and A. M. Shelton. 2005. Insect resistance management
in GM crops: past, present and future. Nat. Biotechnol. 23: 57-62
Bravo, A., and M. Sobero´n. 2008. How to cope with insect resistance to Bt toxins? Trends
Biotechnol. 26: 573-579.
Carpenter JE. 2011. Impact of GM crops on biodiversity. GM Crops 2:7-23.
CBD. 2011. "Parties to the Protocol and Signatories to the Supplementary Protocol."
Available at http://bch.cbd.int/protocol/parties/
Chassy B, Carter C, McGloughlin M, McHughen A, Parrott W, Preston C et al. (2003) UK
field-scale evaluations answer wrong questions. Nat Biotechnol. 21:1429–1430
CIP 2009. Confined Field Trial application submitted to Uganda NBC. National Agricultural
Research Organization in partnership with International Potato Center.
CIP, 2008. Weevil resistant sweetpotato through biotechnology. Sweetpotato Action for
Security and Health in Africa. International Potato Center.
CDO, 2006. The cotton sector in Uganda: Progress made in the sector and recommendations
for achieving further progress. Cotton Development Organisation, pp 1-14.
Dana G. V.; Kapuscinski A. R.; Donaldson J. S. 2012. Integrating diverse scientific and
practitioner knowledge in ecological risk analysis: A case study of biodiversity risk
assessment in South Africa. J. Environ. Manage. 98: 134–146
Dively GP, 2005. Impact of transgenic VIP3A × Cry1Ab Lepidopteran-resistant field corn on
the nontarget arthropod community. Environmental Entomology 34: 1267–1291.
Downes S, Mahon R, 2012. Evolution, ecology and management of resistance in Helicoverpa
spp. to Bt cotton in Australia. J Invertebr Pathol 110:281-6
Duan JJ, Head G, McKee MJ, Nickson TE, Martin JW, Sayegh FS. 2002. Evaluation of
dietary effects of transgenic corn pollen expressing Cry3Bb1 protein on a non-target ladybird
beetle, Coleomegilla maculata. Ent Exp Appl 104:271–280.
Duke SO, Powles SB, 2008. Glyphosate: a once-in-a-century herbicide. Pest Manag Sci
64:319–325
Edmeades G. O. 2008. Drought Tolerance in Maize: An Emerging Reality; Companion
Document to Executive Summary ISAAA Briefs 39.
EFSA 2010. Guidance on the environmental risk assessment of genetically modified plants.
European Food Safety Authority Panel on Genetically Modified Organisms (GMO), 100pp.
173

Ekobu, M., Solera, M., Kyamanywa, S., Mwanga, R.O.M., Odongo, B., Ghislain, M., Moar,
W.J., 2010. Toxicity of seven Bacillus thuringiensis Cry proteins against Cylas puncticollis
and Cylas brunneus (Coleoptera: Brentidae) using a novel artificial diet. Journal of Economic
Entomology 103: 1493-1502.
EPA's Regulation of Bacillus thuringiensis (Bt) Crops, 735-F-02-013
May 2002, http://www.epa.gov/pesticides/biopesticides/pips/regofbtcrops.htm
Ervin DE, Carriere Y, Cox WJ, Fernandez-Cornejo J, Jussaume RA Jr, Marra MC, Owen
MDK, Raven PH, Wolfenbarger LL, Zilberman D. 2010. The impact of genetically
engineered crops on farm sustainability in the United States. National Research Council,
Washington, DC
Evans, P. 2003. Development as institutional change: The pitfalls of monocropping and
potentials of deliberation. Manuscript, Department of Sociology,University of California,
Berkeley
FAO, 2007. Report of Panel of Eminent Experts on Ethics in Food and Agriculture. Food and
Agriculture Organization of United Nations, Rome. November 2007.
Fedoroff NV, Battisti DS, Beachy RN, Cooper PJM, Fischoff DA, Hodges CN, Knauf VC,
Lobeii D, Mazur BJ, Molden D, Reynolds MP, Ronald PC, Rosegrant MW, Sanchez PA,
Vonshak A & Zhu J-K 2010. Radically rethinking agriculture for the 21st century. Science
327: 833-834.
Freckleton RP, Sutherland WJ, Watkinson AR (2003) Deciding the future of GM crops in
Europe. Science 302:994–996
Gassmann, Aaron J., Jennifer L. Petzold-Maxwell, Ryan S. Keweshan, and Mike W. Dunbar.
(2011). Field-Evolved Resistance to Bt Maize by Western Corn Rootworm. PLoS ONE 6 (7):
p e22629
Godfray HCJ, Beddington JR, Crute I, Haddad L, Lawrence D, Muir JF, Pretty J, Robinson S,
Thomas SM & Toulmin C 2010. Food security: The challenge of feeding 9 billion people.
Science 327: 812-818.
Gray A., 2012. Problem Formulation in Environmental Risk Assessment for Genetically
Modified Crops: A Practitioner’s Approach International Centre for Genetic Engineering and
Biotechnology (ICGEB) Collection of Biosafety Reviews 6: 10-65
Gray, M. 2011. Severe Root Damage to Bt Corn Observed in Northwestern Illinois. Available
at http://bulletin.ipm.illinois.edu/article.php?id=1555
Grüneberg W., R. Mwanga, M. Andrade, and H. Dapaah 2009. Unleashing the potential of
sweetpotato in Sub-Saharan Africa: Current challenges and way forward. CIP Social
Sciences Working Paper.
Guillette, L., & Iguchi, T. (2012). Life in a contaminated world. Science 337:1614–1615.
Heap, I. 2009. International survey of herbicide resistant weeds. http://www.weedscience.com
Accessed: January 8, 2009.
174

Heckel D.G., 2012. Insecticide Resistance After Silent Spring. Science 337:1612-14
Hilbeck A, Moar WJ, Pusztai-Carey M, Filippini A & Bigler F(1999) Prey-mediated effects
of Cry1Ab toxin and protoxin and Cry2A protoxin on the predator Chrysoperla
carnea.Entomologia Experimentalis et Applicata 91: 305–316.
Hill RA 2005. Conceptualizing risk assessment methodology for genetically modified
organisms. Environmental Biosafety Research 4: 67-70.
Hodgson, E., and A. Gassmann. (2011). "First Iowa Confirmation of Resistance to Bt Corn by
Western Corn Rootworm." Corn & Soybean Digest.
http://cornandsoybeandigest.com/corn/first-iowa-confirmation-resistance-bt-corn-westerncorn-rootworm
Hofte H, Whiteley HR. 1989. Insecticidal crystal proteins of Bacillus thuringiensis.
Microbiol Rev 53:242–55.
Horna D., M. Kyotalimye, J. Falck-Zepeda, 2009. Cotton Production in Uganda: Would GM
technologies be the Solution? Paper prepared for presentation at the International Association
of Agricultural Economists Conference, Beijing, China, August 16-22, 2009.p. 28
James C 2012. Global Status of Commercialized Biotech/GM Crops: 2010. ISAAA Brief 44.
Available at http://www.isaaa.org/resources/publications/briefs/44/executivesummary/
James C., 2009. Global Status of Commercialized Biotech/GM Crops: The first fourteen
years, 1996 to 2009. ISAAA Brief 41-2009.
Klinger T., “Variability and uncertainty in crop-to-wild hybridization,” in Genetically
Engineered Organisms: Assessing Environmental and Human Health Effects, D. K.
Letourneau and B. E. Burrows, Eds., CRC Press, Boca Raton, Fla, USA, 2002.
Langyintuo, A.S., W. Mwangi, A.O. Diallo, J. MacRobert, J. Dixon, and M. Bänziger. 2008.
An Analysis of the Bottlenecks Affecting the Production and Deployment of Maize Seed in
Eastern and Southern Africa. Harare, Zimbabwe: CIMMYT.
Leipzig,1996. Uganda: Country Report to the FAO International Technical Conference On
Plant Genetic Resources. Leipzig, Germany, 17-23 June 1996.p.38
Losey JE, Rayor LS & Carter ME (1999) Transgenic pollen harms monarch larvae. Nature
399: 214–214.
Lu Y, K. Wu, Y. Jiang, Y. Guo and N. Desneux 2012. Widespread adoption of Bt cotton and
insecticide decrease promotes biocontrol services Nature doi:10.1038/nature11153
Mahon R, Olsen K, Downes S, Addison S, 2007. Frequency of alleles conferring resistance to
the Bt toxins Cry1Ac and Cry2Ab in Australian populations of Helicoverpa
armigera (Lepidoptera: Noctuidae). J Econ Entomol 100: 1844–1853
Naranjo S. E. 2009 Impacts of Bt crops on non-target organisms and insecticide use
patterns. Perspect. Agr. Vet. Sci. Nutr. Nat. Resour. 4: 1–23.

175

National Agricultural Statistics Service.National agricultural statistics service. Available at
(verified 10 Sept. 2010). USDA-NASS, Washington, D.C. National agricultural statistics
service. Available at http://www.nass.usda.gov/ (verified 10 Sept. 2010). USDA-NASS,
Washington, D.C. 2009
Newcomb, R.D., Campbell, P.M., Ollis, D.L., Cheah, E., Russel, R.J.and Oakeshott, J.G.,
1997. A single amino acid substitution converts a carboxylesterase to an organophosphorus
hydrolase and confers insecticide resistance on blowfly. Proc Natl Acad Sci USA 94: 7464–
7468.
Nickson T.E. 2008. Planning Environmental Risk Assessment for Genetically Modified
Crops: Problem Formulation for Stress-Tolerant Crops. Plant Physiology, 147: 494–502
NRC, 2009. Science and Decisions. National Research Council, The National Academies
Press 403pp.
Obrist, L. B., A. Dutton, J. Romeis, and F. Bigler. (2006a). "Biological activity of Cry1Ab
toxin expressed by Bt maize following ingestion by herbivorous arthropods and exposure of
the predator Chrysoperla carnea." BioControl 51 (1): p 31-48.
OECD 1993. Safety considerations for biotechnology: Scale-up of crop plants.Organisation
for Economic Cooperation and Development (OECD), Paris,France. Available at
http://www.oecd.org/dataoecd/26/261958527.pdf.
Onstad, D.W. 2008. Insect resistance management: biology, economics, and prediction.
Academic, London, United Kingdom.
Owang, J.; Sekamatte, B.; Tindyebwa, A. 2005. Aspect on the organic cotton sub-sector in
Uganda. Report on the ground situation of the organic cotton production in selected areas of
the Lango Sub-Region. 29p
Owen, M.J.; Michael, P.J.; Renton, M.; Steadman, K.J.; Powles, S.B. Towards large-scale
prediction of Lolium rigidum (annual ryegrass) emergence. II. Correlation between dormancy
and herbicide resistance levels suggests an impact of cropping systems. Weed Res. 2011, 51,
133–141.
Ponsard S, Gutierrez AP & Mills NJ (2002) Effect of Bt-toxin (Cry1Ac) in transgenic cotton
on adult longevity of four heteropteran predators. Environmental Entomology 31: 1197–1205.
Quemada H, 2011. Environmental biosafety issues associated with genetically engineered
crop. In: Environmental safety of genetically engineered crops. Grumet R, Hancock J,
Maredia K, Weebadde C (eds). Michigan State Univ. Press, East Lansing, p. 31-46.
Raybould AF 2011. The bucket and the searchlight: Formulating and testing risk hypotheses
about the weediness and invasive potential of transgenic crops. Environmental Biosafety
Research 9: 123-133.
Republic of Uganda. (2008). National Biotechnology and Biosafety Policy. Kampala:
Ministry of Finance, Planning and Economic Development.
Republic of Uganda. (2010). National Development Plan 2010/11 – 2014/15. Kampala:
Ministry of Finance, Planning and Economic Development.
176

Romeis J, Bartsch D, Bigler F, Candolfi MP, Gielkens MMC, Hartley SE, Hellmich RL,
Huesing JE, Jepson PC, Layton R, Quemada H, Raybould AF, Rose RI, Schiemann J, Sears
MK, Shelton AM, Sweet J, Vaituzis Z & Wolt JD 2008. Assessment of risk of insect-resistant
transgenic crops to nontarget arthropods. Nature Biotechnology 26: 203-208.
Sarfraz M., Dosdall L.M., Keddie B.A. (2006) Diamondback moth-host plant interactions:
Implications for pest management. Crop Prot. 25:,625–639.
Tabashnik B, Gassman A, Crowder D, Carriere Y (2008). Insect resistance to Bt crops:
evidence versus theory. Nat. Biotechnol. 26
Tester M & Langridge P 2010. Breeding technologies to increase crop production in a
changing world. Science 327: 818-822.
Thierry B., S. Heuberger, M. Zhang, C. Ellers-Kirk, X. Ni, L. Masson, X. Li, B.E. Tabashnik,
and Y. Carrière 2013. Potential shortfall of pyramided transgenic cotton fori insect resistance
management. Proc Natl Acad Sci USA 110 (15): 5806–5811
UN, 1992. Convention on Biological Diversity: text and annexes. Montreal: Secretariat of the
Convention on Biological Diversity, United Nations.
UN, 2000. Cartagena Protocol on Biosafety to the Convention on Biological Diversity: text
and annexes. Montreal: Secretariat of the Convention on Biological Diversity, United
Nations.
USEPA 1998. Guidelines for Ecological Risk Assessment. EPA/630/R-95/002F. April 1998
Final. United States Environmental Protection Agency (USEPA), Washington DC, USA.
Available at http://www.epa.gov/raf/publications/pdfs/ECOTXTBX.PDF.
Van Leeuwen T, et al. 2012. Population bulk segregant mapping uncovers resistance
mutations and the mode of action of a chitin synthesis inhibitor in arthropods. Proc Natl Acad
Sci USA109(12):4407–4412.
Veasey EA, Borges A, Rosa MS, Queiroz-Silva JR, Bressan EDA, Peroni N. 2008. Genetic
diversity in Brazilian sweetpotato (Ipomoea batatas (L.) Lam., Solanales, Convolvulaceae)
landraces assessed with microsatellite markers. Genet. Mol. Biol. 31:725–33
Wang YM, Zhang GA, Du JP, Liu B, Wang MC. 2010. Influence of transgenic hybrid rice
expressing a fused gene derived from cry1Ab and cry1Ac on primary insect pests and rice
yield. Crop Prot. 29:128–33
Williams, G.M., Kroes, R., & Munro, I.C. (2000). Safety evaluation and risk assessment of
the herbicide roundup and its active ingredient, glyphosate, for humans. Regulatory
Toxicology and Pharmacology, 31(2):117-165.
Wolfenbarger, L. LaReesa, Steven E. Naranjo, Jonathan G. Lundgren, Royce J. Bitzer, and
Lidia S. Watrud. (2008). "Bt Crop Effects on Functional Guilds of Non-Target Arthropods: A
Meta-Analysis." PLoS ONE 3 (5): p e2118

177

Wolt JD, Keese P, Raybould AF, Fitzpatrick JW, Burachik M, Gray AJ,Olin SS, Schiemann
J, Sears M & Wu F 2010. Problem formulation in the environmental risk assessment for
genetically modified plants. Transgenic Research 19:425-436.
Zhang D.P., Golmirzaie A., Cipriani G., Panta A., Ghislain M., Smit N., Rety I., and Michaud
D. 1996. Developing weevil resistance in sweetpotato with genetic transformation.
International Potato Center (CIP) Program Report 95-96.
Zhao JZ, Cao J, Li YX, Collins HL, Roush RT, Earle ED, Shelton AM(2003) Transgenic
plants expressing two Bacillus thuringiensis toxins delay insect resistance evolution. Nat
Biotechnol 21:1493–1497.

178

CONCLUSIONS AND FUTURE DIRECTIONS
PROBLEM FORMULATION FOR THE ENVIRONMENTAL RISK ASSESSMENT
OF GE WEEVIL RESISTANT SWEETPOTATO

Introduction
Over the years, many problem formulation studies have been conducted for GE crops
that have provided frameworks for conducting relevant environmental risk assessments
(ERAs) for particular trait (s) and crops (Raybould and Cooper, 2005; Lu, 2008; Nickson,
2008; Romeis et al., 2008; Hokanson et al., 2010; Wolt et al., 2010; Raybould, 2011; 2012;
Gray 2012). Information about the crop biology, the modified trait (s) and the receiving
environment is used to answer the following questions: “can we envision a way in which
valuable entities can be harmed?” and “what is the probability that valuable entities will be
exposed to detrimental amounts of the stressor in the GE organism under the proposed
receiving environment?” (Gray, 2012). Diverse consultations may be used to identify possible
sources of the harm, which addresses the first question (Dana et al., 2012; Gray, 2012;
Zawedde, Chapter 3).
Addressing the second question requires construction of exposure scenarios. These are
best described in the form of a conceptual model providing a clear a sequence of events that
show how the valuable entity is likely to be exposed to the stressor(s) (Raybould, 2011).
Each step in the sequence can be expressed in the form of a risk hypothesis. These risk
hypotheses form the basis for an analysis plan because they portray the kinds of information
necessary for a comprehensive ERA. In many cases the evidence necessary to falsify the risk
hypotheses will exist in previous studies. The analysis plan, which is the last step in problem
formulation, will involve providing evidence to address some of the risk hypotheses and,
where necessary, suggest the additional kinds of data and experimental designs that are
required to assess the identified risk.
179

The diverse consultations conducted (Zawedde, Chapter 3) identified development of
resistance by weevils to Bt toxins, loss of sweetpotato varietal diversity, and changes in crop
management practices leading to reduced crop diversity or emergence of secondary pests as
the potential concerns that warrant primary consideration in the regulatory approval process
for open cultivation of weevil resistant sweetpotato in Uganda. These concerns will each be
examined in turn. The information generated so far in this work, will aid in evaluating the
probability that weevil resistant sweetpotato will have these potential detrimental effects.

Development of resistance to Bt toxins by sweetpotato weevils
Development of resistance by pests is a common phenomenon associated with
pesticide use, and has also been confirmed to occur in some cases for Bt toxins (Gray, 2011;
Hodgson and Gassmann, 2011; Downes and Mahon, 2012). Therefore it seems most
appropriate to focus attention on considering appropriate and feasible strategies that have
been adopted to delay development of resistance by pests for other Bt crops (Tabashnik et al.,
2008; Storer et al., 2010; Gray 2011). It should be noted that evolution of resistance to the
expressed Bt proteins does not itself introduce a new environmental harm; instead it would
lead to loss of an environmental benefit conferred by cultivation of the Bt crop, if production
of the Bt crop allowed for reduced pesticide use. In this case, very few farmers in Uganda
apply pesticides for sweetpotato production, and when applied, it is to manage foliar pests, not
weevils that attack the roots (Abidin et al., 2005; Zawedde, Chapter 2). Thus cultivation of Bt
sweetpotato is unlikely to significantly reduce pesticide use.
Although not an environmental harm per se, minimizing development of resistance is
valued to prolong the value of the Bt trait to reduce production and quality losses caused by
the sweetpotato weevil. To delay evolution of resistance by pests to GE crops producing Bt
toxins, one of the approaches that is most widely applied is the ‘high dose/refuge” strategy
180

(Tabashnik et al., 2008). Some national regulatory bodies, for example the US EPA, require
this. The refuge promotes survival of susceptible pests, which mate with resistant insects and
slow the evolution of resistance (Tabashnik et al., 2008). However there is need for a better
understanding about the genetics of resistance in the weevils (dominant vs. Recessive, allele
frequencies) and the mating behaviour of the weevil. Implementing the refuge requirement in
Uganda will also be problematic because of the small and fragmented plots used for
sweetpotato cultivation. Similar challenges were reported when applying the refuge
requirement in China (Zhao et al., 2010). However, natural refuges derived from mixed
cropping that is practiced by many small-scale farmers have been reported to delay
development of resistance by the pests to Bt crops in China (Wu and Guo, 2005; Huang et al.,
2010). Similarly, in many regions of Uganda people settle in small communities and most of
them produce comparable kinds of crops (Zawedde, Chapter 2), therefore a mixture of crops
and presence of non-Bt sweetpotato in close proximity is likely to occur. Establishing the
alternative hosts of sweetpotato weevils so that they are included in the mixed cropping
systems will also help to delay development of resistance by the weevil. However, if GE
weevil resistant sweetpotato is cultivated as a monocrop on a large scale by commercial
farmers or by a community of small-scale farmers, then a refuge may be warranted (EFSA,
2012; US EPA, 2002). Prior to release of the Bt sweetpotato, the technology developer must
commit to facilitate the responsible regulatory agency to conduct occasional post-release
monitoring to determine the level of adoption and monitor for refuge requirement compliance.
An alternative method to delay development of resistance by the pest to Bt toxins may
be gene pyramiding (introducing more than one type of Bt protein) (Oswald et al., 2012).
Stacking two or more Bt toxins is likely to be more durable with the assumption that the
additional toxin(s) will kill insects that have evolved resistance to the first toxin, a
phenomenon known as “redundant killing” (Thierry et al., 2013). However, the pyramid
181

strategy may not be effective for pests that have gained resistance to one or more of the toxins
in the pyramid (Mahon et al., 2012; Thiery et al., 2013). Three Bt proteins (Cry3Ca1,
Cry7Aa1, and ET33) with high toxicity levels to the African sweetpotato weevil species
(Cylas puncticollis and C. brunneus) (Ekobu et al., 2010) are being used for gene pyramiding
in various paired combinations (CIP, 2009). This measure is likely to contribute to delaying
development of resistance because the weevils have not yet been exposed to any of these
toxins
Efficacy of gene pyramiding can be prolonged by incorporating an integrated pest
management (IPM) system that involves additional control measures such as crop rotation,
and insecticide use with insect scouting (Gray et al., 2009). Intercropping, which reduces pest
incidence by increasing diversity of an ecosystem (Rao et al., 2012), has also been suggested
as a strategy that can be implemented by small-scale farmers especially for crop pests with
limited mobility at the harmful stage (Bates et al., 2005; Christou et al., 2006). These
recommended IPM measures including crop rotation and intercropping are currently practiced
by some sweetpotato farmers in Uganda (Zawedde, Chapter 2).

Loss of sweetpotato varietal diversity
Our consultation with scientific experts and practitioners identified displacement of
the landraces and conventional varieties as an important possible risk in Uganda (Zawedde,
Chapter 3). We proceeded to carry out a problem definition for this potential problem. The
measurable assessment endpoints for determining impact on varietal diversity include a
reduced number of landrace and conventional varieties within communities, and a reduced
level of genetic variability within and among communities (US EPA, 2003). Table 5.1 shows
exposure scenarios and risk hypotheses for assessing the possibility of loss of landrace
diversity due to continuous displacement by weevil resistant sweetpotato. In the absence of
182

Table 5-1 Exposure scenarios, risk hypotheses and action plan for assessing the harm through reduced abundance or displacement of landrace
varieties resulting from increased planting of GE weevil resistant sweetpotato.
Valuable entity: Sweetpotato varietal diversity
Assessment endpoint: Reduced abundance or displacement of landrace and conventional varieties
Potential harm: Significant decrease in the valuable genetic diversity of the crop
Conceptual model:
Exposure scenarios
GE weevil resistant
sweetpotato will be widely
cultivated.

Risk hypotheses

Analysis plan

1. GE weevil resistant
sweetpotato will not be
widely cultivated.

Use existing literature or if necessary, perform studies to examine the
likelihood for adopting GE weevil infestation sweetpotato in the receiving
environment. If the likelihood for widespread adoption is high then reject risk
hypothesis and proceed to the next exposure scenario.

Farmers will replace other
varieties with GE weevil
resistant sweetpotato.

2. Adoption of GE weevil
resistant sweetpotato
does not result in
discarding landrace
varieties.

Use literature, or where necessary, perform studies to determine whether
adoption of GE weevil resistant sweetpotato or any valued new modern
cultivar is likely to result in reduced numbers of landrace varieties in the
country. If risk hypothesis rejected, proceed to next exposure scenario.

Replacement of other varieties
results in significant/important
genetic diversity loss.

3. No significant genetic
diversity loss will be
observed.

Harm

Review literature or conduct experiments to characterize the existing varietal
diversity. If diverse, then estimate the decrease in genetic variability that
results from various scenarios of discarding landrace varieties. If risk
hypothesis rejected, then harm identified.
Consider risk mitigation or management options.

183

cultivation of GE sweetpotato in any part of the world, predicting the rate of adoption and the
impacts of weevil resistant sweetpotato in Uganda is difficult. An alternative approach is to
use data generated from other modern cultivars to estimate the likely impact (Zawedde,
Chapter 2). I recognize the limitation of using this approach because none of the modern
cultivars have high resistance to weevil infestation; however, most of these cultivars have
high resistance to virus diseases, which are an equally important constraint to sweetpotato
production in many parts of the country.

Risk hypothesis 1: Weevil resistant sweetpotato will not be widely cultivated in Uganda
The first risk hypothesis is that weevil resistant sweetpotato will not be widely
cultivated in Uganda. This risk hypothesis cannot be corroborated because weevil infestation
is a major constraint to sweetpotato production with very high yield losses recorded (Mwanga
et al., 2011; Muyinza et al., 2012; Zawedde, Chapter 2); therefore it is a trait likely to be
widely valued by farmers. However, Zawedde (Chapter 2) reported that adoption of a variety
is affected by many other factors besides its resistance to environmental stresses, therefore an
ex-ante study for the likely adoption rate for the GE weevil resistant sweetpotato that is
currently under development is necessary.

Risk hypothesis 2: Adoption of weevil resistant sweetpotato does not result in discarding
landrace varieties
The second risk hypothesis is that cultivation of weevil resistant sweetpotato does not
result in discarding landrace varieties. Decisions made by the farmers regarding whether to
maintain some of the existing varieties when they adopt a modern cultivar will influence the
level of landrace diversity. To make a better prediction of the impact of introducing weevil
resistant sweetpotato on varietal diversity, we carried out a survey to establish how adoption
184

of modern cultivars has influenced maintenance of varietal diversity in farmers’ fields in
Uganda. We explored how farmers make decisions under two scenarios: 1) New cultivars
introduced without awareness campaigns and incentives packages; and 2) new cultivars
introduced with awareness campaigns and incentives packages.
In the first scenario where modern cultivars are released without awareness campaigns
and incentives packages, we explore the usual procedure used to introduce newly-bred
cultivars released by the national sweetpotato program. Under this scenario, the National
Agricultural Research Organization generates improved cultivars. Promotion and
dissemination of the new cultivars to farmers is the responsibility of a different government
agency, the National Agricultural Advisory Services (NAADS). However, not all crops are
promoted by NAADS; in a recent farmers’ survey, the majority of respondents (63%)
indicated that they have rare interactions with extension service (Zawedde, Chapter 2). Most
of the farmers (22% out of 37%) who interacted more frequently with extension services
indicated their discussions focused on other crops and livestock. Currently, the goals for
NAADS focus on promotion production of bananas, groundnuts, and rice, followed by vanilla
and maize (Benin et al., 2011).
Owing to inadequate extension services, modern sweetpotato cultivars are mostly
grown in districts close to the research institutes where farmers have access to the
demonstration plots established for on-station and on-farm evaluation of the cultivars
(Zawedde, Chapter 2). The majority of the respondents (73%) still rely on landrace diversity
to reduce risk of yield losses due to adverse environmental conditions. Under this scenario, it
appears that adoption of weevil resistant sweetpotato is less likely to affect the diversity of
landraces.
In the second scenario where modern cultivars are released in conjunction with
awareness campaigns and incentives package, we explore the impact of introducing orange185

fleshed sweetpotato (OFSP) by HarvestPlus. HarvestPlus introduced an OFSP project in
Uganda in 2007. The purpose of this project is to promote production, consumption and
marketing of OFSP containing elevated beta carotene so as to alleviate vitamin A deficiency
in Uganda (Isubikalu et al., 2009). Promotion activities were implemented in collaboration
with two local NGOs: FADEP and VEDCO, and the pilot target districts were Bukedea,
Kamuli and Mukono. Four OFSP cultivars: NASPOT 9, NASPOT 10, Kakamega and
Ejumula were promoted by HarvestPlus. The NGOs interacted frequently with the farmers in
these districts, and they conducted good agricultural practices training and also provided
access to quality plant materials and market opportunities.
For our study, we selected Bukedea because the adoption rate of the OFSP varieties in
this district reached 100% in 2008 (Wamaniala, 2008). A total of 35 farmers who were
growing sweetpotato at the time of the study (December 2012) were randomly selected. We
conducted face to face individual interviews and focus group discussions to explore the
impacts of introducing new cultivars followed by awareness campaigns and incentives
package on diversity of other varieties in the receiving environment.
Under this project, farmers were encouraged to carry out practices like irrigation
during the dry season; soil fertility management including fertilizer application, especially for
farmers engaged in planting material multiplication; and pesticide application combined with
insect scouting. The majority of the respondents (71%) indicated that they adopted these
practices and implemented them diligently because they appreciated the yield increases and
the net gain obtained due to easy access to market. During 2008-2009, 76% of the respondents
dropped all the other sweetpotato varieties and only grew OFSP cultivars.
However, in the second growing season of 2009, FADEP stopped their activities in the
district. This resulted in many farmers incurring high yield losses due to lack of market access
at the end of the season. One farmer from Kachumbala sub-county reported that he had hired
186

three hectares for sweetpotato production that season and he lost approximately 88% of his
yield because he was only able to sell to the local markets that were already saturated due to
high supply. Owing to this, he had to sell some of his personal belongings including his
livestock to cover the cost. He indicated that he will never adopt OFSP cultivars again no
matter how good the market prices because he is now risk averse. Similar experiences were
echoed by approximately 30% of the respondents.
Stopping the activities of FADEP resulted in 86% of the respondents dropping the
OFSP cultivars except Kakamega. They indicated that they dropped the OFSP cultivars
because they were less adapted to the environment, and their higher productivity was inputdependent. They maintained Kakamega, although they preferred the taste of Ejumula,
because it had relatively better environmental adaptability characteristics, and they
appreciated the benefits of consuming increased beta carotene especially for women and
children. The OFSP cultivars were replaced by landrace varieties with higher environmental
adaptability characteristics and acceptable yields. However, only a limited number (6) of
landrace varieties were regrown by the respondents.
At the time of this study, 48% of respondents were growing the OFSP cultivars. Some
of these farmers readopted the OFSP cultivars in 2010/11 after establishment of a farmers’
group known as Kakibu Association. This farmers’ group was established in 2010 by some of
the farmers who continued OFSP production after FADEP. The farmers’ group sought
guidance from HarvestPlus, NARO and other farmers’ groups involved in sweetpotato
production in the surrounding area. Access to information regarding better crop management
options and market opportunities has resulted in recovery of OFSP production in Bukedea.
The Chairman of this farmers’ group reported that he has been hiring six hectares for OFSP
cultivation for vine and storage root production since 2010, and he has gained tremendous
social and economic progress due to OFSP production. He indicated that due to unpredictable
187

weather conditions and taste preference he still grows at least a quarter of a hectare of his
favorite landrace varieties for home consumption. Most of the other farmers (62%) growing
OFSP also indicated that they maintain 2-4 landrace varieties on a small portion of land for
home consumption. Other farmers were not able to maintain landrace varieties because of
land limitations.
Owing to the activities of the farmers’ group, an additional 25% of the respondents
indicated that they were planning to readopt some of the OFSP for commercial production.
All these respondents were planning to maintain some landrace varieties for crop insurance.
Overall, introducing new cultivars followed by awareness campaigns and incentives package
appeared to reduce the diversity of other varieties in the receiving environment. Therefore
introducing GE weevil resistant sweetpotato under this scenario is likely to result in
discarding some of the landrace varieties.

Risk hypothesis 3: Displacement of landrace varieties does not result in significant genetic
diversity loss
Displacement of varieties is a common phenomenon that occurs in agricultural
systems but it does not always result in loss of genetic diversity (Louette and Smale 2000; van
Heerwaarden et al., 2009). Estimating the effect of modern cultivars on genetic diversity is far
from obvious because disappearance of named landrace varieties is not sufficient proof for
genetic diversity loss. Adoption of modern cultivars can actually increase diversity if the
cultivars are genetically more heterogeneous or they possess traits that are not present in
landraces (Wood and Lenne 1997; Louette and Smale 2000). The best way to address this
hypothesis is by defining and measuring the level of genetic diversity in the field (van
Heerwaarden et al., 2009). This requires quantifying the number of distinct genotypes

188

available, the evenness of distribution of these genotypes, and the extent of the difference
between genotypes (FAO/IPGRI, 2002).
Assessment of the level of genetic diversity in Ugandan sweetpotato germplasm using
microsatellite markers indicated that there were large numbers of distinct genotypes in the
country (Zawedde, Chapter 1). The majority of genetic variability was among sweetpotato
varieties within a region (97%, on average) rather than among accessions from different
regions of the country. This suggests that the genetic diversity is evenly distributed between
various parts of the country. Similarly, very low genetic variability (6%) was observed
between Uganda’s germplasm and varieties from other East African countries (Kenya and
Tanzania), which suggests that the genetic diversity is also evenly distributed across the
region. If adoption of weevil resistant sweetpotato is widespread, then loss of diversity is
likely, as would occur with widespread adoption of any valued modern variety. Losses of
varietal diversity can be mitigated by implementing a combination of appropriate and
effective ex situ and in situ conservation of the varietal diversity. Findings by Zawedde
(Chapter 1 and 2) showed that currently both methods have major limitations. Further studies
are necessary to explore the different ex situ and in situ conservation options so as to
determine the most appropriate and effective measures for conserving sweetpotato and other
vegetatively propagated crops in Uganda.

Changes in crop management systems
The concern about changes in crop management systems stems from the perceived
requirement for a refuge, and a possible increase in mono-cropping rather than intercropping,
either of which could result in reduced abundance of other crop species (Zawedde, Chapter 3).
The refuge requirement has been discussed above. Because the majority of sweetpotato in
Uganda is typically grown as a mono-crop rather than inter-cropped (Bashasha et al., 1995,
189

Abidin et al., 2005; Zawedde, Chapter 2), the use of weevil resistant sweetpotato is unlikely
to significantly affect the level of inter-cropping practiced by farmers.
Another concern is that cultivation of weevil resistant sweetpotato will result in
reduction in pesticide application that, in turn, will lead to emergence or increase of secondary
pests. There is evidence that cultivation of Bt crops in China has resulted in increased
abundance of other pests due to reduced application of broad spectrum pesticides (Naranjo et
al., 2008; Wang et al., 2010; Lu et al., 2012). Exposure scenarios, risk hypotheses and an
analysis plan to assess the likely impact of changes in pesticide application for weevil
resistant sweetpotato are provided in Table 5.2.

Risk hypothesis 1: Cultivation of weevil resistant sweetpotato does not cause changes in
rates of pesticide applications
As indicated above, very few farmers in Uganda apply pesticides for sweetpotato
production, and when applied it is to manage foliar pests (Abidin et al., 2005; Zawedde,
Chapter 2). Cultivation of GE sweetpotato targeting weevils, which are mainly storage root
pests, is less likely to reduce pesticide application. The impact of Bt sweetpotato on
abundance and diversity of secondary pests may be improved by establishing whether there
are other Coleopteran pests of sweetpotato.

190

Table 5-2 Exposure scenarios, risk hypotheses and actions plan for assessing the harm caused by changes in rates of pesticide application in GE
weevil resistant sweetpotato plots.
Valuable entities: Sweetpotato production and performance of other host plant species my understanding was that the valuable entities here
are the crops (and perhaps other biota) that will be harmed if pests erupt because of the cessation of pesticide application . . .
Assessment endpoints: Reduced yields/quality parameters for sweetpotato and or other valued plant species
Potential harm: Reduced pesticide application causes emergence or increase in abundance of secondary pests
Exposure scenarios

Risk hypotheses

Analysis plan

Cultivation of GE weevil
resistant sweetpotato causes
reduction of pesticide
applications.

1. Cultivation of GE
weevil resistant
sweetpotato does not
cause changes in rates of
pesticide applications.

Use literature or where necessary conduct studies to determine whether
cultivation of GE weevil resistant sweetpotato is likely to result in significant
changes in pesticide application compared to existing modern cultivars. If risk
hypothesis rejected, proceed to next exposure scenario.

Reduced pesticide applications
cause emergence of secondary
pests in the sweetpotato plots.

2. Changes in pesticide use
do not cause increase in
abundance of secondary
pests and their damage
to the host plant species.

Use literature or where necessary conduct experiments to develop an inventory
of secondary pests; determine whether their damage is likely to become more
severe; establish whether these pests are currently controlled by pesticide
application; and if there are no effective alternative methods for reducing their
damage. If risk hypothesis rejected, then harm identified.

Harm
Consider risk mitigation or management options.

191

Conclusion
All new GE crops are required to be subjected to an environmental risk assessment
(ERA) before open release. The recommended framework for conducting an ERA (Wolt et
al., 2010; Gray, 2012) was employed by starting with problem formulation. The efficiency of
the problem formulation process was enhanced by integrating expertise and experience of
scientists and practitioners involved in the regulatory process to identify the most important
concerns associated with the cultivation of GE weevil resistant sweetpotato in Uganda.
Utilizing a combination of existing literature and diverse expertise of scientists and
regulators, and by conducting studies of germplasm diversity and farmer practices we were
able to identify the possible risks requiring the greatest consideration in the risk assessment
process, determine aspects where additional information is needed; and suggest risk
management strategies to be considered prior to open cultivation of weevil resistant
sweetpotato varieties. The primary identified concerns were: development of resistance to Bt
toxin by the sweetpotato weevils; loss of varietal diversity; loss of crop diversity; and
emergence of secondary pests. Based on analysis of these concerns the following
recommendations may be considered.
Development of resistance by the weevils to Bt toxin is considered likely to occur
over time if integrated resistance management strategies are not adopted. Strategies to delay
development of resistance can be employed by combining the pyramided gene constructs
under development with currently practiced integrated pest management measures such as
crop rotation and timely harvesting (for commercial farmers), and mixed cropping and/or
intercropping (for subsistence farmers). A refuge requirement may be warranted in case of
large-scale cultivation of weevil resistant sweetpotato in a given location, however efficiency
of this method requires a better understanding of the genetics and mobility of the sweetpotato
weevil. Post-release monitoring of the adoption rate and crop management systems can help
192

to ensure that there is refuge resulting from diverse cropping systems or planting of non-Bt
sweetpotato, and that the recommended integrated resistance management measures are
implemented by farmers.
Loss of landrace and varietal diversity is also a likely concern that will occur with any
highly valued new variety, whether or not it is genetically engineered, emphasizing the
importance of implementing effective ex situ and in situ conservation of the varietal diversity.
We observed that the ex situ conservation is currently facing many challenges causing loss of
some of the collected germplasm (Zawedde, Chapter 1). Further research is necessary to
identify the most effective approaches for ex situ conservation of vegetatively propagated
species. Finally, the analysis indicates that loss of crop diversity and emergence of secondary
pests are not likely concerns for cultivation of weevil resistant sweetpotato.
In summary, prior to open cultivation of weevil resistant sweetpotato, there is need for
clear definition of the risk management strategies to mitigate development of resistance by
the pests and loss of sweetpotato varietal diversity. While this study was conducted for
cultivation of weevil resistant sweetpotato in Uganda, the information and analysis can be
utilized when considering cultivation of this crop in other developing countries where there is
high varietal diversity and sweetpotato production is mainly done by small-scale farmers.

193

LITERATURE CITED

194

LITERATURE CITED

Abidin, P.E., van Eeuwijk, F.A., Stam, P., Sturik, P.C., Malosetti, M., Mwanga, R.O.M.,
Odongo, B., Hermann, M., Carey, E.E., 2005. Adaptability and stability of sweetpotato
varieties for low-input systems in Uganda. Plant Breed. 124:491–497
Baffes J., 2009. The Cotton Sector Of Uganda. Africa Region Working Paper Series No.
123. Paper prepared for the World Bank Group. pp.30
Bashaasha B, Mwanga ROM, Ocitti p’Obwaya C, Ewell PT, 1995. Sweetpotato in the
farming and food system in Uganda: A farm survey report. CIP, sub –Saharan African region,
Nairobi, Kenya /National Agricultural Research organization, Kampala, Uganda. pp.63.
Bates, S. L., J. Z. Zhao, R. T. Roush, and A. M. Shelton. 2005. Insect resistance management
in GM crops: past, present and future. Nat. Biotechnol.23:57-62
Benin S., E. Nkonya, G. Okecho, J. Randriamamonjy, E. Kato, G. Lubadde, M. Kyotalimye,
and F. Byekwaso 2011. Impact of Uganda’s National Agricultural Advisory Services
Program. International Food Policy Research Institute (IFPRI) research monograph
Chassy B, Carter C, McGloughlin M, McHughen A, Parrott W, Preston C et al. 2003. UK
field-scale evaluations answer wrong questions. Nat Biotechnol 21:1429–1430
Christou P, Capell T, Kohli A, Gatehouse J, Gatehouse A, 2006. Recent developments and
future prospects in insect pest control in transgenic crops. TRENDS Plant Sci. 11:302-308
CIP 2009. Confined Field Trial application submitted to Uganda NBC. National Agricultural
Research Organization in partnership with International Potato Center.
Dana G. V.; Kapuscinski A. R.; Donaldson J. S., 2012. Integrating diverse scientific and
practitioner knowledge in ecological risk analysis: A case study of biodiversity risk
assessment in South Africa. J. Environ. Manage. 98:134–146
Downes S, Mahon R., 2012. Evolution, ecology and management of resistance in
Helicoverpa spp. to Bt cotton in Australia. J Invertebr Pathol 110:281-286.
Ekobu, M., Solera, M., Kyamanywa, S., Mwanga, R.O.M., Odongo, B., Ghislain, M., Moar,
W.J., 2010. Toxicity of seven Bacillus thuringiensis Cry proteins against Cylas puncticollis
and Cylas brunneus (Coleoptera: Brentidae) using a novel artificial diet. Journal of Economic
Entomology 103:1493-1502.
ESFA, 2012. Scientific opinion on the annual post-market environmental monitoring
(PMEM) report from Monsanto Europe S.A. on the cultivation of genetically modified maize
Mon810 in 2009. European Food Safety
Gould, F. 2000. Testing Bt refuge strategies in the field. Nat. Biotechnol. 18:266–276.

195

Gray A., 2012. Problem Formulation in Environmental Risk Assessment for Genetically
Modified Crops: A Practitioner’s Approach International Centre for Genetic Engineering and
Biotechnology (ICGEB) Collection of Biosafety Reviews 6:10-65
Gray, M. 2011. Severe Root Damage to Bt Corn Observed in Northwestern Illinois. Available
at http://bulletin.ipm.illinois.edu/article.php?id=1555
Gray, M.E., T.W. Sappington, N.J. Miller, J. Moeser, and M.O. Bohn., 2009. Adaptation and
Invasiveness of Western Corn Rootworm: Intensifying Research on a Worsening Pest.
Annual Review of Entomology 54 (1): p 303-21.
Hodgson, E., and A. Gassmann, 2011. "First Iowa Confirmation of Resistance to Bt Corn by
Western Corn Rootworm." Corn & Soybean Digest
Hokanson KE, Ellstrand NC, Ouedraogo JT, Olweny PA, Schaal BA and Raybould AF 2010.
Biofortified sorghum in Africa: using problem formulation to inform risk assessment. Nature
Biotechnology 28(9): 900-903.
Horna D., M. Kyotalimye, J. Falck-Zepeda, 2009. Cotton Production in Uganda: Would GM
technologies be the Solution?Paper prepared for presentation at the International Association
of Agricultural Economists Conference, Beijing, China, August 16-22, 2009.pp. 28
Huang, J., Mi, J., Lin, H., Wang, Z., Chen, R., Mu, R., et al. (2010). A decade of Bt cotton in
Chinese fields: Assessing the direct effects and indirect externalities of Bt cotton adoption in
China. Science China Life Sciences, 53(8), 981-999.
Ismael, Y., Bennett, R., & Morse, S. (2002). Benefits from Bt cotton use by smallholder
farmers in South Africa. AgBioForum, 5(1), 1-5.
Isubikalu P., S. Magezi, C. Musoke, H. Nsubuga, S. Namanda, A-M. Ball, 2009. Promoting
Production, Consumption, and Marketing of the Orange Sweetpotato Trainer’s guide.
HarvestPlus
James C, 2012. Global Status of Commercialized Biotech/GM Crops: 2010. ISAAA Brief 44.
Available at http://www.isaaa.org/resources/publications/briefs/44/executivesummary/
Louette D, Smale M, 2000. Farmers’ seed selection practices and traditional maize varieties
in Cuzalapa, Mexico. Euphytica 113:25–41
Lu B.L 2008. Transgene escape from GM crops and potential biosafety consequences: An
environmental perspective. Collection of Biosafety Reviews 4:66-141.
Lu Y, K. Wu, Y. Jiang, Y. Guo and N. Desneux 2012. Widespread adoption of Bt cotton and
insecticide decrease promotes biocontrol services Nature doi:10.1038/nature11153
Lu Y., Wu K., Jiang Y., Xia B., Li P., Feng H., Wyckhuys K.A.G., Guo Y., 2010. Mirid Bug
Outbreaks in Multiple Crops Correlated with Wide-Scale Adoption of Bt Cotton in China.
Science 328 (5982): 1151–1154.
Mahon R, Olsen K, Downes S, Addison S, 2007. Frequency of alleles conferring resistance to
the Bt toxins Cry1Ac and Cry2Ab in Australian populations of Helicoverpa
armigera (Lepidoptera: Noctuidae). J Econ Entomol 100: 1844–1853
196

Muyinza H., H.L. Talwana, R.O.M. Mwanga and P.C. Stevenson, 2012. Sweetpotato
weevil (Cylas spp.) resistance in African sweetpotato germplasm. International Journal of
Pest Management 58(1):78-81
Mwanga R.O.M., M. Ghislain, J. Kreuze, G. N. Ssemakula and C. Yencho, 2011. Exploiting
the use of biotechnology in sweetpotato for improved nutrition and food security: Progress
and future outlook. In: Proceedings of the International Conference on Agro-Biotechnology,
Biosafety and Seed Systems in Developing Countries, pp. 25-31.
Naranjo S. E. 2009 Impacts of Bt crops on non-target organisms and insecticide use
patterns. Perspect. Agr. Vet. Sci. Nutr. Nat. Resour. 4:1–23.
Nickson T.E. 2008. Planning Environmental Risk Assessment for Genetically Modified
Crops: Problem Formulation for Stress-Tolerant Crops. Plant Physiology 147: 494–502
Oswald KJ, French BW, Nielsen C, Bagley M (2012) Assessment of fitness costs in
Cry3Bb1-resistant and susceptible western corn rootworm (Coleoptera: Chrysomelidae)
laboratory colonies. J Appl Entomol 136(10): 730-740
Owen, M.J.; Michael, P.J.; Renton, M.; Steadman, K.J.; Powles, S.B. 2011. Towards largescale prediction of Lolium rigidum (annual ryegrass) emergence. II. Correlation between
dormancy and herbicide resistance levels suggests an impact of cropping systems. Weed Res.
2011, 51, 133–141.
Rao MS, Rama Rao CA, Srinivas K, Pratibha G, Vidya Sekhar SM, Sree Vani G,
Venkatswarlu B. 2012. Intercropping for management of insect pests of castor, Ricinus
communis, in the semi–arid tropics of India. Journal of Insect Science 12:14available online:
insectscience.org/12.14
Raybould A, Vlachos D. 2010. Non-target organism effects tests on Vip3A and their
application to the ecological risk assessment for cultivation of MIR162 maize. Transgenic
Res.
Raybould A.F, L.S. Higgins, M. J. Horak, R.J. Layton, N.P. Storer, J. M. DeLaFuente and
R.A. Herman 2012. Assessing the ecological risks from the persistence and spread of feral
populations of insect-resistant transgenic maize. Transgenic Research 21:655–664
Raybould AF 2011. The bucket and the searchlight: Formulating and testing risk hypotheses
about the weediness and invasive potential of transgenic crops. Environmental Biosafety
Research 9:123-133.
Romeis J, Bartsch D, Bigler F, Candolfi MP, Gielkens MMC, Hartley SE, Hellmich RL,
Huesing JE, Jepson PC, Layton R, Quemada H, Raybould AF, Rose RI, Schiemann J, Sears
MK, Shelton AM, Sweet J, Vaituzis Z & Wolt JD 2008. Assessment of risk of insect-resistant
transgenic crops to nontarget arthropods. Nature Biotechnology 26:203-208.
Storer NP, Babcock JM, Schlenz M et al. 2010. Discovery and characterization of field
resistance to Bt maize: Spodoptera frugiperda (Lepidoptera: Noctuidae) in Puerto Rico. J
Econ Entomol 103:1031–1038

197

Tabashnik B, Gassman A, Crowder D, Carriere Y, 2008. Insect resistance to Bt crops:
evidence versus theory. Nat. Biotechnol. 26
Thierry B., S. Heuberger, M. Zhang, C. Ellers-Kirk, X. Ni, L. Masson, X. Li, B.E. Tabashnik,
and Y. Carrière 2013. Potential shortfall of pyramided transgenic cotton fori insect resistance
management. PNAS 110 (15): 5806–5811
US EPA's Regulation of Bacillus thuringiensis (Bt) Crops, 735-F-02-013
May 2002, http://www.epa.gov/pesticides/biopesticides/pips/regofbtcrops.htm
Wamaniala, M. 2008. Orange fleshed sweetpotato; reaching end users project in Uganda.
HarvestPlus Program Project No. 6010, 2008 Semi-Annual Report.
Wang YM, Zhang GA, Du JP, Liu B, Wang MC. 2010. Influence of transgenic hybrid rice
expressing a fused gene derived from cry1Ab and cry1Ac on primary insect pests and rice
yield. Crop Prot. 29:128–33
Wolt JD, Keese P, Raybould AF, Fitzpatrick JW, Burachik M, Gray AJ,Olin SS, Schiemann
J, Sears M & Wu F 2010. Problem formulation in the environmental risk assessment for
genetically modified plants. Transgenic Research 19: 425-436.
Wood D, Lenne JM, 1997. The conservation of agrobiodiversity onfarm: questioning the
emerging paradigm. Biodivers Conserv. 6:109–129
Wu KM, and Guo YY, 2005. The evolution of cotton pest management practices in China.
Annual Review of Entomology, 50:31–52
Yada B., P. Tukamuhabwa, Villordon A., A. Alajo, and R.O.M. Mwanga 2010a. An online
database of sweetpotato germplasm collection in Uganda. Hortscience 45(1):153–153. 2010.
Zhang D.P., Golmirzaie A., Cipriani G., Panta A., Ghislain M., Smit N., Rety I., and Michaud
D. 1996. Developing weevil resistance in sweetpotato with genetic transformation.
International Potato Center (CIP) Program Report pp. 95-96.
Zhao JH, Ho P, Azadi H, 2011. Benefits of Bt cotton counterbalanced by secondary pests?
Perceptions of ecological change in China. Environ. Monit. Assess. 173:985-994.
Zhao JZ, Cao J, Li YX, Collins HL, Roush RT, Earle ED, Shelton AM, 2003. Transgenic
plants expressing two Bacillus thuringiensis toxins delay insect resistance evolution. Nat
Biotechnol 21:1493–1497.

198