IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF THE DEGREE OF MASTER OF SCIENCE IN CROP SCIENCE OF MAKERERE UNIVERSITY MAY 2016 i… [601795]
V
ARIETY SCREENING FOR
STRIGA
HERMONTHICA
RESISTANCE IN UPLAND
RICE IN EASTERN UGANDA
KAYONGO
NICHOLAS
BSc.
(
Horticulture
) Hons. MAK
2012/HD02/105
U
A RESEARCH THESIS
SUBMITTED TO THE
DIRECTORATE OF RESEARCH AND
GRADUATE TRAINING
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE AWARD
OF THE DEGREE
OF MASTER OF SCIENCE IN CROP
SCIENCE
OF MAKERERE UNIVERSITY
MAY
2016
i
DECLARATION
I
,
Nicholas
Kayongo,
do hereby declare that the
work presented in this thesis
is original and has
not been submitted to any institution of learning by any one for
an
Academic award.
Signed………
…………………………………Date……………………………………
KAYONGO NICHOLAS
This thesis
has been submitted with
our approval as University
supervisor
s
:
Signed………………………………………………Date …………………………………….
DR. JENIPHER BISIKWA
Department of Agricultural production
Makerere University, Kampala
Signed……………………………………………… Date………………………………………
DR. JAMES
M. SSEBULIBA
Department of Agricultural production
Makerere University, Kampala
ii
DEDICATION
This thesis is dedicated to
the
people of Namutumba district.
I hope the research findings will help
you improve your livelihood
s
through rice production. Also
,
to my p
arents a
nd friends. Thank you
all for the
moral support and prayers.
iii
ACKNOWLEDGEMENTS
In the first place, I than
k the A
lmighty God for all the wisdom, strength and courage he provided
me during the course of the experiments and writing of the thesis.
I would
like to give my heartfelt
appreciation
to my
Academic superv
isors, Dr. Jenipher Bisikwa,
Dr. James
M. Ssebuliba,
School of Agricultural Sciences,
Makerere University
, Kampala
and
Professor
Julie Scholes,
Department
of Plant and Animal Sciences,
University of Sheffield,
U.K
for all the technical advice as well as the wonderful supervision. I am
very grateful that you
continued supervising me despite all your other commitments.
Also, special thanks go
to Dr. Jonne Rodenburg
and Dr. Mamadou Cissoko, Africa Rice center,
Tanzania
,
for providing the Rice germplasms, helping with trial setup and
for all the guidance
during w
riting
of this thesis.
Thank you so much for
you
r help, I would not have made it without
your help.
I would also like to
thank the
members of C45
lab
at the Department of Plant and
Animal sciences
,
University of Sheffield U.K, for all the assistance
they
rendered
during my
Laboratory experiments
. I would also like to
thank
all
the
members of Abendewoza farmers group
for
providing me with land and participating in the field experiments in Nsinze sub
–
county,
Namutumba district.
Last but not least,
I
am also
very grateful to the Seohyun Foundation that provided me
with
the
scholarship for the two academic years
. Special thanks also go
to the Department
of International
Development (DF
ID) and
to the
Biotechnology and Biological Sciences Research Co
uncil
(BBSRC)
for funding
this research.
I am very grateful for all the financial assistance rendered to
me during the course of this research.
iv
LIST OF ACRONYMS
FAOSTAT
Food and Agricultural Organization statistics division
MAAIF
Ministry of Agriculture, A
nimal industry and Fisheries
NAADS
National
Agricultural Advisory Services
NERICA
New Rice for Africa
PMA
Plan for modernization of agriculture
WARDA West Africa Rice Development Association
UBOS Uganda Bureau of Statistics
PVS
Participatory Variety Selection
AATF
African Agricultural Technology Foundation
FAO Food and Agricultural Organization
NPA National Planning Authority
v
TABLE OF CONTENTS
DECLARATION
…………………………..
…………………………..
…………………………..
……………………….
i
DEDICATION
…………………………..
…………………………..
…………………………..
………………………….
ii
ACKNOWLEDGEMENTS
…………………………..
…………………………..
…………………………..
……..
iii
LIST OF AC
RONYMS
…………………………..
…………………………..
…………………………..
……………
iv
TABLE OF CONTENTS
…………………………..
…………………………..
…………………………..
………….
v
LIST OF TABLES
…………………………..
…………………………..
…………………………..
…………………..
xi
LIST OF FIGURES
…………………………..
…………………………..
…………………………..
……………….
xiii
LIST OF APPENDICES
…………………………..
…………………………..
…………………………..
………..
xiv
ABSTRACT
…………………………..
…………………………..
…………………………..
…………………………..
xvi
CHAPTER ONE
…………………………..
…………………………..
…………………………..
………………………
1
1.0 INTRODUCTION
…………………………..
…………………………..
…………………………..
……………….
1
1.1 Background
…………………………..
…………………………..
…………………………..
………………………
1
1.2 Overview of rice production in the world
…………………………..
…………………………..
………….
2
1.3 Rice production in Uganda and importance
…………………………..
…………………………..
……….
2
1.4 Problem statement
…………………………..
…………………………..
…………………………..
……………..
4
1.5 Justification
…………………………..
…………………………..
…………………………..
………………………
5
1.6 Objectives of the
study
…………………………..
…………………………..
…………………………..
……….
6
1.7 Hypotheses
…………………………..
…………………………..
…………………………..
……………………….
7
CHAPTER TWO
…………………………..
…………………………..
…………………………..
……………………..
8
vi
2.0 LITERATURE REVIEW
…………………………..
…………………………..
…………………………..
……
8
2.1 Major weeds of rice
…………………………..
…………………………..
…………………………..
……………
8
2.1.1
Striga
…………………………..
…………………………..
…………………………..
…………………………..
..
8
2.1.1.1 Origin and distribution of
Striga
…………………………..
…………………………..
…………………
9
2.1.1.2 Taxonomy and Botany of
Striga
…………………………..
…………………………..
……………….
10
2.1.1.3 Life Cycle of
Striga
…………………………..
…………………………..
…………………………..
…….
11
2.1.1.4 Ecology of
Striga
…………………………..
…………………………..
…………………………..
……….
12
2.1.1.5 Nature of
Striga
damage
…………………………..
…………………………..
………………………….
13
2.1.1.6 Control of
Striga
…………………………..
…………………………..
…………………………..
…………
14
2.1.1.6.1 Cultural control
…………………………..
…………………………..
…………………………..
……….
14
2.1.1.6.2 Hand weeding
…………………………..
…………………………..
…………………………..
………
15
2.1.1.6.3 Crop rotation and trap crops
…………………………..
…………………………..
……………..
15
2.1.1.6.4 Use of fertilizers
…………………………..
…………………………..
…………………………..
…..
16
2.1.1.6.5 Chemical control
…………………………..
…………………………..
…………………………..
……..
16
2.1.1.6.6 Use of resistant varieties
…………………………..
…………………………..
……………………….
17
CHAPTER THREE
…………………………..
…………………………..
…………………………..
………………..
20
3.0 MATERIALS
AND METHODS
…………………………..
…………………………..
………………….
20
3.1 Study one: Resistance of upland rice varieties against
Striga hermonthica
under field
conditions
…………………………..
…………………………..
…………………………..
…………………………..
…..
20
3. 1.1Experimental site
…………………………..
…………………………..
…………………………..
…………..
20
vii
3.1.2 Rice varieties used
…………………………..
…………………………..
…………………………..
…………
21
3.1.3. Experimental design and Layout
…………………………..
…………………………..
…………………
23
3.1.4.2.1 Cultural practices
…………………………..
…………………………..
…………………………..
……..
25
3.1.4 Data collection
…………………………..
…………………………..
…………………………..
………………
26
3.1.4.
1 Soil data
…………………………..
…………………………..
…………………………..
…………………….
26
3.1.4.2 Crop data
…………………………..
…………………………..
…………………………..
…………………..
27
3.1.4.3
Striga
data
…………………………..
…………………………..
…………………………..
…………………
28
3.1.4.4
Data analysis
…………………………..
…………………………..
…………………………..
……………..
28
3.2 Study two: Farmer participatory variety selection of upland rice varieties
……………….
29
3.2.1 Data collection
…………………………..
…………………………..
…………………………..
………………
29
3.2.2 Da
ta analysis
…………………………..
…………………………..
…………………………..
………………..
30
3.3 Study three: Evaluation of post germination attachment resistance of upland rice
varieties to
Striga
hermonthica
under controlled environment conditions
……………………….
30
3.3.1. Plant materials
…………………………..
…………………………..
…………………………..
……………..
30
3.3.2 Growth and infection of rice plants with
Striga hermonthica
…………………………..
……….
31
3.3.3 Non
–
destructive measurements of growth
…………………………..
…………………………..
……..
32
3.3.4 Destructive measurement of above ground rice biomass
…………………………..
……………..
32
3.3.4 Quantification of post
–
attachment resistance of the rice cultivars
…………………………..
…
32
3.3.5 The phenotype of resistance
…………………………..
…………………………..
………………………..
33
3.3.6 Statistical analysis
…………………………..
…………………………..
…………………………..
…………
33
viii
CHAPTER FOUR
…………………………..
…………………………..
…………………………..
……………….
34
4.0 RESULTS AND DISCUSSION
…………………………..
…………………………..
…………………..
34
4.1 Study one: Resistance of upland rice varieties against
Striga hermonthica
under field
conditions
…………………………..
…………………………..
…………………………..
…………………………..
…..
34
4.1.1 Soil characteristics of the field trial.
…………………………..
…………………………..
……………..
34
4.1.2
The growth parameters of upland rice varieties grown under
Striga
infested fields.
.
34
4.1.2.1 Plant height per variety
…………………………..
…………………………..
…………………………..
.
34
4.1.2.2 Number of tillers
…………………………..
…………………………..
…………………………..
………..
38
4.1.3 Grain yield components of rice varieties grown under
Striga
infested conditions.
…..
42
4.1.3.1 Rice biomass
…………………………..
…………………………..
…………………………..
……………..
42
4.1.3.2 Panicle dry weight
…………………………..
…………………………..
…………………………..
………
44
4.1.3.3 Harvest index
…………………………..
…………………………..
…………………………..
……………..
45
4.1.3.4 Grain recovery
…………………………..
…………………………..
…………………………..
……………
47
4.1.3.5 Rice grain yield
…………………………..
…………………………..
…………………………..
………….
48
4.1.4
Striga hermonthica
and its effect on growth of rice varieties grown under field
conditions.
…………………………..
…………………………..
…………………………..
…………………………..
….
54
4.1.4.1
Striga
counts per variety
…………………………..
…………………………..
………………………….
54
4.1.4.2 Days to
Striga
emergency and flowering
…………………………..
…………………………..
……
57
4.1.4.3 Maximum number of above
Striga
plants
…………………………..
…………………………..
…..
58
4.1.4.4
Striga
dry weight
…………………………..
…………………………..
…………………………..
………..
62
ix
4. 2 Study two: Farmer participatory variety selection of upland rice varieties
………………
65
4.2.1 Social demographic factors and Rice farming practices
…………………………..
…………..
65
4.2.2 Rice variety selection; the most preferred varieties by farmers in Eastern Uganda
…..
67
4.2.3 Ri
ce variety selection by gender; the most preferred varieties by male and female farmers
in Eastern Uganda
…………………………..
…………………………..
…………………………..
………………
68
4.2.4 The worst perf
orming varieties of rice according to farmers.
…………………………..
……
70
4.2.5 Farmer selection criteria for upland rice varieties.
…………………………..
…………………..
71
4.2.6 Variety ranking based on farmer selection criteria
…………………………..
…………………..
73
4.3
Study three:
Evaluation of post germination attachment resistance of upland rice
varieties to
Striga hermonthica
under controlled environment conditions
……………………
75
4.3.1
How does
Striga hermonthica
alter the morphology of host
…………………………..
………..
75
4.3.1.1 Plant height
…………………………..
…………………………..
…………………………..
…………….
75
4.3.1.2
Stem width
…………………………..
…………………………..
…………………………..
……………..
77
4.3.2 Effect of
Striga hermonthica
on tillering and leaf number of rice varieties
……….
80
4.3.2.1 Number of tillers
…………………………..
…………………………..
…………………………..
……..
80
4.
3.2.2
Number of leaves
…………………………..
…………………………..
…………………………..
…….
82
4.3.3 Rice cultivars exhibit differential resistance to
Striga hermonthica
.
…………………….
85
4.3.3.1 Quantification of post
–
attachment
Striga
germination
…………………………..
………….
85
4.3.3.2
Striga
hermonthica
dry weight
…………………………..
…………………………..
………………….
86
x
4.3.4 Effect of
Striga hermonthica
on above ground rice biomass of upland rice varieties with
respect to the uninfected rice plants.
…………………………..
…………………………..
…………………….
90
4.3.5 Impact of
Striga hermonthic
a on the biomass of rice varieties
…………………………..
……..
91
5.0 GENERAL DISCUSSION AND CONCLUSION
…………………………..
………………………..
94
REFERENCES
…………………………..
…………………………..
…………………………..
……………………..
102
APPENDICES
…………………………..
…………………………..
…………………………..
………………………
116
xi
LIST OF TABLES
Table 1: List of varieties screened in Nsinze, Namutumba District, Uganda
………………………….
21
Table 2
:
Plant height of selected rice varieties grown under
Striga
i
nfested field conditions for the
first and second
seasons.
…………………………..
…………………………..
…………………………..
……………
37
Table 3
:
Number of tillers of selected rice varieties grown under
Striga
infested field conditions
for the first and s
econd seasons.
…………………………..
…………………………..
…………………………..
…
41
Table 4
:
Yield components of rice varieties under
Striga hermonthica
infected conditions for 2014
in Namutumba E
astern Uganda.
…………………………..
…………………………..
…………………………..
…
52
Table 5
:
Yield components of rice varieties under
Striga hermonthica
infected conditions for 2015
in Namutumba Eastern Uganda.
…………………………..
…………………………..
…………………………..
…
53
Table 6
:
Number of
Striga
plants per rice variety for 2014
–
2015 in E
astern Uganda.
……………..
56
Table 7
:
Striga
components per rice variety in 2014 and 2015 Namutumba district Eastern Uganda
…………………………..
…………………………..
…………………………..
…………………………..
…………………..
64
Table
8
:
Farmer participant characterization: gender ratios, age
–
groups and rice farming
experiences of farmers in Namutumba district Eastern Ug
anda in 2015.
…………………………..
…..
65
Table 9
:
Reasons for rice variety selection among the least
–
liked, as indicated by farmers in Eastern
Uganda in
2015.
…………………………..
…………………………..
…………………………..
……………………….
71
Table 10
:
Criteria rice farmers use in selection of the best upland rice varieties in Eastern Uganda
in 2015
…………………………..
…………………………..
…………………………..
…………………………..
……….
72
Table 11
:
Ranking of the nine varieties of upland rice based on farmer selection criteria in Eastern
Uganda in 2015.
…………………………..
…………………………..
…………………………..
……………………….
74
Table 12
:
Height of rice varieties infected with
Striga hermonthica
and their respective uninfected
(control) grown under controlled conditions
…………………………..
…………………………..
…………….
77
xii
Table 13
:
Stem width of the infected and uninfected control rice plants of rice varieties grown
under controlled conditions.
…………………………..
…………………………..
…………………………..
……….
79
Table 14
:
Number of tillers of the infected and uninfected control rice plants per variety of selected
rice varieties under cont
rolled conditions
…………………………..
…………………………..
…………………
82
Table 15
:
Maximum number of leaves of the infected and uninfected control rice plants per variety
under cont
rolled conditions
…………………………..
…………………………..
…………………………..
………..
84
Table 16
:
Number of
Striga
plants and
Striga
dry weight taken at 21 days after
Striga hermonthica
infection under controlled conditions
…………………………..
…………………………..
………………………
88
Table 17
:
Above
–
ground rice dry biomass of infected rice plants and their respective uni
nfected
control plants under controlled conditions
……………………………
…………………………..
……..
91
xiii
LIST OF FIGURES
Figure 1:
Life cycle of
Striga
…………………………..
…………………………..
…………………………..
……..
11
Figure 2: Schematic representation of the field trial in Nsinze, Namutumba district, Uganda with
the ind
ication of the path, the homesteads and the water pump.
…………………………..
………………
24
Figure 3: The field trial in Nsinze, Namutumba district, Uganda in the
first season (2014) at 25
days after sowing.
…………………………..
…………………………..
…………………………..
…………………….
25
Figure 4: Farmer participatory preference of upland rice varieties in Namutumba district Eastern
Uganda in 2015, bars indicate the percentage (%) of farmers p
referring a certain variety.
………
68
Figure 5
:
Rice variety selection by gender of rice farmers in Eastern Uganda (Namutumba district)
in 2015, data presented as percentage responses by farmers.
…………………………..
…………………..
70
Figure 6:
Parasitic plants attached on the host plant roots 21 day
s after infection with
Striga
hermonthica.
…………………………..
…………………………..
…………………………..
…………………………..
.
89
Figure 7: Relationship between percentage losses in total rice biomass of infected
plants compared
with control plants and the amount of parasite biomass dry weight on roots of rice plants.
……..
92
xiv
LIST OF APPENDICES
Appendix 1: Initial soil characteristics of the field trial for season 2014 A and 2015 B in Eastern
Uganda.
…………………………..
…………………………..
…………………………..
…………………………..
…….
116
Appendix 2: Analysis of variance for plant height per rice variety for 2014
–
2015 under
Striga
field
conditions
………………………………………………………………………………………
116
Appendix 3: Analysis of variance for the number of tillers per rice variety for 2014
–
2015 under
Striga
infested field conditions.
…………………………..
…………………………..
…………………………..
..
116
Appendix 4: Analysis of variance for the yield components of rice varieties for 2014 and 2015
seasons under
Striga
infested conditions.
…………………………..
…………………………..
……………….
117
Appendix 5: Analysis of variance of number of
Striga
plants rice varieties for 2014
–
2015 in
Namutumba district Eastern Uganda
…………………………..
…………………………..
……………………..
117
Appendix 6: Analysis of variance of
Striga
components recorded per rice variety for 2014
–
2015
in Namutumba district Eastern U
ganda.
…………………………..
…………………………..
…………………
117
Appendix 7: Analysis of variance of days to
Striga
emergency and
Striga
flowering of rice
varieties for 2014
–
2015 in N
amutumba district Eastern Uganda.
…………………………..
……………
118
Appendix 8: Analysis of variance of Number of tillers and plant height rice varieties in 201
4 in
Namutumba district Eastern Uganda.
…………………………..
…………………………..
…………………….
118
Appendix 9: Analysis of variance of yield components of rice varieties in 2014 in Namutumba
district Eastern Uganda.
…………………………..
…………………………..
…………………………..
…………..
118
Appendix 10: Analysis of variance of Number of tillers and plant height rice varieties in 2015 in
Namutumba district Eastern Uganda.
…………………………..
…………………………..
…………………….
119
Appendix 11: Analysis of variance of yield components of rice varieties in 2014 in Namutumba
district Eastern Uganda.
…………………………..
…………………………..
…………………………..
…………..
119
xv
Appendix 12: Analysis of variance of number of
Striga
plants rice varieties for 2014 and 2015 in
Namutumba district Eastern Uganda
…………………………..
…………………………..
……………………..
119
Appendix 13: Analysis of variance of
Striga
components recorded per rice variety in 2014 in
Namutumba district Eastern Uganda.
…………………………..
…………………………..
…………………….
120
Appendix 14: Analysis of variance of
Striga
components recorded per rice variety in 2015 in
Namutumba district Eastern Uganda.
…………………………..
…………………………..
…………………….
120
Appendix 15: Analysis of variance for growth parameters of rice varieties grown under controlled
conditions.
…………………………..
…………………………..
…………………………..
…………………………..
…
120
Appendix 16: Analysis of variance for
Striga
number and Striga dry weight of rice varieties grown
under controlled conditions.
…………………………..
…………………………..
…………………………..
……..
121
Appendix 17: Questionnaire for participatory variety selection
…………………………..
……………..
121
xvi
ABSTRACT
Striga
is a major
biotic
constraint to rice production
in Sub Sharan Africa.
The use of resistant
rice
varieties
is a cost effective way of managing
Striga
in farmers’ fields.
However, because
of
existence
of
different ecotypes of
Striga
hermonthica
,
resistance of a particular variety in one area
does not guarantee its resistance in another area.
Therefore,
rice
varieties have to be tested in
different agro
–
ecological zones
to establish their resistance/ tolerance and susceptibility to these
ecotypes of
Striga
.
This study was therefore aimed at
screen
ing
a selected
rice varieties for
resistance to
Striga hermonthica
in Eastern Uganda
.
A field experiment
, a
pa
rticipatory variety
selection study
were
conducted in
Namutumba district star
ting
in the
first season of 2014 and
a
laboratory experiment
was
conducted at
the
Department of Plant and Animal Sciences
,
University
of Sheffield U.K
.
In the field experiment, a
5×5 lattice design was used, with 25 varieties
constituti
ng the treatments
.
Prior to planting,
the plots were supplied with 0.92g of
Striga
seed
mixed with 200g of white sand to homogenize
Striga
levels in each plot.
In
a
participat
ory selection
exercise,
38 farmers with 2
years’ experience in rice farming were interviewed.
Furthermore
,
i
n
the laboratory
experiment
,
a complete randomized design with a
total of
14 varieties each with
10
plants, 6 infected and 4 uninfe
cted
(
control
)
plants were used.
Each rice plant was infected with
12mg of sterilized
pre
–
germinated
Striga
seeds. Results
from the
field experiment
indicated
a
significant difference
(P<0.001) in
Striga
resistance
and yield
of introduced varieties
compared to
the local variety
.
Varieties
WAB928, Blechai,
SCRID090, WAB935, NERICA
–
10,
–
2,
–
17, IRGC,
IR49, WAB181
–
18, CG14, Anakila, WAB880
were the most
resistant.
These same varieties
,
except WAB928, WAB935
, IR49 and IRGC
produced higher
yield
s
.
In addition
,
the
same varieties
except
WAB935, N
ERICA
–
2, IR49 and Anak
ila demonstrated excellent post germination
attachment resistance to
Striga
under controlled conditions.
Furthermore,
farmers selected
xvii
varieties; SCRID090, WAB880, Blechai, WAB181
–
18 and NERICA
–
17 varieties
as the most
preferred
varieties
.
Striga
resistance, early maturity
and high
yield
ing
were
the
most important
traits in
farmer
variety selection
criteria
.
T
herefore, t
hese results are highly relevan
t to rice
breeders, agronomists and
molecular biologists working on
Striga
resist
ance. Similarly,
those
Striga
resistant varieties combining high yields and excellent adaptability to field conditions can
be recommend to farmers in
Striga
prone areas elsewhere in Uganda.
1
CHAPTER ONE
1.0
INTRODUCTION
1.1 Backgroun
d
Rice belongs to the genus
Oryza
, sub
–
fam
ily Oryzoidaea of family G
ramine
ae. It is a small
genus with approximately twenty two species (Vaughan, 1994). Twenty species of genus Oryza
are wild species and
there are
only two cultiv
ated species
Oryza sativa
L (Asian rice) and
Oryza
glaberrima
Steud (African rice).
Oryza sativa
is the most widely grown of the two cultivated
species.
Oryza glaberrima
can be distinguished from
Oryza sativa
by differences in ligule
shape, lack of seconda
ry branches in the panicle and
an
almost glabrous glume. In many parts
of Africa,
Oryza sativa
and interspecific crosses between
Oryza glaberrima
and Oryza sativa
are replacing
Oryza glaberrima
(Linares, 2002). Among such interspecifics is NERICA upland
ri
ce, which is as a result of backcrossing between the Asian and African rice
species
(
Jones
et
al.,
1997; Wopereis
et al.,
2008
). This is spreading in most parts of Africa because of its
resistance to drought and its ability to produce high yields.
It is t
hought that
African and Asian rice
were domesticated independently
(Khush, 1997). It is
suggested that
Oryza sativa
originated in India from
Oryza perennis
whereas cultivation may
have been earlier in China (Purseglove, 1975).
Oryza sativa
was first culti
vated in south
–
east
Asia, India and China between 8000 and 15000 years ago (Normile, 2004).
Oryza glaberrima
is believed to have been cultivated in the primary area of diversity in West Africa since 1500
BC while in the secondary areas
,
cultivation begun 5
00
–
700 years later (Porteres, 1956).
Rice plays an important role in many ancient customs and religious magical rites in the East
,
a
sign of antiquity of the crop. Its significance is connected with fecundity and plenty for example
rice cakes are eaten at
certain festivals in Asia to symbolize long life, happiness and abundance
(Purseglove, 1975). Rice straw can be fed to livestock, although it is not nutritious as compared
2
to straw of other cereals. It can also be used for the manufacture of strawboards,
for thatching
and brading as well as making of mats and hats. In some countries like China and Thailand rice
straw is used for making mushroom culture (Purseglove, 1975).
1.2 Overview of rice production in the world
Rice is the most important
staple food
for about half of the people in the world. Rice has been
cultivated worldwide for more than 10,000 years, longer than any other crop (Kenmore, 2003).
Globally, the area under rice cultivation is estimated at fifteen million hectares with an annual
producti
on of about 500 million metric tons (Tsuboi, 2004). This accounts for about 29% of
the total output of the grain crop worldwide (Xu and Shen
, 2003
). South and Eastern Asia
produce about 90% of world’s rice crop. China and India are the leading producers a
nd
consumers of rice in the world. In Africa, rice is becoming a major staple food among the diet
of many people. It is the staple food for about ten African countries. Rice is grown in more
than 75% of the countries in the African continent. In 2008, Afri
ca produced an estimated
quantity of 23 million metric tons of unmilled rice on 9.5 million hectares (FAOSTAT, 2010).
The major pr
oducing regions in Africa were Western, Northern, and E
astern with an estimated
production of 10.2 million metric tons, 7.3
metric tons and 5 million metric tons respectively.
These were harvested on 5.8 million hectares, 0.8 million hectares and 2.4
million
hectares
respectively (FAOSTAT, 2010)
. In East Africa, rice production
is considerably increasing due
to establishment o
f upland rice varieties.
1.3 Rice production in Uganda and importance
Rice is becoming a major food crop in Uganda. Changes in consumption trends and increased
population have led to increased production of rice over the years. The per capita consumption
o
f rice is estimated at 8 kg producing a total consumption of 224,000 metric tons by a
population of 28
–
30 million with an annual growth rate of 3.2%. Rice production was
3
introduced by Indian traders in Uganda in 1904 (MAAIF, 2009). Rice production only be
came
economically relevant in the late 1940’s when the government included rice
–
based food rations
in the diet of soldiers (Wilfred, 2006). The establishment of rice schemes in Kibimba in 1960
and Doho in 1976 led to production of lowland rice in Eastern
and Northern parts of the
country. In the 1980’s the production area under rice increased tremendously up to date. Since
1997, the annual unmilled rice production has increased from 80,000 metric tons in 1997 to
over 210,000 metric tons in 2010, representi
ng an annual growth rate of 12% (FAOSTAT,
2010). This increase in production is partly attributed to the introduction of upland rice varieties
particularly NERIC
A varieties in 2002 (Kijima
et al.,
2006
). Since the introduction of the
upland NERICA varietie
s
rice, area under cultivation has increased estimated at 72
,
000
hectares in 20
00 (Uganda Bureau of Statistics,
2002) and currently estimated at 90
,
000 hectares
(Uganda Bureau of Statistics
,
2012)
. About 80% of rice farmers are small
–
scale farmers with
les
s than two hectares with women playing a big role in rice production.
Rice production in Uganda has had a positive influence on the livelihoods of farmers. It has
also contributed greatly to the development of the
country by providing people with income
through
the selling of rice as grains. About 60% of the rice produced
in Uganda
is sold (PMA,
2009). This has improved the livelihoods of rice farmers. The demand for rice especially in the
urban centers has increased tremendously as a result of population
increase. It is estimated that
the population growth rate in Uganda increases by 3.2% annually (PMA, 2009; MAAIF, 2009).
This has acted as potential market for the rice that is produced by the farmers. According to
PMA (2009) and MAAIF (2009), Uganda’s ri
ce imports dropped from 77,600 tonnes in 2000
to 33,000 tonnes in 2010. This reduction has increased domestic market supply by the farmers
increasing their incomes.
4
1.4 Problem statement
In 2003, the government of Uganda encouraged the adoption of NERICA
high yielding upland
rice variet
ies, as one of the strategies for eradicating poverty and increasing
food security. Since
the adoption of NERICA varieties, rice production has risen both in acreage and volume of
production (MAAIF, 2011). However, despite
the rise in volumes, production per unit area is
currently declining in Uganda and one of the major constraints that has been identified as
contributing to this decline is the increase in
Striga
infestation. Most cereals including rice are
attacked by
two
species of
Striga
from the family Orobanchaceae namely;
Striga
hermonthica
(Del.) Benth and
Striga
asiatica
, (L.) Kuntze.
Striga
hermonthica
is however the most
im
portant
specie in
upland rice production in Uganda. It is estimated that 62,000 hectares of
farmland in Uganda is infested with
Striga
(AATF, 2006) and
this
infestation
can cause around
40
–
100% yield loss if not checked in the field (Oswald, 2005).
The problem of
Striga
is accentuated under conditions of low soil fertility especially low
nitroge
n and moisture levels. The recent increase in
Striga
infestation in Eastern Uganda is
attributed to two main reasons; the decline in cotton production, which acted as a trap crop
decreasing infestations since it was not a host crop for
Striga
and declining soil fertility as a
result of continuous cul
tivation of the soil w
ithout replenishment of used up
nutrients from the
soils. This
is partly due to
increase in the population wit
h many of the areas which would
be
under fallow being put to cult
ivation.
Previous studies on rice show that some rice varieties exhibit resistance to either
Striga
hermonthica
or
Striga
asiatica
or both (Johnson
et al.,
1997; Harahap
et al.,
1993; Cissoko
et
al.,
2011; Gurney
et al.,
2006; Rodenburg
et al.,
2015). Ciss
oko
et al
. (2011) and Jamil
et al.
(2011) assessed 18 upland NERICA varieties for pre
–
and post
–
attachment resistance to
Striga
hermonthica
and
Striga
asiatica
under controlled environmental conditions. According to these
5
studies, NERICA varieties includi
ng NERICA1, 10, 17 and 2 exhibited very good post
–
attachment resistance to several ecotypes of
S.hermonthica
(Cissoko
et al.,
2011). Additionally,
NERICA varieties produce different amounts and types of strigolactones in their root exudates,
which will al
ter pre
–
attachment resistance (Jamil
et al.,
2011). Inspite of the above successes
under controlled environmental conditions, there has been no published information about the
effect of the environment on the expression of resistance apart from Rodenburg
e
t al.
(2015)
who evaluated 18 NERICA cultivars and their parents under field conditions in Kyela,
Tanzania and Mbita, Kenya. They showed variation in field resistance among the NERICA
cultivars and their parents to different
Striga
ecotypes. In Uganda, NERICA varieties have not
been evaluated for
Striga
resistance since their introduction in farmer’s fields. Also, there are
a number of upland rice varieties being reported to be resistant to either
S.hermonthica
or
S.asiatica
in diff
erent areas of Sub
–
Saharan Africa (Harahap
et al.,
1993; Johnson
et al.,
1997).
Such varieties and NERICA varieties should be evaluated in different
Striga
infested agro
–
ecosystems to determine the level of resistance in the field.
1.5 Justification
Control of
Striga
hermonthica
presents a challenge and is complex because of the possible
interactions between ecotypes of the parasitic weed, the host and the environment (Oswald,
2004). A number of strategies have been proposed in the management of
Strig
a
in rice fields
for example application of nitrogenous fertilizers (Riches
et al.,
2005), use of chemical
herbicides such as 2, 4
–
D (Carsky
et al.,
1994a), crop rotations and improved fallow
management (Oswald, 2004) and hand weeding. However, none of the
se control approaches
has proven very effective in
Striga
weed management in rice production systems. Use of
resistant varieties is an important approach in
Striga
management due to its genetic nature. It
is also cost effective as farmers do not have to in
vest a lot of time and resources in
6
implementing this management
technology.
However b
ecause of the genetic variability
existing among different species and ecotypes of the parasite, resistance found in some varieties
may be overcome by a small subset of
S
triga
individuals within the seed bank leading to
development of a virulent population of
Striga
overtime (Rodenburg and Bastiaans, 2011).
Also some studies have shown high levels of variability existing within and between
Striga
populations in Kenya, Mali
and Nigeria (Gethi
et al.,
2005). This means that resistance of a
variety in one place with a particular ecotype of
Striga
does not guarantee resistance in another
place with another parasitic ecotype. Therefore varieties have to be tested at multiple /different
agro
–
ecosystems. In
addition NERICA varieties in Uganda have not been evaluated for
Striga
resistance since their
introduction in farmer’s fields.
Likewise farmers in one place may have
different variety preferences compared to farmers in another place. Also farmers need to have
a wide range of variety options so that they choose varieties that are not only resistant
/ tolerant
but also have desired characteristics such as grain yield, grain colour/ size, taste etc. This study
is
therefore aimed at screening
selec
ted
upland rice varieties sourced from different areas in
Africa for
S.hermonthica
resistance and identify
rice varieties that can be adopted by farmers
in Eastern Uganda.
1.6 Objectives of
the
study
The overall objective of this study was to screen selected upland rice varieties for resistance to
Striga
hermonthica
and identify farmer preferences in order to c
ontribute to
Striga
management and overall farmer livelihoods in Eastern Uganda.
The specific objectives were
the following;
I.
To determine
the resistance of twenty five
selected
upland rice varieties sourced
from different areas of
Africa against
Striga
her
monthica
under field conditions
.
7
II.
To identify
Striga
resistant and farmer
–
preferred upland rice varieties that can be
adopted by farmers in Eastern Uganda.
III.
To deter
mine the post
germination
attachment resistance
of
farmer
selected upland
rice varieties against
Striga hermonthica
under controlled conditions.
1.7 Hypotheses
i.
There is a significant variation in resistance among rice varieties to
Striga
hermonthica
.
ii.
Farmer preferred characteristics in
Striga
resistant upland ri
ce varieties don’t differ
from those identified by researchers.
iii.
The resistance of rice varieties to
Striga
hermonthica
expressed under field and
controlled environmental conditions does not differ.
8
CHAPTER TWO
2.0 LITERATURE REVIEW
2.1 Major
weeds of rice
Worldwide
,
weeds are estimated to account for 32% potential and 9% actual yield losses in
rice (Oerke and Dehne, 2004). In Africa, research has shown that there are 130
most common
weed species of which 61 species are found in upland rice fie
lds, 31 species in hydromorphic
and 74 species in lowland rice. The most common weed species in upland rice include;
Rottboellia
cochinchinensis
(Lour.) W. Clayton,
Digitaria horizontalis
Willd,
Ageratum
conyzoides
L; and
Tridax
procumbens
L., while
Agerat
um conyzoides
L., and
Panicum laxum
S.W are common in hydromorphic areas,
Cyperus
difformis
L.,
Echinochloa colona
(L). Link,
are most common in lowland rice (Rodenburg and Johnson, 2009). In general, most weed
species in lowland rice are from families G
ra
minaea 43%, Cyperaceae 37% whereas upland
rice
species from Graminaea
account for
36% and Compositae
only
16% (Rodenburg and
Johnson, 2009). Also important in upland rice production are parasitic weeds of genus
Striga
which include;
Striga
hermonthica
(De
l) Benth and
Striga
asiatica
(L.) Kuntze. Agronomic
factors such as inadequate land preparation, rice seed contamination, broad cast seeding in
lowlands, inadequate water and fertilizer management, mono cropping, delayed herbicide
application and use of p
oor quality seed are responsible for the major weed problems in rice.
2.1.1
Striga
The genus
Striga
is a major limiting biotic constraint in production of cereal crops. Members
of this genus are obligate hemi parasites, they are chlorophyllous but require a host to complete
their life cycle (Musselman, 1987). The genus
Striga
includes over 40
species a
nd of these
eleven species are considered parasitic on agricultural crops.
Striga
species infest an estimated
two
–
thirds of cereals and legumes in Sub
–
Saharan Africa, causing annual crop losses estimated
9
at US$7 billion and negatively affecting the livelih
ood of people in the region (Berner
et al.,
1995). Based on
host preference
,
Striga
species can be split
into two
groups. The first group
parasitizes
Poaceae
,
including major food and forage grains such as maize (
Zea mays
), sorghum
(
Sorghum bicol
or
), rice
(
Oryza sativa
) and millet. The second
group preferentially attacks
legumes including cultivated and wild species (Mohamed
et al.,
2001). According to Ramiah
et al.
(1983), the most important species of
Striga
in Africa include;
Striga
hermonthica
(Del.)
Be
nth;
Striga
asiatica
(L.) Kuntze;
Striga
gesneriodes
(Willd) Vatke;
Striga
aspera
(Willd)
Benth; and
Striga
forbesii
Benth. All
species except
Striga
gesneriodes
, which parasitizes
cowpea and
other wild legumes, parasitize
the major cereal crops in Africa
,
including rice
(Rodenburg
et al.,
2010)
.
2.1.1.1
Origin and distribution of
Striga
The genus
Striga
is predominantly Afri
can in origin and distribution and
about 30 species are
endemic to Africa (Mohamed
et al.,
2001). The prevalence and extent of
genetic diversity of
species in a particular geograph
ic area
are often indicators of
the place of origin
(Gebisa and
Jonathan, 2007). The vast t
r
opical savannah between the Semien Mountains of Ethiopia and
the Nubian hills of Sudan has the greatest biodiv
ersity of sorghum and millet which are in
fested
by
Striga
. These regions are
believed to be the origin of
Striga
hermonthica
(Del.) Benth and
Striga
asiatica
(L.) Kuntze (Gebisa and Jonathan, 2007).
Striga
gesneriodes
is thought to have
originated from Wes
t Africa,
Striga
hermonthica
has had the largest geographical distribution
with obligate out
–
crossing behaviour and it is found in
Sub
–
Saharan Africa with most
prevalence in western, central and eastern
Africa and some parts of south
western part of th
e
Ara
bian Peninsula across the R
ed sea.
The red
–
flowered and weedy ecotype
Striga
asiatica
(L.) Kuntze is mostly found distributed in eastern and southern Africa
, whereas the yellow
–
10
flowered ecotype is found in West Africa (Mohammed
et al.,
2001; Rodenburg
et a
l.,
2010)
.
The latter ecotype has no importance as a weed.
2.1.1.2
Taxonomy and Botany of
Striga
The genus
Striga
traditionally included in the Scrophulariaceae family is now included in
Orobanchaceaae (Olmstead
et al.,
2001). This was based on molecular
evidence with a result
that both genera
Striga
and Orabanche are in the same family. Acco
rding to Mohammed et al.
(2001) of the
approximately forty described species of
Striga
, thirty species are endemic to
Africa. The hemi
–
parasite
Striga
is a succulent, greenish yellow annual herb up to 35 cm tall,
usually branching from the base, glabrous or minutely puberulent. Each plant has a single,
large, tuberous primary haustorium 1
–
3 cm in diameter. The species have numerous
adventitious roots em
erging from subterranean scales; stem square but obtusely angled. Leaves
are opposite, appressed to the stem, scale
–
like, 5
–
10 mm × 2
–
3 mm.
The i
nflorescence
is
a
terminal bracteates spike. Flowers
are
bisexual, zygomorphic, 5
–
merous, not fragrant, sessile
;
the
calyx
is
tubular with 5 teeth at apex, 4
–
6 mm × 2 mm; corolla tubular, 2
–
lipped, up to 15
mm long, bent in upper part of tube, pale blue to dark purple, upper lobes 2, fused, sharply
recurved, up to 2.5 mm long, lower lobes 3, spreading, 3 mm long; s
tamens 4, 2 longer and 2
shorter; ovary superior, tubular, 2
–
celled, style terete, stigma 2
–
fid.
The f
ruit an ovoid capsule
1
–
2 mm × 3 mm, many
–
seeded. Seeds are very small, dust
–
like, with prominent encircling
ridges.
Striga
species are distinguished from
other root parasites by unilocular anthers and bilabiate
corollas with pronounced bend in the corolla tube. In some species such as
Striga
hermonthica
,
the corolla is bent within the calyx teeth,
while
other species have their corolla bent above the
calyx
as in
Striga
asiatica
. Other distinguishing features used include; indumentum type that
is either pubescence, hirsute and whether hairs are ascending or retorse (pointing backward),
11
stem shape with some species having either round, obtusely square and squ
are in cross section;
leaf lobbing and dentations whether leaf margins are lobed, serrate or smooth; inflorescence
types with some species having spike, raceme and the length of the inflorescence relative to
the vegetative stem; number of ribs on the calyx
tubes and length of the calyx teeth relative to
the tubes, corolla color and type of indumentum on corolla.
2.1
.1
.3
Life Cycle of
Striga
Figure
1
: Life cycle of Striga
(
Mweze
et al., 2015
)
Striga
species have a complex life cycle. Seeds are the sole source of inoculum, they are
produced in abundance ranging from 10,000
–
100,000 or more per plant (Pieterse and Pesch,
1983). First dormancy of the seeds has to be broken; seed dormancy can persist for
about six
months (Vallance, 1950). The seeds require warm and humid conditions for a period of one
–
two weeks for them to germinate (pre
–
conditioning or conditioning). The pre
–
germination
requirement of exposure to moisture, combined with temperatures abov
e 20
0
C for a period of
one week or more is probably a survival adaptation which prevents the seed from germinating
before rainy season is well established
(Berner
et al.,
1997)
. Before the rains the host roots are
12
not well established and expande
d to allow
successful
Striga
attachment. The next step is a
signal by a specific
bio
chemical germination stimulant from
the
host roots (Vall
ence, 1950;
Worsham, 1987). Thes
e groups of biochemical
s
have been identified as germination stimulants
for parasitic weeds;
d
ihydrosorgoleones
,
s
esquiterpelactones and
s
trigolactones. The latter are
the
most important biochemicals with respect to cereals (Bouwmeester
et al.,
2003). Production
of the stimulant by the host roots directs the
Striga
radicle to grow directly towards the source
of the stimulant; this growth is chemotropic (Chang and Lynn, 1986). After germination, a
series of chemical signals direct the radicle to the host root where it attaches and penetrates.
However, when the seedli
ng does not atta
ch to a host root within three to
five days, the seedling
dies (Worsham, 1987). Once penetration has taken place, an internal feeding s
tructure
(haustorium) is formed. Through the haustorium,
the parasitic weed establishes a xylem to
xylem
connection with the host roots (Worsham, 1987). This is also thought to require
biochemical triggers (Yoder, 2001). Th
r
ough the xylem to xylem connection with the host root,
the parasite obtains metabolites, water and amino acids from the host plants (Pre
ss
et al.,
1987a). As the host matures, the parasite emerges and begins to produce chlorophyll and
starts
to
photosynthesize (Saunders, 1933).
Reproductive strategies
range from autogamy to obligate
allogamy depending on the species (Musselman, 1987). Spe
cies known to be obligatory
allogamous are
Striga
hermonthica
and
Striga
aspera
(Safa
et al.,
1984; Mohammed
et al.,
2007)
. These require insect pollinators and can hybridize producing viabl
e pollen and virulent
offspring
. The other remaining species of
S
triga
are autogamous
(Musselman
et al
.,
1982).
Following reproduction, seeds are dispersed and then the cycle begins again.
2.1.1.4
Ecology of
Striga
Problems of
Striga
appear to be associated with degrade
d
environments
and freely draining
soils of rain
fed cereal production systems
and are most severe in subsistence farming systems
13
with little option for addition of external inputs
(Gurney et al., 2006; Parker, 2012)
.
Striga
is
prevalent u
nder conditions of
low soil fertility especially low nitrogen an
d moisture levels.
Striga
species have been described as indicators of low soil fertility and their infestation is
linked to low nutrient conditions (Oswald, 2005).
Striga
is common in tropical and subtropical
areas of Africa and some parts of India excep
t in extremely cold climates.
Striga
species require
a suitable temperature of 20
0
–
40
0
C under moist conditions for the seeds to germinate coupled
with host derived signals. Soil type and soil
pH
are probably not critical for growth, as
Striga
occurs in all
types of soils from sandy acidic soils to alkaline clay soils.
2.1.1.5
Nature of
Striga
damage
Host damage due to
Striga
infestation occurs
already
before the emergence of the parasite
, and
continues thereafter
. Initial symptoms occur while the parasite i
s still subterranean; they appear
as water soaked leaf lesions, chlorosis and eventual leaf and plant desiccation and necrosis,
severe stunting and drought like symptoms such as leaf margin curling also occur (Kim
et al.,
1991).
Parasitism by
Striga
specie
s reduce
s
the yields in upland rice in two main ways;
The parasite directly derives water and mineral nutrients from the ro
ot vascular system
retarding the host
growth and development (Press and Stewart, 1987
; Rodenburg
et al.,
2006;
Atera
et al.,
2011
). A
lthough
Striga
is chlorophyllous, its rate of photosynthesis i
s low;
therefore as much
of its
carbon is host derived (Pageau
et al.,
2003
). The nutrients flow from
the host’s vascular system to the parasites via
the haustorium;
this flow is facilitated by
the
high rate of transpira
tion in
Striga
that exceeds transpiration in the
host. One of the most
important
causes for
Striga
damage
is its effective competitive ability in depriving the host
plant of carbon, nitrogen and inorganic salts. This happens whil
e at the same time inhibiting
the growth and impairing photosynthesis of its host (Khan
et al.,
2006).
14
The parasite pathologically affects the growth and development of upland rice. This is
associated with the phytotoxic effects of
Striga
within days of at
tachment. It is hypothesized
that the parasite produces phytotoxic substances that affect the crop’s growth with even low
levels of infection resulting in
host
dehydration. The parasitic weed also alters the hormone
balance of the host plant (Frost
et
al.,
1997). High concentration of abscisic acid inhibits growth
by reducing the rate of photosynthesis. It was observed that introduction of abscisic acid to the
xylem stream of plants affects photosynthesis by suppressing ribulose biphosphate
carboxylatio
n (Fischer
et al.,
1986). Some reports show a decrease in the concentrations of
cytokinins and gibberellins with an increase in the concentration of abscisic acid in
Striga
infested hosts (Drennan and El Hiweris, 1979).
2.
1.1.6
Control of
Striga
Management
of
Striga
presents a challenge because of
the
competitive ability of the weed, high
seed production, mechanisms of dormancy and possible interactions between the parasitic
ecotype, host and environment. Varying degrees of control has been achieved throug
h cultural
methods such as crop rotations, intercropping, hand weeding, trap cropping and fertilizer
application. Use of biological control, chemical control and use of resistant cultivars have also
been applied to control
Striga
species
(Rodenburg
et al.,
2010)
.
2.1
.1
.6.1
Cultural control
There are a number of methods described in the management of
Striga
in the field
. These are
described below
;
15
2.1.1.6.2
Hand weeding
This is the most used method of
Striga
management by farmers in
Africa
. Weeding removes
emerging
Striga
shoots, preventing them from flowering and seed production, though
the
damage normally occurs before the parasitic weed emerges. For this reason the benefit of hand
weeding may not be realized in the cu
rrent season, but rather
in the subse
quent seasons
as
it
can reduce the
Striga
seed bank over the long term. Use of hand weeding reduced
Striga
infestation by 12% in the maize fields in the one season, when applied for four consecutive
seasons it reduced infestations by 26.6% in the fourth y
ear (Ransom
et al.,
1997). Timing of
weeding operations is essential in management of
Striga
, for example Parker and Riches,
(1993) recommend that hand pulling
should
be done
before
or right after
Striga
hermonthica
begins to flower to avoid seed
sett
ing.
The effectiveness of hand weeding can be increased in
combination with other methods of
Striga
control. Ransom and Odiambo (1994) reported
improved maize yields after integrating soil fertility
amendment measures with
hand weeding.
2.1.1.6.3
Crop rotation
and trap crops
Under suitable conditions successful control can be obtained by use of trap crops. Various grass
species parasitized by
Striga
can be sown
to stimulate germination of the seed and then
ploughed back before
the seed reaches maturity which
can
reduce the soil seed bank if
continuously carried out. For example Sudan grass (
Sorghum sudanense
) has been used in
management of
Striga
hermonthica
and ploughed 5 weeks after sowing (Ivens, 1989).
Other
trap crops like cowpea (e.g. Carsky
et al.,
1994b)
and pigeon pea (e.g. Oswald and Ransom,
2001) are possible means of lowering
Striga spp
. infestations in cereal production systems.
Crop rotation cycles disrupt the parasitic weed life cycles
by reducing the soil seed bank
.
R
otations including trap crops s
uch as cow pea (
Vigna unguiculata
), soy bean (
Glycine max
),
groundnut (
Arachis hypogeal
) and pigeon pea (
Cajanas cajan
) have all proved effective, by
16
stimulating and contributing greatly to reduction in soil seed bank (Eplee and Langston, 1991).
Rotations
with cow pea resulted in reduced infestation of
Striga
asiatica
in upland rice fields
in Tanzania (Riches
et al.,
2005). Use of the trap crops encourages suicidal germination of
Striga
seeds (Ramaiah, 1983) therefore use of the crops can be important to co
ntrol
Striga
.
2.1.1.6.4
Use of fertilizers
Several studies have reported success with the use of fertilizers in the control of
Striga
.
Striga
species are more prevalent in low fertility soils. Soil fertility technologies and mineral nutrients
stimulate the
growth of the host, at the same time affecting the germination of the weed
(Aberyena and Padi, 2003). Use of nitrogen fertilizers can reduce
Striga
infections, for example
application of Urea three weeks after sowing reduced
Striga
asiatica
infestations o
n upland
rice in Tanzania (Riches
et al.,
2005). Also Adagba
et al.,
(2002a) reported delayed and
reduced
Striga
infection in upland rice fields in Nigeria after using 90
–
120kgN ha
–
1
. Reduction
in
Striga
infestation as a result of use of fertilizer is believed to be
partly
due to decreased
biosynthesis of germination stimulants strigolactones (
Yoneyam
et al.,
2007a; 2007b)
.
Therefore use of nitrogen and phosphorous fertilizers can greatly reduce parasiti
c weed
germination (Lopez
–
Raez
et al.,
2009)
2.1.1.6.5
Chemical control
There are few herbicides that have been proved effective in control of
Striga
. Herbicides such
as 2, 4
–
D or MCPA can be used to kill established weeds (Ivens, 1989). Herbicide 2, 4
–
D
has
been used against
Striga
hermonthica
(Carsky
et al.,
1994a),
Striga
asiatica
(Delassus, 1972).
The herbicides control the weed post
–
emergenc
e
, but germinations continues after the residues
loose effectiveness
(Ivens, 1989). Also the weed exerts its ha
rmful effects before it emerges
above the ground. Therefore coating of seeds with herbicides can be effective at improving the
17
chemical control, for example use of herbicide coated Imazapyr maize seeds in East Africa has
been used as an effective method to
manage
Striga
hermonthica
and
Striga
asiatica
(Kanampiu
et al.,
2003). Imazapyr is an Imidazolinone herbicide, based on ALS inhibition with broad
spectrum and residual effects. Use of such herbicides can have an effect on the soil biology and
suppression
of
Striga
species (Ahonsi
et al.,
2004).
To date, no herbicide
–
tolerant rice varieties
are identified for the rain
–
fed uplands that can be used in combination with these ALS inhibiting
chemicals as seed coating (Rodenburg and Demont, 2009)
2.1.1.6.6
Use of
resistant varieties
Use of resistant crop cultivars is probably the most economically feasible and environmentally
friendly means of
Striga
control. In general cultivars of the African rice species (
Oryza
glaberrima
) show more
Striga
resistance than Oryza
sativa genotypes (Riches
et al.,
1996;
Johnson
et al.,
1997). An example is CG14 a cultivar of African rice that showed resistance
against
Striga hermonthica
and
Striga aspera
(Johnson
et al.,
2000). Mechanisms of resistance
in cereals can be categorized as post
–
attachment or pre
–
attachment resistance. Pre
–
attachment
resistance involve mechanisms that prevent the parasitic weed from attaching on the host, such
mechanisms include; the absence or
reduced production of germination stimulants (Hess
et al.,
1992), this was observed by Jamil
et al.,
(2011) who assessed eighteen NERICA cultivars and
their parents,
Oryza sativa
L. (WAB 56
–
50, WAB 56
–
104 and WAB 181
–
8) and
Oryza
glaberrima
Steud parent C
G14.The results of his study showed a significant variation among
NERICA cultivars and their parents for strigolactone production and
Striga
germination
confirming feasibility of this approach in rice. NERICA cultivars 7, 8, 11 and 14 represented
the top
five highest strigolactone producers (Jamil
et al.,
2011) and NERICA cultivars
–
6,
–
10,
–
15,
–
2,
–
9 and
–
5 showed an intermediate production levels and NERICA
–
1 and CG14
produced the smallest amount of strigolactones meaning highly resistant cultivars (Jam
il
et
18
al.,
2011). Other mechanisms are; inhibition of germination by the host (Rich
et al.,
2004),
inhibition or reduction in the formation of the haustorium (Rich
et al.,
2004), thickened host
root cell walls resulting in mechanical barrier to infection b
y the weed (Maiti
et al.,
1984;
Olivier
et al.,
1991). Post
–
attachment mechanisms involve mechanisms that prevent
attachment of the parasitic weed, these mainly include; failure of the
Striga
to establish the
xylem
–
xylem connections with the host as a resu
lt of blockage in vascular continuity an
inability to penetrate through the endodermal barrier (Gurney et al., 2006) or in some cases due
to a blockage in vascular continuity e.g. as observed in NERICA 10 following infection with
S. hermonthica (kibos isol
ate) (Cissoko et al., 2011). Other mechanisms are; hypersensitivity
reactions resulting in the death of host root tissue around the point of attachment discouraging
parasite penetration (Mohammed
et al.,
2003); also plants can have natural ability to relea
se
phenolic compounds such as the rice phytoalexins in infected host cells. Cissoko
et al.
(2011)
studied post
–
attachment resistance of eighteen NERICA cultivars and their parents, NERICA
cultivars and their parents exhibited a range of susceptibility to t
he
Striga
species. They showed
that some NERICA cultivars such as NERICA
–
7,
–
8,
–
9,
–
11 and
–
14 were very susceptible
supporting between 100 and 150 parasites per root system, while NERICA
–
1, 2 and 10
exhibited the greatest resistance supporting a smaller
number of parasites per root system
(Cissoko
et al.,
2011). It is clear, as with many host
–
pathogen systems, that different cultivars
of rice may be resistant to some ecotypes of
Striga
but susceptible to others. This is because
of genetic variation in b
oth
Striga
and the host; different ecotypes of
S.hermonthica
with
different levels and mechanisms of virulence exist and may overcome the genetic resistance on
some cultivars but not others (Mohamed
et al.,
2007; Scholes and Press, 2008). However,
Cissoko
et al,
(2011) discovered that NERICA
–
1 and
–
10 exhibited a broad spectrum of
resistance when subjected to different ecotypes of
Striga hermonthica
and
Striga asiatica;
this
shows that broad
–
spectrum resistance exists and may play an important role in
Strig
a
control
19
in rice. Also recently Rodenburg
et al
. (2015) evaluated the resistance of 18 NERICA cultivars
and their parents under field conditions. Interestingly nine of the 18 NERICA cultivars
(NERICA
–
1,
–
5,
–
10,
–
12,
–
13 and
–
17) and two of the NERICA pa
rents (WAB181
–
18 and CG14
showed excellent resistance to
S. hermonthica
ecotype from Mbita. These same varieties had a
good post attachment resistance to
S. hermonthica
ecotype from Kibos Western Kenya
(Cissoko
et al.,
2011) and pre
–
attachment resistance
to
S. hermonthica
ecotype Medani Sudan
(Jamil
et al.,
2011). This is a clear indication that these varieties have a broad
–
spectrum of
resistance to different ecotypes of
Striga hermonthica.
Most cultivars of the African rice
species
Oryza glaberrima
show
yield advantages under weedy conditions due to vigorous
growth, high tillering ability and large phanophile leaves (Johnson
et al.,
1998; Saito
et al.,
2010). Combining pre and post
–
attachment resistance is necessary in future if varietal control
of
Striga
is to become an important component in integrated
Striga
management (Cissoko et
al., 2011)
20
CHAPTER THREE
3.0
MATERIALS AND METHODS
The research undertaking involved three studies
conducted as follows (i) to determine the
resistance of
selected
upland rice varieties against
Striga hermonthica
under field conditions
(ii)
to
identify in
a
farmer participatory
manner
Striga
resistant
upland rice varieties
to be
adopted by farmers
(iii) To evaluate the post
–
germination
attachment resistance of select
ed
upland rice varieties under controlled conditions.
3.1
Study one: Resistance of upland rice varieties against
Striga h
ermonthica
under field
conditions
3
.
1
.1
Experimental site
The study was
carried out in
Nsinze, a small village in
Namutumba district i
n the eastern region
of Uganda. It is borde
re
d by Pallisa district to the north, Butaleja district to the east, Bugiri
district to the south, Iganga district to the South east and Kaliro district to the North West. The
district headquarters at Namutumba ar
e located approximately 88 kilometers by road North
West of Jinja, the largest city in the sub
–
region. It is located 00
0
51
’
N, 33
0
41
’
E. Namutumba
district has one county and six sub
–
counties, thirty six parishes and 233 villages. The district
has a tot
al population of 167,691 people with a population density of 208 persons per
k
m
2
. The
district receives a bimodal type of rainfall, which averages at 1,250mm. The topography
ranges
from
1,167
m a.s.l
to 1
,
249
m a.s.l (
Namutumba
Census report, 2007).
21
3.1
.
2
Rice varieties
used
The experiment
involve
d
twenty
–
four varieties of upland rice obtained from Africa Rice
Center, Tanzania and one local variety
used
as a control. These varieties were
selected
because
of their putative
different levels of resist
a
nce and yield (Table 1)
.
The
l
ocal check, Superica 1
(WAB165) is a high yielding but
Striga
–
susceptible
variety
.
Table
1
:
List of varieties screened in Nsinze, Namutumba District, Uganda
Variety names
Short name
Rice species
Striga
reaction
(resistance)
Information
ACC102196
ACC
O.glabberima
S.aspera
(Johnson
et al.,
1997)
Agee
Agee
O.glabberima
S.hermonthica
(Jamil
et al.,
2012)
Anakila
Anakila
O.glabberima
S.hermonthica
(Jamil
et al.,
2012)
WAB56
–
50
WAB50
O.sativa
Susceptible to
S.hermonthica
(Rodenburg
et al
2015)
CG14
CG14
O.glabberima
S.hermonthica and
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011)
IAC165
IAC
O.sativa
Susceptible to
S.hermonthica
(Cissoko
et al.,
2011;
Jamil
et al.,
2012)
IR49255
–
B
–
B
–
5
–
2
IR49
O.sativa
S.hermonthica
and
S.aspera
(Harahap
et al.,
1993; Johnson
et al.,
1997)
IRGC78281 (Ble
Chai)
BleChai
O.sativa
S.hermonthica
(Harahap
et al.,
1993)
IRGC81712 (IR
38547
–
B
–
B
–
7
–
2
–
2)
IRGC
O.sativa
S.hermonthica
(Harahap
et al.,
1993)
Makassa
Makassa
O.glabberima
Tolerant to
S.hermonthica
and
S.aspera
(Johnson
et al.,
1997)
MG12
MG12
O.glabberima
Tolerant to
S.hermonthica
and
S.aspera
(Johnson
et al.,
1997)
NARC3 (ITA257)
NARC3
O.sativa
Susceptible to
S.hermonthica
(Harahap
et al.,
1993)
NERICA
–
1
N1
Interspecific
hybrids
S.hermonthica
and
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011
;
22
Rodenburg
etal.,
2015
)
NERICA
–
10
N10
Interspecific
hybrids
S.hermonthica
and
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011
;
Rodenburg
etal.,
2015
)
NERICA
–
17
N17
Interspecific
hybrids
S.hermonthica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011
;
Rodenburg
etal.,
2015
)
NERICA
–
2
N2
Interspecific
hybrids
S.hermonthica
and
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011
;Rodenburg
etal.,
2015
)
NERICA
–
4
N4
Interspecific
hybrids
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011
;
Rodenburg
etal.,
2015)
SCRID090
–
60
–
1
–
1
–
2
–
4
SCRID090
O.sativa
No information
Superica1
(WAB165)
Superica
O.sativa
Susceptible to
S.hermonthica
Local check
UPR
–
103
–
80
–
1
–
2
UPR
O.sativa
Tolerant to
S.hermonthica
(Harahap
et al.
,
1993)
WAB 181
–
18
WAB181
O.sativa
S.hermonthica
and
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011)
WAB 56
–
104
WAB104
O.sativa
T
olerant to
S.asiatica
(Cissoko
et al.,
2011;
Jamil
et al.,
2011)
WAB 928
–
22
–
2
–
A
–
A
–
B
WAB928
O.sativa
S.hermonthica
and
S.aspera
(
Johnson
et
al.,
2000
)
WAB 935
–
5
–
A
–
2
–
A
–
A
–
B
WAB935
O.sativa
S.hermonthica
and
S.aspera
(
Johnson
et
al.,
2000
)
WAB880
–
1
–
32
–
1
–
1
–
P2
–
HB
–
1
–
1
–
2
–
2
WAB880
O.sativa
No information
23
3.1.3
Experimental design
and Layout
The field experiment was
laid out as a 5
×
5 lattice
square
design replicated six times.
The
twenty
–
five
varieties constitute
d the
treatments
. The above design
(a lattice square) and the
high number of replicates (6) was
used
to account for the inherent high heterogeneity of natural
Striga
infestation
in a farmer’s field (Rodenburg
et al.,
2015). Such spatial variation affects the
cor
rect interpretation of
Striga
resistance
screening based on the
number of emerged
Striga
plants.
The experimental area measured 50m ×15m
.T
he upland rice cultivars (treatments) constituted
the sub
–
plots of the experiment. Each
net sub
–
plot
, containing an in
dividual cultivar, was
1.25m
×
2.75 m
(3.
44m
2
) and was separated from the adjacent sub
–
plot
by
an open row (0.50 cm)
to
avoid neighbor effects and to allow easy access
. A distance of 1
m was
left between each
replicate for access and
separation. These plot
s were
maintained
throughout the two se
asons of
2014 and 2015 (Figure 2
).
24
Figure
2
: Schematic
representation of the field trial in Nsinze, Namutumba district,
Uganda with the indication of the path, the homesteads and the water pump.
Plots
1
2
3
4
5
Reps
Blocks
3m
3m
3m
3m
3m
1
1
SCRID090
WAB181
WAB50
N17
IR49
2
IAC
Anakila
N10
N4
WAB880
3
BleChai
CG14
WAB104
MG12
Superica
4
Agee
N1
Makassa
N2
WAB935
5
NARC3
IRGC
UPR
WAB928
ACC
1m
2
1
N1
UPR
WAB50
WAB104
IAC
2
IRGC
SCRID090
Agee
MG12
WAB880
3
IR49
N2
WAB928
CG14
N10
4
Anakila
Superica
Makassa
NARC3
WAB181
5
ACC
WAB935
N17
BleChai
N4
1m
3
1
Anakila
N1
BleChai
WAB928
SCRID090
2
N17
UPR
WAB880
CG14
Makassa
3
WAB50
N10
Superica
IRGC
WAB935
4
N2
MG12
WAB181
ACC
IAC
5
WAB104
NARC3
IR49
Agee
N4
1m
4
1
WAB880
WAB104
WAB935
WAB181
WAB928
2
WAB50
Agee
CG14
Anakila
ACC
3
NARC3
N10
N1
N17
MG12
4
UPR
N4
Superica
SCRID090
N2
5
IR49
IAC
BleChai
IRGC
Makassa
1m
5
1
Anakila
N17
N2
IRGC
WAB104
2
WAB935
IAC
NARC3
CG14
SCRID090
3
WAB50
MG12
WAB928
Makassa
N4
4
BleChai
UPR
WAB181
N10
Agee
5
N1
IR49
WAB880
Superica
ACC
1m
6
1
WAB104
Makassa
ACC
N10
SCRID090
2
CG14
IRGC
WAB181
N4
N1
3
NARC3
WAB50
N2
WAB880
BleChai
4
Anakila
IR49
MG12
WAB935
UPR
5
IAC
Agee
N17
WAB928
Superica
Farmers'
homesteads
Water
pump
PATH
15 m
7.5 m
7.5 m
7.5 m
7.5 m
50 m
7.5 m
7.5 m
25
Figure
3
:
The field trial in Nsinze, Namutumba district, Uganda in the first season
(2014)
at 25
days after sowing.
3.
1.4
Cultural practices
Field preparation was
don
e
two weeks before the beginn
ing of the rains. The field was prepared
using an oxen plough
. Field preparatio
n was
done
on 2 March 2014 for the first season and on
3 March 20
15
for the second season
.
Striga
seed collected from farmers’ fields
were
used for artificial infestation of each sub
–
plot
to supplement
and homogenize
the
existing
Striga
seed bank
of the soil
. Ea
ch sub
–
plot
receive
d
a
similar amount of seed (~0.92g) mixe
d with 200ml of white construction sand.
This s
and
–
seed mixture
was
applied to designated sub
–
plots and incorporated in the upper 5
–
10cm of soil
using a hoe. Only the part where rice is sown 1.25
×
2.75m
, hence
3.44m
2
was
infested
.
Sowing was
done at the onset of rains
,
on
12 March 2014
and
on
13 March 2015
at a rate
of
six seeds per hill to a depth of 2
–
3 cm
following a spacing of
0.25m between rows and 0.25m
between plants
in the row
.
A total of 55 hills were maintained per sub
–
plot constitu
ting 5 rows
and 11 hills of plants per row. Thinning was done
at
two weeks after sowing leaving 3 plants
per hill. Weeding was done
regularly every 10 days after
Striga
infestation
to
remov
e all
weeds
26
other than
Striga
.
A b
asal application of
(17:17:17)
N
PK fertilizer
, at a rate equivalent of
50kg
ha
–
1
, was applied at 35 days after planting. Each sub
–
plot received 69g of NPK fertilizer.
T
he major pests of upland rice such as stalked
–
eyed flies, sting bug and rice bug w
ere
controlled using contact insectic
ide
Dursban (Chlorpyrifos),
every after 2 weeks starting 30
days after
planting un
til grain formation.
N
ematodes and termites
were
controlled
using
F
uradan
(Carbofuran) applied in plant holes before sowing
.
3.1.5
Data collection
3.1.5
.1
Soil data
Soil
of the upper 20 cm
w
as
sampled prior to
sowing
in each block
.
T
hree
randomly selected
but
distinct points
were sampled
per
replicate and
then
combined into one sample
equivalent
to 200g
. This was
used for
the
analysis of nitrogen, phosphorous, potassium,
organic carbon
and soil texture
.
Soil
analysis w
as
done at Makerere
U
niversity
, in the
soil science laboratory
.
Organic carbon content of the soil
were
determined by reduction of potassium dichromate by
organic carbon compounds and oxidation
–
reduction titr
at
ions with ferrous ammonium sulpha
te
solution
(Okalebo
et al.,
2002)
.
Soil texture was
determined by a hydrometer method
(differential settling within the water) using particles less than 2mm diameter (
Okalebo
et al
.
,
2002
). This procedure measures
percentage of sand (0.05
–
2.0mm), silt (0.002
–
0.05mm) and
clay (<0.002mm) fraction in soils
.
T
otal nitrogen content was
determined by the Kjeldahl
method (Dewis and Freitas, 1975).
Soil pH was
determined using a ratio of 1:2
.5
soil to water
ratio (Okalebo
e
t al.,
2002) and read from
a glass
electrode
attached to a digital pH meter
after
shaking samples for 1 hour.
Potassium was
determined using
Ammonium acetate extraction
method
a
nd
flame photometer
as described in Okalebo et al., (2002)
.
Available phosphoro
us
27
was determined using the Bray 1 method
since the
pH
of the soil was below 7 (Oisen and
Sommers, 1982)
.
3.1
.
5
.
2
Crop
data
Data
collected on rice included
;
plant height, above ground
dry
biomass
at harvest, number of
tillers
, panicle dry weight, harvest
index, grain recovery efficiency
and grain yield
(expressed
at 14% grain moisture content)
.
T
his was collected on a random sample of 9
hills per
sub
–
plot
.
Tiller numbers
were assessed
starting
43
,
57
and 113
days after
rice sowing.
Plant height was
determi
ned
with a
tape measur
e at 43,
57
and 113
days
after
sowing
. The
height
of the tallest
plant was measured from the soil surface to the tip of the tallest leave or the tip of the tallest
panicle
.
Above
–
ground
rice
biomass dry weight
was
determined at
harvest
.
For each subplot
the rice straw of the
plants of
9
hill
s
was
harvested, air dried for about 1
–
2 weeks and then oven
dried for 72 hours at 70
0
C. Directly after oven drying the weights
were
assessed using a
weighing scale and recorded.
Grain yield
at
grain
maturity
was
determined from a random sample of 27
rice hills
per sub
–
plot, this
was
expressed in kg
per
hectare
, corrected at 14% grain moisture content
. Panicles
were
cut and collected for each plot and air
–
dried. After 1
–
2 weeks of air
–
drying t
he panicles
were
weighed
to get panicle dry weight
. The panicle
s
were then
threshed and the grains
obtained after threshing
were
weighted. Then the grains
were
winnowed (removing all empty
grains) and weighted again, immediately followed by grain moisture
content measure
ment
s
which
will enable correction of all grains to a standard 14
% moisture content. All
measurements were
recorded.
Rice grain recovery efficiency was calculated from the
ratio of
unwinnowed and winnowed grains
multiplied by 100,
following
Grain recovery efficiency =
∗
100
28
Harvest in
dex was computed as a ratio of
grain dry weight and total
above ground
dry matter
using the formula
,
Harvest index
=
3.1
.
5
.
3
Striga
data
Data
collected on
Strig
a include;
Striga
emergence, days to
Striga
flowering,
Striga
dry weight
at rice harvest and
above
–
ground
Striga
numb
ers
as described below
. Above
–
ground
Striga
numbers
were
assessed by close examination of rice
varieties in the field
. R
egular
Striga
counts
were
done at
2
–
weekly
intervals, starting after
Striga
emergence
,
to assess maximum above
–
ground
Striga
numbers (NS
max
).
These counts
include
d
the number of living plants within the
central
observation area con
taining 27 hills (hence excluding the border rows) and the number
of dead plants, recorded separately.
Striga
counts were
taken
at
43, 57, 71, 85 days
after rice
sowing
and at rice maturity.
In each sub
–
plot, d
ays to
Striga
emergence
were
monitored and
rec
orded
every two days
starting at 28 days after
sowing
. Once
Striga
had
emerged in a plot, it
was
marked with a colo
ured stick to exclude it from
t
he
next round of observations.
Days to
Striga
flowering
were estimated based on observations done every 3
days,
starting at 21 days
after first
Striga
emergence in each sub
–
plot
. For assessment of
Striga
biomass dry weight, in
each subplot, all above
–
ground
Striga
plants
were
collected from the central observation area
containing 27 hills. During the season
S
triga
plants that die
d
before rice harvest
were
collected
and put in an envelope designated to the associated
sub
–
plot. At harvest all above
–
ground parts
of all of the
collected and
remaining
Striga
plants (dead or living)
were bulked
and oven
–
dried
for 72
hours at 70°C after which dry weights
were
obtained using an electronic
weighing scale
.
3.1.6
Data analysis
Prior to analysis data was subjected to tests of normality and homogeneity of variance (Sokal
and Rohlf, 1995). For analysis of above
–
ground
Striga
numbers (NSmax) and
Striga
dry
29
weight,
data were transformed using Natural logarithms, log (x+1) to meet the requirement of
normality of data distribution for classical ANOVA (Sokal and Rohlf, 1995). Spearman rank
correlations were also calculated between
Ls means of NSmax and rice yields, NSmax and
tiller number and NSmax and
Striga
dry weight. All the d
ata
were
subjected to statistical
analysis of variance
(
ANOVA
) using GenStat computer software (V12).
The means were
separated using the least significant
difference (LSD) test at 5% probability level.
3.2 Study two
:
F
armer participatory variety selection of upland rice varieties
This study was undertaken at the grain maturity stage in the variety screening trial in Nsinze,
Subcounty, Bulagala village, Namu
tumba district in the main rainy season of 2015. The field
screening trial was set up as described before in Chapter three. An open
–
ended questionnaire
was set up to avail information on variety preferences and traits that are important to farmers.
3.2.1
D
ata collection
Information was collected using q
uestionnaires with open
–
ended and structured questions
administered through personal interviews
(Appendix 1
7
)
. A purposive
sample was used,
selecting
upland rice farmers
with 2 years and above experience in r
ice production. These were
selected
from
four rice farmer
’
s groups in the district i.e. Abendowoza Ndala group in
Bulagala, Bukonte mixed farmers group, Ivukula integrated farmers’ association in Ivukula
Subcounty, Emberi Enkula
mixed group. Questionnaires were first pre
–
tested so as to get
acclimatized with the contents of the questionnaires.
Participants were asked to select the five best
va
rieties among 25 varieties, in terms of their
appearance in the field. In order to sele
ct the five best varieties, scores were assigned from 1 to
10; with 10 meaning ‘superior’ and score 1 meaning ‘inferior,’ these were used to rank the
most preferred varieties. Farmers were also asked to provide the five worst performing varieties
30
in the fi
eld trial.
To avoid biased c
hoices, participants were not
allowed to communicate to
each other during the selection exercise.
After making the choices and considerations,
participants were engaged in a one
–
on one discussion with extension officers and rese
archers
to avail data that would explain why the choices were made. This was aimed at getting diverse
views on preferences and selection criteria by farmers. Farmers were also asked to provide
information on the most preferred traits in rice varieties and
rank them from 1 to 5, with
1
indicating ‘not important at all’ and 5 indicating ‘very important’. Farmers evaluated the
expression of these traits in the selected varieties, to avail whether these varieties would
correspond well enough to these criteria
to disseminate them among Ugandan farmers.
3.2.2
Data analysis
Data obtained from questi
onnaires and interviews was
coded, entered in
a database
and
analyzed using descriptive analysis procedures of the statistical package for social scientists
(SPSS
,
2000
)
version 16 computer package
. Graphs, frequency tables
, means
a
nd percentages
generated were
used to summa
rize responses from respondents. To find whether there was a
significant difference in variety selection with respect to gender, data was restructure
d keeping
sex as a fixed variable and varieties as target variables. Then an independent sample t
–
test was
performed.
3.3 Study three
:
E
valuation of post
germination
attachment resista
nce of upland rice
varieties to
Striga hermonthica
under controlled en
vironment conditions
3.3
.1. Plant materials
An evaluation of resistance
of upland rice varieties against
Striga hermonthica
under controlled
environmental conditions
was carried out at the University of
Sheffield
, Department of Animal
and Plant Sciences. This was intended to understand the
Striga
resistance of a selected group
of upland rice varieties using the
Striga
hermonthica
ecotype
found
in
Namutumba district,
31
Uganda
. Rice seeds were provided by the Genetic
Resources Unit of Africa Rice.
Striga
hermonthica
(Del.) Benth seeds were obtained from farmer fields in Namutumba district in
Eastern Uganda.
3.3
.
2
Growth and infection of rice plants with
Striga hermonthica
Rice seeds were germinated between blocks
of mo
istened horticultural rock wool for six days
after which a s
ingle rice seedling was
transferred to a root observation chamber (
r
hizotron)
as
described previously by Gurney
et al.
(2006)
. Each
r
hizotro
n
consist
s
of a 25
×
25
×
2
cm
3
p
erspex container packed
with vermiculite onto w
hich a 100µm polyester mesh was
placed.
Roots of the rice seedlings gre
w down the mesh, and openings at the top and bottom of the
rhizotron
allow
ed
for shoot growth and water drainage
respectively. Rhizotrons were
covered
with
alumi
num
foil to prevent light from reaching the roots.
Rhizotrons were
supplied with
25ml of 40% long Ashton (Hewitt, 1960), nutrient solution containing
2mM ammonium nitrate
three
times each day via an automatic watering system.
Striga
seeds were
sterilized
in 10% bleach, washed thoroughly
with dH
2
O
and then incubated
on moistene
d
glass
–
fiber filter paper
(Whatman),
in petri
–
dishes for 12
–
15 days at 30
0
C (Gurney
et al
., 2006).
E
ighteen hours before infection of rice seedlings, 1ml of 0.1ppm solution of an
arti
ficial germination stim
ulant GR24 was
added to
P
etri dishes containing conditioned
Striga
seeds to stimulate germination.
T
wo weeks
after
sowing, rice plants were
infected with 12mg
of germinated
Striga
seeds by aligning them along the roots using a paint
brush (Gurney
et al.,
2006). Infection of rice roots with germinated
Striga
seeds eliminates differences due to
variation in production of germination stimulants by the different rice cultivars (Cissoko
et al.,
2011; Jamil
et al.,
2011
; Gurney
et al.,
2006
). Uninfected rice cultivar
s acting as the control
were
treated in a similar way as described above without the infestation step.
A total of 10
plants were sown per variety with four controls (uninfected) and six infected plants. Plants
32
were grown in a tem
perature controlled growth chamber with 60% relative humidity and day
and night temperatures of 28
0
C and 24
0
C respectively. The irradiance at plant height was 500
μmol m
–
2
s
–
1
3
.
3
.
3 Non
–
destructive measurements of growth
A series of
non
–
destructive growth
measurements were made each week from the day of
infecting rice seedlings with germinated
Striga
hermonthica
seeds including the height of the
main stem determined using a ruler measured from the stem base until the attachment of the
new leaf, number of l
eaves on the main stem, stem diameter, using a digital Vernier caliper
(Mitutoyo England UK), number of tillers and tiller leaves
3.3.4
Destructive
measurement of above ground rice biomass
Above
–
ground rice biomass was
determined
three weeks after infecti
on of the rice cultivars.
Plant materials were
separated into main stem, tiller stem, main stem leaves and tiller stem
leaves. Plant material was dried at 70
0
C for 7 days and then weighed to determine dry biomass.
3.3.5
Quantification of post
–
attachment resistance of the rice cultivars
Po
st
–
attachment resistance was
quantified 21 days after infection of the rice roots. Before
harvest, the root system of each
rhizotron was
photographed using Canon EOS 300D digital
camera.
Striga
seedlings growing on the roots
of each infected plants were
harvested and placed
in
P
etri dishes and photographed using a Canon EOS 300 digital camera. The number and
length of
Striga
seedl
ings from each rice plant wer
e determined from the
P
etri di
shes
photographs using Image
–
J.
Striga
plants were then
dried at 48
0
C for 2 days and the amount
of
dry biomass per host was determined.
33
3
.
3
.
6
The p
henotype of resistance
The
phenotype of resistance was
investigated by photographing parasites developing on
the
root systems of each cultivar at different stages after infection using a Leica MZFLIII stereo
microscope and a diagnostic i
nstruments camera Model 7.4 and
by cutting small sections of
root with attached parasite and mounting on a glass slid
e in water
. The root tissue was
observed
using an Olympus BX51 microscope employing differential interference contrast microscopy
and photographed using a digital
c
amera
.
3
.
3.7
Statistical analysis
Data w
ere
subjected
to
A
nalysis of
V
ariance
(
ANOVA
) using GenStat
computer software
(v12), m
ean val
u
es
were
separated using LSD at (
P
=0.05)
and Tukey multiple comparison test
(P< 0.05)
.
To meet the requirement of normality of data distribution for classical ANOVA
(Sokal and Rohlf, 1995),
Striga
emergence counts
and
Strig
a
dry weight
were transformed
using Natural logarithms, log (x+1), and stem width was square
–
root transformed.
34
CHAPTER FOUR
4.0 RESULTS
AND DISCUSSION
4.1 Study one: Resistance of upland rice varieties against
Striga h
ermonthica
under field
conditions
4.1.1
Soil characteristics of the field trial
.
Results of the chemical and texture a
nalysis of
the
soil
of the experimental field at the onset of
the season in 2014 and 2015
(Appendix 1).
The pH was 6.1, and within range of optimum soils
for
upland rice established by Somado et al. (2008).
Availa
ble nitrogen was highest in 2015
,
phosphorous was highest in 2014 while potassium was highest in 2015.
According to Gerakis
and Baer. (1999) the soil can be described as sand clay loam.
The remaining m
icronutrie
nts
were highest in 2014 (Appendix 1
).
Organic matter of the soil was above the minimum >2%
in 2014, but below this threshold in 2015. The nitrogen content was below the minimum
threshold (<1.0N/kg) established by Oikeh et al. (2008) in both year
s. Oikeh et al. (2008),
described nitrogen levels below 1.0g N/kg as low, between 1.0 and 2.0 N/kg as medium and
above 2.0g N/kg as high. The soil phosphorus level was above the minimum of 3 mg/kg,
suggested by Wopereis et al. (2009) and Oikeh et al. (200
8), in 2014 but below this minimum
in 2015
(Appendix 1)
. The potassium content of the soil was below the minimum of 0.2
cmole/kg in 2014, however it was above the minimum in 2015 (Oikeh
et al.,
2008). The
remaining micronutrients except Calcium, Sodium and
Magnesium recorded low values in both
seasons.
4.1.2
The growth parameters of upland rice varieties grown under
Striga
infested fields.
4.1.2.1
Plant height per variety
For both seasons,
variety x season interaction
effects
on plant height was highly sig
nificant
(P<0.001) (Appendix 2)
.
Rice varieties ha
d a highly significant effect (P
<0.001) on plant height
35
at 43, 57 and 113 days after sowing in both seasons
(
Appendix 2). In 2014, varieties differed
significantly (P< 0.001) in plant height (Appendix 8). V
ariety MG12 had significantly P< 0.001
Shortest stems at 43 days (21.13cm) and 57 days (28.47cm) after sowing (Table 2). On the
other hand Narc3 (53.26cm) and UPR (53.76 cm) had significantly P<0.001 shortest stems at
113 days after sowing compared to othe
r varieties with tallest stems than the above (Table
2).Variety Superica had the highest plant height at 43 days and WAB56
–
50 at 57days after
sowing. In the same season, Blechai produced significantly P< 0.001 the tallest plants across
varieties at 113days
after sowing (Table 2). Plant height ranged from 50.25 cm for Superica to
21.13 cm for MG12 43 days after sowing, 66.47cm for WAB56
–
50 to 28.47cm for MG12 57
days after sowing and 128.30cm for Blechai and 53.26cm for UPR 113 days after sowing
(Table 2).
In 2015
varieties differed significantly (P<0.
001) in plant height (Appendix 10
).
Variety
Blechai was the tallest at 43 days, 57 days and 113 days after sowing in 2015 (Table 2).
Varieties CG14 (20.95cm), Agee (21.43 cm), UPR (22.56 cm) had significantly
(
P<0.001
)
the
shortest stems at 43 days after sowing compared to other varieties except Narc 3 which was
not significantly (P> 0.001) from UPR (Table 2). In the same season, vari
eties UPR (31.70
cm)
and CG14 (33.22 cm)
produced s
ignificantly ( P< 0.001) the shortest stems at 57 days
after sowing compared other varieties except Narc3 which was not significantly different form
UPR (Table 2). Varieties IAC 165 (38.09 cm), Narc 3 (48
.32 cm) produced significantly
(
P<
0.001
)
the shortest s
tem compared to other varieties in the field
(Table 2). Plant height
ranged
from
36.12cm for Blechai to 20.95cm for CG14 43 days after sowing, 58.73cm for Blechai to
31.70cm for UPR 57 days after sowi
ng and 123.60cm for Blechai to 34.55cm for Superica 113
days after sowing (Table 2
).
It was observed that the tallest varieties were also the most resistant and some of the shortest
varieties were among the most
susceptible varieties (Table 2
). This could
be partly attributed
36
to differences in
Striga
infestation levels among varieties. For better assessment of the effects
of
Striga
on stem stunting of rice plants, you would need
Striga
–
free control plants which are
provided in
study 3
. Varieties IAC 165, S
uperica had short stems probably due to high
Striga
infestation.
This was very evident when their stems were reduced from 48.76 cm
to 38.09 cm
for IAC 165
and 50.17 cm to 34.55cm for Superica in 2015 at 57 days and 113 days after
sowing respectively.
Simil
ar results were reported by Atera et al. (2012) who also showed a
reduction in plant height of the susceptible rice varieties in Western Kenya. Johnson et al.
(1997) also reported high levels of
Striga
infestation reflected in increased levels of stunting
in
rice plants of the susceptible cultivars, including IAC 165.
However these varieties had an
increase in plant height in 2014 113 days after sowing. This could be related to low
Striga
numbers in this this season than 2015, hence these varieties were not highly affected as in 2015.
Other varieties such as Makassa, IR49, ACC, with the highest plant height as per this study
were reported by Johnson et al. (1997) as having the lowest stun
ted levels in Ivory coast. The
other reason for the shortness of the rice plants could be related to the genetic make
–
up of the
varieties. For example varieties WAB928 and WAB935 inspite of being short were highly
resistant to
Striga
in both seasons. The
O
ryza glaberrima
varieties, on the other hand, with
intermediate resistance levels, were all relatively tall. Previous studies have shown that
O.glaberrima
cultivars, have a different plant type than
O sativa
cultivars and are therefore
often more competiti
ve to weeds (Johnson
et al.,
1995). NERICA varieties and Other
O.sativa
varieties were tall mainly because of reduced
Striga
infestation or simply because they are
genetically tall varieties.
37
Table 2: Plant height of selected rice varieties grown under
S
triga
infested field conditions
for the first and second seasons.
`
Plant height (cm)
2014
2015
Days after sowing
Days after sowing
Rice variety
43
57
113
43
57
113
ACC
37.70
54.67
102.2
0
27.42
47.40
102.9
0
Agee
26.31
44.04
84.53
21.43
43.70
80.79
Anakila
32.74
48.53
89.89
25.69
41.86
88.81
Blechai
43.77
60.23
128.3
0
36.12
58.73
123.6
0
CG14
23.03
35.00
93.31
20.95
33.22
95.59
IAC
45.58
59.78
77.49
31.89
48.76
38.09
IR49
39.24
50.00
82.05
36.00
53.11
90.65
IRGC
35.48
44.29
75.36
31.69
45.54
86.01
Makassa
36.05
52.35
108.50
26.47
43.59
105.26
MG12
21.13
28.47
78.62
28.92
45.61
96.72
NARC3
27.14
35.59
53.72
24.60
35.96
48.32
NERICA
–
1
43.62
54.76
87.82
28.60
47.31
82.87
NERICA
–
10
43.05
60.23
87.41
29.20
47.76
87.80
NERICA
–
17
42.78
57.53
93.80
30.82
51.87
94.11
NERICA
–
2
41.27
58.51
90.45
28.25
46.88
95.45
NERICA
–
4
42.15
58.57
98.56
29.75
48.98
90.64
SCRID090
40.73
57.00
96.43
30.42
53.46
90.60
Superica
50.25
62.08
76.40
33.37
50.17
34.55
UPR
27.04
32.47
53.26
22.56
31.70
49.62
WAB181
–
18
46.20
53.57
85.02
28.86
49.68
82.91
WAB56
–
50
49.37
66.47
93.98
33.78
55.44
83.16
WAB56
–
104
45.50
59.22
88.45
33.68
52.83
72.95
WAB880
46.28
62.38
98.81
33.11
55.80
100.4
WAB928
37.13
43.97
75.04
31.66
41.63
69.84
WAB935
39.75
46.14
73.12
32.50
46.12
73.80
Means
38.53
51.43
86.90
29.51
47.08
82.62
LSD (0.05)
4.53
5.32
9.79
2.56
4.35
13.91
S.E
3.47
4.34
7.8
2.02
3.44
11
CV%
8.89
8.24
8.98
6.86
7.33
13.37
38
4.1.2
.2
Number of tillers
Variety x season interaction effects on tiller number
was highly significant (P<0.001)
(Appendix 3).
There
was a significant difference (P<0.001) in the number of tiller produced
per variety
at 43 days, 57 days and 113 days after sowing
for both seasons
(Appendix 3
)
.
The
mean maximum above ground
Striga
numbers (NSmax) per variety correlated negatively with
tiller number per variety for both 2014 and 2015 (Spearman correlation coefficients r
2014
=
–
0.265 and r
2015
=
–
0.077).
In 2014,
there was a significant difference (P< 0.001) in number of tillers
produced per
variety
(Appendix 8
). Variety WAB 928
had the highest number of tillers per plant while
Variety
MG12 produced
no tiller 43 days after sowing. Additionally,
varieties WAB928 and CG14
produced the significantly (P<0.001) the highest number of ti
llers 57 days after sowing
compare
d to other varieties except UPR
whic
h was not different from CG14
(Table 3). In the
same period
(57 days)
, WAB56
–
104 produced the lowest number of tillers.
At 113 days after
sowing, variety CG14 and IAC 165 had the highest
and lowest number of tillers respectively
(Table 3).
Generally
there was a reduction in the
number of tillers produced from 57 days to
113 days after sowing except for varieties AGEE, Makassa, MG12, NERICA
–
1, NERICA
–
10,
NERICA
–
2, SCRID090, WAB880 and WAB5
6
–
104
with
no reduction in tiller number (Table
3).
Varieties WAB928,
WAB 935
, Narc3 and UPR
were
highly
effected with
a reduction of 16
and 7 tillers respectively from 57 days to 113 days after sowing (Table 3).
Similarly i
n 2015,
varieties differed
significantly (P< 0.001 in the number
of tillers produced
(Appendix 8
).
Varieties
CG14
and Superica
had the
highest
and lowest
number of
tillers
at 43
days,
57 days
after sowing
(Table 3
)
.
On the other hand at 113 days after sowing Narc 3 had
the highest n
umber of tillers (27
tillers
) while local check (Superica) had
the lowest number of
tillers (2 tillers) (Table 3).
Additionally
there was
a high reduction in the number of tillers
39
produced per variety
except for varieties AGEE, Blechai, CG14, Narc 3, NERIC
A
–
10,
NERICA
–
2, NERICA
–
4 and WAB880 with no reduction from 57 days to 113 days after sowing
(Table 3). Varieties IR49, WAB928, IAC 165, WAB935 and Superica (local variety) had the
highest reduction in
number of tillers compa
red to other varieties
from 57 d
ays and 113 days
after sowing (Table 3
)
.
Tiller number
tillers
varied highly
(P<0.001) across varieties
for 43, 57 and 113 days after
sowing
(Appendix 3
).
There was a reduction in tiller number produced per variety. This
difference is more likely related
to stress levels
(
Striga
)
and timing this stress sets in. Normally
the plant will adjust the number of tillers it produces with available resources. Studies have
indicated that stress occurring in the initial stages will not tremendously affect yield as stress
in following stages (
Boonjung and Fukai, 1996). Stress occurring at the vegetative stage is
expressed in decreased number of tillers (Boonjung and Fukai, 1996). However if the
Striga
attaches on the host later after tiller initiation, it is more likely that because of nutrient
diversion
by the parasite (Parasitism), the plant will not be able to produce panicles on all the tillers.
Additionally with
respect to the tiller production, it has been shown that strigolactones have
an effect on tiller production as well as on
Striga
infection (Jamil
et al.,
2012). Strigolactones
inhibit tiller/shoot branching (Umehara
et al.,
2010) and also triggers
Striga
germination (Jamil
et
al.,
2011). This will therefore mean that varieties that produce high amounts of strigolactones
are more li
kely to produce
fewer
tillers. Also such varieties will be susceptible to
Striga
.
This
same reason would explain the lower number of tillers produced by varieties IAC165, Superica,
WAB56
–
50 and WAB56
–
104 since they were the most susceptible to
Striga
as pe
r this study.
Additionally, WAB56
–
50 and IAC 165 were ranked by Jamil et al. (2011) and Jamil et al.
(2012) as the highest strigolactone producers. Studies by Jamil et al. (2011) ranked variety
WAB56
–
104 as the lowe
st strigolactone producer. This would ind
icate differences in virulence
levels in the Striga ecotype in this study and Jamil et al, (2011) study.
The study has shown
40
that
O.glaberrima
varieties (ACC, CG14, Makassa, Anakila, AGEE and MG12) produced
higher number of tillers per plant
in both season
s
113 days after sowing (Table 3)
. Varieties
Agee, Anakila and CG14 have also been reported to produce a high number of tillers (Jamil
et
al.,
2011; Jamil
et al.,
2012). This high number of tillers among these varieties could be partly
related to low strig
olactone production. This is consistence with studies by Jamil et al. 2012
who also showed varieties AGEE, Anakila and CG14 as low strigolactone producers.
However
these varieties also had reduction in tiller number from 57 days to 113 days after sowing.
This
could be due to tiller death as a result of increased competition among tillers produced.
The differences in tiller production among rice varieties cannot entirely be accounted by
Striga
infestation due to increased strigolactone production. Genetic differences across varieties can
also explain this occurrence. Similarly Fageria et al. (1997) acknowledged tillering
characteristics to be related to genetic characteristics of the variety. F
or example
O.glabberima
varieties tend to produce more tillers than
O.sativa
varieties (Jamil
et al.,
2012). This would
explain the low tiller numbers on NERICA varieties, and
O.sativa
varieties SCRID090,
WAB880, WAB181
–
18, IRGC and IR49 compared with
O.gl
abberima
varieties with similar
or high resistance levels. The high tiller production in
O.sativa
varieties WAB928 and
WAB935 could be related to low strigolactones since these varieties were highly resistant to
Striga
or simply high tillering potential.
S
imilarly the high reduction in tiller number among
these varieties is related to tiller death as a result of excessive competition among the many
tillers produced.
This study also observed a negative correlation between NSmax and the
number of tillers in 2
014 and 2015. This would indicate a negative effect of
Striga
on tiller
production. However this is across varieties with some varieties more affected than others.
This
also suggests
that lower tillering varieties are higher strigolactone producers and are
more
susceptible to
Striga
.
41
Table 3
:
Number of tillers of selected rice varieties grown under
Striga
infested field
conditions for the first and second seasons.
Number of
tillers
2014
2015
Days after sowing
Days after sowing
Rice variety
43
57
113
43
57
113
ACC
10.5
0
21.2
0
18
.00
5.56
20
.00
16.09
Agee
5.46
19.6
0
22
.00
4.42
18
.00
18.79
Anakila
13.4
0
24.1
0
23
.00
5.5
0
19
.00
17.28
Blechai
3.89
11.2
0
9.8
0
2.53
6
.00
7.93
CG14
11.4
0
31
.00
25
.00
7.93
25
.00
26.68
IAC 165
4.87
9.75
8.7
0
2.16
8
.00
2.32
IR49
8.73
18.3
0
14
.00
4.54
15
.00
9.76
IRGC
7.3
0
19
.00
16
.00
5.29
14
.00
13.74
Makassa
6.75
16
.00
16
.00
4.54
18
.00
16.3
0
MG12
0
.00
11.9
0
19
.00
4.97
20
.00
19.06
NARC3
9.23
24.7
0
18
.00
6.29
24
.00
27.1
0
NERICA
–
1
6.28
10.4
0
11
.00
2.46
8
.00
7.33
NERICA
–
10
7.48
12.4
0
13
.00
2.64
7
.00
9.69
NERICA
–
17
5.94
12.1
0
11
.00
4.11
12
.00
11.29
NERICA
–
2
5.22
9.6
0
12
.00
2.31
8
.00
9.94
NERICA
–
4
5.14
11.6
0
11
.00
2.95
7
.00
8.83
SCRID090
4.43
10.3
0
12
.00
2.72
8
.00
7.89
Superica
6.82
11.4
0
10
.00
2
.00
5
.00
1.67
UPR
12.8
0
27.3
0
20
.00
5.58
18
.00
16.54
WAB181
–
18
6.58
13
.00
12
.00
3.16
9
.00
8.64
WAB56
–
50
6.43
10.3
0
11
.00
2.36
7
.00
6.69
WAB56
–
104
4.78
8.69
11
.00
3.65
8
.00
6.46
WAB880
5.83
10.9
0
12
.00
2.87
9
.00
9.05
WAB928
14.6
0
35.7
0
20
.00
6.84
23
.00
16.26
WAB935
12.2
0
24.5
0
18
.00
5.4
0
19
.00
14.35
Means
7.75
16.6
0
15
.00
4.11
13
.00
12.39
LSD (0.05)
3.45
5.72
3.3
0
1.32
4
.00
3.81
S.E
2.74
4.55
2.7
0
1.04
3
.00
3.01
CV%
35.3
27.5
18
.0
25.4
24
.0
24.39
42
4.1.3
Grain yield components of rice varieties grown under
Striga
infested conditions.
4.1.3.1
Rice biomass
For both seasons, variety x season interaction effects on rice biomass was highl
y significant
(P<0.001) (Appendix 4)
. Rice biomass was signific
antly different (P<0.001) across the
different varieties of rice
fo
r 2014 and 2015 seasons (Appendix 4
). In 2014, varieties
ACC
(246.10g) Makassa (238.50g) and
WAB928
(230.70g)
had
si
gnificantly (P< 0.001)
the highest
rice biomass compar
ed to other varieties (Table 4
). Addition
ally UPR (110.80g)
had the lowest
rice bio
mass in the same season. However for 2015
season, varieties IRGC (293.90g),
WAB935 (251.40g) and ACC (243.40g)
had the highest rice biomass while IAC
165
and local
check (Superica) had
significantly (P< 0.001)
the lowest rice b
iomass per rice plant
compa
red
to the remaining varieties
(Table 5
).
Based on the rice biomass produced by each variety, varieties can be categorized into four
groups: (1) high rice biomass producers, i.e.
WAB935, WAB928, ACC and
Makassa
in 2014,
ACC,
WAB935
, MG12, IR49, Makassa, W
AB928
and IRGC in 2015,
(2)
intermediate rice
biomass producers, i.e. WAB880, CG14, Blechai, IR49, IRGC
, SCRID090, WAB 56
–
50,
NERICA
–
1, AGEE, Anakila
and NERICA
–
4 (2014) and WAB880, NERICA
–
17, NERICA
–
2,
NERICA
–
4 and Narc3
(2015),
(3) intermediate low rice
bi
omass producers, i.e.
NERICA
–
2,
NERICA
–
17, NERICA
–
1
0, NARC3, WAB 181
–
18 and WAB 56
–
104
(2014
)
and
WAB56
–
50
,
SCRID090, NERICA
–
10, NERICA
–
1, AGEE, Anakila, WAB181
–
18 and WAB56
–
104
(2015)
and
(4) the low rice b
iomass producers, i.e. local check (Superica),
UPR and
IAC
165
(
both
2014
and 2015) (Table 4, 5)
Results from the current study indicate differences in rice biomass production across varieties.
Varieties such as IAC 165 and Superica with lower rice biomass were also the most susceptible
varieties. It is therefore possible that parasitism lowered their
biomass. Additionally studies by
43
Johnson et al. (1997) have reported
Striga
–
inflicted biomass reduction in IAC 165.
However
these same varieties had high biomass in 2014. This could be related to lower
Striga
number
in this season compared to 2015.
Also
the high rice biomass among the
O.glabberima
,
NERICA varieties and
O.sativa
varieties IRGC, IR49, WAB928, WAB880, SCRID090 and
Blechai is partly related to lower parasitic load on these varieties. It has also been proposed
that a grass plant is a collectio
n of tillers (Nelson
et al.,
1992). This could imply that those
varieties with high tillering capacity could also have high rice biomass. This would also explain
the difference in rice biomass among
O.glabberima
varieties, NERICA varieties and
O.sativa
var
ieties because former produce a lot of tillers. Another observation from the study was some
varieties such as NARC3 (ITA 257), WAB56
–
50,
O.glabberima
va
rieties ACC, Makassa,
MG12 and v
arieties NERICA
–
1 and UPR categorized as susceptible and intermediate
re
spectively had a high rice biomass. Similarly Johnson
et al.,
(1997) also reported a high rice
biomass in
O.glaberrima
varieties Makassa, ACC with relatively high
levels of
Striga
hermonthica
infection. Varieties combining high
Striga
numbers with high hos
t biomass are
deemed tolerant. Tolerance is the ability of a variety to withstand
Striga
infection with
minimum yield losses (Rodenburg and Bastiaans, 2011). It is also true that highly resistant
varieties can also have parasites developing on them. Therefore a combination of tolerance and
resistance is a very important strategy to improve cr
op yields (Rodenburg and Bastiaans, 2011).
Tolerance has also been reported in sorghum (Gurney
et al.,
1995; Van Ast
et al.,
2000;
Rodenburg
et al.,
2006). However tolerance cannot be easily assessed in the field as it would
require infected and uninfected
control plants which was not been provided in this study (see
Chapter five).
44
4.1.3.2
Panicle dry weight
V
ariety x season interaction effects
on panicle dry weight
was highly
significant (P<
0.001)
(Appendix
4
).
Panicle dry weight
was significantly diff
ere
nt (P<
0.001) across rice
varieties
fo
r 2
014 an
d 2015 seasons (Appendix 4
).
In 2014, there was a significant (P< 0.001) difference in Panicle dr
y weight per variety
(Appendix 9
). Variety NERICA
–
10 (685g) had the highest panicle dry weight
while v
arieti
es
WAB935, WAB928 and NARC
3 had significantly (P<
0.001) the lowest panicle dry weight
compared to other varieties with high panicle dry weight
(Table 4)
Similarly varieties varied significantly (P<
0.001) in panicl
e dry weight in 2015 (Appendix 11
).
Varieties NERICA
–
2 and WAB880 had the highest panicle dry
weight i.e
. 732.50g and 725.50g
respectively (Table 5). Additionally, varieties IAC165 (49.60g)
,
NARC3 (67.30g) and
Superica (69.30g) had significantly (P< 0.001) the lowest panicle dry weight compa
red to other
varieties (Table 5). However there was no significant difference (P>0.001) in panicle dry
weight produced by variety Superica and WAB928 (Table 5).
Results from the study indicate a reduction in panicle dry weight for varieties Superica and
I
AC 165 especially in 2015. These same varieties
had the highest
Striga
numbers per variety
in this same season
. Therefore since these varieties are very susceptible to
Striga hermonthica
it is possible that their panicle dry weight was decreased.
However t
hese varieties had their
panicle
dry
weight not significantly reduced in 2014. This could due a lower
Striga
number in
this season and hence less parasitism on these varieties.
Also other varieties such as WAB928
and WAB935 despite being highly resistant t
o
Striga
had lower panicle dry weight. This could
be related to lower yielding potential of these varieties or simply limited adaptability to local
conditions.
The high panicle dr
y weight of NERICA
–
varieties (NERICA
–
2, NERICA
–
10,
NERICA
–
17 and NERICA
–
4),
Oryza sativa
varieties WAB880, SCRID090 and
Blechai,
Oryza
45
glaberrima
varietie
s AGEE, CG14, Anakila and NERICA parent WAB181
–
18 is related to
lower
Striga
number of these varieties or simply due to high yielding potential.
These same
varieties have been re
ported being high yielding (Rodenburg
et al
.,
2015
; Saito
et al.,
2012).
NERICA
–
parents WAB56
–
50 and WAB56
–
104 in spite of high
Striga
numbers produced high
panicle dry weight. This indicates tolerance of these varieties to
Striga hermonthica
infection
compared to the local variety Superica.
Similarly, Rodenburg et al. (2015) reported a high
yielding potential of varieties WAB56
–
104 and WAB56
–
50 under
Striga
infested conditions in
Kenya.
The panicle dry weight cannot be entirely attributed to
Striga
inf
estation. It could also be related
to differences in rainfall patterns in 2014 and 2015
(Appendix 18)
. For e
xample varieties
WAB935, WAB928, IRGC and IR49
produced a high panicle dry weight in 2015 and lower
one in 2014.
In 2014, these varieties were affec
ted by dry spell that s
et in at anthesis
. Therefore
it is possible that some of their grains were not fully filled leading to lower panicle dry weights.
4.1.3.3
Harvest index
For both seasons, variety x season int
eraction effects on harvest index
was high
ly s
ignificant
(P<0.001) (Appendix 4
). Harvest index per variety
was significantly different (P
<0.001) across
varieties
in 2
014 and 2015 seasons (Appendix 4
).
In 2014, there was a signif
icant difference (P< 0.001) in h
arvest
index
a
cross rice varieties
(Appendix 9
). Variety NERICA
–
10 (0.55
) had significantly (P< 0.001) the highest harvest
index compared
to the rest of the
varieties
(Table 4). However this same variety was not
significantly (P> 0.001) in harves
t index from varieties WAB181
–
18
, SCRID090,
N
ERICA
–
4,
NERICA
–
1, AGEE, WAB56
–
50,
NERICA
–
2 and WAB56
–
104 (Table 4).
Additio
nally
varieties NARC 3, WAB928 and WAB935 had significantly (P< 0.001) the lowest harvest
46
index compared to other varieties though these were not significantly different (P> 0.001)
form
IR49
, CG14
and IRGC (Table 4)
In 2015, varieties varied significantly (P< 0.001) in harvest index
across varieties
(Appendix
11
).
Variety WAB18
–
18 (0.57)
had significantly (P< 0.001) the highest harvest index compared
to the remaining varieties. Howe
ver, this variety’s harvest index was not significantly (P>
0.001)
different from variety NERICA
–
10, WAB800, NERICA
–
2, SCRID090, NERICA
–
4,
NERICA
–
17, NERICA
–
1, AGEE, Blechai and CG14
(Table 5). On the other hand, variety
NARC 3 had sign
ificantly (P<0.001)
the lowest
harvest index compar
ed to the rest of the
varieties
(Table 5).
Harvest index is very fundamental in portioning of assimilates
between the shoot,
grains
and
consequently grain yield (Ishii, 1995)
.
Therefore improving on harvest index can greatly
increase the yield of varieties. Grain
harvest index
varies greatly across locations, varieties,
seasons and ecosystems ranging from 0.35
–
0.62 (Kiniry
et al.,
2001)
. Therefore any varietie
s
with a harvest index below 0.3
would indicated limited assimilate portioning to the grains and
vice versa.
The lower harvest index of varieties IAC 165, NARC3 and Superica especially in
2015
compared in 2014
is related
lower grain yield as a result of
Striga
parasitism.
Since in
this s
ame season, the
Striga
numbers were higher than in 2014.
Striga hermonthica
also
depends on the host nutrients before its emergency from the soil. Since there is a relationship
between nutrient uptake and assimilate portioning, any alteration in this proce
ss can alter
assimilate portioning and hence lowering the harvest index.
Additionally, for varieties
WAB928, WAB935, IRGC and IR49 in spite of being
Striga
resistant,
they registered lower
harvest index. These varieties are late maturing, during t
he grain
filling phase
/post
–
anthesis
,
they were
affected by a dry spell
especially in 2014
which significantly affected dry matter
portioning to the grains. The higher harvest in
dex among varieties
SCRID090, WAB181
–
18,
NERICA
–
2
, NERICA
–
4, WAB880
and NERICA
–
10 is
du
e to increased grain yield. This is also
47
an indication of increased portioning of assimilates to the grains. It is therefore not surprising
that these
va
rieties produced high yields as they also had
lower
Striga
number
s
in the two
seasons.
It is also possi
ble that rainfall difference
among the two
seasons is
responsible for lower harvest
index among the varieties WAB928, WAB935, IRGC and
IR49. This so because in 2014 these
varieties had a lower harvest index as compared in 2015. The prolonged dry spell
in 2
014
during the post
–
anthesis stage affected the harvest index in these varieties.
It is also possible
that these varieties are simply low yielding despite the high rice biomass produced, therefore
lowering harvest index.
4.1.3.4
Grain recovery
For both
seasons, variety x season interaction ef
fects rice grain recovery efficiency was
signif
icant (P<0.05) (Appendix 4
). Grain recovery efficiency
was significantly different
(P<0.001) among different varieties of rice fo
r 2
014 and 2015 seasons (Appendix 4
).
I
n 2014, varieties varied significant
ly in grain recovery (Appendix 9
). Variety
AGEE and
NARC 3 had the highest and lowest rice grain recovery efficiency respectively compared to
the remaining varieties (Table 4).
Similarly in 2015, there was a significant
difference (P<
0.001) in rice grain recovery efficie
ncy across varieties (Appendix 11
).
Variety SCRID090 had
the highest rice grain recovery efficiency while variety NARC3 had significantly (P< 0.001)
the lowest rice grain recovery efficiency compared to
the rest of the varieties (Table 5).
Results
from the current study indicate
difference in
recovery efficiency across rice varieties.
Varieties NARC3 with the lowest recovery efficiency
in both seasons
indicated this variety had
a lot of empty grains
and c
onsequently lower yield
. This could be related lower portioning of
assimilates to the developing grains resulting from increased parasitism of
Striga
on this
variety. It should also be noted that even the most resistant varieties had lower recovery
48
efficiency simply because of lower yielding potential of these varieties.
Varieties WAB928,
WAB935, IRGC and IR49 with lower grain recovery can only imply a high number of unfilled
grains. It is possible since these varieties produce a lot of tillers, assi
milates are partitioned to
the most active sinks (grains) leaving others unfilled. Also these varieties are late maturing,
only carrying out grain filling when the rains are at the end or about to end. Therefore due to
limited moisture it can greatly affec
t partitioning of assimilates to the grains. Other varieties
with a high recovery efficiency only indicate increased partitioning of dry matter to the grains
despite the high
Striga
numbers among such varieties like IAC165, WAB56
–
50, WAB56
–
104
and Superica
.
4.1.3.5
Rice grain yield
For both seasons, variety x season interaction effect was highly significant (P<0.001)
(Appendix 4
)
Rice grain dry weight varied significantly (
P
<0.001) among rice var
ieties for
both
seasons (Appendix 4
)
.
Generally t
here was also
an increase in yield from 2014 to 2015
across
varieties
except
for
varieties IAC165, Superica, WAB56
–
50, WAB56
–
104, Anakila and
NERICA
–
10 (Table 4, 5). Varieties WAB928 and WAB935 had the highest increase in yield
from 2014 to 2015
(Table 4, 5)
.
Varieties
Superica and IAC 165 had the highest
decrease in
yield compared to
other varieties with a decrease in yield (Table 4, 5).
In 2014
, there was a significant difference (P< 0.001) in yield produ
ced across varieties
(Appendix 11
). Variety
SCRID090
(3.7 t ha
–
1
)
produced
significantly (P< 0.001)
the highest
yield
compared to the remaining varieties except varieties NERICA
–
10, WAB181
–
18,
NERICA
–
4
, WAB880, NERICA
–
1,WAB56
–
104
and NERICA
–
2 whose yield was not
significantly (P > 0.001) different form SCRID090 (Table
4). Varieties
NARC3 and
WAB935
prod
uced
significantly
the lowest yield
as compared to the other varieties. However the yield
of these varieties was not significantly different (P> 0.001) form WAB928
, IRGC and IR49
49
(Table 4).
In the same season, y
ield per
variety ranged from 0.04
t ha
–
1
to 3.7 t ha
–
1
with an
average yield of 2.0 t ha
–
1
(Table 4).
In 2014, the infestation level being low, the correlation
between NSmax and grain yield though negative was very weak and not significant (r
2014
=
–
0.10; P= 0.640).
Similarly in 2015, there was a significant difference (P> 0.001) in yield across varieties. Variety
NERICA
–
2
(
3.8 t ha
–
1
)
produced the highest yield
.
NARC 3 p
roduced
significantly (P< 0.001)
the lowest yield
across varieties
though it was not
different f
orm Superica, IAC165, WAB935
and WAB 928
(Table 5
).
Yield per
variety in 2015 ranged from 0.2
t ha
–
1
to 3.8
t ha
–
1
with an
average yield of 2.4
t ha
–
1
.
In 2015, there was a significant correlation between the above
–
ground emerged
Striga
plants (NSmax) and grain yield (r
2015
=
–
0.465; P=0.02
) with more
resistant varieties showing th
e highest grain yield (Table 5
).
Rice varieties combining excellent resistance to
Striga hermonthica
with high yields and
environmental adaptability would be v
ery useful to rice farmers growing in
Striga
prone areas.
Results of this study showed varying levels of yield among the 25 rice varieties and a negative
correlation between number of emerged
Striga
plants (NSmax) and grain yield in 2015.
Normally it is in
conditions of high
Striga
infestation that such correlation occurs (Rodenburg
et al.,
2005; Rodenburg
et al.,
2015). This would imply that highly resistant varieties provide
a yield advantage under conditions of high
Striga
infestation.
The results also
reported high yield performance among the NERICA’s varieties and one
O.sativa
parent, WAB181
–
18. The latter variety was reported before by Rodenburg et al. (2015)
as high yielding under
Striga
–
infested conditions. The relative high yields among the
NERICA’
s can be attributed to the relative low
Striga hermonthica
infection levels and good
yield potential and environmental adaptability. Previous studies have demonstrated NERICA’s
high yielding ability (Gridley
et al.,
2002; Saito
et al.,
2012). All the
O.glaberrima
varieties,
50
CG14, Makassa, AGEE, ACC and Anakila were categorized as intermediate high to low
yielding. These varieties are not only high tillering (Jamil
et al.,
2012) but also have a greater
competitive ability with weeds (Johnson
et al.,
199
5). This characteristic, combined with lower
Striga
numbers (NSmax) recorded per variety could be the reason for this yield. Studies by
Johnson et al. (1997) also reported
Striga
hermonthica
number among
O.glaberrima
varieties
though the rice biomass was n
ot reduced highly. This could imply a reduced effect of
Striga
on these varieties compared to
O.sativa
varieties IAC 165 and Superica. These varieties can be
regarded as tolerant to
Striga
, a situation where varieties with high parasitic load still produce
high yields. Varieties SCRID090, WA
B880, Blechai produced a higher
yields under high
Striga
pressure in 2015 and low
Striga
pressure in 2014 compared to other varieties. This could
be attributed to relatively low
Striga
infestation.
Results show that NERI
CA
–
1, WAB56
–
50 and WAB56
–
104 which had high
Striga
numbers
produced a high yield under these conditions. This could imply that these varieties are
tolerant
to
Striga hermonthica
producing significantly high yields even under high
Striga
levels.
Tolerance h
as also been reported in sorghum (Gurney
et al.,
1995; Rodenburg
et al.,
2006).
Understanding of this trait can help plant breeders to incorporate it in the most resistant
varieties in order to improve yield (Rodenburg
et al.,
2008; Rodenburg and Bastiaans, 2011).
However to further exploit this phenomenal, one needs to have
Striga
–
free plants of the same
varieties looking at relative yield losses (Rodenburg
et al.,
2006; Rodenburg
et
al.,
2008).
Another reason could be that t
hey have a high yielding potential. The lower yields of local
check (Superica) and IAC165
especially in 2015
can be attributed to high
Striga
infection
levels since these two supported the highest number of parasitic plants.
Additionally these
varieties ha
d high relatively high yields in 2014 probably due reduced
Striga
numbers in this
season.
51
The yielding ability of the varieties cannot be solely attributed to
Striga
infection. For example
inspite of the excellent resistance levels of varieties WAB928, WAB935, IR49 and IRGC, these
produced a lower yield. This could be attributed to limited adaptability to the prevailing
weather conditions or simply because of an inher
ent lower yield potential. These varieties
progress
ed to flowering when the rains were over especially in 2014 and therefore were
affected by too much sunlight. This significantly affected their yield potential. Also the
differences in yield ranking across
the varieties in 2014 and 2015could be partly attributed to
rainfall levels in these years
(Appendix 15)
. For example, varieties Makassa, AGEE, ACC,
Blechai
, WAB935, WAB928 and CG14 that produced
low
er yields
2015 produced high yields
in 2015.
52
Table
4
:
Yield components of rice varieties under
Striga hermonthica
infected conditions
for 2014 in Namutumba Eastern Uganda.
*MG12
variety is not included because the land was cultivated by farmers before harvesting it.
Rice variety
Rice biomass
(g)
Panicle dry
weight
(
g)
Harvest index
Grain
recovery (%)
Grain yield
(t
/ha)
ACC
246.1
0
433.60
0.29
82.94
1.82
Agee
154.8
0
541.00
0.44
99.83
2.57
Anakila
152.6
0
554.20
0.38
86.74
2.46
Blechai
176.5
0
459.10
0.35
83.78
1
.9
8
CG14
184.1
0
362.50
0.22
74.74
1.37
IAC
111.7
0
377.00
0.30
89.65
1.54
IR49
181.1
0
223.9
0
0.17
61.97
0.78
IRGC
177
.00
237.1
0
0.20
74.1
0
0.89
Makassa
238.5
0
493.7
0
0.26
78.11
2.01
NARC3
132.8
0
89
.00
0.08
45.98
0.06
NERICA
–
1
157
.00
583
.00
0.47
92.03
2.88
NERICA
–
10
137.9
0
685
.00
0.55
91.72
3.39
NERICA
–
17
144.3
0
500.4
0
0.39
90.84
2.33
NERICA
–
2
134
.00
628.4
0
0.52
91.93
3.09
NERICA
–
4
155
.00
682.6
0
0.50
92.21
3.41
SCRID090
157
.00
721
.00
0.51
95.41
3.71
Superica
116.5
0
352.7
0
0.34
87.22
1.61
UPR
110.8
0
310.2
0
0.30
80.86
1.11
WAB181
–
18
124.4
0
680.9
0
0.53
91.96
3.33
WAB56
–
50
150.1
0
535.3
0
0.46
89.66
2.59
WAB56
–
104
122.8
0
575.8
0
0.50
94.61
2.91
WAB880
170.1
0
584.9
0
0.43
90.14
2.82
WAB928
230.7
0
85.8
0
0.13
66.28
0.20
WAB935
214
.00
54.6
0
0.05
55.2
0
0.04
LSD (0.05)
43.33
169.40
0
.12
12.56
0.94
S.E
34.33
124.40
0.77
8.76
0.67
CV%
21.22
26.45
34.26
10.30
30.86
53
Table
5
: Yield components of rice varieties under
Striga hermonthica
infected conditions
for 2015 in Namutumba Eastern Uganda.
Rice variety
Rice
biomass
(g)
Panicle dry
weight g)
Harvest
index
Grain
recovery (%)
Grain yield
(
t
/ha
)
ACC
243.4
0
610.2
0
0.41
88.89
2.93
Agee
139.2
0
610.9
0
0.48
94.99
3.15
Anakila
108
.00
519.1
0
0.43
87.42
2.34
Blechai
211.8
0
580.4
0
0.47
96.44
3
.10
CG14
204.8
0
694.7
0
0.47
93.57
3.
46
IAC 165
44
.00
49.6
0
0.22
88.62
0.
25
IR49
210.9
0
455.8
0
0.37
85.05
2.24
IRGC
293.9
0
358
.00
0.23
78.97
1.
51
Makassa
225.2
0
572.8
0
0.40
84.98
2.72
MG12
231.1
0
414.3
0
0.24
72.95
1.
78
NARC3
155.6
0
67.3
0
0.04
53.06
0.
20
NERICA
–
1
126.6
0
543.2
0
0.50
93.28
2
.89
NERICA
–
10
131.5
0
585.3
0
0.55
89.79
3.
01
NERICA
–
17
160.7
0
649.2
0
0.49
90.64
3.36
NERICA
–
2
172.5
0
732.5
0
0.55
92.99
3.81
NERICA
–
4
150.6
0
715.9
0
0.52
89.45
3.
72
SCRID090
140.4
0
685.2
0
0.51
98.35
3.
75
Superica
37.4
0
69.3
0
0.21
83.37
0.
35
UPR
99.5
0
317.9
0
0.36
85.02
1.
61
WAB181
–
18
115.5
0
716.1
0
0.57
89.1
0
3.
66
WAB56
–
50
133
.00
417
.00
0.44
90.41
2.
14
WAB56
–
104
103.9
0
361.6
0
0.44
92.32
1.8
4
WAB880
187.4
0
725.5
0
0.50
93.1
0
3.7
0
WAB928
223.3
0
266.1
0
0.18
66.01
1.10
WAB935
251.4
0
335.6
0
0.16
66.43
1.22
LSD (0.05)
49.48
198.2
0.11
11.66
1.
08
S.E
39.15
156.8
0.084
9.22
0.
85
CV%
23.93
32.62
21.68
10.76
3
5.81
54
4.1.4
Striga hermonthica
and its effect on growth of rice
varieties grown under field
conditions.
4.1.4.1
Striga
counts
per variety
For both seasons, variety x season interaction effects on
Striga
number
was high
ly significant
(P<0.001) at 43 days,
57 days after sowing (Appendix 5
).
However, 71 days after sowing,
variety x season interaction was significant (P< 0.01) and
at harvest (P
<0.05) (Appendix 5
).
There was no significant interaction eff
ect between season and variety at 85 days after sowing
although the seasons were highly significant (P< 0.001)
(Appendix 5
).
Striga
counts were
highly significant
different (P<0.001) among diffe
rent varieties of rice fo
r 2
014 and 2015
seasons (Appendix 5
). In 2014 at 43, 57, 71, 85
days after sowing
and at rice harvest
,
v
ariety
NERICA
–
10 had the lowest
Striga
numbers while
Superica and IAC 165 had the highest
Striga
numbers
compared to the rest of the
varieties (
Table 6
).
In 2015, varieties differed significantly (P< 0.001) in the number of
Striga
plants support
ed
per
plant at 43, 57, 71, 85 days after sowing
and at rice harvest (Appendix 12
).
In the same season
at 43 days
after sowing, NERICA
–
10 and WAB935 had the lowest
Striga
plants while Superica
(local variety), IAC 165 and WAB56
–
104 had the highest
Striga
plants (Table 6). At 57 days
after sowing, NERICA
–
2, NERICA
–
10 and WAB928 had the lowest
Striga
plants while
Superi
ca, IAC 165 and WAB56
–
50 had the highest
Striga
plants (Table 6). At 85 days after
sowing and rice harvest, NERICA
–
2 had the lowest
Striga
plants with Superica, IAC 165
having the highest
Striga
plants (Table 6).
The
results indicate differences in suscept
ibility and resistance levels among rice varieties.
Varieties I
AC 165, Superica
, WAB56
–
50
with high
Striga
numbers
at
43 days
in both seasons
is an indication of earlier
Striga
infestation on these varieties
.
Normally an earlier infestation
of
Striga
can tremendous affect the growth of the plants
(Frost
et al.,
1997)
. Also these varieties
55
are deemed very susceptible to
Striga hermonthica
since there is an increase in
Striga
numbers
from 43 days up to harvest time.
These except Superica were reported t
o be highly susceptible
to
Striga
under field conditions (Johnson
et al.,
1997; Rodenburg
et al.,
2015).
The lower
Striga
numbers among varieties NERICA
–
10, NERICA
–
2, WAB928, WAB880, SCRID090,
NERICA
–
4, NERICA
–
17 and IRGC
in both seasons
is a clear indicat
ion of the resistance of
these varieties to
Striga
. The fact that the lower number of
Striga
plants attached at 43, 57 and
71 days could indicate lower number of strigolactones produced by these varieties. Other
varieties with an increase in
Striga
plants
from 43 days to harvest were NARC3, WAB56
–
104
and NERICA
–
1 all of which categorized as being susceptible to
Striga
.
It was also observed
that there was a greater increase in
Striga
counts at 43 days, 57days, 71 days, 85
days and at
rice harvest in 2015 compared in 2014. This could be as result of an increased
Striga
seed bank
since more
Striga
was added in the same season.
56
Table 6
:
Number of
Striga
plants (
Striga
counts) per rice variety for 2014
–
2015 under field
conditions in Eastern Uganda.
Number of above ground
Striga
plants were log transformed before analysis
Number of
Striga
plants
2014 2015
Days after sowing
Days after sowing
Rice variety
43
57
71
85
Harvest
43
57
71
85
Harvest
ACC
0.33
0.36
0.76
1.10
0.86
0.30
0.66
0.66
1.07
1.37
Agee
0.14
0.26
0.55
0.97
0.90
0.05
0.23
0.23
0.77
1.00
Anakila
0.30
0.38
0.45
0.61
0.67
0.45
0.82
0.82
1.12
1.3
0
Blechai
0.10
0.14
0.43
0.77
0.82
0.42
0.65
0.65
0.91
0.94
CG14
0.18
0.38
0.47
0.72
0.53
0.15
0.77
0.77
1.11
1.23
IAC
0.83
1.25
1.95
2.15
2.37
1.52
2.10
2.10
2.08
2.22
IR49
0.34
0.35
0.63
0.75
0.45
0.68
0.74
0.75
0.95
1.10
IRGC
0.05
0.21
0.30
0.37
0.46
0.26
0.33
0.33
0.67
0.68
Makassa
0.30
0.29
0.65
1.00
0.88
0.62
1.10
1.10
1.36
1.53
MG12
0.09
0.07
0.53
1.03
1.12
0.08
0.76
0.76
1.25
1.43
NARC3
0.38
0.74
1.01
1.30
1.30
0.68
1.41
1.41
1.63
1.68
NERICA
–
1
0.32
0.54
1.10
1.42
1.45
0.58
0.88
0.88
1.10
1.26
NERICA
–
10
0.02
0.02
0.09
0.15
0.17
0.01
0.10
0.10
0.18
0.37
NERICA
–
17
0.08
0.16
0.36
0.65
0.51
0.53
0.65
0.63
0.73
0.98
NERICA
–
2
0.04
0.15
0.18
0.25
0.27
0.05
0.10
0.11
0.14
0.15
NERICA
–
4
0.08
0.19
0.39
0.59
0.52
0.25
0.52
0.52
0.70
0.82
SCRID090
0.22
0.26
0.41
0.48
0.52
0.18
0.52
0.52
0.62
0.69
Superica
0.99
1.31
1.98
2.1
0
2.07
1.52
2.06
2.06
2.18
2.32
UPR
0.20
0.57
0.62
0.90
1.03
0.36
0.70
0.70
1.08
1.15
WAB181
–
18
0.21
0.22
0.97
1.13
1.12
0.47
0.46
0.46
0.57
0.74
WAB56
–
50
0.74
1.16
1.58
1.84
1.93
1.15
1.85
1.85
1.99
2.00
WAB56
–
104
0.41
0.90
1.38
1.58
1.74
1.26
1.63
1.63
1.71
1.81
WAB880
0.12
0.19
0.63
0.69
0.74
0.38
0.50
0.50
0.67
0.82
WAB928
0.01
0.05
0.11
0.35
0.17
0.08
0.11
0.11
0.22
0.27
WAB935
0.20
0.19
0.57
0.61
0.47
0.02
0.42
0.42
0.62
0.65
Means
6.68
10.34
18.10
23.51
23.07
12.05
20.07
20.07
25.43
28.51
LSD (0.05)
0.28
0.31
0.35
0.34
0.39
0.34
0.45
0.45
0.40
0.41
S.E
0.22
0.25
0.28
0.27
0.31
0.27
0.35
0.36
0.32
0.32
CV%
82.92
60.09
38.03
28.74
33.21
56.7
43.76
35.12
27.94
29.64
57
4.1.4.2
Days to
Striga
emergency and flowering
Variety x season interaction effect was highly significant (P< 0.001) for days to
Striga
emergency and days to
Striga
flowering
(Appendix 7
)
Similarly, rice varieties v
aried
significantly (P< 0.001) in the days
to
Striga
flowering
and days to
Striga
emergency
for
both
season
s (Appendix
7
)
.
In 2014, there was a significant (P< 0.001) difference in days to
Striga
emergency and
flower
ing across varieties (Appendix 12
). Variety WAB56
–
104 had the lowest days to
Striga
emergency (35 days) while variety
NE
RICA
–
2 had
Striga
plants emerging latter than the rest
of the varieties at
96 days (Table 6). Additionally, variety Superica had
Striga
plants flowering
earlier at 64 days while variety MG12 flowering
later
at 105 days (Table 6)
Also the most
resistant suc
h as NERICA
–
2, NERICA
–
10 and WAB928 had
Striga
emerging later
and
no
flowering of
Striga
plants
(Table 6
).
Similarly, in 2015, there was a significant (P<0.001) difference in days to
Striga
emergency
and flower
ing across varieties (Appendix 13
). Variety IAC 165 had the lowest days to
Striga
emergency and flowering i.e. 49 days and 79 days respectively (Table 6). Additionally, variety
NERICA
–
10 had
Striga
plants emerging later at 96 days and flowering later at 109 days
(Table
6)
. Additionally,
NERICA
–
2
did not have any
Striga
plants flowering in the all season.
Rice varieties varied significantly in days to first
Striga
emergency and days to
Striga
flowering
(Tabl
e 6
). The average time taken by a
Striga
plant to complete its cycle from emergenc
e to
flowering vari
ed greatly among seasons from 68 days for 2014 and 90
days for 2015. Studies
by Atera et al. (2012) reported the first
Striga
plant emerging 42 after rice emergence with
minimum of 56 days to complete life cycle. The current study howeve
r reports a shorter time
to first
Striga
emergence (35
days) in 2014, and a longer tim
e to first emergence in 2015 (49
days). The difference in emergence days between Atera et al. (2012) study and this study could
58
be attributed in differences in rainfall,
Striga
seed bank and genotype. Several studies have
reported variation in first
Striga
emerging for example 21 days for sorghum in Sudan (Bebawi,
1981), 54 days for sorghum in Mali (Clark
et al.,
1994), and 35 days for sorghum and maize in
Kenya (Gurney
et
al.,
1995). Low soil moisture at cr
itical stages, caused by low
rainfall levels,
can prevent the germination of
Striga
seed (Ransom and Njorge, 1991). This same reason can
account for the late emerging of
Striga
plants in 2015, the rains appeared about tw
o weeks later
and lasted longer than
in 2014.This partly explains the higher
Striga
numbers in this season
because even after rice physiological maturity, new
Striga
plants were emerging. Generally
even the most susceptible cultivars in 2015 had their firs
t
Striga
plants emerging later than in
2014.
There were also differences in
Striga
flowering across rice varieties with more resistant
varieties i.e. WAB928, WAB935, NERICA
–
2 and NERICA
–
10 having the
Striga
plants
flowering later
or not flowering at all
.
In comparison with the susceptible varieties i.e. Superica,
IAC 165, WAB56
–
104 and WAB56
–
50 with earlier emergence and flowering
Striga
plants.
This can be attributed to differences in susceptible to resistance levels among rice varieties.
Similarly, Gebre
medhin et al. (
2000) reported earlier flowering and emergence
of
Striga
in
susceptible sorghum cultivar compared to the resistant one.
Varieties without flowering of the
Striga
plants can greatly reduce on the
Striga
seed bank. Such can be incorporated wit
h
integrated
Striga
management methods.
4.1.4.3
M
aximum number of above
Striga
plants
For both seasons, the variety x season interaction effect
s were not significant (P> 0.05),
however the seasons were highly significant (P< 0.
001) (Appendix 6
).
Rice
varieties ha
d a
highly significant effect (P
<0.001) on the maximum number of emerged
Striga hermonthica
plants
(NSmax) in both seasons
(Appendix 6
).
In all seasons, mean maximum above
–
ground
59
Striga
numbers (NSmax) per variety correlated positively and high
ly significantly (P<0.001)
with the mean
Striga
dry weights at harvest per variety (Spearman correlation coefficients were
r
2014
=0.89
9
and r
2015
=0.898
).
In 2014, there was a significant difference (P< 0.001) in the maximum number of
Striga
plan
ts
produced
per variety (Appendix 13
).
Varieties Superica and IAC 165 produced significantly
(P< 0.001) the highest maximum
Striga
number though it was not significantly different (P>
0.001) from WAB56
–
50. Also, NERICA
–
10, NERICA
–
2 and WAB928 had the lowest
maximum
number (NSmax) of
Striga
plants.
Based on the maximum above
–
ground
S.
hermonthica
numbers (NSma
x
) observed in the field in 2014;
varieties can be categorized i
nto
five separate groups:
(1) very resistant
, (2) moderately resistant, (3) intermediate, (4)
s
usceptible, (5) very susceptible. Varieties
NERICA
–
2, NERICA
–
10, Blecahai,
WAB928,
WAB935, Anakila, CG14, IR49, IRGC, NERICA
–
4 having
on average
a mean
NSmax
of >
10
per m
2
, were classified as very resist
ant while varieties MG12, UPR, ACC, NERICA
–
17,
WAB
880, SCRID090, WAB181
–
18, AGEE
with NSmax of 10
–
20
per m
2
, were
moderately
resistant (Table 7
). Other varieties suc
h as NARC3 and
N
ERICA
–
1 with NSmax of 20
–
40
per
m
2
were classified as intermediate (between resistant and susceptible).Varieties WAB56
–
50
and WAB56
–
104 with an NSmax of 40
–
100
per m
2
were classified
as susceptible. The last
group
consisted of very susceptib
le varieties with NSmax of 100
–
250
per m
2
i.e. IAC165 and
loca
l check (Superica) (Table 7
).
Similarly in 2015, varieties differed signif
icantly (P< 0.001) in the
maximum
number
(NSmax)
of
S
triga
plants
produced per variety (Appendix 14
).
Varieties IAC 165, Superica and WAB56
–
50 had the highest number of maximum above ground
Striga
plants (Table 7). Additionally,
NERICA
–
2 had the lowest num
ber of above ground
Striga
plants (Table 7). Based on the
maximum above
–
ground
S.hermonthica
numbers (NSmax) observed in the field in 2015;
varieties
WAB928, WAB935, NERICA
–
2, NERICA
–
4, SCRID090, Anakila and NERICA
60
having
an NSmax
>10
per m
2
parasites wer
e considered very resistant. Varieties WAB880,
WAB181
–
18, AGEE, CG14, IRGC, NERICA
–
17, UPR and Blechai with an
NSmax of
10
–
20
per m
2
were considered moderately resistant. Also varieties Makassa, MG12, NERICA
–
1, IR49
and ACC with
NSmax
of 20
–
40
per m
2
were
considered intermediate (between resistant and
susceptible). More so varieties WAB56
–
50, WAB56
–
104 and Narc3 with an
NSmax of
40
–
100
per m
2
were considered susceptible. Lastly varieties IAC165 and Superica (local variety) with
an
NSmax
of 100
–
250
per m
2
considered as very susceptible in the same season.
Results from the current field study indicate a highly significant difference in the number of
emerged
Striga
plants (NSmax) and
Striga
dry weight among the 25 varieties screened in both
seasons. All the
NERICA varieties (NERICA
–
10, NERICA
–
17, NERICA
–
2 and NERICA
–
4)
except NERICA
–
1, seven of
O.sativa
varieties (WAB928, WAB935, IR49, IRGC, Blechai,
WAB880 and SCRID090), one
O.sativa
parent (WAB181
–
18) and three
O.glaberrima
(CG14,
AGEE and Anakila) showed
an excellent resistance to the
S.hermonthica
ecotype from
Namutumba, Uganda. All the NERICA varieties, one
O.glaberrima
variety CG14 and one
O.sativa
parent WAB181
–
18 in this study have also been reported to resistant to
Striga
hermonthica
ecotype from Mbi
ta, in western Kenya approximately 211 km south east of
Namutumba, under field conditions (Rodenburg
et al.,
2015). Similarly rhizotron experiments
by Cissoko et al. (2011) have also ranked these same varieties as having an excellent post
–
attachment resist
ance to
S.hermonthica
ecotype found in Kibos, western Kenya. Similar results
reported by Jamil et al. (2011) in his pre
–
attachment resistance study on an ecotype of
Striga
hermonthica
from Medani (Sudan). This implies that these varieties have excellent re
sistance
levels to various ecotypes of
Striga
. Broad spectrum resistance of rice varieties to
Striga
have
also been reported by Cissoko et al. (2011). Additionally varieties AGEE and Anakila
categorized as resistant as per this study were also reported by
Jamil et al. (2012) as having a
lower number of
Striga
plants emerging on them. Varieties WAB928 and WAB935 with an
61
excellent
Striga
resistance as found in this study have also been reported to have an excellent
resistance to
Striga hermonthica
by Johnson et al. (2000) in Ivory Coast. We also carried out a
rhizotron study
on
a
selection of varieties,
whereby
WAB928 showed an excellent post
attachment resistance to
Striga hermonthica
(see study three
). On the other hand resistance
found with vari
eties SCRID090 and WAB880 have not been reported before.
IAC 165 and Superica were the most susceptible varieties in this study. For IAC 165, it was
used as a susceptible check variety in this study and it exhibited high susceptible levels
confirming stu
dies by Johnson et al. (1997); Gurney
et al.
(2006); Cissoko
et al.
(2011); Jamil
et al.
(2012); Rodenburg
et al.,
(2015) where it was grouped as very susceptible to
Striga
.
Superica, the local check variety, was very susceptible probably due to continuou
s cultivation
of the same variety in the region. Therefore it is likely that the virulence levels of local
Striga
population against this variety increased in time. Similar insights were provided by Rodenburg
et al. (2015) on variety Supa India which was v
ery susceptible in Kyela but resistant at Mbita
where it had not been grown before. The
O.glaberrima
varieties (Makassa, MG12, and ACC)
had an intermediate level of resistance to
S.hermonthica.
Similar results were reported by
Johnson et al. (1997) who als
o showed varieties ACC and Makassa having a partial resistance
to
S.hermonthica
. Studies have shown that
O.glaberrima
have an excellent competitive ability
over weeds (Johnson
et al.,
1995). This could explain the lower
Striga
plants on these varieties.
Al
so Johnson et al. (1997) reported later attachment of
Striga
on
O.glabberima
varieties and
because they are vigorous growers, it is possible that they can out
compete
Striga
. Similarly
Johnson et al. (1997) reported that
O.glabberima
varieties are less affected by
Striga
compared
to
O.sativa
varieties. Other varieties such as Blechai and UPR have been reported in Harahap
et al. (1993) to be
Striga
resistant. This study has however demonstrated partial resistance of
UPR. Varieties NAR
C 3 (ITA 257), WAB56
–
50 and WAB56
–
104 were classified as
susceptible in this study. Studies by Harahap et al. (1993) also classified variety NARC3 (ITA
62
257) as being susceptible to
S.hermonthica
ecotype in Kenya. On the other hand, WAB56
–
50
was ranked by J
amil et al. (2011) as the highest strigolactone producers. Since strigolactones
trigger the germination of
Striga
, it is therefore not surprising that these supported a high
number of
Striga
plants as per this study.
The high
Striga
numbers on NERICA
–
1 cla
ssified as
intermediated in this study. However studies by Cissoko et al. (2011) and Rodenburg et al.
(2015) have classified it as resistant to
Striga
. This could be related to differences in ecotype
virulence of
Striga hermonthica
to NERICA
–
1 in this stud
y and their studies.
4.1.4.4
Striga dry weight
For both seasons, variety x season interaction
effects
on
Striga
dry weight
was significant
(P<0.001
)
(Appendix 6
).
Striga
dry weight varied significantly (P<
0.001) among the rice
varieties for the two seasons
(Appendix 6
)
.
In 2014
,
varieties varied significantly (P< 0.001) in
Striga
dry weight
(Appendix 13)
.
Varieties
IAC
165 (1.89)
local check (Superica)
(1.80
g
) and WAB56
–
50 (1.59
g
)
produced
Significantly
(P
<0.001)
the highest
Striga
dry weight at harvest
compared to remaining varieties though the
Striga
dry weight of WAB56
–
50 was not significantly different from NARC 3 and WAB56
–
104
(Table 7
).
Also variety
NERICA
–
1
0 (0.03
g
) had the lowest
Striga
dry weight a
t harvest
in
the same season (Table 7)
.
Similarly in 2015, there was a significant different (P<0.001) in
Striga
dry weight
produced
per variety (Appendix 14
). Varieties Superica (2.11
g
), IAC 165 (1.98
g
) and WAB56
–
50 (1.90
g
)
had significantly (P< 0.001) th
e highest
Striga
dry weight at harvest compared to the rest of
the
varieties in the study (Table 7
). However
Striga
dry weight from varieties WAB56
–
50 and
IAC 165 was not significantly (P>0.001) different from varieties NARC 3 and WAB56
–
104
(Table
7
).
Variety NERICA
–
2 had significantly (P< 0.001) the lowest
Striga
dry weight
63
compared to the rest of the varieties except varieties WAB928 and NERICA
–
10 with similar
Striga
dry weight
(P> 0.001) as NERICA
–
2 (Table 7
).
Results showed that there was a highly
significant positive correlation in NSmax and
Striga
dry weight at harvest
in both seasons
, confirming Rodenburg et al. (2015). The highly
susceptible varieties such as IAC 165 and Superica did not only have the highest
Striga
numbers but also had the high
est
Striga
dry weight at harvest. This can only be explained by
earlier infection
Striga
as a result of increased strigolactone production. Similarly Van Ast and
Bastiaans. (2006) reported an increase in
Striga
dry weight when sorghum plants were infected
with in
Striga
7 days after sowing compared to 21 days after sowing. However this only holds
true for the most susceptible varieties. Varieties WAB56
–
50, WAB56
–
104 and NARC 3 also
with high numbers of emerged
Striga
plants had a high
Striga
dry weight at h
arvest though not
as high as IAC 165 and Superica. NERICA
–
1 and UPR also had relatively high
Striga
dry
weight compared to the rest of the varieties though not as high as the earlier mentioned
varieties. The rest of the varieties had a lower
Striga
dry weight at harvest. This could be
attributed to only a lower number of
Striga
plants but also to the small sized parasitic plants on
these varieties.
64
Table
7
:
Striga
components per rice variety in 2014 and 2015 Namutumba district Eastern
Uganda
2014
2015
Rice variety
DSE
(days)
DSF
(days)
NSmax
Striga
dry
weight (g)
DSE
(days)
DSF
(days)
NSmax
Striga
dry
weight (g)
ACC
51
71
1.11
0.73
66
96
1.38
0.87
Agee
58
74
1.06
0.66
71
92
1.01
0.62
Anakila
47
71
0.72
0.40
66
94
1.25
0.77
Blechai
78
100
0.88
0.46
59
94
0.95
0.87
CG14
65
80
0.73
0.49
78
97
1.23
0.60
IAC 165
41
68
2.32
1.89
49
79
2.23
1.98
IR49
55
75
0.74
0.46
62
95
1.09
1.05
IRGC
66
80
0.55
0.46
70
97
0.67
0.55
Makassa
59
78
1.09
0.47
70
93
1.54
1.12
MG12
72
105
1.16
0.79
70
91
1.44
0.74
NARC3
52
78
1.36
1.28
55
84
1.71
1.67
NERICA
–
1
51
78
1.49
0.94
62
94
1.26
1.03
NERICA
–
10
85
0
0.28
0.03
94
109
0.36
0.27
NERICA
–
17
79
82
0.71
0.45
68
96
1.00
0.93
NERICA
–
2
96
0
0.31
0.17
86
0
0.24
0.05
NERICA
–
4
54
73
0.62
0.33
67
94
0.83
0.63
SCRID090
45
68
0.55
0.51
70
98
0.70
0.52
UPR
52
74
1.03
1.01
57
82
1.16
1.12
WAB181
–
18
42
78
1.19
0.64
55
87
0.85
0.54
WAB56
–
50
59
72
1.93
1.59
53
101
2.02
1.90
WAB56
–
104
35
73
1.76
1.23
57
88
1.77
1.66
WAB880
69
86
0.81
0.43
63
96
0.84
0.73
WAB928
78
0
0.32
0.09
79
106
0.32
0.31
WAB935
64
72
0.65
0.40
84
98
0.63
0.51
Superica
38
64
2.14
1.80
51
81
2.33
2.11
LSD (0.05)
17.39
13.03
0.34
0.41
8.83
11.39
0.43
0.41
S.E
11.95
7.33
0.27
0.33
6.04
6.06
0.34
0.32
CV%
20.70
9.61
26.53
45.98
9.13
6.62
36.81
27.74
Maximum number of above ground
Striga
plants (NSmax) and
Striga
dry weight were log transformed before
analysis to meet the assumption of Analysis of Variance. DSE (Days to
Striga
emergency and DSF (Days to
Striga
flowering)
65
4. 2
Study two
: F
armer participatory variety selection of upland rice varieties
4.2.1
Social demographic factors and Rice farming practices
A total of 21 (55%) men and 17 (45%) women were involved in the
rice selection exercise
(Table
8
). The largest share of these farmers was in the age
–
group 42
–
54 years (37%) followed
by the age
–
group 31
–
41 (28%), 18
–
30 (24%) and lastly 55 and above (11%) (Table 6). Out of
38 farmers, 17 farmers (45%) had between 4 and 6 years’ of experience in growi
ng upland rice,
11 farmers (29%) had 1
–
3 years, 9 farmers (24%) had 7
–
9 years and only one farmer (3%) had
between 10
–
12
years of experience (Table 8
).
Table
8
:
Farmer participant characterization: gender ratios, age
–
groups and rice farming
experiences of
farmers in Namutumba district Eastern Uganda in 2015.
Variable
Number of respondents N=38
Percentage of respondents
Gender
Male
21
55
Female
17
45
Age distribution
18
–
30
9
24
31
–
41
11
28
42
–
54
14
37
55 and above
4
11
Years of experience
in rice
farming
1
–
3 years
11
29
4
–
6 years
17
45
7
–
9 years
9
24
10
–
12 years
1
3
66
This study revealed that the largest number of upland rice farmers who turned up for the
participatory variety selection (PVS) were
men (Table 8
), these results are similar to Nanfumba
et al.
(2013) and Adekunle et al. (2013) who also found a lower number of women who turned
up for the variety selection for rain fed lowland ecologies and upland rice respectively. The
lower involvement of female fa
rmers is attributed to social economic constraints including
resource endowment, capital and land (Adekunle
et al.,
2013). Addison et al. (2014) who also
found a lower number of women rice farmers attributed it to rice being labour intensive
considering ot
her roles by women. It was also revealed that a high number of farmers in the
PVS were in age
–
group 42
–
54 and 31
–
41 years, these two groups constituted the largest
perce
ntage share of farmers. (Table 8
). This suggests that there is low involvement of youth
and elderly in rice farming. These findings were similar to Addison et al. (2014) who also
reported a lower number of youth and elderly involved in rice farming in low land rice
ecologies in Ghana. On other hand, the study indicated, out of 38 farmers in
the PVS, the largest
percentage of farmers had experience of 4
–
6 years and 1
–
3 years (Table 8
), this confirms a
survey carried out in 2005 by Advanced studies on international development (FASID) and
Makerere University which also revealed that in Eastern
Uganda household experience in rice
production was highest for years 1
–
3 (Kijima and S
erunkuma
, 2008). The lower number of
farmers with low experience in rice farming could be attributed to high dropout rate of
NERICA’S which the most cultivated upland ric
e varieties in Uganda and the lack of
functioning seed distribution system in these areas (Kijima
et al.,
2011). According to Kijima
et al. (2006) for areas which had upland rice introduced for first time, these got seed supply
from NGO’S such as Africa 2
000 Network, such areas lacked seeds even in input stores. From
the discussion with farmers, they claim to have got their seeds from NGO’S and that they
lacked seeds for cultivating in their fields.
67
4.2.2
Rice variety selection; the most preferred varietie
s by farmers in Eastern Uganda
The most preferred variety of rice was SCRID090 (
SCRID090
–
60
–
1
–
1
–
2
–
4)
(18%), this was
followed by NERICA
–
17 (16%) and an equal share of farmers (11%) preferred Blechai
(
IRGC78281
) and WAB 880 (
WAB880
–
1
–
32
–
1
–
1
–
P2
–
HB
–
1
–
1
–
2
–
2
).
About 13% of the farmers
selected NERICA
–
10, 10% selected NERICA
–
4, 5% NERICA
–
2 and only 2% selected
NERICA
–
1 as the m
ost preferred variety (Figure 4
)
The study showed that out of the 25 varieties of rice in the trial, the best preferred varieties of
rice
in ascending order were SCRID090, NERICA
–
17, NERICA
–
10, Blechai and WAB880
(Figure 4
). These constituted the five best preferred verities of rice, interestingly all the
varieties except NERICA
–
10 were introduced for the first time in Uganda and they
demons
trated suitability to local Ugandan conditions. Nanfumba et al. (2013) also reported
high preferences for improved rice varieties compared to local varieties by farmers in PVS for
rain
–
fed lowland ecologies in Uganda. Other varieties selected by farmers we
re WAB181
–
18,
NERICA
–
4, NERICA
–
2 and NERICA
–
1 (Figure 10), these also except WAB181
–
18 and
NERICA
–
2 have been grown in Uganda. From the study all the NERICA varieties featured
among the preferred varieties by farmers, these results are similar to Gridley e
t al. (2002) who
also found out in PVS (Participatory Variety Selection) in 2000 that NERICA’s were highly
preferred in Ivory coast. Among the 58 varieties they screened in their PVS trial in 2000 in
West Africa, the most frequently selected varieties were
designated by code WAB, with WAB
18
–
18 among the most preferred varieties which is also the case in our study. NERICA varieties
normally show stable yields under low and high input conditions and therefore are expected to
reduce risks and increase product
ivity in farmers’ fields (Gridley
et al.,
2002), this could be the
reason why they were highly selected by farmers.
68
Figure 4: Farmer participatory preference of upland rice varieties in Namutumba district
Eastern Uganda in 2015, bars indicate the percent
age (%) of farmers preferring a certain
variety.
4.2.3
Rice variety selection by gender; the most preferred varieties by male and female
farmers in Eastern Uganda
A t
otal of 18% and 16% of women selected SCRID090 and NERICA
–
10 respectively as the
best two
most preferred varieties, accounting 1/3 of the all the most
–
preferr
ed varieties by
women (Figure 5
). Among men, 18% selected NERICA
–
17 as their most
–
preferred variety and
17% selected SCRID090 as their second best va
riety (Figure 5
). Generally there were
significant differences (t= 4.803 P<0.001) in variety preferences between men and women, for
example, 12% of women preferred NERICA
–
4, compared to 9% of men selecting this variety
(Figure 2). NERICA
–
4, was preferred by 13% of the men and only 10% of the w
omen, Blechai
was preferred by 13% of the men and 15% of the women, while WAB181
–
18 was preferred by
13% of the men and
only 9% of the women. (Figure 5
). Sorted from the most to the least
frequent choices by women, the most
–
preferred varieties are SCRID090
, NERICA
–
10,
WAB181
–
18, NERICA
–
17, NERICA
–
4, Blechai, WAB 880, NERICA
–
2 and NERICA
–
1. For
0
2
4
6
8
10
12
14
16
18
20
SCRID090
NERICA-17
WAB181-18
NERICA-10
BLECHAI
WAB880
NERICA-4
NERICA-2
NERICA-1
Percentage of rice farmers
Upland rice varieties
69
men the order was:
NERICA
–
17, SCRID090, Blechai, WAB880, WAB181
–
18, NERICA
–
10,
NERICA
–
4, N
ERICA
–
2 and NERICA
–
1 (Figure 5
).
The results of the current study also r
evealed differences in variety selection among men and
women. Men selected NERICA
–
17 as their best rice variety whereas women selected
SCRID090 as their best variety
.
There were mainly three common varieties among the two
groups i.e. SCRID090, NERICA
–
17, W
AB181
–
18, it was not surprising to see that the first two
constituted the best varieties of rice selected by farmers. The selection could be related to the
women’s and men’s roles in crop valve chain. For example since men take a leading role in
marketing
(Wanyonyi
et al.,
2008), they prefer varieties that are high yielding with big grain
size which are characteristics of the chosen varieties in their group.
Dorward et al. (2007) also
found similar selection criteria used by men in PVS in Ghana. On the
other hand, women are
often involved in farm activities contributing over 50% of agricultural labour besides other
reproductive roles (FAO, 2011), and they prefer a variety that is early maturing to reduce on
farm work load. It is therefore not surprising
that they picked SCRID090, WAB181
–
18 and
NERICA varieties since these varieties are all early maturing. These results are similar to
Dorward et al. (2007), who also stated early maturity and yield as the most traits women used
in selection of upland rice v
arieties in Ghana. Also Addison et al. (2014) reported early maturity
as the most important trait in varietal preferences among women in low land rice ecologies in
Ghana.
70
Figure
5
:
Rice variety selection by gender of rice farmers in Eastern Uganda
(Namutumba district) in 2015, data presented as percentage responses by farmers.
4.2.4 The worst performing varieties of rice according to
farmers.
The worst performing varieties according to farmers were Superica and Anakila, these two
were selected by 20
%
of far
mers as least
–
preferred (Table 9
). These were followed by varieties
UPR
(
UPR
–
103
–
80
–
1
–
2)
and IAC 165 which
were
least preferred
by
(19%)
and 17%
of farmers
respectively (Table 9
).
Lastly
among the least preferred varieties were NARC 3 (ITA 257) and
Makassa
selected by 16%
and 8% of the farmers respectively
(Table 9
).
The study also revealed the five worst varieties, which included Anakila, Superica, Makassa
,
IAC165, UPR and NARC3 (Table 9)
Superica (local variety) and IAC165, were very
susceptible t
o
Striga
. These would therefore give low yields especially in these areas which are
highly infested with
Striga.
On the other hand Anakila and Makassa were low yielding with
weak stems hence could be attached by soil pests when they lodge. These two variet
ies also
have small seeds, this could explain the low yields. However Anakila was early maturing while
Makassa was late maturing. Other varieties like UPR
(
UPR
–
103
–
80
–
1
–
2) and NARC3
were not
only judged for being late maturing with short stature but also h
ad small seed sizes and were
0
5
10
15
20
NERICA 1
NERICA 2
NERICA 4
NERICA 10
WAB 880
BLECHAI
WAB 181-18
SCRID090
Percentage of rice farmers
Rice cultivars
Female
Male
71
susceptible to
Striga
. The short stature makes it difficult to harvest these varieties especially
in areas where farmers use knives for harvesting. NARC3, despite having a good aroma, was
negatively judged for being susceptibl
e to drought. These results show that farmers dislike low
yielding, late maturing, rice varieties with weak stems, small grains and very susceptible to
Striga
, drought and lodging. These results are similar to those by Namufumba et al. (2013) who
also indi
cated similar reasons for farmers disliking rice varieties for low land ecologies except
Striga
resistance.
Table
9
:
Reasons for rice variety selection among the least
–
liked, as indicated by farmers
in Eastern Uganda in 2015.
4.2.5
Farmer se
le
ction criteria for upland rice varieties.
Striga resistance and maturity period, with
the highest mean score 4.86 and 4.81 respectively,
were the most important characters for farmers t
o select rice varieties (Table 10
). Other
characters such as yield (4.62), tillering ability (4.42), drought resistance (4.38) and height
(4.33) were also co
nsidered by farmers as very important
attributes in selection (Table 10
).
Resu
lts
indicated that when farmers get new varieties, they often compare them with those
currently grown on the basis certain characteristics. These were mainly
Striga
resistance,
Variety
% of
farmers
Reasons for disliking the varieties
Superica
20
Very susceptible to Striga, low yielding due to Striga
UPR
19
Late maturing, small seeds, short variety (difficult to harvest, easily overwhelmed by
weeds), susceptible to Striga
IAC 165
17
Very
susceptible to Striga, sensitive to much sunshine, low yielding
Anakila
20
Weak stems making it susceptible to lodging, labour intensive, small sized grains, low
yielding
Makassa
8
Late maturing, small sized grains, fewer panicles
, has awns
NARC3
16
Late maturing, susceptible to Striga, short variety (difficult to harvest, easily
overwhelmed by weeds), small sized grains, susceptible to drought
72
maturity period, yield, height, tillering, drought t
olerance and grain size (Table 10
). Among
these traits,
Striga
resistance, early maturity, tillering and yield were the most important criteria
for rice variety selection in Eastern Uganda. Farmers prefer
high yielding,
Striga
resistant rice
varieties that mature early with good tillering ability to provide high incomes even in short
rainy seasons. All the selection criteria as per this study, except
Striga
resistance, were similar
and consistent amongst
farmers in 17 countries PVS in 1999 in West Africa (Gridley
et al.,
2002). Surprisingly, in most studies yield is seen as the most important criteria in rice variety
selection (Nanfumba
et al.,
2013; Gridley
et al.,
2002; Kimani
et al.,
2011). In the curre
nt
study,
Striga
resistance emerged as an important trait and this could be due to this area being
infested with
Striga
. A survey done in 2008 indicated
Striga
as a serious problem in upland
rice in Eastern Uganda (Pittchar and Mbeche, 2008
unpublished inf
ormation
). The local rice
variety (Superica) inspite of being high yielding was rejected by farmers for being susceptible
to
Striga
. Therefore new sources of resistance were seen as the only way to solve the problem.
Table
10
:
Criteria rice farmers use
in selection of the best upland rice varieties in Eastern
Uganda in 2015
Character
Mean score
*
Standard deviation
Rank
Striga resistance
4.86
0.35
1
M
aturity
4.81
0.4
0
2
Y
ield
4.62
0.63
3
Good tillering ability
4.42
0.50
4
Drought resistance
4.
3
8
0.52
5
Good seed colour
4.00
0
.00
7
Height
4.33
0.48
6
*Scores 1
–
5 where1 = not important 2 = not important at all3 = more or less important 4=
important, and 5= very important. Ranking was performed from most important
–
least important
trait, with 1
indicating the most important trait. Data was presented as mean scores for different
characters.
73
4.2.6
Variety ranki
ng based on farmer selection criteria
Table 11
shows farmers’ evaluation on the expression of traits in the most preferred varieties.
Resul
ts
indicated
SCRID090 with
the highest
phenotypic
acceptability among rice farmers
and
with
mean score of 2.76 ranked as the first variety. This was followed by WAB880 with (2.75)
and NERICA
–
17 (2.63) which were ranked as second
and third respectively (Table 11
).
BLECHAI with mean score of 2.60 and WB 18
–
18 with mean score of 2.58 were ranked as
fourth and fifth respectively. NERICA
–
4 with (2.54) and NERICA
–
1 with (2.52) were ranked
as sixth and seventh respectively. NERICA
–
10w
ith mean score of 2.43 and NERICA
–
2 with
mean score of 2.37 were ranked as eighth
and ninth respectively (Table 11
).
In relation to farmer selection criteria, farmers’ ranked varieties SCRID090, WAB880,
NERICA
–
17, Blechai and WAB181
–
18 as the best variet
ies in ascending order, these were
ranked as the five best rice
varieties in this study (Table 11
). Other varieties also ranked by
farmers were NERICA
–
4, NERICA
–
1, NERICA
–
2 and NERICA
–
10, they were ranked sixth’s,
seventh’s, eighth’s and ninth’s respective
ly. Farmers ranked SCRID090 and WAB800 as their
best varieties owning to their unique attributes, these varieties were not only early maturing but
also with good grain size and tillering ability which are all traits related to yield. These
constitute the m
ost sought attributes in rice varieties (Gridley
et al.,
2002; Dorward
et al.,
2007;
Efisue
et al.,
2008; Nanfumba
et al.,
2013). The inherent ability of these varieties to resists or
perform despite
Striga
infestation is another reason for s
electing them
(see study one
). Farmers
selected NERICA
–
17 because of its big grains as well as being drought tolerant and
Striga
resistant. Blechai was chosen due to its height and
Striga
resistance. It has been reported by
several studies that farmers prefer tall varie
ties because they reduce the burden of bending
when harvesting (Efisue
et al.,
2008; Kimani
et al.,
2011). WAB 181
–
18 was chosen among
the five best varieties owning to its big sized grains and early maturity, early maturity varieties
are desired because t
hey are seen as drought escaping options (Nanfumba
et al.,
2013; Hill,
74
2004). On the other hand NERICA
–
4 and NERICA
–
1 were ranked sixth and seventh
respectively due to both being high yielding and drought resistant, all of which are
characteristic traits i
n NERICA varieties (Gridley
et al.,
2002; Rodenburg et al., 2015),
however NERICA
–
1 had a better tillering ability compared to NERICA
–
4. NERICA
–
10 and
NERICA
–
2 ranked eighth and ninth respectively because both were early maturing and thus
could be cultivat
ed in short rainy season which are characteristics of the Ugandan conditions,
however NERICA
–
10 was high yielding though not as resistant as NERICA
–
2 to
Striga
.
Table
11
:
Ranking of the nine varieties of upland rice based on farmer selection criteria
in
Eastern Uganda in 2015.
*
Scale
(1
–
3) where 1= Bad, 2=Good/sufficient and 3=Very good. Ranking was performed from
most preferred
–
least preferred
among the most selected varieties, with 1 indicating the most
preferred variety, data was presented as mean scores
Variety
Striga
resistance
*
Maturity
time
Yielding
ability
Tillering
ability
Drought
resistance
Height
Grain
size
Across
trait
means
Rank
NERICA
–
10
2.61
2.74
2.67
2.25
2.25
2.47
2.00
2.43
8
SCRID090
2.77
2.84
2.73
2.87
2.63
2.50
3.00
2.76
1
NERICA 4
2.44
2.78
2.80
2.33
3.00
2.42
2.00
2.54
6
WAB
880
2.75
2.75
2.82
2.89
2.57
2.44
3.00
2.75
2
WAB
181
–
18
2.50
2.70
2
.5
0
2.43
2.50
2.40
3.00
2.58
5
Blechai
2.81
2.68
2.59
2.70
2.43
2.67
2
.3
3
2.60
4
NERICA
–
17
2.65
2.56
2.52
2.71
2.71
2.42
2.83
2.63
3
NERICA
–
2
2.67
2.50
2.40
0.00
2.00
2.13
2.50
2.37
9
NERICA
–
1
2.33
2.33
3.00
3.00
3.00
2
.00
2.00
2.52
7
75
4.3
Study three
evaluation of post
germination
attachment resistance of upland rice
varieties to
Striga hermonthica
under controlled environ
ment conditions
4.3.1
How does
Striga hermonthica
alter the morphology of host
4.3.1.1
Plant height
Plant height of the infected rice plants were compared with those of the uninfected plants to
evaluate the impact of
Striga hermonthica
on the growth of the rice plants
(Table 12)
. Rice
plant height was significantly (P<0.001) affected by
Striga
for varieties IAC 165, Superica
(local variety), WAB181
–
18 and WAB880, these had significantly shorter stems compared to
their respective control
s
(Table 2)
. The rest of the varieties were not significantly reduced
(P>0.001) in height with r
espect to their controls (Table
12). Varieties
Superica, IAC 165,
WAB181
–
18 andWAB880 had a great reduction in stem height 44.6%, 33.9%, 24.8% and
23.6% respect
ively when infected with
Striga hermonthica
(Table
12).
Additionally varieties
SCRID090, NERICA
–
1 and WAB56
–
50
registered no reductions in plant height when infested
with
Striga
(Table 12).
Irrespective of the treatments, variety SCRID090
and Blechai had
t
he
high
est plant height while NERICA
–
17
had the lowest
plant height
(Table 12).
The study has indicated
v
arieties Superica, IA
C 165, WAB880 and WAB181
–
18
a
s having
significantly high reductions in stem height following infection with respect to control pl
ants.
These results are consistent with studies by
Cechin and Press, 1994
;
Walting and Press, 2000;
Swarbrick
et al.,
2008; Atera
et al.,
2012 who also indicated that cereals such as rice when
infected with
Striga
species exhibit characteristic changes in
plant morphology and architecture
as compared to uninfected plants including stunting /reduction in stem length of infected plants
.
Stunting could be as result of lack of internode elongation rather than increase in internode
numbers. Internode elongation
is based on increased cell elongation in well
–
delineated zones
of the internode (Kende
et al.,
1998).
76
Alteration of growth regulators metabolism is one hypothesis that may account for this change
in stem height following infection. A number of growth regu
lators are involved in elongation
of stems in plants including Gibberellins, cytokinins and auxins (Kende
et al.,
1998; Sakamoto
et al., 2006; Ikeda et al., 2001). Siliva and Jorge, (2010) reviewed a number of studies on the
effect on the growth regulators
on tillering, these concluded that existence of separate
gibberellin
–
mediated pathways control tillering and plant height. Gibberellins are involved in
cell wall expansion by induction of cell wall loosening enzymes (Kende
et al.,
1998). Many
rice mutants
have been described to a lack of gibberellins or to be insensitive to this hormone
(Ishikawa
et al.,
2005). Therefore any alteration in the synthesis of gibberellins will eventually
lead to reduced height of infected plants. According to study by Swarbric
k et al. (2008) many
genes involved in auxin and gibberellin signaling are down regulated following
Striga
infection. Therefore reduction in internode extension in
Striga
infected plants could be the
result of an alteration in gibberellin and auxin signal
ing, hormones important in stem
elongation.
Another explanation for stunting of rice plants after infection has been suggested to be related
to translocation of toxic compound from
S.hermonthica
to the host (Musselman and Press,
1995). Also some studies ha
ve suggested the involvement of secondary metabolites toxic to
cereals to have an effect on host morphology (Ejeta and Butler, 1993). These together with
Rank et al. (2004) revelation of irioid glucosides and their suppression of cell division can also
exp
lain this impact of
Striga
on plant’s height. However such toxins have not been identified
and evidence of this mediated effect of this host has not been presented.
77
Table
12
:
Height of rice varieties infected with
Striga hermonthica
and their respective
uninfected (control) grown under controlled conditions
Plant height (cm)
Treatment
Rice variety
Striga
infected
Uninfected
(
control
)
Variety means
% loss in plant
height
BLECHAI
17.8
0
18.18
17.99
c
2.1
CG14
14.02
14.6
0
14.31
abc
4.0
IAC 165
11.5
0
17.4
0
14.45
abc
33.9
IRGC
14.07
16
.00
15.04
bc
12.1
NERICA
–
1
14.1
0
14.03
14.06
ab
0.0
NERICA
–
17
10.57
11.53
11.05
a
8.3
NERICA
–
4
15.03
15.53
15.28
bc
3.2
SCRID090
18.05
17.95
18
.00c
0.0
SUPERICA
10.47
18.9
0
14.69
abc
44.6
WAB
181
–
18
15.8
0
17.37
16.59
bc
24.8
WAB 56
–
104
13.07
13.88
13.47
ab
5.8
WAB 56
–
50
15.65
14.23
14.94
bc
0.0
WAB 880
14.28
18.68
16.48
bc
23.6
WAB 928
12.72
13.95
13.34
ab
8.8
Means
16.07
a
14.12
b
15.1
0
LSD
3.04
S.E
2.16
CV%
14.3
Means with the
same letters are not significantly different us
ing Tukey multiple comparison
test (P< 0.05)
.
4.3.1.2
Stem width
The stem width per variety of rice were significantly reduced (P<0.05) for varieties IAC 165,
Superica, WAB 181
–
18, Blechai and WAB 880
(Table 1
3)
. These varieties had significantly
thinner stems compared to their r
espective control plants (Table
13). There was no significant
difference in stem width (P>0.05) of the infected plants of the remaining varieties with respect
to their non
–
infected cont
rols.
Additionally, variety NERICA
–
17 and Superica had significantly
78
(P< 0.05) thinner stems compared to SCRID090 with a bigger stem diameter (Table 13).
Also
there was a reduction in stem width when rice plants were infected with
Striga
. However
varieties varied greatly with some having significantly reduced stem diameters such as
WAB880 (22.3%), IAC 165 (20.0%) and Superica
(26.3%) (Table
13). Other varieties with
moderate reductions in stem diameter following infection were Blechai (16.2%), WAB1
81
–
18
(14.9%), IRGC (13.4%) and CG14 (13.0%). The rest of the varieties had low reduction in stem
diameter (<10%) with WAB56
–
50 having registered no
reduction in stem width (Table
13).
Study results indicated reduction in stem diameter of rice plants infec
ted with
Striga
hermonthica
when compared to uninfected control
plants across varieties (Table
13). Varieties
Superica, IAC165 and WAB880 had a high
reduction in stem width (Table
13). These except
WAB880 supported a high parasite load. This is only an ind
ication of great effect of
Striga
on
these varieties when compared to other varieties in the study. It appears that this high reduction
in stem diameter of these varieties could be attributed to earlier infection of the parasite. Studies
by Cechin and pres
s, (1993) demonstrated that earlier attachments of parasites had greater
effect on host growth than later attachments. Other varieties with less reduction in stem
diameter such as WAB928, WAB56
–
50, WAB56
–
104, NERICA
–
1, SCRID090, NERICA
–
4
and NERICA
–
17 is a
ttribute to less effect of the parasite on these varieties.
Another possible explanation for reduction in stem width of infected plants relative to the
uninfected control plants is related to limited assimilate partitioning to other activities of the
plant
including stem enlargement.
Striga hermonthica
survives by diverting essential nutrients
which could otherwise be used by plants (Rodenburg
et al.,
2006; Atera
et al.,
2011). In other
words the parasite may act as an alternate sink for the assimilates (Gu
rney
et al.,
1999). These
nu
trients are responsible for the entire
essential processes of the
plant;
it could therefore be
possible that the thinner stems in susceptible cultivars such as IAC 165 and Superica is due to
diversion of nutrients to parasite w
hich could be used for stem enlargement.
79
Table
13
:
Stem width of the infected and uninfected control rice plants of rice varieties
grown under controlled conditions.
Stem width (mm)
Treatment
Rice variety
Striga
infected
Uninfected
control
Variety means
% loss in
width
BLECHAI
1.96
2.34
2.15
ab
16.2
CG14
1.88
2.16
2.01
ab
13
.0
IAC 165
1.92
2.4
0
2.16
ab
20
.0
IRGC
2.06
2.38
2.22
ab
13.4
NERICA
–
1
2.11
2.13
2.12
ab
0.9
NERICA
–
17
1.94
1.97
1.96
a
1.5
NERICA
–
4
1.97
2.12
2.05
ab
7.1
SCRID090
2.34
2.54
2.44
b
7.9
SUPERICA
1.68
2.28
1.98
a
26.3
WAB 181
–
18
2.11
2.48
2.29
a
b
14.9
WAB 56
–
104
2.16
2.24
2.2
0ab
3.6
WAB 56
–
50
2.17
2.11
2.14
ab
0
.0
WAB 880
1.99
2.56
2.28
ab
22.3
WAB 928
2.18
2.24
2.21
ab
2.7
Means
2.03
a
2.28
b
2.16
LSD
0.37
S.E
0.27
CV%
12.3
Data was first transfo
rmed
to meet the assumption for analysis of variance.
Means with the
same letters are not signi
ficantly different using Tukey multiple comparison test (P< 0.05)
.
80
4.3.2
Effect of
Striga hermonthica
on
tillering and leaf number of rice varieties
4.3.2.
1
Number of tillers
There was no significant difference (P> 0.05) in number of tillers produced by infected plants
compared to the non
–
infected controls when analyses across the varieties
(Table 14)
. Howeve
r
after
Striga
infection the local variety (Superica) produced a significantly (P< 0.05) lower
number of tillers compared to the u
ninfected control plants (Table
14). Number of tillers for
the control plants ranged from ~3tillers for WAB 181
–
18 to ~1 tille
r for Blechai. For infected
plants, tiller number ranged from highly susceptible varieties with no tillers for the local variety
to highly resistant with 2 ti
llers per plant for CG14 (Table
14). The results show that
irrespective of the treatments variety
WAB181
–
18 produced significantly (P< 0.05) the highest
number of tillers compared to Blechai, Superica and WAB56
–
104 with a lower number of tillers
(Table 14).
Also,
control plants produced
significantly
more tillers
(P<
0.05)
than
infected
plants
(Table
14)
.
This study has indicated less tiller production among rice plants infected with
S.hermonthica
compared to u
ninfected control plants (Table
14). The results are similar to Cissoko
et al.,
(2011) who showed that infection of rice plants with
Striga her
monthica
and
Striga asiatica
suppressed tillering of infected plants compared to the uninfected plants across rice varieties.
Also since tillers play a major role in determining plant architecture and yield. These results
could suggest reduction in yield
of
Striga
infected rice varieties especially those with high
Striga
numbers such as IAC 165 and Superica which had highly reduced tiller number.
Reduction in the number of tillers of infected rice plants compared with the control is
attributed
to both suppression of outgrowth of tiller buds and inhibition of the formation of tiller buds by
Striga
. Out of 14 varieties tested, only three varieties WAB56
–
104, NERICA
–
1, IRGC had a
high number of tillers compared to the uninfected plants.
The remaining varieties had lower
81
number of tillers when compared with the uninfected plants. This is a clear indication that
Striga
had an effect on the tillering performance of these varieties.
One possible explanation for lower number of tillers in inf
ected rice plants is attributed to
growth regulators just like in plant height. Hormones such as auxins and cytokinins have been
known to control tillering (Garba
et al.,
2007; Kamato
et al.,
2006). Auxins have an indirect
inhibitory action on tillering wh
ile cytokinins directly promote tillering (Ongaro and leyser
2008). Therefore any changes in synthesis of these hormones will affect tillering of rice.
Further evidence has showed
Striga
infected sorghum tissues having greater amounts ABA and
ethylene and
lower amounts of Cytokinins (Drennan and ELhiweris, 1979) hormone that is
important in tiller formation.
Another possible explanation for the lower tiller production for the infected plants could be
due to Nutrition imbalances especially with respect to n
itrogen. Cruz and Boval, (2000);
Mckenzie, (1998) found a positive effect of nitrogen availability and tillering.
Striga
hermonthica
also depends on host for all its nitrogen demands once a xylem connection has
been established with the host (Pageau
et al
.,
2003). It is therefore possible that tillering is
lowered because the nitrogen is diverted to the parasite
82
Table
14
:
Number of tillers of the infected and uninfected control rice plants per variety
of selected rice varieties under controlled
conditions
Number of tillers
Treatment
Rice variety
I
nfected
with
Striga
Un
infected
(Control)
Variety means
BLECHAI
0.25
0.25
0.25a
CG14
2
.00
2
.00
2
.00
bc
IAC 165
0.5
0
1.25
0.88
abc
IRGC
1.5
0
1.25
1.38
abc
NERICA
–
1
1.5
0
1.25
1.38
abc
NERICA
–
17
0.5
0
1.25
0.88
abc
NERICA
–
4
0.75
1.25
1
.00
abc
SCRID090
1
.00
1.75
1.38
abc
SUPERICA
0
.00
1.5
0
0.75ab
WAB 181
–
18
2
.00
2.5
0
2.25c
WAB 56
–
104
0.75
0.75
0.75ab
WAB 56
–
50
1
.00
1.75
1.38
abc
WAB 880
1
.00
1.25
1.13
abc
WAB 928
1.5
0
2.25
1.88
bc
Means
1.02
a
1.45
b
1.23
LSD
1.12
S.E
0.8
0
CV%
64.8
Means with the same letters are not significantly different us
ing Tukey multiple comparison
test (P<0.05)
.
4.3.2
.2
Number of leaves
Number of leaves was significantly different (P<0.05) between the infected
and uninfected
control plants for local variety (Superica)
(Table 15)
. However there was no significance
difference (P>0.05) in the number of leaves produced per plant for the infected and uninfected
controls of all the rem
aining varieties of rice (Table
1
5). For the infected plants varieties WAB
181
–
18 and CG14 produced the highes
t number of leaves per plant (13 and 12
leaves
83
respectively
) while Superica produced the lowest number of leaves per plant (6 leaves). On
av
erage control plants produced
significantly (P< 0.05) a high number of leaves
(
9
leaves)
compared with
infected plants
which
produced 8
leaves.
Also varieties WAB181
–
18 and CG14
produced the significantly (P< 0.05)
the highest number of leaves compared to the rest of the
varieties exce
pt IRGC, SCRID090, WAB928, WAB56
–
50 and NERICA
–
1 (Table 15).
Variety
WAB56
–
104 had the smallest number of leaves (Table 15).
There was a reduction in number of leaves between the infected rice plants and uninfected
control
plants across varieties (Table
1
5). These results agree with Cissoko et al. (2011) who
also indicated a lower leaf dry weight across varieties when they were infected with
S.hermonthica
and
S.asiatica
. All the rice varieties when infected with
S.hermonthica
produced
lower number of leave
s compared to their respective uninfected plants. However varieties
NERICA
–
1 and WAB56
–
104 had a high number of leaves compared to the uninfected plants.
The lower leaf number in these varieties except these two is related to the genetic resistance
levels
difference among varieties. This led to some varieties more significantly affected than
others. The high number of leaves among the two varieties could
be
related to late
Striga
attachment on host roots. It should however be noted that even though these pr
oduced a high
number of leaves they were not significantly different from control plants. Leaves are the main
photosynthetic organs of the plant, therefore their reduction can affect assimilate manufacture
and consequently yield of the plant. This only im
plies that those varieties that are severely
affected such as Superica and IAC 165 can ultimately have a low yield in the
Striga
prone
areas.
Changes in the number of leaves and leaf area of infected as compared to the uninfected plants
have
also been rep
orted (Frost
et al.,
1997; Cechin and Press, 1994; Atera et al., 2012). The
reduction in leaf number of the infected compared to uninfected plants could be due to nutrient
diversion from the host plant to the parasite, there is less dry matter partitioned
in the growth
84
of new leaves. Studies have also shown rates of photosynthesis are usually lower in infected
plants than the uninfected plants (Gurney
et al.,
1997; Frost
et al.,
1997). It is possible that the
reduced photo
–
assimilate production due to the c
hange in plant architecture after infection is
responsible for the reduced number of leaves in infected plants compared to the control plants.
Table
15
:
Maximum number of leaves of the infected and uninfected control rice plants
per variety under
controlled conditions
Number of leaves
Treatments
Rice variety
Striga
Infected
Uninfected
(control)
Variety means
BLECHAI
7
.00
7.5
0
7.25
ab
CG14
11.5
12.25
11.88
c
IAC 165
6
.00
7.5
0
6.75
ab
IRGC
9.75
10.75
10.25
bc
NERICA
–
1
10
.00
8.5
0
9.25
abc
NERICA
–
17
7.5
0
8
.00
7.75
ab
NERICA
–
4
7
.00
8.25
7.62
ab
SCRID090
10.25
10.5
10.38
bc
SUPERICA
6
.00
9.75
7.88
ab
WAB 181
–
18
11.5
0
13.75
12.62
c
WAB 56
–
104
6.5
0
6.25
6.38
a
WAB 56
–
50
8.25
9.75
9
.00
abc
WAB 880
7.25
8.25
7.75
ab
WAB 928
8.25
10
.00
9.12
abc
M
eans
8.32
a
9.42
b
8.87
LSD
3.101
S.E
2.207
CV%
24.9
Means with the same letters are not significantly different using T
ukey multiple comparison
test (P<0.05)
.
85
4.3.3
Rice cultivars exhibit differential resistance to
Striga
hermonthica
.
4.3.3.1
Quantification of post
–
attachment
Striga
germination
There was a significant difference (P< 0.001) in the number of
Striga
parasites atta
ched per
rice plant across
rice varieties
(Appendix 16
)
.
IAC 165 had significantly (P<0.001) the
highest
number of
Striga
plants per host root system though it was not significantly different from
Superica and NERICA
–
1 (Table 16). Additionally, CG14 had the lowest
Striga
plants per host
root system (Table 16).
Parasitic attachments per host plant rang
ed from 30
–
45 on the very
susceptible varieties Superica and IAC 165, supporting the highest number of
Striga
plants, to
<1 on the most resistant variet
ies IRGC and CG14 (Table 16
). Also cultivars such NERICA
–
1,
NERICA
–
4 and NERICA
–
17 supported relatively large number of
Strig
a plants (6
–
15) parasi
tes
per host root (Table 16
). The remaining cultivars exhibited a good level of post
–
attachment
resistance supporting 1
–
5 paras
itic plan
ts per host (Table 16
). Generally varieties Superica and
IAC 165, had on average the largest well developed parasites on root system compared to the
rest of the varieties (Figure 6
a, b, c, d, e and f).
Post
–
attachment resistance levels varied significant
ly P<0.001 acr
oss varieties tested
(Appendix
16
).Varieties IAC165 and Superica were the most affected both during the course
of the experiment and at Harvest. Similarly studies by Gurney et al. (2006); Swabrick et al.
(2008); Cissoko et al. (2011) have all
reported IAC 165 as susceptible and has been used as a
susceptible check in these studies. Variety CG14 was the most resistant variety as per this study
having on average 1
–
2 developed parasites. Field and pot experiment studies by Johnson et al.
(1997) s
tressed the high
Striga
resistance potential of
O.glabberima
varieties compared to
O.sativa
varieties. This therefore agrees with results from this study indicating
O.glabberima
variety CG14 as highly resist
ant. Similarly Kaewchumnog and P
rice (2008) repor
ted CG14 as
one of the most resistant variety in their study. On the other hand post
–
attachment resistance
86
of some varieties from this study was similar to observations by Cissoko
et al
, (2011) who also
found out that varieties NERICA
–
4, WAB56
–
50, WAB181
–
18 and CG14 exhibited a good level
of post
–
attachment resistance to
Striga hermonthica
(Sh
–
Kibos) and
Striga asiatica
(Sa
–
USA)
and IAC165 was the most susceptible as per this study. This implies that these varieties are
resistant to a wide number of ecotyp
es of
Striga
including the one used in this study. Varieties
such as WAB880, WAB928, SCRID090, Blechai, and IRGC were evaluated for post
–
attachment resistance for the first time. These varieties exhibited an excellent post
–
attachment
resistance to
S.hermon
thica
. Field experiments by Harahap et al. (1993) also showed that
Blechai and IRGC were highly resistant to
Striga hermonthica
in field trails in Kenya. Also
studies by Johnson et al., (2000) reported WAB928 as a resistant variety to
S.hermonthica
parasit
ism. However no documented studies have indicated the reaction of varieties SCRID090
and WAB880 to
Striga
. It is therefore the first time we report the resistance of these vari
eties
in both filed (study one
) and under controlled conditions. Varieties such
as NERICA
–
1 and
–
17 reported to be resistant to
Striga hermonthica
(Cissoko
et al.,
2011; Jamil
et al.,
2011;
Rodenburg
et al.,
2015) supported a moderately number of
Striga hermonthica
in this study,
though the
Striga
plants were relatively small thus contributing less to the total
Striga
biomass
per plant. The small sized parasites could be attributed to time of emergency of the parasite as
they could have emerged later or due to defense mechanisms in these varieties.
4.3.3.2
Striga
hermonthica dry weight
Striga
biomass was significantly (P<0.001) higher for varieties IAC 165 and SUPERICA (21
–
24mg), compared to other varieties
(Appendix 16
). However the
Striga
biomass of Superica
was not significantly different from N
ERICA
–
1 and NERICA
–
17 (Table 16). Varieties
NERICA
–
17,
–
1 and WAB56
–
50 had moderate biomass of
Striga
ranging from 3
–
5mg per plant
on average (Table 16). There was no significant difference (P>0.001) in
Striga
biomass
between the remaining varieties (T
abl
e 16
). These varieties had a relatively small
Striga
87
biomass ranging from 0.02mg
–
1.58mg per plant (Table 16). The results characterize varieties
into susceptible (IAC165, SUPERICA) intermediate (NERICA
–
1,
–
17 and WAB56
–
50),
resistant (WAB800, WAB56
–
104, NE
RICA
–
4, SCRID090, WAB181
–
18) and very resistant
(WAB928, BLECHAI, IRGC and CG14) (Ta
ble 16
)
The results show differences in resistance among the rice varieties. Varieties with high
Striga
biomass such as IAC 165 and Superica also had the highest
Striga
nu
mbers. Similarly, Cissoko
et
al. (2011) a high
Striga
biomass
on variety IAC165 attributing it to increased
Striga
infestation. Other
varieties i.e. NERICA
–
1,
NERICA
–
17 recorded a higher number
of
Striga
biomass
though not compared to the above two varieties. This is a clear indication of an
intermediate number of
Striga
parasites per root system. The highly resistant varieties such as
WAB928, CG14, Blechai, IRGC and CG14 with a lower
Striga
biomass was possibly
due to
limited
Striga
parasite development.
Varieties NERICA
–
1, WAB
56
–
50, WAB880 and
NERICA
–
4 with a
Striga
biomass not correlating to the
Striga
number could only mean a
reduced size of the
Striga
plants attached on these varieties. This could also be as
result of a
defense mechanism from
Striga
establishment on
these varieties.
88
Table
16
:
Number of
Striga
plants and
Striga
dry weight taken at 21 days after Striga
hermonthica infection under controlled conditions
Data was first transfo
rmed before analysis log (x+1) to meet the assumption for analysis of
variance. Means with the same letters are not significantly
different using Tukey multiple
comparison test (P<0.05)
.
Rice variety
Striga
number
Striga
dry weight
(mg)
BLECHAI
0.416abc
0.032
a
CG14
0.060a
0.008
a
IAC 165
1.594e
1.387
c
IRGC
0.156ab
0.024
a
NERICA
–
1
1.024cde
0.593
ab
NERICA
–
17
0.823bcd
0.608
ab
NERICA
–
4
0.703abcd
0.162
a
SCRID090
0.381abc
0.136
a
SUPERICA
1.486de
1.179
bc
WAB
181
–
18
0.476abc
0.119
a
WAB 56
–
104
0.385abc
0.188
a
WAB 56
–
50
0.657abc
0.466
a
WAB 880
0.711abcd
0.351
a
WAB 928
0.264abc
0.038
a
LSD
0.435
0.354
S.E
0.335
0.272
CV%
52.5
74.2
89
Figure 6
:
Parasitic plants attached on the host plant roots 21 days after infection with
Striga hermonthica.
These show differences in
Striga
attachments with IAC165 and
Superica having the highest number of attachments followed by NERICA
–
1 and
–
17 and
lastly WAB5
6
–
50 and WAB928 where parasites failed to attach on the host plants.
d) NERICA
a) IAC 165
b) SUPERICA
c) NERICA 1
d) NERICA 17
e) WAB 56
–
50
f) WAB 928
90
4.3.4
Effect of
Striga hermonthica
on
above ground
rice biomass of upland rice varieties
with respect to the uninfected rice plants.
There was
a significant difference (P<0.01
) in above
–
ground dry biomass of infected rice plants
for varieties WAB181
–
18, WAB 880, CG14, local variety (Superica) and IAC165 c
ompared to
their control plants
(Table 17, Appendix 16)
. Rice biomass of the remaining varieti
es was not
significantly (P>0.01
) affected by
Striga
infection (Table 17
). With respect to control plants,
WAB181
–
18 had the highest mean biomass (1.144g) and NER
ICA
–
17 had the lowest rice
biomass (0.347g). Infected rice varieties ranged from very susceptible cultivars, such as
Superica having biomass of 0.254g compared to the non
–
infected control plant biomass of
0.983g, to resistant varieties such, as IRGC with 0
.721g biomass of infected plants compared
to 0.922g of unin
fected control plants (Table 17
).
Irrespective of the infected and uninfected
control plants, varieties WAB181
–
18, SCRID090 and IRGC produced the highest rice biomass
which was significantly (P<0.0
1
) different from biomass produced by WAB56
–
104 and
NERICA
–
17 (Table 17)
91
Table
17
:
Above
–
ground rice dry biomass of infected rice plants and their respective
uninfected control plants under controlled conditions
Rice biomass (
m
g)
Treatment
Rice variety
Striga
infected
Uninfected
(control)
Variety means
% loss in biomass
WAB181
–
18
0.70
1.14
0.92c
39.19
WAB880
0.46
1.00
0.73abc
54.17
SUPERICA
0.25
0.98
0.62abc
74.15
SCRID090
0.85
0.97
0.90c
12.82
IRGC
0.72
0.92
0.82bc
21.84
CG14
0.53
0.80
0.67abc
33.27
IAC 165
0.30
0.75
0.52abc
60.17
WAB56
–
50
0.54
0.66
0.60abc
19.09
BLECHAI
0.55
0.58
0.57abc
6.50
WAB928
0.43
0.55
0.49abc
22.81
NERICA
–
1
0.48
0.49
0.49abc
1.91
NERICA
–
4
0.46
0.46
0.46abc
0.41
WAB56
–
104
0.43
0.45
0.44ab
5.09
NERICA
–
17
0.33
0.35
0.34a
6.06
Means
0.50a
0.72b
0.61
LSD
0.37
S.E
0.26
CV%
42.4
Means with the same letters are not significantly different u
sing T
ukey multiple comparison
test (
P< 0.05)
.
4.3.5
Impact of
Striga
hermonthic
a
on the biomass of
rice varieties
Generally there was a loss in biomass of all infected plants especially for the most susceptible
varieties such as Superica (74.2%
) and IAC 165 (60.2%) (Table 17
). Others varieties with high
loss in biomass included WAB800 (54.2%), WAB181
–
1
8 (39.2%) CG14 (33.3%) and
IRGC
92
28.2% (Table 17
). The remaining varieties except SCRID090 and WAB56
–
50 with 10
–
20%
loss in biomass, the rest had <10% loss in rice biomass.
Figure 7
: Relationship between percentage losses in total rice biomass of infected
plants
compared with control plants and the amount of parasite biomass dry weight on roots of
rice plants.
There was a significant negative linear relationship between the effect of
S.hermonthica
on the
host biomass and the amount of parasite biomass (R
2
=
0.417, P=0.007). The most susceptible
varieties Superica and IAC 165 were the most affected
of
all the varieties (Figure 7
). This
negative relationship would indicate that as you increase the
Striga
there is a significant effect
on th
e biomass of the host
(Figure 7
).
However this is due to the effect of two very susceptible
varieties. Many varieties had similar levels of infection but lost biomass to differing extents.
For example
WAB 56
–
104 and WAB181
–
18 had
similar levels
of
Striga
biomass (Figure 7
).
However the percentage loss in
host
was 5% compared to 39% this illustrates a difference in
tolerance between the two rice varieties to
Striga hermonthica
infection.
y =
–
2.0047x + 83.419
R² =
–
0.417
0
10
20
30
40
50
60
70
80
90
100
0
5
10
15
20
25
30
Biomass of
Striga
infected plants as a
percentage of the control biomass
Striga
biomass (mg)
Superica
IAC 165
WAB880
WAB181
–
18
CG14
WAB928
WAB56
–
50
N1
N17
SCRID090
N4
WAB104
IRGC
Blechai
93
Infection of rice plants with
Striga hermonthica
altered the partit
ioning of assimilate
s
to the
leaves and stems
of rice plants. This was revealed in the above ground rice biomass of
infected
rice plants (Table 17, Figure 7
).Similar results were reported for sorghum parasitized by
S.hermonthica
(Cechin and Press, 1993; Frost
et al.,
1997), m
aize (Taylor
et al.,
1996) and rice
(Cissoko
et al.,
2011). Rice biomass of infected rice plants was high for varieties IRGC,
SCRID090, BLECHAI, NERICA
–
1,
–
4,
–
17 and WAB928 compared to their respective
uninfected rice plants. This could indicated less effec
t of
Striga
on these rice varieties and is
attributed to the fact that these varieties were very resistant to
Striga hermonthica
(Harahap
et
al.,
1993; Johnson
et al.,
1997; Johnson
et al.,
2000; Cissoko
et al.,
2011) except NERICA
–
1and NERICA
–
17
.
The effect of
Striga
on rice biomass can be ultimately understood looking
at the percentage loss in biomass following
Striga
infection. Generally there was a reduction
in rice biomass across varieties following infection, it was however very pronounced in
varieties Superica, IAC165 and WAB880. These except WAB880 supported the highest
number of
Striga
plants. It is also possible that alteration in plant morphology of
Striga
infected
plants depended upon the variety genetic differences in resistance and tol
erance. Surprisingly
varieties such as CG14, WAB928, IRGC, WAB181
–
18 and WAB880 inspite of having lower
number of parasites per root system registered tremendous reduction in host biomass. On
contrally varieties NERICA
–
1, NERICA
–
17 with high number of para
sites when compared to
the above
varieties
registered a less reduction in host biomass. This can only be explained by
variety differences in tolerance levels. Similar insights were reported in sorghum by Rodenburg
et al. (2006); in rice Cissoko et al. (201
1); Rodenburg et al. (2015). This can only imply that
inspite of high
Striga
numbers, such varieties can still provide high yields. Also the fact that
percentage reduction in host biomass was linearly related to the host
Striga
biomass per root
system wou
ld suggest that the highly resistant varieties can produced high yield potentials in
Striga
prone areas.
94
CHAPTER
FIVE
5
.0 GENERAL DISCUSSION AND CONCLUSION
Striga
is a major biotic constraint in cereal production systems of Sub
–
Saharan Africa (Atera
et
al.,
2012). Despite all efforts to manage
Striga,
it has persisted and increased in number. A
number of
Striga
management technologies have been developed for a long time (Oswald
et
al.,
2005; Rodenburg and Johnson, 2009; Atera
et al.,
2011). However
,
farmers have not
universally adopted
these
technologies because either they are too costly or laborious (Gressel
et al.
,
2004
).
The development of tolerant and resistant varieties in upland rice growing areas
is a viable option to manage
Striga
. It is also
cost
–
effective as farmers don’t have to invest in
this technology. This study therefore sought to evaluate the resistance of rice varieties against
Striga hermonthica
in
Eastern Uganda
.
In this study two experiments were conducted on
–
farm
and under contro
lled conditions. The first experiment had two studies, (1) Evaluated the
performance of selected rice varieties under
Striga
infested field conditions and (2) Identified
in a farmer participatory manner rice varieties that could be adopted by farmers. The
second
experiment evaluated the post
–
attachment resistance of farmer selected upland rice varieties to
Striga hermonthica
.
The results of the first study indicated variation in
Striga
resistance and yield across varieties.
Out of the 25 varieties evaluated
, three
O.glabberima
varieties AGEE, CG14, Anakila, seven
O.sativa
varieties IR49, IRGC, WAB935, WAB928, SCRID090, WAB880, Blechai, one
O.sativa
parent WAB181
–
18 and NERICA varieties NERICA
–
2,
–
10,
–
17,
–
4 had an excellent
resistance to
Striga
compared to t
he locally grown variety Superica and IAC 165. All the above
varieties except WAB880 and SCRID090 have been reported as having an excellent resistance
to
S.hermonthica
elsewhere.
For example Blechai and IRGC by (Harahap
et al.,
1993) in
Kenya, IR49 (Harahap
et al.,
1993; Johnson
et al.,
1997) in Kenya and Ivory Coast, WAB928
95
and WAB935 (Johnson
et al.,
2000) in Ivory Coast. Jamil et al. (2012) reported
O.glabberima
varieties AGEE, Anakila and CG14 have having an excellent resista
nce to
Striga
. The NERICA
varieties
–
2,
–
10,
–
17,
–
4 and the
O.sativa
parent WAB181
–
18 was also reported by Rodenburg
et al. (2015) as having an excellent resistance to
S.hermonthica
in Mbita, Kenya. This can only
indicate resistance of the above varieties
to a number of
Striga
ecotypes.
Also these same
varieties demonstrated an excellent resistance
to Striga hermonthica
under controlled
conditions.
The susceptibility of the local variety (Superica) can be attributed to continuous
cultivation of this variet
y. Therefore virulence levels of local
Striga
population against this
variety increased in time.
Similarly this variety was very susceptible under controlled
conditions
(study three)
.
Results also indicated partial resistance of
O.glabberima
varieties
ACC,
Makassa and MG12 confirming results by Johnson et al (1997) who also reported them
as partially resistant to
Striga
hermonthica
in
Ivory coast. Varieties NERICA
–
1 and UPR
reported to be resistant in Mbita (Rodenburg
et al.,
2015) and (Harahap et al 1993) respectively
supported moderate number of
Striga
plants in this study. This could be related to differences
in virulence levels of the
Striga
ecotype in this area as compared to ecotype in Mbita against
these varieties. This
study has indicated susceptibility of varieties WAB56
–
50, WAB56
–
104,
NARC 3 (ITA 257) to
Striga hermonthica
.
Although these
except NARC3
were resistant under
controlled conditions.
These results are consistent with Rodenburg et al. 2015 who also
confirmed
WAB56
–
50 as susceptible to
Striga hermonthica
in Mbita
.
This could be due to lack
of pre
–
attachment resistance mechanisms in these varieties.
Varieties that combine
Striga
resistance and high yields is important to improve rice
production. However
Striga
resistant varieties don’t necessarily produce high yields. The study
has indicated variation in yield production across varieties under
Striga
infested field
conditions. The best performing variet
ies yielded an equivalent of 3
.
1
t ha
–
1
–
3.7
t ha
–
1
i.e.
SCRID0
90,
NERICA
–
2,
WAB181
–
18, NERICA
–
10 and NERICA
–
4 in 2014A. Additionally in
96
2015, the best performi
ng varieties yielded between 3.0
t ha
–
1
–
3.8
t ha
–
1
i.e. NERICA
–
2,
NERICA
–
4, NERICA
–
10, Blechai, SCRID090, W
AB181
–
18, CG14, AGEE and NERICA
–
17.
These results are
similar to Rodenburg et al. (2015) who also indicated varieties W
AB 181
–
18
and all the NERICA varieties as high yielding varieties in Mbita, Kenya. Also some studies
have reported the h
igh yielding potential of NERICA
varieties (Gridley, 2002; Saito
et al.
,
2012).
This is the same reason why all the above varieties except AGEE and CG14 were
selected by farmers in study two.
O.sativa
varieties WAB880 and SCRID090 evaluated for the
first did not only have good resistance to
Striga
but also produced high
yields. This could also
be related to a high
yielding potential of these varieties. The results
also indicated
O.glabberima
varieties Makassa, ACC, MG12, NERICA
–
1
O.sativa
parents WAB56
–
104,
WAB56
–
50 with partial resistance to
Striga
had high yield compare
d to varieties WAB928,
WAB935, IRGC and IR49 which were very resistance to
Striga
. This suggests variety
differences in tolerance to the deleterious effects of infection by
S. hermonthica
. Tolerance is
the ability of a give variety to produce yield even un
der high
Striga
infestation levels
(Rodenburg
et al.,
2006; Rodenburg
et al.,
2008; Rodenburg and Bastiaan
et al.,
2011). This
has also been reported in Sorghum (Gurney
et al.,
1995; Rodenburg
et al.,
2006).
Understanding the expression of this trait can b
e important to improve yield in the most
resistant varieties with low yields (Rodenburg and Bastiaans, 2011). The results also reported
a significant negative correlation between NSmax and grain yield in 2015. This indicated a
high
Striga
pressure in this
year compared to 2014. This also implied that the highly resistant
varieties produced high yields. The local variety Superica produced the lowest yield in this
2015
implying that
the
increase in
Striga
affected the performance of this variety.
This same
re
ason why farmers selected it as the worst performing variety.
In the second study, we sought to understand a selection criteria farmers’ use in identification
of the best preferred varieties. The study indicated
Striga
resistance, yield, tillering ability,
97
drought tolerance/resistance, maturity period and height of the varieties. Consequently, farmers
ranked SCRID090 as their best variety, this was followed by WAB880, NERICA
–
17, Blechai
and WAB181
–
18 as the five best varieties out of 25. These varieties wer
e as ranked among
those with high yields and excellent resistance to
Striga hermonthica
in the first study
and third
study
. Other varieties selected by farmers were NERICA
–
4, NERICA
–
1, NERICA
–
10 and
lastly NERICA
–
2 in ascending order of preference. All the
NERICA varieties included in this
study one were all selected by farmers
probably due to high yields as demonstrated in study
one
. Similar results were shown by Gridley et al. (2002) where farmers highly selected these
varieties in Ivory Coast. Also sever
al studies have indicated high yields and good
Striga
resistance among these varieties (Gridley
et al.,
2002; Saito
et al.,
2012; Rodenburg
et al.,
2015)
.
A number of studies have always indicated yield as the most sought trait in rice variety selection
(Namufumba et al., 2013; Gridley et al., 2002; Kimani et al., 2011). This study has however
indicated
Striga
resistance as the most considered trait in varie
ty selection. Having
Striga
resistance as the most sought trait in decision making to adopt a variety only qualifies the effect
of
Striga
on rice production in this area. Also from study one, results indicated the local variety
(Superica) as a highly susce
ptible variety to
Striga
. This is one reason why it was rejected by
farmers.
Results also indicated other
worst preferred varieties of rice which included NARC 3
(ITA 257), IAC 165, Makassa, Anakila and UPR. Surprisingly, varieties Anakila had an
excellent
resistance to
Striga
as well as high yields
(see study one)
.
This indicates difference
in traits sought by farmers and researchers.
Varieties Makassa and UPR has partial resistance
levels to
Striga
. The
O.glabberima
varieties Makassa and Anakila are susce
ptible to lodging,
this negative trait led farmers to reject all the
O.glabberima
varieties. The
O.sativa
variety
NARC3 (ITA 257) inspite of its good aroma was rejected by farmers for being short,
susceptible to
Striga
and being low yielding, this same var
iety produced low yields in study
98
one.
On other hand UPR was rejected for it short suture and low yields. It was also shown in
some studies that farmers prefer tall varieties which reduce on the burden of bending while
harvesting (Efisue
et al.,
2008; Kima
ni
et al.,
2011).
This study also indicated significant difference in variety selection among men and women.
For example Men selected NERICA
–
17 and women selected SCRID090 as the best variety for
each group. This is related to the roles of men and women i
n the crop life cycle. Women
normally carry out the initial roles in the crop life cycle such as sowing, weeding while men
normally involve in the marketing chain of the crop. Studies have shown differences in traits
men and women sought in the newly intro
duced varieties (Dorward
et al.,
2007). Variety
SCRID090 selected by women is tall,
Striga
resistant and early maturing
(see study one and
three)
. However though NERICA
–
17 is high yielding and
Striga
resistant it is not as earlier
maturing as SCRID090.
Addison et al. (2014) acknowledged early maturity as one of the most
important trait women sought in selecting varieties in Ghana.
In study three, a selection of varieties in the first and second study were evaluated for their post
attachment resistance un
der controlled conditions. Also the study investigated whether the
resistance observed in the field can be replicated under controlled conditions. Results indicated
significant effects of
Striga
on rice varieties when compared to control plants. This was
m
anifested in the reduced rice biomass. Generally there was a loss in biomass across varieties,
however
some varieties i.e. Superica, IAC165, WAB880, and WAB181
–
18, CG14, WAB928
and IRGC were
more
affected when compared to varieties NERICA
–
1, WAB56
–
50, WAB5
6
–
104, SCRID090, Blechai,
NERICA
–
4
and NERICA
–
17 with respect to their control plants
.
It
has been proposed that
the
loss of biomass of
Striga
infected plants to some extent is due to
acquisition of carbohydrates,
nitrogen and other solutes from the host.
This results in change
in host architecture leading to stunting of plants especially for the most susceptible varieties,
lowering in tillering (Cissoko
et al..
2011; Jamil
et al.,
2011), leaf area, leaf number, thinning
99
of stems and shortening of stem inte
rnodes (Swarbrick
et al.,
2008; Gurney
et al.,
1999;
Ransom 2004) which ultimately reduces the total biomass of the infected plants. The second
notion to the reduction in infected biomass could be due to lowering of photosynthesis in
infected plants. Studi
es have shown rates of photosynthesis are usually lower in
Striga i
nfected
leaves due to stomatal closure following infection (Gurney
et al.,
1997; Frost
et al.,
1997).
Photosynthesis plays a greater role in the regeneration of new plant tissues that event
ually
increases on the plant biomass.
Striga hermonthica
alters the partitioning of dry matter to the
different parts of the plant (
Cechin and Press, 1994).
Therefore reductions or interruptions in
this process can reduce the biomass and ultimately the yi
eld of the varieties.
This study has indicated excellent post
–
attachment resistance for varieties CG14, WAB928,
IRGC, SCRID090, WAB181
–
18, WAB880, Blechai, WAB56
–
104 and NERICA
–
4. These
same varieties except WAB56
–
104 had an excellent resistance to
Striga
under field conditions
in study one. Also Cissoko et al. (2011) reported varieties WAB181
–
18, NERICA
–
4 and CG14
as having an excellent Post
–
attachment resistance to
S.hermonthica
ecotype (Sh.Kibos). The
hypothesis set that the resistance under field and c
ontrolled conditions does not differ holds
only true for the above varieties. However for varieties WAB56
–
104 and WAB56
–
50
considered susceptible under the field conditions edged out
Striga
by having a good resistance
to
Striga
. This however is not surpris
ing for variety WAB56
–
50 because it was reported as
highly resistance under controlled conditions (Cissoko
et al.,
2011; Rodenburg
et al.,
2015).
However when tested under field conditions in Kenya by Rodenburg et al. (2015) it was
susceptible. This can on
ly be expressed by lack of pre
–
attachment resistance mechanism, no
wonder Jamil et al. (2011) reported it as the highest strigolactone producer. Results indicated
varieties
NERICA
–
17 and NERICA
–
1 produced
moderate levels of
Striga
plants per root
system ye
t the former had excellent resistance
and the latter had partial resistance
to
Striga
under field conditions.
This implies that the resistance of NERICA
–
17
was not replicated under
100
controlled conditions. Results contradict with Cissoko et al. (2011) who re
ported these two
varieties as having an excellent post
–
attachment resistance to
Striga
. This could be related to
differences in the ecotypes of
Striga
used
in these
two studies.
This study also clearly examined the concept of tolerance, where some varieti
es perform better
than others under similar parasitic load (Cissoko
et al.,
2011). Also the most resistant varieties
can also have well developed
Striga
plants on their root system. Some of the most resistant
varieties were also less affected by
Strig
a i.e. Blechai, SCRID090 was 6.5% and 12.8% less
than the biomass of uninfected control plants compared to Superica and IAC165 whose
biomass was 74.1% and 60.2% less than their respective control plants. This study also
indicated genetic variation for tole
rance to
Striga
among rice varieties. For example varieties
CG14, WAB880 with lower
Striga
attached on the host roots had a high reduction in rice
biomass 33.3% and 54.2% respectively.
This can only indicate that small numbers of parasites
attached on thes
e varieties in the field can reduce the yield.
However varieties NERICA
–
1 and
NERICA
–
17 with a high parasitic number had less reduction in rice biomass 1.9% and 6.1%
respectively compared to uninfected control plants.
This certifies reduced effect of
Stri
ga
on
these varieties irrespective of the high parasite numbers.
This concept can therefore be
exploited such that tolerance genes can be incorporated in the resistant varieties.
101
From the results of the whole study, the following conclusions can be d
rawn
I.
Rice varieties that combine high
Striga
resistance/tolerance levels and yield such as
SCRID090, WAB800, BLECHAI, WAB181
–
18, and NERICA
–
17,
–
2,
–
4,
–
10
,
–
1
can be
incorporated in integrated
Striga
management packages to improve rice production in
these a
reas.
II.
Rice varieties that were selected by farmers such as WAB880, SCRID090, NERICA
–
17, WAB181
–
18 and Blechai can be recommended for dissemination in all
Striga
prone
areas in Uganda. This
is
because these varieties also had an excellent resistance and
hi
gh yields both under field and controlled conditions.
III.
Use of invitro methods based on a multiple mechanisms either pre
–
attachment /post
–
attachment resistance is important in selection of resistance for
Striga
amongst a large
accessions of rice.
IV.
Understan
ding the tolerance genes in rice varieties such as WAB56
–
50, NERICA
–
1
and
NERICA
–
17
can be important to incorporate these genes in the most resistance varieties
such as WAB880 and CG14 to improve their yielding potential.
V.
The
O.glabberima
varieties Anakila, ACC, AGEE, CG14 and Makassa with high yields
under
Striga
prone areas can be improved by incorporating lodging resistance genes to
improve their suitability among
Striga
prone areas.
VI.
Varieties WAB928, WAB935, IRGC and IR49 with lower yie
ld potential despite the
excellent resistance levels to
Striga
can also be improved or crossed with other
susceptible varieties to improve on these varieties.
102
REFERENCES
Abbasher, A.A., Kroschel, J., Sauerborn, J., 1995.
Micro
–
organisms of
Striga
hermonthica
in northern Ghana with potential as bio
–
control agents. Bio
–
control Science and
Technology 5, 157
–
161.
Abunyewa, A.A. & Padi, F.K. 2003.
Changes in soil fertility and
Striga
hermonthica
prevalence associated with legume and cereal cultivation in the Sudan savannah zone
of Ghana. Land Degradation and Development 14, 335
–
343.
Adagba, M.A., Lagoke, S.T.O., Imolehin, E.D., 2002a.
Nitrogen effect on the incidence of
Striga
hermonthica
(Del.)
Benth in upland rice. Acta Agronomica Academiae
Scientiarum Hungaricae 50, 145
–
150.
Addison, M., Edusah, S. E., & Sarfo
–
Mensah, P. 2014.
Gender Constraints and Rice Varietal
Characteristics Preferences in Lowland Rice Ecosystem in Ghana.
Developing Country
Studies
,
4
(15), 92
–
98.
Adekenle, W. (2013).
Improving smallholder incomes through intensification of upland rice.
A
case study produced by WRENmedia, funded by the Swiss Agency for Development and
Cooperation (SDC) and implemented by the European
Initiative on Agricultural
Research for Development (EIARD)
.
Africa Rice Center (WARDA). 2009.
The growing NERICA Boom in Uganda. WARDA
Publications: Cotonou, Benin
Ahmed, N.E., Sugimoto, Y., Babiker, A.G.T., Mohamed, O.E., Ma, Y., Inanaga, S.,
Nakajima, H
., Ma, Y.Q., 2001
. Effects of Fusarium solani isolates and metabolites on
Striga
germination. Weed Science 49, 354
–
358.
Ahonsi, M.O., Berner, D.K., Emechebe, A.M., Lagoke, S.T., 2004.
Effects of ALS inhibitor
Analytical Report February 2007
and Tanzania. A
sp. Appl. Biol. 75, 127
–
138.
Assuero, S. G., & Tognetti, J. A. 2010.
Tillering regulation by endogenous and environmental
factors and its agricultural management.
Am. J. Plant Sci. Biotechnol
,
4
, 35
–
48.
103
Atera, E. A., Itoh, K., & Onyango, J. C. 2011.
Evaluation of ecologies and severity of Striga
weed on rice in sub
–
Saharan Africa.
Agriculture and Biology Journal of North
America
,
2
(5), 752
–
760.
Atera, E. A., Itoh, K., Azuma, T., & Ishii, T. 2012.
Response of NERICA rice to Striga
hermonthica infectio
ns in western Kenya.
International Journal of Agriculture and
Biology
,
14
(2), 271
–
275.
Baguma,
1996.
Striga
research activities in Uganda, 1994
–
1995.
Striga
Newsletter, 6, 5
–
6.
Becker, M., Johnson, D.999
. The role of legume fallows in intensified upland
rice based
systems of West Africa. Nutrient Cycling in Agro
–
ecosystems 53, 71
–
81
Berner, D. K., Kling, J. G., and Singh, B. B. 1995.
Striga
research and control
–
a perspective
from Africa. Plant Disease 79:652
–
660.
Berner, D. K., Winslow, M. D., Awad, A. E.
, Cardwell, K. F., Raj, D. M., & Kim, S. K.
1997
. Striga research methods.
Manual, The Pan
–
African Striga Control Network
(PASCON) and the International Institute of Tropical Agriculture,
13
–
20.
Bouwmeester, H.J., Matusova
, R., Sun, Z.K., Beale, M.H., 2003.
Secondary metabolite
signaling in host
–
parasitic plant interactions. Current Opinion in Plant Biology 6, 358
–
364.
Carsky, R.J., Singh, L., Ndikawa, R., 1994a.
Effect of herbicide and hand weeding on current
and subsequent season
Striga
hermonthica
density on sorghum. International Journal
of Pest Management 40, 111
–
116.
Cechin, I., & Press, M. C. 1994.
Influence of nitrogen on growth and photosynthesis of a C3
cereal, Oryza sativa, infected with the root hemi
–
parasite
Striga hermonthica
.
Journal
of Experimental Botany
,
45
(7), 925
–
930.
Chang M and Lynn DG, 1986.
Houstoria and the chemistry of host recognition in parasitic
angiosperms.
Journal of Chemistry and
Ecology
12:561
–
579
104
Cissoko M, Boisnard A, Rodenburg J, Pres M.C, Scholes J.D. 2011.
New Rice for Africa
(NERICA) cultivars exhibit different levels of post
–
attachment resistance against the
parasitic weeds
Striga
hermonthica
and
Striga
asiatica
. New
Phytologist
192
: 952
–
963.
Clark, L. J., Shawe, K. G., Hoffmann, G., & Stewart, G. R. 1994.
The effect of
Striga
hermonthica
(Del.) Benth. infection on gas
–
exchange characteristics and yield of a
sorghum host, measured in the field in Mali.
Journal of Expe
rimental Botany
,
45
(2),
281
–
283.
Cruz, P., & Boval, M. 2000.
Effect of nitrogen on some morphogenetic traits of temperate and
tropical perennial forage grasses.
Lemaire, G.; Hodgson, J
, 151
–
168.
Delassus, M., 1972.
Methods of controlling
Striga
spp.
Agronomie
Tropicale 27, 249
–
254.
Dorward, P., Craufurd, P., M
arfo, K., Dogbe, W., & Bam, R. 2007
.
Improving participatory
varietal selection processes: participatory varietal selection and the role of informal
seed diffusion mechanisms for upland rice in Ghana.
Euphytica
,
155
(3), 315
–
32
Drennan, D.S.H., El Hiweris, S.O., 1979.
Changes in growth regula
tor substances in Sorghum
vulgare infected with
Striga
hermonthica.
In: Musselman, L.J., Worsham, A.D., Eplee,
R.E. (Eds.), Proceedings of the Second International Symposium on Parasitic Weeds,
pp. 144
–
155.
Efisue, A., Tongoona, P., Derera, J., Langyi
ntuo,
A., Laing, M., & Ubi, B. 2008
.
Farmers'
Perceptions on rice varieties in Sikasso region of Mali and their implications for rice
breeding.
Journal of Agronomy and Crop Science
,
194
(5), 393
–
400.
El Hiweris, S.O., 1987.
Nature of resistance to
Striga
–
hermont
hica
(Del.) Benth. Parasitism in
some Sorghum vulgare (Pers.) cultivars. Weed Research 27, 305
–
312.
Eplee, R. and Langston, M. 1991
. Inter
grated control methodology for
Striga
in the USA. In:
Ransom, J.K.,
Musselman, L.J., Worsham, A.D.
and Parker, C.
(Eds.), Proceedings of
the 5
th
international symposium of parasitic weeds, Nairobi: CIMMYT. Pp 397
–
399.
FAO STAT, 2010
(
http://faostat.fao.org
).
105
Fisher, E., Raschke, K., and Stitt, M. 1986.
Effects of Abscisic acid on
photosynthesis in
whole leaves: changes in carbondioxide assimilation levels of carbon
–
reduction
–
intermediates and activity of ribulose 1
–
5
–
phosphate carboxylase Planta 169,536
–
545.
Frost, D.L., Gurney, A.L., Press, M.C., Scholes, J.D., 1997.
Striga
herm
onthica
reduces
photosynthesis in sorghum: the importance of stomatal limitations and a potential role
for ABA. Plant Cell and Environment 20, 483
–
492.
Gebisa Ejeta and Jonathan Gressel. 2007.
In
tegrating new technologies for
Striga
control:
towards ending
the witch hunt; World Scientific Publishing Co.Pte
–
Ltd.
Gerakis, A., & B
aer, B. 1999
.
A computer program for soil textural classification.
Soil Science
Society of America Journal
,
63
(4), 807
–
808.
Gitau, R., Mburu, S., Mathenge, M. and M. Smale. 2011.
Trade and agricultural
competitiveness for growth, food security and poverty reduction: A case study of wheat
and rice production in Kenya. WPS 45/2011. Tegemeo Institute of Agricultural Policy
and Development. Nairobi, Kenya. Pp. 69.
Graves JD, Press MC,
Stewart GR. 1989.
A carbon balance model of the sorghum
–
Striga
hermonthica host
–
parasite association. Plant, Cell Environment 12: 101
–
107.
Greathead, D.J. and Milner, J.E.D. 1971.
A survey of
Striga
species (Scrophulariaceae) and
their insect natural enemies in East Africa with discussion of the possibilities of
biological control. Tropical Agriculture, 48: 111
–
124
Gridley, H. E., Jon
es, M. P., & Wopereis
–
Pura, M.
2002, March.
Development of new rice
for Africa (NERICA) and participatory varietal selection. In
breeding rain
fed rice for
drought
–
prone environments: integrating conventional and participatory plant
breeding in South and Southeast Asia: proceedings of a DFID Plant Sciences Research
Programme
/IRRI Conference
(pp. 12
–
15).
Gurney, A. L., Press, M. C., & Ransom, J. K. 1995.
The parasitic angiosperm Striga
hermonthica can reduce photosynthesis of its sorghum and maize hosts in the
field.
Journal of Experimental Botany
,
46
(12), 1817
–
1823.
106
Gurney, A. L., Press, M. C., & Scholes, J. D. 1999.
Infection time and density influence the
response of sorghum to the parasitic angiosperm Striga hermonthica.
New
Phytologist
,
143
(3), 573
–
580.
Gurney, A.L., Slate, J., Press, M
.
C., Scholes, J.D., 2006.
A novel form of resistance in rice
to angiosperm parasite
Striga
hermonthica.
New Phytologist 169, 199
–
208
Harahap, Z., Ampong
–
Nyarko, K., & Olela, J. C. 1993
. Striga hermonthica resistance in
upland rice.
Crop Protection
,
12
(3), 229
–
231.
Haussmann, B. I
., Hess, D. E., Welz, H. G., & Geiger, H. H. 2000
. Improved methodologies
for breeding
Striga
–
resistant sorghums.
Field Crops Research
,
66
(3), 195
–
211.
Hess, D.E., Ejeta, G., Butler, L.G., 1992
. Selecting sorghum genotypes expressing a
quantitative biosynthetic trait that confers resistance to
Striga
hermonthica.
Photochemistry 31, 493
–
497
Huang, K., Whitlock, R., Press, M. C., & Scholes, J. D. 2012.
Variation for host range within
and among pop
ulations of the parasitic plant
Striga hermonthica
.
Heredity
,
108
(2), 96
–
104.
Ishii, R. 1995.
Roles of photosynthesis and respiration in the yield
–
determining process. In:
Science of the Rice Plant: Physiology, Vol. 2, eds. T. Matsuo, K. Kumazawa, R.
Ishii,
K. Ishihara, and H. Hirata, pp. 691
–
703. Tokyo: Food and Agricultural
Policy
Research Center
.
Ivens G.W. 1989.
East A
frican weeds and their control
2
nd
Edition, Oxford University press
Jamil M, Rodenburg J, Charnikhova T, Bouwmeester HJ. 2011
. Pre
–
attachment
Striga
hermonthica
resistance of NERICA cultivars based on low strigolactone production.
New Phytologist
192
: 964
–
975
Jamil, M., Charnikhova, T., Houshyani, B., van Ast, A., & Bouwmeester, H. J. 2012.
Genetic variation in strigolactone production and tillering in rice and its effect on
Striga
hermonthica
infection.
Planta
,
235
(3), 473
–
484.
107
Jimmy P, Mbeche I, 2008.
A report on farmer’s
evaluation of the
Striga
threat and potential
of push
–
pull technology in controlling
Striga
weed in upland rice in East Africa;
www.icipe.org
Johnson, D.E., Dingkuhn, M., Jones, M.P., Mahamane, M.C., 1998.
The influence of rice
plant type on the effect of weed competition on
Oryza sativa
and
Oryza glaberrim
a
.
Weed Res. 38, 207
–
216
Johnson, D.E., Riches, C.R., Diallo, R., Jones, M.J., 1997.
Striga
on rice in West Africa; crop
range and the potential of host resistance. Crop Protection 16, 153
–
157.
Johnson, D.E., Riches, C.R., Jones, M.P., Kent, R., 2000.
The potential for host resistance
to
Striga
on rice in West Africa. In: Haussmann, B.I.G., Hess, D.E., Koyama, M.L.,
Grivet, L., Rattunde, H.F.W., Geiger, H.H. (Eds.), Breeding for
Striga
Resistance in
Cereals: Proceedings of a Work
shop Held at IITA, pp.
139
–
146.
Jones, M. P., Dingkuhn, M.,
Aluko, G. K., &Semon, M. 1997
.
Interspecific Oryza sativa L.
x O. glaberrima Steud. Progenies in upland rice improvement.
Euphytica
,
94
(2), 237
–
246.
Kaewchumnong, K., & Price, A. H. 2008.
A study on the susceptibility of rice cultivars to
Striga hermonthica
and mapping of
Striga
tolerance quantitative trait loci in rice.
New
Phytologist
,
180
(1), 206
–
216.
Kanampiu, F.K., Kabambe, V., Massawe, C., Jasi, L., Friesen, D., Ransom, J.K.,
Gressel,
J., 2003.
Multi
–
site, multi
–
season field tests demonstrate that herbicide seed
–
coating
herbicide
–
resistance maize controls
Striga
spp. and increases yields in several African
countries. Crop Protection 22, 697
–
706.
Kenmore P, 2003.
Sustainable rice
production, food security and enhanced livelihoods,
in
“Rice Science: Innovations and Impact for Livelihood, pg. 27
–
34.
Khan, Z. R., Midega, C. A. O., Hassanali, A., Pickett, J. A., Wadhams, L. J., and Wanjoya,
A. 2006.
Management of witch
weed,
Striga
h
ermonthica,
and stemborers in sorghum,
Sorghum bicolor, through intercropping with green leaf desmodium, Desmodium
intortum. Int. J. Pest Manag. 52(4), 297
–
302.
108
Khush, G. S. (1997).
Origin, dispersal, cultivation and variation of r
ice. Plant Mol. Biol. 35,
25
34.
Kijima, Y.,
Otsuka, K., &Sserunkuuma, D. 2008
.
Assessing the impact of NERICA on
income and poverty in central and western Uganda.
Agricultural Economics
,
38
(3),
327
–
337.
Kijima, Y.,
Otsuka, K., &Sserunkuuma, D. 2011
.
An inquiry into constraints on a green
revolution in Sub
–
Saharan Africa: the case of NERICA rice in Uganda.
World
Development
,
39
(1), 77
–
86.
Kijima, Y., Sserunkuuma, D
., & Otsuka, K. 2006
.
How revolutionary is the “NERICA
revolution”? Evidence from Uganda
.
The Developing Economies
,
44
(2), 252
–
267.
Kim, S. K. (Ed). 1991.
Combating
Striga
in Africa. Proceedings, International Workshop
organized by IITA, ICRISAT, and IDRC, 22
–
24 August 1988. IITA, Ibadan, Nigeria.
151 pp.
Kimani, J. M., Tongoona, P
., Derera, J
., &Nyende, A. B. 2011
.
Upland rice varieties
development through participatory plant breeding.
ARPN Journal of Agricultural and
Biological Science
,
6
(9), 1990
–
6145.
Kiniry, J. R., G. McCauley, Y. Xie, and J. G. Arnold. 2001.
Rice parameters describing
crop
performance of four U. S. cultivars. Agronomy Journal 93: 1354
–
1361.
Lendzemo, V.W., Kuyper, T.W., Kropff, M.J., Ast, A., 2005
. Field inoculation with
arbuscular mycorrhizal fungi reduces
Striga hermonthica
performance on cereal crops
and has the
potential to contribute to integrated
Striga
management. Field Crop
Research 91, 51
–
61.
Linare, O.F 2002.
African rice (Oryza glaberrima) history and future potential. Proceedings
of the National Academy of Science of United States of America 99: 16360
–
163
65.
Lopez
–
Raez, J.A., Matusova, R., Cardoso, C., Jamil, M., Charnikhova, T., Kohlen, W.,
Ruyter
–
Spira, C., Verstappen, F., Bouwmeester, H., 2009.
Strigolactones: ecological
significance and use as a target for parasitic plant control.
Pest Management Scie
nce
65, 471
–
477.
109
Maiti, R.K., Ramaiah, K.V., Bisen, S.S., Chidley, V.L., 1984.
A comparative study of the
haustorial development of
Striga
asiatica (L.) Kuntze on sorghum cultivars. Annales
Botanici 54, 447
–
457
Marley, P.S., Shebayan, J., Aba D.A., Idem,
N., 2004.
Possibilities for control of
Striga
hermonthica
in sorghum (Sorghum bicolar) using neem (Azadiracthaindica) and Parkia
(Parkiabiglobosa) based products. International Journal of pest management 50, 291
–
296.
Ministry of Agriculture, Animal Industr
y and Fisheries (MAAIF). 2001.
Agicultural
Production Statistics, Ministry of Agriculture, Animal Industry and Fisheries: Kampala,
Uganda.
Ministry of Agriculture, Animal Industry and Fisheries (MAAIF). 2009.
Uganda National
Rice Development Strategy. Seco
nd draft. Ministry of Agriculture, Animal Industry
and Fisheries, Entebbe, Uganda
Ministry of Agriculture, Animal Industry and Fisheries (MAAIF). 2011.
Statistical
Abstract. Ministry of Agriculture, Animal Industry and Fisheries, Entebbe, Uganda.
Mohammed,
A., Ellicot, A., Housley, T.L., Ejeta, G., 2003.
Hypersensitive response to
Striga
infection in sorghum. Crop Science 43, 1320
–
1324.
Mohammed, K.I., Bolin, J.F., Musselman, L.J., Townsend Peterson, A., 2007.
Genetic
diversity of
Striga
and implications for control and modeling future distribution. In:
Ejeta,G., G
ressel, J. (Eds.), Integrating
New Technologies for
Striga
Control
–
Towards
ending the witch hunt World Scientific Singapore, Pp. 71
–
84
Mohammed, K.I., Musselman, L.J., Riches, C
.R., 2001.
The genus
Striga
(Scrophulariaceae)
in Africa. Annals of the Missouri Botanical Garden 88, 60
–
103
Musselman, L. J., Press, M. C., Press, M. C., & Graves, J. D. 1995.
Parasitic
plants.
Parasitic plants
.
Musselman, L.J. (Ed) 1987
. Taxonomy of witch weeds. Page 3
–
12. In: Parasitic plants in
Agriculture
Striga
, CRC Press. Boca Raton. USA 317Pp
110
Namutumba Census Report. 2007.
Namutumba District 2002 Population and Housing
Census
Nanfumba, D., Turyahabwe, N., Ssebuliba, J., Kakuru, W.,
Kaugu
le, J., Omio, S.,
&Samuka, M. 2013
.
Participatory identification of farmer acceptable improved rice
varieties for rain
–
fed 683 lowland ecologies in Uganda.
African Crop Science
Journal
,
21
(1), 683
–
692.
National planning authority (NPA). 2010
. The National Development Plan (2010/11
–
2014/15) Government of Uganda. Ministry of Finance Planning and Economic
Development,
www.htttp://npa.ug/wp
–
content/themes/npat
heme/documents/NDP2.pdf
Normile, D, 2004.
Yangtze seen as earlier rice site Science 275:309
Odhiambo, G.D., Ransom, J.K., 1994
. Long term strategies for
Striga
control. In:
Proceedings of the Fourth Eastern and Southern Africa Regional Maize Conference, 28
March
–
1 April 1994, pp. 263
–
266.
Oerke, E.C and Dehne, H.W. 2004.
Safeguarding production losses in major crops and role
of Crop protection, Crop prot.23, 275
–
285
Oikeh, S.
O., Nwilene, F. E., Agunbiade, T. A., Oladimeji, O., Ajayi, O., Mande, S
&Samejima,
H. 2008.
Growing upland rice: a production handbook.
Africa Rice
Center (WARDA)
.
Okalebo, J. R., Gathua, K. W., &Woomer, P. L. 2002.
Laboratory methods of plant and soil
analysis: a working manual.
Tropical Soil Biology and Fertility Programme, Nairobi
.
Olivier, A., Benhamou, N., Leroux, G.D., 1991.
Cell surface interactions between sorghum
roots and parasitic weed
Striga
hermonthica
: cytochemical aspects of cellulose
distribution in resistant and susceptible host tissues Canadian Journal of Botany 69,
16
79
–
1690.
Olmstead, A.,
Depamphilis, C.W., Wolfe, A.D., Young, N.D., Elisens, W.J and Reeves,
P.A. 2001.
Disintegration of Scrophulariaceae. American Journal of Botany 88, 348
–
361.
111
Ongaro, V., & Leyser, O. 2008
.
Hormonal control of shoot branching.
Journal
of
Experimental Botany
,
59
(1), 67
–
74.
Oswald A., 2004.
Striga
control
–
technologies and their dissemination; Crop Protection Journal:
24, 333
–
342.
Pageau, K., Simier, P., Le Bizec, B., Robins, R. J., &Fer, A. 2003.
Characterization of
nitrogen relationships
between Sorghum bicolor and the root
‐
hemiparasitic angiosperm
Striga hermonthica (Del.) Benth. Using K15NO3 as isotopic tracer.
Journal of
Experimental Botany
,
54
(383), 789
–
799.
Parker, C. 2012.
Parasitic weeds: a world challenge.
Weed Science
,
60
(2),
269
–
276.
Parker, C., Riches, C.R., 1993.
Parasitic Weeds of the World: Biology and Control. Cab
International, Wallingford, Oxon.
Plan for Modernization of Agriculture (PMA) Secretariat. 2009.
Rice Value Chain Study
in Acholi and Lango Sub
–
regions. Plan fo
r Modernization of Agriculture (PMA)
Secretariat. Kampala, Uganda.
Porteres, R. 1956.
Taxonomic agrobotanique der rizcultives O. sativa Linneet
O. Glaberrima
Steudelo, J. Agri.
Press, M.C and Stewart, G.R, 1987.
Growth and photosynthesis in
Sorghum bicolar
infected
with
Striga
hermonthica. Annals of Botany 60: 657
–
662.
Press, M.C., Shah, N., Tuohy, J.M., Stewart, G.R. 1987a
. Carbon isotope ratios demonstrate
carbon flux from C4 host to C3 parasite. Plant physiology 85, 1143
–
1145.
Purseglove, J.W. 1975.
Trop
ical crops; Monocotyledons vol.1 and 2. Published by Longman
Group UK Ltd.
Ramaiah, K.V., Parker, C., Vasudeva Rao, M.J., and Musselman, L.J., 1983
.
Striga
identification and control handbook. Information Bulletin No. 15. Patancheru, A.P.,
India: Interna
ti
onal crop Research Institute f
or the Semi
–
Arid Tropics
112
Ransom, J. K., &
Njoroge, J. 1991.
Seasonal variation in ethylene
–
induced germination of
Striga hermonthica
in Western Kenya. In
5. International Symposium of Parasitic
Weeds, Nairobi (Kenya), 24
–
30
Jun 1991
. CIMMYT.
Ransom, J.K., Odiambo G.D., Abayo G., and Oswald, A. 1997.
An update on
Striga
control
research in
Africa In: Adipala, E., Tusiime
, G., and Okori P (eds) Proceedings of the
16
th
Biannual Weed Science Society Conference for Eastern Africa
Kampala Pp 215
–
219.
Rich, P., Grenier, C., Ejeta G., 2004.
Striga
resistance in the wild relatives of sorghum.
Crop
Science 44, 2221
–
2229.
Riches, C.R, Johnson, D.E., Jones, M.P., 1996
.
The selection of resistance to
Striga
species
in upland rice. In: Moren
o, M.T., Cubero, J.L., Berner, D., Musselman, L.J., Parker, C.
(Eds.) Advances in Parasitic Research Proceedings of the Sixth International Parasitic
Weed Symposium Cordon Pp. 673
–
680.
Riches, C.R., Mbwaga, A.M., Mbapila, J., and Ahmed, G.J.U. 2005
.
Improved weed
management delivers increased productivity and farm incomes from rice in Bangladesh
and Tanzania Asp. Appl. Biol. 75
–
127
–
128.
Rodenburg J; Charles R. Riches; Jumam K. 2010.
Addressing current and future problems
of parasitic weeds in rice Jou
rnal of Crop Protection; 29: 210
–
221.
Rodenburg, J., Bastiaans, L., &Kropff, M. J. 2006
. Characterization of host tolerance to
Striga hermonthica
.
Euphytica
,
147
(3), 353
–
365.
Rodenburg, J., Bastiaans, L., Weltzien, E., & Hess, D. E. 2005.
How can field sel
ection for
Striga
resistance and tolerance in sorghum be improved?
Field Crops Research
,
93
(1),
34
–
50.
Rodenburg, J., Cissoko, M., Kayeke, J., Dieng, I.,
Khan, Z. R., Midega, C. A.,
&
Scholes,
J. D. 2015
.
Do NERICA rice cultivars express resistance to
Striga hermonthica
(Del.)
Benth.and
Striga asiatica
(L.) Kuntze under field conditions?.
Field Crops
Research
,
170
, 83
–
94.
113
Rodenburg, J., Demont, M., 2009.
Potential of herbicide resistant rice technologies for sub
Saharan Africa. AgBioForum 12, 313
–
325.
Rodenburg, J., Johnson, D.E., 2009.
Weed management in rice based cropping systems in
Africa. Advances in Agronomy 103, 149
–
218.
Rodenburg, J., Saito, K., Kakaï, R. G., Touré, A., Mariko, M., &Kiepe, P. 2009.
Weed
competitiveness of the lowland rice variet
ies of NERICA in the southern Guinea
Savanna.
Field Crops Research
,
114
(3), 411
–
418.
Safa, S. B., Jones, B. M. G., & Musselman, L. J. 1984.
Mechanisms favouring outbreeding
in
Striga hermonthica
[Scrophulariaceae].
New phytologist
,
96
(2), 299
–
305.
Saito, K., Azoma, K., Rodenburg, J., 2010.
Plant characteristics associated with weed
competitiveness of rice under upland and lowland conditions in West Africa. Field
Crop. Res. 116, 308
–
317
Saito, K., Sokei, Y., & Wopereis, M. C. S. 2012.
Enhancing rice productivity in West Africa
through genetic improvement.
Crop Science
,
52
(2), 484
–
493.
Sakamoto, T. 2006.
Phytohormones and rice crop yield: strategies and opportunities for genetic
improvement.
Transgenic research
,
15
(4), 399
–
404.
Samado
, E. A., Guei, R. G., & Nguyen, N. 2008.
NERICA: Origin, nomenclature and
identification characteristics.
NERICA®: The New Rice for Africa
–
a Compendium.
Africa Rice Center, Cotonou, Benin
, 10
–
19.
Saunders, A.R., 1933.
Studies in phanerogamic parasitism wit
h particular reference to Striga
LuteaLour. South Africa departments of Agriculture, Science Bulletin 128, 1
–
56.
Scholes, J. D., & Press, M. C. 2008.
Striga
infestation of cereal crops
–
an unsolved problem in
resource limited agriculture.
Current opinion in
plant biology
,
11
(2), 180
–
186.
Seck, P. A., Diagne, A., Mohanty, S., & Wopereis, M. C. 2012.
Crops that feed the world 7:
rice.
Food Security
,
4
(1), 7
–
24.
114
Shikawa, S., Maekawa, M., Arite, T., Onishi, K., Takamure, I., &Kyozuka, J. 2005.
Suppression of tiller bud activity in tillering dwarf mutants of rice.
Plant and Cell
Physiology
,
46
(1), 79
–
86.
Sokal, R.R., Rohlf, F.J., 1995.
Biometry. W.H. Freeman and Company, New York, NY.
Swarbrick, P. J., Huang, K., Liu, G., Slate, J., Press, M. C.
, & Scholes, J. D. 2008.
Global
patterns of gene expression in rice cultivars undergoing a susceptible or resistant
interaction with the parasitic plant Striga hermonthica.
New Phytologist
,
179
(2), 515
–
529.
Tatsushi Tsuboi, 2005.
Paper presented at the WAR
DA
–
NERICA rice Workshop, Ivory
Coast, 8thOctober, 2005.
Uganda Bureau of Statistics (UBOS), 2002.
Uganda National House hold Survey 1999/2000
report on the crop Survey module. Entebbe, UBOS.
Uganda Bureau of Statistics (UBOS), 2012.
Statistics Abstract 2012, Kampala UBOS
statistical house available at
www.ubos.org
(accessed May 2013).
Vallence, K.B., 1950.
Studies on the germination of the seeds of
Striga hermonthica
. The
influence of moisture trea
tment, stimulant dilution and after ripening on germination
Annales Botanici 14, 347
–
363.
Vaughan, D.A, 1994.
The wild relatives of rice; A genetic hand book, International Rice
Research Institute, Manila.
Wanyonyi, M.W., Wanyama, J.M., Ommbakho, G
.A.,
Ligeyo, D O. and Rano, S. 2008
.
Gender perception in the selection of newly released highland maize varieties.
Proceedings of the 11th KARI Biennial Scientific Conference, KARI Headquarters,
Nairobi, Kenya, 10
–
14th November 2008. Online Proceeding. http:
//
www.kari.org/fileadmin/publications/ conference11. /Gender Perception
_inthe_selection.pdfý.
Watling, J. R., & Press, M. C. 2000.
Infection with the parasitic angiosperm
Striga
hermonthica
influences the response of the C3 cereal Oryza sativa to elevated
CO2.
Global Change Biology
,
6
(8), 919
–
930.
115
Wilfred, O. 2006.
Final survey report on the status of rice production, processing and
marketing in Uganda, a report submitted to the Embassy of Japan in Uganda through
JICA and Sasakawa
Africa Association
–
Uganda pp.90
Wopereis, M. C. S., Diagne, A., Rodenburg, J., Sié, M., & Somado, E. A. 2008.
Why
NERICA is a successful innovation for African farmers: A response to Orr et al from
the Africa Rice Center.
Outlook on AGRICULTURE
,
37
(3),
169
–
176.
Worsham, A.D.1987.
Germination of witch weed seeds. Pages 45
–
62. In Musselman, L.J.
(ed.), parasitic weeds in Agriculture, Vol.1, Striga. CRC Press, Inc. Boca Raton, FL, pp
45
–
61
Xu Kuangdi and Shen Guofang
,
2003.
Promoting Chinese rice production
through
innovative science and technology, pg 11
–
18 of the Proceedings of the International
Rice Research Conference, 16
–
19 September 2002, Beijing, China.
Yoder, J.I., 2009.
Host
–
plant recognition by parasitic Scrophulariaceae. Current opinion in
Plant
Biology 4, 359
–
365.
Yoneyama, K., Xie, X., Kusumoto, D., Sekimoto, H., Sugimoto, Y., Takeuchi, Y., &
Yoneyama, K. 2007
a
. Nitrogen deficiency as well as phosphorus deficiency in
sorghum promotes the production and exudation of 5
–
deoxystrigol, the host recog
nition
signal for arbuscular mycorrhizal fungi and root parasites.
Planta
,
227
(1), 125
–
132.
Yoneyama, K., Yoneyama, K., Takeuchi, Y., & Sekimoto, H. 2007
b
. Phosphorus deficiency
in red clover promotes exudation of orobanchol, the signal for mycorrhizal
symbionts
and germination stimulant for root parasites.
Planta
,
225
(4), 1031
–
1038.
116
APPENDICES
Appendix 1: Initial soil characteristics of the field trial for season 2014 A and 2015 B in
Eastern Uganda.
Appendix 2: Analysis of variance for plant height per rice variety for 2014
–
2015 under
Striga
infested field conditions
.
***
Significantly different at 0.001
Appendix 3: Analysis of variance for the number of tillers per rice variety for 2014
–
2015
under
Striga
infested field conditions.
Source of
variation
df
43 days
57 days
113 days
Variety
24
57.991***
518.34***
320.23***
Seasons
1
976.335***
750.70***
503.83***
Variety x
Seasons
24
11.219***
40.25***
26.66***
Residual
220
4.152
16.14
11.30
***
Significantly different at 0.001
Season
PH
N (%)
P
(ppm)
K
(ppm)
OC
(%)
S
and:
Silt: C
lay
Na
(ppm)
Mg
(ppm)
Ca
(ppm)
2014A
6.1
0.061
7.41
135
2.31
71:9:20
63
146
977
2015B
6.1
0.095
6.72
174
1.22
66:14:20
36
125
890
Source of
variation
df
43 days
57 days
113 days
Variety
24
358.604***
709.73***
3341.3***
Seasons
1
6386.327***
1415.59***
1377.7***
Variety x Seasons
24
71.167***
114.65***
434.7***
Residual
220
7.933
17.40
101.5
117
Appendix 4: Analysis of variance for the yield components of rice varieties for 2014 and
2015 seasons under
Striga
infested conditions.
Source of variation
df
Straw dry
weight (g)
Panicle dry
weight (g)
Rice grain
recovery
(%)
Harvest
index
Grain dry
weight (t
/ha)
Variety
24
26214***
42129
***
1104.99***
0.2515
***
14.1747
***
Seasons
1
42ns
13480*
706.34*
0.2524
***
20.9456
***
Variety x Seasons
23
4247***
9391
***
116.69*
0.0216
***
2.4992
***
Residual
215
1582
3572
81.93
0.0071
0.7861
***
Significantly different at 0.001
,
*
significantly different at 0.05
,
ns
–
not significantly different
Appendix 5: Analysis of variance of number of
Striga
plants
of
rice varieties for 2014
–
2015 in Namutumba district Eastern Uganda
Source of variation
df
.
43 days
57 days
71 days
85 days
At harvest
Variety
24
1.265***
2.490***
3.142***
3.325***
3.406***
Seasons
1
3.280***
11.394***
6.254***
2.924***
1.992***
Variety x Seasons
24
0.158***
0.194**
0.198**
0.145ns
0.181*
Residual
220
0.065
0.091
0.101
0.095
0.114
*
Significantly different at 0.05,
**
significantly different at 0.01,
***
significantly
different at 0.001
Appendix 6: Analysis of variance of
Striga
components recorded per rice variety for
2014
–
2015 in Namutumba district Eastern Uganda.
Source of variation
df
NSmax
Striga
dry weight (g)
Variety
24
3.360***
3.011***
Seasons
1
1.254***
3.464***
Variety x Seasons
24
0.121ns
0.109ns
Residual
220
0.094
0.124
***
Significantly different at 0.001
,
ns
–
not significantly different
,
NSmax
–
maximum number of aboveground Striga plants
118
Appendix 7: Analysis of variance of days to
Striga
emergency and
Striga
flowering of rice
varieties for 2014
–
2015 in
Namutumba district Eastern Uganda.
***
Significantly different at 0.001
, SED
–
Striga emergency days, SFD
–
Striga flowering days.
Appendix 8: Analysis of variance of Number of tillers and plant height rice varieties in
2014 in Namutumba district Eastern Uganda.
***
Significantly
different at 0.001
,
*
significantly different at 0.05
Appendix 9: Analysis of variance of yield components of rice varieties in 2014 in
Namutumba district Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
Source of variation
df
SED (days)
Df
SFD (days)
Variety
24
1290.00***
24
317.61***
Seasons
1
3785.46***
1
10028.17***
Variety x Seasons
24
234.28***
22
126.05***
Residual
172
90.99
122
46.91
Source of
variation
df
43 days
57 days
113 days
Plant height
Variety
24
287.27***
530.92***
1236.60***
Residual
95
12.02
17.92
60.83
Number of
tillers
Variety
24
24.34***
286.46***
106.93***
Residual
95
7.49
20.74
7.09
Source of variation
df
Straw dry
weight (g)
Panicle dry
weight (g)
Rice grain
recove
ry
(%)
Harvest
index
Grain dry
weight (t/ha)
Variety
23
7508***
145042***
56.83***
0.064***
4.62***
Residual
82
1179
15463
76.81
0.007
0.45
119
Append
ix 10: Analysis of variance of n
umber of tillers and plant height rice varieties in
2015 in Namutumba district Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
Appendix 11: Analysis of variance of yield comp
onents of rice varieties in 2015
in
Namutumba district Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
Appendix 12: Analysis of variance of number of
Striga
plants rice varieties for 2014 and
2015 in Namutumba district Eastern Uganda
***
Significantly different at 0.001
,
*
significantly different at 0.05
Source of
variation
df
43 days
57 days
113 days
Plant height
Variety
24
86.48***
219.40***
2174.1***
Residual
95
4.09
11.85
121.1
Number of
tillers
Variety
24
13.59***
203.58***
210.04***
Residual
95
1.091
10.35
9.10
Source of variation
df
Straw dry
weight (g)
Panicle dry
weight (g)
Rice grain
recovery
(%)
Harvest
index
Grain dry
weight (t/ha)
Variety
24
20438***
221899***
575.54***
0.104***
6.55***
Residual
95
1533
24592
85.06
0.007
0.73
Source of variation
df.
43 days
57 days
71 days
85 days
At harvest
2014
Variety
24
0.317***
0.709***
1.348***
1.456***
1.785***
Residual
96
0.050
0.061
0.076
0.073
0.095
2015
Variety
24
1.016***
1.692***
1.591***
1.626***
1.457***
Residual
95
0.074
0.125
0.128
0.102
0.105
120
Appendix 13: Analysis of variance of
Striga
components recorded per rice variety in 2014
in Namutumba district Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
,
**
significantly different at 0.01
, NSmax
–
maximum
number of above ground Striga plants.
Appendix 14: Analysis of variance of
Striga
components recorded per rice variety in 2015
in Namutumba district Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
, NSmax
–
maximum number of above ground Striga
plants
Appendix 15: Analysis of
variance for growth parameters of rice varieties grown under
controlled conditions.
***
Significantly different at 0.001
,
*
significantly different at 0.05
,
**
significantly different at
0.01
, ns
–
not significantly
different
Source of variation
df
S
triga
emergency days
Striga
flowering
days
NSmax
(m
2
)
Striga
dry
weight (g)
Variety
24
681.1*
2575**
1.545***
1.272***
Residual
96
420.3
1103
0.733
0.106
Source of variation
df
S
triga
emergency days
Striga
flowering
days
NSmax
(m
2
)
Striga
dry
weight (g)
Variety
24
623.5***
166.04***
1.573***
1.511***
Residual
95
128.1
59.07
0.103
0.116
Source of variation
df
Plant height
(cm)
Stem
diameter
(mm)
Tiller
number
Leaf
number
Rice
biomass (g)
Variety
13
27.59***
0.130*
2.382***
28.28***
0.231***
Treatment
1
113.88***
2.099***
5.143**
36.30**
1.366***
Variety x treatment
13
13.95***
0.093*
0.489ns
2.943ns
0.156**
Residual
84
4.68
0.071
0.639
4.872
0.070
121
Appendix 16: Analysis of variance for
Striga
number and
Striga
dry weight of rice
varieties grown under controlled conditions
.
Source of variation
d
f
.
Striga
number
Striga
dry weight
Variety
13
0.984***
0.892***
R
esidual
53
0.112
0.074
***
Significantly different at 0.001
,
*
significantly different at 0.05
Appendix 17: Questionnaire for participatory variety selection
The questionnaire is intended for upland rice farmers in Namutumba district. It is intended to
establish variety preferences and to assess the criteria of selection used by farmers in the
district.
Date:
……………………………………………………………………………………………………
Enumerator:
………………..
………………………………………………………………………..
1. Name of farmer:
…………………………………………………………………………………
2. Age:
…………………………………………..
……………………………………………………..
3. Sex:
………………………………………………………………………………………………….
4. Village (origin):
……………………………………………
……………………………………..
5. For how long have you grown upland rice?
…………………………………………………………………………………………………………….
…………………………………………………………………………………………………………….
122
6
. Which rice variety/ies from the demonstration plot would you wish to sow in your
exploitation next year and why (see variety list under Annex 1)?
Variety (name and
number from Annex 1)
Reasons of choices
Detailed explanations
Score (1
–
10)*
1.
1.
2.
3.
1.
2.
3.
2.
1.
2.
3.
1.
2.
3.
3.
1.
2.
3.
1.
2.
3.
4.
1.
2.
3.
1.
2.
3.
5.
1.
2.
3.
1.
2.
3.
123
7
. Which varieties according to you are NOT good or which varieties do you not wish to sow
in your exploitation and why
(see variety list under Annex 1)
?
Variety (name and
number from Annex 1)
Why don’t you like those varieties
Detailed explanations
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
1.
2.
3.
8
. What are the most important characters for you when you must choose rice varieties
to be
sown in your exploitation?
Give a score to each character.
Character
Score (1, 2, 3, 4, 5)
*
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
124
*
1
= not
important
at all
; 2
=
not
important; 3
= more or less important; 4
=
important
; 5
=
very
important
9
. Can you evaluate at witch level these characters listed
under Question 13 are expressed by
the varieties you have just selected under Question 11? Put ‘’X’’ in the column ‘’a, b or c’’
which corresponds to the choice for each variety.
Variety names and numbers (see list Annex 1)
1.
2.
3.
4.
5.
Character
A
b
c
a
b
C
a
b
c
a
b
c
a
b
c
1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
a= very good; b= good/sufficient;
c= bad
*** END***
125
Annex 1. List of varieties screened in Uganda (2014
–
2015)
Variety name
Short name
Variety number
ACC102196
ACC
1
Agee
Agee
2
Anakila
Anakila
3
WAB56
–
50
WAB50
4
CG14
CG14
5
IAC165
IAC
6
IR49255
–
B
–
B
–
5
–
2
IR49
7
IRGC78281 (Blue
Chai)
BleChai
8
IRGC81712 (IR 38547
–
B
–
B
–
7
–
2
–
2)
IRGC
9
Makassa
Makassa
10
MG12
MG12
11
NARC3 (ITA257)
NARC3
12
NERICA
–
1
N1
13
NERICA
–
10
N10
14
NERICA
–
17
N17
15
NERICA
–
2
N2
16
NERICA
–
4
N4
17
SCRID090
–
60
–
1
–
1
–
2
–
4
SCRID090
18
Superica1 (WAB165)
Superica
19
UPR
–
103
–
80
–
1
–
2
UPR
20
WAB 181
–
18
WAB181
21
WAB 56
–
104
WAB104
22
WAB 928
–
22
–
2
–
A
–
A
–
B
WAB928
23
WAB 935
–
5
–
A
–
2
–
A
–
A
–
B
WAB935
24
WAB880
–
1
–
32
–
1
–
1
–
P2
–
HB
–
1
–
1
–
2
–
2
WAB880
25
126
Appendix 18: Rain fall data for Namutumba district in Eastern Uganda for 2014 and
2015
0
50
100
150
200
250
300
350
2014
2015
Rainfall (mm)
Seasons
J
F
M
A
M
J
J
A
S
O
N
D
Data source: Tororo
district meteorological station Eastern Uganda
Copyright Notice
© Licențiada.org respectă drepturile de proprietate intelectuală și așteaptă ca toți utilizatorii să facă același lucru. Dacă consideri că un conținut de pe site încalcă drepturile tale de autor, te rugăm să trimiți o notificare DMCA.
Acest articol: IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF THE DEGREE OF MASTER OF SCIENCE IN CROP SCIENCE OF MAKERERE UNIVERSITY MAY 2016 i… [601795] (ID: 601795)
Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.