IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF THE DEGREE OF MASTER OF SCIENCE IN CROP SCIENCE OF MAKERERE UNIVERSITY AUGUST 2016 i… [600713]
V
ARIETY SCREENING FOR
STRIGA
HERMONTHICA
RESISTANCE IN UPLAND
RICE IN EASTERN UGANDA
BY
KAYONGO
NICHOLAS
BSc.
(
Horticulture
) Hons. MAK
2012/HD02/105
U
208006518
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
AUGUST
2016
i
DECLARATION
I
,
Kayongo
Nicholas
,
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
a
cademic 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
P
roduction
Makerere University, Kampala
Signed……………………………………………… Date………………………………………
DR. JAMES
M. SSEBULIBA
Depart
ment of Agricultural
P
roduction
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 and 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 Sci
ences,
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 C
issoko,
both of
Africa Rice
center, Tanzania
,
for providing
me with the Rice germplasm
, helping with
the
trial setup and
for
all the guidance during writing
of this thesis.
Thank you so much for
you
r help, I would not have
made it without your help.
I wou
ld 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
to
me
during my Laboratory experiments
in the UK
. 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 assistanc
e rendered to
me during the course of this research.
iv
TABLE OF CONTENTS
DECLARATION
…………………………..
…………………………..
…………………………..
……………………….
i
DEDICATION
…………………………..
…………………………..
…………………………..
………………………….
ii
ACKNOWLEDGEMENTS
…………………………..
…………………………..
…………………………..
……..
ii
i
LIST OF TABLES
…………………………..
…………………………..
…………………………..
……………………
x
LIST OF FIGURES
…………………………..
…………………………..
…………………………..
………………..
xii
LIST OF APPENDICES
…………………………..
…………………………..
…………………………..
………..
xiii
LIST OF ACRONYMS
…………………………..
…………………………..
…………………………..
…………..
xv
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 productio
n in Uganda
…………………………..
…………………………..
…………………………..
….
2
1.4 Problem statement
…………………………..
…………………………..
…………………………..
……………..
4
1.5 Justification
…………………………..
…………………………..
…………………………..
………………………
5
1.6 Objectiv
es of the study
…………………………..
…………………………..
…………………………..
……….
6
1.7 Hypotheses
…………………………..
…………………………..
…………………………..
……………………….
7
CHAPTER TWO
…………………………..
…………………………..
…………………………..
……………………..
8
2.0 LITERATURE REVIEW
…………………………..
…………………………..
…………………………..
……
8
v
2.1 Majo
r 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
…………………………..
…………………………..
…………………………..
……….
13
2.1.1.5 Nature of
Striga
damage
…………………………..
…………………………..
………………………….
13
2.1.1.6 Control of
Striga
…………………………..
…………………………..
…………………………..
…………
14
2.1.1.6.1 Cultural control
…………………………..
…………………………..
…………………………..
……….
15
2.1.1.6.1.1 Hand weeding
…………………………..
…………………………..
…………………………..
……
15
2.1.1.6.1.2 Crop rotation and trap crops
…………………………..
…………………………..
…………….
15
2.1.1.6.1.3 Use of fertilizers
…………………………..
…………………………..
…………………………..
..
16
2.1.1.6.2 Chemical control
…………………………..
…………………………..
…………………………..
……..
16
2.1.1.6.3 Use of resistant varieties
…………………………..
…………………………..
……………………….
17
2.2 Participatory variety selection
…………………………..
…………………………..
………………………..
20
CHAPTER THREE
…………………………..
…………………………..
…………………………..
………………..
22
3.0 MATERIALS AND METHODS
…………………………..
…………………………..
…………………….
22
3.1 Study one: Resistance of upland rice varieties against
Striga hermonthica
under field
conditions
…………………………..
…………………………..
…………………………..
…………………………..
…..
22
3.1.1 Experimental site
…………………………..
…………………………..
…………………………..
…………..
22
vi
3.1.2 Rice varieties used
…………………………..
…………………………..
…………………………..
…………
22
3.1.3 Experimental design and Layout
…………………………..
…………………………..
………………….
24
3.1.4 Cultural practices
…………………………..
…………………………..
…………………………..
…………..
26
3.1.5 Data collection
…………………………..
…………………………..
…………………………..
………………
27
3.1.5.1 Soil data
…………………………..
…………………………..
…………………………..
…………………….
27
3.1.5.2 Crop data
…………………………..
…………………………..
…………………………..
…………………..
28
3.1.5.3
Striga
data
…………………………..
…………………………..
…………………………..
…………………
29
3.1.6 Data analysis
…………………………..
…………………………..
…………………………..
………………..
29
3.2 Study two: Farmer participatory variety selection of upland rice varieties
……………….
30
3.2.1 Data collection
…………………………..
…………………………..
…………………………..
………………
30
3.2.2 Data analysis
…………………………..
…………………………..
…………………………..
………………..
31
3.3 Stud
y three: Evaluation of post
–
germination attachment resistance of upland rice
varieties to
Striga hermonthica
under controlled environment conditions
……………………….
31
3.3.1. Plant materials
…………………………..
…………………………..
…………………………..
……………..
31
3.3.2 Growth and infection of rice plants with
Striga hermonthica
…………………………..
……….
32
3.3.3 Data collection
…………………………..
…………………………..
…………………………..
………………
33
3.3.3.1 Growth measurements
…………………………..
…………………………..
…………………………..
…
33
3.3.3.2 Above ground rice biomass
…………………………..
…………………………..
………………………
33
3.3.3.3 Quantification of post
–
germination attachment resistance of the rice cultivars
…………
33
3.3.3.4 The phenotype of resistance
…………………………..
…………………………..
……………………..
34
vii
3.3.4 Statistical analysis
…………………………..
…………………………..
…………………………..
…………
34
CHAPTER FOUR
…………………………..
…………………………..
…………………………..
……………….
35
4.0 RESULTS AND DISCUSSION
…………………………..
…………………………..
………………………
35
4.
1 Study one: Resistance of upland rice varieties against
Striga hermonthica
under field
conditions
…………………………..
…………………………..
…………………………..
…………………………..
…..
35
4.1.1 Soil characteristics
of the field trial.
…………………………..
…………………………..
……………..
35
4.1.2 The growth parameters of upland rice varieties grown under
Striga
infested fields.
……….
36
4.1.2.1 Plant height per variety
…………………………..
…………………………..
…………………………..
.
36
4.1.2.2 Number of tillers
…………………………..
…………………………..
…………………………..
………..
39
4.1.2.3 Rice biomass
…………………………..
…………………………..
…………………………..
……………..
43
4.1.3 Grain yield and yield components of upland rice varieties grown under
Striga hermonthica
infested field conditions.
…………………………..
…………………………..
…………………………..
……………
46
4.1.3.1 Panicle dry weight
…………………………..
…………………………..
…………………………..
………
46
4.
1.3.2 Harvest index
…………………………..
…………………………..
…………………………..
……………..
49
4.1.3.3 Grain recovery
…………………………..
…………………………..
…………………………..
……………
52
4.1.3.4 Rice grain yield
…………………………..
…………………………..
…………………………..
………….
54
4.1.
4 Reaction of upland rice varieties to
Striga hermonthica
…………………………..
…………………
58
4.1.4.1
Striga
counts per variety
…………………………..
…………………………..
………………………….
58
4.1.4.2 Days to
Striga
emergence and flowering
…………………………..
…………………………..
……
62
4.1.4.3 Maximum number of emerged
Striga
plants (Nsmax)
…………………………..
………………
65
viii
4.1.4.4
Striga
dry weight
…………………………..
…………………………..
…………………………..
………..
68
4.
2 Study two: Farmer participatory variety selection of upland rice varieties
………………
70
4.2.1 Social demographic factors and Rice
farming practices
…………………………..
…………..
70
4.2.2 Rice variety selection by farmers basing on morphological appearance in the field
…
72
4.2.4 Characteristics of the least preferred varieties of rice according to farmers.
……………
75
4.2.5 Characteristics of preferred rice varieties according to farmers
…………………………..
…
76
4.2.6 Variety ranking based on farmer selection criteria
…………………………..
…………………..
78
4.3 Study three: Evaluation of post
–
germination attachment resistance of upland rice
varieties to
Striga hermonthica
under controlled environment conditions
……………………….
81
4.3.1 Effect of
Striga
infection on the morphology and growth characteristics of the host rice
plants.
…………………………..
…………………………..
…………………………..
…………………………..
…………
81
4.3.1.1 Effect on plant height
…………………………..
…………………………..
…………………………..
81
4.3.1.2 Effect on stem diameter
…………………………..
…………………………..
………………………..
84
4.3.1.3 Effect on number of tillers
…………………………..
…………………………..
…………………….
86
4.3.1.4 Effect on number of leaves
…………………………..
…………………………..
……………………
89
4.3.3 Post
–
germination attachment levels of
Striga hermonthica
to selected rice varieties.
……..
91
4.3.3.1 Quantification of post
–
germination attachment of
Striga
.
…………………………..
………..
91
4.3.3.2
Striga
hermonthica
dry weight
…………………………..
…………………………..
………………….
95
4.3.4 Phenotype of resistance of selected rice var
ieties against
Striga hermonthica
(Namutumba
ecotype)
…………………………..
…………………………..
…………………………..
…………………………..
…..
96
ix
4.3.5 Effect of
Striga hermonthica
on above ground rice biomass of upland rice varieties with
respect to the uninfected rice plants.
…………………………..
…………………………..
…………………….
98
CHAPTER FIVE
…………………………..
…………………………..
…………………………..
………………….
103
5.0 GENERAL DISCUSSION
…………………………..
…………………………..
…………………………..
.
103
CHAPTER SIX
…………………………..
…………………………..
…………………………..
…………………….
110
6.0 CONCLUSION A
ND RECOMMENDATIONS
…………………………..
…………………………
110
6.1 CONCLUSION
…………………………..
…………………………..
…………………………..
……………..
110
6.2 RECOMMENDATIONS
…………………………..
…………………………..
…………………………..
..
112
REFERENCES
…………………………..
…………………………..
…………………………..
……………………..
113
APPENDICES
…………………………..
…………………………..
…………………………..
………………………
133
x
LIST OF TABLES
Table 1: List of rice varieties screened for
Striga
resistance in Nsinze, Namutumba District,
Uganda and their characteristics.
…………………………..
…………………………..
…………………………..
..
23
Table 2: Plant height of upland rice varieties gro
wn under
Striga hermonthica
infested field
conditions during the first rainy season of 2014 and 2015 at 43, 57 and 113 days after sowing.
37
Table 3: Number of tillers of upland rice varieties grown under
Striga hermonthica
infested field
conditions during the first rainy season of 2014 and 2015 at 43, 57 and 113 days after sowing.
40
Table 4: Plant biomass of upland rice varieties grown under
Striga hermonthica
infested field
conditions in Namutumba, Eastern Uganda, 2014 and 2015.
…………………………..
…………………..
44
Table 5: Panicle dry weight of upland rice varieties grown under
Striga hermonthica
infested field
conditions in Namutumba, Eastern Uganda, 2014 and 2015.
…………………………..
…………………..
47
Table 6: Harvest index of upland rice varieties grown under
Striga hermonthica
infested field
conditions in Namutumba, Eastern
Uganda, 2014 and 2015.
…………………………..
…………………..
50
Table 7: Grain recovery efficiency of upland rice varieties grown under
Striga hermonthica
infested fi
eld conditions in Namutumba, Eastern Uganda, 2014 and 2015.
…………………………..
.
53
Table 8: Grain yield of upland rice varieties grown under
Striga
hermonthica
infested field
conditions in Namutumba, Eastern Uganda, 2014 and 2015.
…………………………..
…………………..
55
Table 9: Number of
Striga
plants (
Striga
cou
nts) per rice variety under
Striga
infested field
conditions in Namutumba, Eastern Uganda, 2014 and 2015.
…………………………..
…………………..
60
Table 10:
Striga
com
ponents per upland rice variety in 2014 and 2015, Namutumba district,
Eastern Uganda.
…………………………..
…………………………..
…………………………..
……………………….
63
Table 11: Characterization of upland
rice farmers’ interms of gender, age and rice farming
experience, Namutumba district, Eastern Uganda, 2015.
…………………………..
………………………..
71
xi
Table 12: Reasons
for disliking some rice varieties as indicated by farmers in Eastern Uganda in
2015.
…………………………..
…………………………..
…………………………..
…………………………..
…………..
76
Table 13: Criteria rice farmers used in se
lection of the best upland rice varieties in Eastern Uganda
in 2015
…………………………..
…………………………..
…………………………..
…………………………..
……….
77
Table 14: Ranking of selected upland rice varieties based
on farmer selection criteria in
Namutumba district, Eastern Uganda, 2015.
…………………………..
…………………………..
…………….
79
Table 15: Effect of
Striga hermonthica
infection on plant height of upland rice varieties grown
under controlled (growth chamber) conditions at 21 days after infection.
…………………………..
…
82
Table 16: Effect of
Striga hermonthica
infection on stem diameter of upland rice varieties grown
under controlled (growth chamber) conditions at 21 days after infection.
…………………………..
…
85
Table 17: Effect of
Striga hermonthica
infection on tillering of selected upland rice varieties grown
under controlled (growth chamber) conditions at 21 days after infection.
…………………………..
…
87
Table 18: Effect of
Striga hermonthica
infection on number of leaves of upland rice varieties
grown under controlled conditions (growth chamber) at 21 days after infection.
……………………
90
Table 19: Number of
Striga
plants and
Striga
dry weight taken at 21 days after
Striga hermonthica
infection under controlled growth chamber conditions
…………………………..
…………………………..
92
Table 20: Effect of
Striga hermonthica
on above
–
ground rice biomass of infected rice plants under
controlled (growth chamber) conditions at 21 days after infection.
…………………………..
………….
99
xii
LIST OF FIGURES
Figure 1: Life cycle of
Striga
…………………………..
…………………………..
…………………………..
……..
12
Figure 2: Schematic representation of the field trial in Nsinze, Namutumba district, Uganda in
2014.
…………………………..
…………………………..
…………………………..
…………………………..
…………..
25
Figure 3: The field trial in Nsinze, Namutumba district, Uganda in the first season (2014) at 25
days after sowing.
…………………………..
…………………………..
…………………………..
…………………….
26
Figure 4: Farmer preference of upland rice varieties in Namutumba district, Eastern Uganda in
2015, bars indicate the percentage (%) of farmers preferring a certain var
iety.
………………………
73
Figure 5: Rice variety selection by gender of rice farmers in Namutumba district, Eastern Uganda,
2015, data presented a
s percentage responses by farmers.
…………………………..
………………………
73
Figure 6: Parasitic plants attached on the host plant roots 21 days after infection with
Striga
hermonthica
.
…………………………..
…………………………..
…………………………..
…………………………..
.
93
Figure 7: Characteristics and phenotype of resistance in selected rice varieties against
Striga
hermonthic
a
(Namutumba ecotype) in Eastern Uganda. PR (phenotype of resistance), (a)
susceptible and (b) resistant variety compared.
…………………………..
…………………………..
…………
97
Figu
re 8: 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.
……
101
xiii
LIST OF APPENDICES
Appendix 1: Initial soil characteristics of t
he field trial for 2014 and 2015
in
Namutumba district,
Eas
tern Uganda.
…………………………..
…………………………..
…………………………..
……………………..
133
Appendix 2: Analysis of variance
summary
for plant height per rice variety for 2014
–
2015 under
Striga
infested field
conditions
……………………………………………………………
……
13
3
Appendix 3: Analysis of variance
summary
for the number of tillers per rice
variety for 2014
–
2015
under
Striga
infested field conditions.
…………………………..
…………………………..
……………………
133
Appendix 4: Analysis of variance
summary
of rice biomass, grain
yield and
yield components of
rice varieties for 2014 and 2015 seasons under
Striga
infested conditions.
………………………….
134
Appendix 5: Analysis of variance
summary
of
Striga
counts
per
rice variety
for
2014
–
2015 in
Namutumba district,
Eastern Uganda
…………………………..
…………………………..
…………………….
134
Appendix 6: Analysis of variance
summary
of
Striga
components recorded per rice variety for
2014
–
2015 in Namutumba d
istrict,
Eastern Uganda.
…………………………..
…………………………..
..
134
Appendix 7: Analysis of variance
summary
of days to
Striga
emergency and
Striga
flowering of
rice varieties for 2014
–
2015 in Namutumba district
,
Eastern Uganda.
…………………………..
…….
135
Appendix 8: Analysis of variance
summary
of
Number of tillers and plant height
of
rice varieties
in 2014 in Namutumba district
,
Eastern Uganda.
…………………………..
…………………………..
…….
135
Appendix 9: Analysis of va
riance
summary
of
rice biomass, grain yield and
yield components of
rice varieties in 2014 in Namutumba district
,
Eastern Uganda.
…………………………..
……………..
135
Appendix 10: Analysis of variance
summary
of Number of tillers and plant height
of
rice varieties
in 2015 in Namutumba district
,
Eastern Uganda.
…………………………..
…………………………..
…….
136
Appendix 11: Analysis of variance
summary
of
rice biomass, grain yield and
yield components of
rice varieties in 2014 in Namutumba distric
t Eastern Uganda.
…………………………..
………………
136
xiv
Appendix 12: Analysis of variance
summary
of number of
Striga
plants
per rice variety
for 2014
and 2015 in Namutumba district Eastern Uganda
…………………………..
…………………………..
……
136
Appendix 13: Analysis of variance
summary
of
Striga
components record
ed per rice variet
y in
2014 in Namutumba district,
Eastern Uganda.
…………………………..
…………………………..
………..
137
Appendix 14: Analysis of variance
summary
of
Striga
compo
nents recorded per rice variety in
2015 in Namutumba district
,
Eastern Uganda.
…………………………..
…………………………..
………..
137
Appendix 15: Analysis of variance
summary
for
growth parameters of rice varieties grown under
controlled conditions.
…………………………..
…………………………..
…………………………..
………………
137
Appendix 16: Analysis of variance
summary
for
Striga
n
umber and
Striga
dry weight of rice
varieties grown under controlled conditions.
…………………………..
…………………………..
…………..
138
Appendix 17: Questionnaire for participatory
variety selection
…………………………..
……………..
138
Appendix 18:
Rain fall data for Namutumba district in Eastern Uganda for 2014 and 2015
…
..
14
3
xv
LIST OF ACRONYMS
FAOSTAT
–
Food and Agricultural Organization statistics division
MAAIF
–
Ministry of Agriculture, Animal industry and Fisheries
NERICA
–
New Rice for Africa
PMA
–
Plan for modernization of agriculture
UBOS
–
Uganda Bureau of Statistics
PVS
–
Participatory Variety Selection
AATF
–
African Agricultural Technology Foundation
xvi
ABSTRACT
The Government of Uganda
encouraged the adoption of NERICA high yielding upland rice
varieties as one of the strategies to eradicate poverty and increase food security. However, rice
production per unit area has been declining due to production constraints such as
Striga
infestati
on.
Striga
weed is a major biotic constraint to rice production in Sub
–
Sharan Africa. The use of
resistant varieties of rice could be 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 the
se ecotypes of
Striga
.
The main objective
of this research
was to screen a
selection of
upland
rice varieties for
Striga hermonthica
resistance in Eastern Uganda.
A field
experiment
and
a
participatory variety selection study
were
conducted in
Namutumba district
star
ting
in the
first season of 2014
and a followup
laboratory experiment
was
conducted at
the
Department of Plant and Animal Sciences
,
University of Sheffield
,
United Kingdom
.
In the field
experiment, a 5×5 lattice design was used, wit
h 25
upland rice
varieties constituti
ng the treatments
.
Prior to planting, 0.92g of
Striga
seed mixed with 2
00g of white sand were added per
plot to
enhance
Striga
seed levels bank in the soil.
At the time of harvesting a
participat
ory
variety
selection
(P
VS)
exercise involving
38
farmers
, 21 males and 17 female
with 2
y
ears or more
experience
in rice farming were interviewed.
Furthermore
,
i
n the laboratory
experiment
,
a
complete randomized design with a
total of
14
rice
varieties each with
10 plants, 6
infected and 4
uninfe
cted
(
control
)
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
grain
yield
of introduced
rice
varieties
compared to the local
xvii
variety
.
Varieties
WAB928, Blechai,
SCRID090, WAB935, NERICA
–
2,
–
4,
–
10,
–
17, IRGC, IR49,
WAB181
–
18, CG14, Anakila, WAB880
were the most
resistant.
These same varieties
,
except
WAB928, WAB935
, IR49 and IRGC
also
produced higher
grain
yield
s
.
In addition
,
the
same
varieties except
WAB935, NERICA
–
2, IR49 and Anak
ila demonstrated excellent post germination
attachment resistance to
Striga
under controlled conditions.
Furthermore,
farmers selected
varieties; SCRI
D090, WAB880, Blechai, W
AB181
–
18 and NERICA
–
17
as the most preferred
varieties
.
Striga
resistance, early maturity
and high
yield
were
the
most important traits for
farmer
variety selection
criteria
.
T
herefore, t
hese results are highly relevan
t to rice breeders, agronomists
and
molecular biologists working on
Striga
resistance. Similarly,
those
Striga
resistant varieties
combining high yields and excellent adaptability to field conditions can be recommend
ed
to
farmers in
Striga
prone areas else
where 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 cultivated 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 secondary branches in the panicle and
an
almost glabrous glume. In many parts
of Africa,
Oryza sativa
and i
nterspecific crosses between
Oryza glaberrima
and Oryza sativa
are replacing
Oryza glaberrima
(Linares, 2002). Among such interspecifics is NERICA upland
rice, which is as a result of backcrossing between the Asian and African rice
species
(
Jones
et
al.,
1
997; 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 thought that
African and Asian rice
were domesticated independently
(Khush, 1997). It is
suggested t
hat
Oryza sativa
originated in India from
Oryza perennis
whereas cultivation may
have been earlier in China (Purseglove, 1975).
Oryza sativa
was first cultivated in south
–
east
Asia, India and China between 8000 and 15000 years ago (Normile, 2004).
Oryza gl
aberrima
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 500
–
700 years later (Porteres, 1956).
Rice plays an important role in many ancient customs and relig
ious 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
2
as compared to straw of other cereals. It can also be used for the man
ufacture 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
as a substrate in
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 15
million hectar
es with an annual
production of about 500 m
illion metric tons (Tsuboi, 2005
)
.
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 ar
e the leading producers and
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, Africa produced an estimated
quantity of 23 million metric tons of unmilled rice on 9.5 million hectares (FAOSTAT, 2010).
The major pr
oducing regions
were Western, Northern, and E
astern
in Africa
with an estimated
produc
tion of 10.2, 7
.3
and 5
million metric tons,
h
arvested on 5.8, 0.8
and 2.4
million
hectares
respectively (FAOSTAT, 2010)
. In East Africa, rice production
is considerably increasing due
to establishment of upland rice varieties.
1.3 Rice production in Uganda
Rice is becom
ing 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
of 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
introduced by Indian traders in Uganda in 1904 (MAAIF, 2009). Rice production only became
3
economically relevant in the late 1940’s when the government included rice
–
bas
ed 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 inc
reased 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, representing an annual growth rate of 12% (FAOSTAT,
2010). This increase in production is par
tly 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
rice
varietie
s,
the
area under cultivation
increased
and was
estimated 72
,
000
hectares in 2
0
00 (Uganda Bureau of Statistics,
2002) and
is
currently estimated at 90
,
000
hectares (Uganda Bureau of Statistics
,
2012)
. About 80% of rice farm
ers are small
–
scale
farmers owning
less 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 commercially
traded (PMA, 2009), which
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 ra
te in Uganda increases by 3.2% annually (PMA, 200
9;
MAAIF, 2009). This has created a
potential market for the rice produced by the farmers.
According to PMA (2009) and MAAIF (2009), Uganda’s rice imports dropped from 77,600
tonnes in 2000 to 33,000 tonnes
in 2010. This reduction has
been due to
increased domestic
market supply by the farmers
, thus
increasing their incomes.
4
1.4 Problem statement
In 2004
, 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 are
a is
currently declining in Uganda and one of the major constraints
attributed
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 h
ectares of farmland in Uganda are
infested
with
Striga
(AATF, 2006) and
this
infestat
ion
can cause between
3
0
–
100% yield loss if not
checke
d in the field (Oswald, 2005).
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 cro
p decreasing infestations. Cotton
was
not a host crop for
Striga
.
S
econd
is the
declining soil fertility as a result of continuous
cul
tivation of the soil w
ithout replenishment of used up
nutrients from the soils
(Olupot
et al.,
2003).
This
is partly due t
o
increase in the population wit
h many of the areas which would
be
under fallow being put to cultivation.
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). Cissoko
et al
. (2011) and Jamil
et al.
(2011) assessed 18 upland NERICA varieties for pre
–
and post
–
attachment resistance to
Striga
hermonthica
and
Striga
asi
atica
under controlled environmental conditions. According to these
studies, NERICA varieties including NERICA1, 2, 10 and 17 exhibited very good post
–
5
attachment resistance to several ecotypes of
S.hermonthica
(Cissoko
et al.,
2011). In addition
,
NERICA v
arieties produce different types and amounts of strigolactones in
their root exudates,
which
alter pre
–
attachment resistance (Jamil
et al.,
2011). Inspite of the above successes under
controlled environmental conditions, there has been no published informa
tion about the effect
of the environment on the expression of resistance apart from Rodenburg
et al.
(2015) who
evaluated 18 NERICA cultivars and their parents under field conditions in Kyela, Tanzania
and Mbita, Kenya
. They observed variation
in field res
istance 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
numbe
r of upland rice varieties
repo
rted to be resistant to either
S.hermonthica
or
S.asiatica
in
different 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
ir
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 a
nd the
environment (Oswald,
2005
). A number of strategies have been proposed in the management of
Striga
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), cr
op rotations and improved
fallow
management (Oswald, 2005
) and hand weeding. However, none of these control approaches
has proven very effective in
Striga
weed management in
rice production systems. Use of
resistant varieties is an important approach in
St
riga
managemen
t due to its genetic nature. It
is also cost effective as farmers do not have to invest a lot of time and resources in
implementing this management
technology.
However
,
b
ecause of the genetic variability
existing among different species and e
cotypes of the parasite, resistance found in some varieties
6
may be overcome by a small subset of
Striga
individuals within the seed bank leading to
development of a virulent population of
Striga
overtime (Rodenburg and Bastiaans, 2011).
Also some studies h
ave 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 w
ith another parasitic ecotype.
Therefore
rice varieties need
to be tested at multiple
/different agro
–
ecosystems.
Likewise farmers in one place may have different variety
preferences compared to farmers in a
nother 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
therefore
aimed
at screening
selec
ted
upland rice varieties sourced from different areas in Africa for
S.hermonthica
resistance and identify rice varieties that
could
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
to
identify farmer preferences in order to contribute to
Striga
management and overall farmer livelihoods in Eastern Uganda.
The specific objectives were
aimed at establi
shing the following
:
I.
To determine
the
field
resistance
to
Striga hermonthica
of twenty five
selected
upland rice varieties sourced from different areas of
Africa.
II.
To identify
Striga
resistant
farmer
–
preferred upland rice varieties that can be
adopted by
farmers in Eastern Uganda.
III.
To evaluate
the post
–
germination
attachment resistance
to
Striga hermonthica
of
farmer
–
preffered
upland rice varieties
under controlled conditions.
7
1.7 Hypotheses
i.
There is a significant variation in resistance
to
Striga
hermonthi
ca
among
upland
rice
varieties
ii.
Far
mer preferred characteristics of
Striga
resistant upland rice 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 estima
ted to account for 32%
and 9%
potential and
actual yield losses
respectively
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 fields, 31 species in
hydromorphic and 74 species in lowland rice. The most common weed species in upla
nd rice
include;
Rottboellia
cochinchinensis
(Lour.) W. Clayton,
Digitaria horizontalis
Willd,
Ageratum conyzoides
L; and
Tridax
procumbens
L., while
Ageratum 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
raminaea (43%) and
Cyperaceae
(
37%
)
whereas
in
upland rice
, weed
species
are from Graminaea (
36%
)
and Compositae
(
16%
)
(Rodenburg and Johnson, 2009).
Also important in upland rice production are parasitic weeds
of genus
Striga
which include;
Striga
hermonthica
(Del) Benth and
Striga
asiatica
(L.) K
untze.
Agronomic factors such as inadequate land prepar
ation, rice seed contamination, broad cast
seeding in lowlands, inadequate water and fertilizer management, mono cropping, delayed
herbicide application and use of poor quality seed are responsible for the major weed problems
in rice.
2.1.1
Striga
The ge
nus
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 and of these
eleven species are considered parasitic on agricultural crops.
Striga
species infest an estimated
9
two
–
thirds of cereals and legumes in Sub
–
Saharan Africa, causing annual crop losses estimated
at US$7 billion and negatively affecting t
he livelihood 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
Ramaiah
et al. (1983) and
Parker
.
(20
12
)
the most important species of
Striga
in Africa include;
Striga
hermonthica
(Del.) Benth;
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 Suda
n has the greatest biodiversity 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 (Ejeta,
2007).
Striga
hermonthica
has had the largest geogra
phical
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 Arabian Peninsula across the R
ed sea
(Ejeta, 2007)
.
The re
d
–
flowered and weedy ecotype
Striga
asiatica
(L.) Kuntze is mostly found distributed in eastern and southern Africa
, whereas
the yellow
–
flowered ecotype is found in West Africa (Mohammed
et al.,
2001; Rodenburg
et
10
al.,
2010)
.
The latter ecotype has no impo
rtance as a weed.
Striga
gesneriodes
is
also
thought
to h
ave originated from West Africa
(Ejeta, 2007).
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
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. The
stem
is
square
but obtusely angled. Leaves are
opposite, appressed to the stem, scale
–
like, 5
–
10 mm × 2
–
3 m
m.
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; stamens 4, 2 longer and 2 shorter; ovary
superior, tubular, 2
–
celled, style terete, stigma 2
–
fid.
The f
ruit
is
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
S
triga
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 hai
rs are ascending or retorse (pointing backward),
stem shape with some species having either round, obtusely square and square in cross section;
11
leaf lobbing and dentations whether leaf margins are lobed, serrate or smooth; inflorescence
types with some spe
cies 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
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 mor
e per plant (Scholes and Press,
2008
). 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 exp
osure
to moisture and
temperatures above 20
0
C for a period of one week or more is probably a
survival adaptation which prevents the seed from germina
ting before
the
rainy season is well
established
(Berner
et al.,
1997)
. Before the rains
,
the host roots are 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 (
Scholes and Press, 2008
; Cardoso
et
al.,
2011
). Thes
e groups of biochemical
s
have been identified as germination stimulants for
parasitic weeds;
dihydrosorgoleones
,
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 c
hemotropic (
Chang and Lynn, 1986;
Westwood
et al.,
2010;
Cardoso
et al.,
2011
).
After germination, a series of chemical signals direct the radicle to the host root where it
attaches and penetrates. However, when the seedling does not atta
ch to a host root within three
to
five days, the
s
eedling dies (Worsham, 1987).
Once penetration has taken place, an internal
feeding s
tructure
, the haustorium
is formed. Through the haustorium,
the parasitic weed
12
establishes a xylem
–
to
–
xylem connection with the host roots
(Scholes and Press, 2008
; Cardo
so
et al.,
2011
). 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 (Press
et al.,
1987a
; Cardoso
et
al.,
2011
). As the host m
atures,
the parasite emerges,
b
egins to produce chlorophyll,
starts to
photosynthesize
and
flowers
(Boumeester
et al.,
2003; Scholes and Press, 2008
).
Reproductive strategies
range from
autogamy to obligate allogamy depending on
th
e species (Musselman, 1987).
Species 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
.
T
he other remaining species of
Striga
are autogamous
(Musselman
et al.,
1982)
.
Following reproduction, seeds are dispersed and then the cycle
begins again.
Figure
1
:
Life cycle of
Striga
(Mweze
et al.,
2015)
13
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
with little option for addition of external inputs
(
Ejeta,
2007
; Parker, 2012)
.
Striga
is prevalent
u
nder conditions of
low soil fertility especially low nitrogen and moisture levels.
Striga
species
have been described as indicators of low soil fertility and their infestation is linked to low
nutri
ent conditions (Oswald, 2005).
Striga
is common in tropical and subtropical areas of
Africa and some parts of India except in extremely cold climates.
Striga
species require a
suitable temperature of 20
0
–
40
0
C under moist conditions for the seeds to germina
te 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 o
ccurs
already
before the emergence of the parasite
, and
continues thereafter
. Initial symptoms occur while the parasite is 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, 1991
).
Parasitism by
Striga
species reduce
s
the yields in rice
in two main ways;
In the first instance, t
he 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
). Although
Striga
is chlorophyllous, its rate of
photosynthesis i
s low; therefore most
of its
carbon is host derived (Press
et
al.,
1991
; Pageau
et al.,
1998
; Parker, 2012
). 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 m
ost important
causes for
Striga
damage
is its effective
14
competitive ability in depriving the host plant of carbon,
nitrogen and inorganic salts.
This
happens while at the same time inhibiting the growth and impairing photosynthesis of its host
(Khan
et al.
,
2006).
Secondary, t
he parasite pathologically affects the growth and development of upland rice. This
is associated with the phytotoxic effects of
Striga
within days of attachment. 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
(Musselman and Press, 1995; Gurney
et al.,
199
9
; Gurney
et al.,
2006)
. 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 carboxylation (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
pre
sents 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 have
been achieved through cultural
met
hods 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
(
Oswald
et al.,
2005;
Rodenburg
et al.,
2010)
.
15
2.1
.1
.6.1
Cultural control
A
number of
cultural
methods
are
described in the management of
Striga
in the field
. These
are described below
;
2.1.1.6.1.1
Hand weeding
This is the most used method of
Striga
management by farmers in
Africa
. Weedi
ng removes
emerging
Striga
shoots, preventing
them from flowering and producing seed
, 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 subsequent seasons
as
it
can reduce the
Striga
seed bank over the long term.
Use of hand weeding reduced
Striga
infestation by 12% in maize fields in one season and
when applied for four consecutive seasons
it reduced infestations by 26.6% in the fo
urth year (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
set
t
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.
2.1.1.6.1.2
Crop r
otation 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 w
hich
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.
16
Crop rotation cycles disrupt the parasitic weed life cycles
by reducing the soil seed bank
.
R
otations including trap
crops such as cow pea (
Vigna unguiculata
), soy bean (
Glycine max
),
groundnut (
Arachis hypogeal
) and pigeon pea (
Cajanas cajan
) have all proved effective, by
stimulating and contributing greatly to reduction in soil seed bank (Eplee and Langston, 1991).
Ro
tations 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 importa
nt to control
Striga
.
2.1.1.6.1.3
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
sti
mulate 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
infe
stations on 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 fertiliz
er 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 parasitic weed germination (Lopez
–
Raez
et al.,
200
9)
2.1.1.6.2
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) and
Striga
asiatica
(Delassus,
1972). The herbicides control the weed post
–
emergenc
e
, but germinations continues after the
17
residues
loose effectiveness
(Ivens, 1989). Also the weed exerts its harmful effects before it
emerges above the ground. Therefore coating of seeds with herbicides can be effective at
improving the chemical control, for example use of herbicide coated Imazapyr maize seeds in
East Afri
ca 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
Acetolactate Synthase (
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 upl
and
rice
that can be used in combination with these ALS inhibiting c
hemicals as seed coating
(Rodenburg and Demont, 2009)
.
2.1.1.6.3
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)
and
yield advantages under weedy conditions due to vigorous growth,
high tillering ability and large phanophile leaves (Johnson
et al.,
1998; Saito
et al.,
2010).
An
example is CG14 a cultivar of African rice that showed resistance against
Striga hermonthica
a
nd
Striga aspera
(Johnson
et al.,
2000
;
Kaewchumnong and Price, 2008
). 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 fro
m 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 CG14.
The results of his study
showed a
significant
18
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
–
2,
–
5
–
6,
–
9
,
–
10
and
–
15
showed an intermediate production levels and NERICA
–
1
and CG14 produced the smallest am
ount of strigolactones meaning
that they are
highly
resistant cultivars (Jamil
et 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 by the weed
(Maiti
et al.,
1984; Olivier
et al.,
1991).
Post
–
attachment mechanisms
mainly include; failure of the
Striga
to establish the xylem
–
to
–
xylem connections with the host
as a result 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 isolate)
from Kenya
(Cissoko
et al.,
2011).
Other
post
–
attachment
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 p
lants
can have natural ability to release 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 e
xhibited a range of
susceptibility to the
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 g
reatest 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.
Th
is is because of genetic variation in both
19
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; Scho
les and Press, 2008
; Huang
et al.,
2012
).
For example when
Johnson
et al,
(1997) screened a selection of upland rice cultivars, he reported that rice
cultivars IR38547
–
B
–
B
–
7
–
2
–
2 and IR47097
–
4
–
3
–
1 remained parasite
–
free at two sites in Kenya
but these supp
orted more parasite emergence in Ivory Coast. Also in this same study a number
of cultivars showed greater susceptibility to
Striga hermonthica
in West Africa than when
observed in field trails in Kenya by Harahap
et al.,
1993.
However, Cissoko
et al.
(201
1)
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
Striga
control i
n rice.
R
ecently 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 parents (WAB181
–
18
and CG14
)
showed excellent resistance to
S. hermonthica
ecotype from Mbita
Kenya
. 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. herm
onthica
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.
Combining pre and post
–
attachment resistance is necessary in
future if varie
tal control of
Striga
is to become an important component in integrated
Striga
management (Cissoko
et al.,
2011)
.
Evaluation of
pre
–
or
post
–
attachment resistance under field conditions is a challenge as
resistance is usually measured as the number of parasite attachments that emerge above ground
over time (Rodenburg
et al.,
2015). This reflects the combination of pre
–
and post
–
attachments
mec
hanisms that are operating in each cultivar but it is not possible to separate these. In
20
addition, the expression of resistance in the field can be affected by environmental factors. For
example the nutrient status of the soil, particularly nitrogen and
phosphorus levels will affect
the exudation of strigolactones and thus germination of parasite seeds (Yoneyama
et al.,
2007a,
b). The distribution of
Striga
seeds in the soil can also be heterogeneous and environmental
conditions including temperature and
rainfall can affect the resistance levels observed
(Haussman
et al.,
2000). Thus in the field, the resistance level of a cultivar results from both
Genetic and Environmental factors, often referred to as the G x E interaction (Rodenburg
et
al.,
2015).
To clearly evaluate post
–
attachment resistance, cultivars have to be grown under controlled
environment conditions so that the resistance level observed can be attributed to genetic factors
rather than to environmental variables. Many studies to evaluate p
ost
–
attachment resistance of
different cereal hosts to
Striga
have been carried out using soil
–
less rhizotrons that allow easy
access to the host root system for quantification of host resistance reactions (Gurney
et al.,
2006; Swarbrick
et al.,
2008; Ciss
oko
et al.,
2011). These growth systems also allow roots to
be inoculated with germinated
Striga
seeds, which by
–
passes the differences in production of
strigolactones by the roots of different cultivars (Gurney
et al.,
2006; Cissoko
et al.,
2011).
2.2
Par
ticipatory variety selection
Partici
patory variety selection
(PVS)
involves
selection of finished or near
–
finished products
by farmers on farmer’s fields (
Joshi
and
Witcombe
,
1998
).
PSV
can be used in identification
of farmer
–
acceptable varieties. This can overcome the problems associated with growing of old
or obsolet
e varieties (Joshi and Witcombe,
1996
). Normally costs of conducting the research
are reduced and adoption rates increase
d when farmers’ participate actively in variety selection
exercises (Joshi
et al.,
1995). It is therefore possible that there would be
an
increase in rice
production since farmers are growing varieties they desire (Witcombe
et al.,
1999). Studies by
21
Joshi
et al. (2002); Dorward et al. (2007); Nanfumba et al. (2013) have all indicated the
importance of participation of farmers in variety selection according to their preferences, needs
and other expected characteristics in rice.
There are mainly three phas
es in participatory variety
selection to identify farmer preferred varieties including identification of farmer’s needs,
searching for suitable material to test with farmers and experimentation on farmer’s fields
(
Witcombe and Joshi, 1998
).
Several
studie
s on rice have indicated high yield
ing ability, tillering capacity and short
growing
life cycle as the most important traits attached t
o farmers
–
preferred
varieties (
Joshi and
Witcombe
,
1996; Gridley
et al.,
2002; Dorward
et al.,
2007; Nanfumba
et al.,
201
3; Rahman
et al.,
2015). Also farm
ers have shown great preference
for
varieties with more spikelets per
panicle, resistance to pests and diseases, uniform tillers, good panicle length, uniform plant
height and lodging resistance
(
Gridley
et al.,
2002;
Nanfumba
et al.,
2013;
Rahman
et al.,
2015).
Apart from these traits there are also
agoleptic
characteristics farmers attach to rice
varieties such as good taste, good aroma and go
od cooking ability among others.
In upland
rice
ecologies
,
less has been don
e on variety selection especially on
Striga
resistance.
A number of
varieties have been tested for
Striga
resistance (
Harahap
et al.,
1993
;
Johnson
et al.,
1997;
Atera
et al.,
2012
; Rodenburg
et al.,
2015) however there has been no involvement of farmers
in these studies. In
line with this issue, evaluation of selected rice varieties was conducted in
e
astern Uganda
with hope of availing
Striga
resistant and other desirable characteristics of
rice
vari
eties to overcome the
Striga
scourge in this area.
22
CHAPTER THREE
3.0
MATERIALS AND METHODS
The research undertaking involved three studies
conducted as follows
:
(i)
T
o determine the
resistance of
selected
upland rice varieties against
Striga hermonthica
under field conditions
,
(ii)
T
o
identify
using
a
farmer participatory
approach,
Striga
resistant
upland rice varieties
to
be adopted by farmers
, and
(iii) To evaluate the post
–
germination
attachment resistance of
selected 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 in 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
distri
ct headquarters at Namutumba are 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 v
illages. The district
has a total 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 (
Namutum
ba
Census report, 2007).
3.1
.
2
Rice varieties
used
The experiment
involve
d
evaluation of
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
ance
to
Striga
and
good
grain
yield
23
(Table 1)
.
The
l
ocal check, Superica
(WAB165) is a high yielding but
Striga
–
susceptible
variety
.
Table
1
:
List of rice varieties screened for
Striga
resistance in Nsinze, Namutumba
District, Uganda and their characteristics
.
Variety names
Short
name
Rice species
Striga
reaction
(resistance)
Stature
Information
ACC102196
ACC
O.glabberima
Tolerant to
striga
hermonthica
Dwarf
(Johnson
et al.,
1997)
Agee
Agee
O.glabberima
S.hermonthica
Dwarf
(Jamil
et al.,
2012)
Anakila
Anakila
O.glabberima
S.hermonthica
Dwarf
(Jamil
et al.,
2012)
WAB56
–
50
WAB50
O.sativa
Susceptible to
S.hermonthica
Tall
(Rodenburg
et al
2015)
CG14
CG14
O.glabberima
S.hermonthica
and S.asiatica
Intermediate
(Cissoko
et al.,
2011; Jamil
et al.,
2011)
IAC165
IAC
O.sativa
Susceptible to
S.hermonthica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2012)
IR49255
–
B
–
B
–
5
–
2
IR49
O.sativa
S.hermonthica
and
S.aspera
Dwarf
(Harahap
et al.,
1993; Johnson
et
al.,
1997)
IRGC78281 (Ble
Chai)
BleChai
O.sativa
S.hermonthica
Tall
(Harahap
et al.,
1993)
IRGC81712 (IR
38547
–
B
–
B
–
7
–
2
–
2)
IRGC
O.sativa
S.hermonthica
Dwarf
(Harahap
et al.,
1993)
Makassa
Makassa
O.glabberima
Tolerant to
S.hermonthica
and
S.aspera
Tall
(Johnson
et al.,
1997)
MG12
MG12
O.glabberima
Tolerant to
S.hermonthica
and
S.aspera
Intermediate
(Johnson
et al.,
1997)
NARC3
(ITA257)
NARC3
O.sativa
Susceptible to
S.hermonthica
Dwarf
(Harahap
et al.,
1993)
NERICA
–
1
N1
Interspecific
hybrids
S.hermonthica
and
S.asiatica
Intermediate
(Cissoko
et al.,
2011; Jamil
et al.,
2011
; Rodenburg
etal.,
2015
)
NERICA
–
10
N10
Interspecific
hybrids
S.hermonthica
and
S.asiatica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2011
; Rodenburg
etal.,
2015
)
NERICA
–
17
N17
Interspecific
hybrids
S.hermonthica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2011
; Rodenburg
etal.,
2015
)
NERICA
–
2
N2
Interspecific
hybrids
S.hermonthica
and
S.asiatica
Intermediate
(Cissoko
et al.,
2011; Jamil
et al.,
2011
;Rodenburg
etal.,
2015
)
24
NERICA
–
4
N4
Interspecific
hybrids
S.asiatica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2011
; Rodenburg
etal.,
2015)
SCRID090
–
60
–
1
–
1
–
2
–
4
SCRID090
O.sativa
No information
Tall
Superica
(WAB165)
Superica
O.sativa
Susceptible to
S.hermonthica
Tall
Local check
UPR
–
103
–
80
–
1
–
2
UPR
O.sativa
Tolerant to
S.hermonthica
Dwarf
(Harahap
et al.
,
1993)
WAB 181
–
18
WAB181
O.sativa
S.hermonthica
and
S.asiatica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2011)
WAB 56
–
104
WAB104
O.sativa
T
olerant to
S.asiatica
Tall
(Cissoko
et al.,
2011; Jamil
et al.,
2011)
WAB 928
–
22
–
2
–
A
–
A
–
B
WAB928
O.sativa
S.hermonthica
and
S.aspera
Dwarf
(
Johnson
et al.
,
2000)
WAB 935
–
5
–
A
–
2
–
A
–
A
–
B
WAB935
O.sativa
S.hermonthica
and
S.aspera
Dwarf
(
Johnson
et al.,
2000)
WAB880
–
1
–
32
–
1
–
1
–
P2
–
HB
–
1
–
1
–
2
–
2
WAB880
O.sativa
No information
Tall
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
correct interpretation of
Striga
resistance
screen
ing based on the
number of emerged
Striga
plants.
The experimental area measured 50m ×15m
.
T
he upland rice varieties
(treatments) constituted
the sub
–
plots of the experiment. Each
net sub
–
plot
, c
ontaining an individual variety
, 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 plots were
maintained
throughout the two se
asons of
2014 and 2015 (Figure 2
and 3
).
25
Figure
2
:
Schematic representation of the field trial in Nsinze, Namutumba district,
Eastern
Uganda in 2014.
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
26
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
on 2 March 2014 and 3 March 2015 using an oxen plough
two
weeks before the beginn
ing of the rains.
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) mixed 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 at
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 constituting 5 rows
a
nd 11 hills of plants per row.
Thinning was done
at
two weeks af
ter sowing leaving 3 plants
per hill. Weeding was done
regularly every 10 days after
Striga
infestation
to
remov
e all
weeds
27
other than
Striga
.
A b
asal application of
(17:17:17)
NPK 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 insecticide
Dursban (Chlorpyrifos
48
% EC
, 0.75liters/ha
),
every after
2 weeks starting
30
days after
planting un
til grain formation.
N
ematodes and termites
were
controlled
using F
uradan
(Carbofuran
, 7.5kg/acre
) 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
organic carbon
,
soil texture
,
nitrogen,
pH,
potassium
and
phospho
rous.
Soil
analys
is w
as
done at Makerere
U
niversity
Soil Science
L
aboratory
.
Organic carbon content of the soil
was
determined by reduction of potassium dichromate by
organic carbon compounds and oxidation
–
reduction titrat
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
de
termined by the Kjeldahl
method.
Soil pH was
determined using a ratio of 1:2
.5
soil to water ratio (Okalebo
et 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 phosphorous was de
termined using the Bray 1
method.
28
3.1
.
5
.
2
Crop
data
Data
collected o
n rice included
;
plant height,
number of tillers
,
above ground
dry
biomass
at
harvest and
harvest index these were collected on a random
sample of 9 hills per sub
–
plot while
p
anicle dry weight, grain recovery efficiency
and grain yield
were
determined from a
sample
of 27
rice hills per sub
–
plot
as follow;
Plant height was
determined
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 leaf
or the
tip of the tallest panicle
.
Tiller numbers
were assessed starting
43
,
57
and 113
days after
rice sowing
by counting all the tillers
per plant
.
Above
–
ground
dry
biomass
was
determined at harvest
.
For each subplot the rice straw of the
plants of
9
hill
s
were
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.
Panicle dry weight was detrermined by
cut
ting and collecting panicles for each p
lot and air
–
drying them
. After 1
–
2 weeks of air
–
drying
,
the panicles
were
weighed
using a weighing scale
and recorded.
Grain yield at
grain maturity was
expressed in kg
per
hectare
, first, the
panicle
s
were
threshed and the grains obt
ained
were
weighted. T
hen the grains
were
winnowed
(removing all empty grains) and weighted again, immediately followed by grain moisture
content measure
ment
s
which
enable
d
correction of all grains to a standard 14
% moisture
content. All measurements were
recorded.
Rice grain r
ecovery efficiency was
calculated from
the
ratio of
unwinnowed and winnowed grains
mu
ltiplied by 100
as follows;
Grain recovery efficiency =
∗
100
Harvest in
dex was computed as a ratio of
grain dry weight and
total
above ground
dry matter
multiplied by 100
as follows
;
Harvest index
=
∗
100
29
3.1
.
5
.
3
Striga
data
Data
collected on
Strig
a
weed
include
d
;
days to
Striga
emergence, days to
Striga
flowering,
above
–
ground
Striga
numb
ers and
Striga
dry weight at rice harvest
as follows
;
D
ays to
Striga
emergence
in each sub
–
plot, were monitored and recorded 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 visual
observations done every 3 days,
starting at 21 days after first
Striga
emergence in each sub
–
plot.
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 centra
l
observation area containing 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 harvest
.
For assessment of
Striga
biomass dry
weight, in
each subplot, all above
–
ground
Striga
plants
were
collected from the central
observa
tion area containing 27 hills.
During the season
,
Striga
plants that die
d
before rice
harvest
were
collected and put in an envelope designated to th
e associated
sub
–
plot. A
t harvest
all above
–
ground
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 subj
ected to tests of normality and homogeneity of variance (Sokal
an
d Rohlf, 1995). For analysis of
Striga
counts, maximum
above
–
ground
Striga
numbers
(NSmax) and
Striga
dry 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
means of
maximum above
30
ground
Striga
n
umbers (
NSmax
)
and rice yields, NSmax and
number of
tiller
s
and NSmax and
Striga
dry weight. All the d
ata
were
subjected to statistical analysis of variance
(
ANOVA
)
usin
g GenStat computer software (V14
)
(
VSN International, Hemel Hempstead, UK
)
.
The
means w
ere 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, Namutumba district in the main rainy season of 2015. The field
screening trial w
as set up as described above
(Study one)
.
3.2.1
Data collection
Information was collected using q
uestionnaires with open
–
ended and structur
ed questions
administered through personal interviews
(Appendix 1
7
)
. A purposive
sample was used,
selecting
upland rice farmers
with 2 years or more
experience in rice production. These were
selected
from
four rice farmer
’
s groups in the district i.e.
(
Abe
ndowoza Ndala group in
Bulagala, Bukonte mixed farmers group, Ivukula integrated farmers’ a
ssociation in Ivukula
Subcounty and
Emberi Enkula mixed group
)
. Questionnaires were first pre
–
tested so as to get
acclimatized with the contents of the questionnaire
s.
Participants were asked to select the five
(5)
best
va
rieties among
the
25 varieties
in the
field
trial
, in terms of their
morphological appearance in the field, using a scale of 1to 10
with 10
meaning ‘superior’ and
1 meaning ‘inferior’.
Farmers were also asked to
s
e
lect
the five worst
performing varieties in the field trial.
To avoid bias
, participants were not
allowed to
communicate to
each other during the selection exercise.
After making the choices and
considerations, partic
ipants we
re engaged in a one
–
on
–
one discussion with extension officers
and researchers to explain
reasons for
the
ir
choices. This was aimed at getting diverse views
31
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 stati
stical 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 select
ion with respect to gender, data was restructured 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 environment 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
S
heffield
, Department of Animal
and Plant Sciences
UK
. 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,
Uganda
. Rice seeds were provided by t
he Genetic Resources Unit of Africa Rice.
Striga
hermonthica
(Del.) Benth seeds were obtained from farmer fields in Namutumba district
in Eastern Uganda.
A total of 14 varieties were used including WAB56
–
50, WAB56
–
104,
WAB181
–
18, WAB880, WAB928, NERICA
–
1,
–
4,
–
17, SCRID090, Blehai, IRGC, CG14,
32
IAC165 and the local check (Superica). Th
ese were slected according good field resistance to
Striga
in
Namutumba in
2014
(season one)
and
farmer preference
.
3.3
.
2
Growth and infection of rice plants with
Striga hermonthica
Rice seeds were germinated between blocks
of moistened horticultural rock wool for six days
after which a s
ingle rice seedling was
transferred to a r
oot 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.
Rhizotron
s
were
covered
with
aluminum
foil to prevent light from reaching the roots.
Rhizotrons were
supplied with
25ml of 40% long Ashton (Hewitt
, 1966
), 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 mo
istene
d
glass
–
fiber filter paper
(Whatman),
in petri
–
dishes for 12
–
15 days at 30
0
C (Gurney
et al
., 2006
; Cissoko
et al.,
2011
).
E
ighteen hours before infection of rice seedlings, 1ml of
0.1ppm solution of an artificial germination stim
ulant GR24 was
added
to
P
etri dishes
containing
pre
–
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 o
f rice roots with germinated
Striga
seeds eliminates differences due to variation in production of germination stimulants by the
different rice cultivars (
Gurney
et al.,
2006;
Cissoko
et al.,
2011
). Uninfected rice cultivar
s
acting as the control were
treated in a similar way as des
cribed above without the infec
tion
step.
A total of 10 plants were sown per variety with four controls (uni
nfected) and six infected
arranged in completely randomized design.
Plants were grown in a temperature controlled
33
grow
th 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
Data collection
3.3.3.1 Growth measurements
A series of
non
–
destructive growth measurements were made each week from the day of
infecting rice seedlings with germinated
Striga
hermonthica
seeds includ
ing plant height
,
number of leaves, stem diameter
and
number of tillers. Plant height was
determined using a
ruler measured from the stem base until
the attachment of the new leaf. Number of leaves
were
determined by counting all the leaves emerging
from the tillers and main stem.
Stem
diameter
was determined from the base of the stem
, 5mm above the root crown
us
ing a digital Vernier
caliper (Mitutoyo England UK). N
umber of tillers
were counted and recorded
3.3.3.2 A
bove ground rice biomass
Above
–
ground rice biomass was
determined
three weeks after infection of the rice cultivars.
Plant materials were
separated i
nto main stem, tiller stem, main stem leaves and tiller stem
leaves
for esy drying
. Plant material was dried at 70
0
C for 7 days and then weighed to determine
dry biomass.
3.3.3.3
Quantification of post
–
germination
attachment resistance of the rice cultiva
rs
Po
st
germination
–
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
(Canon UK, Regate Surrey)
.
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
D
digital camera. The number of
Striga
seedl
ings from each rice plant wer
e determined
34
from the
P
etri dishes 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.
3
.
3
.
3.4
The p
henotype of resistance
The
phenotype of resistance was
investigated by photographing parasites developing on the
root syste
ms of each cultivar at different stages after infection using a Leica MZFLIII
(Leica
Microsystems Ltd, Heerbugg, Switzerland)
stereo microscope and a diagnostic i
nstruments
camera Model 7.4 and
by cutting small sections of root with attached parasite and m
ounting
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
.
3.4
Statistical analysis
Data w
ere
subjected
to
A
nalysis of
V
ariance
(
ANOVA
) usin
g GenStat computer software
(v14
)
(
VSN International, Hemel Hempstead, UK
)
, 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 cl
assical ANOVA (Sokal and Rohlf, 1995),
Number of emerged
Striga
plants
and
Striga
dry weight
were transformed usin
g Natural logarithms, log (x+1)
.
35
CHAPTER FOUR
4.0 RESULTS
AND DISCUSSION
4.1 Study one: Resistance of upland rice varieties against
Str
iga 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 exp
erimental field at the onset of
the season in 2014 and 2015
(Appendix 1).
The pH was 6.5 in 2014 and
5.8 in 2015. This was
within range of optimum soils for upland rice
as
established by Somado
et al.
(2008).
Availa
ble
nitrogen was highest
(
0.12%) in 2014 and lowest (0.10%) in 2015. The nitrogen content was
below the minimum thres
hold (<10%
) established b
y Oikeh
et al
. (2008) in both years.
Phosphorous
was
highest in 2015
(2.89
ppm) compared to (2.34
ppm) in 2014
.
The soil
phosphorus level
was below the minimum of 3 ppm
, suggested by Wopereis et al. (2008) and
Oikeh
et al
. (2008), in both seasons
(Appendix 1).
Potassium was highest
(174
ppm) in 2015 and lowest (135
ppm) in 2014. The pota
ssium
content of the soil was above the minimum of 78
ppm in both seasons
(Oikeh
et al.,
2008). The
phosphorous and nitrogen levels were low to
affect the growth
of
Striga
plants in both years
.
Organic matter of the soil was highest (1.42%) in 2014 and lowest in 2015 (1.24%). This was
below th
e minimum <2% in both seasons.
This organic matter was too low to cause suicidal
germination of
Striga
seeds due to ethylene
production hence did not affect the
Striga
seed
bank in the two seasons.
The texture of the soil in 2014 (sand: silt: caly) was (75: 8:7) while in
2015 it was (74:11:15) (Appendix 1).
According to Gerakis and Baer. (1999) the soil can be
described as sand
clay loam.
36
4.1.2
The growth parameters of upland rice varieties grown under
Striga
infested fields.
4.1.2.1
Plant height per variety
Rice varieties ha
d a highly significant effect
on plant height
at 43, 57 and 113 days after sowing
in both seasons
(
Appendix 2).
For both seasons,
variety x season interaction
effects
on plant
height was highly significant
(Appendix 2)
.
In 2014, varieties differed significantly in pl
ant height (Appendix
8). Variety
M
G12
(21.13
cm)
was
significantly
the
s
hortest
at 43 d
ays after sowing
compared to rest of the varieties
except CG14 (23.03cm)
(Table 2). At
57 days after sowing
MG12
was significantly
the
shortest
(28.47cm)
compared to the rest of the varieties
except UPR (32.47cm)
(
Table 2).
Variety
NARC
3 at
53.26
cm and UPR at
53.
76 cm
had significantly
the
shortest stems at 113
days after sowing
(Table 2).
Variety Superica
was the tallest (50.25 cm)
43 days
and WAB56
–
50
(66.47cm)
at 57days
after sowing. In the same season, Blechai produced significantly
(
P
<
0.001)
the tallest plants
(128.30 cm)
across varieties at 113days after sowing (Table 2).
In
2014,
Plant height ranged from
21.13 cm for MG12 to
50.25 cm for Superica 43 days after
sowing,
28.47cm for MG12 to
66.47cm for WAB56
–
50
,
57 days after sowing and
53.26
cm for
UPR to
128.30cm Blechai
113 days after sowing (Table 2).
In 2015
variet
ies differed significantly
in plant height (Appendix 10
).
Variety Blechai was the
tallest
36.12cm
,
58.72cm and 123.60cm at 43 days,
57 days
and 113 days
after sowing
respectively
(Table 2).
Varieties CG14 (20.95cm), Agee
(21.43 cm), UPR (22.56 cm) were
significantly
(
P<0.001
)
the shortest at 43 days after sowing compared
to other varieties except
NARC
3
(24.60 cm)
which was not significantly (P>
0.001) short
er than
UPR
(22.56 cm)
(Table
2).
37
Table
2
:
Plant height of upland rice varieties grown under
Striga hermonthica
infested
field conditions during the first rainy season of 2014 and 2015 at 43, 57 and 113 days after
sowing.
`
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
0
2.02
3.44
11
.00
CV%
8.89
8.24
8.98
6.86
7.33
13.37
38
In the same season (2015), varieties UPR (31.70 cm) and CG14 (33.22 cm)
were
significantly
(P<
0.001) the shortest at 57 days after sowing compared other varieties except
NARC3 (35.96
)
which was not significantly (P> 0.001) different form
them
(Table 2).
Variety Superica
(local
check)
produced significantly
the
shortest
(34.55cm)
stem
s
at 113 days
after sowing
compared
to other varieties in the field
except IAC 165
(38.09 cm)
and NARC3
(48.32 cm)
(Table 2).
Plant height ranged from
20.95cm for CG14 to
36.12cm for Blechai 43 days after sowing,
31.70cm for UPR to
58.73cm for Blechai 57 days after so
wing and
34.55cm for Superica
to
123.60cm for Blechai 113 days after sowing (Table 2).
Striga
has an effect on the height of susceptible rice varieties reflected in increased stunting of
stems (Jonson
et al.,
1997; Atera
et al.,
2012). The chan
ge in
hormonal imbal
ances may be
responsible for this
differences in allometry observed (Taylor
et al.,
1996
).
Varieties IAC 165
and
Superica
(Local check)
that are genetically tall varieties (Table 1),
had
short stems
especially in 2015
. This was very evident w
hen 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.
These varieties are also categorized as being very susceptible to
Striga hermonthica
(
Johnson
et al.,
1997;
Rodenburg
et al.,
2015). It
is
therefore possible that
their stems were reduced due to
Striga
parasitism.
Similar results were reported by Atera et al.
(2012) who also showed a reduction in plant height of rice varieties
susceptible t
o
S
triga
in
Western Kenya. However, these varieties had an increase in plant height in 2014
at
113 days
after sowing. This could be related to low
Striga
numbers in
this this season than 2015
since
more
Striga
was added in the plots in 2015
.
Oryza glab
er
r
ima
varieties such as Makassa, ACC and MG12 with intermediate resistance
levels to
Striga
(Johnson
et al.,
1997)
had their stems not reduced
in both seasons
. These results
are similar to those
by Johnson et al. (1997)
who reported these varieties
as having
the lowest
stunted levels in Ivory
Coast
.
V
arieties WAB928 and WAB935
that are dwarf type had shor
t
39
stems in this study (Table1, 2
)
.
These varieties are also very resistant to
Striga
(
Johnson
et al.,
2000).
This is a clear indication that their short ste
ms are attributed to genetic makeup as
opposed to
Striga
infestation.
Other
Oryza glaberrima
varieties
Agee
, CG14, Anakila
had
intermediate stem heights.
Previous studies have shown that
O.glaberrima
cultivars, have a
different plant type than
O sativa
cul
tivars and are therefore often more competitive to weeds
(Johnson
et al.,
1998
).
NERICA varieties and Other
O.sativa
varieties
(SCRID090, Blechai,
WAB880, IRGC, IR49 and WAB181
–
18) were mainly tall because of
reduced
Striga
infestation
.
These
varieties
have been
reported to be resistant to
Striga hermonthica
(Harahap
et al.,
1993; Johnson
et al.,
1997; Rodenburg
et al.,
2015).
4.1.2
.2
Number of tillers
There
were
significant
(P< 0.001)
difference in the number of tiller produced per variety
at 43
days, 57 days and 113 days after sowing
for both seasons
(Appendix 3)
.
Variety x season
interaction effects on tiller number
was highly significant (Appendix 3).
In 2014,
there was a significant difference in number of tillers produced per
variety (
Appendix
8
). Variety WAB
928
had the highest number of tillers
(15
)
per plant while
Variety MG12
had
no tiller 43 days after sowing.
V
ariety
WAB928
produced
significantly (P<
0.001) the highest
number of tillers
(
36
)
at
57 days after sowing compare
d to
other varieties except CG14
with
31tillers per plant
(Table 3).
In the same period
(57 days)
, WAB56
–
104 produced the lowest
number of tillers
(9 tillers
per plant
)
.
At 113 days after sowing, variety CG14 and IAC 165
had the highest
(25
)
and lowest
(9
)
nu
mber of tillers respectively (Table 3).
The number of
tillers produced and present gen
e
rally increased form 43days to 57 days but decre
a
sed from
57days to 113 days after sowing. V
arieties Agee
, Makassa, MG12, NERICA
–
1, NERICA
–
10,
40
Table
3
:
Number of tillers of upland rice varieties grown under
Striga hermonthica
infested field conditions during the first rainy season of 2014 and 2015 at 43, 57 and 113
days after sowing.
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
3.30
1.32
4.00
3.81
S.E
2.74
4.55
2.70
1.04
3.00
3.01
CV%
35.3
27.5
18.0
25.4
24.0
24.39
41
NERICA
–
2, SCRID090, WAB880 and WAB56
–
104 had no reduction in tiller number (Table
3).
Similarly in 2015, varieties differed significantly (P< 0.001
)
in the number of tillers produced
(Appendix 10). Varieties CG14 and Sup
erica had the highest (8
and 25
)
and lowest (2
and 5
)
number of tillers
at 43 da
ys and
57 days after sowing respectively (Table 3). At 113 days after
sowing NARC3 had the highest number of tillers (27) while Superica
(local check) had the
lowest number of tillers (2
) (Table 3).
As in 2014
,
there was a high reduction in the number of
tillers produced per variety
from 57 days to 113 days after sowing
except for vari
eties Agee
,
Blechai, CG14, NARC
3, NERICA
–
10, NERICA
–
2, NERICA
–
4 and WAB88
0
(Table 3).
However, v
arieties IR49, WAB928, IAC 165,
WAB935 and Superica (local variety) had the
highest reduction in number of tillers compared to other varieties from 57 days and 113 days
after sowing (Table 3).
There was also a general decrease in the number of tillers produced per
pla
nt across varieties
in 2015 com
p
a
red to 2014.
The mean maximum above ground
Striga
numbers (NSmax) per variety correlated negatively with tiller number at 113days after sowing
per variety for both 2014 and 2015 (Spearman correlation coefficients r
2014
=
–
0.265 and r
2015
=
–
0.077
).
Number of tillers varied highly across varieties for 43, 57 and 113 days after sowing (Appendix
3). There was a reduction in tiller number produced per variety especially in 2015.
This
difference is more
to be
likely related to
Striga
parasitism since
more
Striga
was
a
dded in the
plots in
2015.
I
t 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
; Jamil
et al.,
20
12
) and also triggers
Striga
germination (Jamil
et
al.,
2011). This will therefore mean that varieties that produce high amounts of strigolactones are
more likely 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,
42
WAB56
–
50 and WAB56
–
104 since they were the most susceptible to
Striga
as per this study.
Additionally, WAB56
–
50 and IAC 165 wer
e ranked by Jamil et al. (2011)
as the highest
strigolactone producers. Studies by Jamil et al. (2011) ranked variety WAB56
–
104 as the lowest
strigolactone producer
though it supported high
Striga
numbers in this study
. This would
indicate differences in virulence levels in the
Striga
ecotyp
e in this
study and Jamil et al.
(2011)
study.
The study has shown that
O.glaberrima
varieties (A
CC, CG14, Makassa, Anakila, Agee
and
MG12) produced higher number of tillers per plant in both
seasons
(Table 3). Varieties Agee,
Anakila and CG14 have also been report
ed to produce a high number of tillers (Jamil
et al.,
2011; Jamil
et al.,
2012).
Jamil et al. (2012)
attributed the
high number of tillers among these
varieties to low strigolactone production. However varieties
Anakila and CG14
also showed a
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.
This because these varieties
were reported to be resistant to
Striga
(Jamil
et al.,
2012).
Notably varietie
s ACC, Makassa
and MG12 had a reduction in tillers form 57days to 113days after sowing
especially in 2015.
This re
d
uction could be related
to
Striga
infestation. These were also reported to have partial
resistance levels to
Striga
in Ivory Coast (Johnson
et al.,
1997).
The differences in tiller production among rice varieties cannot entirely be accounted
for
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
43
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
as opposed to
Striga
infestation
.
This study also indicated
a negative
correlation between
ma
ximum above ground
Striga
number (
NSmax
)
and the number of tillers
in 2014 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 t
hat
lower tillering varieties are higher strigolactone producers and are more susceptible to
Striga
(
Jamil
et al.,
2012
)
.
4.1.2.3
Rice biomass
Rice biomass was significantly (P<0.001)
different
across the different varieties of rice for
2014 and 2015 seasons (Appendix 4).
For both seasons, variety x season interaction effects on
rice biomass was highl
y significant (Appendix 4)
.
Generally there was
decrease
in rice biomass
in 2015
compared to 201
4 across varieties except for varieties
Blechai, CG14, IR49, IRGC,
NERICA
–
17, NERICA
–
2, WAB880, WAB935, MG12 and NARC3
In 2014, varieties
ACC with biomass of
246.10g
, Makassa (238.50g), WAB928 (230.70g)
had
significantly (P< 0.001) the highest rice bioma
ss compared to rest of the varieties except
WAB935 (214.00g)
while variety
UPR (110.80g)
had the lowest rice biomass in the same
season. However
,
for 2015
season, varieties IRGC (293.90g), WAB935 (251.40g) and ACC
(243.40g)
ha
d the highest rice biomass
while v
arieties
IAC
165
and
Superica
(
local check
) with
44g and 37.40g per rice plant
had
significantly (P< 0.001)
the lowest
rice b
iomass
compared
to the remaining varieties
(Table 4
).
44
Table
4
:
Plant biomass of upland rice varieties grown under
Striga hermonthica
infested
field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Rice biomass (g)
Rice variety
2014
2015
ACC
246.1
0
243.4
0
Agee
154.8
0
139.2
0
Anakila
152.6
0
108
.00
Blechai
176.5
0
211.8
0
CG14
184.1
0
204.8
0
IAC 165
111.7
0
44
.00
IR49
181.1
0
210.9
0
IRGC
177
.00
293.9
0
Makassa
238.5
0
225.2
0
MG12
–
231.1
0
NARC3
132.8
0
155.6
0
NERICA
–
1
157
.00
126.6
0
NERICA
–
10
137.9
0
131.5
0
NERICA
–
17
144.3
0
160.7
0
NERICA
–
2
134
.00
172.5
0
NERICA
–
4
155
.00
150.6
0
SCRID090
157
.00
140.4
0
Superica
116.5
0
37.4
0
UPR
110.8
0
99.5
0
WAB181
–
18
124.4
0
115.5
0
WAB56
–
50
150.1
0
133
.00
WAB56
–
104
122.8
0
103.9
0
WAB880
170.1
0
187.4
0
WAB928
230.7
0
223.3
0
WAB935
214
.00
251.4
0
LSD (0.05)
43.33
49.48
S.E
34.33
39.15
CV%
21.22
23.93
1
variety MG12 excluded from the analyis because the land was cultivated by farmers before
harvesting it.
45
Based on the rice biomass produced by each variety, varieties can be categorized into four
groups ie. High (200
–
300g), intermediate (150
–
200g), intermediate low (100
–
150g) and low
(<100g) rice biomass producers:
(1) high rice biomass producers, i.e.WAB935,
WAB928,
ACC and Makassa in 2014, ACC, WAB935, MG12, IR49, Makassa, Blechai, CG14, WAB928
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) an
d
WAB880, NERICA
–
17, NERICA
–
2, NERICA
–
4 and NARC3 (2015),
(3) intermediate low
rice biomass producers, i.e. NERICA
–
2, NERICA
–
17, NERICA
–
10, NARC3, WAB 181
–
18
and WAB 56
–
104 (2014) and WAB56
–
50, SCRID090, NERICA
–
10, NERICA
–
1, AGEE,
Anakila, WAB181
–
18 and W
AB56
–
104 (2015) and (4) the low rice biomass producers, i.e.
local check (Superica), UPR and IAC 165 (bot
h 2014 and 2015) (Table 4
)
Results from the current study indicate differences in rice biomass production across varieties.
Varieties such as IAC 165 a
nd Superica
(local check)
with lower rice biomass were also the
most
susceptible rice varieties
. It is therefore possible that
Striga
parasitism lowered their
biomass. Additionally, studies by Johnson et al. (1997) have reported
Striga
–
inflicted biomass
re
duction in IAC 165. However, these same varieties had high
er
biomass in 2014
compared to
2015
. This could be related to lower
Striga
number in this season compared to 2015
since more
Striga
seeds were added in the plots in 2015
. Also the high rice biomass among the
O.glabberima
varieties ACC,
Agee,
Maksssa and CG14,
NERICA varieties and
O.sativa
varieties IRGC, IR49, WAB928, WAB880, SCRID090 and Blechai is
related to reduced effect
of
Striga
on these varieties.
These varieties w
ere reported to resistant to
Striga hermonthica
(Harahap
et al.,
1993; Johnson
et
al.,
1997; Rodenburg
et al.,
2015).
It has also been proposed that a grass plant is a collection of tillers (Skinner and Nelson, 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,
46
NERICA varieties and
O.sativa
varie
ties because the former produce
a lot of tillers. Another
observation f
rom the study was that some varieties such as NARC3 (ITA 257), WAB56
–
50,
WAB56
–
104 and
O.glabberima
varieties ACC, Makassa, MG1
2
categorized as susceptible and
intermediate
to
Striga
respectively
(Table 1
and 10
)
had a high rice biomass
. Similarly, Johnson
et al.,
(1997) also reported a high rice biomass in
O.glaberrima
varieties Makassa and
ACC
with relatively high levels of
Striga hermonthica
infection. Varieties combining high
Striga
numbers with high host 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 dev
eloping on them.
Therefore
a combination of tolerance
and resistance is a very important strategy to improve rice crop
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
a
ssessed in the field as it would require infected and uninfect
ed control plants which was
provided for in another
study (see study three).
4.1.3
Grain yield
and yield
components of
upland
rice varieties grown under
Striga
hermonthica
infested
field
conditions.
4.1.3.1
Panicle dry weight
Panicle dry weight was significantly (P< 0.001) different across rice varieties for 2014 and
2015 seasons (Appendix 4).
Variety x season interaction effects on panicle dry weight was
highly significant (Appendix 4).
Generally there was an increase in panicle dry
weight in 2015
compared to 2014 except for varieties Anakila, SCRID090, NERICA
–
10, NERICA
–
1,
WAB56
–
104, WAB56
–
50, IAC165, NARC3 and Superica (local check) (Table 5).
47
Table
5
:
Panicle dry weight of upland rice varieties grown under
Striga hermonthica
infested field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Panicle dry weight (g)
Rice variety
2014
2015
ACC
433.60
610.20
Agee
541.00
610.90
Anakila
554.20
519.10
Blechai
459.10
580.40
CG14
362.50
694.70
IAC 165
377.00
49.60
IR49
233.90
455.80
IRGC
237.10
358.00
Makassa
493.70
572.80
MG12
–
414.30
NARC3
89.00
67.30
NERICA
–
1
583.00
543.20
NERICA
–
10
685.00
585.30
NERICA
–
17
500.40
649.20
NERICA
–
2
628.40
732
.5
0
NERICA
–
4
682.60
715.90
SCRID090
721.00
685.20
Superica
352.70
69.30
UPR
310.20
317.90
WAB181
–
18
680.90
716.10
WAB56
–
50
535.30
417.00
WAB56
–
104
575
.8
0
361.60
WAB880
584.90
725.50
WAB928
85.80
266.10
WAB935
54.60
335.60
LSD (0.05)
169.40
198.20
S.E
124.40
156.80
CV%
26.45
32.62
1
variety MG12 excluded from the analyis
in 2014
because the land was cultivated by farmers
before harvesting it.
48
In 2014, there was a significant difference in panicle dry weight per variety (Appendix 9).
Variety NERICA
–
10 (685g) had the highest panicle dry weight while variety WAB935 had
significantly (P< 0.001) lower panicle dry weight (54.60g) compared to other v
arieties except
WAB928 (85.80g) and NARC3 (89.00g) ( (Table 5).
In 2015,
varieties varied significantly (P< 0.00
1) in panicle dry weight
(Appendix 11).
Varieties NERICA
–
2
(732.50g)
and WAB880
(725.50g)
had the highest panicle dry weight
(Table 5). Notably, varieties IAC165 (49.60g), NARC3 (67.30g) and Superica (69.30g) had
significantly (P< 0.001) lower panicle dry weights compared to other varieties (Table 5).
However, there was no significant (P>
0.00
1)
difference
in panicle dry weight produced by
variety Superica
(local check)
(69.30g) and WAB928 (266.10g) (Table 5).
Results from the study indicate a reduction in panicle dry weight for varieties Superica and
IAC 165 especially in 2015. These same var
ieties had the highest
Striga
numbers per variety
in this same season
(Table 9 and 10)
. Therefore
,
since these varieties are very susceptible to
Striga hermonthica
(Table 1)
it is possible that their
panicle dry weight was reduced by
Striga
infestation
. Ho
wever
,
these varieties had their panicle dry weight not significantly reduced in
2014. This could
be
due
to
a lower
Striga
number in this season and hence less parasitism on
these varieties. Also
,
other varieties such as WAB928 and WAB935 despite being hig
hly
resistant to
Striga
(Johnson
et al.,
2000)
had lower panicle dry weight
especially in 2014
. This
could be related to lower yielding potential of these varieties or simply limited adaptability to
local conditions. The high panicle dry weight of NERICA
–
varieties
(NERICA
–
2, NERICA
–
10, NERICA
–
17
and NERICA
–
4),
Oryza sativa
varieties
(
WAB880, SCRID090 and Blechai
)
,
Oryza glaberrima
varieties
(
AGEE, CG14, Anakila
)
and NERICA parent WAB181
–
18 is
related to lower
Striga
number recorded on
these varieties or
simply due to thier
high yielding
potential. These same va
rieties have been reported as
high yielding (Rodenburg
et al
.,
2015;
Saito
et al.,
2012). NERICA
–
parents WAB56
–
50 and WAB56
–
104 in spite of
recording
high
49
Striga
numbers
(Table 9 and 10)
produced h
igh 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
Stri
ga
infested conditions in Kenya.
The panicle dry weight cannot be entirely attributed to
Striga
infestation. It could also be related
to differences in rainfall patterns in
the year
2014 and 2015 (Appendix 18). For example
varieties WAB935, WAB928, IRGC a
nd IR49 produced a high panicle dry weight in 2015 and
lower one in 2014. In 2014, these varieties were affected by
a
dry spell that set in at
post
–
anthesis. Therefore it is possible that some of their grains were not fully filled leading to lower
panicle
dry weights.
4.1.3.2
Harvest index
Harvest index per variety
was significantly different
across varieties in 2014 and 2015 seasons
(Appendix 4).
For both seasons, variety x season interaction effects on harvest index was highly
significant (Appendix 4).
Generally harvest index increased in 2015 compared to 2014 except
for varieties IAC165, Superica (local check), WAB56
–
50, WAB56
–
104 and NARC3
(Table 6)
.
In 2014, there was a signif
icant difference
in harvest index across rice varieties (Appen
dix 9).
Vari
ety NERICA
–
10 had significantly (P< 0.001)
the highest
(55%)
harvest index compared
to th
e rest of the varieties (Table 6
). However this same variety was not significantly (P> 0.001)
different
in harvest index from varieties
WAB181
–
18
(52.8%)
,
NERICA
–
2 (51
.5%),
SCRID090
(50.7%)
, NERICA
–
4
(
49.9%)
,
WAB56
–
104 (49.8%)
NERICA
–
1
(47.4%),
WAB56
–
50 (45.6%), Agee (43.7%) and
WAB880
(43.2%)
(
Table 6
). Additionally variety
WAB
935
had significantly
the lowest
(5%)
harvest index compared to o
ther varieties though
it
no
was not
significantly
(P> 0.001)
different
form
NARC 3 (7.8%) and WAB928 (12.5%)
(Table 6
)
.
5
0
Table
6
:
Harvest index of upland rice varieties grown under
Striga hermonthica
infested
field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Harvest index (%)
Rice variety
2014
2015
ACC
28.8
40.4
Agee
43.7
47.5
Anakila
38.0
43.3
Blechai
34.8
46.5
CG14
22.1
46.6
IAC 165
29.9
22.1
IR49
17.2
37.3
IRGC
19.9
23.1
Makassa
26.3
39.7
MG12
–
23.8
NARC3
7.8
4.0
NERICA
–
1
47.4
49.8
NERICA
–
10
55.0
55.3
NERICA
–
17
38.6
48.8
NERICA
–
2
51.5
55.4
NERICA
–
4
49.9
52.0
SCRID090
50.7
51.4
Superica
34.2
21.4
UPR
30.0
36.0
WAB181
–
18
52.8
56.8
WAB56
–
50
45.6
44.0
WAB56
–
104
49.8
44.2
WAB880
43.2
50.2
WAB928
12.5
18.0
WAB935
5.0
16.4
LSD (0.05)
12.0
10.7
S.E
8.3
8.4
CV%
21.8
21.7
1
variety MG12 excluded from the
analysis in 2014
because the land was cultivated by farmers
before harvesting it.
51
In 2015, varieties varied significantly in harvest index (Appendix 11). Variety WAB181
–
18
had significantly (P< 0.001) higher harvest index (56.8%) compared to the remaining varieties.
However, this variety’s harvest index was not significantly (P> 0.001)
different from variety
NERICA
–
10 (55.3%), WAB800 (50.2%), NERICA
–
2 (55.4%), SCRID090 (51.4%), NERICA
–
4 (52.0%), NERICA
–
17 (48.8%), NERICA
–
1 (49.8%), Agee (47.5%), Blechai (46.5%) and
CG14 (46.6%)
(Table 6
). On the other hand, variety NARC 3 had significant
ly (P<0.001) lower
(4.0%) harvest index compared to the rest of the varieties (Table 6).
Harvest inde
x is very fundamental in par
rti
ti
oning of assimilates between the shoot, grains and
consequently grain yield (Ishii, 1995). Therefore, improving on harvest
index can greatly
increase the yield of rice varieties. Grain harvest index varies greatly across locations, varieties,
seasons and ecosystems ranging from 35%
–
62% (Kiniry
et al.,
2001). Therefore any varieties
with a harvest index below 30% woul
d indic
ate limited assimilate pa
rti
ti
oning to the grains and
vice versa.
The lower harvest index of varieties IAC 165
(22.1%)
, NARC3
(4.0%)
and Superica
(21.4%)
especially in 2015
compared to
2014 is related
to the
lower grain yield
. These varieties are
suscepti
ble to
Striga
(Harahap
et al.,
1993; Rodenburg
et al.,
2015)
. Therefore their harvest
index was reduced by
Striga
parasitism. Since in this same 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 n
utrient uptake and assimilate pa
r
ti
tioning,
any alteration in this
process can alter assimilate pa
rti
ti
oning 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 the grain
filling phase/post
–
anthesis, they were affected by a dry spell especial
ly in 2014 which
sign
ificantly affected dry matter pa
r
ti
tioning to the grains. The higher harvest index among
varieties SCRID090, WAB181
–
18, NERICA
–
2, NERICA
–
4, WAB880 and NERICA
–
10 is due
52
to increased grain yield. This is also
an indication of increased p
a
r
ti
tioning of assimilates to the
grains. It is therefore not surprising that these varieties produced high yields as they also had
lower
Striga
numbers in the two seasons.
It is also possible that
the
rainfall difference
between
the two seasons is respons
ible for lower
harvest index among the varieties WAB928, WAB935, IRGC and IR49. This
is
so because in
2014 these varieties had a lower harvest index as compared in 2015. The prolonged dry spell
in 2014 during the post
–
anthesis stag
e affected the harvest in
dex of
these varieties. It is also
possible that these varieties are simply low yielding despite the high rice biomass produced,
therefore lowering harvest index.
4.1.3.3 Grain recovery
Grain recovery efficiency was significantly different among different
varieties of rice for 2014
and 2015 seasons respectively (Appendix 4).
For both seasons,
variety x season interaction of
rice grain recovery efficiency was significant (Appendix 4).
Generally grain recovery
efficiency was higher in 2015 compared to 2014 fo
r most varieties
except for varieties Agee,
IAC 165, NERICA
–
10, NERICA
–
17, NERICA
–
4, Superica (local check), WAB18
1
–
18,
WAB56
–
104 and WAB928
(Table 7)
In 2014, varieties varied significantly in grain recovery (Ap
pendix 9). Variety Agee and
NARC
3 had the h
ighest
(99.83%)
and lowest
(45.98%)
rice grain recovery efficiency
respectively compared to the
remaining varieties (Table 7
).
Similarly in 2015, there was a
significant
(P< 0.001)
difference in rice grain recovery efficiency across varieties (Appendix
11)
. Variety SCRID090 had the highest
(
98.35%)
rice grain recovery efficiency while variety
NA
RC3 had
the lowest
(53.06%)
rice grain recovery efficiency compared to th
e rest of the
varieties (Table 7
).
53
Table
7
:
Grain recovery efficiency of upland rice varieties grown under
Striga
hermonthica
infested field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Grain recovery (%)
Rice variety
2014
2015
ACC
82.94
88.89
Agee
99.83
94.99
Anakila
86.74
87.42
Blechai
83.78
96.44
CG14
74.74
93.57
IAC 165
89.65
88.62
IR49
61.97
85.05
IRGC
74.10
78.97
Makassa
78.11
84.98
MG12
–
72.95
NARC3
45.98
53.06
NERICA
–
1
92.03
93.28
NERICA
–
10
91.72
89.79
NERICA
–
17
90.84
90.64
NERICA
–
2
91.93
92.99
NERICA
–
4
92.21
89.45
SCRID090
95.41
98.35
Superica
87.22
83.37
UPR
80.86
85.02
WAB181
–
18
91.96
89.10
WAB56
–
50
89.66
90.41
WAB56
–
104
94.61
92.32
WAB880
90.14
93.10
WAB928
66.28
66.01
WAB935
55.20
66.43
LSD (0.05)
12.56
11.66
S.E
8.76
9.22
CV%
10.30
10.76
1
variety MG12 excluded from the
analysis in 2014
because the land was cultivated by farmers
before harvesting it.
54
Results
from the current study indicate difference in recovery efficiency across rice varieties.
Varieties NARC3 categorized as susceptible to
Striga
(Harahap
et al.,
1993) had the lowest
recovery efficiency in both seasons
. This is a clear indication that
this va
riety had a lot of empty
grains and consequently lower yield.
This could be related lower partitioning of assimilates to
the developing grains resulting from increased parasitism of
Striga
on this variety. It should
also be noted that even the most
Striga
resistant varieties had lower recovery 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 v
arieties produce a lot of tillers, assimilates 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 t
o limited moisture it can
greatly affect 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
(Table 10
and 7)
.
4.1.3.4 Rice grain yield
Rice grain yield
varied significantly among rice var
ieties for both seasons (Appendix 4
)
.
For
both seasons, variety x season interaction
effect
on grain yield
was highly significant
(Appendix 4)
.
Generally there was also an increase in yield from 2014 to 2015 across varieties
except for varieties
NERICA
–
10
,
Anakila
,
WAB56
–
50, WAB56
–
104,
Superica (local check)
and IAC165
(Table 8
). Varieties
WAB928 and WAB935 had the highest increase in yield from
2014
to 2015 (Table 8
). Varieties Superica and IAC 165 had the highest decrease in yield
compared to
other varieties
from 2014 to 2015
(Table 8
).
55
Table
8
:
Grain yield of upland rice varieties grown under
Striga hermonthica
infested
field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Grain yield (t/ha
)
Rice variety
2014
2015
ACC
1.82
2.93
Agee
2.57
3.15
Anakila
2.46
2.34
Blechai
1.98
3.10
CG14
1.37
3.46
IAC 165
1.54
0.25
IR49
0.78
2.24
IRGC
0.89
1.51
Makassa
2.01
2.72
MG12
–
1.78
NARC3
0.06
0.20
NERICA
–
1
2.88
2.89
NERICA
–
10
3.39
3.01
NERICA
–
17
2.33
3.36
NERICA
–
2
3.09
3.81
NERICA
–
4
3.41
3.72
SCRID090
3.71
3.75
Superica
1.61
0.35
UPR
1.11
1.61
WAB181
–
18
3.33
3.66
WAB56
–
50
2.59
2.14
WAB56
–
104
2.91
1.84
WAB880
2.82
3.70
WAB928
0.20
1.10
WAB935
0.04
1.22
LSD (0.05)
0.94
1.08
S.E
0.67
0.85
CV%
30.86
35.81
1
variety MG12 excluded from the
analysis in 2014
because the land was cultivated by farmers
before harvesting it.
56
In 2014, there was a significant difference in yield produced across varieties (Appendix 9
).
Variety SCRID090 with
3.71 t ha
–
1
produced significantly (P< 0.001) the highest y
ield
compared to the remaining varieties except
for
varieties NERICA
–
10 (3.39 t ha
–
1
), WAB181
–
18 (3.33 t ha
–
1
), NERICA
–
4 (3.41 t ha
–
1
), WAB880
(2.82 t ha
–
1
), NERICA
–
1 (2.88
t ha
–
1
),
WAB56
–
104 (2.91
t ha
–
1
) and NERICA
–
2 (3.09
t ha
–
1
) whose yield was not significantly (P >
0.001) different form
SCRID090 (Table 8). Varieties NARC3 and WAB935 with yield of 0.06
t ha
–
1
and 0.04
t ha
–
1
respectively
produced significantly the lowest yield compared to the other
varieties.
However the yield
of these varieties was not significantly different (P> 0.001) form
WAB928 (0.20
t ha
–
1
), IRGC (0.89
t ha
–
1
) and IR49 (0.78
t ha
–
1
) (Table 8).
In the same season,
yield per variety ranged from 0.04 t ha
–
1
to 3.71 t ha
–
1
with an average yield of 2.0 t ha
–
1
(
Table
8
). In 2014, the
Striga
infestation level was
low, the correlation between maximum above
ground
Striga
numbers (NSmax) and grain yield though negative was very weak and not
significant (r
2014
=
–
0.10; P= 0.640).
I
n 2015, there w
as a significant (P<
0.001)
difference
in yield across varieties
(Appendix 11)
.
Variety NERICA
–
2 with
3.8
1
t ha
–
1
p
roduced the highest yield. NARC
3 produced significantly
(P< 0.001) the lowest yield
(0.2 t ha
–
1
)
across varieties though it was not different form Superica
(0.35
t ha
–
1
)
, IAC165
(0.25
t ha
–
1
)
, WAB935
(1.22
t ha
–
1
)
and WAB
928
(1.10 t ha
–
1
)
(Table 8
).
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
maximum
above
–
ground emerged
Striga
plants (NSmax) and grain yield (r
2015
=
–
0.465; P=0.02) with more resistant varieties
showing t
he highest grain yield
.
Rice varieties combining excellent resistance to
Striga hermonthica
with high yields and
environmental adapta
bility would be very 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
maximum
number of emerged
Striga
plants (NSmax) and grain yield i
n
57
2015
(
(Spearman correlation coefficient
r
2015
=
–
0.465; P=0.02)
. 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 provi
de 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,
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, CG14,
Makassa, Agee
, ACC and Anakila were categorized as
high
to
intermediate high yielding.
These varieties are not only high tillering (Jamil
et al.,
2012) but also have a greater competitive
ability with weeds (Johnson
et al.,
1998
). The
s
e
characteristic
s
, combined with lower
Striga
numbers (NSmax) recorded per v
ariety 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 not reduced highly. This could imply a reduced effect of
Striga
on these
varieties
compared to
O.sativa
varieties IAC 165 and Superica.
Varieties SCRID090,
WAB880, 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 NERICA
–
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 has 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
58
varieties in order to improve yield (Roden
burg 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
). Another reason could be that they 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
(Table 9 and 10)
. Additionally these varieties had relatively high y
ields in
2014 probably due
to
reduced
Striga
infestation
.
The yielding ability of the varieties cannot be solely attributed to
Striga
infection. For example
inspite of the excellent resistance levels of varietie
s WAB928, WAB935, IR49 and IRGC
described in
studies (Harahap
et al.,
1993; Johnson
et al.,
2000)
. T
hese
varieties produced
lower yield
s
. This could be attributed to limited adaptability to the prevailing weather
conditions or simply because of an inherent lower yield potential. These varieties
progressed
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 2015
could be pa
rtly attributed to rainfall le
vels in
these years (Appendix 18
). Fo
r example, varieties Makassa, Agee
, ACC, Blechai, WAB935,
WAB928
, IRGC, IR49
and CG14
that produced lower yields in 2014
produced high yiel
ds in
2015.
4.1.4
Reaction of upland
rice varietie
s
to
Striga hermonthica
4.1.4.1
Striga
counts per variety
Striga
counts were highly significant (P<0.001) among different varieties of rice for 2014 and
2015 seasons (Appendix 5). For both seasons, variety x season interaction effects on
Striga
number was highly significant at 43 days and 57 days after sowing (Appendix 5).
59
However, at 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 effect bet
ween season and
variety at 85 days after sowing although the seasons were highly significant (P< 0.001)
(Appendix 5).
In 2014 at 43, 57, 71, 85 days after sowing and at rice harvest, variety NERICA
–
10 had the
lowest
Striga
numbers while Superica (local ch
eck) and IAC 165 had the highest
Striga
numbers compared to the rest of the varieties (Table 9).
In 2015, varieties differed significantly (P< 0.001) in the number of
Striga
plants supported per
plant at 43, 57, 71, 85 days after sowing and at rice harves
t (Appendix 12). In the same season
at 43 days after sowing, NERICA
–
10, NERICA
–
2 and WAB935 had no
Striga
plants
developing on them while Superica (local variety) and IAC 165 had the highest count of
Striga
plants (33) (Table 9). At 57 days after sowing,
NERICA
–
2, NERICA
–
10 and WAB928 had the
lowest count of
Striga
plants while Superica (local check) and IAC 165 with count of
Striga
plants of 115 and 127
had significantly the highest count of
Striga
plants though wasnot
significantly (P>
0.001) different from WAB56
–
50
(71)
. At 71 days after sowing, NERICA
–
10
and WAB928 had the lowest count of
Striga
plants whereas Superica (local check) had
significantly the highest count of
Striga
plants
. However,
Striga
plants recor
ded on
Superica
(loca
l check) was not significantly (P> 0.001) different from IAC165 (121) and WAB56
–
50
(151) (Table 9). At 85 days after sowing and rice harvest, NERICA
–
2 had the lowest (1) count
of
Striga
plants with Superica (local check), IAC 165 having the highest count o
f
Striga
plants
(164) and (208) at 85days and (148) and (115) at harvest respectively (Table 9).
The
results indicate differences in susceptibility and resistance levels among rice varieties.
Varieties IAC 165, Superica (local check), WAB56
–
50 with high
St
riga
numbers at 43 days in
both seasons is an indication of earlier
Striga
infestation on these varieties
.
60
Table
9
:
Number of
Striga
plants (
Striga
counts) per rice variety under
Striga
infested
field conditions in Namutumba, Eastern Uganda, 2014 and 2015.
Number of above ground
Striga
plants (
Striga
counts) were log transformed
(logx+1)
before
analysis to meet the assum
ption of Analysis of Variance.
Means were presented as original
data.
M
ean comparisons were done
basing on the transformed data.
Number of
Striga
plants
(m
–
2
)
2014 2015
Days after sowing
Days after sowing
Rice variety
43
57
71
85
Harvest
43
57
71
85
Harvest
ACC
1
2
6
12
7
2
5
12
23
13
Agee
0
2
4
9
10
1
2
6
10
12
Anakila
1
2
3
4
5
3
7
13
18
19
Blechai
0
0
3
6
7
3
4
8
9
7
CG14
1
1
3
5
3
1
6
13
17
13
IAC
6
18
89
141
234
33
127
121
164
115
IR49
1
2
4
6
3
5
6
9
13
10
IRGC
0
2
2
2
3
2
2
5
5
4
Makassa
1
2
4
10
8
4
13
23
34
28
MG12
0
0
3
11
13
1
6
18
27
13
NARC3
1
5
10
20
20
5
25
42
48
44
NERICA
–
1
1
4
13
26
28
4
8
13
18
18
NERICA
–
10
0
0
1
1
1
0
1
2
2
2
NERICA
–
17
0
0
2
4
3
3
4
5
10
13
NERICA
–
2
0
0
2
2
2
0
1
1
1
1
NERICA
–
4
0
2
2
4
4
2
3
5
7
7
SCRID090
1
2
3
3
3
2
3
4
5
4
Superica
20
21
94
118
118
33
115
151
208
148
UPR
3
4
4
8
11
2
5
12
14
13
WAB181
–
18
1
2
9
14
13
3
3
4
5
5
WAB56
–
50
14
15
38
70
84
14
71
98
101
97
WAB56
–
104
7
8
24
38
55
18
43
52
65
67
WAB880
1
2
4
5
6
2
3
5
7
10
WAB928
0
0
1
2
1
0
1
2
2
3
WAB935
1
2
4
4
3
1
3
4
4
4
Means
2
4
13
21
26
5
19
25
33
27
CV%
82.92
60.09
38.03
28.74
33.21
56.7
0
43.76
35.12
27.94
29.64
61
Normally an earlier infestation of
Striga
can
have tremendous affect on
growth of the plants
(Cechin and Press, 1993
;
Gurney
et al.,
199
9
;
Atera
et al.,
2012). Also these varieties are
deemed very susceptible to
Striga hermonthica
since there is an increas
e in
Striga
numbers
from 43 days up to harvest time. These varieties except Superica
(local check)
were reported
to be highly susceptible to
Striga
under field conditions (Johnson
et al.,
1997; Rodenburg
et
al.,
2015). The lower
Striga
numbers among variet
ies NERICA
–
10, NERICA
–
2, WAB928,
WAB880, SCRID090, NERICA
–
4, NERICA
–
17, IR49 and IRGC in both seasons is a clear
indication of the resistance of these varieties to
Striga
. These except WAB880 and SCRID090
were proved to be resistant to
Striga
in Kenya, Tanzania and Ivory Coast (Harahap
et al.,
1993;
Johnson
et al.,
1997; Johnson
et al.,
2000; Rodenburg
et al.,
2015).
The fact that a lower
number of
Striga
plants were attached to rice plants at 43, 57 and 71 days could indicate a
lower amount o
f strigolactones produced by these varieties.
Oryza glaberrima
varieties (MG12, ACC and Makassa) with a moderate count of
Striga
especially in 2015 were also reported by Johnson et al. (1997) as having partial resistance to
Striga
in Ivory Coast. The rema
ining
Oryza glaberrima
varieties (CG14, Anakila and Agee)
regesitered lower
Striga
counts in both seasons. These have
also been reported to be resistan
t
to
Striga hermonthica
(Jamil
et al.,
2012; Rodenburg
et al.,
2015). Other varieties with an
increase in
Striga
plants from 43 days to harvest were NARC3 and WAB56
–
104 all of which
were
categorized as being susceptible and intermediate
to
Striga
respectively (Harahap
et al.,
1993; Rodenburg
et al.,
2015)
.
It was also observed that there was a greater increas
e in
Striga
counts at 43 days, 57days, 71 days, 85 days and at rice harvest i
n 2015 compared in 2014 (Table
9
). This could be as a result of an increased
Striga
seed bank since more
Striga
was added in
each plot in 2015.
62
4.1.4.2
Days to
Striga
emergenc
e
and flowering
Rice varieties v
aried significntly (P< 0.001) in the days
to
Striga
flowering
and days to
Striga
emergency
for
both
season
s (Appendix 7)
.
Variety x
season interaction effect was highly
significant (P< 0.001) for days to
Striga
emergeny and d
ays to
Striga
flowering (Appendix 7).
Generally there was an increase in days to
Striga
emergency and flowering in 2015 compared
to 2014 (Table 10)
In 2014, there was a significant
(P< 0.001)
difference in days to
Striga
emergency and
flowering across vari
eties (Appendix 13). Variety WAB56
–
104 had the shortest time (35 days)
to
Striga
emer
gency while variety NERICA
–
2 (96 days)
took the longest time for
Striga
to
emerge
(Table 10). Additionally, variety Superica (local check) had
Striga
plants flowering
earlier at 64 days while variety MG12 flowered later at 105 days after sowing (Table 10).
Furthermore, the most resistant rice varieties to
Striga
such as NERICA
–
2, NERICA
–
10 and
WAB928 had
Striga
emerging later, with
no flowering of
Stri
ga
plants
recorded (Table 10).
In 2015, there was a significant (P<0.001) difference in days to
Striga
emergency and flowering
across varieties (Appendix 14). Var
iety IAC 165 had the lowest time
to
Striga
emergency and flowering i.e. 49 days and 79 days r
espectively (Table 10). On the other hand,
variety NERICA
–
10 had
Striga
taking the longest time to emerge and flower at 96 and 109
days after sowing respectively
(Table 10). Notably variety NERICA
–
2 did not have any
Striga
plants flowering during both sea
sons (Table 10).
Rice varieties varied significantly in days to first
Striga
emergency and days to
Striga
flowering
(Table 10). The average time taken by a
Striga
plant to complete its life cycle from emergence
to flowering varied greatly among seasons fro
m 68 days for 2014 and 90 days for 2015.
63
Table
10
:
Striga
components per upland rice variety in 2014 and 2015, Namutumba
district, Eastern Uganda.
2014
2015
Rice variety
DSE
(days)
DSF
(days)
NSmax
(m
–
2
)
Striga
dry
weight (g)
DSE
(days)
DSF
(days)
NSmax
(m
–
2
)
Striga
dry
weight (g)
ACC
51
71
13 (1.11)
4.37 (0.73)
66
96
24 (1.38)
6.41 (0.87)
Agee
58
74
11 (1.06)
3.57 (066)
71
92
10 (1.01)
3.17 (0.62)
Anakila
47
71
5 (0.72)
1.51 (0.40)
66
94
18 (1.25)
4.89 (0.77)
Blechai
78
100
8 (0.88)
1.88 (0.46)
59
94
9 (0.95)
6.41 (0.87)
CG14
65
80
5 (0.73)
2.09 (0.49)
78
97
17 (1.23)
2.98 (0.60)
IAC 165
41
68
208 (2.32)
76.62 (1.89)
49
79
171 (2.23)
94.50 (1.98)
IR49
55
75
5 (0.74)
1.88 (046)
62
95
12 (1.09)
10.22 (1.05)
IRGC
66
80
4 (0.55)
1.88 (0.46)
70
97
5 (0.67)
2.55 (0.55)
Makassa
59
78
12 (1.09)
1.95 (0.47)
70
93
35 (1.54)
12.18 (1.12)
MG12
72
105
15 (1.16)
5.17 (0.79)
70
91
28 (1.44)
4.50 (0.74)
NARC3
52
78
23 (1.36)
18.05 (1.28)
55
84
51 (1.71)
45.77 (1.67)
NERICA
–
1
51
78
31 (1.49)
7.71 (0.94)
62
94
18 (1.26)
11.72 (1.03)
NERICA
–
10
85
0
2 (0.28)
0.07 (0.03)
94
109
2 (0.36)
0.86 (0.27)
NERICA
–
17
79
82
5 (0.71)
1.82 (0.45)
68
96
10 (1.00)
7.51 (0.93)
NERICA
–
2
96
0
2 (0.31)
0.48 (0.17)
86
0
2 (0.24)
0.12 (0.05)
NERICA
–
4
54
73
4 (0.62)
1.14 (0.33)
67
94
7 (0.83)
3.27 (0.63)
SCRID090
45
68
4 (0.55)
2.24 (0.51)
70
98
5 (0.70)
2.31 (0.52)
UPR
52
74
11 (1.03)
9.23 (1.01)
57
82
14 (1.16)
12.18 (1.12)
WAB181
–
18
42
78
13
(1.19)
3.37 (0.64)
55
87
7 (0.85)
2.46 (0.54)
WAB56
–
50
59
72
86 (1.93)
37.90 (1.59)
53
101
105 (2.02)
78.43 (1.90)
WAB56
–
104
35
73
57 (1.76)
15.98 (1.23)
57
88
58 (1.77)
44.71 (1.66)
WAB880
69
86
6 (0.81)
1.69 (0.43)
63
96
7 (0.84)
4.37 (0.73)
WAB928
78
0
2 (0.32)
0.23 (0.09)
79
106
2 (0.32)
1.04 (0.31)
WAB935
64
72
4 (0.65)
1.51 (0.40)
84
98
4 (0.63)
2.24 (0.51)
Superica
38
64
138 (2.14)
62.10 (1.80)
51
81
214 (2.33)
127.82 (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
27.74
36.81
Maximum number of above ground
Striga
plants (NSmax) and
Striga
dry weight were log transformed (log x+1)
before analysis. Means presented as original data and results in the parenthesis was (log x+1) transformed. Mean
comparisons were done basing on the transformed data, the LSD and S.E presented is for the transforme
d scale.
DSE (Days to
Striga
emergency) and DSF (Days to
Striga
flowering)
64
Studies by Atera et al. (2012) reported the first
Striga
plant emerging 42
days
after rice
emergence with minimum of 56 days to complete the life cycle.
The current study however
recorded a shorter time of first
Striga
emergence (35 days) in 2014, and a longer time to first
Striga
emergence in 2015 (49 days). The difference in days to
Striga
emergence between Atera
et al. (2012) study and t
his study could
be attributed to
differences in rainfall amounts,
Striga
seed bank and rice genotypes.
Several studies have reported variation in dates of first
Striga
emergence. For example 21 days have been reported for sorghum in Sudan (Bebawi, 1981), 54
days have been
reported for sorghum in Mali (Clark
et al.,
1994), and 35 days have been
reported for sorghum and maize in Kenya (Gurney
et al.,
1995).
Low soil moisture content at
critical 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 emergence of
Striga
plants in 2015, the rains were about two weeks late
and lasted longer than in 2014. This partly
explains the higher
Striga
numbers counted in this season because
even after rice physiological
maturity, new
Striga
plants were still emerging. Overall, even the most susceptible cultivars in
2015 had their first
Striga
plants emerging later than in 2014.
There were also significant differences in
Striga
flowering dat
es across rice varieties with the
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, the susceptible varieties i.e.
Superica, IAC 165, WAB56
–
104 and WAB56
–
50
exhibited earlier emergence and flowering
of
Striga
plants.
This can be attributed to differences in susceptibility and resistance levels
among rice varieties. Similarly, Gebremedhin et al. (2000) reported earlier flowering and
emergence of
Striga
plants
in a susceptible sorghum cultivar compared to a resistant one.
Varieties without flowering of the
Striga
plants can greatly reduce on the
Striga
seed bank in
the soil. Such varieties can be incorporated in the farming systems with integrated
Striga
management methods.
65
4.1.4.3 M
aximum number of emerged
Striga
plants
(Nsmax)
Rice varieties ha
d a highly significant effect
on the maximum number of emerged
Striga
hermonthica
plants
(NSmax) in both seasons (Appendix 6).
For both seasons, the variety x
se
ason interaction effects were not significant, however the seasons were highly significant
(Appendix 6). In all seasons, mean maximum above
–
ground
Striga
numbers (NSmax) per
variety correlated positively and highly significantly (P<0.001) with the mean
Str
iga
dry
weights at harvest per variety (Spearman correlation coefficients were r
2014
=0.899 and
r
2015
=0.898).
In 2014, there was a significant (P< 0.001)
difference
in the maximum number of
Striga
plants
counted per variety (Ap
pendix 13).
Variety IAC165 rec
orded significantly the highest
maximum number
Striga
plants (210 m
–
2
) though it was not significantly
(P> 0.001)
different
from Superica
(138 m
–
2
)
.
Notably
, NE
RICA
–
10, NERICA
–
2 and WAB928 recorded
the lowest
maximum number (NSmax) of
Striga
plants. Based on the maximum above
–
ground
S.
hermonthica
numbers (NSmax) observed in the field in 2014; varieties can be categorized into
five separate groups: (1) very resistant, (2) moderately resistant, (3) intermediate, (4)
susceptible and (5) very s
usceptible. Varieties NERICA
–
2, NERICA
–
10, NERICA
–
17,
SCRID090, WAB880, Blecahai, WAB928, WAB935, Anakila, CG14, IR49, IRGC and
NERICA
–
4 having on average a mean
NSmax
of > 10 per m
2
, were classified as very resistant
while varietie
s MG12, UPR, ACC, W
AB181
–
18, Agee
and Makassa with NSmax of 10
–
20
per m
2
, we
re moderately resistant (Table 10
). Other varieties such as NARC3 and NERICA
–
1
with NSmax of 20
–
40 per m
2
were classified as intermediate (between resistant and
susceptible).Varieties WAB56
–
50 and W
AB56
–
104 with an NSmax of 40
–
100 per m
2
were
cl
assified as susceptible (Table 10
). The last group consisted of very susceptible varieties with
NSmax of 100
–
250 per m
2
i.e. IAC165 and
local check (Superica) (Table 10
).
66
I
n 2015
,
varieties differed significantly (P< 0.001) in the maximum number (NSmax) of
S
triga
plants produced per variety (A
ppendix 14). Variety
Superica
(local check) with NSmax of 214
m
–
2
recorded significantly (P< 0.001)
the highest number of maximum above groun
d
Striga
plants
though it was not significantly (P> 0.001) different from IAC 165 (171m
–
2
) and WAB56
–
50 (105m
–
2
)
. Varieties
NERICA
–
2, WAB928 and NERICA
–
10 had the lowest number of above
ground
Striga
plants (Table 10
). Based on the maximum above
–
ground
S.
hermonthica
numbers (NSmax) observed in the field in 2015; varieties WAB928, WAB935, NERICA
–
2,
NERICA
–
4, SCRID090, IRGC, WAB880, WAB181
–
18, Blechai and NERICA
–
10 having an
NSmax >10 per m
2
parasites were considered very resistant (T
abl
e 10
). Varieties Anak
ila,
Agee
, CG14, NERICA
–
17, NERICA
–
1, UPR and IR49 with an NSmax of 10
–
20 per m
2
were
considered moderately resistant. Also varieties Makassa, MG12 and ACC with NSmax of 20
–
40 per m
2
were considered intermediate (between resistant and s
usceptible) (Table 1
0
). More
so variet
ies WAB56
–
50, WAB56
–
104 and NARC
3 with an NSmax of 40
–
120 per m
2
were
considered susc
eptible. Lastly varieties IAC165
and Superica (local variety) with an NSmax
of 120
–
250 per m
2
were considered as very suscept
ible in the same season
(Table 10
).
Results from the field study indicated a highly significant difference in the number of emerged
Striga
plants (NSmax) among the 25 varieties screened in both seasons. This difference not
only depended on genetic makeup but also prevailing cli
matic conditions. 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 be resistant to
Striga
hermonthica
ecotype
under field conditions
from Mbita, in western Kenya, approximately
211 km south east of Namutumba
(Rodenburg
67
et al.,
2015). Similarly rhizotron experiments by Cissoko et al. (2011) have also ranked these
same varieties as having an exce
llent post
–
attachment resistance to
S.hermonthica
ecotype
found in Kibos, western Kenya and similar results have been reported by Jamil et al. (2011) in
his pre
–
attachment resistance study on an ecotype of
Striga hermonthica
from Medani (Sudan).
This impli
es that these varieties have excellent resistance levels to various ecotypes of
Striga
.
NERICA varieties being progenies of CG14 which was reported to be highly
resistant
to
Striga
in s
e
veral studies (Kaewchumnong
and Price
,
2008
; Rodenburg
et al.,
2015) must have
inheret
e
d the resistant genes
from this variety.
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.
Vari
eties WAB928 and WAB935 with an 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.
WAB928
has
showed an excellent post
germination
–
attach
ment
resistance to
Striga hermonthica
in our current study
(see study three). On the other hand
resistance found with varieties SCRID090 and WAB880 had not been reported before.
Varieties
IAC 165 and Superica were the most susceptible in this study.
For IAC 165, was used
as a susceptible check variety in this study and it exhibited high susceptibility levels confirming
studies 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
. On ther hand
Superica (
local check
)
variety, was very susceptible probably due to continuous cultivation of
the same variety in the region. Therefore it is likely that the virulence levels of local
Striga
populat
ion against this variety increased with time. Similar insights were provided by
Rodenburg et al. (2015) on variety Supa India which was very susceptible in Kyela but resistant
at Mbita in Kenya where it had not been grown before. The
O.glaberrima
varieties
(Makassa,
MG12, and ACC) had an intermediate level of resistance to
S.hermonthica
especially in 2015
.
68
Similar results were reported by Johnson et al. (1997) who also showed varieties ACC and
Makassa having a partial resistance to
S.hermonthica
. Studies ha
ve also shown that
O.glaberrima
varieties have an excellent competitive ability over weeds (Johnson
et al.,
1998).
This could partly explain the lower
Striga
plants on these varieties. Furthermore, 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
–
103
–
80
–
1
–
2 have been reported by
Harahap et al.
(1993) to be
Striga
resistant. This study has however demonstrated moderate resistance of
UPR
–
103
–
80
–
1
–
2. Varieties NARC 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 257)
as being susceptible to
S.hermonthica
ecotype in Kenya. On the other hand,
WAB56
–
50 was ranked by Jamil et al. (2011) as the highest strigolactone producers. Since
strigolactones trigger the germination of
Striga
, it is not surprising that this variety su
pported
a high number of
Striga
plants as per this study. Variety NERICA
–
1 classified as intermediate
and moderately resistance to
Striga
in 2014 and 2015 respectively in this study was classified
as resistant to
Striga
by Cissoko et al. (2011), Atera et a
l. (2012) and Rodenburg et al. (2015).
This could be related to differences in ecotype virulence of
Striga hermonthica
to NERICA
–
1
in this study.
4.1.4.4
Striga
dry weight
Striga
dry weight varied significantly among the rice varieties for the two seasons
(Appendix
6).
For both seasons, variety x season interaction effects on
Striga
dry weight was not
significant though seasons were highly significant (Appendix 6).
Generally
there was an
increase
in
Striga
dryweight
from 2014 to 2015
.
69
In 2014, rice variet
ies varied significantly (P< 0.001)
regarding dry weight of collected
Striga
plants (Appendix 13).
Varieties IAC 165
with
76.62g
Striga
dry weight, Superica (local check)
with 62.10g and WAB56
–
50 with 37.90g produced significantly
the
highest
Striga
dry
weight
at harvest c
ompared to remaining varieties
. However
Striga
dry weight for
WAB56
–
50 was not
significantly different from NARC 3 (18.05g) and WAB56
–
104 (15.98g)
(
Table 10
).
NERICA
–
10 had the lowest (0.07g
)
Striga
dry weight at har
vest in the same sea
son (Table 10
).
I
n 2015, t
here was a significant difference
in
Striga
dry weight recorded per variety (Appendix
1
4). Variety
Superica
(local check) (
127.82g
)
had significantly (
P< 0.001) the highest
Striga
dry weight at harvest compared to the rest of the
varie
ties
.
However,
Striga
dryweight of
variety
Superica
(local check) was
not significantly different from varieties IAC 165
(94.50g)
and
WAB56
–
50
(78.43g)
(Table 10
). Variety NERICA
–
2
with 0.12g
had
significantly the lowest
Striga
dry weight at harvest
compared to the rest of the varieties except NERICA
–
10 (0.86g)
and WAB928 (1.04g) with
similar
Striga
dry weight (P> 0.001)
as
NERICA
–
2
(Table 10
).
Results showed that there was a highly significant positive correlation in NSmax and
Striga
dry weight at h
arvest in both seasons (Spearman correlation coefficients were r
2014
=0.889 and
r
2015
=0.898), confirming Rodenburg et al. (2015)
who also demonstrated a similar relationship
.
The highly susceptible varieties such as IAC 165 and Superica
(local check)
did no
t only have
the highest
Striga
numbers but also had the highest
Striga
dry weight at harvest. This can be
explained by the earlier infection of
Striga
as a result of increased strigolactone production.
S
imilarly, Van Ast and Bastiaans,
(2006) reported an increase in
Striga
dry weight when
sorgh
um plants were infected with
Striga
at 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 with high numbers of emerged
Striga
plants had a high
Striga
dry
weight at harvest though not as high as IAC 165 and Superica
(local check)
. NERICA
–
1 and
UPR
–
103
–
80
–
1
–
2 also had relatively high
Striga
dry weights compared to the rest of the
70
varieties th
ough not as high as WAB56
–
50, WAB56
–
104 and NARC3 (ITA 257). The rest of
the varieties had a lower
Striga
dry weight at harvest. This could be attributed to
not
only a
lower number of
Striga
plants but also to the small size of the parasitic plants on thes
e varieties
.
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 38 rice farmers were involved in the upland rice variety selection exercise of these,
21 (
55%
)
were men and
17 (
45%
)
women (Table 11). The largest number 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 above 55 years (11%) (Table 11).
There was a wide variation in their experience in rice
farming with 45%
having
4
–
6 years’ of experience in g
rowing upland rice, 29% (
1
–
3 years
experience)
,
24%
(
7
–
9 years
experience) and 3%
with (
10
–
12 years’ experience
)
(Table 11).
This study revealed t
hat the largest number of upland rice farmers who turned up for the
participatory variety selection (PVS) were men (Table 11). These results are similar to
Nanfumba
et al.
(2013) and Adeke
nle et al. (2013) who also found lower numbers of women
participatin
g in variety selection for rain fed lowland ecologies and upland rice respectively.
The lower involvement of female farmers is attributed to social economic constraints including
resource en
dowment, capital and land (Adeke
nle
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 other 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 grou
ps constituted the largest percentage share of farmers (Table 11). This
suggests that there is low involvement of the youth and elderly in rice farming.
These findings
are similar to those of Addison et al. (2014) who also reported a lower youth and elderl
y
involvement in rice farming in low land rice ecologies in Ghana.
71
Table
11
:
Characterization of upland rice farmers’ interms of gender, age and rice
farming experience, Namutumba district, Eastern Uganda,
2015.
On the other hand, the study indicated that, out of 38 farmers in the PVS, the largest percentage
of farmers had experience of 4
–
6 years followed by those with 1
–
3 years
(Table 11). This
finding is in agreement with a survey carried out in 2005 by Advanced Studies on International
Development (FASID) and Makerere University which revealed that experience in rice
production ranged between 1
–
3 years
in Eastern Uganda (Kijima
and S
erunkuma
, 2013).
L
ow
experience in
upland rice growing
could be attributed to dropout of NERICA’S which are the
most cultivated upland ric
e varieties in Uganda or to
lack of
a
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
2000 Network and these areas lacked seeds even in input stores. From the discussion with
Variable
Number of respondents
N=38
Proportion
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
Rice farming
experience
1
–
3 years
11
29
4
–
6 years
17
45
7
–
9 years
9
24
10
–
12 years
1
3
72
farmers, they claim to ha
ve got their seeds from NGO’S and that they lacked seeds for
cultivating in their fields.
4.2.2 Rice
variety selecti
on
by farmers
basing on morphological appearance in the field
The most preferred variety of rice was SCRID090 (
SCRID090
–
60
–
1
–
1
–
2
–
4)
(18%), this was
followed by NERICA
–
17 (16%), WAB181
–
18 (14%) and NERICA
–
10 (13%). About 11%
preferred Blechai
(
IRGC78281
) and WAB880 (
WAB880
–
1
–
32
–
1
–
1
–
P2
–
HB
–
1
–
1
–
2
–
2
). 10%
selected NERICA
–
4, 5% NERICA
–
2 and only 2% selected NERICA
–
1 as the most preferred
var
ieties
(Figure 4)
Pereference of upland rice varieties in Eastern Uganda by geneder is presented in Figure 5.
A
t
otal of 18% and 16% of women selected SCRID090 and NERICA
–
10 respectively, as the best
two most preferred varieties, accounting for
1
3
of all
the most
–
preferred varieties by women
farmers (Figure 5). The most
–
preferred varieties by women were SCRID090, NERICA
–
10,
WAB181
–
18, NERICA
–
17, NERICA
–
4, Blechai, WAB 880, NERICA
–
2 and NERICA
–
1 in
that order. Among male farmers, 18% selected NERICA
–
17
as their most
–
preferred variety and
17% selected SCRID090 as their second best variety (Figure 5). For men the order of
preference was: NERICA
–
17, SCRID090, Blechai, WAB880, WAB181
–
18, NERICA
–
10,
NERICA
–
4, NERICA
–
2 and NERICA
–
1 (Figure 5). Generally there
were significant
differences (t= 4.803 P< 0.001) in variety preferences between men and women, for example,
12% women preferred NERICA
–
4, compared to 9% of men selecting this same variety (Figure
2). NERICA
–
4, was preferred by 13% of the men and only 10%
of the women. 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).
73
Figure
4
:
Farmer preference of upland rice
varieties in Namutumba district, Eastern
Uganda in 2015, bars indicate the percentage (%) of farmers preferring a certain variety.
Figure
5
:
Rice variety selection by gender of rice farmers in Namutumba district, Eastern
Uganda,
2015, data presented as percentage responses by farmers.
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
Proportion of rice farmers selecting a variety (%)
Upland rice varieties
0
5
10
15
20
NERICA 1
NERICA 2
NERICA 4
NERICA 10
WAB 880
BLECHAI
WAB 181-18
SCRID090
Proportion of rice farmers selecting a variety (%)
Rice varieties
Female
Male
74
The study showed that out of the 25 varieties of rice in the field trial, the most preferred
varieties of rice in descending 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
demonstrated suitability to local Ugandan conditions. Nanfumba et al. (2013) also report
ed
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 were WAB181
–
18,
NERICA
–
4, NERICA
–
2 and NERICA
–
1 (Figure 4). These varieties also exc
ept WAB181
–
18
and NERICA
–
2 have been grown in Uganda before. From this study all the NERICA varieties
featured among the preferred varieties by farmers. These results are similar to Gridley et al.
(2002) who also found out in PVS (Participatory Variety Sel
ection) 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 was 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 productivity in farmers’ fields (Gridley
et
al.,
2002), this could be t
he reason why they were highly preferred by farmers.
Overall, men selected NERICA
–
17 as their best rice variety whereas women selected
SCRID090 as their best variety
(Figure 5)
.
This difference in
preferences could be related to
women’s and men’s roles in the rice 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 were important cha
racteristics of the varieties men chose. Dorward et al.
(2007) also found that similar selection criteria was 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 and they prefer a variety that is early maturing to reduce
75
on farm work load. It is therefore not surprising that they picked SCRID090, WAB181
–
18 and
NERICA varieties since these varieties were all early maturing. Dorward et al. (
2007), also
reported that early maturity and yield were the most important traits women used in selection
of upland rice varieties in Ghana. Whereas Addison et al. (2014) reported early maturity as the
most important trait in varietal preferences among wom
en in low land rice ecologies in Ghana.
4.2.4 Characteristics of the least preferred
varieties of rice according to
farmers.
The least preferred
varieties according to fa
rmers were Superica and Anakila.
These
two
varieties
were selected by 20%
of
the
far
me
rs as least
–
preferred (Table 12
). These were
followed by varieties
UPR
(
UPR
–
103
–
80
–
1
–
2)
and IAC 165 which
were
least preferred
by
(19%)
and 17%
of
the
farmers respectively (Table 12
).
Lastly
among the least preferred
varieties were NARC 3 (ITA 257) and Ma
kassa
selected by 16%
and 8% of the farmers
respectively
(Table 12
).
This study also revealed
the five least preferred varieties, which included Anakila, Superica,
Makassa
, IAC165, UPR and NARC3 (Table 12
).
Superica (local variety) and IAC165, were
very susceptible to
Striga
. These varieties would therefore give low yields especially in these
areas which are highly infested with
Striga.
Furthermore, Anakila and Makassa were low
yielding with weak stems and he
nce they could be easily attached by soil pests as they lodge
easily. These two varieties also produced small seeds which could explain the low yields.
However, Anakila was early maturing while Makassa was late maturing.
Other varieties like
UPR
(
UPR
–
103
–
8
0
–
1
–
2) and NARC3
were not only judged for being late maturing with short
stature but also had small seed sizes and were susceptible to
Striga
.
According to the farmers,
the short stature makes it difficult to harvest these rice varieties especially in are
as where
farmers use knives for harvesting. NARC3, despite havi
ng a good aroma, was
less preferred
for being susceptible to drought.
76
Table
12
:
Reasons for disliking some rice varieties as indicated by farmers in Eastern
Uganda in 2015.
These results show that farmers dislike low yielding, late maturing, varieties with weak stems,
varieties with small grains and those susceptible to
Striga
, drought and lodging. These results
are similar to those by Nanfumba et al. (2013) who also indicate
d similar reasons for farmers
disliking rice varieties for low land ecologies in Uganda.
4.2.5
Characteristics of preferred rice varieties according to farmers
Striga
resistance and maturity period, with the highest mean score
of
4.86 and 4.81
respe
ctivel
y, were the most preferred characters for farmers’
select
ion criteria of
rice varieties
(Table 13
). Other characters such as yield (4.62), tillering ab
ility (4.42), drought tolerance
(4.38) and height (4.33) were also considered by farmers as very importan
t
a
ttributes in
selection (Table 13
).
Resu
lts
indicated that when farmers get new varieties, they often compare them with those
currently grown on the basis
of
certain characteristics. These were mainly
Striga
resistance,
maturity period, yield, height, tillering
ability
, drought t
olerance and grain
colour and
size
Variety
% of farmers
indicating
dislike
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
,
Weak stems making
it susceptible to lodging
NARC3
16
Late maturing, susceptible to Striga, short
variety (difficult to harvest, easily
overwhelmed by weeds), small sized grains, susceptible to drought
77
(Table 13
). Among these traits,
Striga
resistance, early maturity, tillering and
high
yield were
the most important criteria for rice variety select
ion in Eastern Uganda.
Table
13
:
Criteria rice farmers used 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
Maturity
4.81
0.40
2
Yield
4.62
0.63
3
Good tillering ability
4.42
0.50
4
Drought tolerance
4.38
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 all
3 = 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.
Farmers prefer high yielding,
Striga
resistant rice varieties that mature ea
rly 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
rice
farmers in 17
countries PVS in 1999 in West Africa (Gridle
y
et al.,
2002). Surprisingly, in most studies high
yield is presented as the most important criteria in rice variety selection (Nanfumba
et al.,
2013;
Gridley
et al.,
2002; Kimani
et al.,
2011).
However, in this study,
Striga
resistance emerged as
the most important trait and this could be attributed to this area being infested with
Striga
. A
survey done in 2008 indicated that
Striga
was a serious problem in upland rice production in
Eastern Uganda (Pittchar and Mbeche, 2008
u
npublished information
).
The local rice variety
78
(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.
4.2.6
Variety ranki
ng based on farmer selection criteria
Table 14
shows farmers’
ranking of rice varieties based on most preferred traits.
Results
indicated
that
SCRID090 with
the highest
phenotypic
acceptability among rice farmers
and
with
mean score of 2.76
was ranked as t
he best preferred
variety. This was followed by
WAB880 with (2.75) and NERICA
–
17 (2.63) which were ranked as second
and third
best
preferred
respectively (Table 14
).
Blechai
with mean score of 2.60 and W
AB
18
1
–
18 with mean
score of 2.58 were ranked as fourth and fifth
preferred varieties
respectively. NERICA
–
4 with
(2.54) and NERICA
–
1 with (2.52) were ranked as sixth and seventh
preferred varieties
respectively. NERICA
–
10
with mean score of 2.43 and NERICA
–
2
with mean score of 2.37
were ranked as eighth
and ninth
preferred varieties
respectively (Table 14
).
In relation to farmer selection criteria, farmers’ ranked varieties SCRID090, WAB880,
NERICA
–
17, Blechai and WAB181
–
18 as the be
st variet
ies in ascending order.
These
were
ranked as the five best
preferred
rice
v
arieties in this study (Table 14
). Other varieties also
ranked by farmers were NERICA
–
4, N
ERICA
–
1, NERICA
–
2 and NERICA
–
10.
These
were
ranked sixth’s, seventh’s, eighth’s and ninth’s
respectively. Farmers ranked SCRID090 and
WAB800 as their best
preferred varieties because of
their
unique attributes. T
hese varieties
were not on
ly early maturing but also had
good grain size and tillering ability which are all
traits related to
high
yie
ld.
These constitute the most 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 rice varieties to resists or perform despite
Striga
infestation is ano
ther reason
for farmers selecting them (see study one). Farmers selected NERICA
–
17 because of its big
grain size as well as being drought tolerant and
Striga
resistant.
79
Table
14
:
Ranking of selected upland rice varieties based on
farmer selection criteria in Namutumba district, Eastern Uganda, 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
variety
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
80
Variety Blechai was chosen due to its height and
Striga
resist
ance. It has been reported by
several studies that farmers prefer tall varieties 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 because of its big si
zed grains and early maturity. Early maturing varieties
are desired because they are seen as drought escaping options (Nanfumba
et al.,
2013). On the
other hand NERICA
–
4 and NERICA
–
1 were ranked sixth and seventh respectively due to both
being high yieldin
g and drought resistant. All these are characteristic traits of 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 cultivated in the short rainy
seasons. These are characteristics
of the Ugandan conditions. However, NERICA
–
10 was high
yielding though not as resistant as NERICA
–
2 to
Striga
.
81
4.3
Study three:
E
valuation of post
–
germination
attachment resistance of upland rice
varieties to
Striga hermonthica
u
nder controlled environment conditions
4.3.1 Effect of
Striga
infection on the morphology and growth characteristics of the
host
rice plants.
4.3.1.1 Effect on p
lant height
Plant heights of the infected rice plants were compared with those of the uninfected plants to
evaluate the impact of
Striga hermonthica
on the gro
wth of the rice plants (Table 15
). Rice
plant heights were significantly (P<0.001) affected by
Striga
infecti
on for varieties IAC 165,
Superica (local variety) and WAB880 (Appendix 15
and Table 15
). These varieties had
significantly shorter stems compared to the
ir respective controls (Table 15
). The plant heights
of the rest of the varieties
were not significantl
y affected
comp
ared to their controls (Table 15
).
Varieties Superica, IAC 165 and WAB880 had
a significant reduction in plant height
of 44.6%,
33.9%, and 23.6% respectively when infected with
Striga hermonthica
(Table 15
).
Notably,
varieties SCRID090,
NERICA
–
1 and WAB56
–
50 registered no significant reductions i
n their
plant heights when infec
ted with
Striga
(Table 15
). Irrespective of the treatments, variety
SCRID090 and Blechai had the tallest plants while NERICA
–
17 had shortest plants (Table
15
).
Thi
s study
indicated that varietie
s Superica
(local check)
, IAC 165 and WAB880
had
significantly high reductions in their stem height following
Striga
infection compared to
control plants.
These results are consistent with studies by
Cechin and Press,
(
1994
)
;
Walting
and Press,
(
2000
)
; Swarbrick
et al.
(
2008
)
and
Atera
et al.
(
2012
)
who also indicated that cereals
such as rice exhibit characteristic changes in plant morphology and architecture
including
stunting /reduction in stem length when infected with
Striga
species
as compared to uninfected
plants
.
82
Table
15
:
Effect of
Striga hermonthica
infection
on
plant h
eight of
upland
rice varieties
grown under controlled
(growth chamber)
condition
s
at 21 days after infection
.
Plant
height (cm)
Treatment
Rice variety
Striga
infected
Uninfected
(
control
)
Variety means
% reduction
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
9.0
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
14.08
a
15.78
b
15.1
0
LSD
3.04
S.E
2.16
CV%
14.3
Means followed by the same letter
are not significantly different us
ing Tukey multiple
comparison
test (P< 0.05)
.
83
Stunting could be as a result of lack of internode elongation rather than decrease in internode
numbers. Internode elongation is based on increased cell elongation in well
–
delineated zones
of the internode (Kende
et al.,
1998).
Alteration
of growth regulators metabolism is one hypothesis that may account for this change
in stem heights following infection. A number of growth regulators are involved in elongation
of stems in plants, including g
ibberellins, cytokinins and auxins (Kende
et
al.,
1998; Sakamoto
et al.,
2006; Ikeda
et al.,
2001). Assuero and Tognetti
, (2010) reviewed a number of studies on
the effect of growth regulators on tillering. These two 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 lack 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 a
Swarbrick et al. (2008) many genes
involved in auxin and gibberellin signaling are down regulated following
Striga
infectio
n.
Therefore, reduction in internode extension in
Striga
infected plants could be the result of an
alteration in gibberellin and auxin signaling, hormones important in stem elongation.
Another explanation for stunting of rice plants after infection
with
S
triga
has been suggested
to be related to translocation of toxic compound from
S.hermonthica
to the host (M
usselman
and Press, 1995). Some
studies have
also
suggested the involvement of secondary metabolites
toxic to cereals to have an effect on host morph
ology (Ejeta and Butler, 1993). These together
with Rank et al. (2004) revelation of irioid glucosides and their suppression of cell division can
also explain this impact of
Striga
infection on plant
height
s
. However
,
such toxins have not
been identified and evidence of this mediated effect of this host has not been presented.
84
4.3.1.2
Effect on s
tem
diameter
The stem diameter of rice varieties
IAC 165, Superica (local check), WAB181
–
18 and
WAB880 were significantly (P<
0.05) effected by
Striga
infection (Table 16). These varieties
had significantly thinner stem diameters compared to their respective control plants (Table 16).
There was no significant difference in stem diameter (P> 0.05) of the infected plants of the
re
maining varieties compared to their non
–
infected controls. Notably, there was a reduction in
stem diameter when rice plants were infected with
Striga.
However,
varieties varied greatly
with some having significantly reduced stem diameters such as WAB880 (4
2.7%), IAC 165
(39.1%)
WAB181
–
18 (30.5%) and Superica (48.4%) (Table 16). Other varieties with moderate
reductions in stem diameter following
Striga
infection were Blechai (29.1%), IRGC (24.5%)
and CG14 (27.4%) (Table 16)
.
The rest of the varieties had sma
ll reductions in stem diameter
(<20%) with WAB56
–
50 having regis
tered no reduction in stem diameter
(Table 16).
Results presented indicate reductions in stem diameter of rice plants infected with
Striga
hermonthica
when compared to uninfected control p
lan
ts across varieties (Table 16
). However,
varieties Superica, IAC165
, WAB181
–
18
and WAB880 had a significant r
eduction in stem
diameter
(Table 16
). These varieties except WAB880
and WAB181
–
18
supported a high
parasite load. This is only an indication of the great effect of
Striga
on these varieties when
compared to other varieties in
the study. The
high reduction in stem diameter of these rice
varieties could be attributed to earlier infection
by the parasite. Studies by Cechin and press,
(1993) demonstrated that earlier attachments of parasites had a greater effect on host growth
than later attachments. Other varieties with less or no reduction in stem diameter such as
WAB928, WAB56
–
50, WAB56
–
104, NERICA
–
1, SCRID090, NERICA
–
4 and NERICA
–
17
exhibited less effect of the parasitic weed.
85
Table
16
:
Effect of
Striga hermonthica
infection on s
tem diameter
of
upland
rice
varieties
grown under controlled
(growth chamber)
conditions
at 21 days after infection
.
Stem diameter
(mm)
Treatment
Rice variety
Striga
infected
Uninfected
control
Variety means
% reduction
in diameter
BLECHAI
3.54
4.99
4.27
ab
29.1
CG14
3.05
4.20
3.62
ab
27.4
IAC 165
3.22
5.29
4.26
ab
39.1
IRGC
4.06
5.38
4.7
2
ab
24.5
NERICA
–
1
4.03
4.09
4.06
ab
1.5
NERICA
–
17
3.28
3.45
3.37
a
4.9
NERICA
–
4
3.41
4.03
3.72
ab
15.4
SCRID090
5.00
5.95
5.48
b
16.0
SUPERICA
2.43
4.71
3.57
a
48.4
WAB 181
–
18
3.94
5.67
4.80
a
b
30.5
WAB 56
–
104
4.17
4.59
4.38
ab
3.9
WAB
56
–
50
4.21
3.94
4.07
ab
0
.0
WAB
880
3.51
6.12
4.82
ab
42.
7
WAB
928
4.30
4.53
4.41
ab
5.1
Means
3.72
a
4.78
b
4.25
LSD
1.54
S.E
1.10
CV%
25.8
Means followed by
the
same letter
are not signi
ficantly different using Tukey multiple
comparison test (P
< 0.05)
.
86
Another possible explana
tion for reduction in stem diameter
of infected
rice plants compared
to the uninfected control plants is related to limited assimilate
partitioning to other parts
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 (Gurney
et al.,
1999).
These
nutrients
are responsible for the entire
growth
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
div
ersion of nutrients
which could be used for stem enlargement
to parasiti
c use
.
4.3.1
.
3
Effect on n
umber of tillers
There was generally a significant (P< 0
.05
) reduction in the number of tillers produced by the
infected plants compared to uninfected ones (Table 17). A
fter
Striga
infection the local variety
(Superica)
had
significantly
a high reduction in number
tillers
compared to the rest of the
varieties
(Table
17
).
There was no significant (P> 0.05) difference in the number of tillers
produced by the remaining rice varieties when infected with
Striga
c
ompared to
their
u
ninfected control plants
.
Number of tillers for t
he control plants ranged from
0.25 tillers for
Blechai to 2.5 tillers for
WAB 181
–
18
. For infected plants, tiller number ranged from no tillers
for the local variety
(Superica)
to 2 ti
llers per plant for CG1
4
which was highly resistant
(Table
17
). The results show
ed
that
irrespective of the treatments
,
variety WAB181
–
18 produced the
highest
(2.25)
number of tillers compared to
Blechai, Superica and WAB56
–
104
(Table 17
).
This study has indicated less tiller
production among rice plants infected with
S.hermonthica
compared to u
ninfected control plants (Table
17
). The
se
results are similar to Cissoko
et al.,
(2011) who showed that infection of rice plants with
Striga hermonthica
and
Striga asiatica
suppressed tillering of infected plants compared to the uninfected plants across rice varieties.
87
Table
17
:
Effect of
Striga hermonthica
infection on tillering of selected upland rice
varieties grown
under controlled
(growth chamber)
conditions at 21 days after infection.
Number of tillers
Treatment
Rice variety
I
nfected
with
Striga
Un
infected
(Control)
Variety
means
%r
eduction
in tillers
BLECHAI
0.25
0.25
0.25a
0.0
CG14
2
.00
2
.00
2
.00
bc
0.0
IAC 165
0.5
0
1.25
0.88
abc
60
.0
IRGC
1.5
0
1.25
1.38
abc
0.0
NERICA
–
1
1.5
0
1.25
1.38
abc
0.0
NERICA
–
17
0.5
0
1.25
0.88
abc
60
.0
NERICA
–
4
0.75
1.25
1
.00
abc
40.0
SCRID090
1
.00
1.75
1.38
abc
42.9
SUPERICA
0
.00
1.5
0
0.75ab
100.0
WAB 181
–
18
2
.00
2.5
0
2.25c
20.0
WAB 56
–
104
0.75
0.75
0.75ab
0.0
WAB 56
–
50
1
.00
1.75
1.38
abc
42.9
WAB
880
1
.00
1.25
1.13
abc
20.0
WAB
928
1.5
0
2.25
1.88
bc
33.3
Means
1.02a
1.45b
1.23
LSD
1.12
S.E
0.8
0
CV%
64.8
Means followed by the same letter
are not significantly different us
ing Tukey multiple
comparison test (P<0.05)
.
88
Since tillers play a major role in determining plant architecture and yield, these results could
suggest reduction in grain yield of
Striga
infected rice varieties especially those with high
Striga
numbers. Such varieties include IAC 165 and Superica (local check) which had highly
reduced tiller numbers.
Reducti
on in the number of tillers of infected rice plants compared
with the control is
attributed to both suppression of growth of tiller buds and inhibition of the
formation of tiller buds by
Striga
. Out of the 14 varieties tested, only two varieties NERICA
–
1
and IRGC had a high number of tillers compared to the uninfected plants. The remai
ning
varieties had lower 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 infected rice
plants is attributed to
growth regulators just like in plant height
s
. Hormones such as auxins and cytokinins have been
known to control tillering (Garba
et al.,
2007; S
a
kamo
to
et al.,
2006). Auxins have an indirect
inhibitory action on tillering while cyt
okinins directly promote tillering (Ongaro and leyser
2008). Therefore any changes in synthesis of these hormones will affect tillering of rice.
Further evidence
has shown that
Striga
infected sorghum tissues have
greater amounts
of
ABA
and ethylene and lo
wer amounts of Cytokinins (Drennan a
nd ELhiweris, 1979) hormone which
is important in tiller formation.
Another possible explanation
for lower tiller production of
i
nfected plants could be due to
n
utrition
al
imbalances especially with respect to nitrogen.
Cruz and Boval, (2000); Mckenzie,
(1
998) found a positive effect between
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 pos
sible that tillering reduced as
the nitrogen
is diverted to the parasite
89
4.3.1.4
Effect on n
umber of leaves
Generally, there was a significant (P<
0.05) difference in the number of leaves produced by
infected rice plants compared to uninfected control plants (Table 18).
Number of le
aves were
significantly
reduced for local variety (Superica) after
Striga
hermonthica
infection
(Table 18
)
.
However,
th
ere was no significance (P>
0.05)
difference
in the number of leaves produced per
plant for the infected and uninfected controls of all the rem
aining rice varieties
(Table
18
). For
the infected plants varieties WAB 181
–
18 and CG14 produced the highes
t num
ber of l
eaves
per plant (
12
leaves
) while Superica and IAC
165 produced
the lowest number of le
aves per
plant (6 leaves).
Notably, varieties WAB181
–
18 and CG14 p
roduced a significantly
higher
number of leaves compared to the rest of the varieties except IR
GC, SCRID090, WAB928,
WAB56
–
50 and NERICA
–
1 (Table 18
). Variety WAB56
–
104 had the
least number of leaves
(Table 18
).
There was a reduction in number of leaves on the infected rice plants compared with the
uninfected control p
lants across varieties (Table
18
). These results agree with those of Cissoko
et al. (2011) who also reported 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 less number
of leaves compared to their respective uninfected plants.
The lower
number of leaves exhibited by
infected rice plants
is attributed to the genetic resistance levels
of the different rice varieties. This led to some varieties to be more significantly affec
ted than
others.
However, varieties NERICA
–
1 and WAB56
–
104 produced a high number of leaves on
their infected plants compared to the uninfected plants
.
The high number of leaves on the
infected plants of NERICA
–
1 and WAB56
–
104 could be related to late
Str
iga
attachment on
host roots of these varieties.
It should however be noted that even though these varieties
produced a higher number of leaves, they were not significantly different from control plants.
90
Table
18
:
Effect of
Striga
hermonthica
infection on
number of leaves
of upland rice
varieties
grown
under controlled conditions
(growth chamber)
at 21 days after infection
.
Number of leaves
Treatments
Rice variety
Striga
Infected
Uninfected
(control)
Variety means
%reduction
in leaves
BLECHAI
7
.00
7.5
0
7.25
ab
6.7
CG14
11.5
12.25
11.88
c
6.1
IAC 165
6
.00
7.5
0
6.75
ab
20
.0
IRGC
9.75
10.75
10.25
bc
9.3
NERICA
–
1
10
.00
8.5
0
9.25
abc
0.0
NERICA
–
17
7.5
0
8
.00
7.75
ab
6.3
NERICA
–
4
7
.00
8.25
7.62
ab
15.2
SCRID090
10.25
10.5
10.38
bc
2.4
SUPERICA
6
.00
9.75
7.88
ab
38.5
WAB 181
–
18
11.5
0
13.75
12.62
c
16.4
WAB 56
–
104
6.5
0
6.25
6.38
a
0.0
WAB 56
–
50
8.25
9.75
9
.00abc
15.4
WAB
880
7.25
8.25
7.75
ab
12.1
WAB
928
8.25
10
.00
9.12
abc
17.5
M
eans
8.32
a
9.42
b
8.87
LSD
3.101
S.E
2.207
CV%
24.9
Means followed by the same letter
are not significantly different using T
ukey multiple
comparison test (P<0.05)
.
91
Leaves are the main photosynthetic organs of the rice plant.
Therefore their reduction can affect
assimilate manufacture and consequently grain yield of the plant. This only implies that those
varieties that are severely affected such as Superica and IAC 165 can ultimately have a low
grain 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 reported (Frost
et al.,
1997; Cechin and Press, 1994; Atera
et al.,
2012). The
reduction in leaf number of the
infected plants
could be du
e to nutrient diversion from the host
plant to the parasite. There is less dry matter partitioned to the growth of new leaves. Other
studies have also shown that rates of photosynthesis are usually lower in
Striga
infected plants
than their uninfected cou
nterparts (
Gurney
et al.,
1995;
Frost
et al.,
1997
; Gurney
et al.,
1999
).
It is possible that the reduced photo
–
assimilate production due to the change in plant
architecture after infection is responsible for the reduced number of leaves in infected plants
compared to the control plants.
4.3.3
Post
–
germination
attachment lev
els of
Striga hermonthica
to selected rice varieties
.
4.3.3.1 Quantification of post
–
germination
attachment
of
Striga
.
There was a significant difference in the number of
Striga
parasites attached per rice plant
across rice varieties (Appendix
16).
Variet
y
IAC 165
(42)
had significantly
(P< 0.001)
the
highest number of
Striga
plants per host root system though it was not significantly different
from Superica
(39)
and NERICA
–
1
(12)
(Table 19
). Whereas
, CG14
(1)
had the lowest
Striga
plant
s per host root
system (Table 19
). Parasitic attachments per host plant ranged
from 39
–
42
Striga
plants
on the very susceptible varieties
with
Superica and IAC 165, supporting the
highest number of
Striga
plants, to less than two
on the most resistant varieties
like
IRG
C and
CG14 (Table 19
).
92
Table
19
:
Number of
Striga
plants and
Striga
dry weight taken at 21 days after
Striga
hermonthica
infection under controlled
growth chamber
conditions
Data was first transfo
rmed before analysis log (x+1) to meet the assumption for analysis of
variance.
Means
were
presented as original data
and results in the parenthesis was (log x+1)
transformed
.
Means followed by the same letter
are not significantly
different using Tukey
multiple comparison test (P<
0.05)
.
Rice variety
Striga
number
Striga
dry weight
(mg)
BLECHAI
3
(0.42
abc
)
0.08
(0.32
a
)
CG14
1
(0.06
a
)
0.02
(0.01
a
)
IAC 165
42
(1.59
e
)
24.18
(1.39
c
)
IRGC
1
(0.16
ab
)
0.06
(0.02
a
)
NERICA
–
1
12
(1.02
cde
)
5.38
(0.59
ab
)
NERICA
–
17
10
(0.82
bcd
)
4.52
(0.61
ab
)
NERICA
–
4
5
(0.70
abcd
)
0.54
(0.16
a
)
SCRID090
2
(0.38
abc
)
0.42
(0.14
a
)
SUPERICA
39
(1.49
de
)
21.45
(1.18
bc
)
WAB 181
–
18
3
(0.48
abc
)
0.38
(0.12
a
)
W
AB
56
–
104
2
(0.39
abc
)
0.72
(0.19
a
)
WAB
56
–
50
5
(0.66
abc
)
3.1
(0.47
a
)
WAB
880
5
(0.71
abcd
)
1.62
(0.35
a
)
WAB
928
2
(0.26
abc
)
0.09
(0.04
a
)
LSD
0.44
0.35
S.E
0.34
0.27
CV%
52.5
74.2
93
Varieties
such
as
NERICA
–
1
and NERICA
–
17 supported relatively large number of
Strig
a
plants (10
–
12
) p
arasites per host root (Table 19
). The remaining cultivars exhibited a go
od level
of post
germination
–
at
tachment resistance supporting 2
–
5
parasitic plants per host
root system
(Table 19
).
Generally
the susceptible
varieties Superica
(local check)
and
IAC 165, had the
largest
number of
well
–
developed parasites on
their
root sy
stem
s
compared to the rest of the
varieties (Figure 6 a, b, c, d, e and f).
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 WAB9
28
and WAB
56
–
50
a)
superica
b) IAC165
c) NERICA
–
1
d) NERICA
–
17
e) WAB928
f) WAB56
–
50
94
Post
–
germination
attachment resistance levels varied significantly across varieti
es t
e
sted
(Appendix 16).Varieties IAC165 and Superica were the most
affected both during the course
of
the experiment and at h
arvest. Similarly studies by Gurney et al. (2006); Swabrick et al. (2008)
and Cissoko et al. (2011) have all reported IAC 165 as susceptible and was used as a susceptible
check variety in these studies. Variety CG14 was the most resistant variety as per th
is study
having an average of 1
–
2 developed parasites. Field and pot experiment studies by Johnson et
al. (1997) stressed 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 resistant
to
Striga
. Similarly Kaewchumnog and Price (2008) reported
CG14 as one of the most resistant variety
to
Striga
in their study. On the other hand post
germination
–
attachment resistance of
some varieties from this study was similar to
observations by Cissoko
et al
,
(2011) who also found
varieties NERICA
–
4, WAB56
–
5
0,
WAB181
–
18 and CG14 with
good level
s
of post
–
germination
attachment resistance to
Striga
hermonthica
(Sh
–
Kibos) in Kenya and
St
riga asiatica
(Sa
–
USA) in
USA.
This implies that
these varieties are resistant to a wide number of ecotypes of
Striga
including the one used in
this study. Varieties such as WAB880, WAB928, SCRID090, Blechai, and IRGC were
evaluated for post
germination
–
at
tachment resistance for the first time
in this study
. These
varieties exhibited an excellent post
–
germination
attachment resistance to
S.hermonthica
.
Field experiments by Haraha
p et al. (1993) also showed Blechai and IRGC as
highly resistant
to
Striga
hermonthica
in field
trails in Kenya.
Johnson et al. (2000)
also
reported WAB928 as
a resistant variety to
S.hermonthica
parasitism. However no documented studies have indicated
the reaction of varieties SCRID090 and WAB880 to
Striga
. This study
therefore
,
the first time
to
report the resistanc
e of these varieties in both fi
e
l
d (study one) and under controlled
growth
chamber
conditions.
Varieties such as NERICA
–
1 and
–
17
were
reported to be resistant to
Striga hermonthica
(Cissoko
et al.,
2011; Jamil
et al
.,
2011; Rodenburg
et al.,
2015) supported
95
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 t
o 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
dry weight
was signi
ficantly higher for variety
IAC 165
(24.18mg)
compared to other
varieties though it wasnot significantly
(P>0.001)
different from variety
Superica (local check)
(21
.45mg)
(Tabl
e 19
).
Varieties NERICA
–
17,
–
1 and WAB56
–
50 had moderate
Striga
biomass
ranging from 3
–
5m
g per plant
(Table 19
). There was no significant differ
ence (P
>
0.001) in
Striga
biomass
of
the remaining
rice
varieties (Table 19
). These varieties had a relatively small
Striga
biomass ranging from 0.02mg
–
1.62
mg per plant (Table 19
).
The results show differences in resistance among the rice varieties. Varieties with high
Striga
dry weight
such as IAC 165 and Superica also had the highest
Striga
numbers. Similarly,
Cissoko et al. (2011)
reported
a high
Striga
dry weight
on variety IAC16
5 attributing it to
increased
Striga
infestation. O
her varieties i.e. NERICA
–
1, NERICA
–
17 recorded a high
Striga
dry weight
though not compared to the above two varieties. This is a clear indication of an
intermediate number of
Striga
parasites per root sy
stem. The highly resistant varieties such as
WAB928, CG14, Blechai, IRGC and CG14 with a lower
Striga
dry weight
was possibly due
to limited
Striga
parasite development. Varieties NERICA
–
1, WAB56
–
50, WAB880 and
NERICA
–
4 with a
Striga
dry weight
not correla
ting 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.
96
4.3.4
Phenotype of resistance
of selected rice vari
eties against
Striga hermonthica
(Namutumba ecotype)
Figure 7
shows striking difference in ability to support
Striga hermonthica
o
n the very
susceptible variety Superica (local check)
and its resistant counterpart WAB928. In the
susceptible variety Superica
,
by day 21 the parasites were fully developed implying an earlier
establishment of the xylem
–
to
–
xylem connection
with the parasite.
The bigge
r size of the
parasites on the host root
system of Superica (local check)
simply indicate
s
an extende
d effect
of the parasite on
it. O
n
ce a xylem
–
to
–
xylem connection is
formed, the parasite gr
o
w
s
rapidly
(Gurney
et al.,
2006; Cissoko
et al.,
2011).
According to Gurney et al. (2006) b
y 21 days a
fter
infection
, the parasite haustorium
of the susceptible varieties
have
a well differentiated vascular
core and hyaline body
(
regulates
supply nutrients to developing parasite
or metabolise the
solutes
)
.
On the most resistant variety WAB928,
some parasi
tes failed to attach on the host roots
whereas those that
attached
on the root system (Figure 4c),
failed to penetrate the root
evidenced by the hypersentivity reaction around the point of attachment
(host necrosis at the
site of attachment)
. This left som
e
parasite
s either dead or showing signs of necrosis.
These
findings are similar to what was observed by Mohammed et al. (2003) on a sorghum variety
Framida infested with
Striga asiatica
, Nipponbare infected with
Striga hermonthica
(
Gurney
et
al.,
2006) and (Cissoko
et al.,
2011) on NERICA varieties infected with
S.
hermonthica
and
S.asiatica
.
The evidence of necrosis at the point of parasite
–
attachment to host root could be a resistance
mechanism from the host. It is not clear whether this reactio
n is as a result of accumulation of
reactive oxygen species or the expression of NADPH oxidases and peroxidases as observed in
sunflower resistant to
Orabanche cumana
(Lestousey
et al.,
2007).
97
Figure
7
:
Characteristics and phenotype of resistance in selected rice varieties
against
Striga hermonthica
(Namutumba ecotype) in Eastern Uganda. PR
(phenotype of resistance), (a) susceptible and (b) resistant variety compared
.
a) Superica
b)
WAB928
C) WAB928
PR
PR
98
Secondary the necrosis at the point of attachment could also be due to up
–
regulation of
hypersensitive
–
induced response protein as observed in Nipponbare (Swarbrick
et al.,
2008).
This therefore requires further research to understand the molecular basis
of this
resistance to
Striga hermonthica
. Notably a few successful attachments were seen on
WAB928 which were very small (Figure 6b). This is a clear indication of late
establishment of the xylem
–
to
–
xylem connections with the host roots. It is not clear
whether this is a resistance mechanism in this variety. Cissoko et al. (2011) also
reported fewer successf
ully xylem
–
to
–
xylem connection in NERICA
–
10 and NERICA
–
1 when infected with
Striga
.
4.3.5
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 in above
–
ground rice bio
mass of infected rice plants
for varieties WAB181
–
18, WAB 880,
Superica (local variety)
and IAC165 compared
t
o their control pla
nts (Table 20
, Appendix 16). Rice biomass of the remaining varieties
was not significantly (P > 0.05
) affected by
Striga
infection (Table 20
). With respect to
control plants
, WAB181
–
18 had the highest
rice
biomass (1.144g) and NERICA
–
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 uninfected control plants (Table 20).
Irrespective of the treatments, varieties WAB181
–
18, SCRID090 had the highest rice
biomass whereas NERICA
–
17 had the lowest biomass. Notably control plants produced
significantly the highes
t rice biomass compared to
Striga
infected plants (Table 20)
99
Table
20
:
Effect of
Striga hermonthica
on a
bove
–
ground rice biomass of infected rice
plants under controlled
(growth chamber)
conditions
at 21 days after infection
.
Rice
biomass (
m
g)
Treatment
Rice variety
Striga
infected
Uninfected
(control)
Variety means
% reduction
in
biomass
WAB
181
–
18
0.696
1.14
4
0.920c
39.16
WAB880
0.459
1.00
1
0.730ab
c
54.15
SUPERICA
0.254
0.98
3
0.619abc
74.16
SCRID090
0.845
0.969
0.907c
12.80
IRGC
0.721
0.92
2
0.821bc
21.80
CG14
0.533
0.798
0.665abc
33.2
1
IAC 165
0.342
0.746
0.544abc
54.16
WAB
56
–
50
0.553
0.665
0.601abc
16.84
BLECHAI
0.547
0.585
0.566abc
6.50
WAB928
0.427
0.55
3
0.490abc
22.78
NERICA
–
1
0.480
0.493
0.487abc
2.64
NERICA
–
4
0.461
0.46
3
0.46
2ab
0.43
WAB
56
–
104
0.427
0.45
0
0.439
ab
5.11
NERICA
–
17
0.326
0.347
0.336
a
6.05
Means
0.504
a
0.730
b
0.61
3
LSD
0.361
S.E
0.256
CV%
41.9
Means followed by
the same lette
r
are not significantly different u
sing T
ukey
multiple
comparison test (
P< 0.05)
.
100
Generally there was a loss in biomass of all infected plants especially for the most susceptible
varieties such as Superica (74.16%) and IAC 165 (54.16%) (Table 20). Others varieties with
high loss in biomass included WAB800 (54.15%), WAB181
–
18 (39.16%), I
RGC (21.80%)
WAB928 (22.78%) and (CG14) (33.21%) (Table 20). The remaining varieties except
SCRID090 and WAB56
–
50 with 10
–
20% loss in biomass had less than 10% loss in rice
biomass.
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
=0.376, P=0.012
).
The most susceptible varieties Superica
(local check)
and IAC 165 were the most affected
of
all the varieties (Figure 8
). This negative relationship would indicate t
hat as you increase the
Striga
population
,
there is a significant
negative
effect on th
e biomass of the host (Figure 8
).
Many varieties had similar
levels of infection but lost biomass to differing extents.
For example
WAB56
–
104 and WAB181
–
18 with almost
similar
infection
levels
had
Striga
biomass
of 0.72g
and 0.38g respectively
(Figure 8).
However the percentage loss in
host
was 5% compared to
39%.
T
his illustrates a difference in tolerance between the two rice varieties to
Striga
hermonthica
infection.
I
nfection of rice plants with
Striga hermonthica
altered the partit
ioning
of assimilates
to the leaves and stems
of rice plants. This was revealed in the above ground rice
biomass o
f infected rice plants (Table 20
, Figure 8). Similar results were reported f
or sorghum
parasitized by
S.hermonthica
(Cechin and Press, 1993; Frost
et al.,
1997), maize (Taylor
et al.,
1996) and rice (Cissoko
et al.,
2011).
Rice biomass of infected rice plants for varieties WAB56
–
104, WAB56
–
50, SCRID090,
BLECHAI, NERICA
–
1,
–
4,
–
17 an
d WAB928 was not highly reduced compared to other rice
varieties. This could indicated less effect 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.,
2000; Cissoko
et al.,
2011)
.
101
Figure
8
:
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.
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, WAB880 and WAB181
–
18. These except WAB880 and WAB181
–
18 supported the
highest number of
Striga
pl
ants. It is also possible that alteration in plant morphology of
Striga
infected plants depended upon the variety genetic differences in resistance and tolerance.
Surprisingly varieties such as CG14, WAB928, IRGC, WAB181
–
18 and WAB880 inspite of
having low
er number of parasites per root system registered tremendous reduction in host
biomass.
On contrally varieties NERICA
–
1, NERICA
–
17
and WAB56
–
50
with high number of
parasites
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. (2011); Rodenburg et al. (2015).
y =
–
1.8689x + 83.367
R² = 0.376
0
10
20
30
40
50
60
70
80
90
100
0
5
10
15
20
25
30
loss in biomass as a percen tage of the
control
Striga
biomass (mg)
Superica
IAC 165
WAB880
WAB181
–
18
CG14
WAB928
WAB56
–
50
N1
N17
SCRID090
N4
WAB104
Blecha
i
IRGC
102
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 would suggest that the highly r
esistant varie
ties can have
high
yield potentials in
Striga
prone areas.
103
CHAPTER
FIVE
5
.0 GENERAL DISCUSSION
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 are
as
is a viable option for
Striga
management
.
It is also
cost
–
effective as farmers do not
hav
e to
invest in this technology. This
thesis
therefore sought to evaluate the resistance of rice varieties
against
Striga hermonthica
in
Eastern Uganda
.
Two studies
were conducted on
–
farm and
one
study was conducted
under
growth chamber
controlled condition
s. The first
study e
valuated
the performance of selected rice varieties under
Striga
infested field conditions
. The second
study was
a farmer participatory
variety selection of upland rice varieties whereas the third
study
evaluated the post
–
germination
at
tachment resistance
to
S.hermonthica
of selected
upland rice varieties to
Striga hermonthica
.
The results of the first study indicated variation in
Striga
resistance and
grain
yield across
varieties. Out of the 25
upland rice
varieties evaluated, three
O.glabberima
varieties
namely
AGEE, CG14 and Anakila. S
even
O.sativa
varieties
namely
IR49, IRGC, WAB935, WA
B928,
SCRID090, WAB880 and Blechai. O
ne
O.sativa
parent WAB181
–
18 and
four
NERI
CA
varieties NERICA
–
2,
–
10,
–
17 and
–
4
showed
excellent resistance to
Striga
compared to the
susceptible variety
IAC 165 and the
locally grown variety Superica. All the above varieties
except WAB880 and SCRID090 have been reported as having an excellent resistance to
S.hermonthica
elsewhere.
Variety Blechai and IRGC in
Kenya
(Harahap
et al.,
1993), IR49
in
104
Kenya and Ivory Coast
(Harahap
et al.,
1993; Johnson
et al.,
1997)
,
WAB928 and WAB935
in Ivory Coast
(Johnson
et al.,
2000)
.
Jamil et al. (2012) reported
O.glabberima
varieties Agee
,
Anakila and CG14 have having an excellent resistance to
Striga
. The NERICA varieties
–
2,
–
10,
–
17,
–
4 and the
O.sativa
parent WAB181
–
18 were
also reported by Rodenburg et al. (2015)
as having an excellent
field
resistance to
S.hermonthica
in Mbita, Kenya
. This indicate
s
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 a
ttributed to continuous cultivation of this
variety. Therefore virulence levels of local
Striga
population against this variety
seem to have
increased
with
time.
T
his variety was
similarly
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. Var
ieties NERICA
–
1 and UPR which were
observed
to be resistant
to
Striga
in Mbita
, Kenya
(Rodenburg
et al.,
2015;
Harahap
et al
.,
1993) supported
a
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 com
pared to
the
ecotype in
Mbit
a
Kenya
.
Results
indicated suscep
tibility of varieties WAB56
–
50 and
WAB56
–
104
under
field conditions but 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
, kenay
.
This
susceptibility to
S.hermonthica
could be due to lack of pre
–
attachment
resistance mechanisms in these varieties.
Varieties that combine
Striga
resistance and high
grain
yields are important in improving rice
production. However,
Striga
resistant varieties do not
necessarily produce high
grain
yields
.
This
study has indicated variation in yield production across varieties under
Striga
infested
field conditions. The best p
erforming variet
ies
i.e. SCRID090,
NERICA
–
2,
WAB181
–
18,
105
NERICA
–
10 and NERICA
–
4 in 2014A
yielded between
3.1t ha
–
1
to 3.7t ha
–
1
. Additionally
,
in
2015, the best performi
ng varieties
i.e. NERICA
–
2, NERICA
–
4, NERICA
–
10, Blechai,
SCRID090, W
AB181
–
18, CG14, Age
e
and NERICA
–
17
yielded between 3.0t ha
–
1
to 3.8t ha
–
1
.
These results are similar to
those reported
Rodenburg et al. (2015) who also indicated varieties
W
AB 181
–
18
and all the NERICA varieties as high yielding
varieties in Mbita, Kenya.
However,
some studies have reported the h
igh yielding potential of NERICA
varieties
(Gridley, 2002; Saito
et al.,
2012).
Therefore, this
is the same reason why a
ll the above
varieties except Agee
and CG14 were selected
for adoption
by farmers in study two.
The
O.
sativa
varieties WAB880 and SCRID090 evaluated for
resistance to
Striga
for
the first
time
in this study
did not only have good resistance to
Striga
but
were
also high
yielding
.
The results
of this study have
also indicated
O.glabberima
varieties
(Makassa
, ACC and
MG12
),
NERICA
–
1
partially resistant to
Striga
and
O.sativa
parents WAB56
–
104,
WAB56
–
50
susceptible
to
Striga
had high yield compared to varieties WAB928, WAB93
5, IRGC and IR49
which showed high
resistance to
Striga
. This
observation
suggests
differential varietal
tolerance to the effects of
S. hermonthica
. Tolerance is
described as “
the ability of a give
n
variety to produce
grain
yield even under high
Striga
infestation levels
”
(Rodenburg
et al.,
2006; Rodenburg and Bastiaan,
2011).
Tolerance
to
Striga
has also been reported in sorghum
(Gurney
et al.,
1995; Rodenburg
et al.,
2006).
Unde
rstanding the expression of tolerance
trait
can be important to improve
grain
yield in the most resistant varieties with low
grain
yields
(Rodenburg and Bastiaan
s, 2011). The results
of this study also showed
a significant
negative
correlation
(Spearman correlation coefficients =r
2015
=
–
0.465, P= 0.02) between
maximum
number of above ground
Striga
plants (
NSmax
)
and gra
in yield in 2015. This suggested
a high
Striga
pressure in 2015
compared to 2014. This
significant negative correlation
also implied
that the highly resistant varieties produced high
grain
yields. The local variety Superica
produced the lowest
grain
yield in
2015
implying that
the
increase in
St
riga
population
affected
106
the performance of this variety.
This
this
may be
the
same reason why farmers selected it as
the worst performing variety.
In the second
study
sought to understand the
selection criteria farmers’ use in identification of
the best p
referred
rice
varieties. The study indicated
that
farmers use
Striga
resistance,
high
grain
yield,
good
tillering ability, drought tolerance/resistance, maturity period and
plant height
when selecting which rice varieties to grow
. Consequently, farmers ran
ked
SCRID090 as their
best variety. T
his was followed by WAB880, NERICA
–
17, Blechai and WAB181
–
18 as the
five
next
best varieties
in that order
out
of 25. These varieties were
ranked among those with
high yields and excellent resistance to
Striga hermonthi
ca
under both field and controlled
conditions
. Other varieties selected by farmers were NERICA
–
4, NERICA
–
1, NERICA
–
10 and
lastly NERICA
–
2 in ascending order of preference. All the N
ERICA varieties included in
study
one were all selected by farmers
probably due to high
grain
yields
.
Similar results were shown
by Gridley et al. (200
2) where farmers highly preferred
these variet
ies in Ivory Coast. Also
other
studies have indicated high yields and good
Striga
resistance among these varieties
(Gridley
e
t al.,
2002; Saito
et al.,
2012; Rodenburg
et al.,
2015)
.
A number of studies have
indicated
high grain
yield as the most sought
after
trait
in rice variety
selection (Nan
fumba
et al.,
2013; Gridley
et al.,
2002; Kimani
et al.,
2011).
However, this
study
has
indicated
Striga
resistance as the most
important
considered trait in variety selection.
Having
Striga
resistance as the most sought
after
trait in decision making to
adopt a variety
emphasizes the importance of
Striga
in
this area. The r
esults
on fiel
d resistance (study one)
also
showed
the local variety (Superica) as a highly susceptible variety to
Striga
. This i
s one
of the
reason
s
why it was not selected
by farmers.
Results also indicated other
non
–
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
grain
yield
(Table 8
and 10
)
.
Varieties Makassa and UPR had
partial resistance leve
ls to
Striga
and the
107
O.sativa
variety NARC3 (ITA 257)
had good aroma. This indicates differences in traits sought
by farmers and researchers.
The
O.glabberima
v
arieties Makassa and Anakila were susceptible
to lodging.
This
negat
ive trait led farmers
not
to
select
th
es
e
rice
varieties. The
O.sativa
variety
NARC3 (ITA 257
was not selected
by farmers for being short, susceptible to
Striga
and being
low
grain yielding.
This
same variety produced low
grain
yields in study one
(Table 8
)
.
On
other hand
, UPR was not selected
for it
s
short suture and low
grain
yield
. 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; Kimani
et al.,
2011).
This study
has
also in
dicated significant difference
s
in variety selection among men and
women. Men selected NERICA
–
17 and women selec
ted SCRID090 as the best
rice
varieties
for
adoption in
each group. The difference in variety selection creteria of men and women
is
relate
d to
the roles each group play
in the
rice crop production
cycle. Women normally carry
out the initial
activities such as
sowing, weeding
and harvesting. M
en
are
normally involve
d
in
the marketing
of the crop.
Other s
tudies have
also
shown differences in traits men and women
sought in the newly introduced varieties (Dorward
et al.,
2007). Variety SCRID090 selected
by women is tall,
Striga
resistant and early maturing
(
study one and three)
. However
,
al
though
NERICA
–
17
selected by men
wa
s
high yielding and
Striga
resistant but
was
not as early
maturing as SCRID090.
Addison et al. (2014) acknowledged early maturity as one of the most
important trait women sought in selecting
rice
varieties in Ghana.
A selection of rice varieties selected f
or high field resistance (study one) and for adoption by
farmers (study two) were evaluated for their post
–
germination
attachment resistance un
der
controlled
(growth chamber)
conditions. T
he study
also
investigated whether the resistance
observed in the fi
eld can be replicated under controlled conditions. Results indicated significant
effects of
Striga
on rice varieties when compared to control plants. This was manifested in the
reduced rice biomass. Generally
,
there was a l
oss in biomass across varieties.
However
some
108
varieties i.e. Superica, IAC165, WAB880, and WAB181
–
18, CG14, WAB928 and IRGC were
more
affected when compared to varieties NERICA
–
1, WAB56
–
50, WAB56
–
104, SCRID090,
Blechai,
NERICA
–
4
and NERICA
–
17 with respect to their control plants
.
It has b
een proposed
that
the
loss of biomass in
Striga
infected plants to some extent is due to acquisition of
carbohydrates,
nitrogen and other solutes from the host
to the S
triga
parasite. Diversion of
resources from the plant
results in change in host architecture leading to stunting of plants
especially for the most susceptible
varieties, reduction in
tillering (Cissoko
et al.,
2011; Jamil
et al.,
2011
; Atera
et al.,
2012
), leaf area, leaf number, thinning of stems and shorten
ing of
stem internodes (Swarbrick
et al.,
2008; Gurney
et al.,
1999) 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. St
udies have shown
lower
p
hotosynthetic rates
in
Striga
in
fected 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 t
hat eventually contributes to increases of
p
lant 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 an
d
ultimately the
grain
yield of rice
varieties.
This study has indicated excellent post
–
germination
attachment resistance for varieties CG14,
WAB928, IRGC, SCRID090, WAB181
–
18, WAB880, Blechai, WAB56
–
104
, WAB56
–
50 and
NERICA
–
4. All these
varieties except W
AB56
–
104
and WAB56
–
50 also showed
excellent
resistance to
Striga
under field conditions in study one. Cissoko et al. (2011)
also
reported
varieties WAB181
–
18, NERICA
–
4 a
nd CG14 as having excellent p
ost
–
germination
attachment
resistance to
S.hermonthica
eco
type (Sh.Kibos)
in Kenya
. The hypothesis set that the resistance
under field and controlled conditions does not differ holds only true for the above
mentioned
varieties
only
. However
,
for varieties WAB56
–
104 and WAB56
–
50 considered susceptible
109
under field
conditions
showed high resistance to
Striga
under controlled conditions
.
This
behavior
however is not surprising for variety WAB56
–
50 because it was reported as highly
resistance under controlled conditions
by
Cissoko
et al.
(
2011
and
Rodenburg
et al.
(
2015).
However
it was found susceptible
when tested under field conditions in Kenya by Rodenburg
e
t al. (2015). This can only be
an indicator
of
lack of pre
–
germination
attachment resistance
mechanism.
Jamil et al. (2011) reported it as the highest strigol
actone producer and
strigolactones trigger the germination of
Striga
.
Results indicated varieties
NERICA
–
17 and
NERICA
–
1 produced
moderate levels of
Striga
plants per root system yet 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 controlled conditio
ns.
Results contradict those of
Cissoko et al. (2011) who reported these two varieties as hav
ing an
excellent post
–
germination
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 varieties perform better
than
others under similar parasitic load (Cissoko
et al.,
2011).
The
most resistant varieties can
also have well developed
Striga
plants on their root system
s
.
The difference is observed on
how they respond
to the attached
Striga
plants.
V
arieties CG14, WAB880 with lower
Striga
attached on the host roots had a hig
h reduction in rice biomass 33.2
% and 54.2% respectively.
However
,
varieties NERICA
–
1,
NERICA
–
17
and WAB56
–
50
with a high parasitic number had
less reduction in rice biomass 2.6
%,
6.1%
and 16.8%
respectively
,
compared to uninfected
control plants.
This confirms
reduced effect of
Striga
on these varieties irrespective of the high
parasite numbers.
This concept can therefore be exploited such that tolerance genes can be
incorporated i
n the resistant varieties.
110
CHAPTER SIX
6.0 CONCLUSION AND RECOMMENDATIONS
6.1 CONCLUSION
Out of the 25 rice varieties
evaluate
d
against
Striga hermonthica
ecotype of Namutumba
(Uganda)
under field conditions
;
seven
Oryza sativa
r
ice varieties
(
WAB928,
WAB9
35,
IRGC, IR49, WAB880, Blechai
and SCRID090
)
,
three
Oryza glaberrima
varieties
(Anakila,
CG14, Agee), four
NERICA varieties
(
NERICA
–
2, NERICA
–
10,
NERICA
–
17 and NERICA
–
4
)
and one NERICA parent WAB181
–
18
had
excellent resistance levels
to
Striga hermon
thica
.
The most susceptible varieties as per this study were Superica, IAC165, WAB56
–
50 and
WAB56
–
104.
Varieties
WAB56
–
50, WAB56
–
104 and NERICA
–
1
with
high grain yield inspite of the high
Striga
numbers
recor
ded
were tolerant to
Striga hermonthica
.
Striga hermonthica
had a great effect on the yield potential of rice varieties
.
The yield of
varieties NARC3, IAC 165 and Superica (local check) was significantly reduced
by
Striga
.
The
high
est yielding varieties were
SCRID090, NERICA
–
4, NERICA
–
2, WAB181
–
1
8, WAB880,
NERICA
–
10
,
Agee, NERICA
–
1, NERICA
–
17
and Blechai. V
arieties
IR49,
IRGC, WAB928
and WAB935
had the lowest yield
despite the high
resistant
levels to
Striga hermonthica
.
Basing on morphological appearance of rice vari
e
ties in the field, farmers
selected varieties
SCRID090, WAB880, WAB181
–
18, Blechai, NERICA
–
17, NERICA
–
4, NERICA
–
2,
NERICA
–
1 and NERICA
–
10 as the most preferred over their local variety
(Superica)
. Notably,
men selected variety NERICA
–
17 as their first choice whereas women selected v
ariety
SCRID090 as their first choice variety.
111
The least preferred varieties of rice according to farmers were the local variety (Superica)
IAC165, Anakila, Makassa, NARC3 (ITA 257) and UPR.
According to farmers, t
hese were
either susceptible to
Striga,
p
rone to lodging, not tolerant to drought, low yielding or with short
stature.
F
armers
identified
Striga
resistance
,
high grain yield, tillering
ability, height, early
maturity,
drought tolerance, seed size and colour as the most important attributes they
looked for
in newly
introduced
rice
var
ieties.
Basing on
this
selection criteria, farmers
ranked
varieties SCRID090,
WAB880,
NERICA
–
17, Blechai and WAB181
–
18
as the most preferred
.
P
ost
–
germination
attachment resistance of
selected
rice
varieties
evaluated
under controlled
conditions indicated v
arieties CG14, IRGC, WAB928,
SCRID090,
WAB56
–
104, BLECHAI,
WAB181
–
18, NERICA
–
4 and
WAB880
as resistant
to
Striga hermonthica
. The most
suscepti
ble varieties as per this
study were Superica (local check) and
IAC
165.
There was a great effect of
Striga hermonthica
on plant height, stem width, number of tillers
and leave
s
(rice biomass)
across rice varieties with
varieties IAC
165, WAB880 and
Superica
,
having
the highest reductions
following infection with
Striga hermonthica
.
R
esis
tant v
arietie
s
WAB181
–
18
, IRGC and
CG14
also
had a high reduction in their rice biomass following
infection with
Striga
compared
NERICA
–
1 and NERICA
–
17
with
moderate number of
parasi
tes attached on the host roots.
112
6.2 RECOMMEND
ATIONS
From the results of the whole
study, the following
recommendations
can be drawn
:
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 improv
e rice production in
Eastern Uganda
.
II.
Rice varieties
with excellent resistance and high yields both under field and controlled
conditions
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.
III.
Use of invitro methods based on multiple mechanisms either pre
–
attachment /post
–
attachment resistance
is important in selection of resistance for
Striga
amongst a large
accessions of rice.
IV.
Understanding the tolerance genes in rice varieties such as WAB56
–
50, NERICA
–
1
and
WAB56
–
104
can be important to incorporate the
se genes in the most resistant
varieti
es
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 yield potential despite the
excellent resistance levels to
Striga
can also be i
mproved or crossed with other
susceptible varieties to improve on these varieties.
113
REFERENCES
AATF, 2006.
Empowering Africa Farmers to Eradicate Striga from maize croplands, P: 17.
African Agricultural Technology Foundation Nairobi, Kenya
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 b
y WRENmedia, funded by the Swiss Agency for Development
and Cooperation (SDC) and implemented by the European Initiative on Agricultural
Research for Development (EIARD)
.
Ahonsi, M.O., Berner, D.K., Emechebe, A.M., Lagoke, S.T., 2004.
Effects of ALS
inhibi
tor
Analytical Report February 2007
and Tanzania. Asp. 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
(1), 35
–
48.
114
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 infections in western Kenya.
International Journal of Agriculture and
Biology
,
14
(2), 271
–
275.
Bebawi, F.F 1981
. Intraspecific physiological variants of
Striga hermonthica
. Exp. Agric.
17, 419
–
423
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
signalling in host parasitic plant interactions. Current Opinion Plant biology 6:358
–
364
Cardoso, C., Ruyter
–
Spira, C., & Bouwmeester, H. J. 2011.
Strigolactones and root
infestation by plant
–
parasitic Striga, Orobanche and Phelipanc
he spp.
Plant
Science
,
180
(3), 414
–
420.
Carsky, R. J., Singh, L., & Ndikawa, R. 1994b.
Suppression of Striga hermonthica on
sorghum using a cowpea intercrop.
Experimental agriculture
,
30
(03), 349
–
358.
115
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.
Cechin, I., Press, M.C. 1993.
Nitrogen relations of the sorghum
Striga hermonthica
host
–
parasite associations: germination, attac
hment and early growth.
New phytologist
124: 681
–
687
Chang M and Lynn DG, 1986.
Houstoria and the chemistry of host recognition in parasitic
angiosperms.
Journal of Chemistry and Ecology
12:561
–
579
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 Experimental 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.
116
Dorward, P., Craufurd, P., Marfo, 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 grow
th regulator 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., La
ngyintuo, 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.
Ejeta, G. 2007.
The
Striga
scourge in Africa: a growing pandemic.
Integrating new
technologies for Striga control: towards ending the witch
–
hunt
, 3
–
16.
Ejeta, G., & Butler, L. G. 1993.
Host
–
parasite interactions throughout the Striga life cycle,
and their contributions to Striga res
istance.
African Crop Science Journal
,
1
(2).
Eplee, R. and Langston, M. 1991
. Intergrated 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.
Fageria, N. K., & Baligar, V. C. 1997.
Upland rice genotypes evaluation for phosphorus use
efficiency.
Journal of Plant Nutrition
,
20
(4
–
5), 499
–
509.
FAO STAT, 2010
(
http://faostat.fao.org
).
117
Fis
c
her, 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 ribulos
e 1
–
5
–
phosphate carboxylase Planta 169,536
–
545.
Frost, D.L., Gurney, A.L., Press, M.C., Scholes, J.D., 1997.
Striga hermonthica
reduces
photosynthesis in sorghum: the importance of stomatal limitations and a potential
role for ABA. Plant Cell and Environm
ent 20, 483
–
492.
Garba, A., Fagam, A.S., Fushion, G.G. 2007.
Contribution of environmental factors to
tillering and yield of rice during the dry season in Bauchi, Nigeria. Int J Trop Agric
Food Syst 1:42
–
47
Gebisa Ejeta and Jonathan Gressel. 2007.
Integrat
ing new technologies for
Striga
control:
towards ending the witch hunt; World Scientific Publishing Co. Pte
–
Ltd.
Gebremedhin, W., Goudriaan, J., & Naber, H. 2000.
Morphological, phenological and
water
–
use dynamics of sorghum varieties (Sorghum bicolor) under Striga
hermonthica infestation.
Crop Protection
,
19
(1), 61
–
68.
Gerakis, A., & Baer, B. 1999.
A computer program for soil textural classification.
Soil
Science
Society of America Journal
,
63
(4), 807
–
808.
Gethi, J
.G
., Smith, M.E., Mitchell, S.E., and Kresovich, S. 2005
.
Genetic diversity of
Striga hermonthica and Striga asiatica population in Kenya. Weed Res. 45:64
–
73
Gressel, J., A., Hanafi, G., Head, W. Marasas,
Obilana A.B., Ochanda, J., Souissi, T
and Tzotzos, G. 2004
. Major heretofore intractable biotic constraints to African
118
food security that may be amenable to novel biotechnological solutions
Crop Prot.,
23: 661
–
689
Gridley, H. E., Jones, M. P.,
& Woperei
s
–
Pura, M. 2002.
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 Exper
imental Botany
,
46
(12), 1817
–
1823.
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., P
ress, 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.
119
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
Hewitt, E.J. 1966.
Sand and water culture methods used in the study of plant nutrition
.
Farnham, UK: Common
wealth Agricultural Bureaux.
Huang, K., Whitlock, R., Press, M. C., & Scholes, J. D. 2012.
Variation for host range
within and among populations of the parasitic plant
Striga
hermonthica
.
Heredity
,
108
(2), 96
–
104.
Ikeda, A., Ueguchi
–
Tanaka, M., Sonoda, Y.
, Kitano, H., Koshioka, M., Futsuhara, Y., &
Yamaguchi, J. 2001.
Slender rice, a constitutive gibberellin response mutant, is
caused by a null mutation of the SLR1 gene, an ortholog of the height
–
regulating
gene GAI/RGA/RHT/D8.
The Plant Cell
,
13
(5), 999
–
1
010.
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
.
Ishikawa, 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.
Ivens G.W. 1989.
East African weeds
and their control 2
nd
Edition, Oxford University press
120
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.
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 glaberrima
.
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., Ra
ttunde, H.F.W., Geiger, H.H. (Eds.), Breeding for
Striga
Resistance in
Cereals: Proceedings of a Workshop 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 i
n upland rice improvement.
Euphytica
,
94
(2),
237
–
246.
Joshi, A., & Witcombe, J. R. 1996.
Farmer participatory crop improvement. II. Participatory
varietal selection, a case study in India.
Experimental Agriculture
,
32
(04), 461
–
477.
Joshi, A., & Witcombe,
J. R. 1998.
Farmer participatory approaches for varietal
improvement.
Seeds of choice: Making the most of new varieties for small farmers
,
171
–
189.
121
Joshi, K. D., Rana, R. B., Subedi, M., Kadayat, K. B., & Sthapit, B. R. 1995.
[Effectiveness of participatory variety testing and dissemination programme: a case
study of Chaite rice in the Western Hills of Nepal].[English].
LARC Working Paper
(Nepal). no. 95/49.
Joshi, K. D., Sthapit, B., Subedi, M., Witcombe, J. R., Cleveland, D.
A., & Soleri, D.
2002.
Participatory plant breeding in 10 rice in Nepal.
Farmers, Scientists, and Plant
Breeding: Integrating Knowledge and Practice
, 239.
Kaewchumnong, K., & Price, A. H. 2008.
A study on the susceptibility of rice cultivars to
Striga her
monthica
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.
Kende, H., Van Der Knaap, E., & Cho, H. T. 1998.
Deepwater rice: a model plant to study
stem elongation.
Plant Physiology
,
118
(4), 1105
–
1110.
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 hermonthica,
and stemborers
in sorghum, Sorghum bicolor, through intercropping with green leaf desmodium,
Desmodium intortum. Int.
J. Pest Manag. 52(4), 297
–
302.
122
Khush, G. S. (1997).
Origin, dispersal, cultivation and variation of rice. Plant Mol. Biol. 35,
25 34.
Kijima, Y., & Sserunkuuma, D. 2013.
The adoption of NERICA rice varieties at the initial
stage of the diffusion process in Uganda.
African Journal of Agricultural and
Resource Economics
,
8
(1), 45
–
56.
Kijima, Y., Otsuka, K., &Sserunkuuma, D. 2011.
An inquiry into constraints on a green
revol
ution 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 varieti
es
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.
Letousey, P., De Zélicourt, A., Vieira Dos Santos, C., Thoiron, S., Monteau, F., Simier,
P., & Delavault, P. 2007.
Molecular analysis of resistance mechanisms to
Orobanche cumana
in sunflower.
Plant Pathology
,
56
(3),
536
–
546.
123
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
–
16365.
Lopez
–
Raez, J.A., Matusova, R., Cardoso, C., Jamil, M., Charnikhova, T., K
ohlen, W.,
Ruyter
–
Spira, C., Verstappen, F., Bouwmeester, H., 2009.
Strigolactones:
ecological significance and use as a target for parasitic plant control.
Pest
Management Science 65, 471
–
477.
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
McKenzie, F. R. 1998.
Influence of applied nitrogen on vegetative, reproductive, and aerial
tiller densities in Lolium perenne L. during the establishment year.
Australian
journal of agricultural research
.
Ministry of Agriculture, Animal Industry and Fisheries (MAAIF). 2009.
U
ganda
National Rice Development Strategy. Second 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:
124
Ejeta, G., Gressel, J. (Eds.), Integrating New Technologies for
Striga
Control
–
Towards ending the witc
h hunt World Scientific Singapore, Pp. 71
–
84
Mohammed, K.I., Musse
lman, L.J., Riches, C.R., 2001.
The genus
Striga
(Scrophulariaceae) in Africa. Annals of the Missouri Botanical Garden 88, 60
–
103
Musselman L.J and Press M.C. 1995.
Parasitic Plants. London,
UK: Chapman and Hall.
Musselman LJ, Parker C, Dixon N 1982
.
Notes on
a
utogamy and flower structure
agronomica
lly important species of Striga
(
Scrophulariaceae) and Orobanche
(Orobanchac
eae). Beitrage zur Biologie der
Pflanzen, 56, 329
–
43
Musselman, L.J.
(Ed) 1987
. Taxonomy of witch weeds. Page 3
–
12. In: Parasitic plants in
Agriculture
Striga
, CRC Press. Boca Raton. USA 317Pp
Namutumba Census Report. 2007.
Namutumba District 2002 Population and Housing
Census
Nanfumba, D., Turyahabwe, N., Ssebuliba, J., Ka
kuru, W., Kaugule, 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.
Normile, D, 2004.
Yangtze seen as earlier rice site
.
Science
275:309
Nweze, A.C., Nweze, J.A., and Nweze J.E. 2015.
Witchweed (Striga asiatica): A destructive
crop plant parasitic. Retrieved
from
https://nwezejustus.wordpress.com/2015/07/28/witchweed
–
striga
–
asiatica
–
a
–
destructive
–
crop
–
plant
–
parasitic
–
weeds/
Oerke, E.C and Dehne, H.W. 2004.
Safeguarding production losses in major crops and role
of Crop protection, Crop prot.23, 275
–
285
125
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 interac
tions between sorghum
roots and parasitic weed
Striga hermonthica
: cytochemical aspects of cellulose
distribution in resistant and susceptible host tissues Canadian Journal of Botany 69,
1679
–
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.
Olupot, J. R., Osiru, D. S. O., Oryokot, J., & Gebrekidan, B. 2003.
The effectiveness of
Celosia argentia (Striga “chaser”) to control Striga on sorghum in Uganda.
Crop
protection
,
22
(3), 463
–
468.
Ongaro, V., & Leyser, O. 2008.
Hormonal control of shoot branching.
Journal of
Experimental Botany
,
59
(1), 67
–
74.
Oswald A., 2005.
Striga
control
–
technologies and their dissemination; Crop Protection
Journal: 24, 333
–
342.
Oswald, A., & Ransom, J. K. 2001.
Striga
control and improved farm productivity using
crop rotation.
Crop Protection
,
20
(2), 113
–
120.
126
Pageau, K., Si
mier, 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.
Pageau, K., Simier, P., Naulet, N., Robins, R., & Fer, A. 1998.
Carbon dependency of the
hemiparasite
Striga hermonthica
on Sorghum bicolor determined by carbon isotopic
and gas exchange analyses.
Functional Plant Biology
,
25
(6),
695
–
700.
Parker, C. 2009.
Observations on the current status of Orobanche and
Striga
problems
worldwide.
Pest management science
,
65
(5), 453
–
459.
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.
P
itcher
J
, 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
Plan for Modernization of Agriculture (PMA) Secretariat. 2009.
Rice Value Chain Study
in Acholi and Lango Sub
–
regions. Plan for Modernization of Agriculture (PMA)
Secretariat. Kampala, Uganda.
Porteres,
R. 1956.
Taxonomic agrobotanique der rizcultives O. sativa Linneet
O. Glaberrima
Steudelo, J. Agri.
Press MC, Smith, S., & Stewart, G. R. 1991.
Carbon acquisition and assimilation in
parasitic plants.
Functional ecology
,
5
(2), 278
–
283.
127
Press, M.C and Stewa
rt, 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.
Tropical crops; Monocotyledons vol.1 and 2. Published by Longman
Group UK Ltd.
Rahman, M. A., Thant, A. A., Win, M., Tun, M. S., MOET, P., Thu, A. M., & Labios,
R. V. 2015.
Participatory varietal selec
tion (PVS): a" bottom
–
up" breeding approach
helps rice farmers in the Ayeyarwady Delta, Myanmar.
SABRAO Journal of Breeding
& Genetics
,
47
(3).
Ramaiah, K.V., Parker, C., Vasudeva Rao, M.J., and Musselman, L.J., 1983
.
Striga
identification and control handbo
ok. Information Bulletin No. 15. Patancheru, A.P.,
India: International crop Research Institute for the Semi
–
Arid Tropics
Rank, C., Rasmussen, L. S., Jensen, S. R., Pierce, S., Press, M. C., & Scholes, J. D. 2004.
Cytotoxic constituents of Alectra and St
riga species.
Weed research
,
44
(4), 265
–
270.
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
. CIM
MYT.
Ransom, J. K., & Odhiambo, G. D. 1994.
Long
–
term effects of fertility and hand
–
weeding
on Striga in maize
(No. 632.58 PIE. CIMMYT.).
128
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: Moreno, 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 inco
mes 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 Journal of Crop Protection; 29: 210
–
221.
Rodenburg, J and Bastiaans L.
2011.
Host
–
plant defence against
Striga
: reconsidering the
role of tolerance. Weed Research, doi: 10.1111/j.1365
–
3180.2011.087.x.
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 selection for
Striga
resistance and tolerance in sorghum be improved?
Field Crops Research
,
93
(1),
34
–
50.
129
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.
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.
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 upla
nd 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.
130
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.
Skinner, R. H., &
Nelson, C. J. 1992.
Estimation of potential tiller production and site usage
during tall fescue canopy development.
Annals of Botany
,
70
(6), 493
–
499.
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 Ts
uboi, 2005.
Paper presented at the WARDA
–
NERICA rice Workshop, Ivory
Coast, 8thOctober, 2005.
Taylor, A., Martin, J., & Seel, W. E. 1996.
Physiology of the parasitic association between
maize and witchweed (Striga hermonthica): is ABA involved?.
Journal
of
Experimental Botany
,
47
(8), 1057
–
1065.
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 UBO
S
statistical house available at
www.ubos.org
(accessed May 2013).
Umehara, M., Hanada, A., Magome, H., Takeda
–
Kamiya, N., Yamaguchi, S. 2010
.
Contribution of strigolactones to the inhibition of tiller bud growth under p
hosphate
deficiency in rice. Plant Cell Physiol 51: 1118
–
1126
131
Van Ast, A., & Bastiaans, L. 2006.
The role of infection time in the differential response of
sorghum cultivars to Striga hermonthica infection.
Weed research
,
46
(3), 264
–
274.
Van Ast, A., Bastia
ans, L., & Kropff, M. J. 2000.
A comparative study on Striga
hermonthica interaction with a sensitive and a tolerant sorghum cultivar.
Weed
Research
,
40
(6), 479
–
493.
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.
Westwood, J. H., Yoder, J. I., & Timko, M. P. 2010.
The evolution o
f parasitism in
plants.
Trends in plant science
,
15
(4), 227
–
235
.
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
132
Witcombe, J. R., Petre, R., Jones, S., & Joshi, A. 19
99.
Farmer participatory crop
improvement. IV. The spread and impact of a rice variety identified by participatory
varietal selection.
Experimental Agriculture
,
35
(04), 471
–
487.
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., 2001.
Host
–
plant recognition by parasitic Scrophulariaceae. Current opinion in
Plant Biology 4, 359
–
365.
Yoneyam
a, K., Xie, X., Kusumoto, D., Sekimoto, H., Sugimoto, Y., Takeuchi, Y., &
Yoneyama, K. 2007a
. Nitrogen deficiency as well as phosphorus deficiency in
sorghum promotes the production and exudation of 5
–
deoxystrigol, the host
recognition signal for arbuscula
r mycorrhizal fungi and root parasites.
Planta
,
227
(1), 125
–
132.
Yoneyama, K., Yoneyama, K., Takeuchi, Y., & Sekimoto, H. 2007b
. Phosphorus
deficiency in red clover promotes exudation of orobanchol, the signal for
mycorrhizal symbionts and germination stim
ulant for root parasites.
Planta
,
225
(4),
1031
–
1038.
133
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
summary
for plant height per rice variety for 2014
–
2015
under
Striga
infested field conditions
.
***
Significantly different at 0.001
Appendix 3: Analysis of variance
summary
for the number of tillers per rice variety for
2014
–
2015 under
Striga
infested field conditions.
Source of
variation
d
f
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
2014A
6.5
0.12
2.34
135
1.42
75:8:17
2015B
5.8
0.10
2.89
174
1.24
74:11:15
Source of
variation
D
f
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
134
Appendix 4: Analysis of variance
summary
for
rice biomass, grain yield and
yield
components of rice varieties for 2014 and 2015 seasons under
Striga
infested conditions.
Source of variation
df
Rice biomass
(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
summary
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
summary
of
Striga
components recorded per rice
variety for 2014
–
2015 in Namutumba district
,
Eastern Uganda.
Source of variation
D
f
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
135
Appendix 7: Analysis of variance
summary
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
summary of n
umber of tillers and plant height rice
varieties in 2014 in Namutum
ba district
,
Eastern Uganda.
***
Significantly different at 0.001
,
*
significantly different at 0.05
Appendix 9: Analysis of variance
summary
of
rice biomass, grain yield and
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
D
f
SED (days)
D
f
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
D
f
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
Rice biomass
(g)
Panicle dry
weight (g)
Rice grain
recovery
(%)
Harvest
index
Grain dry
weight (t/ha)
Variety
23
7508***
145042***
56.83***
639.82***
4.62***
Residual
82
1179
15463
76.81
68.51
0.45
136
A
pend
ix 10: Analysis of variance
summary
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
summary
of
rice biomass, grain yield and
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
summary
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
Rice biomass
(g)
Panicle dry
weight (g)
Rice grain
recovery
(%)
Harvest
index
Grain dry
weight (t/ha)
Variety
24
20438***
221899***
575.54***
1042.03
***
6.55***
Residual
95
1533
24592
85.06
71.17
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
137
Appendix 13: Analysis of variance
summary
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
summary
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
summary
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***
2.816
*
2.382***
28.28***
0.252
***
Treatment
1
113.88***
22.996
***
5.143**
36.30**
1.342
***
Variety x treatment
13
13.95***
2.072*
0.489ns
2.943ns
0.105
*
Residual
84
4.68
1.357
0.639
4.872
0.066
138
Appendix 16: Analysis of variance
summary
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?
…………………………………………………………………………………………………………….
…………………………………………………………………………………………………………….
139
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)?
Variet
y (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.
140
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.
141
*
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***
142
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 (Ble
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
143
Data source: Tororo district meteorological station Eastern
Uganda.
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
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 AUGUST 2016 i… [600713] (ID: 600713)
Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.