Africa Rice Annual Report 1987

CON T E 11" T S
INTRODUCTION
EVOLVING VARIETIES
SOIL AND CROP 111ANAGEMENT
CONTROLLING PESTS AT FARNERS I LEVEL
ON-FARM AND ADAPTIVE TRIALS
_ - - - - - - - - - - - -- 1
- - - - - - - - - - - - - - 7
- - - - - - - - - - - - 13
- - - - - - - - - - - - - 23
INTERNATIONAL AND NATIOnAL CO-OPERATION - - ... - - - - - - - - - 31
TRAINING - - - - - - - - - - - - 33
PUBLICATIONS
PERSONNEL
NOTE:
- - - - - - - - - - - - 34
- - - - - - - - - - - - 36
This report in its present form has not been officially approved by
the West Africa Rice Development Association, Monrovia, Liberia.
INTRODUCTION
An estimated 200,000 he. of tidal and associated swamps are cultivated
in' The Gambia, Guinea, Guinea-Bissau, Nigeria, Senegal and Sierra Leone.
These swamps are relatively more fertile than other rice environments and
farmers' average yield is 2.0 tons/ha.
Over 100,000 farm families in the region are dependent on rioe cultivation
in the mangrove swamps their livelihoM. They are confronted with several
environmental and stl.cio-economic constraints.
Research at the WARDA Rokupr Station is undertak()n to improved
technologies for higher and sustainable for the resource -
riee farmers in West Africa. With a multi-disciplillar7 team
approach, we accomplish this objective in collaboration with the national
agenoies and organisations responsible for research, training
transfer in the various member cormtries.
In northwest Sierra Leone, along the Great Scaroies and southern
Sierra Leone along the Bumpeh river about 53% of the farmers in the short
season areas are now growing ROX 5, WAR 1 and Rohyb 6. This means an increase
of up to 1.5 tons/ha of their production. The use of improved varieties and
the single axle traotor/power tiller as recommended by us formed the basis
the proposal for donor \EEC) sponsored Northwestern Integrated Agricultural
Development Projeot (NWIADP) at Xambia, Sierra Leone. In The Gambia and
Guinea-Bissau farmers grow ROX 5 to the extent of about 20 and 60 peroent,
respectively, along side their traditional varieties. Additionally, four
other varieties also introduoed by WARDA are being multiplied for distribution
in Guinea-Bissau.
Our achievementBover the years have been aooomplished against a
background of limited laboratory facilities and infrastructure, logistio
problems, erra&1b water and electrioity supply, and poor links
with Rokupr, operating base (no telephone/cable/telex/radio).
Looking ahead, emphasis will be placed on responding to the changing
needs and strengths of national programs. We will continue to find solutions
to the environmental land biologioal stresses to rioe in the mangrove swamps
in closer partnerShip with National Agrioultural Researoh Systems (HARS) to
enable low income farmers of this rice environment to have a higher and
tmStainable produotiono
EVOLVING IMPROVED YARIETUS
The varietal improvement program focused on the identification from
existing varieties, and breeding of new higher yielding varieties with improved
tolerance to the major environmental and biological stresses suoh as salinity,
iron toxic! ty, acidity and orab damage.
Germplasm Introduction
Since 1977, approximately 6,000 varieties have been introduced and screened
for their adaptation to the. mangrove swamp environment. Selected
with high-yielding ability, fertilizer responsiveness, toleranoe to
stresses, resistanoe to major inseot pests and diseases, and suitable plant
stature have been identified and are being exploited in the breeding program.
Materials introduoed and soreened in observational trials in 19S7 were
nbtained through. IRTP and oomprised of 680 lines from 1986 IRRSWON, ITPRON,
IRSATON and Aoid lowland soreening set.
IRRSWON: The trial was established on a site subjeoted to shallOW tidal
flooding (.)a 20em) for three or four days at each period of high tide, As in
previous years many of the entries in the nursery proved of interest and 38
were seleoted for further testing. Most of the seleotions were intermediate
statured, short duration lines suitable for mangrove swamp areas of
Guinea-Bissau and southern Senegal.
ITPRON: The test lines were planted on a site subjected to deep tidal tlood1n«
( ,. 700m) • Thirty-five lines performed exceptionally well with higher grain
yield and better phenotypio aoceptability scores than the oheck varieties.
These were selected for further yield testing. Sixteen of these selections
were tall statured, long duration varieties (180-200 days) considered
appropriate for the deep flooded long season areas.
IRSATON: The salinity screening set was transplanted at Balanoera, a salt
affeoted site in northwest Sierra Leone. The seedlings were transplanted
early in July to expose them to severe stress at the seedling stage. However,
severe crab damage was encountered due to severe infestation by crabs at the
trial site, and most test lines were wiped out. The Surviving lines showed
poor adaptation and only small amounts of seeds were harvested from BR9-17-4-2,
CISADANE, IR48, IR32307-107-3-2-2, IR9808-9-2, Pokkali and
PR106.
Acid Lowland Screening Set: The entries were planted on a tidal site with
initial PH of 4.2. The soil pH increased during the course of the trial.
Twelve best entries rated on their reaction to soil problems and ability to
withstand tidal flooding were selected, and will be further tested next season.
2.
Germplasm Cellection and Characterization
Collection and evaluation of traditional mangrove swamp rice varieties
from the region were initiated in 1977 to establish the varietal types favoured
by farmers and to provide a source of well-adapted germplasm for our breeding
pregram. At the end of 1987, the collection had 768 accessions from
Sierra Leone, The Gambia, Guinea, Guinea-Bissau, Nigeria and Senegal.
Breeding Pregram
The chances of identifying varieties that are better than the traditional
varieties from existing improved ones are low. A breeding program was therefore
initiated in 1978 to develop higher and stable yielding varieties for the
different mangrove swamp conditions encountered in the region.
During 1987, 34 new crosses were made to' incorporate desirable
characterist·· 0 such as tolerance to' salinity, iron toxicity and ree.ietane&
major insect pests and diseases into improved varieties.
This season, the breeding nurseries contained over 3,500 segregating lines
on station and 200F
4
and FS populations at off-statiO'n problem sites.
A bulk selection method was used to improve on the selection in deep flooded
and adverse tidal swamps. A rapid generation advance teohnique was used to
speed up the generation rate of long duration photeperiod sensitive
lines. Over 2,000 plant selections were made and advanced to the next
generation.
Adverse Soil Tidal and associated rice fields cover a wide spectrum
of Boils. The development of genetic materials that are adaptable to these
conditions will allow for expansion ef land under rice and provide a means of
increasing annual production of mangrove swamp farmers. In 1987, a total of
260 varieties were screened for salinity (180), iron toxicity (40) and acidity
(40) by growing the entries in specific problem soils contained in concrete
troughs. A total of 4,880 traditional and improved varieties/advanced lines
have so far been screened for this purpose.
Several promising advanced lines including WAR81-2-1-2, ' WAR77-3-2-2,
WAR81-2-1-3-2, WAR98-6-1-9-1 jWAR74-23-2-2-B-2, WAR79-4-Rl-2-1 and
WAR91-2-4-1 have shown multiple tolerance to soil stresses. Under field
testing in moderate salinity (6-8 mmhos/cm) at an iron toxic site at Balancera,
northwest Sierra Leone, some of these advanced lines gave yields over 2.0 .t.
tons/ha (Table 1).
Further, 30 WARDA - Rokupr F 4 lines and 11 cuI ti vars were screened in the
laboratory and field, respectively for tolerance to iron toxicity at the WARDA
Nursery farm at Suakoko, Liberia.
3.
Some of the tolerant F<_ lines identified were WAF\.9a-2-6-1, WAR100-3-2-1,
WAR100-3-4-1, and WAR100-4-7-1. Two of the cultivare that were
rated as tolerant to iron toxicity were WAR81-2-3-1 and Fant 213. These would
serve as donors for iron toxicity tolerance in future breeding programs.
Table 1. Grain yield and tolerance of some promising lines
in saline and mild iron toxic field - Balancera,
Sierra Leone, 1987.
Line
WAR77-3-2-2
WAR 1 (Rohyb 15)
WAR50-3O-1-3-1
WAR81-2-1-2
Rohyb6-WAR-6-2-B-2
WAR29-3-3-1
WAR72-2-1-2
WAR52-284-3-2-1
ROK 5 (check)
Cross
IR4595-4-1-15/Merr 10SA
Wellington/SR 26//CCA
B5141B-Kn-1-1/Konday 225
Miniku 32A/I. Mahsuri
CCA/RH 2
Mahsuri/Pama Bangura/ /
Ginsa Killing
Merr 108A/BG90-2
Faro 15/Tail Kokoyo
Wellington/SR 26
Grain
Yield
(tons/hal
3.2
3.1
3.0
3.0
2.6
2.4
2.2
2.2
1.8
MT
=
Moderately tolerant; f.m = Moderately Susceptible.
Reaction to
Salt Iron
MT
MT
MT
MT
MT
MT
MT
MT
MT
toxicity
MS
MT
MS
MT
MS
MT
MT
MS
MT
Disease. Insect and Crab Damage Resistance: Mangrove swamp rice is attacked by
diseases, insects and crabs. Of these, blast, brown spot, leaf scald, stem
and tidal swamp crabs are most important. Breeding for resistance for
these pests is therefore essential for sustainahle rice yields.
Over 200 have been evaluated by 1987 to identify
sources of resistance to crab attack. Cultivars that have shown some amount of
resistance to crabs such as Boyah 246, Damalai 247A, 2478, and Rolontho
265A were used this season as donors in the breeding program for crabs. Also in
1987, 40 promising rice varieties/advanced lines were screened and their mean
percent crab damage ratings were tested for correlation with some morphological
characteristics of the rice plant. The correlation between percent crab damage
and leaf length was highly significant (r = -O.tJ-.iP**.). Eight varieties including
five Rokupr bred advanced lines were outstanding with crab damage ratings of less
than five percent (Table 2),
4.
Table 2, Screening for resistance to crab damage among promising
rice selections - Rokupr, 1987.
Variety Source/Cross
IR 25429-R-WAR-1 IRRI/WARDA /- RGA
Raden Mas Asia
WAR 72-2-1-1 Merr 1081l/BG90-2
WAR 77-3-2-2 IR 4595-4-1-15/Merr 10BA
WAR 44-5-5-2 Raja Sal/BL 4E
WAR 44-21-3-2 Raja Sal/BL 4E
WAR 49-5-1-3-1 B 541BKN-47-1-1/Yenken Yonkah
SL 22617 RRS - Rokupr
ROK 5 (Susceptible Check) Wellington/SR 26
WAR = Rokupr bred advanced lines.
Yield Trials of Promising Varieties/Advanced Lines,
Crab
damage
level ~ ~ )
4.3
3.3
1.0
3.0
4.3
4.7
4.7
4.3
13.7
In continuation with the testing of varieties/advanced lines selected in
previous seasons, a number of observational yield trials (OYTs), replicated
yield trials (RYTa), farmers' field trials (FFTs) and advanced variety trials
(A VTs) were conducted during 1 987.
~ : Three OYTs were conducted during the season. One trial sited in the
associated swamp tested short duration varieties (120-135 days) for types suited
to bunded swamps in the north of the region. The second and third trials were
sited in the tidal mangrove swamp and tested medium duration varieties (135-155
days) and long duration ones (155-200 days) for suitability for use in Sierra
Leone, Guinea and Nigeria.
In the short duration associated swamp trial, 38 varieties w e ~ e tested
against IR 1073-143-2-3 as check. The trial was located at the edge of a swamp,
with low natural fertility and mild iron toxicity problems. A basal application
of 60 kg N/ha and 20 kg P
2
0
5
/ha were given. Eleven varieties outyielded the
best check plot and were selected on this basis (Table 3). A further 10
varieties were selected on the baSis of their, performance in the field, and were
shorter in duration than the check variety. Their earliness coupled with their
short stature indicated that these materials would be suitable for use in
non-tidal swamps in the short rainy season areas of northern Guinea-Bissau and
southern Senegal.
5.
Table 3. Performance of new selections from" .bservational yield
trials of short duration varieties under associated swamp
oonditions - Rokupr, 1987.
Variety
Yield
( tons/ha)
% Yield Plant Growth
increase height! Duration

over adja-
(cm) (Days)
cent check
variety
WAR 115-1-2-10-5
5.6 68
95
132
WAR 115-2-2-1-1
5.5 36 87 128
BR 50-120-2
5.0 57
81 132
IR 29912-63-2-2 4.7 49 92
128
WAR 115-1-2-11-2 4.6 45 93 135
WAR 100-2-15-1 4.3 100
90 133
WAR 100-2-1 2-1 4.3 44 92 135
IR 1073-143-2-3* (Mean of 20
plots) 3.1 98 135
*
Check variety; WAR = Rokupr bred advanced lines.
In the medium duration tidal mangrove swamp OYT, 47 lines were tested
against ROK 5. A basal dressing of 60 kg N/ha was given. Two entries,
BW 295-5 and BW 295-4, out yielded their adjacent ROK 5 plots and were selected
along with eight more varieties which showed good performance in the field.
The selections were comparable to ROK 5 but up to two weeks longer in duration
and were virtually all intermediate statured. These would seem suitable for
flooded swamps in the long rainy season zones.
Sixty-eight long duration entries were tested on a site g to deep
tidal flooding. Fertilizer application was the same as for the tidal mangrove
swamp OYT. Yields recorded for the trial were low, with the check variety,
Kuatik Kundur averaging 1 .9 tons/ha. Twenty-eight test varieties including 13
advanced lines, out yielded the best check plot were selected and for the shallow
flooding tidal swamps.
RYTs: Five RYTs, associated (2) and tidal mangrove swamp (3), were oonducted
during the season. The varieties tested were selected introductions from 1986
OYTs but included a small number of advanced lines from the breeding program
and the best varieties from last years' trials. A basal dressing of 60 kg N/ha
was given to all trials while 60 kg P205/ha was also basally applied in the
associated swamp trials.
RYT (Short Fifteen varieties were tested against IR 10781-143-2-3
as check in the assocated swamp, BG 380, BR400-1 and IR 21855-53-2-1-2
Significantly outyielded the check variety with average yields of 4.2, 3.8 and
3.7 tons/ha, respectively.
RYT (Medium Duration): Two RYTs, one each in the tidal and associated
mangrove swamps were conducted during 1987.
In the associated swamp, 15 varieties were compared against ROK 5 as check.
The trial was conducted in a fertile swamp in the seepage zone. Five varieties,
WAR72-2-1-1, WAR77-3-2-2, WAR81-2-1-3-2, RTN16-1-1-1 and WAR 1, outyielded ROK 5
although not Significantly. However, WAR72-2-1-1 and WAR77-3-2-2 showed multiple
tolerance to salinity, iron toxicity and acidity.
In the tidal mangrove swamp, 14 varieties were compared with ROK 5 as
check. The best four varieties, 1, Rohyb4-WAR-1-3-B-2, B981d-51-35-1,
Rohyb4-WAR-1-1-B-1 were not significantly different from ROK 5. WARS7-10-2-3-2,
WAR73-1-M4-5-1-1-2, and Haji Haroun yielded significantly the same, but lower
than ROK 5.
RYT (Long Duration): Two RYTs to test long duration varieties were
conducted in tidal mangrove swamp during 1987. Attempt was made to stratify
varieties into two categories according to duration (155-180 days and 180-200
days).
In the longer duration category (180-200 12 entries were compared
with ROK 10 and CP 4. The first 11 varieties were not significantly different
from CP 4 and ROK 10. The top three varieties, WAR73-1-M2-1, WAR73-1-M1-4 and
WAR44-64-2-1 gave yields of 3.8, },7 and 3.5 tons/ha, respectively and
significantly outyielded an iron toxicity tolerant variety selected
from 1986 IRSATON with grain yield of 2.5 tons/ha.
In the second trial (155-180 days) only two varieties WAR42-82-2-3-1 and
WAR 1 02-1-3-1 significantly out yielded the check variety, Kuatik Kundur.
WAR42-82-2-3-1 and WAR 1 02-1-3-1 showed very good seedling blast tolerance
while Kuatik Kundur showed moderate tolerance.
7.
SOIL AND CROP MANAGEMENT
In 1987 emphasis was placed on improvement of soil fertility in acid
sulphate soil conditions. Research to identify nitrogen responsive rice
varieties among newly bred lines and to develop easily adaptable techniques
for improving nitrogen use efficiency in mangrove swamps continued. Studies
on utilization of Azolla in swamp rice production also continued.
Improvement of Soil Fertility
Efficacy of Phosphate Fertilizers: Phosphorus is a major limiting nutrient
for rice crop under acid soil conditions. In the past, moderate application
of Superphosphate (40 kg P
2
0
5
/ha) to rice on these seils improved growth and
brought about significant responses to applied nitrogen but higher levels of
phosphorus have not had significant effects in increasing rice yields. Past
research at Rokupr showed that phosphorus applied to mangrove soils became
rapidly bound to iron and aluminium. Much of the response obtained with the
application of phosphorus on these soils has been attributed to immobilization
of iron and aluminium in the soil, especially at the early stages rice
growth. Increasing the availability of phosphorus is therefore essential for
increasing the productivity of rice in these soil conditions.
A trial was initiated in 1986 to determine the efficiency of superphosphate
and phosphate rock materials, such as Taiba rock phosphate and Matam rock
phosphate from Senegal, for rice production in acid soil conditions and to
evaluate intensive use of phosphorus availability.
In two consecutive seasons of the trial, application of single superphosphate
has been more effective than the rock phosphate materials in improving rice
yields, but phosphorus levels above 40 kg P
2
0
5
/ha have not been effective in
increaSing rice yields irrespective of the form of phosphorus applied (Table 4).
The apparent lack response to application of the rock phosphate materials
presumably resulted from their low P dissolution rates and rapid immobilisation
of by iron and aluminium under acid sulphate conditions. The 1987
season results indicated no residual effects to the application of phosphorus
in previous seasons even at the high levels of application.
Amelioration of Acid Soil Conditions: Soil groups in mangrove swamps range
between constantly flooded pyrite saturated subgroups and seasonally aerated,
leached aluminium saturated clays with reserves of pyrite below 60cm in the
subsoil. Rice grown on the latter are subjected to toxic levels of iron and
aluminium and a deficiency of phosphorus. Multiple nutritional disorders and
severe infestation of brown spot disease are usually manifested by the crop.
8.
Table 4. Rice response to levels of phosphatic fertilizers in acid
soil conditions on the Swamp Catena at Rokupr, 1987
(Var. ROK 5).
Phosphatic Fertilizers
CAntrol - without Phosphorus
Taiba Phosphorus Rock
Ma.tam
,, -
"
Single Superphosphate
Taiba Phosphate Rock
Matam " "
Single Superphosphate
Taiba Phnsphate Rnck
Matam
" "
Single Superphosphate

Level of
applied
P
2
0
S
kg/ba*
40
40
40
80
80
80
120
120
120
Mean
Grain
Yield
( tons/ha)
3.1 c
3.8 abc
3.8 abc
4.1 ab
4.0 abc
3.3
abc
4.2 abc
3.7
abc
3.9
abc
3.8 a.bc
Means followed by the same letter in the column are not significantly
different at 5% DMRT.
* Nitrogen at 80 kg/ha was uniform.
Lime requirements based on test for acid sulphate soil conditions
are very high (16 tons/ha). In the past, regular application of 40 kg P
2
0
S
/ha
has been necessary for improving soil fertility conditions and
rice yield under these conditions at Rokupr. Liming at low levels
(3 tons/ha) plus moderate application phosphorus have also been found
effective for improving rice yields in acid sulphate soils. Research at WARDA
Rokupr in 1985 and 1986 showed that incorporation of decomposed rice husk
(plus bran) as soil amendment at 10 tons/ha was effective in improving nutrient
availability and rice yield under acid soil conditions. With the incorporation
of rice husk, conSistently higher grain yield responses were obtained to
applications of 80 kg and 120 kg N/ha. The organic amendment sustained higher
levels of productivity, yielding 0.6 tons/ha more paddy on average than
treatments without husk in the second season after application.
9.
In 1987, the interaction of graded levels of 0, 2.5, 5 and 10 tons/ha of
decomposed rice husk (plus bran) as soil amendment with 80 kg N/ha and
40 kg P205/ha on yield of rice in acid soil conditions were examined at Rokupr.
Significant improvements in rice yields were obtained with the soil
amendments even at the lowest level of husk application. Nitrogen and
phosphorus were also effective in improving rice yields. Addition ef
40 kg P
2
0,/ha increased grain yield responses to the levels of rice husk applied,
except at 10 tons/ha (Figure 1).
The indications were that incorporation of decomposed rice husk (plus bran)
reduced the toxic effects of aluminium and enhanced nutrient availability to
rjQe. Lew response to applied phosphorus at 10 tons/ha of the decomposed rice
. C •
husk suggested that the nutrient may not be limiting for rice growth under high
levels of soil amendment. Presumably the contents of phosphorus and Silicon in
rice husk, as well as the ability to complex with oxides and hydroxides .f iron
and aluminium may be important attributes of the soil amendment in the
amelioration of the nutritional di&orders encountered on these soils. Future
trials should examine the effect of other organic materials such as rice straw
on nutrient availability and their effect in sustaining rice yields under acid
sulphate soil conditions.
Improvement of Rice Yields with Azolla
Complementary use of Azolla and Chemical Nitrogen: Earlier trials conducted at
ROkupr demonstrat$d the potential of Azolla as a source of nitrogen for rice
production in the associated mangrove swamp. Incorporation of one crop of
Azolla (17-20 tons fresh weight-FW/ha) before transplanting rice increased rice
grain yield significantly above the control. A second incorporation .f Azolla
two weeks after transplanting guaranteed similar yields of rice as 40 kg N/ha,
the recommended dose of nitrogen if applied by injection technique·. Different
methods of Azolla cultivation with rice were evaluated. Cultivation.f Azolla
(ADUL 54 pp) as a mono crop was rapid, yielding a biomass of 2.32 kg-FW/m
2
on
average within 15 days of innoculation. As an intercrop with rice, Azolla
biomass production averaged 1.74 kg FW/m
2
in 28 days. Availability of
phosphorus and sunlight intensity were impotant factors a£fecting Azolla growth,
Despite a slower growth rate of Azolla, intercropning it with rice was
advantageous in reducing the weed infestation of rice considerably. As much as
50% reduction in weed infestation has been achieved with Azolla rice intercrop.
The effect of Azolla in sustaining rice yields were determined in subsequent
trials.
GrSl.i il Yi,.ald
tOilS/ ha • .
4. 0
r; .'l ,. .. ;;: u .. 8%
3.0
2.0
o Without
1. 0
o

o • .5
Fig ' )! I n teraction of I'ice
1 .. 1··· · .
':' ) '>-4 _, . ..
s oi l conditions -
1Q .. ()
in <:i.c i d
10.
One s8il incorporation of is generally feasible under humid tropical
rain-fed conditions. This facilitates an early build up of soil N which
declines as the crop matures necessitating supplementary nitrogen for maximizing
rice yield. A trial was conducted in the last three seasons to determine the
trend in soil fertility and sustainance of rice yields with different methods
Azolla cropping and supplementary levels of applied nitrogen in the associated
swamp at Rokupr • .
ef Azolla before transplanting rice gave consistently higher
yields of rice than Azolla/rice intercrop (Table 5). Over the three seasons of
the trial, the relative affects of Azolla cr"'pping on ric.e yield were stahle.
Residual effects from Azolla cropping were not and nitrogen x Azolla
interactions were also absent presumably because of the level of AzelIa biomass
used.. Incorp"'ration of higher Azolla biomass (2-3 crops of Azolla) would
contribute a greater amount of nitrogen to the rice crop and influence response
to the levels of chemical nitrogen applied.
Jable 5. Rice grain yield under long term Azolla cropping in the
associated mangrove swamp at Rokupr, 1985-1987 (Variety ROK 11).
Azolla
Creppini'
Without
Monecrop
Intercrep
Mean
1985
3.6
4.7 a
4.0 'b
Yield of Grain
1986
c 3.9 b
4.7 a
3.9
b
(tonsL
ha
2
1987
4.1 b 3.9
b
4.8 a 4.7 a
4.1 b 4.0 b
Means fellewed by the same letter in the column are not significantly
different at 5% DMRT.
c. V. C%) 9.5 8.2 9.6 9.1
Response to nitrogen in the range .. f 0-60 kgjha was linear under all
methnds of Az,.,lla cropping (r = 0.99*** without r = 0.97* Azolla monocrop;
r = 0.96* Azel;ta intercrop). The effect .. f Aznlla incorporation plus 60 kg Nlha
on rice yield was additive, and increased grain yield of rice over the
untreated contrel by 82% as compared with an iricrease of in the absence of
Azolla. Application of 60 kg Nlha under intercrepped Azolla increased rice
yield by 55%.
11.
Fertilizer Application and
Varietal Response to Nitrogen: Cellaborative studies with the breeding
section have identified HOK 5, ROK 10, 1:TAR 1, and i'lAR44-50-4-1 as nitrogen
responsive varieties. ROK 5 was found to be better adapted to the
associated swamp environment and efficiently utilized soil lJ for grain
production in earlier trials.
In 1987, trials to identify rices with efficient utilization and high
grain responses to nitrogen among the newly developed varieties for mangrove
swamp environments were conducted on the tidal mangrove and associated swamr s
at Rokupr. Four varieties were evaluated in each sub-ecology over a nitrogen
range of 0-150 kg N/ ha.
On the tidal mangrove swamp, application of 60 kg N/ha was optimum for
increasinff grain yield of rice irrespective of the variety tested. The yield
perforJMD.ces of and ROK 10 (check) were s ;. milar, out yielding
lNAR39-17-2-2 and WAR44-5-5-2 (Table 6).
Table 6. grain yield ef rice showing response of some newly
developed varieties to nitrogen in a tidal mangrove swamp
at Rokupr - 1987 season.
Varieties
Ni
o 30 60 90 1 20 150 Mean
Tv'lAR30-17-2-2
1.7 1 .9 2.7 2.6 2.7 3.0 2.4
\vAR44-5-5-2 1.4 1.9 2.6 2.5 2.8 2.6 2.3
'itT AR44-50-4-1 1.6 2.0 3.2 3.3 3.4 3.2 2.8
ROK 10 (check) 1.8 2.1 2.;1
:2.:2
:2.0 .2,..4
2.8
Means 1.6 2.0 2.9 2.9 3.0 3.1
----
C.V. (%) Nitrogen
=
18.5
C.v. Varieties
=
13.2
L.S.D.
0.05 for differences between nitrogen 0.43 means
=
L.S.D.
0.05 for differences between variety means 0.23
=
In grain yield response to rice varieties grown on the associated
swamp ecolegy was markedly influenced by the levels of nitrogen applied
(Table 7).
level!1l
12.
Table 7, Mean grain yield of rice showing response 'f some newly
devel,ped varieties to nitrogen in an mangrove
swamp at Rekupr - 1987 seasen.
WI

Varieties
aEElied
0 30 60 90 120 150 Mean
ROK 5 (check) 1 .6 c 2.0 'bc 2,5 bc 2.7ab 2.8a 2.6ab 2,4
Rohyb6-WAR-6-2-B-2 2.0a 2.6a 2 ... 4a 2.0a 2.3a 2.2a 2.3
WAR81-2-1-2 1.1 b 1,7b 1.5 b 2.5 b 2.5a 2.8a 2.0
WAR81-2-1-3-2 1,7b 2.5a 2.7a 2.9a 3.2a 3.0a 2.7
Means 1.6 2.2 2.3 2.5 2.7 2.7
Means follewed by the same letter in the row are not significantly
different at 5% DMRT.
C.V. (%) = 22.3
C.V. (%) Variety = 16.2
1YAR81-2-1-3-2 was higher yielding and produced significantly higher grain
yield responses to nitrogen in comparison to the ether varieties.
R.hyb6-WAR-6-2-B-2 was not responsive to the levels nitrogen applied and
WAR81-2-1-2 required relatively high levels ef nItrogen to increase grain
producti,n.
In general the yieldLwere lew and may be indicative ef the lew fertility
,f the seils in the associated swamps. The resuits indicated that
WAR81-2-1-3-2 was adaptable t, the asseciated mangrove swamp environment with
a for increasing rice under prevailing conditions.
CONTROLLING PESTS AT FARMERS' LEVEL
The programts thrust was geared toward the development and testing of
effective, low cost and easy to adopt pest control strategies for the small
farmer. Cultural practices that minimize pest infestation were formulated int.
integrated pest management packages and tested for the of the major
pests.
WEEDS
The weed control pngram in 1987 continued work on the development e6 .,
cultural practices for effective weed control. Research to identify suitable
herbicides for weed control especially in the associated mangrove swamps was
alsecfJJ,tinued at Rokupr.

Cultural Weed C,ijirol
Seedling Establishment Methods: trial was initiated in 1985 to determine
the effect of method of seeding and one handweeding, on stand establishment,
weed and grain yield of rice in the associated swamps. The treatments
,
included broadcast ungerminated seeds, broadcast pre-germinated seeds, direct
seed drilling, and 4, 6 and 8 week-old seedf ing transplants.
The 1987 results confirmed the findings .f previous seasons. Rice stand
establishment was lowest with direct seed and had whe highest weed
infestation compared te the treatments.
The grain yields, .f transplanted rice were higher than those obtained with
direct seeding. Six seedlings (2.4 tons/ha) and 8 week-old seedlings
(2.5 tons/ha) preduced significantly higher yields than the other treatments.
Direct seed broadcast (germinated"' er ungerminated) precludes nursery
preparation and transplanting with a potential for reducing the cost of
production. But, to obtain high yields an effective weed control practict:; ;",
becomes imperative. In this regard, a minimal use of herbicide may be
desirable.
Competi ti ve Ability of Rice Varieties with Weeds: This .,trial was
initiated this season to evaluate the competitive ability of 30 ri.e varieties
(Oryzae sativa L.) of various plant stature two weed management levels
(no weeding, and handweeding at 30 days after transplanting - DT).
Results obtained indicated that the varieties differed greatly in their
ability to supprel!ls weeds. Weed':- dry weight at 50DT varied from 1.4 €!1m
2
for
WAR50-49-1-2-1 (141 .Scm tall) to 9.1 g/m
2
for WAR49-S-1-3-1 (126am tall) with
fi!ne bandweeding.
14.
For the non-weeded control weed dry weight ranged from 91.3 g/m
2
for
WAR81-2-1-3-2 (96.4cm tall) to 12.6 g/m
2
for WAR49-5-1-3-1 (126cm tall).
However, there was no correlation between plant height of the varieties tested
and their abili;ty to suppress weeds. Possibly, the architecture .f the variety
coupled with faster canopy development may suppress weeds more effectively,
owing to shading effects.
Influence of Crop Density on the Competitive Ability of Weed: Rice variety ROK 5
was tested at four densities and two weeding regimes (no weeding, and ' one
handweeding at 30 days after transplanting) to determine their effect on weeds
and rice yields.
The plant spacing of 20 x 15cm (33 hills/m
2
) provided good control of
weeds than plant spacing er 20 x 20cm (25 hills/m
2
), 25 x 25cm (16 hills/m
2
) or
30 x 30cm (9 hills/m
2
). The unweeded plet of the 20 x 15cm plant spacing
exhibited enl,.. 7.6% reductien in yield as compared to 20.CJfo, 25.3% and 24.4%
in the 20 x 20cm, 25 x 25cm and 30 x 30cm plant spacings, respectively (Table 8).
The results s h ~ w e d that the wider the plant spacing, the higher tbe degree of
weed infestation, with resultant decrease in grRin yield. The closer plant
spacing ~ f 20 x 15cm possibly created a less favourable environment for weed
proliferation.
The results indioated that in associated swamps with severe weed problems
the plant density of 33 hills/m2 was ideal for ' minimizing weed competition with
the rice crop.
Table 8. . Effect of four plant spacing on weed growth and grain yield ef
ROK 5 in an associated mangrove swamp at Rokupr, 1987 season.
Plant spacing (cm)
20 x 15 Weeded
20 x 15 Unweeded
20 x 20 Weeded
20 x 20 Unweeded
25 x 25 Weeded
25 x 25 Unweeded
30 x 30 Weeded
30 x 30 Unweeded
L.S.D. 0.05
C.V. (%)
Grain
yield
(tons/ha)
2.2
2.1.
2.1
1.7
2.1
1 .6
2.t
1 .6
0.49
17.3
% Reduction
in yieldc
7.6
20.0
25.3
24.4
Dry weed weight
(g/m
2
)
2.5
16.4
2.0
24.6
3.2
23.9
7.8
30.8
11.8
57.9
15.
Nitrogen Application and Weed Control Methods: A field trial in an associated
mangrove swamp at Rokupr evaluated the effect of nitrogen application method
(broadcast at 14 DT, and deep placement hy injection method at 14 DT) and weed
control methods (one handweeding at 30 DT, two handweedings at 21 and 35 DT,
propanil at 91/ha plus bentazon at 41/ha at 21 DT, and an untreated c h e o ~ ) on
the growth of ROK 5 and weeds.
Method of N application did not significantly affect the degree of weed
infestation. Nitrogen applied by injection method had the lowest weed dry matter
production while plots without N produced the highest weed dry matter. This is
possibly due to better N-efficiency use by the crop in the N-injected plots which
resulted in vigorous crop growth and faster canopy development, thereby
providing better competition with weeds. Weed control methods affected weed
growth significantly. All weed control methods gave lower weed weights than the
untreated check. Two handweedings had the lowest weed weight, followed by
propanil (91/ha) plus bentazon (41/ha), which provided effective grass and sedge
control.
Nitrogen application and weed control methods affected grain J'ield
Significantly. The N-treated plots gave higher yields than when N was not
applied. Nitrogen applied by injection gave the highest yield but was not
significantly higher than N applied by broadcast. Grain yields of all weed
control treatments were higher than the untreated check. The mixture of
propanil plus bentazon gave the highest yield which was similar to that of two
handweedings (Table 9).
The results indicated that the efficiency of N application can be fully
realized with effective weed control, particularly with N-responsive rice
varieties such as ROK 5.
Chemical Weed Control
Promising Herbicides: Research to identify promising herbicides for weed control
in associated mangrove swamp rice continued during the 1987 season, in order to
identify suitable herbicides for this ecology. All the herbicides tested
provided significantly good weed control than the untreated check (Table 10).
The herbicides, Tamariz (propanil + thiobencarb at 91/ha), Basagran PL2 (bentaz on
+ propanil at 81/ha) and Stam F34T (propanil + fenoprop at 9l/ha) produced
significant yield increases over the untreated check and were as effective as
two handweeding at 21 and 35 DT.
1
6
.

T
a
b
l
e

9
.

E
f
f
e
c
t

o
f

N

a
p
p
l
i
c
a
t
i
o
n

a
n
d

w
e
e
d

c
o
n
t
r
o
l

m
e
t
h
o
d
s

o
n

t
o
t
a
l

w
e
e
d

w
e
i
g
h
t

a
n
d

g
r
a
i
n

y
i
e
l
d

o
f

R
O
K

5

i
n

a
n

a
s
s
o
c
i
a
t
e
d

n
a
n
g
r
o
v
e

s
w
a
m
p

a
t

R
o
k
u
p
r

-
1
9
8
7

s
e
a
s
o
n
.

T
o
t
a
l

W
e
e
d

C
o
n
t
r
o
l

M
e
t
h
o
d

o

k
g

N
/
h
a

M
e
a
n

M
e
a
n

O
n
e

h
a
n
d
w
e
e
d
i
n
g

a
t

3
0

D
T

1
3
.
4

6
.
8

2
.
5

7
.
6

2
.
0

3
.
5

3
.
8

3
.
1

T
w
o

h
a
n
d
w
e
e
d
i
n
g
s

a
t

2
1

a
n
d

3
5

D
T

3
.
9

1
.
3

2
.
7

2
.
6

2
.
5

3
.
5

3
.
9

3
.
3

P
r
o
p
a
n
i
l

(
9
l
/
h
a
)

a
t

2
1

D
T

1
0
.
3

6
.
3

1
1
.
4

9
.
3

2
.
7

3
.
6

3
.
8

3
.
4

P
r
o
p
a
n
i
l

(
9
l
/
h
a
)

+

B
e
n
t
a
z
o
n

(
4
l
/
h
a
)

a
t

2
1

D
T

3
.
1

5
.
1

1
1

.
3

6
.
5

3
.
1

3
.
7

4
.
0

3
.
6

U
n
t
r
e
a
t
e
d

c
h
e
c
k

3
0
.
4

3
7
.
9

2
6
.
2

3
1
.
5

1
.
5

2
.
0

2
.
9

2
.
1

M
e
a
n

1
2
.
2

1
1
.
5

7
.
0

'
)

2
.
4

3
.
3

3
.
7

.
.
.

,

L
.
S
.
D
.

0
.
0
5

Y
i
e
l
d

W
e
e
d

W
e
i
J
d
l
t

f
o
r

d
i
f
f
e
r
e
n
c
e
s

b
e
t
w
e
e
n

N

t
r
e
a
t
o
e
n
t

n
e
a
n
s

=

0
.
4
4

N
S

L
.
S
.
D
.

0
.
0
5

f
o
r

d
i
f
f
e
r
e
n
c
e
s

b
e
t
w
e
e
n

w
e
e
d
i
n
g

m
e
t
h
o
d

B
e
a
n
s

=

0
.
3
8

6
.
7

L
.
S
.
D
.

0
.
0
5

f
o
r

d
i
f
f
e
r
e
n
c
e
s

a
B
o
n
g

w
e
e
d
i
n
g

B
e
t
h
o
d
s

f
o
r

t
h
e

s
a
n
e

N

t
r
e
a
t
B
e
n
t

=

0
.
6
5

1
1

•
.

6

L
.
S
.
D
.

0
.
0
5

f
o
r

d
i
f
f
e
r
e
n
c
e
s

b
e
t
w
e
e
n

w
e
e
d
i
n
g

m
e
t
h
o
d
s

f
o
r

d
i
f
f
e
r
e
n
t

N

t
r
e
a
t
m
e
n
t
s

=

0
.
7
3

1
0
.
6

C
.
v
.

(
%
)

(
N

t
r
e
a
t
m
e
n
t
s
)

=

1
8
.
3

9
5
.
5

C
.
V
.

(
%
)

(
W
e
e
d
i
n
g

m
e
t
h
o
d
s
)

=

1
4
.
6

7
0
.
6

17.
Table 10. Effect of herbicides on weed control and yield of ROK 5
in an associated mangrove swamp at Rokupr.
Applic9.tion
Treatment Rate
( l/ha)
Tamariz
9
20 DT Fb 1 hwat 30 DT
Tamariz 10 20 DT
Arozin D 2.5 10 DT Fb 1hw at 30 DT
Basagran PL 2 6 20 DT Fb 1 hw at 35 DT
Basagran PL 2 8 20 DT
Stam F34T
9
21 DT
Stam F34T
9
21 DT Fb lhw at 30 DT
Stam F34T 10 21 DT
Propanil + Bentazon 9 + 4
21 DT
Propanil + Bentazon
9
+ 4 21 DT Fb 1hw at 35 DT
Handweeding 2 times 21 DT and 35 DT
Untreated Check
L.S.D. 0.05
C.V. (%)
11 DT = days after transplanting;
1hw = One handweeding.
Fb = followed by;
DISEASES
Weed
weight

4.3
19.5
17.6
3.8
18.8
11 .6
1 .5
12.9
11 _ 8
3.5
2.6
44.6
18.9
87.9
Yield
( tons/ha)
3.4
2.4
2.7
2.3
3.0
3.0
3.0
2.5
2.5
2.5
3.0
1.7
1.0
22.6
Ecological studies of the 'white tip' nematode, Aphelenchoides besseyi
started in 1986 was continued this season to ascertain on the seriousness of
the problem. Riee blast , Pyricularia oryzae, brown spot, Cochliobolus
miyabeanus,and leaf scald, Monographella albescens are the key diseases already
identified. Crop management practices that minimize the incidence of
spot were formulated into an integrated pest management package and tested in
the field.
Disease Ecology
'White tip'
A study on the tip' disease caused by the nematode
Aphelenchoides besseyi was initiated at Rokupr in 1986. Although 'White, tip'
symptoms have not been evidenced in the field for the last two seasons,
nematode infection was recorded at the panicle stage on ROK 5, 7, 8 and 9, SR 26 ,
and WAR 1 in 1986.
18.
The same observations). were made in 1987 and on Rohyb6-WAR-6-2-B-2. The level
of seed infestation observed in the previous season may warrant some
disinfestation of contaminated seeds in order to forestall the spread of the
pathogen.
Prototype Integrated Pest Management for Rice Diseases: In 1987, the packages
comprising (a) farmers' practice (his variety, method of raiSing seedlings
and transplanting, no fertilization and pest control), (b) Integrated Pest
Management (IPM) i.e. improved long duration cultivar (ROK 10) transplanting
of 6 week-old seedlings at 25 hills/m
2
in early July, and application .f
N
60
P
40
, and (c) IPM 2 i.e. IPM 1 plus fungicide spray at flowering/milk
of rice growth using tricyclazole (Beam) and Cupric hydroxide
. " -.", - . -
(Kocide 101) at 0.2 kg a.L/ha, respectively were compared at two locatiens.
In the Rokupr tidal mangrove swamp, the yields obtained were 2.4; 3.7
and 4.0 tons/ha for the practice, IPM 1 and IPM 2, respectively.
Foliar brown spot incidence was not significantly different between packages.
The two IPM packages out yielded farmers' practice probably because of
improved crop management practices.
In the short season zone at Moribaia, the variety WAR 1 was used in the
IPM packages. On average, the farmer's practice yielded 0.6 t/ha as compared
to 2.5 t/ha and 3.5 t/ha for IPM 1 and 2, respectively. Brown spot incidence
was relatively high at the site. The swamps at Moribaia are saline, a
condition which predisposes the rice plant to severe attack by Q. miyabeanus
probably because of reduced plant vigour, The incidence of brown leaf spot
was reduced by 17.1 and in 1 and IPM 2, respectively.
The low disease incidence in IPM 2 package and the consequent high grain
yield indicated that concomitant fungicide protection would be required in
the IPM package
INSECTS AND CRABS
Studies on the African white rice borer, Maliarpha separatella the most
predominant insect pest species focused on its seasonality in the rain-fed
upland rice nursery and monitoring of the adult moths in the sub-region by
sex pheromone. Grain yield losses in improved rice varieties attributable
to insect pests continued to be assessed under natural field infestation
while cultural means of pest control were investigated in order to develop
a cheap and effective integrated pest management strategy for the control of
the major insect pests and crabs.
19.
Ecological Studies of the African' White Rice Borer
Trials conducted on Maliarpha infestation of rice nurseries at Rokupr
indicated that the rice ?rop was attacked by this species as early as two
weeks after sowing in May/June (Figure 2). Farmers usually transplant eight
week-old seedlings in the Rokupr are, implying that they transplant infested
crop material in their rice swamps. No deadheart symptoms was evidenced in
the seedling crop attacked by Maliarpha inspi te of injury caused to the ::: , _.1::' i:-· .
seedlings. For effective control and management of this pest, crop protection
may therefore have to be initiated at the nursery stage.
Monitoring of the African White Rice Borer by Sex Pheromone
To develop an effective method of controlling Maliarpha, a dependable
means of monitoring needed to be devised, Through collaboration with the
Overseas Development Natural Resources Institute (ODNRI), the female sex
pheromone for this species has identified, synthesized, field tested and
proven effective in attracting male Maliarpha moths. In 1987, the synthesized
lure was used region-wide for the first time to monitor field incidence of
Maliarpha moths in Sierra Leone, Guinea and Guinea-Bissau (Table 11). It is
apparent that the lure could be used to develop efficient trapping techniques
which can be used to monitor and suppress Maliarpha populations.
Insect Pest Infestation and Grain Yield Loss
Stem borer incidence on farmers' fields is usually high, ranging from
60 to 90% hill infestation. Under these levels of incidence in fields, rice
varieties are known to interact with insect pests differently. Improved rice
cultivars have therefore continued to be assessed under maximum insecticide
protection to ascertain on their levels of grain yield losses under fertilized
and management practices to justify insect pest control action.
Table 11. Maliarpha male moth catches by sex pheromone in mangrove
swamp rice fields in Sierra Leone, Guinea and Guinea-Bissau
1987 season
Country
Location Male .. moth catchesLmonth ( totalL!2
June July Aug. Sept. Oct. Nov. Dec.
Sierra Leone Rokupr 0 0
3 9 56 200 233
Guinea Sonfonia 23 49 13*
Guinea-Bissau Bissau 63**
* Catch for 3 weeks; ** Catch for 4 nights.
N
u
:
-
;
:
h
e
-
:
'
3

2

1

1
<
'
-
'
f
?
:
-


.

2








i
J
i

2

o



•














4

6

8

1
0

1
2

1
4

1
6

T
i
m
e

a
f
t
e
r

s
C
9
d
i
n
g

(
V
e
e
k
s
)

I
n
c
i
d
e
n
c
e

o
f

M
a
l
i
a
r
p
h
a













e
g
3
,

l
a
r
v
s
,

a
n
d

p
u
p
a

i
n










:
0

s
e
e
d
l
i
n
g

a
g
e

i
n









u
p
l
a
n
d

r
i
c
e

n
u
r
s
e
r
y

-
R
o
y
u
p
r

1
9
8
7
.

20.
This season maximum crop protection commenced from the nursery stage
involving the improved varieties ROK 5, Rohyb6-WAR-6-2-B-2, WAR44-50-4-1,
WAR 1, WAR81-2-1-2 and ROK 10 under fertilized and unfertilized management
practices. WAR81-2-1-2, WAR 1 and ROK 5 (128-155 days) were most susceptible
to stem borer attack while ROK 10 (190 days) was most tolerant among test
varieties (Table 12). Maliarpha larvae beoomectess active feeders beginning
mid-October to early November which may possibly explain for the tolerance
observed in the long duration ROK 10 variety which matures in late December.
Generally, grain yield losses in the test varieties ranged from 23 to 40%
without fertilization and from 9 to 23% with fertilization. Fertilization
invigorated the crop and allowed it to tolerate pest attack better than the
unfertilized crop possibly through compensation.
Table 12. Average grain yield loss associated with stem borer
infestation in six improved rice varieties - Rokupr 1987.
Duration Percent Grain Percent
Variety (days) stem yield yield
infestation (tons/ha) loss
WAR 81-2-1-2 128
5.7 c 2.5 c 39.9
WAR 1
147 5.9 c 3.4 a 23.3
ROK
5 155 5.6 c 3.3 a 25.7
WAR 44-50-4-1 165 8.3 b 3.3 a 35.3
Rohyb6-WAR-6-2-B-2 165 8.6 b 3.1 ab 25.5
ROK 10 190 10.6 a 2.8 b 28.0
Means followed by the same letter in the column are not significantly
different at 5% DMRT.
Insect Pest Control
Through ecological studies, the seasonality of the major stem borer
species lie saparatella, is now fRirly well known. Insecticidel control for
this and other insect pest species is de-emphasized because of high costs
and for both human and environmental considerations. Preliminary studies
have indicated that cultural practices influence to some extent insect pest
infestation of the rice crop. Early transplanting in July allowed the crop
to escape the peak of moth emergence thereby suffering less injury, while
transplanting at close spacing (25 hills/m
2
) has been shown to minimize stem
borer incidence in the field. Top priority was therefore given to the
manipulation of cultural practices to develop a cheap but effective system
of control for the major insect pests.
21 •
Various insect pest control tactics developed through cultural practices
were tested in combination as an rPM package in the field against the farmers'
practice. This season, the rPM package comprising of transplanting 42-day-
old seedlings at 25 hills/m
2
in July, reduced stemboerincidence by 35% and
boosted grain yield by 16% when top dressed, with 80 kg N/ha urea over the
farmers' practice. Without fertilization, there was no significant difference
between the rPM package and the farmers' practice. Generally, crop
management by fertilization enhanced better crop performance in tolerating
inse.ct pest attack as also evidenced in the crop loss trial.
CRABS
Crabs continue to be serious pests of newly transplanted rice in most
mangrove swamp rice environments. Also, they are noted to burrow
through dikes and bunds and cause leakage of s aline water into empoldered
rice especially in Guinea-Bissau. Out of nine species presently recorded in
this environment, Sesarma huzardi, and Sarmatium curvatum are
partioularly destructive.. Sesarma huzardi is the most prevalent and voracious
species.
Crab Damage and Grain Yield Loss
Crab damage in tidal mangrove swamp rice fields is often heavier near
bunds and at the edges of rivers and creeks that drain these fields. At low
tides, crab intensities are higher at these sites. Weak seedlings (seedlings
raised on marginal soils and diseas e-infected seedlings) are most susceptible
to crab damage •
This season, preliminary studies undertaken to assess the impact of
huzardi intensity and damage on grain yield of ROK 5 showed increasing grain
yield loss with increasing levels of crab intensity/m
2
(Figure 3). Data
obtained indicated that as much as 34% loss in grain yield could be expected
from a 40% loss of crop stand due to crab damage.
Control and Management of Crabs
Tidal mangrove swamp rice farmers address severe crab problems in their
fields by transplanting very old seedlings (8 weeks old and over),
re-transplant devastated fields and in some instances deliberately delay
transplanting after uprooting rice seedlings from their upland rica nurseries.
These practices may not coincide with optimum agronomic requirements for the
rice crop. Re-transplanting one or two times in the season places undue
stress on the farmers' meagre resources.
f
-
6
0

4
0

2
0



p
e
r
c
e
n
t

p
e
r

M





o

D
a
m
a
g
e

G
r
a
i
n

Y
i
e
l
d

L
o
s
s

r
-
-
-







:
;
:
;
















g
-
J




















..































d
h

:
:
.
















!
-
.
t
;
;

















-
f
-
.
\




.• :





)<
'.,





:
:
;
;
:
\

.
.

,


















j














1
,
.
'


















'
;
0
'


























'"

.;
,
-
.







?
,....,.
'
l
.
"
"

.
.

:
.






':...,$
.,
..

t"
':
';

:
;
;
-
:

...:







.t
;
·,

t
"
.... :

'. ... :
'
!



0

y






,
:
0
/

1
1
-
::;'








:
.
.
\.











j













0














{
.
;
.
:





{
¥
"

I
·
J
·

.
•

'"
o
J
:
-
i
:
t

i
.
.
,
l
,






:

.
•
:

.
..
... '












u
..:..'
.

'Y
':
?

,
,
'.
r
.
;
'






















..





L
!
':
1









1




I




"
.--
.'











o

7

1
U

2
1

2
8

"
.

.

2

G
r
a
b
"
J
:
n
t
e
n
s
:
J
:
i
;
Y
-
'p
e
r
-
M






.

3
.

'
T
n
r
o
a
c
-
t
_
o
L






.
i
:
r
.
:
:
0
s
t
a
t
i
o
n

(
S
e
s
a
m
a

h
u
z
a
r
d
i
)

p
e
:
!
:
"
M
2

O
!
C

R
O
K

5






c
:
!
:
'
o
j

.
;
,
.

'
1
o
k
'
J
:
p
r

1
9
8
7
.

22.
The identification and determination of the role played by natural
enemies in regulating crab populations in tidal swamp rice fields are yet to
be determined. Meanwhile control of crabs making use of cultural practices
is given priority especially in the development of an IPM strategy •.
Prototype IPM for Crab Control: Earlier studies indicated that vigorous
seedlings and rice crop transplanted at close spacing (25 hills/m
2
) were
less attacked by crabs. These findings were incorporated into a single
control package (IPM package) this season and tested against the farmers'
practice in an area prone to severe crab problems. Farmers' practice involves
transplanting up to 32 seedlings of old and weak seedlings/hill at wide
spacing (12-16 hills/m
2
). Crab damage was significantly reduced in the
prototype IPM package and boosted yield by 40% over the farmers' practice at
one site prone to severe crab attack. In addition to minimizing crab damage
to the rice crop, the IPM package also imposes a saving in seedling coverage/m
2
compared with the farmers' practice.
23.
O N - F ~ i AND ADAPTIVE TRIALS
In order to ensure adaptability across environments, economic viability
and acceptability of improved technologies, trials continued to be carried out
across the sub-region especially for programs in varietal 'improvement and soil
and crop management.
EVOLVING NEVI VARIETIES
The variety improvement program tested improved varieties in collaboration
with NARS to select materials best suited to their needs.
MULTILOCATION TRIALS
FFTs: Three sets of six varieties each were tested in IDultilocation trials on
farmers' fields during 1987. The varieties were selected from previous
statiort trials that had shown potential for high grain yield.
The results of the medium duration FFTs grown at eight sites in the short
season swamps in Sierra Leone and Guinea are summarized in Table 13. WAR 1
significantly out yielded all test varieties except the new salt tolerant line,
WAR77-3-2-2. BG 400-1 which had performed well in station trials was also
out yielded by WAR77-3-2-2. BG 400-1 matured slightly later than the other
varieties, and this led to greater salt damage at saline sites. This resulted
in increased spikelet sterility compared to the earlier maturing varieties.
In two sets of long duration FFTs, no significant differences in yield
were found between Maung Nyo, Gbassin, Raden Jawa, WAR44-50-4-1, ROK 10 and
CP 4 grown at five sites, and between Rohyb6-WAR-6-2-B-2, WAR44-5-1-3,
WAR44-17-5-3, IR23429-R-WAR •. 1 and Kuatik Kundur al so grown at five sites in
Sierra Leone and Guinea.
Table 13. Farmers' Field Trials of medium duration varieties
in Sierra Leone and Guinea - 1987 season.
Variety Nean yield
( tons/ha)
WAR
3.3 a
WAR 77-3-2-2 3.1 ab
Rohyb 4-1rlAR-1-3-B-2 2.7 b
ROK
5
2.0 ~
WAR 52-384-3-2 1.7 c
ROK 5 1.7 c
Means followed by the same letter in the column are not significantly
different at 5% DMRT.
C.V. (%) 17.8
24.
To further evaluate and distribute germplasm to national Scientists,
two sets I'If AVTs each 14 test varieties were sent to 10
locations: Sierra Leone (2), Guinea (1), The Gambia (2), Guinea-Bissau (1),
Nigeria (2) and Senegal (2). Due to logistic problems, the trials in Guinea,
and The Gambia were not conducted and feedback on the trials in Senegal B;!ld
Nigeria is still pending.
In the medium duration AVT conducted at one location each in Sierra Leone
and Guinea-Bissau, average grain yields were 3.3 tons/ha at Rokupr and 3.7
tons/ha at Carboaanque in southern Guinea-Bissau. On average WAR 1,
Rohyb4-WAR-1-3-B-2, RD 15 and WAR 77-3-2-2 were the four top varieties, with
grain yields of 4.2, 4.0, 4.0 and 3,9 tons/ha, respectivelY, (Table 14). Days
to maturity of the test entries over the two locations ranged from 130 days in
WAR 49-14-2-1 to 156 days in Rohyb 1-1.
Table 14. Grain yield of the highest yielding entries in the region-wide
nedium duration advanced variety trials conducted in 1987.
Grain Yield (tons/ha) Mean
Variety
Rokupr Carboxangue
(Sierra Leone)(Guinea-Bissau)
WAR 1
4.2 4.2 4.2
Rohyb4-WAR-1-3-B-2 3.6 4.3 4.0
RD 15 3.6 4.3 4.0
WAR 77-3-2-2 3.6 4.0 3.8
Grand Mean
3.3 3.7
C.V. (%)
14.6 13.74
L.S.D. (0.05) 0.8 1.0
In the long duration li.VT conducted in Sierra Leone average grain yield was
3.6 tons/ha. The three top varieties were SL 22-617, WAR 44-50-4-1 and
Rohyb6-WAR-6-2-B-2 with grain yields of 4 .2, 4.1 tons/ha, respectively.
Rohyb6-WAR-6-2-B-2 had ranked among the top five varieties for four consecutive
seasons and was officially released in Sierra Leone in March 1988 and renamed
ROK 21.
25.
SOIL itND CROP MANAGEMENT
In collaboration with other institutions and farmers, promising new
technologies on fertilizer and Azolla identified on the Rokupr Station farm
were tested extensively on farmers' fields in Sierra Leone and Guinea.
Improving Nitrogen Use Efficiency on Farmers' Fields: Nitrogen is a major
limiting nutrient for rice cultivated in mangrove swamps. However, movement
of vast volumes of water under tidal influence and leaching make efficient
use of nitrogen in mangrove swamps difficult to accomplish. Several
approaches for increasing nitrogen use efficiency in mangrove swamps,
including split broadcast, point application and use of slow release forms of
nitrogen have been investigated.
Research undertaken at WARDA demonstrated that the method of application
is the most important factor influencing rice response to nitrogen in mangrove
swamps. Placement of 60 kg N/ha as an aqueous solution of urea at 20.ill soil
depth by the injection technique developed at vlilRDA - Rokupr showed a
potential for increasing yield of rice by a margin of 54-143% over the control
without nitrogen, as against increases of 17-70% when the fertilizer was
broadcast in a single dose. The injection technique had the advantage over
similarly effective three split broadcast procedures for nitrogen application,
in requiring only one application of nitrogen at the early phase
of crop growth. This is well suited to the farming practice in mangrove SWaBpS
as farmers are relunctant to make more than one visit to their farms between
transplanting and harvest.
However, appropriate and easily adaptable technique for nitrogen
application in the mangrove swamp still needs to be developed. In 1987,
several methods of nitrogen fertilizer (60 kg/ha) involving
broadcast and incorporation of prilled ure3. CPU) before transplanting;
broadcast of PU at 4 weeks after transplanting (1;vAT); and soil incorporation
(Point application) of urea supergramile (USG) at 4 WAT were evaluated for
grain yield responses under farmers' field conditions on the Scarcies.
Soil incorporation of USG resulted in the highest grain yield response
with yield increases of 1.5 tons/ha over the farmers' practice (Without
nitrogen) with a yield of 1.6 tons/ha of paddy (Figure 4). This increase
represented more in grain yield over basal incorporation of PU, which was
the most promiSing among the other methods of nitrogen application. Basal
incorporation of PU and broadcast PU at early tillering increased yield by
64% and 55%, respectively over the farmers' practice.
4
3
2
1
o
Grain Yield
tons/ha
cv % =
c
z
Fig. 4.'
a
b
(53)
~
_L
•
n·:
H
fj
0
()
s::
+->
H
m
r-l
8 ctS
U) (/)
t) ctS
<
(l:\ (l:\
p
P
Pi Pi
of N (60 Kg/ha) applocation
Rice response to methods of N fertilizer
application on farmers' fields - ;-!, c,'lY.'cicr; 19B?
.; .
( ) Percentage increase over cont rol
26.
To date, soil incorporation is the most effective means for increasing
nitrogen use efficiency in mangrove swamps. However, soil incorporation of
USG at 10-15cm by hand ' seemed a more arduous task than deep placement by the
injection technique. The potential of USG for increasing yield of rice under
the traditional broadcast application and basal incorporation procedures will
be evaluated in future trials.
On-farm Evaluation of Azolla: Trials were established on farmers' fiel ds in
inland and associated mangrove swamps in Sierra Leone and Guinea to assess the
potential of Azolla for rice cultivation.
Uncontrolled flooding, iron toxicity, phosphorus deficiency, and abscence
of adaptable strains of Azolla have been t he major constraints to utilization
of Azolla under farmers' conditions in humid tropical environments.
The Azolla strain ADUL 138 PP was identified as being widely adaptable to
these conditions in preliminary adaptation trials. In Sierra Leone, a biomass
of 1.71 kg FW/m
2
, on average was obtained with cultivation of ADUL 138 PP
under farmers' conditions.
In 1987, cultivation of Azolla resulted in significant improvements in
grain yield of rice, producing similar yields of rice as the full complement
of mineral fertilizers (Table 15). Part of the response observed with Azolla
cultivation may be attributed to its effect on weed growth. Effective soil
incorporation of Azolla prior to transplanting was difficult to achieve on
farmers' fields. Allowing the unincorporated Azolla to intercrop with rice
resulted in the suppression of weed growth and probably accounted for the large
proportion of the grain yield response to Azolla cultivation.
Table 15. Grain yield and weed infestation of rice cultivated with Azolla
and mineral fertilizers on farmers' fields in Sierra Leone -
1987 season.
Treatlilenta
Control - without fertilizers
Mineral fertilizer N60P40K40
Azolla cultivation
Hean grain
yield (t/ha)
1.3 b
2.2 a
2.1 a
Average*
dry weed
weiqht
(gLm2)
411 .1
602.3
209.0
Heans followed by the same letter in the column are not
significantly different at 5% DMRT.
* Average over 7 sites
27.
The reduction of weed infestation in swamps with Azolla suggested that
labour time for weeding can be reduced considerably, especially in the
absence of an appropriate weed control technology. However, the difficulties
with water control on farmers' fields suggested that Azolla technology may be
appropriate only for swamps witll adequate water control systems.
22 ",
Adaptive Trials
Three components of improved technology packages involving (i) improved
varieties such as ROK 5, ROK 10, CP 4, Kuatik Kundur, \Hill 1,
(ii) application of '40 kg N/ha in the form of urea and (iii) mechanical land
preparation by using the single axle power tiller have been identified for
farmer managed trials in Sierra Leone, The Gambia and Guinea. Over the past
fi ve seasons the packages have been found to be successful under farme'r
management and are being adopted by farmers (Table 16). Improved varieties
that fit well in the farming system of most areas are being identified.
Table 16. Percent adoption of i mproved varieties by farmers in
the mangrove swamp rice environment of WARDA, 1978 to 1987.
I mproved variety
Sierra Leone Guinea
Gr. ScarciesNoribaya Satlu (Coyah,
Sonfonia)
Kamsar
Guinea- The
Bissau Gambia
ItOK 5
T.JTAR
]ohyb6-WAR-6-2-B-2
Rohyb4-1,IT AR-1 -3-B-2
Br 5191-6
Kuatik Kundur
ICuatik Hirang
Raden Mas
ROK 10
CP 4
50
*
17
35
90
*
* New adoption (cultivated by a few farmers)
+ Handled by seed Dultiplication unit.
30
*
*
10
+
*+
*+
*
10
10
(Carboxque)
60
*+
*+
*+
*+
(South)
Bank
+15-20
*
*
+
In 1987, packages involving i mproved varieties and fertilizers were
placed in tarmer managed trials at 8 locations each in tidal mangrove and
associated swamps in Sierra Leone, Guinea and The Gambia. In tidal mangrove
swamps, the package comprised WAR 1 and Rohyb6-WAR-6-2-B-2 in short and long
season areas, respectively plus top dressing of 60 kg N/ha of urea at early
tillering. The package in the associated swamps include 40 kg P
2
0S/ha as
single superphosphate in addition to urea and Rohyb6-WAR-6-2-B-2 or KU,atik
Kundur. Data obtained are summarized in Table 17.
29.
Table 17. Performance of mangrove swamp rice technology packages,
under farmer management in Sierra Leone and Guinea, 1987.
Country and
Zone
Sierra Leone
Balancera
Kychum
Mambolo
Guinea
Coyah
Yattaya/Sonfonia
Xattaya/Sonfonia
Grain yield (t!ha)
Ecology New Farmers' Yield Additional Net benefit
TMS
TMS
AMS
AMS
TMS
AMS
Technology Practice Gap input cost due to new
(tZha) technology
2.957

7."96
2,917
2.426
3.381
2.557*
2.206*
1 .701
2.027
1.850
2.456**
0.400
0.667
1.595
0.890
0.606
0.925
Leones Leones
822.30 2,Cf57.70
822.30 3,980.10
2,9
1
7.60 9,842.40
QE QE
55,849 10,901
21,422 24.028
49,628 19,747
TMS = Tidal Mangrove Swamp
AMS = Non-tidal Mangrove Associated Swamp
New Technology - TMS = Improved variety + Urea (60 kg N/ha)
AMS = Improved variety + Urea (60 kg N/ha) and SSP
(40 kg P 205/ha).
Farmers' Practice - Local variety only.
(* Adoption of improved varieties by large number of famers)
(** Local variety with better adaptation to soil stresses)
-1
Labour cost day Sierra Leone = UQ5.30
Guinea: Coyah, GF 687; Yattaya, = GF 493.8
t
-1
Fer ilzer cost kg
-1
Paddy Price kg

Sierra Leone
Balancera
Kychum
Mambolo
Guinea
Coyah
NPK 1 5 :1 5 : 1 5
1e3.80
3.80
3.20
GF145
Yattaya/Sonfonia GF132.5
Sierra Leone Leones
Balancera 7.20
Kychum 7.20
Mambolo 8.00
LE4.00
4.00
3.90
GF120
GF120
Guinea
Coyah
Yattaya/
sonfonia
UQ.80
2.80
2.80
GF125.0
GF125.0
QE
75
75
30.
Sierra Leone: In the tidal mangrove swamps. of Balancera (short season zone)
and Kychum (medium season zone) along the great Scarcies, most of the
co-operating farmers have adopted improved rice varieties. This accounted
for the comparatively high yields obtained even with the farmers' practice
in these areas. The yield gap between the farmers' practice and the new
technology ranged from 0.40 to 0.67 tons/ha. In the associated mangrove swamp
in the Mambolo area (medium season zone) Kuatik Kundur out yielded the farmers'
practice by a yield gap of 1.6 tons/ha. Consequently, higher returns were
realized from the improved package with net benefits in the range of U2000 -
1.e10,000/ha.
This is the third season lillli 1 has performed well in adaptive trials in
the Balancera zone and would seem well suited for this area. In the Kychum
and Mambolo area, Kuatik Kundur may be promising material.
Demonstrations: The package of ROK 10 plus urea (60 kg N/ha) placed in long
duration areas, out yielded farmers' practice by 21.2 and on the Scarcies
and along the BUL1peh in northwest and southwestern regions of Sierra Leone,
respectively.
Guinea: 1 and Rohyb6-WAR-6-2-B-2 performed well both in the tidal and
associated mangrove swamps in the Yattaya area with a yield gap range of 0.61 to
0.93 tons/ha. The improved package carried an additional input cost of
GF 21,422 to GF49,628 but brought in a net benefit of GF19,747 to GF24,028/ha.
In the Coyah zone, the improved package comprising of Rohyb6-WAR-6-2-D-2
yielded significantly higher than the farmers' practice but had a lower net
benefit GF10,901. In the Coyah zone, input costs were higher than in the
Yattaya/Sonfonia zone.
Seed Multiplication and Distribution
To assist in the availability and spread of improved varieties, the
station has been multiplying and distributing small quantities of seeds of
improved varieties/advanced lines to scientists and extension workers in the
region. Since 1985, a total of 1.1 tons of breeder seeds of 54 recommended
and promising lines; and 1.0 ton of improved seeds of seven varieties
(ROK 10, CP 4, Kuatik Kundur, ADNY 301, Rohyb6-VIAR-6-2-B-2, ROK 5, and WAR 1)
have been produced.
In collaboration with the Northwestern Integrated Rural Development
Projwct (NWIADP) in Sierra Leone, 15 tons ' varieties (ROK 5, ADNY 301
and ROK 10) are being made available for d_ .oution to over three hundred
mangrove swamp rice farmers in Northwest and Southwest Sierra Leone in 1988.
The seeds will be recovered at harvest for further distribution to more
farmers in 1989.
31.
INTERNATIONAL AND NATIONAL CO-OPERATION
The station worked in close partnership with some international
agricultural research and developed strong linkages and cooperation
agricultural research systems (NARS) to help them utilize
improved rice technologies in the \,lARDA sub-region.
A. Cooperation with International Institutions
Since its inception, this station has collaborated actively with IRRI
and has tested over 6,000 lines from several IRTP nurseries. Several of
these lines have shown promise on the basis of soil stress tolerance, plant
type, grain yield and seed characteristics. In 1985, the station made
available to IRRI ROK 5, ROK 10, and CP 4 for somaclonal varietal studies and
produced 17 variants. These are currently being tested for tolerance to
various and biological stresses and adaptation to the nangrove
rice conditions in the sub-region. This station also collaborate:d::
with IITA in a netwDrk of varietal screening activity for iron toxicity in the
1-JARDA region.
In collaboration with ODNRI the ·female sex pheromone of li. seaparatella
the predominant stem borer pest has for the first time been identified and
synthesized. The pheromone was deployed for region-wide monitoring of
Maliarpha in the sub-region and elsewhere.
B. Cooperation with National Institutions
Member states of the region differ in their needs and capabilities for
research. We tlS·OO c);i;a ted; ourselved with NARS to respond to the precise
needs of any particular country. This partnership has facilitated the
rapid spread of improved rice varieties among mangrove swamp rice farmers in
the region. In Sierra Leone, we have developed a close and strong working
relationship between our station and the National Rice Research Station (RRS).
vi e co-operate with RRS in planning research activity, exchanging and testing
promiSing varieties. A breeding program on resistance to RYMV has been
initiated in collaboration with RRS.
''Te held joint monthly seminars with RRS uhere Scientists alternated in
making presentations. Our collaborative effort strengthened field days and
training courses for field assistant and extension workers organized by RRS.
In 1987, our station participated fully in the annual field-day activities
organized by RRS at Rokupr, and displayed charts and photographs of relevant
research findings.
32.
We have collaborated with other agricultural institutions like the seed
mul tiplication project, Bo-Pujehun Rural Develo"pment Project and t !le Bayande
Snall Farm Development Project in supplying promising varieties such as WAR 1,
Rohyb6-\vAR-6-2-B-2 and Kuatik Kundur or gi ven assistance to these Projects in
the identification, selection, and testing of Azolla as a bio-fertilizer.
Both our station and NWIADP have jointly planned seasonal field activity and
carried out on-farm demonstrations of i TIproved technology in northwest
Sierra Leone.
In Guinea, partnership with the Ministry of Agriculture has led to the
secondment of their field staff to enhancG opportunity for continuous and
effective participation of on-going collaborative activity. Through
researcher managed pre-extension and adaptive trials conducted at multiloca-
tions "TAR 1, Rohyb6-vlAR-6-2-B-2, Kuatik Kundur, Kuda Hirang and Raden Nas have
Gained acceptance by mangrove swamp rice farmers in the Sonfonia and Coyah
regions in preference to their traditional varieties. Over 5 ha in Coyah
region alone was under multiplication of these varieties in 1987.
In The Gambia, like in Guinea, we plan with the Ninistry of Agriculture
and provided logistic support to field staff to supervise our collaborative
activities. "In cooperation with the Freedom from Hunger Campaign (FFHC)
Project and the Ministry of Agriculture, several adaptive trials of improved
packages were carried out at several sites. By this,ROK 5 ga. ined popularity
and is now" widely cultivated in t ] ~ e country. Three other varieties, VliLT-?. 1,
Rohyb6-'IlAR-6-2-B-2 and Kuatik Kundur have been identified by the Ninistry of
Agriculture for demonstration in 1988 and for further multi plication.
Over 400 WARDA varieties have been tested in Guinea-Bissau in collaboration
wi th the Ninistry of Agriculture and ILural Development (DEPA). Of these, viAll. 1,
Rohyb6-WAR-6-2-B-2, Rohyb4-vJAR-1-3-B-2 and BR5191-6 have been considered for
release. ROK 5, the most widely grown variety was made popular by WARDA
through its coordinated variety trials in the early 1980's.
The research team at Djibelor in t he CaS8LlanCe (southern Senegal) has
made some selections from the breeding populations supplied them by. our
station since 1983.
In Nigeria, a procram has been des it",'ned to develop collaboration 'vi th the
National Cereals Research Inst j.tute in testing and eval uating Ilromising
varieties/advanced lines in regional advanced variety trials.
Within the next two cropping seasons, this station will "work in closer
partnership vTi th NARS in identifying, planning and carrying our research
strategies in areas of concern 1tT i th t he aim of strene,-thening national
research capabilities.
33.
:.r,RA.1
N 1 U G
WARDA - Rokupr oontinued t o strengthen t he man-power resources of NARS in
the areas of rice research, development and production. This was
accomplished through various trai ning activities in rice research and
production methodologies ineluding training of young scientists
and technicians. Exchnnge visits with Nlu'tS staff .. Tas also an integral part of
the Station's training activities.
Students enrolled at accredited uni versi ties \,li thin
and outside the region conducted t heir research at t he Station under the
guidance and supervision of station scientists. The selected areas of
researoh were within the scope of the Station's research mandate, and served
to train the students and young scientists in rice research methodologies.
As at 1987, a total of 34 students and scientists have conducted their
researoh at the Station as shown in Table 18.
In 1987 tl'lo B.Sc. students from Njala University worked on the following
dessertat10ns under collaborative supervision of Station Scientists and
University lecturers.
(i) Potential of Azolla s pecies as source of nitrogen for rice.
(1i) Impact of crab damage on grain yield of rice.
Table 18. Students supervised and/or supported by WARDA - Rokupr
for degree-related progra ,. 1977 - 1987.

Country University B.Sc. M.Sc. Ph.D.
Total
.
--- -- ---
Britain Reading
4 4
West GerLlany Justus/Liebig
Cameroon Sierra Leone
Sierra Leone Birmingharu 1
Sierra Leone Sierra Leone 24 2 27
------ - - -
._- _.-
TOTAL
25 7
2 34
34.
1987 PUBLICATIONS
1. Agyen-Sarnpong,H. (1987). Assessment of on-far.n losGes in rice due
to insect pests. Paper presented the Regional Study Workshop
on On-farm and Post-harvest Losses of Cereal Crops in Africa
due to p'ests and Diseases. ICIPE, Mbita Point, Kenya,
October 11-15, 1987.
2. Agyen-Sarnpong,M. and Fannah, SoJ. (1987). Seasonality of the vlhite
borer, Maliarpha separatella Rag. (Lepidoptera: Pyralidae)
in mangrove swamp rice ecology of northwest Sierra Leone.
\JARDA Technical Newsletter 1 (1): 4-6.
30 Akinsola,E.Ao and Agyen-Srunpong,M. (1987). Irriga ted rice pest
control in v/est Africa.Paper presented c.t the International
Rice Research Conference, Hangzahou, China, September 21-25,
1987.
40 Bangura,FoJo, Jones,M.P. and Bernard,H.M. (1987). Deep flooded
rice growing areas near Tormabum, Sierra Leone. Deep Water
Rice E : 3-6.
Bernard,H.M. (1987). The biogology and control of Paspalum
.. . -e .
vaginatum (Swo) a weed destructive to mangrove swamp rice in
Sierra Leone.. H.Sc. Thesis, University of Sierra Leone. 98po
60 Cork,Aoand Agyen-Sarnpong,Mo (1987)0 Sex pheromone of the white
rice stemborer, !ialiarpha separatella. Paper presented at
the 4th Annual Heeting of the International Society of
Chemical Ecology. University of Hull, Englru1d, July 13-17,
1987.
7. Dixon,C.A. and Agyen-Sampong,M. (1987). On-farm research at WARDA-
Rokupr. Paper presented at the VJorkshop on On-farm Research
in Sierra Leone. Njala University College, Sierra Leone,
November 23-25, 1987.
8. Fannah,SoJo (1987). Elophila sp. ? africalis Hwnpson (Lepidoptera:
Pyralidae): A new pest of Azolla in Sierra Leone. International
Rice Research Newsletter (3): 300
9. Fomba,SoN. (1987). Promising fungicides for the control of seertling
blast, Pyricularia oryzae Cav. in upland rice nursery in
Sierra Leone. VJARDA Technical Newsletter 7 (1): 14-15.
10. FOinba,S.N. (1987). The ecological role of
(Ito et Kuribayashi) Drechsler ex Dastur in reducing rice yielrts
in mangrove swamps of Sierra Leone. Ph.D .. 'l'hesis, University
of Sierra Leone. 235p.
11. Jones,MoP. (1987). Nain and ratoon rice crop performance in mangrove
swamps. International Rice Research Newsletter (2) : 11-12.
12. Jones,M.P. and Agyen-Sampong,M. (1987). The current status of varietal
improvement in mangrove swamp rice. Paper presented at the
vlorkshop on IRTP-Africa, st. Louis, Senegal, October 14-16,1987"
13. Jones,M.Po (1987). Screening mangrove swamp rice varieties for salinity
tolerance. Paper presented at the vJorkshop on IRTP-Africa, St.
Louis, Senegal, October 14-16, 1987.
36.
PERSONNEL
R ESEATICH STAFF Po.SITION
H. Agyen-Sampong Ph.D. Entomolo[::i ,s t/Bta hon Director
N. P. Jones Ph.D. Associa t e Br ·3eder
s. N. Fomba Ph.D.
C. Ao Dixon HoSe. Ass ocia t e E>o:i.l Scientist
s. J. Fa,nnah H.Sc.
Associate EntoL1clogist
H. H. Bernard MoSe. Associate \Teed
SUPPORTING ;)Tt'\FF
..... _' ... _----'.-------
s . Kamara B .. Sc.. (Hons.) Admin. As;3t./Fin. Officer
M. S. Sufi A.li HND. Supervisor
26 Field Assi.stant/Clerks/Drivers
60 La bourers/l1'a tchmen/N e ssengers