Citation
Annual report, International Institute of Tropical Agriculture annual report

Material Information

Title:
Annual report, International Institute of Tropical Agriculture annual report
Spine title:
International Institute of Tropical Agriculture annual report
Creator:
International Institute of Tropical Agriculture
Place of Publication:
Ibadan Nigeria
Publisher:
The Institute
Frequency:
Annual
regular
Language:
English
Physical Description:
8 v. : ill. ; 23-28 cm.

Subjects

Subjects / Keywords:
Agriculture -- Periodicals -- Tropics ( lcsh )
Tropical crops -- Periodicals ( lcsh )
Genre:
serial ( sobekcm )
international intergovernmental publication ( marcgt )

Notes

Dates or Sequential Designation:
Began in 1974; ceased in 1981.
General Note:
Description based on: 1977; title from cover.
Funding:
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
International Institute of Tropical Agriculture.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services (UFDC@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
002203897 ( ALEPH )
03806543 ( OCLC )
ALE3829 ( NOTIS )
82642611 //r85 ( LCCN )

Related Items

Preceded by:
Report
Preceded by:
Report
Preceded by:
Report
Preceded by:
Report
Succeeded by:
IITA annual report

Downloads

This item has the following downloads:


Full Text
-dli
rr
- 71W,
hi:M
'113
-~ ~l........
- I
~ ~I, <,
Zip ~




7,
E,
Ow
Aboutil
het lnternationa4iiistituit 4.6t -".k.. Tropicai*Agricult6j6* (11.rXg.v
-orWJUly#
T was established 7!ft'
19677 as& arwautonomouo, nc P rofiftcorporationsbyrattdecreeoo . . . . .
thft Federalt Militaryw GoVernment 0f NigeriaLit'formalikiorganized, 61.""4 v
thet first meeting* of B.oards"obA Trustees ins lbadarw duHngv ujyj k 1968tV Thm Federal; Republicto geriaoprovide&11, 000thectarest of lanchfootheti[TA site!randtthciiFord" Foundation- the initial capita buildingstandidevelopmenfl- ,,,
Oak,
Support fooresearchtandtdaywt6 w dayr opera tionsi in, 1980, m the* For& Foundationrs. the!,, Canatv _,dianit international DevelopmentK. Agenc)w(CIDA),,.thetiOversea4pe, velopmentzMinistryw oft thet United,&., Kingdorrr. (ODM); thetUiS. Y#
1 Agefic
for* Internationak'. D V iopmenW.,-, (USAID)" the.' WorldfBankp international Funds, for- Agriculftirallil" DevelopmenV(IFAD) '-an OthetgovV1.1 a
ernmentsr oV AustraliamwtBeigium Japan;; The Netherlandig,
nk*
Norway .and.the, FederahRepublic, 7
of Germanyi
The- institute& is. governed bjvan, in- ttemational.Board. otTrustees;.,-.,..,.
. . . . . .
Correct citation: International Institute of Tropical Agriculture, 1981, Annual Report for 1980. lbadan, Nigeria.
Published by the
International Institute of Tropical Agriculture
Oyo Road, PMB 5320, lbadan. Nigeria




1980
ANNUAL REPORT
INTERNATIONAL INSTITUTE OF TROPICAL AGRICULTURE







Contents
v Board of Trustees
vi Foreword
1 Farming Systems Program
1 Introduction
1 Baseline data collection and analysis
1 General weather conditions
3 Agroclimatic analysis
5 Soil and land characterization and evaluation
7 Socio-economic analysis
10 Land and soil management
10 Land clearing and development
11 Tillage systems and small tools development
21 Management of kaolintic Alfisols 28 Management of siliceous Ultisols
32 Cropping systems
32 Intercropping agronomy and meso/micro-climatic studies
36 Alley cropping
i 42 Live mulch systems
43 Weed biology and herbicide research
44 Plantain improvement
48 Eastern Cameroon Farming Systems Project
49 Small farms systems research (Atebubu District, Ghana)
51 Root and Tuber Improvement Program
5.1 Introduction
51 Cassava
51 Genetic improvement
54 Entomology
59 Pathology
61 Nematology
62 Tissue culture and virus indexing
62 Zaire's National Manioc Program (PRONAM)
64 Cameroon National Root Crops Improvement Program (CNRCIP)
64 Yam
65 Genetic improvement
66 Pathology
70 Nematology
72 Quality evaluation
72 Tissue culture and virus indexing
72 Cameroon National Root Crops Improvement Program (CNRCIP)
72 Sweet Potato
73 Genetic improvement
74 Entomology 76 Nematology
77 Quality evaluation
78 Tissue culture and virus indexing
80 Cameroon National Root Crops Improvement Program (CNRCIP)
83 Cocoyam
83 Genetic improvement
83 Agricultural economics
84 Cameroon National Root Crops Improvement Program (CNRCIP)
III




85 Cereal Improvement Program
85 Introduction
85 Maize
86 Genetic improvement 95 Entomology 96 Semi-Arid Food Grains Research and Development Project (SAFGRAD)
100 Rice
100 Genetic improvement 109 Pathology 109 Physiology
117 Grain Legume Improvement Program
117 Introduction
117 Cowpea 117 Genetic improvement 124 Entomology
- 130 Upper Volta Food Legumes Program 131 Semi-Arid Food Grains Research and Development Project (SAFGRAD) 134 Tanzania Food Crops Research 137 Brazil Cowpea Program
137 Soybean 137 Genetic improvement 143 Entomology 149 Agronomy 155 Tanzania Food Crops Research
155 Microbiology
161 Training Program
163 Research Support 163 Introduction 163 High-rainfall substation 163 Research station operations 163 Screenhouses and the plant growth facilities 163 Genetic Resources Unit 166 Virology Unit 171 Analytical Services Laboratory 171 Biometrical Unit 171 Conference Center 171 Library and Documentation Center 172 Communications and Information Office
173 Special Projects 173 Introduction 174 Training and collaborative research projects 174 Core supplementing projects
175 Personnel
177 Collaborators and Training
180 Publications
182 Conference and Seminar Papers
185 Major Seminars
IV




Board of Trustees
Mr. Guy Vallaeys (Chairman) Dr. Willi Ehman
Institute de Recherches Agronomiques Tropicales et de Ministry of Economic Cooperation, 53 Bonn 12,
Cultures Vivrieres (IRAT), Paris, France Germany
Mr. D. lyamabo (Vice Chairman) Prof. R. J. Olembo
Ministry of Science and Technology, Ibadan, Nigeria Div. of Ecosystems and Natural Resources, UNEP, Nairobi, Kenya
Dr. R. C. McGinnis
Dean of Agricultute, University of Manitoba, Winnipeg, Citoyen Mukendi Mbuyi Tshingoma
Canada FAO, Bamako, Mali
Dr. Hidetsugu Ishikura Dr. F. F. Hill
Director General, Institute of Environmental Toxicology, Ford Foundation, Ithaca, New York 14850
Tokyo, Japan
Dr. John J. McKelvey, Jr.
Alhaji M. Liman Associate Director, Agricultural Sciences, The
Federal Ministry of Agriculture, Lagos, Nigeria Rockefeller Foundation, New York, N.Y. 10036
Dr. E. H. Hartmans Prof. C. A. Onwumechili
Director General, IITA, Ibadan, Nigeria Vice Chancellor, University of Ife, Ile-lfe, Nigeria
V




Foreword
T he year 1980 was one of financial uncer- Farming systems research was reorganized tainty for the International Institute of to focus on three major components-land
Tropical Agriculture. A high rate of infla- clearance and development, post-clearance tion, a doubling of the minimum wage, and a management, and cropping systems. The strengthening of the Nigerian naira against maize improvement program included some the American dollar had a very serious and exploratory work on the development of hynegative impact on the real research capac- brids. The development of lowland rice vari
- ity of IITA.i. eties and the problems of production in
Research program budgets and staff had to central and eastern Africa were given greater be reduced, and a general belt tightening left emphasis. Breeding for resistance to insect theremaining scientists with insufficient sup- pests of cowpeas was intensified, and casporting staff and operating allocations to en- sava research was concentrated on, host ab~them to work in a fully effective and plant resistance to the green spider mite and productive manner. mealybug.
It was through a very generous special con- In the resolution of the etiology of virus distribution of 1 million naira, equivalent to 1.9 eases, techniques of indexing the presence million U.S.- dollars, by the Federal Republic of sweet potato virus was. developed.. As a of Nigeria in December 1980 that this precar- result, it is now possible to export virus-free ious financial situation could be redressed. sweet potato clones in tissue, culture form Despite.. the disruptive economic condi- internationally.
tions, IITA's research community produced a Previously, the Institute's research was alscientific product of commendable quality most exclusively pointed toward the small, and quantity as reflected in this Annual traditional, resource-poor African farmer. But Report. because of increasing interest in commercial
As in previous years, the 1980 Annual Re- farming in many African countries, the Instiport is a compendium of salient, research tute is devoting a greater share of its reconducted in cereal, root, tuber, and grain search effort toward the solution of problems legume improvement programs, and in farm- encountered by medium and large-scale ing systems for the humid and subhumid farming enterprises. tropics. This report and its sister publication, We are sure you will find this Annual Report 1980 Research Highlights, together provide a informative, and we will be pleased to recomprehensive summary of findings for both spond to all requests for additional informathe scientist and the generalist., tion about on-going research at the Institute.
This year also marks the first time the An- A listing of journal articles and IITA's scinual Report is being published in French. entific staff is appended to this report to Long overdue, we are confident it will assist you. be enthusiastically received by our Frenchspeaking colleagues.
The structure of the Institute's research
program remained unchanged in 1980, but
the focus of the work was somewhat Dr. E. H. Hartmans
modified. Director General
VI




Farming Systems Program
Introduction
A primary goal of the Farming Systems Program is B sln aac leto
the development of methods of land use and crop B s ln aac le to
management that will enable efficient, economic and analysis and sustained production of food crops for the humid
and subhumid tropics. Research is directed primarily at Research on baseline data collection and analysis-inproblem solving the constraints of small farmers, many cludes the following areas: agroclimatology, soil and of whom still rely on bush fallow systems for producing ln hrceiainadeauto n oieooi
the bulk of the food in the humid and subhumid tropics landaratrzainansvlatoindscoeooi of Africa and elsewhere. Emphasis is given to developing anls. technologies that are scale neutral so that they can be Research in 1980 emphasized crop water requirements used by a range of farmers. as well as the development of a soil evaluation system
for highly weathered soils in West Africa and studies of
Because of the wide range of farming system forms in soil erodibility. Socioeconomic analyses included food the humid and subhumid tropics, the program will not crop and agro-forestry farming systems surveys. develop specific local blueprints for improved methods
of land, soil and crop management; rather, the program's
concern is to develop and make available preliminary 1ITA general weather conditions technologies and subsystems that can be modified and
adapted by national and regional organizations to the The nonoccurrence of the late *July-August dry season agronomic and economic as well as political conditions was the dominant feature of the weather in 1980. As with of their own areas. Considering the diversity of the farm- similar cases in 1973 and 1979, the only other 2 of the ing systems in the Institute's mandate area, baseline data decade, this feature was again associated with a general collection and analysis are undertaken to better deline- shift in rainfall pattern; the second rather than the first ate the major benchmark areas with relevant bioclimatic season constituted the major rainy season. There were and soil parameters in relation to prevailing cropping consonant departures in isolation, the heavy rainfall pesystems. The benchmark sites will be used for determin- riod generally receiving less than normal global radiaing typologies and testing principles of land and soil tion. The mean temperatures were comparatively high management and cropping systems. along with the mean relative humidity while the evapoThe eserchempasi oftheproramis n asising rative demand was lower than the long-term mean. A farmers in the move from the subsistence shifting culti- smayo h anvralsi ie nTbe1
vation and particularly bush fallow systems, to more continuous and productive systems of cultivation with Rainfall and evaporation. A streak of rainless days appropriate land and soil management practices, which extending back to November, 1979, was broken with a will maintain soil productivity and minimize soil erosion 26.2 mm rain on 14 February. There were 4 other rains and soil degradation. before the end of March (Table 1, Fig. 1), and the first
quarter of the year was, thus, marked by a pronounced
To achieve more impact in the immediate future, the pro- moisture deficit with a -44 percent departure in cumugram focused its activities in the following research lative rainfall in conjunction with a -7 percent differareas: baseline data collection and analysis, land and ence in cumulative evaporation. soil management and cropping systems.
No rain of substance fell in April, and total rainfall was
Cooperative programs have been established in Came- fully 66 percent below normal. It was a clear case of a roon and Ghana. Reports of these projects are included "false start" in the rains, causing most of the late March in cooperative projects, and individual reports are avail- plantings to succumb to drought stress (Fig. 2). Favorable from the respective organizations. able water balance was first established at the beginning




Table 1. Summary of climatic data (IITA, 1980).
Total Total Solar Temperature Relative humidity
rain- evapo- radiation 0C %
fall ration G"-cal/
Month mm mm cm2/day Min. Mean Max. Min. Mean Max.
Jan. 0 121.22 391.90 23.3 27.8 32.2 46 71 96
Feb. 26.2 150.51* 504.14 23.8 28.9 33.9 39 68 39
Mar. 52.0 174.22 513.70 24.3 28.7 33.0 46 72 98
Apr. 50.1 164.70" 490.16 23.8 28.4 33.0 49 72 95
May 193.0 126.17" 401.85 23.3 26.7 30.1 61 79 97
June 132.2 129.50 421.16 23.2 26.3 29.4 64 80 95
Jul. 203.8 88.04* 318.53 22.2 24.6 27.0 72 84 96
Aug. 367.8 92.07* 325.16 22.4 24.9 27.3 73 85 97
Sept. 248.9 104.10* 384.56 22.3 25.5 28.7 67 83 98
Oct. 208.4 107.88 418.96 22.5 26.0 29.4 64 81 97
Nov. 66.7 107.10* 408.28 22.5 26.3 30.0 56 76 97
Dec. 0 117.58 386.25 20.1 25.3 30.5 41 67 92
Walues adjusted for days with missing data.
400- IITA, 0 24"---- University of Ibadan 20-year ----o RAINFALL
380-~~ avrg x EVAPORATION
average
360- .20340 "E
S 16
320' 4I.,
3_ 0
2z 8
260" 6
240
- 2 22 JAN F'EB MAR APR MAY' 'JN JU;L' 'AUIG ~ 'C E
T IME 180 Fig. 2. Weekly mean rainfall and evaporation (IITA,
I. \ 1980).
o-\ This spell of moisture deficit was followed by a period of
/ V excessive and persistent rainfall, which culminated in
100- the highest monthly peak for the year and the third highest on record (367.8 mm for.August). A singular rain in80- / cidence on 31 August resulted in an unprecedented and
catastrophic flood in the general area.
/ With the failure of the 'normal' August break in the rains,
40- first-season crops, with potential yields already depressed, suffered further damages through lodging, im2 / proper drying and delayed harvesting. The excessive
moisture also caused considerable delay in land prepa'JAN 'FEB MAPR'MAY'JUN JUL USEP' CT NOVDEC ration and second season planting and persisted TIME through September and October. The rainy season
Fig. 1. Mean monthly rainfall, ended in early November. The cumulative total for the
year was 1,549 mm or 22 percent above the long-term mean. Pan evaporation, on the other hand, was 7 percent of May, about a full month later than normal (Fig. 2) with less than its mean, totalling 1,483.09 mm. a corresponding delay in the start of the first cropping Sky conditions and solar radiation. January and Februseason.
ary were unusually cloudy, but solar radiation (Table 1, Precipitation remained quite regular through the first 2 Fig. 3) was higher than normal. Cloudiness increased, weeks of June but tapered off again in the last third of particularly in the second half of April, in spite of the the month and continued so through the first of July. limited rainfall.
2




100
90
80
6. 00" 70.
,5 350- _660
- WEEKLY MEAN W o
<250"
--K-. MONTHLY MEAN
200- ..30.
_0__ MAXIMUM
JA FB'AI lY U UL AGSE C NO CI20 -- MEAN
T.,% IM E 10 M IN IM U M
Fig. 3. Weekly and monthly mean solar radiation (IITA, -A Y
1980). 'JJUI EII AG ''AS i OCT ' 0 i C'
Fig. 5. Mean, maximum, and minimum relative humidity During the period May through August, which in effect (IITA, 1980). covers the first cropping season, incident radiation dropped to a mean of 363.7 g-cal cm-2 day-1 or 6 percent siderable lodging and stalk breaking observed during the less than the multiannual mean for these 4 months. The period. rest of the year, except for the month of November, saw Dew/light precipitation. The contribution of dew to the a return to higher than normal insulation regimes. water budget was much less than in the previous year. Temperature and relative humidity. Higher tempera- This is not surprising in view of the record high minimum tures prevailed on the average, mainly as a result of temperatures and the high incidence of rainfall. The higher night-time temperatures, an apparent result of the measured total for the year was 11.59 mm, ranging from more humid and cloudier conditions (Fig. 4). Monthly a monthly maximum of 1.57 mm in November to 0.48 mm average minimum temperatures were record-high in in August. many cases.
Agroclimatic analysis
40.
Agroclimatic zones of West Africa
30 ... To provide a uniform framework for research to facilitate
Z the choice of priority areas and determine the range of
,W applicability of research results, an agroclimatic zona20 tion of West Africa was undertaken. It is based on the
MAXIMUM concept of water-balance and Franguin's method of in...... MEAN terception of rainfall potential evapotranspiration cure.
Months of positive water balance were defined as months with rainfall greater than or equal to potential JAIN'FES I MAR "APR IMAY 'JUN I JUL I AUG I SEP OCT I NOV I DECI evapotranspiration. On the basis of the number of the TIME months determined for the network of approximately 88
stations over the region, 7 generalized agroclimatic Fig. 4. Weekly mean, maximum, and minimum mean air zones were defined (Fig. 6). temperature (IITA, 1980). Commonly used terminologies were retained in describing these zones. It is believed that the basic definition Daytime hours were by contrast cooler. The only above- used in this classification is not only more meaningful average maximum temperature occurred in April (depar- from the cropping point of view but also avoids complex ture: +0.60CY, a month of subnormal rainfall. With the indices and is, thus, easier for the user to follow. delay in the onset of the rains, soil temperatures remained rather high well into the month of April with a mean maximum of 41.30C at 5 cm. Crop water requirements.
Observed relative moisture contents of air were also There is a scarcity of research information regarding the higher than normal, except for April (Fig. 5). These water requirements for tropical food crops. It is imporhigher humidities evidently account, in part, for the tant, therefore, to define the optimum water requirelower values of evaporative demand. ments of these crops to maximize their productivity
An unusual prevalence of southerly and southwesterly under different ecological conditions. winds in January and February constituted the basis for Evapotranspiration and cowpea yields. Many studies on the observed cloudiness noted earlier. In general, wind a variety of crops have tended to show a linear relationspeeds also remained above average throughout the ship between actual evapotranspiration and, more paryear. The incidences of gusts during the wet period, par- ticularly, actual transpiration and yield. Using cowpea, ticularly in late July and in August, account for the con- TVu 3629 and TVu 4557, yield figures from a series of
3




15W low 5W 0 5E IOE 15E
LEGEND 20N 1 -20N
( PERHUMID (P!ETp for months or more) uI
HUMID: (P2ETp for period between 6-8months)
TRANSITION HUMID/SUBHUMID: (P_ ETp for
period 5-6 months)
SUBHUMIDO. (PETp for period between 4-5 months)
SEMI-ARID:(P>ETp for period between 2-4
months) 4.5
ARID: (PETp for period between 1-2 ION + ION
months)
DESERT:(P ETp for period less thanone 7 month)
NATIONAL BOUNDARIES
ISOLINES OF NUMBER OF HUMID (P>ETp) 5N 4 + 5N
MONTHS
BOUNDARY OF BIMODAL RAINFALL/CROPP- t O7
IN0R A 5 ow 5w 0 I E8 7 ISE
ING AREA
* STATIONS
Fig. 6. Generalized agroclimatic map of West Africa. (Source: T. L. Lawson, IITA, 1979).
previous sequential planting experiments (IITA Annual Pest and disease problems are believed to be partly reReport, 1976), a preliminary attempt was made to model sponsible for this. Possible inaccuracies in the evapocowpea yield using estimated evapotranspiration. transpiration estimates may be equally important.
Evapu ranspiration estimates were obtained by assuming that the potential evapotranspiration is equal to Class A pan evaporation, and the actual mean weekly evapo- oo1600 1977PLANTINGSEASON
transpiration is proportional to the ratio of the actual VARIETY ,, IOWEEKS
available soil moisture to the available soil moisture at 400.oo. field capacity. The 1976 results show a much better relationship than the 1977 results between the estimated zoo "
ev'Apotranspiration and cowpea yields (Figs. 7 and 8).
2000 000 *
0800
1976 PLANTING SEASON x
VARIETY I, ISt -10 weeks a *- o
VARIETY2, I -I0 weekS --1600 400
1400 200 $oo
1200 20 40 60 8o 100 120 1410 160 leo 200
TOTAL EVAPOTRANSPIRATION (SET)
Z X
oo000 fig. 8. Cowpea yield vs. evapotranspiration (1977).
W
600
- Dry season evaporative demand and cassava yields.
Cassava planted in a given year normally grows through 600. / the dry season and is harvested at the beginning of the
SA rains the following year or thereafter. Although the plant
/
400- o o reduces its canopy during the intervening dry season, no
/ e sign of wilting is observed on the leaves of the remaining
/ foliage. On the hypothesis that the plants probably
200 o. achieve this only by drawing on reserve resources, it
100 ,would appear that yields might tend to decline in propor20 ..2 2' tion to the length and severity of the dry season. Using
0 20 40 60 80 100 120 140 160 180 200 220 240 260
TOTAL EVAPOTRANSPIRATION ( ET ) cumulative Class A pan evaporation (Ep) as a measure of
the intensity of the dry season, an analysis was made to Fig. 7. Cowpea yield vs. evapotranspiration (1976). relate the reported decline in the average yield of the 10
4




most promising cassava clones at IITA between 1973/ the last 2 columns refer to: 1974 and 1977/1978 (IITA Annual Report 1977). The re- Physical condition modifiers: suits, indeed, show a negative and highly significant lin- w -low available water reserve ear relationship between the cumulative pan evaporation r -high soil erosion hazard from 1 December through 28 and 29 February (the dry c -high soil compaction hazard season for IITA) and the cassava yield. The relevant equa- Chemical condition modifiers: tion is as follows: Y = 176.69-0.3421, r = -.978*, where k -low potassium reserve
is the yield of cassava and I (1 = Dec., n i -high phosphate fixation
d (Ep) (1 t -secondary/micronutrient deficiencies and/or imbalFeb. 28/29). This relationship could prove useful in esti- ances nation of the performance of cassava across climatic a -aluminum toxicity for most legume crops m -manganese toxicity to most legume crops
zones. -potential soil toxicity and/or secondary and micronutrient deficiencies and imbalance due to continuous cultivation with conventional chemical fertilSoil and land characterization and ization.
evaluation Tentative quantitative limits of the "soil condition modifiers" have been defined, and refinement of these can be introduced in the future on the basis of further research Technical soil evaluation system based information.
on soil mineralogy It is proposed to integrate this soil evaluation system in
Two distinctive approaches have evolved in recent years the land types and agroclimatic information so as to esregarding the management of tropical soils for higher tablish comprehensive guidelines for land clearing and agricultural productivity. The high-energy input, exten- management for different regions of the tropics. Close sive food and cash crop production systems are well collaboration with FAO and other development agencies suited for the fine-textured oxidic Oxisols, Alfisols, Ulti- and interested national soils research institutions will be sols and Inceptisols; whereas, agroforestry and low-en- fruitful in further development of the system. ergy input, food crop production systems are more suitable for the kaolinitic and siliceosis Alfisols, Ultisols and Inceptisols. Soil erodibility characterization
A key factor differentiating these 2 major categories of Field experiments have been established to directly monsoils and their response to agricultural exploitation is itor soil erodibility at 3 locations in Nigeria-Onne, Ikom soil mineralogy. A technical soil evaluation system using and Jos. In addition, soil detachability and transportabilmineralogical characteristics as the main criterion is ity measurements on 20 soils collected from different being developed at IITA with the primary objective to parts of Nigeria are being made with a laboratory rainfall provide agricultural planners in the tropics with a set of simulator. The accumulative infiltration measured for 3 simple guidelines for agricultural soil utilization. It is in- locations prior to establishing the runoff plots are detended to provide supplementary information to the es- scribed by the following equations: tablished soil classification systems with special I = 30.8t,/2 + 26.1t ... Onne reference to agricultural soils in which variable charge 1= 16.0t,/2 + 93.3t ... Ikom colloids dominate. Vertisols, Alfisols, Ultisols and Incep- I = 2.1t1/2 + 6.3t ... Jos tisols dominating in high activity clays or constant Where I is the accumulative infiltration in centimeters, charge minerals are excluded from this system since the and t is the time in minutes. agricultural soil evaluation systems for such soils are well established for the temperate and subtropical re- Field measurements of soil bulk density for 20 soils indigions. cated a range of 0.70g cm-3 for Ikom to 1.5g cm-3 for
Bakura and Tumu, Nigeria. Similarly, the soil-water The proposed system may be briefly summarized as fol- transmissivity ranged from 236 cm/hr for Ikom to 1.4 cm/ lows: hr for Samaru, Nigeria. A plot of the accumulative infiltration vs. time for some Nigerian soils is shown in Fig. 9. 'Condition modifiers Estimates of erodibility for some Nigerian soils by using (or soil fertility the USDA Nomogram indicate a range of 0.05 f/acre/footlimitation) ton for Ikom to 0.56 t/acre/foot-ton for steep lands near
Soil group Sub-group Physical Chemical Abakaliki, Nigeria. A majority of the soils, however, have a low erodibility of about 0.1.
Kaolinitic soils Eutric w, r, c t*, (m*) Dystric w, r, c t, k, a, (m)
Siliceous soils Eutric w, c t*, k*, (m*) Hydromorphic soils in West Africa Dystric w, c t, k, a (m) A study was carried out to assess the quality and distriEutric w i bution of hydromorphic soils in various wetland areas in
AIophanicsoils Dystric w i, t, k, a West Africa and their suitability and limitations for
Dyssoils trick i, krainfed and irrigated rice production. Dystric i, t, k
Hydromorphic soils could be defined simply as those The eutric subclass refers to high "base" status, and dys- soils where water can gather in sufficient volume and tric to low "base" status. The soil fertility limitations in time to produce the effects of gleying or reducing re5




inputs and good water control for intensive wetland rice RVOM production. More fertile rice soils in terms of soil-texture
2. LAFIA and nutrient status comprise about 15-20 percent of the
4.TUAGU 37 soils studied (Tables 2 and 3). Wetland areas with
, GMANA
"T 30 S. Wu, such soils may be developed into productive rice land.
3Zo 6. TUMBIu More detailed and systematic soil surveys by national
7. SAU organizations are needed in order to determine the extent and distribution of the more fertile wetlands and their feasibility for rice production.
Table 2. Texture, organic C and pH of 37 hydromorphlc soils from the forest region of West Africa
(southern Nigeria, Sierra Leone and Liberia).
Properties Distribution, %
and range Surface soil Subsurface soil
pH (H20, 1:1)
4.0-5.0 78 73
5.1-5.5 14 22
5.6-6.0 8 5
1 2 3 & 5 676910 20 30 405060 80100 200 AWO am B O
TIME (I) Organic C, %
0.5-1.5 38 97
Fig. 9. Accumulative infiltration for some Nigerian soils. 1.6-2.5 24 3
2.6-5.0 30 0
> 5.0 8 0
gime. While hydromorphic soils mean the same to all soil scientists in terms of morphological and chemical prop- Clay content, % erties, the different soil classification systems are still 0-15 46 19
not very interchangeable because their respective con- 16-25 19 35
cepts are based on different premises. 25-35 14 16
According to U.S. Soil Taxonomy, hydromorphic soils 35-45 8 16
have an aquic soil moisture regime, and they appear 45-55 5 3
under the suborder as well as the subgroup levels of 55-65 8 11
classification. The FAO/UNESCO soil classification emphasizes the concept of gleization, and Gleysols and the Silt content, % "gleyic" subunits constitute the major hydromorphic 0-10 16 24
soils of the "FAO Soil Map of the World." For both the 11-20 22 24
French and Belgian classification systems, the pro- 21-30 16 14
cesses of gley and pseudogley play prominent parts in 31-40 24 19
the definition of hydromorphism. Moreover, Vertisols in 41-50 12 14
West Africa are also subjected to seasonal waterlogging 51-80 10 5
(pseudogley) and should come under the realm of hydro- Source of Data: IITA Soil Information Bank, Njala Univermorphic soils. From the agronomic viewpoint, wherever sity College and University of Illinois Soil Survey Report excess soil moisture is the most important feature in land for Sierra Leone (1974), Ministry of Agriculture and use for crop production, soils in such areas will come USAID Soil Survey Report for Liberia (1977), Manor River underthe realm of hydromorphicsoils. Union Soil Survey Report, Liberia (1979).
An approximate correlation of the hydromorphic soils among the 3 international soil classification systems has Forest/savanna transition zone (southern Nigeria). Hybeen prepared at IITA. dromorphic soils in the forest/savanna transition zone
A preliminary inventory of some chemical and physical occur mainly in inland valleys and depressions. As the properties of hydromorphic soils from Nigeria, Sierra region is characterized by a rolling topography and Leone and Liberia may be summarized as follows: quartzose-rich basement complex rocks, the wetland valleys or depressions are generally small, and the soil Forest zone (Liberia, Sierra Leone and southern Ni- depths are shallow. All the same, they occupy about 10geria). Hydromorphic soils occur in coastal, fresh-water 15 percent of the total land surface. Most of the hydroswamps, inland swamps and small river valleys. Some morphic land in this region presently is not used for food areas are subject to deep flooding during the rainy sea- crop cultivation. son. An extensive study on the soil and land characteristics of
Approximately 80 percent of the hydromorphic soils (ex- several inland valleys in the region was completed in cluding acid sulfate soils in this region) are acidic (pH 1979. In this study, groundwater regime throughout the below 5.0) and coarse-textured (loamy sand and sandy year was used as an important criterion for the hydroloam) with low P and K reserves and low clay activity morphic land quality evaluation. N deficiency and Fe tox(Tables 2 and 3). Such soils would require high fertilizer icity associated with groundwater or seepage water were
6




Table 3. Nutrient status of 37 hydromorphic soils from Table 4. Properties of 27 hydromorphic soils from the
the forest region of West Africa (southern Ni- savanna zone and the forest/savanna transigeria, Sierra Leone and Liberia). tion zone of Nigeria.
Properties Distribution, % Surface soil Subsurface soil
and range Surface soil Subsurface soil Properties Range Mean Range Mean
Exch. Ca, meqIlOOg Forest/savanna transition zone (17 soils)
0-0.50 41 59 pH (H20, 1:1) 4.7-6.4 5.6 5.0-6.5 5.8
0.51-2.00 35 22 Organic C, % 0.37-1.96 0.91 0.07-1.36 0.42
2.01-4.00 5 5 Clay, % 4-68 15 3-70 21
4.01-8.00 11 8 Silt, % 4-51 21 5-53 18
> 8.00 8 5 Sand, % 13-89 64 10-84 61
Exch. Mg, meqIlOOg Exch. cations, meq/100g
0-0.50 70 78 Ca 0.67-12.83 3.48 0.47-11.13 3.34
0.51-1.50 11 5 Mg 0.29-7.40 1.53 0.21-7.89 1.67
1.51-2.50 8 5 K 0.03-0.26 0.13 0.02-0.46 0.11
2.51-3.50 8 3 ECEC, meq/100g 1.51-23.60 5.99 1.06-21.26 6.14
> 3.50 3 8 Bray P1, ppm 2-14 5 0.2-4 2
Savanna zone (10 soils)
Exch. K, meqIlOOg pH (H20, 1:1) 4.2-6.1 5.3 4.6-7.4 5.7
0-0.05 24 65 Organic C, % 0.29-2.70 1.04 0.07-0.75 0.42
0.06-0.15 35 22 Clay, % 6-50 22 5-70 28
0.15-0.25 30 14 Silt, % 8-51 37 6-55 29
0.25-0.35 11 0 Sand.% 4-82 42 5-89 43
Extractable P (Bray 1), ppm Exch. cations, meq/100g
0-5 51 89 Ca 0.55-7.94 3.50 0.40-6.84 3.34
6-15 35 8 Mg 0.20-3.36 1.32 0.06-3.25 1.30
16-25 8 0 K 0.04-1.92 0.40 0.04-0.29 0.14
25-35 0 3 ECEC, meq/100g 1.73-11.31 6.26 0.99-10.61 6.13
> 35 5 0 Bray, P1, ppm 2-39, 9 0.4-5 2
Source of Data: IITA Soil Information Bank, Njala University College and University of Illinois Soil Survey Report
for Sierra Leone (1974); Ministry of Agriculture and this region are montmorillonitic and, thus, of high clay USAID Soil Survey Report for Liberia (1977), Manor River activity. Hydromorphic soils with more favorable clayey Soil Survey Report, Liberia (1979). and silty texture are found in some extensive areas in the
river basins of central Nigeria. But, because of generally
low nutrient status and seasonal flooding, effective water
also stressed. Utilization of these lands at traditional and management and fertilization are required in order to improved levels of management for year-round crop pro- develop such areas for rice production. duction using wetland rice as the main season crop was
proposed. An estimate is being made of the distribution of hydroFurther inventory of properties of 17 selected soils from morphic soils of selected areas of West Africa using the region are given in Table 4. Again, the predominant LANDSAT imageries and a few large-scale soil maps that hydromorphic soils in this region are coarse and medium are available. The selected study areas include the textured, but their chemical quality is slightly better than wetland tracts along rivers of West Africa and inland their counterpart in the high-rainfall region in terms of swamps, such as those in Liberia and Sierra Leone and available Ca, Mg and K in the surface horizons. Available the wet sedimentary belts of southern Nigeria. Vertisols P status is invariably low although these soils have very and the vertic subgroups are widely scattered but not low P fixation capacity. fully documented on a national and regional basis.
Savanna zone (central and northern Nigeria). Hydromorphic soils in the savanna zone occur in a wide range Socio-economic analysis of land forms such as inland depressions, river valleys
and ancient and recent flood plains. The sedimentary Labor utilization plains of the Niger and Kaduna Rivers and their tributaries comprise a significant portion of potential rice Literature was reviewed on labor utilization in cassava, land in central Nigeria, which is yet to be fully developed yam, maize and upland rice production. Additionally, and utilized. from short regional crop production surveys-preliminary labor utilization data were derived for cocoyam and
The inventory of hydromorphic soils in this region is ex- soybean production in Nigeria (Table 5). ploratory due to lack of soil survey information. A preliminary inventory of soil properties of 10 selected sites in In 1980, further progress was made in the collection of central Nigeria (excluding Vertisols) is given in Table 4. time series data for food crop prices in West African Hydromorphic soils at these locations show wide vari- countries. In addition to food crop prices in Nigeria and ability in texture, organic matter and soil reaction. Miner- Cameroon, market prices were obtained from Ghana, alogical studies showed that many hydromorphic soils in Ivory Coast and Togo.
7




'Table 5. Estimated average labor utilization for differ- their landuse. The dominant and most widespread types
ent food crops per operation in Nigeria. of traditional farming systems are permanent compound farming, rudimentary sedentary agriculture (with or withFrom literature out short fallow periods), bush fallow cultivation and
sources From field surveys traditional permanent tree crop farming. Shifting Crop mandays/ha mandays/ha cultivation hardly exists, and taungya is hardly used in
Up- the region. Livestock is relatively insignificant. The bush
Cas- land Coco- Soy- fallow cultivation and the rudimentary sedentary agriculOperation Maise Yam sava rice yam bean ture are the most widespread systems of land use. Permanent compound farming is comprised of tree and Land preparation 24 95 40 55 36 35 arable crop farming enterprises with livestock as a third Planting 10 35 13 30 14 15 and minor activity. Land allocation is essential for homeFert. application 15 15 5 12 steads, and tree and arable crop farming with livestock
Weeding 25 70 45 50 38 28 is confined to homesteads and villages.
Harvesting 16 60 70 35 60 35
Outside the permanent compound farms, except for reTotals 90 275 168 175 148 125 tained (wild) economic tree crops such as oil palm,
planted tree crops and arable crop farming are not spatially integrated. Hence, tree crops, once planted, occupy Agroforestry surveys the land permanently while the arable crops are largely
An agroforestry field survey was undertaken in southern grown in the traditional bush fallow and/or rudimentary Nigeria to identify major types of landuse in traditional sedentary agriculture with a land use factor for uplands farming systems and establish the degree of integrationI between (a) forest and tree crop plantations and small (L ranging from 2.5 to 3.6 (Table 6). Thus, outholder tree crop farming and (b) traditional arable crop farming and livestock production systems. Economic side the permanent compound farms and traditional perwoody species, both cultivated and noncultivated, and C
their association with traditional arable crop farming manent tree crop farms, the land use ratio (R = C x received particular emphasis during the survey. The
6 survey locations-Ezzamgbo, Umudike, Onne, Ikom, 100) in eastern Nigeria is less than 34 percent for uplands /$Imkpa and Uyo-are in the humid tropical zone, ex- and 51 percent for bottomlands (Table 6). cept Ezzamgbo, which is in the transition from the tropi- In terms of land allocation, planted tree crops (including cal wet to the tropical wet-and-dry climatic zone. plantains and bananas) accounted for about 67 percent Landuse in traditional farming systems. Both traditional of the cultivated land; tree and arable crops mixture for agriculture and rural settlements are largely confined to 7 percent; and arable crops for only 25 percent (Table 7). uplands with bottomlands seldom used for agriculture. About 76 percent of the farmers had arable crop farms All farmers in the survey areas have rain-fed upland under 4 ha, 45 percent of which were under 2 ha. Simifarms. Only 23, 13 and 23 percent of the farmers in the larly, about 56 percent of the farmers had tree crop farms derived savanna, lowland forest and coastal lowlands, under 4 ha. Tree and arable crop mixtures were largely respectively, have additional farms located in bottom- confined to farmers with the smallest amount of cultilands. Permanent compound farm plots, common in the vated land. central zone of the region and in the settlement sites, are Cultivated areas devoted to the important tree and arable typically located on uplands or well-drained sites. The crops areasesed forah trey ation. most common type of settlement is the dispersed home- crops grown were assessed for each survey location. stead type, which is also correlated with the prevalence Tree and arable crops most frequently grown with considerable acreage, in the order of importance for each of permanent compound farms. group, were (a) oil palm, cocoa, plantain/banana, kola,
Major types of traditional systems of agriculture and rubber and citrus and (b) cassava, yam, maize and rice.
Table 6. Small holder land use in eastern Nigeria: Type of land cropped and cropping/fallow duration (1980).
Derived Savanna Lowland Forest Coastal Lowland
Ezzamgbo Ikom Umudike Akamkpa Onne Uyo
Bot- Bot- Bot- Bot- Bot- BotUp- tom Up- tom Up- tom Up- tom Up- tom Up- tom land land land land land land land land land land land land Enterprise/Farmer Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr.
Cropping year, average 2 1.5 1.8 2.20 1 2.10 1 2 1.3 2.15
Cropping year, range 1-5 1-8 1-3 1-3 1 2-3 1 1-3 2-8 2-3
Fallow year, average 3.2 1.3 5 4.9 1.25 4.9 1.35 4.9 1.3 4.1
Fallow year, range C + F' 2-5 1-3 3-7 5 6 3-7 1-5 2-7 1-8 2-7 Land use factor (L = C ) 2.5 1.9 3.6 3.25 2.25 3.33 2.35 3.45 2.08 2.9
Frequency of upland and bottom
lands (% of farmers using) 100 35 100 100 5 100 20 100 50 100 5 '(See Okigbo and Greenland, 1977 for classification)
8




Table 7. Small holder land use in eastern Nigeria: Land allocation to enterprises per farmer (1979-80).
Derived Savanna Lowland Forest Coastal Lowland Row Total Enterprise Ezzambgo Ikom Umudike Akamkpa Onne Uyo Av. %
Average No. of plots 4.5 3.7 5.20 4.95 4.45 5.50 4.91
Area under Tree Crops (ha) 1.34 7.10 15.05 13.79 6.30 11.87 9.24 67.0
Area under Arable Crop (ha), 3.00 4.43 2.30 2.47 5.00 3.78 3.50 25.0
Area under Tree/Arable
Crop (ha)' 0.12 1.01 0.56 0.81 3.03 0.51 1.01 7.0
Av. total Cult. land (ha)l 4.46 12.54 17.81 17.07 14.33 16.16 13.75
I Fallow periods are often short and contain some cassava resulting in some short fallows being considered as cultivated plots by farmers. This has tended to put the total cultivated land slightly higher along with the problem of double counting of same plots due to intercropping and relay cropping.
Small ruminant (goats and sheep) and poultry pro- Proposed agroecological regions of the survey area: a duction is common with a general trend of more small synthesis. A map at a scale of 1:3,000,000 delineating ruminant production in areas of higher demographic agroecological regions has been prepared using the data densities where goats are more common and important of this survey (Fig. 10). Climate, geomorphology and in both humid and subhumid zones, while sheep and soils, agricultural land use (farming systems), demopoultry are more important in the subhumid (derived sa- graphic conditions (density and settlement patterns) and vanna) zone. The number of domestic animals kept by the interaction of these have been considered in identieach family is small. fying these regions.
Resource allocation to enterprises and production con- Accordingly, 5 major agroecological regions were identraints. Family labor and arable land are the 2 principal tified: resources. Farmers are dependent on family labor and
only use hired labor during labor peak periods. A crop- I. Basement complex (cocoa/forest) with 4 subping calendar of the 13 most important tree and 18 ara- regions. ble crops showed that upland tree and arable crop I1. Sandstone and coastal sand complex (oil palm/ farming, as parallel systems of production, compete for root crop) with 5 subregions. farm labor during land clearing, preparing and weeding. Ill. Niger delta and coastal swamp complex with 2 It was observed that there is a better utilization of family subregions. labor by having both tree and arable food crops than by IV. Upland moist savanna complex with 4 subreeither alone, and the 2 enterprises fulfill both cash and gions. food needs of the family. The competition for labor be- V. Mangroves and coastal sand complex. tween the tree and arable food crop enterprises could be
reduced, if not eliminated, if both are integrated on the Region V was not surveyed, for it is considered a nonsame land and, when possible, simultaneously operated. agricultural zone. Similarly, region Ill was not included in However, the practice of burning the bush in land prep- the survey because it is nonzonal in that land use is govaration induces the physical or spatial separation of the erned by the Niger River as the environmental factor. 2 enterprises and causes arable crops to be grown in Only the lower part of Region IV was included in the distant farms away from the homesteads (Table 7). survey, the derived savanna part, and this region, often
referred as the middle belt is agriculturally (crop and
State of tree and arable crop farming. In general, there livestock) important and is currently experiencing inappears to be less expansion of tree crops on new lands. creased agricultural developments with a corresponding Any increase, especially in cocoa, kola and citrus, is due increase in population. to rehabilitation and an increase in density of tree crops.
Most farmers in the drier savanna and coastal lowlands, Region I and II are the most important for rain-fed tradiparticularly under high-population conditions, thought tional farming. The major distinction between these is tree crop farming to be declining. Major constraints to their difference in geology and soils. In general, soils of tree crop replanting and expansion were credit and lack Region I are geologically better soils (mostly Alfisols) of production inputs. and can support both tree and arable crop farming. The
r zone is currently used for cocoa, kola, citrus, etc., and a Farmers' view toward farming. Many traditional farmers great variety of food crops is grown. On the other hand, themselves view farming as a nonprofitable enterprise Region II has soils that are chemically poor but with and will abandon it if given other options. Eighty-six per- good physical properties (mostly Ultisols) that make tillcent of the farmers stated that they will continue farming. age easy. Both soils are erodable and easily degradable Of these, 68 percent stated that they will continue be- with intensive cropping, particularly soils of Region II cause they have no other option available to them, 12 where the situation is already serious. percent stated that they have great family responsibility
that requires them to continue farming and 6 percent The agroforestry potentials (alley or strip cropping, stated that they do not wish to buy foods from the mar- planted fallows or taungya, multi-storey farming and ket. Only 14 percent stated that they will continue farm- agro-silvo-pastoral) are higher for Regions I and IV ing because it is profitable. Those wishing to discontinue where such new or improved systems of land use may farming, especially arable crops farming, often gave old not be difficult to adopt. But, the same cannot be said age as their main reason. for Region I1.
9




Fig. 10. Agro-ecological regions in the humid and sub-humid zones of Nigeria. 1
, Basement Complex Regions (Cocoa/Forest) llCb Cross River Basin-sedimentary sands-/ow
IA Northern sub-humid zone (W. Nigeria)-pop. pop.
high liD (Coastal) Sedimentary Sandy Plains-(E. NiI~a Central Humid, Sub-humid Uplands-(W. Ni- geria)-high pop.
geria)-pop. high. "
IBb Eastern Humid Uplands (E. Nigeria) base- IIl. Niger Delta and Coastal Swamps
ment complexlvolcanic-low pop. IliA Middle-Upper Delta-medium to high pop.
IC Southern Humid Lowlands (W. Nigeria)- IllB Lower Delta and Coastal Swamps-low pop.
basement complex/coluvial, alluvial deposits
. high pop. IV. Upland Moist Savanna
IVAa Western Moist Savanna-mostly Basement II. Sandstone and Coastal Sand (Qilpalm/Root Crops! Complex, medium-high pop.
Forest Complex) IVAb Eastern Moist Savanna (sandstone hills)IIA S.W. Sub-humid lowlands-sandy lowlands high pop.
-medium pop. IVAc Eastern Moist Savanna-medium pop.
liB Lagos and Adjacent Lowlands-sedimentary IVB. North Eastern (Guinea) Savanna-Sandsands-high pop. stone medium to low pop.
IICa Benin Lowlands-Sedimentary sands-medium to low pop. V. Man groves/Coastal Sands"
Landandsoi ma age ent Effect on soil bulk density. Deforestation resulted in a
~Lan an sol ma age ent significant increase in soil bulk density (Table 8), and
there were slight differences in bulk density among varResearch on land and soil management includes the fol- ious methods of land clearing. Maize, being an open-row lowing areas: land clearing and development, tillage sys- crop grown immediately after deforestation, increased terns and small tools development and management of soil bulk density (data of 1979), while cassava, being a kaolinitic Alfisols and siliceous Ultisols. close canopy crop, decreased soil bulk density. Moreover, tuber development just beneath the soil surface may have contributed to decreasing the bulk density of Land clearing and development the layer above the tuber and increasing the bulk density
Hydrological investigations and characterization of soil oftelyrbowheues.Dfrncsisiluk
physcalproprtis asinfuened b mehodsof and density were'also reflected in the infiltration rate and clearing and post-clearing soil management were con- pntoee eitne tinued in 1980 for cassava planted in the 1979 second Effect on total water yield. Water yield from the cleared season and harvested toward the end of 1980. watershed treatment was 259 mm while the forested
10




e7,,.,nt was only 2 mm. Deforestation contributed to a Effect on cassava growth and yield. Seedling mortality
.-. ::nt amount of seepage or ground water flow. The was about 50 percent in these treatments because of .-..., :en: stream with only traces of flow during pe- shading by trees in the traditionally managed plots and
-.,eavy rains is now a perennial stream with mea- by maise in the no-tillage plots (Table 10). Since cassava
..ow throughout the year. The maximum surface was planted in no-tillage treatments through 6-8 week&as observed in September; the subsurface flow old maize, row spacing was often more than 1 m (some.' er was almost equal to surface runoff. Continu- times 1.5 m), which also contributed to a low plant pop-.:,..:..through the dry season also indicates the pos- ulation. Cassava tuber yield varied by a factor of 2-2.5
cf sizeable ground water storage that may be among treatments with the lowest yield in the tradition.... nt at least in analyzing the hydrological balance ally managed plots. Conventional tillage plots were ce~~red watershed. planted at least 6 weeks later and harvested about a
month after no-tillage plots. Tuber yield in conventional tillage plots would have been even more if they had been Table 8. Effects of land clearing methods on soil bulk harvested 2-3 months later. Nevertheless, tuber and stalk
density in glcm for the 0-5 cm depth (IITA, yield from the no-tillage plots were comparable with
1978-80). those from the plowed and ridged plots.
T, itment 19781 1979 1980
" -.1ni.onal farming 0.64 1.06 1.07 Table 10. Effects of land clearing methods on cassava
',' inual clearing-no tillage 0.68 1.14 1.05 growth and yield (1ITA, 1980)
,,'inual clearing-conventional tillage 0.68 1.20 1.29
Sretar blade 0.70 1.19 1.37 Yield
Tr,?e pusher-no tillage 0.66 1.25 1.37 Plants! Tubers! t/ha
Tree pusher-conventional tillage 0.53 1.22 1.32 Treatment ha ha Tubers Stalks
,'ean of 25 replications. Traditional farming 4,540 31,570 7.7 14.6
1)78 data prior to clearing. Manual clearingno-tillage 5,230 36,640 15.0 22.4
Elfect on surface runoff and erosion. Soil erosion and Manual clearing,.atter runoff were lower for cassava than maize. More- conventional-tillage 4,850 36,180 11.7 26.1 ;';er. there were no significant differences among treat- Shear blade-.fnts either for runoff or erosion, except for the tree no-tillage 5,420 30,550 14.1 14.6 pusher-conventional tillage treatment. In spite of the Tree pusherprotective cover of the cassava canopy, this treatment no-tillage 5,800 39,580 16.8
had 48 mm of water runoff and 4.2 t/ha/annum of soil Tree pushere2rosion (Table 9). In comparison to this, runoff and ero- conventional-tillage 12,700 59,940 17.5 23.0 sion from the shear blade-no-tillage cassava was 5.0
Table 9. Effects of land clearing methods on surface Tillage systems and small tools
runoff and erosion (IITA, 1980) development
Run- Erooff, sion, Effects of tillage methods on maize production. NoTreatment mm tha tillage methods with residue mulch have proven useful
Traditional farming 0 0 for some row crops in kaolinitic Alfisols in the forest
Manual clearing-no tillage 0.2 0.001 zone of Western Nigeria. However, long-term studies at
Manual clearing-6.n0 tlag 0 0.06 IITA have indicated that soil compaction can be a probSha clearing-conventional tillage .0 0.0 lem in no-tillage plots within 3-4 years. Moreover, soil Sre blade-no tillage 5.0 0.08 compaction is more severe on mechanized than manTree ousher-no tillage 3.0 0.04 ually cultivated plots. Crop residue mulch is needed for
Tree pusher-conventional tillage 48.0 4.2 many other uses (fodder, building houses and fences,
fuel, etc.) and, therefore, may not be always available in mm and 0.08 t/ha. On the contrary, runoff and erosion the quantity required for effective soil and water conserfrom the shear blade-no-tillage maize treatment was 53 vation. Chiseling in the row zone rather than plowing the mm and 1.9 tlha'annum, respectively. These results on entire field, which makes the soil vulnerable to erosion, the effects of methods of land clearing and post-clearing may be an alternative to ameliorate the soil of the comsoil management have important practical agronomic paction hazard. Plowing at the end of the rainy season Implications. It seems that, in the long run, post-clearing may be another method. With this background, an expersoil management has the most important effect on soil iment was carried out at IITA with the following treaterosion, runoff and decline of soil physical and chemical ments: properties. This is not to say that the methods of land A. No-tillage with residue mulch. clearing are not important because a combination of B. No-tillage with chiseling in the dry season. harmful land clearing and post-clearing soil manage- C. Moldboard plowing followed 2 harrowings (residue ment methods, such as land clearing with tree pusher/ plowed in). root rake followed by conventional plowing and harrow- D. Disc plowing (residue disced). ing, results in the most losses in water runoff and soil E. No-tillage with residue removed. erosion and in rapid degradation of soil physical and F. Moldboard plowing at the end of rainy season and chemical properties. harrowing at planting.
harwn t lnig




G. Moldboard plowing followed by 2 harrowings (res- Maize grain yield declined sharply in a field with its
idue on the surface). twelfth continuous crop of maize (2 crops per year).
H. Moldboard plowing followed by 2 harrowings, fol- However, the rate of decline was much greater in conlowed by ridging (residue plowed in). ventional than in no-tillage plots (Fig. 11). Soil erosion
The effects of these treatments were investigated for soil has been a severe problem for yield decline in convenphysical properties and crop response. Preliminary re- tional-tillage plots while soil compaction has been a sesuits indicated that for a sandy loam soil, tillage methods vere problem in no-tillage plots. had no significent effects on grain yield (Tables 11 and 12). However, it will be useful to conduct a similar study for different soils and ecological conditions to provide 50 guidelines for appropriate tillage methods. BLOCK
00 NO TILLAGE UNTERRACED
Table 11. Effects of tillage methods on grain and stover CONVENTIONAL TILLAGE-TERRACED
yield (IITA, 1980 First season). 40Grain yield Stover yield Harvest index Treatment tlha tlha %
A 2.52 A 6.3 AB 39.8 A
B 2.00 A 5.2 B 39.8 A
C 2.69 A 7.7 A 35.0 AB 030
D 2.41 A 6.7 AB 36.0 AB 0
E 2.14 A 5.9 AB 36.2 AB
F 2.62 A 7.3 AB 36.5 AB
G 2.07 A 7.7 A 27.0 B
H 2.25 A 6.9 AB 33.0 AB
LSD (5%) 0.82 2.1 10.5 9 20
Table 12. Effects of tillage methods on grain and stover
yield (IITA, 1980 Second season).
Grain yield Stover yield Harvest index o Treatments t/ha t/ha %
A 2.40 A 4.80 A 51.8 A
B 1.81 A 4.32 A 43.5 A
C 2.00 A 3.65 A 53.3 A
D 2.12 A 3.53 A 60.2 A
E 1.80 A 4.22 A 42.3 A 1975 1976 4977 1978 1979 1960 1981 1982 1983 1984
F 1.81 A 3.58 A 49.9 A YEAR
G 1.97 A 4.54 A 43.6 A Fig. 11. Maize grain yield (Cv TZPB) under conventional
H 2.05 A 4.27 A 49.3 A and no tillage on a kaolinitic Alfisol cleared from secLSD (5%) 1.01 1.81 17.7 ondary forest.
Wt
D11. M|Il (
Poor stand In conventionally tilled maize with water Maize planted at the same time as the conventionally standing for up to 2 weeks after rain because of com- tilled maize, it has a good crop stand and compaction paction and low permeability, was not severe enough to reduce infiltration.
12




E!,ecI of tillage methods and mulches on crop produc- Table 15. Effects of tillage methods and N on rice grain
-.on in heavy-textured soils in Zanzibar. In collaboration yield (IITA, 1980 Second season).
- e University of Morogoro, Tanzania, tillage studies
7 :-ducted in Zanzibar to investigate the effects of Grain yield t/ha
-s of seedbed preparation and mulching on crop N rate Conventional
,r Different treatments investigated were black and kg/ha No Tillage Tillage
,,. coiythene mulch, no tillage, ridges and bare flat
i,.e Differences in plant growth among different 0 1.99 3.75
-rents were attributed to differences in soil temper- 30 2.94 4.79
.1!,v' ., and moisture regimes. These soils have high clay 90 4.02 5.05
c --,nt and low permeability. High soil moisture content
.1a :} oor aeration were more serious in no-tillage and LSD (5%) 0.82
rucned of ots during periods of frequent and heavy !a.-_ consequently, crops were more chlorotic than in
-cnventional-tillage plots. Effects of different treatments Effects of tillage methods and mulches on yam producen yield indicate that the highest maize yield was tion. An experiment was initiated at Onne to investigate .;,mined under white polythene mulch treatments and the effects of tillage methods and mulches on yam pro,c'vwest under no-tillage treatments (Table 13). These duction. Treatments consisted of planting on either the
c fencess in grain yield were attributed to the number flat or ridge with and without residue mulch. Observa-, !;rains per row and the unit grain weight. The yields of tions were made for soil temperature, moisture and bulk co uea and soybean also observed a similar trend. density and root and shoot growth and tuber yield. Leaf area was generally more in mulched than unmulched plots with the least leaf area measured in the unmulched Table 13. Effects of seedbed preparation on soybean
grain yields in Zanzibar.
Grain yield t/ha
K Treatment 1979 1980
* lack polythene 0.8 1.1
Vhite polythene 1.5 1.0 W
Nc-tillage 0.5 0.4
Ridges 0.6 0.5
Bare 0.8 0.6
LSD (506) 0.4 0.2
Effect of tillage on paddy rice production. In 1976, grain i. \ ;*
yields of rice were similar in conventional and non-tillage va.,
methods. Yields were significantly affected only by the
level of N application. The yields were not affected by the .
tillage treatments for about 3 years. Since 1979, the plant ,
height, tillers and grain and straw yield have been less in conventional-tillage than no-tillage for sandy soils (Table
14). In clayey soil, the differences due to tillage treat- No-till rice (top) had a poor stand and low yield after 6 ments were not pronounced (Table 15). Comparisons of consecutive crops compared to good stand and high
rice yield in Tables 14 and 15 indicate that an optimum yield in conventionally puddled paddy (below).
level of N for rice for a clayey soil is about 60 kg N/ha; whereas, the effect of N on rice yield in a sandy soil was not significant. The adverse effect of no-tillage on rice may be due to many factors, including nutrient imbalance. In a sandy soil, leaching losses of applied fertilizer on an untilled paddy may be more than in a puddled soil.
Table 14. Effects of tillage methods and N on rice grain
yield (IITA, 1980 First season).
Grain yield t/ha
Conventional
Fertilizer No Tillage Tillage P11
With 2.50 3.82
Without 1.98 3.13
LSD (5%) 1.00 .
13




ridged treatment. Table 16 shows that 20 percent more volumetric heat capacity, generally increases with an inyam tubers were produced in mulched than unmulched crease in soil moisture content up to a certain value. The plots with least tuber yield measured in the unmulched exact relationship depends on soil texture. Thermal difridge treatment. The beneficial effects of residue mulch fusivity measurements were made for 20 different soils, may partly be attributed to favorable soil temperature and the relationship between soil moisture content and and moisture regimes. Observations on soil moisture thermal diffusivity for some soils is shown in Fig. 12. In content made 40 and 90 days after planting indicated general, thermal diffusivity is higher in coarse-textured that unmulched ridges had the least soil moisture re- sandy soils than heavy-textured clayey soils. The regresserves. sion equations between sand, silt and clay on thermal
diffusivity are given in Table 18.
Table 16. Effects of tillage methods and mulches on
yam production (Onne, 1980).
Ridge Flat Ridge plus Flat plus
no mulch no mulch mulch mulch
Length of 3 2
tuber 20.7 20.7 23.5 21.7 6
cm
Diameter
of tuber 13.3" 13.9a 16.7b 17.1b Cly o.C
cm I so" Clay Loam 61 27 12
Yed2 Clay Loom 47 33 :.56
Yield 14.1 16.3 19 18.9 2c3. Loom 33 22 47
tlha o 2 ,4
4 Lom 27 23 1.61
, 5 Lo, 78 Is 039
ow 67 020 030 0. o40 045
Effects of soil bulk density on root and shoot growth of WATER CONTENT, C._.m3
cassava. A root-box study was conducted to investigate the effects of 3 soil bulk density treatments, 1.4, 1.6 and Fig. 12. Thermal differsivity of various soils as a function 1.8 gcm-3, on cassava root and shoot growth. Cassava of the water content. Size of glass beads used 50-75 U. was grown in collapsable boxes 48 cm x 46 cm x 100 cm, with sides that could be removed for monitoring root growth at different stages. Measurements on root growth were made for cassava cultivar TMS30572 at 48, 78, 108, 132 and 185 days after planting. Root density, measured. Table 18. Regression equations relating thermal diffuas length and dry weight, was not significantly affected sivity, D (cm2 sec-') to percent sand, silt, clay by the soil bulk density treatments investigated. Neither or organic matter in a soil sample. were there significant differences in plant height, leaf area nor shoot dry weight. Cassava can withstand soil Simple compaction more than grain crops such as maize, cowpea or soybean. Nevertheless, the shoot: root ratio was Multiple generally higher at soil bulk density of 1.6 gcm 3 than 2. D = 0.01182 01099 X 10-3 r =0.844** other densities (Table 17). The optimum density for the cl 0.1680 x 102 cm high feeding: tuberous root ratio was 1.6 gcm-3. These 3. D = 0.0121 0.3066 x 10-4 r = 0.860** results imply that effects of soil compaction on cassava Si 0.1104 x 10-3 ci 0.1479 x 10-2 may be highly dependent on soil texture as the latter Om affects both the intensity and capacity factors for nu- 4. D = 0.001025 + 0.1104 x 10-3 r = 0.60** trient and water availability. Sa + 0.798 x 10 4Si 0.1479 x 10-2
Om
Effects of tillage methods and mulches on soil temper-. ature regime and its effects on crop production. Ther- Sa = Sand, Si = Silt, Cl = Clay, and Om = Organic mal diffusivity, the ratio of thermal conductivity and matter.
Table 17. Effects of soil bulk density on cassava's shoot: root ratio and feeding root: tuberous root ratio.
Bulk density Bulk density Bulk density
1.4 1.6 1.8
Root dry wt Root dry wt Root dry wt
Days after Shoot dry wt Tuber dry wt Shoot dry wt Tuber dry wt Shoot dry wt Tuber dry wt
planting Root drywt x 10 1 Root drywt x 10 4 Root drywt x 10 1
48 134.72 58.89 50.00
77 188.24 84.53 258.89 305.08 126.04 185.90
106 140.13 67.47 128.55 70.40 110.4 98.81
134 150.17 42.44 267.41 41.90 348.91 12.60
185 401.96 299.79 249.39 537.66 200.01 329.94
14




A knowledge of the thermal diffusivity is necessary in 56predicting the temperature profile of a soil under differ- 5ent systems of management. Soil temperature affects 2 -"
crop growth directly as well as indirectly through its in- 48- Rodiotion 59Ogr/Cal cm2
teraction with soil moisture regime and its effects on root .o. Flat bare / growth, biotic activity and plant nutrients. Soil tempera- Z Sybean (196) ture measurements at different depths and times can be 40. -- Cowpeo (1.80) cumbersome. With the knowledge of soil thermal prop- 936 -&- Maize (1-41) erties, moisture regime and boundary conditions, the A temperature profile of a soil can be predicted with a rea- 32 sonable degree of accuracy. An attempt was made to 28 compare the measured soil temperature with that esti- 24 37 days after seeding
mated by the Fourier series solution and the Hanks 24 model. The initial and boundary conditions were mea- 20sured for the following treatments: bare flat, flat mulched 1'6 2,0 i4
with crop residue ( ut/ha, ridged and flat surface cov- Ho,, of the day
ered with transparent polythene. 36 . "Soybean (0-37) Radiation z 230g/Col cm-2
-x- Cowpeo (0.41) 1 1'"
Soil temperature measurements were made for maize, F 4 -34 Maize (047) 1 cowpea and soybean. The predicted soil temperature at 32 20 cm depth under maize and cowpea was within 0.6C of the measured soil temperature. Soil temperature can E also be predicted from the air temperature (Table 19). 28
0
The effect of canopy cover on diurnal fluctuations in soil 29 doW after seeding
temperature at 29 and 37 days after planting for a sunny and a cloudy day, respectively, is shown in Fig. 13. The 24- -X ........
higher temperature under soybean compared to the 22 lower temperature under maize and cowpea is partly 1'6 2'0 24
attributed to poor stand because of a supraoptimal temperature regime during the periods of seedling Fig. 13. Diurnal temperature variation at 5-cm depth in emergence and establishment, a flat base soil under 3 crop canopies at 2 occasions
The values of amplitude and phase angle for the first 3 following seeding. Numbers in the brackets denote leaf harmonics of the temperature wave for different surface area index on the day of temperature measurements. conditions and crop covers are shown in Table 20. It is
Table 19. Calculated and measured soil temperature in control (bare flat) treatment.
Air
temperature
at 1-r' Soil temperature (0C)
height
Soil (C)
depth Maxi. Mini. Maximum Minimum Mean
Date (cm) Meas. Calc. Meas. Calc. Meas. CaIc. Difference
4.12.80 1 34.3 22.9 44.2 47.8 26.7 26.0 35.4 36.9 1.5
15 35.5 34.2 29.5 30.0 32.5 32.1 0.4
8.12.80 1 35.5 23.2 45.8 49.2 27.2 26.3 36.5 37.7 1.2
15 35.6 34.4 29.6 30.3 32.6 32.3 0.3
Maxi. = maximum; Mini. = minimum; Meas. = Measured; Calc. = Calculated.
Table 20. Total variance (OC) of the soil temperatures at various depths under different crops grown on bare flat,
mulched flat or bare ridged soil surface and its percentage accounted for by different harmonics.
Percentage of variance accounted for by the harmonic
Cover Crop-- No Crop Soybean Cowpea Maize
Harmonic Soil --*
Number depth (cm) 5 20 35 5 20 35 5 20 35 5 20 35
BARE FLAT
Variance (C) 115.53 19.90 4.45 140.83 28.75 6.45 19.30 3.49 0.56 18.12 3.20 0.57
1 94.8 95.6 92.6 93.2 96.0 93.3 92.3 94.0 91.1 94.4 95.3 89.5
2 3.8 2.1 4.5 5.3 1.7 4.2 4.4 4.3 5.4 3.2 2.5 7.0
3 0.5 0.7 0.7 0.6 0.8 0.5 1.3 0.3 0.4 0.9 0.9 1.2
Total
1 to 3 99.1 98.4 97.8 99.1 98.5 98.0 98.1 98.6 96.9 98.5 98.7 97.7
15




obvious that 90-95 percent of the variability in soil tern- Table 21. Effects of crop cover on runoff and erosion. perature regime under bare, maize, cowpea and soybean Runoff Erosion
covers can be explained by the first harmonic while 97- Crop cover (mm) (t/ha)
99 percent of the variability can be explained by the first
3 harmonics (Table 20). In general, the second and third Maize with sweet potato 154 2.2 harmonics represented 1.2-7 percent and 0.1-3.8 percent Cassava with melon 380 3.7 of the total variance, respectively. This implies that the Yam with melon 314 4.0 first 2 harmonics alone can explain the diurnal fluctua- Sweet potato 223 4.4
tion in soil temperature up to about a 30 cm depth. The Maize 197 4.4
third harmonic explains about 1 percent of the total var- Melon 303 6.3
iance. Weed fallow 251 4.4
Yam 186 2.0
Runoff and erosion under root crops. Table 21 shows Cassava with maize 457 9.2
the effects of canopy cover of tropical root crops at dif- Cassava 462 8.0
ferent growth stages on runoff and erosion. Cassava, although a closed canopy crop, takes a relatively longer time to provide a complete ground cover than maize or trol, the yield was significantly higher in conventional cowpea. Staked yams leave the ground surface exposed than no-tillage plots (Table 23b). to raindrop impact even when the canopy is fully devel- Weed control in conventional and no-tillage cassava was oped and, therefore, renders the soil more susceptible to evaluated in an Alfisol in a subhumid climate. The vegeerosion. Sweet potato, with quick and effective ground nation was a 2-year fallow consisting of perennial grass, cover, is a soil-conserving cover since it results in mini- Panicum maximum; and broad leaves, Eupatorium odormal runoff and erosion compared to cassava and yam atum, Alchornia laxiflora and Ficus spp. The fallow veg(Table 21). Mixed croppings of maize with cassava, cas- elation was first slashed in late February before the onset sava with melon and maize with yam resulted in a signif- of rains, and its regrowth was then sprayed in April with icant decrease in both runoff and erosion. Simultaneous 3.0 kg glyphosate/ha a.i. In the conventional-tillage secmeasurements of canopy cover made at different growth tion, the fallow vegetation was plowed under, and the stages will provide necessary information for computing cassava was planted at 1.0 m x 1.0 m spacing on the flat "C" values that can be used in the "Universal Soil Loss after harrowing. In the no-tillage section, the cassava Equation" for predictive purposes. was planted directly into sprayed regrowth. Cassava
Table 22a. Effect of residue management and tillage Weed control in no-tillage system practices on maize yield and crop perforThe effect of residue management and tillage on maize mance (kenne, 1980). production using chemical weed control was investi- Weed Grain
gated on a 1-year Eupatorium odoratum fallow. (Tables Management Plant Lodging F.W. Yield 22a and 22b). Maize yield was significantly lower in the practice HI. cm % g/m2 t/ha no-tillage plot than conventional (plow and harrow) and
reduced tillage (disc harrow) plot. Also, significantly Residue present 81.4 33 a 6.1 2.06 a more weeds grew at 4 weeks after planting where the Residue removed 81.7 29 a 6.7 2.06 a
moewed ge a wesResidue burnt 87.9 a' 35 a 12.2 a 1.98 a
residue was burnt off than where the residue was re- Plow and harrow 86.7 a 41 a 8.5 ab 2.12 a tained or raked off. The method of weed control affected Disc harrow 89.8 a 34 a 11.4 a 2.20 a grain yield but not lodging in maize. The maize yield was No tillage 74.5 22 5.0 b 1.78 significantly lower in the unweeded than weeded plots (Table 23a). When plots were hand weeded, the crop Means followed by the same letter in the same column yield was identical in the 3 tillage practices. However, are not significantly different at the 5% level of Duncan's when a preemergence herbicide was used for weed con- New Multiple Range Test.
Table 22b. Effect of residue management and tillage practices on maize yield and crop performance (Ikenne, 1980)
Grain
Residue Tillage Plant ht. Lodging Weed F.W. Yield
Management Practice cm. % g/m2 t/ha
Residue present Plow and harrow 83.8 ab' 45 a 5.8 b 2.13 ab
Disc harrow 88.7 ab 31 abc 7.2 b 2.20 ab
No tillage 71.7 c 24 bc 5.2 b 1.84 ab
Residue removed Plow and harrow 87.0 ab 38 ab 8.8 b 2.17 ab
Disc harrow 86.3 ab 32 abc 7.0 b 2.34 a
No tillage 71.9 c 18 c 4.2 b 1.68 b
Residue burnt Plow and harrow 89.3 ab 41 ab 10.8 ab 2.07 ab
Disc harrow 94.3 a 39 ab 20.2 a 2.05 ab
No tillage 79.9 bc 25 bc 5.7 b 1.83 ab
'Means followed by the same letter in the same column are not significantly different at the 5% level of Duncan's New Multiple Range Test.
16




Table 23a. Effect of tillage and weed control on maize tillage. The hand weed plot under no-tillage compared
yield and crop performance (Ikenne, 1980). favorably with the preemergence herbicide treatments Stand under conventional-tillage. This is an indication that
count good cassava yield is possible under no-tillage condi(x 103 Lodging Grain yield tions provided that weeds are controlled. The best weed Treatment pl/ha) % t/ha control in cassava was obtained when a mixture of metolachlor and fluometuron was used. This yield did not Plow and harrow 25 a' 40 a 1.91 ab differ significantly from that of the formulated mixture of Disc harrow 25 a 37 a 2.06 a atrazine and metolachlor, which has previously been
No tillage 18 21 1.81 b shown to be safe for maize/cassava intercrops.
Herbicide 22 33 a 2.03 a
Hand weeding x 2 25 a 34 a 1.99 a
Unweeded check 20 32 a 1.76 Table 24b. Effect of weed control on weed biomass
'Means followed by the same letter in the same column and cassava root yield in an Alfisol (IITA, are not significantly different at the 5% level of Duncan's 1980) New Multiple Range Test. Cassava
Weed D. Wt. root yield
Table 23b. Effect of tillage and weed control on maize Weed control t/ha t/ha
yield and crop performance (Ikenne, 1980). Atrazine plus metolachlor 3.18a' 19.53 b Stand Fluometuron plus metocount Grain lachlor 3.88 a 20.89 b
Tillage Weed X 103 Lodging yield Diuron plus paraquat 3.97 a 19.58 b
practice Control pl/ha) % t/ha Weed free 0 31.79 a
Unweeded check 3.39 a 12.27 c
Plow and 'Means followed by the same letter in the same column
harrow Herbicide 26 a, 41 a 2.12 a are not significantly different at the 5% level of Duncan's
Handweeding New Multiple Range Test.
x 2 27 a 40 a 1.96 abc
Weedy 22 bc 40 a 1.66 c
Disc harrow Herbicide 25 ab 34 a 2.20 a
Handweeding Small tools development and evaluation
x 2 27 a 39 a 2.08 ab
Weedy 22 bc 39 a 1.92 abc Rolling injection planter. The rolling injection planter
No tillage Herbicide 16 e 22 b 1.78 bc was modified to improve its performance in both the
Handweeding conventional and no-tillage systems. First, it was prox 2 20 cd 23 b 1.94 abc vided with a small metal cover to hold the seed on the
Weedy 18 de 19 b 1.71 c metering wheel to prevent uneven distribution of seed.
This cover holds the seed in the metering wheel until the 'Means followed by the same letter in the same column seed is just over the opener, and the seed is dropped are not significantly different at the 5% level of Duncan's almost directly on the ground through the opener (Fig. New Multiple Range Test. 14). Previously, the cutoff device on the rolling injection
planter had a tendency to pinch maize seeds in such a storage root yield was significantly depressed by tillage way that the seeds jumped out of the seed hole in the (Tables 24a and 24b). However, preemergence herbi- metering wheel. This happened just as the seed had alcides were more effective in conventional than no-tillage most completely passed under the cut-off device. Also, cassava. Crop yield was significantly higher in the hand the vibration of the machine had a tendency to shake weeded than in the chemically weeded plot under no- seeds out of the metering wheel prematurely. This
Table 24a. Effect of tillage and weed control on weed biomass and cassava root yield in an Alfisol (IITA, 1980).
Treatment Cassava root
Tillage Weed control Weed D. Wt. t/ha yield t/ha
Conventional tillage Atrazine + metolachlor 3.0 kg/ha 1.85 de' 25.08 b
Fluometuron + metolachlor 2.0 + 2.0 kg/ha 1.94 de 28.62 ab
Diuron + paraquat 3.0 kg/ha 2.71 bcd 27.27 b
Weed free 0 e 35.83a
Unweeded check- 2.13 cde 16.43 c
Mean 1.73 26.65
No tillage Atrazine + metolachlor 3.0 kg/ha 4.51 abc 13.98 cd
Fluometuron + metolachlor 2.0 + 2.0 kg/ha 5.81 a 13.17 cd
Diuron + paraquat 3.0 kg/ha 5.24 ab 11.90 cd
Weed free 0 e 27.75 b
Unweeded check- 4.65 abc 8.10 d
Mean 4.04 14.98
'Means followed by the same letter in the same column are not significantly different at the 5% level of Duncan's New Multiple Range Test.
17




(c) A single-row rolling injection planter with a fertilizer band applicator attached behind it is too heavy to balance.
____(d) A single-row rolling injection planter with a fertilizer band applicator attached beside it acting as a row marker as well as placing basal fertilizer works very well and eliminates using strings or poles for
______marking the rows. This system does not work well
on heavy mulch, which contains a lot of fine grasses or weeds and hides the fertilizer.
Neither the single-row rolling injection planter with a fertilizer band applicator nor the 2-row rolling injection planter are able to work on traditionally cleared land that has stumps left standing more than 15 cm in height. This
- is because the amount of deviation from the row to be
able to go around the stump is too great. The working rates of these machine combinations are shown in Table 25. Table 26 gives the amount of time to move row marker stakes when planting or spraying.
Table 25. Rate of work of planter combinations (IITA, Fig. 14. Two-view drawing of the metering wheel cover 1980).
for the rolling injection planter. Machine Man-h r/ha*
caused increased numbers of seeds in some hills and Single row rolling injection planter 10
none in others. For crops requiring 3 or 4 seeds to be Single row rolling injection planter plus
planted per hill, the small metal plate is not necessary. 2-rwrlingr ianjeto aplatr 63
The seeds distribute themselves fairly evenly between 2-row rolling injection plusftzer ban
*the hole openers giving good crop stands. appwrliato ineto lsfetlzrbn
A simple guard was designed that covers the opening Single row rolling injection with fertilizer
between the opener lever and the injection wheel to pre- band applicator row marker 13
vent trash from passing between. The guard does not *These times to not include time to move row marking
add significantly to the cost or weight of the machine, stakes.
Previously, the small pieces of wood left in the field after clearing fallow were found to catch in the hole opener
lever and wedge between the injector wheel and the han- Table 26. Time for machine operator to move row
dles. This makes the planter almost impossible to push. marker stakes per hectare (IlTA, 1980).
The solution to the planter sinking in soft soil was to fit a Width of machine man-hr/ha
wider hole opener lever. This lever is 5 cm wide, adding 0.75 m 5
3 cm to the width of the previous hole opener lever. The 1.5 m 4
extra width provides for more contact with the soil, thus, 20m distributing the weight over a greater area and keeping2.m3
the planter from sinking. Finally, the requirement of *Estimate.
planting in the row at other than 250 mm spacing has partially been solved. Seed rollers were made that give 2 and 3 seeds per revolution of the rolling injection wheel. This will give seed spacing of 500 mm (3-hole roller) and 750 mm (2-hole roller). A single hole roller could be made to give 1,500 mm spacing. This development brings a much wider range of activities for the rolling injection planter, giving it adaptability to various crops. In an attempt to increase the rate of planting, differentV configurations of planters and fertilizer band applicators were also designed and tested as follows:
(a) A 2-row rolling injection planter was designed to
in crease the field capacity and make a more stable planter by eliminating the sideways tipping of the
machine.
(b) A 2-row rolling injection planter with a fertilizer
band applicator attached behind it acting as a
press wheel worked very well, but lumpy fertilizer ~
slows down the planting operation as the lumps Guard on opener lever. Planter on right has guard inmust be broken up before being used. stalled.
18




t- ..ft opener /ever. Planter on left has 50 mm wide lever,
-jni .:er on right a 25 mm wide lever.
A-,." :.. Single-row rolling injection planter with the fertilizer
.... ... 7, -:band applicator attached.
: : Single-row rolling injection planter with fertilizer band
. "..' !applicator row marker.
Two-row rolling injection planter. : i '
U"
AI
/ -I,-.,.
Swo-row rolling injection planter with the fertilizer band
applicator attached. ac
19
Sigerwrligijcinpatrwt2etlzrbn




All machines were designed to either push or pull. Op- The results on the no mulch treatments for both maize erators generally chose to pull. When conditions made and cowpea support the conclusion of the earlier test on penetration a problem, it was improved if the planter was covering of seed. When the seeds are well covered, there pushed. Other trials were carried out to judge the vari- is higher germination. During planting, it was noted that ability of the rolling injection planter in stand establish- seeds are covered better when planting on soil with conment with no tillage. This variability may be caused by ventional than no-tillage. The soil falls back better onto the amount of mulch cover on the soil or by how well the the seed with conventional than no-tillage. On no-tillage seeds are covered. To determine whether the planter was planting, there is usually a small indentation above the covering seeds well enough, maize was planted with no seed where the opener is inserted. This may allow birds tillage. Ten pairs of rows were selected at random. One and rodents to find the seeds more easily and, thus, reof the pairs of rows was carefully dug up and the seeds duce the stand establishment. Further tests will be concounted, then carefully replaced in their holes and.cov- ducted to determine the causes of lower stands. ered with soil. The next row was left and not counted. Out of the total number of rows tested, the average stand The hole openers of the rolling injection planter were count for the seeds dug up was 82 percent. The count tested to know the maximum speed for pushing before for the rows left alone was only 51 percent. It is clear seeds are left on the ground's surface and the advanfrom this test that more work is required on seed cover- tages and disadvantages of the different types of hole ing. It was also noted that the heavier the mulch cover, openers. Table 29 shows the maximum speed for a the lower the stand count when using the rolling injec- formed hole opener, a wedge hole opener and a split tion planter. A test to determine the effect of the amount type hole opener. (Also see Fig. 15.) Both wedge and the of mulch cover on the effectiveness of the rolling injec- split hole openers had a maximum speed of 3.4 kph while tion planter was carried out on small plots with maize the formed hole opener had only 2.6 kph. This lower and cowpea, separately, replicated 20 times. speed for the formed hole opener was because a small cup near the end of the opener held the seed after it was Tables 27 and 28 show that there is a consistent reduc- opened causing a delay before the seed was dropped. tion in stand establishment as the amount of mulch cover increases for both maize and cowpea. This reduction in stand as mulch weight increased could be attributed to greater bird and rodent activity on the heavily i
mulched fields and/or shading from the mulch.
Table 27..Percent stand of maize on varying mulch +
weight per hectare (IITA, 1980). =
Mulch t/ha Stand percent
0 (Control) 74.4* Fig. 15. Side view of different hole openers for the roll0 47.4 lng injection planter; left to right, formed, wedge and
4 35.2 split.
6 32.4
-8 23.2
10 22.1 Table 29. Performance test of 3 types of hole openers
*The control seeds were planted by hand and well cov- at different speeds of planting. ered between the rows planted and the rolling injection % of Seed placement'
planter. Wedge" Split,
Kph Cup type 2 Type Type Remarks
Table 28. Percent stand of cowpeas on varying mulch 1.69 100 100 100
weight per hectare (IITA, 1980). 2.14 100 100 100
Mulch t/ha Stand percent 2.61 100 100 100
0 (Control) 84.0* 3.00 75.02 97.85 100
0 84.4 3.43 50.07 98.60 97.37 Limiting
4 59.7 4.00 3.2 80.66 83.31 -speed
6 52.9 4.29 13.05 15.00 17.00
8 51.5 'Percentage (%o) seeds deposited inside the hole.
10 41.1 2Formed hole opener carries the seed out, harder to pencontol eedswer platedby hnd nd wll oy- etrate the soil; clogged with soil very often.
*The coto ed eepatdb adadwl o-3Wedge hole opener frequently clogged with soil.
ered between th Ie rows planted by the rolling injection 4 Split hole opener; split opener squeezes out dirt. planter.
In the cowpea plots, no mulch treatments were rotovated CDA sprayer. A 2 head, hand-carried sprayer, which had before planting. The control seeds, which were well coy- been built previously at IITA, was modified to give more ered, had the same germination rate as the seeds planted even flow between the heads. The original sprayer had 1 with the rolling injection planter. In the maize plots, the feed line from the bottle to the cross supporting the no mulch treatments were not rotovated before planting. heads. The tube had a T placed in it at the cross support The control seeds had 27 percent more germination than to feed each of the heads (Fig. 16). The combination of the seeds planted with the rolling injection planter. low pressure in the spray bottle and the resistance to
20




SPRAY BOTTLE 3 Poleustalt
II TA, Ibadan
After 8years
DELIVERY TUBE DELIVERY TUBES 2--/
E /
U
MAIZE- RESIDUE
Li 2-0-"
2,A
TEE I
FIRST MESXG SECOND DESIGN E
MAIZE. RESIDUE
Fig. 16. Schematic diagram of delivery tubes for 2 head I~ CDA sprayer. 2:
S .'0- PANICUM MAX
0
flow in the 6 mm tubing would at times allow 1 head to uJ take all the flow, leaving the other with no flow at all. w This would happen normally when 1 head was higher nL 0.5than the other. This was mostly overcome by having a BUSH FALLOW
separate tube for each head coming from the spray bottle. The sprayer was set up with the spray heads 0.75 m apart to match the row width when working on standing 0 maize stubble. This 2 head sprayer was used in conjunc- 5 10 1,5 2)
tion with the 2-row planter with every other row of maize stover knocked down. The time required for using the 2 SOIL DEPTH, cm
heads was approximately 0.6 of the single head. This Fig. 17. Penetrometer readings of the surface layers of increase in the rate of work would probably justify the kaolinitic Alfisols under fallow and after 8 years of conextra expense of the 2 head machine, approximately 1.5 tinuous cultivation. times the price of a single head machine.
crop residue return resulted in an increase in soil bulk Management of Kaolinitic Alfisols density and acidity and a decrease in soil organic matter
levels, and, consequently, a reduction in CEC and exA long-term experiment established at IITA since 1972 changeable Ca and Mg status in the soil (Table 30). shows that fallow remains a vital component in maintain- Penetrometer readings of the cropped plots after 8 years ing the soil productivity of the highly erosive kaolinitic indicated a severe compaction of the surface horizons Alfisols on a predominantly rolling topography in the for- compared to the soils under grass and bush fallow (Fig. est/savanna transition zone of West Africa. This is, in 17). part, demonstrated by the alarming decline of some Continuous no-tillage maize with crop residue returned chemical and physical properties of the soils after 8 as surface mulch twice a year was able to maintain soil years of continuous cropping compared to those under organic matter (i.e., total N) levels comparable to that of natural and planted fallow, bush fallow. But, the decline in soil pH, CEC, exchangeEffect on soil chemical and physical properties. Contin- able Ca and Mg levels and the increases in soil bulk uous cropping for 8 years with insufficient amounts of density and penetrometer readings were considerable
Table, 30. Properties of surface soil (0-15 cm) under 8-year continuous cultivation and fallow after clearing of
secondary forest (Oxic Paleustalf, Egbeda Series, IITA, 1980).
Bulk density
Total (Fine earth)
pH N ECEC Exch. cations, meg/100g g/cm3
Treatment (H20) % meq/1OOg Ca Mg K 0-5 cm
Continuous cropping with minimum tillage
Maize, residue returned 5.0 0.18 3.23 2.19 0.41 0.35 1.20
Maize, residue removed 4.7 0.11 1.81 1.13 0.24 0.11 1.31
Maize/Cassava 5.6 0.15 3.40 2.04 0.42 0.32 1.25
Soybean, residue returned 5.0 0.11 3.05 1.65 0.42 0.28 1.23
Natural and planted fallow
Natural regrowth 6.5 0.19 5.14 3.53 0.91 0.41 0.88
Guinea grass 6.7 0.26 7.69 4.75 1.28 0.91 1.01
Pigeon pea 6.0 0.23 3.42 2.18 0.64 0.32 1.10
21




although to a much lesser extent compared to the no 10 crop residue plots (Table 30). Moreover, properties of soils under continuous mulch, no-tillage maize for 8 9 years were by no means comparable to that under bush No-,il Stow, Muhed
and grass fallow. Planted fallows such as Lucaena with V.7182.23756667X--I2X2
both leaves and branches returned as surface mulch compared favorably to that of natural bush fallow; whereas, results of pigeon pea (bush type) were inferior. 7 The latter result is mainly due to severe disease problems that cause the crop to die off after 1 year. Lack of effec- o6 NO-1,S .. w ... tive ground cover during or before the reestablishment .629 25.M56X652592 '_1 of the new crop may have caused the loss of Ca, Mg and R2. 0,7%9 K from the surface soil by eroding and leaching. \
Effect on earthworm activity. There was a remarkable N difference in earthworm activity between the fallow and cultivated plots. Earthworm activity monitored during a 4-day period in July, 1980 (Table 31), showed that the natural bush fallow plot had a significantly higher Hyper- 2 iodillus activity (columnar casts) than the 2 planted fallow plots (Leucaena and Guinea grass). The extremely low earthworm activity in the continuously cropped plots (8.5 years) is apparently due to many factors. Soil compaction, continued use of pesticides and herbicides and k2 Pin Ik 175 197 d77 198 079 980 the decline in soil organic matter are probably among the important ones. The results of Eudrillus activity Fig. 18. Calculated yield curves of maize for a period of (granular casts) taken during the same period were less 9 years under no-tillage with and without stover mulch indicative. Although the difference among the fallow and on a kaolinitic Alfisol. cropped plots was not statistically significant, the Eudrillus activity under Leucaena was considerably greater 0an that under bush and Guinea grass as well as the Mineralization of soil organic matter. cropped plots (Table 31). A small-scale, manual, land clearing project (0.25 ha)
was carried out to study the soil organic matter decomRole of crop residue mulch on maize yield under no- position and nutrient release after forest clearing. A sectillage system. Respectable grain yields under a mulch- ondary objective of the experiment was to demonstrate no-tillage system on manually cleared land were main- whether improved soil and crop management practices tained up to 4 years; but yield declined steadily, there- at the small farmer's level could, in fact, increase the after, despite adequate fertilization and plant protection length of the period under cultivation before the land is (Fig. 18). The reason for the yield decline is a complex returned to bush fallow. The improved practices inone. The decline in soil biomass activity due to cultiva- eluded no burning, no tillage and the use of mulches, tion, soil compaction and possible Mn toxicity due to soil fertilizers, preemergence herbicides, pesticides and an acidification are probably among the more important improved crop variety. Soybeans were planted immedigrowth-limiting factors that cannot be simply remedied ately after land clearing during the 1979 second season by conventional fertilization. The beneficial effect of crop followed by maize during the 1980 first season. Fertilizer residue mulch in the no-tillage maize system is evident, was not applied to either crop. Excellent growth of both particularly in areas of ustic soil moisture regime and the soybean and maize was observed, and tissue analysis frequent incidences of dry spells during the early crop- indicated no nutrient deficiency. Results of weekly monping season. These results suggest that to avoid perma-, itoring of NO,-N and NH4-N in the surface (0-10 cm) soil nent degradation of the "superficially" fertile Alfisols in are given in Figure 19. Substantial amounts of soil orthe forest/savanna transition zone of West Africa, culti- ganic N were mineralized during the onset of the rainy vated land needs to be returned to an effective fallow or season. These data further suggest that, in spite of riskresting period after 4-5 years of cropping under recom- ing a drought stress, early planting of maize is preferred mended soil management practices, i.e., the mulch no- so that the crop can effectively utilize the high levels of tillage system. mineralized N in the soil during the second season, both
Table 31. Earthworm activity in fallow and cropped plots taken during a 4-day period In July, 1980 (IITA).
Hyperiodillus (columnar cast) Eudrillus (granular cast)
Treatment g/250 cm2 No./250 cm2 g/250 cm2 No./250 cm2
Natural bush 10.46 a 12.98 a 0.01 a 7.66 a
Guinea grass 7.83 ab 10.02 ab 0.04 a 3.22 a
Leucaena 3.98 bc 6.85 abc 2.36 a 29.30 a
Maize/Cassava 2.15 c 5.78 abc 0.01 a 8.48 a
Maize + residue 0.60 c 1.79 c 0.02 a 2.35 a
Maize residue 0.004 c 0.02 c 0.00 a 2.67 a
22




Table 32. Changes in soil properties of a forest Alfisol 1 year after forest clearing (0-10 cm) (IITA, 1979-80).
NOF3-N Property and At clearing After one year #
standard error (1979) (1980)
. Bulk Density, g/cm 1.04 0.05 1.22 0.02
Organic C,% 1.55 0.07 1.24 0.06
,
Total N, % 0.171 0.007 0.146 0.007
Organic P, ppm 1947 1607
4- ,' #Crop residues were removed after each harvest.
Feb Mar Apr.I My Jue July Se Oct Nov
mo,. 98 Table 33. Average grain and dry matter yields of
Fig. 19. Seasonal fluctuation of nitrate and ammonium soybean and maize from 32 subplots (IITA,
-N in a newly cleared forest Alfisol (0-10 cm). 1979-80).
Grain Yield Dry Matter
(12% moisture) (Oven Dry)
the N03-N and NH4-N content in the soil were low, indi- Crop kg/ha kg/ha
cating that a higher rate of N fertilization would be required for the second maize crop. There were significant Soybean (TGM 294)
changes in soil organic C, N and P as well as bulk density (2nd season, 1979) 1759 5 4605202# 1 year after clearing (Table 32). The decline of soil or- Maize (TZPB) 2639 183 4709 20"
ganic C, N and P is primarily due to mineralization as soil #Total dry matter yield sampled at maturing stage.
erosion and surface run-off on this carefully cleared plot **Stover yield only at harvest.
(64 subplots) are minimal. The second season maize,
however, suffered severe streak virus disease.
SGrain and dry matter yields of soybean and first season Continuous fertilization and cropping
maize are given in Table 33. The grain yield of the first Continuous fertilization and cropping
season maize was considerably lower than would be pre- The long-term fertility trials initiated in 1972 on an Alfisol dicted from its early growth condition. This is prob- (Egbeda Series, Oxic Paleustalf) and Entisol (Apomu Seably because of the long dry spell during the grain-filling ries, Psammentic Usthorthent) to investigate the long stage that severely affected the grain yield of the 1980 term N, P, K, Mg, S and Zn responses of these soils folfirst season crop without supplementary irrigation, lowing land clearing were continued in 1980. Maize
LSD (5%) LSD (5%)
I I I I I I Il I II J I I I I
8-0
7.0
6'60 N2 P2 K2
S50 N2P2K1
5.0N N2 P2 KI
5- 4"0-- N 2 P2 K2
0. N2N2 P2Ko
SNP2K0 2 P2 K0
S 30W 20 EGBEDA SOIL APOMU SOIL
N(OXIC PALEUSTALF). (PSAMMENTIC USTHORTHENT)
1.0"
0
72 73 7i4 75 76 77 78 79 86 72 73 74 75 76 77 78 79 806
YEAR YEAR
Fig. 20. Potassium response of maize grown on Egbeda and Apomu soils in long-term fertility trials at Ibadan, Nigeria (K-rates; K, = 0; K, = 40 and K, = 80 kg K/ha).
23




(main season crop) followed by cowpea (minor season Data on the soil K-status with K application and cropping crop) annual rotation was practiced during the last 6 showed significant changes on both soil types (Fig. 21). years. Though P and N responses on the sandy loam The soil K-status showed faster depletion with continuEgbeda soil were observed, respectively, since the first ous cropping compared to a bare uncultivated treatand third years following land clearing, a significant re- ment, particularly on the Egbeda soil. Annual application sponse to K application on the maize crop was not ob- of 80 kg K/ha was able to maintain the soil K-status of the served until 1980, the ninth year after land clearing (Fig. Egbeda soil at a level comparable to that observed under 20). bush fallow. However, on the Apomu soil, the K-status
The significant response to an annual application of 40 was observed to be lower than that observed under bush kg K/ha was rather unexpected since the soil K-status fallow even with an annual application of 80 kg K/ha. with this treatment at the start of the trial in 1980 was The effect of maize crop residue retention and removal more than adequate at 0.32 me K/100g soil (Fig. 21). This with and without fertilizer application in this long-term may result from an early drought affecting the crop, trial, which initially did not show any particular trend on which may have less effect on the crop receiving the the maize grain yield in the last few years began to result higher K application. in definite effects (Fig. 22). On the Egbeda soil, removal
On the sandy textured Apomu soil, where significant K of the maize residue, particularly without fertilizer appliresponse was observed already in the fourth cropping cation, significantly reduced maize grain yield. On the year, the main yield response was observed to be consis- Apomu soil, removal of maize residue significantly retently higher at 40 kg K/ha than 80 kg K/ha. Though no duced maize grain yield, particularly with fertilizer applidefinite explanation can be given for this observation, it cation. may in part be due to the higher acidity build up with the The detrimental effect of continuous removal of maize higher K rate, which could have an indirect negative ef- -crop residue with continuous fertilizing and cropping fect on the maize crop. was observed to be more pronounced with the cowpea
crop. In 1980, the cowpea variety used in the experiment
was changed from VITA-4 to VITA-5. VITA-5 gave a lower
yield (Fig. 23) and is apparently more sensitive to Mn
0. Toxicity. With continuous application of NPK fertilizers,
There was a significant soil pH depression with both the 2! o Egbeda and Apomu soils. The lower pH resulted in
S .... 0 higher mobilization and extractable Mn levels, particu0 larly on the Apomu soil, resulting in lower cowpea yields
P _- mainly due to Mn toxicity combined with K deficiency.
0-2-. 0 0V
CD 03
2? Plant residue management
2t Despite the importance of maintaining adequate soil orE -. ganic matter under the traditional techniques of shifting
S .cultivation or bush-fallowing, the plant residue during
, o01: seedbed preparation is commonly burnt off. The longterm effect of this practice compared to plant residue
management has not been adequately studied. A trial
z was, therefore, initiated on an Alfisol at Ikenne, Nigeria,
I EGBEDA SOL 0 to study the effects of burning and mulching of plant
o residue in relation to fertilizer application on the perfor, ,mance of maize. Results of the trials during 1978-80 are 0-. shown in Table 34. Though some responses to N and P
S ...... were observed following land clearing from fallow, sig- nificant responses to P and N were only observed, re04 spectively, in the second and third years following
Clearing. The main response was to P application. 0-3 Though the mean maize grain yields do not show any
0 significant differences between burning and mulching of
2!! the plant residue during the 3 cropping years, some ef02 fects were observed among the various fertilizer treatlments. Without fertilizer application, PK, NP and NPK
treatments combined with mulching gave a higher maize
0.1 yield, particularly in the third cropping year. However,
O 40 80 BARE BUSH with the NK treatment, burning gave a higher maize yield,
K-RATE, KG/HA SOIL FALLOW which may be attributed to a quick release of P from
burning of the plant residue.
Fig. 21. Changes in Soil potassium status of Egbeda Nitrogen-tillage interaction on maize (Oxic Paleustalf) and Apomu (Psammentic Usthorthent)
soils in long-term fertility trials at Ibadan, Nigeria, with Despite the importance of no-tillage systems in soil conK application (i = 0; K, = 40 and K, = 80 kg K/ha) and servation, little information is available on the best nuwithout cultivation. trient management aspects of the no-tillage system of
24




LSD (5I) LSD (5%)
III III
N2 P2 K2 R
N 2 P2 I, R
N2 P2 KI- N2 P2 K
No P6 Ko + R
No Po Ko-R P0
SEGBEDA SOIL APOMU SOIL No Po Ko R
I p p I I I
_7 2 73 74 75 76 77 78 79 80 2 73 74 75 76 77 78 79 80
YEAR YEAR
1" 22 Ftfect of continuous removal of maize crop residue and fertilizer application on maize grain yield in longoii s pial on Egbeda (Oxic Paleustalf) and Apomu (Psammentic Usthorthent) soils.
LSD (5%) LSD (50/%)
600
10 2 22 K % R
20 N2 P2 K R 4N2 P2KI- R
:*30 N2 P2K4+ R. No Po Ko R
SNoPoKo R No Po Ko +N R
600
"\No Po Ko- R
'400
200 EGBEDA SOIL POMU SOIL
01
76 77 78 79 80 76 77 78 79 80
YEAR YEAR
Fq 23. Effect of continuous removal of maize crop residue and fertilizer application on grain yield of cowpeas
9 ongqterm trial on Egbeda (Oxic Paleustalf) and Apomu (Psammentic Usthorthent) soils.
25




Table 34. Effect of plant residue management and fer- maize production. Most of the fertilizer trials carried out
tilizer application on main season grain yield in the tropics have dealt with conventional tillage. Studof maize variety TZPB grown on Alfisol (Oxic ies were, therefore, carried out on an Alfisol (Alagba sePaleustalf), Ikenne, 1980. ries, Oxic Paleustalf) at Ikenne and on an Entisol (Apomu
1978 1979 1980 series, Psammentic Usthorthent) at Ogbomosho, Nigeria, to determine the N requirements for maize producFertilizer M* B*" M B M B tion under conventional and no tillage. At both locations,
treatment kg/ha maize yields from the control plots without N application
Control 4228 3706 2864 2675 2841 3055 were higher with conventional tillage than no tillage. this
PK 5069 4830 4493 4386 4341 4080 was also observed during the first cropping year in 1979
NK 4939 4812 3016 3385 3243 3601 (Fig. 24). However, maize yields with N application, parNP 5205 4720 4360 4526 5378 4857 ticularly at high N rates, were higher with no tillage than
NPK 5327 5275 4454 4448 5289 4936 conventional tillage.
NPK Mg Zn 5838 5824 4316 4509 4968 4724
Mean 5101 4861 3852 3988 4343 4208 Planted fallow as an alternative N source
LSD (5%) (1)** 835 612 567
(11) 822 536 802 Use of Leucaena prunings. In looking at low cost alter(111) 738 764 1263 natives to N sources, experiments were carried out using
Leucaena leuocephala top prunings. Leucaena not only
*M= Plant residue applied as mulch. can serve as a potential N source but also can supply
**B = Plant residue burnt before each cropping. fuel wood. In a trial carried out on an Apomu soil series
***LSD () = Between plant residue management at IITA, the effects of rate and placement methods of the
means. Leucaena top prunings were compared to urea. LeuLSD (11) = Between fertilizer treatments within resi- caena top prunings were banded or broadcasted once at
due management. maize planting, while urea was banded twice, one-third
LSD (111) = Between any two fertilizer treatments with N at maize planting and two-thirds N at 4 weeks after
different residue management. maize planting. Figure 25 shows that banding the Leucaena top prunings at 25 cm widths was most effective, and banding at rates of 5t or 10t was, respectively, 52 and 67 percent as effective as urea. Though the prunings can be used as an N source, application at planting time only was apparently not too effective.
LSD (.05) LSD (.05)
5.0.
0
NT
'o NT
< 4.0 CT NT
Y- CT
>"3.0
z
IKENNE OGBOMOSHO
J OXIC PALEUSTALF PSAMMENTIC
XJ O
N 1980 USTORTHENT
1980
S2-0
b 30 60 90 .120 150 0 30 60 90 120 150
N-RATE, KG N/HA
Fig. 24. Effect of tillage and N rates on maize grain yield.
26




2000
2000 plication at the rate of 60 kg K/ha. N application only
LSO (5%) slightly increased tuber yield. TMS 30572 also gave
higher tuber yield than TMS 30555 at this particular location. There is no difference in the percentage dry matIter yield between the 2 varieties.
Differential P response of cowpea and soybean varieties. Investigations were carried out at Ikenne -in the A humid region of Nigeria to determine differences in the
- internal and external P requirements of 4 cowpea varinz
I000:
-0-- Table 35. Effect of fertilizer and preemergence herbicide on maize grain yield (IITA, 1980).
ABkg/ha
c Full basal & side dressing with Primextra 3333 a.
Basal P only and side dressing with Primextra 2970 a b No fertilizer and No Primextra 2727 b c
Side dressing alone-No Primextra 2545 b c Full basal & side dressing without Primextra 2303 c Basal P only & side dressing without
SPrimextra 2182 c d
0 o50 K s T5ONS 0 TONS No Fertilizer and with Primextra 1727 d
N. RATE LEUCAENA RATEHA Side Dressing alone with Primextra 1091 e
KG N/HA Desn
*Means having common letters are not significantly difFig. 25. Effect of N and Leucaena rates and application ferent at 5% level according to Duncan's Test.
methods (A and B Leucaena tops banded at, respectively, 25 and 50 cm width and C Leucaena tops broadScasted) on maize grain yield on Apomu soil series Table 36. Effect of fertilizer and preemergence herbi(Psammentic Usthorthent). cide on weed weight at maize harvest (IITA,
1980).
In-situ mulch from cover crops and fertilizer levels on kg/ha
maize yield. In-situ mulch from cover crop residue has No fertilizer with Primextra 1803 a' been found to be a feasible method of establishing Side dressing alone with Primextra 1741 a mulch for effective no tillage or mulch-conventional till- Basal P only and side dressing with Primextra 1437 a age production. Using Mucuna utilis as the cover crop, Full basal & side dressing-No Primextra 1144 a
the effect of fertilizer and preemergence herbicide on Basal P only and side dressing alone-No
maize yield in mulch-conventional tillage was evaluated. Primextra 1112 a
The design was a split plot in which main plots were (a) Side dressing alone-No Primextra 1061 a no herbicide and (b) a basal application of 60 kg N/ha, 60 No fertilizer and without Primextra 1032 a kg P/ha and 60 kg K/ha with a side dressing of 30 kg N/ Full basal & side dressing with Primextra 877 a
ha at 4 weeks after planting; (ii) a basal application of 60 Means having common letters are not significantly difkg P/ha with a side dressing of 30 kg N/ha; (iii) a side *Means having common letters are not significantly difdressing of 30 kg N/ha only and (iv) no fertilizer. The ferent at 5% level according to Duncan's Test.
cover crop was killed with 2.0 kg glyphosate/ha a.i. Side
dressing alone and no fertilizer were found to produce of N and P application on tuber yd
significantly lower yields than a full application of basal Table 37. Effect of N and P application on tuber yield with a side dressing and a basal application of P with a of cassava cultivars TMS 30555 and TMS side dressing (Table 35). The preemergence herbicide 30572 on an Alfisol (Oxic Paleustalf), Ikenne,
had no significant effect on yield. The results indicate 1980.
that some fertilizer is needed in mulch-conventional till- Fertilizer Treatment TMS 30555 TMS 30572
age with in-situ mulch from a leguminous cover crop. A kg/ha Tuber yield t/ha
preemergence herbicide is not necessary as weed N K Fresh Dry Fresh Dry
suppression by the mulch is an effective weed control
(Table 36). 0 0 19.52 8.42 19.50 8.02
0 120 21.48 9.42 23.85 9.51
60 120 23.29 9.39 21.14 9.07
120 120 21.35 9.60 25.42 9.95
Nutrient requirement of crops on Alfisols 120 0 18.29 8.47 21.83 9.79
N and K responses of cassava. The N and K response 120 60 21.56 9.78 26.92 11.68 Study of cassava conducted on an Alagba series (Oxic Mean 20.92 9.18 23.11 9.67
Paleustalf) was initiated in 1978 at Ikenne. The 1979-1980 LSD (5%) Variety means fresh tuber 3.08; dry tuber 1.85.
cropping season results are shown in Table 37. On this Between fertilizer treatments within variety:
plot, which is in the second cycle of continuous crop- fresh tuber 3.09; dry tuber 2.11.
ping, the fresh tuber yields of both varieties, TMS 30555 Between fertilizer treatments among variety
and TMS 30572, showed a significant response to K ap- fresh tuber 4.37; dry tuber 3.74
27




eties with different growing habits. The relationship be- eties have similarly low external P requirements of about tween adjusted P concentrations in the soil at planting .018 ppm P in soil solution. These low external P requiretime and cowpea yields is shown in Figure 26. The 4 ments may be attributed to other growth-limiting factors varieties showed distinct differences in their external P as indicated by grain yields. In spite of the low yield, data requirement. Without P, a local variety, Shaki, gave the in Fig. 27 indicate the relatively higher efficiency of P highest yield and has an external P requirement of about utilization by variety TGM 51 x TGM 344 over variety
0.06 ppm P while with P application, TVx 1193-7D, an TGM 479. improved erect variety, gave the highest yield and has the lowest external P requirement of about 0.016 ppm p. Management of siliceous Ultisols Ife Brown and VITA-4 gave the lowest yields, and Ife Brown has the highest external P requirement of about
0.10 ppm P while VITA-4 has an external P requirement Maize-cowpea rotation of about 0.06 ppm P. From these results, it thus appears A 5-year management trial at Onne shows that a producthat though variety TVx 1193-7D is quite efficient in utiliz- tive maize-cowpea rotation system is possible on coarseing applied P in low P status soil, the local variety Shaki textured deep and permeable Ultisols (Typic Paleudult) does better. on a predominantly flat to gently undulating coastal land
The differential phosphate requirements of 2 soybean form. Maize was sown in early March and cowpea in late varieties, TGM 51 x TGM 344 and TGM 479, were com- September or early October. Maximum grain yield pared in a field experiment conducted on an Alfisol at throughout the 5 years ranges from 3.5 to 4.5 t/ha for Mokwa in the subhumid region of Nigeria. Results of the maize and from 1.3 to 1.5 t/ha for cowpea. trial as shown in Fig. 27 clearly indicate that variety TGM A well-balanced fertilization scheme is required for the 479 has a significantly lower yield than variety TGM 51 maize crop, N, P, K, S. Mg and Zn, and fertilization is x TGM 344. Despite their differences in yield, both vari- generally not needed for cowpea in the second season as the residual fertility from the maize crop with residue return as surface mulch is adequate to support the cowpea crop. Reduced tillage with residue mulch is recomvx IM-70 mended because the coarse-textured, kaolin-dominated
,oo o (sssoil may be easily compacted.
It is important to point out that the maize, TZPB, and VA .cowpea cultivars, VITA-1 and VITA-4, tested are fairly tolerant to soil acidity. The critical level of exchangeable Al 10 saturation, i.e., a level required to attain 90 percent of
IFE maximum yield, for the maize cultivar ranges between
< I0 30-45 percent depending upon the rate of chemical ferSI I tilizer used (Fig. 28). The critical level of AI saturation for
60 .TRIAL At IKENNE
OXIC PALEUSTALF
MINOR SEASON 1980
00 95. OF 100 X-.
- MAXIMUM I -.
200 y ,YIELD '_ _' _ ,_ 90"
0Os oI O so 1
ADAIUSTED P CONCENTRATION AT PLANTING. ppm 80
60\ 1
Fig. 26. Relationships between adjusted P concentra. \ \
tions and cowpea yield.. 70\
50kg K
_ 60 (1980)
150kg K
N 50 (1979)
140 TOM SI (X) TGM 3
A W 40 Maize (TZPB)
Max Yield
4-30- 398 tlha 1979 ist season
o000 4.56 tihq 1980 istseason
TM 47920 S800-0 0 0
TRIAL At MOKWA
600- OXIC PALEUSTALF 0
1980
/95% OF MAXIMUM 980
s005 0 0 0 10 20 30 40 60' 70 90
ADJUSTED P CONCENTRATION Al PLANTING, ppm EXCH. AL SATURATION, %
Fig. 27. Relationships between adjusted P concentra- Fig. 28. Relative yield of maize as affected by percent tions and soybean yield. exchangeable AI saturation in the soil.
28




both cowpea cultivars is as high as 60 percent when Residual effect of lime chemical fertilizer is not applied in the second season (Fig. 28). But, with moderate rate of fertilization, the crit- Because lime is scarce in many parts of the humid tropical level of Al saturation is reduced considerably, and ics such as southeastern Nigeria, it is necessary to know the 2 cowpea cultivars show differential response to soil what minimum rates of lime are needed for optimum acidity. i.e., 30 percent for VITA-4 and 43 percent for crop production and what residual values may be exVITA-1 (Fig. 29). pected. Results from an Onne trial are summarized as
follows:
Relatively low rates of lime would be adequate to sustain 100 crop yields under a maize/cowpea rotation system. It is
interesting to point out that with balanced fertilization, 90" respectable levels of maize and cowpea yields were
Vita-I maintained for 5 years in the unlimed plots (pH 4.3)
80. where no severe A toxicity effects were observed. Both
80 toict
cowpea cultivars nodulated well with indigeneous acid70" Fertilized tolerant rhizobia, and nodulation was only slightly re1979, 2nd Season duced without liming.
Vita-I 1.28Btlha Applied Ca in the form of lime leached readily from the
60- Vita-4 1-50 tlha Vta surface layer, which is accompanied by the subsequent
reappearance of exchangeable Al (Table 39). The in50 crease in CEC values due to liming is also short-lived.
The downward movement of Ca has little effect on the 100- subsoil pH and exchangeable Al, suggesting that Ca
Bleached in the form of neutral salts. The vertical distribution of exchangeable Ca in the field profile measured 90- 3 years after lime application follows the theory and.formulation of ion-exchange chromatography (Fig. 30). This 60- No Fertilizer is in agreement with a previous laboratory leaching study
1980 2nd Season using undisturbed soil columns. Over 90 percent of the
70- Vita -t 1-48 tlha applied Ca may be found between 0-90 cm depths 3
Vita-4 1.36 1lha years after liming because of the strong subsoil acidity
(pH 4.3 and greater than 50 percent Al saturation). Re60 covery of the subsoil Ca, however, would require deep10 20 i0 40 50 60 7b 90 rooting species tolerant to high exchangeable Al levels.
EXCH. AL SATURATION,% Moreover, because a high lime rate in the surface soil
has little effect on the subsoil acidity, there would be no Fig. 29. Relative yield of cowpea as affected by percent advantage of applying a rate of lime more than required exchangeable Al saturation in the soil. to reach the exchangeable Al levels for the crop to be
Thus, liming is not necessary unless the exchangeable Table 38. Concentration of Al in soil solution of a Ultisol Al saturation of the soil has reached a level beyond the (Typic Paleudult) as affected by fertilization critical level of the crop to be grown. The critical level of (Onne, 1980). exchangeable Al saturation in such coarse-textured, ka- Al in
olinitic Ultisols depends to some extent upon the rate of Exch. At saturation
fertilizer to be applied. The variability in the critical level Soil pH saturation extract
of exchangeable Al saturation is because of the large (H20) % ug/ml
increase in soluble Al in the soil solution after fertilization. Such effects have not been taken into account be- Unfertilized 4.3 60 2.7
cause soil sampling is normally done before planting. Fertilized and incubated The effect of fertilizer salts on soil solution Al is shown 6 weeks at field in Table 38. capacity 3.9 46 17.1
Table 39. Changes in soil properties in the surface layer (0-15 cm) 5 years after lime application (IITA Onne substation, 1980).
Initial lime pH (H20) Exch. Ca, meq/100g Exch. A, meq/100g ECEC, meq/100g
rate (1976)
t/ha. Initial # 5 Yrs. Initial 5 Yrs. Initial 5 Yrs. Initial 5 Yrs.
0 4.7 4.2 0.57 0.27 1.26 1.51 2.77 2.47
0.5 5.0 4.3 1.37 0.31 0.82 1.43 3.08 2.40
1 5.2 4.4 1.58 0.42 0.45 1.27 2.77 2.36
2 5.6 4.5 2.36 0.61 0.06 1.01 3.10 2.28
4 6.3 4.8 5.71 1.03 0.04 0.40 6.53 2.12
#Initial soil samples taken one month after lime application.
29




EXCH. CA, meq/100g grown. Thus, it is recommended that for Ultisols (Typi
0 0.5 1-0 1.5 2.0 2.5 3.0 3.5 Paleudult) or for Ultisols with similar mineralogy and tex
0-15 ture, annual rates of 200-400 kg/ha should suffice t(
maintain maize and cowpea yield. Under such circum stances, lime could be regarded as a form of fertilizer
E 15-30. rather than a major soil amendment.
" 30-45- Initial Lime Rate (1976) Leaching of nitrate under maize.
/An experiment was established at Onne in 1980 to study
S-- 2 tlh the pattern of nitrate leaching under field conditions,
S5"60 --- Z4 t/ha Maize was planted in the first season and upland rice in
- jthe second season. Three methods of N application were
X 60-75 used, 1 application at planting, 2 split applications: one0. half at planting and one-half at 4 weeks after planting
and 3 split applications: one-third at planting, one-third 75-901 at 4 weeks after planting and one-third at 8 weeks after
planting. A N rate of 150 kg/ha as calcium amonium nitrate was used. A bare fallow treatment was included. EXCH. AL, meq/100g The experiment was a split-plot design with 4 replica0-150 0.5 0 1. 2.0 25 30 3 tions. Lime and unlimed treatments were the main plot.
Downward movement of inorganic N (NH4-N and NO-N) 150 in the soil was monitored periodically up to a 120 cm
15-30 depth (Fig. 31). At the end of 4 weeks (129 mm rainfall),
E the pattern of nitrate movement when all N was applied
Ul
. 30-45 at planting in the unlimed plots was similar to the bare
S3-fallow and the crop plots. The peak concentration oc2: curred between 30-60 cm. At the end of 8 weeks (477 mm
45-60 rainfall), the nitrate peak in the cropped plots occurred
z at a depth of 60-90 cm, but the nitrate peak in the bare
fallow plots was considerably deeper and broader, indi60-75- cating greater leaching. At the end of the first season
W (940 mm rainfall), the nitrate peak moved to a depth of
0 1105-120 cm. Liming increased the rate of nitrate leaching 75-90. in both bare fallow and cropped plots. When N was split
into 2 applications, nitrate peaks occurred in the upper Fig. 30. Vertical distributions of exchangeable Ca and Al layer between 0-30 cm both at 4 and 8 weeks after plantat 3 years after lime application. ing, indicating less leaching.
Splitting N into 2 applications significantly increased grain yield and N-uptake by the plant (Table 40), but furN03-N in Soil,jg/g
0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25
X-- 00o 0
15 ii t \ ,,
15 0 "
-U.S I,
60 J > / t>
U I I I '
45. 0 X 0
o O / C
E 60. ,
,os5 0) I o,',,. '" o~ 0 '.
NYC > I/t
90 > t
I
Bore Follow, No Bare Follow, N, Maize, N, Maize ,N2
Fig. 31. Leaching of nitrate in unlimed plots under cropping and bare fallow. (No- no N applied, N,-150 kg N/ha applied
at planting, N,-150 kg N/ha split application, one-half at planting, one-half at 4 weeks after planting.)
30




Table 40. Maize grain yield and N recovery as affected by methods of application (IITA Onne substation, 1980).
Estimated fertilizer
N removed by crop N in soil (0-120 cm) Estimated recovery of Grain yield (grain + stover) at harvest + applied N**
Treatment# kgiha kg/ha kg/ha %
LoN, (Bare) 70.6 47.1
LoN, 2,492 c 54.5 d 58.9 75.5
LoN2 3,208 ab 73.7 bc 43.0 78.5
LoN, 3,376 ab 82.7 ab 55.3 92.0
L N, 2.853 bc 61.4 cd 32.9 62.5
L N, 3,499 a 81.6 ab 32.6 76.3
L N, 3,717 a 92.4 a 59.1 100.8
#Lo = unlimed: L = limed, 2t/ha; N rate 150 kg/ha; N, = one application at planting; N2 = 2 splits; N3 3 splits.
+ + Corrected for mineral N (NH4-N and NO3-N) in soil (0-120 cm) from unfertilized bare fallow plot (190.2 kg/ha).
**Crop removal plus estimated fertilizer N in soil (0-120 cm) at harvest.
ther splitting into 3 applications had no significant effect the Ultisol (Typic Paleudult) from Onne was completed on yield though it reduced leaching loss. in early 1980. The primary objectives are to study leachEstimated percentage recovery of applied N is also given ing of nutrients under high-rainfall conditions and deterin Table 40. Loss of applied N in the form of calcium mine crop water use. The high-rainfall condition, which ammonium nitrate under cropping was not as high as prevails at Onne, was simulated during the experimental expected although more reliable measurements on N re- period by supplementing natural rainfall with irrigation cover can only be obtained with N-i5 tagged fertilizers, using the 1979 rainfall regime at Onne as a basis. Nutrient losses over the 2 cropping seasons were deterWhen N was applied in 3 split applications, nearly all the mined by measurements of fertilizer ions in the leachate applied N could be accounted for by plant uptake and collected as samples of the drainage through the profile, the amount present in the soil (0-120 cm) at the end of and evaporation/evapotranspiration was assessed by the
* season. At 2 split applications, about 80 percent of the water balance method.
applied N was recovered in plant and soil.
Maize was planted as a test crop on 3 of the lysimeters It is important to point out that the unfertilized bare fal- and in the surrounding area to maintain adequate fetch low plot contained 60-80 kg/ha of mineralized N (NH4-N while the remaining 3 were left bare. All except 1 of the plus N03-N) in the surface 30 cm layer for both the un- lysimeters was tension drained by applying a suction of limed and limed treatments, suggesting relatively high 0.2 atmosphere at the base of the column. Fertilizers rates of mineralization during the early part of the crop- were applied in each of the 2 seasons to all 6 lysimeters
ping season. It appears that application of N fertilizer as follows:
may not be necessary at the time of planting during the N (Calcium nitrate): 150 kg/ha, 3 split applications
first season if sufficient mineralized N is already present P (Monocalcium phosphate): 60 kg/ha, basal
in the surface layer. K (Potassium chloride): 150 kg/ha, basal
MG (Magnesium sulfate): 30 kg/ha, basal
Lysimeter studies No lime was applied during this preliminary run. As a
result, there was poor growth of maize inside the lysiThe installation at ITA of 6 monolith lysimeters (80 cm in meters due to strong soil acidity. On the basis of the diameter and 130 cm deep) with undisturbed profiles of preliminary results, total drainage through the seasons averaged 1,153 mm or 58 percent of the total equivalent rainfall (2,012 mm) received by the bare (uncropped) lysimeters and 977 mm or 49 percent of the total equivalent rainfall received by the cropped lysimeters. The corresponding values of evaporation and evapotranspiration were 832 mm and 1,035 mm, respectively, averaging 3.70 mm/day and 4.60 mm/day, respectively, over the period. Preliminary leaching data of nutrient cations and anions calculated as percentage of total amount applied are given in Table 41.
These data indicated that nitrate and magnesium ions leached readily; whereas, potassium ions, were preferentially retained. Further studies are being conducted under maize and upland rice cropping where adequate amounts of lime have been applied to ensure normal crop growth.
i& =i' ; :. *, K and Mg response of cassava
Monolith lysimeter cutting and installation at IITA Onne Investigations on the K and Mg response of cassava vasubstation. rieties TIMS 30395 and 30211 were initiated on an Ultisol
31




Table 41. Percentage of applied cations and anions in mixed cropping systems. As a follow up on previous
the leachate after 2,012 mm of water under studies showing or establishing a range of modifications bare fallow (uncropped, weed free) IITA, that may be brought about in the light regime in maize 1980). canopies through modifications of planting geometry
Tension drained and density (IITA Annual Reports 1978 and 1979), studies Free drained (0.2 Atm) were carried out in 1980 with the aim of quantifying and
lons Lysimeter 1 lysimeter 6 standardizing some of these relationships partly to provide a basis for optimal design in mixed cropping systems. Attempts were also made to relate the effects NO,- 78 70 of the modified light climates within the established
C" 35 30 crop (maize) on the lower or slower growing intercrop
Mg 64 60 (cassava).
K, 9 9
Results show that at full development of the maize canopy in a maize/cassava mixture, the percentage of inciat Onne in 1978. Results of the second year cropping dent radiation depleted at cob height is a power function during the 1979-1980 season are shown in Table 42. In of the combined plant population. The relevant equation the second year of cropping, the soil K-status even with is as follows: K application was very low. There was a distinct effect of Y., = 1.735Xos8t r = .917"* Mg application in increasing the soil Mg status. The tuber yields of both varieties showed a more pronounced where Y., is the amount of light intercepted (to cob level), and significant response to K application. Despite the and X is the combined plant population (Fig. 32A). A presence of Mg deficiency symptoms without Mg appli- similar relationship was obtained at ground level (Fig. cations, similar to the first year results, there was no 32B). definite response of tuber yield to Mg application. As previously reported (IITA Annual Report 1979), plant populations in this experiment were varied from 10,000 Cropping systems plants/ha in the pure cassava to 80,000 plants/ha in the
maize/cassava mixtures. The maize (TZPB) population In 1980, research in cropping systems or crop manage- varied from 10,000 to 70,000 plants/ha by increasing the ment focused on the following areas: mixed cropping, number of maize plants per stand from 1 to 7. Spacing alley cropping, live mulch system and the role of agro- was maintained at 1 m x 1 m with the plants along the forestry in food crop systems. The ultimate goals are to same row. achieve high and stable crop yields while maintaining For the lower or slower growing cassava, which was long-range soil productivity. Several projects were initi- shaded through much of the growth of the maize, the ated to investigate the inclusion of leguminous cover crop yields were very significantly affected by the crops and managed tree and shrubs into food crop pro- amount of light reaching it through the maize canopy duction systems in an attempt to find more efficient, low- (Fig. 33). Again, a similarly linear relationship was obenergy input and stable alternative systems to traditional tained between the cassava yield and the total light bush fallow cultivation. Special emphasis was given to transmitted through the combined maize/cassava canthe production system of plantain in the humid and per- opy at full maize development (Fig. 34). These amounts humid regions. of transmitted light can also be systematically related to
the combined plant population as shown in Fig. 35. The Intercropping agronomy and relationship between the 2 factors is expressed:
meso/micro-climatic studies Y = 82.198e 001sx r = -0.903*
Light regime and productivity in mixed crops. Competi- where Y. is the amount of light transmitted through the tion for light and moisture are clearly 2 key factors in canopy, and X is the combined plant population.
Table 42. Effect of K and Mg application on tuber yield of cassava cultivars TMS 30395 and TMS 30211 and K and
Mg status of an Ultisol (Typic Paleudult), Onne, 1980.
Fertilizer Treatment Soil K and Mg
status TMS 30395 TMS30211
kg/ha Me/i OOg Tuber yield t/ha
K Mg K Mg Fresh Dry Fresh Dry
0 40 0.09 0.33 15.84 6.47 14.99 5.66
30 40 0.09 0.27 19.41 8.08 18.67 6.11
60 40 0.11 0.33 19.97 8.84 18.20 6.36
120 40 0.10 0.30 20.13 8.52 19.08 6.73
120 0 0.11 0.22 21.47 9.39 17.33 6.69
120 20 0.11 0.22 19.47 8.06 17.50 5.97
Mean 19.38 8.23 17.63 6.25
LSD (5%) Variety means fresh tuber 0.63; Dry tuber 0.78.
Between fertilizer treatments within variety: fresh tuber 3.48; dry tuber 1.83.
Between fertilizer treatments among variety: fresh tuber 5.29; dry tuber 2.88.
32




100 Y 8.846 + 0.397X
38 (r=889*)
90
x
Y 1735 X0851 34
80
so..
TO 30
60 X
S26 50 hi
> 22
1- 40
KU)
30 18.
20
14
10
10 20 O30 40 50 60 70 o80 90
PLANT POPULATION (inlOOO's) ,I0 20 30 40 50 O 80
Fig. 32A. Light intercepted by mixed maize/cassava crop PERCENT LIGHT TRANSMITTED (%)
canopy (to cob level) at full maize development (10 wap) as a function of plant population. Fig. 33. Cassava yield in a maize/cassava mixed crop as
affected by light transmission through the mixed crop canopy at maximum maize height.
Y 5 694+0'3617X
100 38 (r -0888 *)
Yg .5501X0 622 x
90 34
so 30
30.
To.
70 26
a a 26
60
3 22
- "I
6 V)
50 U x
1840
1441
3010
20
0to 20 30 40 50 60 70 80 90 PERCENT LIGHT TRANSMITTED(%)
10
0 FIg. 34. Cassava yield in a maize/cassava mixed crop as
S0o 2b SO b do o 7 s o So affected by light transmission through the upper maize
PLANT POPULATION (n OOO's) canopy at maximum maize height.
Fig. 32B. Light intercepted by mixed maize/cassava crop canopy (to ground level) a full maize development (10 wap) as a function of plant population.
33




80 ItoII0
60 I-/ /,
too.
0 L.
06.\Y z9i I-./.,,,X X
A z - ,Z MAIZE CASSAVA
650 80 goo4o. =3\ CASSAVA
-..
8o-: \
A N -A' .;MAIZE
2 2 w 70. .,
I- \o
z 60 .
0 'O 20 30 40 5o 60 7o so 9o oo PLANT POPULATION 2n100O's)
50,
Fig. 35. Light transmission through mixed cassaval maize crop canopy at full maize development as a function of plant populations.
6 12 18 24 30 32 Induced micro-climate and crop response. In a related 10/0 12/7 2/9 14/10 25/11 0/I
experiment in 1980, using cassava of relatively prolific WEEKS AFTER PLANTING
growth, plant populations were maintained constant in Fig. 36. Relative soil moisture content in different crop 3 treatments at 12,500 plants/ha (cassava: 3001) and cix ture ote in dieferiod of
Q 25,000 plants/ha (maize: TZPB). In a fourth treatment, the crop mixtures as a function of time during the period of
maize population was increased to 40,000 plants/ha, and growth
the cassava population was reduced to 10,000 plants/ha.
45.
Pure cassava with 12,500 plants/ha constituted a sixth
treatment, and pure maize at 25,000 and 50,000 plants/ MAIZE/CASSAVA
ha were treatments 5 and 7, respectively. 40. A- ... SOLE MAIZE
Analyses of the partial results show that the amount of SOLE CASSAVA
light intercepted by the upper canopy of the maize (to G35
cob level) is significantly related to yield. ""
Observations on soil moisture and temperature in the o /
experiment also show marked differences in both varn- ,. / ....
ables under the different crops and crop combinations.
Fig. 36 compares the 6 weekly average mean relative soil o25
moisture contents under the pure maize, pure cassava and maize/cassava combinations, respectively. Evidently, the presence of cassava in the mixture has a ben- Q 20
eficial effect on the moisture available to the maize while maize helps reduce the soil temperature as a result of J
rapid canopy development (Fig. 37). I 15
Within the mixed crops, appreciable differences were 0
observed with changes in planting geometry (Table 43). o i2 1'6 2o 28 32 36
Similarly, air temperature and humidity as reflected by WEEKS AFTER PLANTING
the wet bulb thermometer were also affected by the
planting pattern (IITA Annual Report, 1979). Fig. 37. Soil temperature in different crop/crop mixtures as a crop function of time during the period of growth.
Table 43. Comparative mean relative mixture content*. Partial mechanization of maize/cassava intercropping.
Yields of maize planted mechanically on ridges were not Treatment 1 2 4 affected by cassava planted on the same ridges at 0, 7,
Period* 1 2 3 1 2 3 1 -2 3 14 and 21 days after maize (IITA Annual Report, 1979).
Moisture content 72 67 102 94 85 123 83 70 112 Herbicide, however, depressed maize yield significantly *The moisture content is in % of value at FMC. at the 5 percent probability level. The effects on cassava *Period 1 = 2-6 WAP; Period 2 = 7-12 WAP; Period 3 show that delay of up to 21 days in planting cassava was
= 13-18 WAP. not critical to cassava yield (Table 44). However, signifi34




Table 44. Effect of relay times and herbicides on cas- cropping and highlight the potential hazard of herbicide
sava yield in maizelcassava intercropping. failure.
Cassava relay Herbicide No Herbicide Mean N response in maize/cassava intercropping. The N rein days M/C' C2 M/C C M/C C sponse of inlercropped maize/cassava on an Alagba soil
0 16.8 7.6 22.6 25.7 19.7 16.6 (Oxic paleustalf) at Ikenne that was initiated during the
7 22.4 12.2 22.9 27.0 22.7 19.6 1978/1979 season was repeated during the 1979/1980
14 16.7 13.9 23.5 22.3 20.1 18.1 season. During the second year, the maize crop was af21 13.0 9.1 23.1 20.6 18.0 14.8 fected by drought resulting in low yield (Table 45). DeMean 13.9 23.5 spite the low grain yields, the sole and intercropped
LSD between herbicide 2.7 maize showed significant responses to N application. As
Relay time 5.2 observed during the first year of cropping, intercropping
Relay time same herbicide 7.3 with cassava had no effect on maize grain yield. In casRelay time different herbicide 7.3 sava, high tuber yields were observed with later harvesting (13 months after planting), but no significant Maizelcassava response to N application was observed. This indicates
2 Sole cassava that the N requirement is lower for the cassava than the
associated maize. High N rates (120 kg N/ha)also tend to decrease the tuber yield of cassava.
cant cassava yield reduction was observed where herbicide was used. This yield reduction was ascribed not Effect of maize population on cassava yield. An experionly to the sensitivity to the herbicide but also to weed meant was conducted to determine effects of maize competition resulting from failure of the herbicide. This population on cassava development and yield in observation is substantiated by a tendency toward lower intercropping where maize is planted in hills or clusters. yields by herbicide treated pure cassava. Shading by It simulated farm situations where many seeds are maize appeared to have retarded weed growth, thus, placed in 1 position, and the number of surviving plants reducing the competitive effect on cassava. The results depends on chance or are deliberately selected by the confirm the compatibility of maize/cassava in inter- farmers. The maize was spaced at 100m x 100 m with
Table 45. Effect of N application on yield of intercropped maize (variety TMS 30395) grown on Alagba soil (Oxic paleustalf) (1979-80).
Cassava fresh
N-Rate tuber yield Maize grain
kg N/ha Cropping mixture tlha yield LER*
Maize (1 x .33 m,
0 1 plant/hill) 1.80
Maize (1 x 1 m,
3 plants/hill) 1.74
Cassava(1 x 1 m) 30.12
Maize (1 x .33 m) + cassava
(1 x 1 m) 29.11 1.93 1.90
Maize (1 x 1 m) + cassava
(1 x 1 m) 28.89 1.87 2.00
Maize (1 x .33 m,
60 1 plant/hill) 2.30
Maize (1 x 1 m,
3 plants/hill) 2.40
Cassava (1 x 1 m) 30.90
Maize (1 x .33 m) + cassava
(1 xl m) 27.10 2.45 1.95
Maize (1 x 1 m) + cassava
(1 x 1 m) 29.50 2.39 1.95
Maize (1 x .33 m,
120 1 plant/hill) 2.28
Maize (1 x 1 m,
c plants/hill) 2.12
Cassava (1 x 1 m) 28.65
Maize (1 x .33 m) + cassava
(1 x 1 m) 24.85 2.52 2.05
Maize (1 x 1 m) + cassava
LSD (1 x 1 m) 29.68 2.10 2.03
3.40 0.40
Maize crop slightly affected by drought.
Cassava harvested at 13 months.
Land equivalent ratio.
35




1-7 plants hill to give 10,000-70.000 plants/ha. Cassava Alley cropping
was planted at 100 m x 100 m in the same rows as maize but spaced 50 m from the maize along the same row to
give a population of 10,000 plants/ha. The land was Alley cropping systems in forest zone
plowed and harrowed but not ridged. The fertilizer rec- Effect of N application on maize/Leucaena alley cropommendation used was for maize and consisted of a ping. Results of a field experiment conducted at IlTA on basal application of N.P.K. 15:15:15 at 400 kg/ha and a the effect of N application in the maize/Leucaena leucoside dressing of 30 kg N;ha as urea 4 weeks after planting cephala alley cropping system are shown in Fig. 39. Subof maize. stantial Leucaena dry matter and N yields are produced
The cassava growth was significantly affected by maize with a total of 6 prunings. The total annual N yield of over
population with tuber yields being significantly lower at 180 kg N/ha was quite remarkable.
the higher maize populations. The marked differences During this cropping year, there was a significant effect between maize populations suggested that maize popu- of the low N rates applied to the maize on the Leucaena lations should not exceed 30,000 plants/ha if good cas- dry matter and N yields. There was, however, only a slight
sava yields are expected. The maize yield indicates no effect from the high N rate applied. serious yield reduction of maize population if kept at 30,000 plants;ha. The results also support the practice of The effect of N rates and removal of Leucaena prunings thinning maize to 3 plants hill in maize/cassava inter- on maize grain yield is shown in Figure 40. The results cropping in West Africa. clearly show that despite the high amount of N yield,
there was still a need for application of low N rates for Weed control in maize/cowpea intercropping. The effect obtaining high yields. Removal of the Leucaena prunings of early weed interference in maize/cowpea intercrops is reduced the yield to about 54 percent. Without N applimodulated by the growing season. During the 1980 first cation but with Leucaena prunings only, the total maize season, a maize/cowpea intercrop was more sensitive to weed interference than each of the crops grown in pure culture (Fig. 38.) However, during the 1980 second season, crop sensitivity to early weed interference peaked in both the sole and intercrop at 2 weeks after planting. This may be related to moisture conditions. Weed biomnis at the 2-week stage was higher in the second seasWhan the first season, thus, accounting for the more severe weed interference during the second season.
16 EARLY 1979
Sole Maize (40 x 103 pit/ho)
o-o Sole Cowpeo (50 x IO pit/ho) 14
SMoize/Cowpeo Intercrop
(30/40xO3 pit/ho)
12
10 o10 LATE 1979 Pod of Leucaena leucocephala, a tree type legume
S] being used for soil fertility restoration.
66 6 60
2
0 2 4 6 8 6
WEED INTERFERENCE DURATION (WAP).
Fig. 38. The effect of weed interference on food energy
values of maize/cowpea intercrop. .
Collecting leaves from Leucaena leucocephala. The dried leaves will be used as a source of plant nutrient and organic matter, the stems as stakes and firewood.
36




grain yield was maintained at about 3.5 t/ha for the 2 LSD(5%) seasons (total yield in 1979 was 3.6 t/ha.) The addition of 60the Leucaena prunings has a distinct effect on soil or- I MAIN SEASON ganic matter and N levels (Table 46.)
MINOR SEASON
Effect of legume species and alley width. Legumes es- 50- MINOR SEAON. tablished between maize did not reach sizes at which the ( quantity of leaves was regarded as sufficient to contribute significantly to soil nutrient levels. Alley width had no a 40significant effect on legume development during early stages. The leaf yield and potential N contributions are z shown in Table 47. The leaf and N yield is related to alley 3 owidth. Because of the low potential N contribution, the plot was fertilized with 60 kg N/ha, 20 kg P/ha and 30 kg K/ha after = 2 oa soil test. The plot was Rome plowed and maize planted with a single-row rolling injection planter in the 225 cm alley and with a 4-row rolling injection planter in the 10 wider alleys. The spacing was 75 cm x 25 cm. The legumes were 75 cm from the nearest maize. Maize yields calculated with a correction for land devoted to legumes ON ON 40N 8ON
are shown in Fig. 41. Neither alley width nor legume spe- LEUCAENA 3N 6N
cies had a significant effect. It would appear that it is too PRUNINGS early to observe the effects of the different treatments. N -RATE, KG/HA
Selection and evaluation of woody species for alley Fig. 40. Effect of N rates on maize grain yield from cropping systems. The usefulness of fast growing maize/Leucaena alley cropping system on Apomu soil woody and herbaceous legumes in alley cropping sys- series (Psammentic Usthorthent). (Main season N rates: tems is being evaluated on the Alfisols at IITA and on the 0, 40 and 80 kg N/ha; Minor season N rates: 0, 30 and 60 kg N/ha).
LSD (5%) LSD (5%)
MAIN SEASON MAIN SEASON 250
6.0- MINOR SEASON MINOR SEASON
5.0- 200
2.0
_150
LU3O0
2 IU00>50
1'O
ON 40+ 80+ ON 40+ 80+
30N 60N 30N 60N
DRY MATTER YIELD NITROGEN YIELD
Fig. 39. Leucaena dry matter yield and N yield from maize/cassava alley cropping on Apomu soil series (Psammentic Usthorthent). (Main season N rates: 0, 40 and 80 kg N/ha; minor season N rates: 0, 30 and 60 kg N/ha).
37




Table 46. Some characteristics of Apomu surface (0-15 cm) soil before and after alley cropping.
Treatment* Org. C Total N Extr. P IN Am. Acetate Extr.
kg N/ha pH-H,20 (%) (%) (ppm) K Ca. Mg. Na ECEC
me/100g
1976 (Before alley cropping)
6.2 0.98 0.12 24.7 0.25 2.63 1.02 0.06 4.40
End 1980 (with alley cropping) ON" 5.7 0.96 0.11 19.4 0.16 5.07 0.35
ON 5.7 1.47 0.12 21.5 0.16 5.33 0.43
30N 5.5 1.24 0.13 18.7 0.14 5.72 0.34
60N 5.3 1.41 0.13 22.8 0.18 4.43 0.28
LSD (5%) 0.1 0.67 0.01 3.3 0.07 1.18 0.08
*Major season 1980 N rates 40 and 80 kg N/ha; Minor season 1980 N rates 30 and 60 kg N/ha.
**Leucaena prunings removed from this treatment.
Table 47. Legume leaf yield and potential N yield in (2) Fast growing pulp, fuelwood and/or soil amelioratearly stages of alley cropping. ing species, such as Cordia alliodora, Albizia fa/Alley width Leaf yield N yield cataria, Cassia siamea, Sesbenia grandiflora,
Legume cm kg/ha dry wt. kg/ha Flemingia congesta, Samanea saman and Gmelina
arborea.
Pigeon pea 225 1321 47.6 (3) Important multipurpose trees and shrubs, such as
375 793 28.6 Treculia africana, Irvingia gabonensis, Dacryodes
675 440 15.9 edulis, Pterocarpus spp., Chrysophyllum albidum,
Leucaena 225 1084 45.5 Afzelia spp. Blighia sapida and Parkia clapperton375 651 27.3 iana.
675 361 15.2
Tephrosia 225 600 22.8
375 360 13.0
675 200 7.65O- T Tephrosio Condido
C Coanu.s Cajon /
L Leucoeno Leucocephalo .
G Glircidio Sepium
40
2"25 (m) alley width
20
W -J
30 .,. U 675(m)
z -.J
0
W
)~O
,.0
0 T C L G
Fig. 41. Effect of shrub legume species and alley widths on maize grain yield.
strongly acidic Ultisols at Onne. Emphasis is given to those species that are established easily and can be maintained through basal sprouts and coppices when cut back periodically. The screening and evaluation includes more than 30 woody specimens as listed below:
(1) Planted and/or retained woody fallow species, such as Acioa barteri, Anthonatha macrophylla, Alchornea cordifolia, Dalium guineense, Gliricidia
sepium, Leucaena leucocephala and Tephrosia Acioa baterii, one of the shrubs being tested for imcandida. proved bush fallow.
38




5-00
-., IBADAN I ITA
4-OO- ONNE
3-00
0*00
, 5 .j .9,U 20
Africa, associated in natural bush fallow for soil fertility of
restoration. IC)
The results of the first 2 years of screening and evaluation indicate that, in general, the establishment and Fig. 42a. Relative growth performance of selected tree growth performance of the shrub species were better at species at IITA and Onne after 18 months of establishIITA while the tree species were better at Onne (Figs. 42 ment. a & b). Some of the shrub species that performed well are being included in the alley cropping trials. Light regimes in potential tree-shrub species for alley 500- IBADAN IITA
cropping. Competition for light remains a key factor in i
the suitability of different species in mixed stands. Tests -- 4-00 ONNE
were made to determine the light characteristics within E alleys of different tree-shrub species established over the 3.00 past 2 years. The tree rows are oriented roughly east- west. C
m 2-00.100
. .. SESENtA GRANDIFLORA 1 OO9 0,
SALCHo RNEA CORDoFOLIA O 0
000
_ 0 *
ALBIEUCAENA LCATAR'
'00 z <~C
. o -z U)
o GMLN IAOA g th -Al i Z ani
rQ .j (D IL U m
-. Fig. 42b. Relative growth performance of selected treeshrub species at IITA and Onne after 18 months of es- ,tablishment.
ALBIZIA FALCATARIA
GM -E Figure 43 shows that Albizia falcataria and Gmelanai ar* borea planted in 4 m rows are clearly incompatible with
alley cropping once they are fully established because of 20 their strong depletion of incident light. They could be
maintained possibly only as planted fallow. ,C. Crops planted at less than 2 m from established Alchoro 2 3 nea cordifolia will also suffer severely from shading. ReaDISTANCE FROM TREE ROWS (m) sonable performance can only be expected at distances
of at least 3 m from the tree rows. This is also true of Fig. 43. Light transmission through the canopies of es- Leucaenea leucocephala. The distance from Sesbania tablished fallow/alley cropping species in relation to grandiflora can be reduced to 2 m and even 1 m for crops distance from tree rows (trees in 4 m-rows, E-W; OBS in with a fair degree of shade tolerance. In general, for all November). the species, except the latter, cutting back after the first
39




year of growth may be the only practical way of using
them for alley cropping.
Establishing leguminous trees or shrubs for alley crop- Pure Maize
ping in the derived savanna. Establishment and early 2.0- Maize- Pigeon pea
maintenance costs appear to be major deterrents in the Moize Leuceeno
use of leguminous trees or shrubs for nutrient recycling Maize Tephrosia
in alley cropping. To overcome these problems, the le- Maize Gliricidia
gumes are being established by interplanting through
maize. A trial evaluating the potential of this method in .: the derived savanna was initiated in 1980. In the inter-
planting, the legumes are spaced at 400 cm x 50 cm and
the maize at 100 cm x 100 cm with 3 plants/hill. Seeds "iC
were used for Tephrosia candida, Cajanus cajan and 0 1Leucaena leucocephala and cuttings for Gliricidia.
Because of adverse rainfall during the first season, maize 0 yields were lower than expected (Fig. 44). The legumes
had no significant effect on maize yield, and the maize had no significant effect on legume yield; legumes planted without maize were the same size as those
planted with maize.
Fig. 44. Effect of shrub legumes on single maize crop.
Establishment of Tephrosia was very poor while that of
the other legumes was good. Growth of both Leucaena and Gliricidia was poor even though adequate weed control was practised during the intercropping phase. By A' ~.- ~ November (6 months after sowing), these 2 legumes
were growing slowly with no marked effect on the adjacent vegetation. On the other hand, Cajanus cajan grew N~ 4 vigorously and by November dominated the field. A well
A/ developed canopy shaded out weeds preventing many
from flowering (Fig. 45). A closer between row spacing
(about 300 cm) could produce better weed control although a closer spacing may hamper tractor drawn implements.
The slow growth of Leucaena and Gliricidia suggests that these legumes may require more than 1 growing
season to significantly affect field ecology. Vigorous growing Cajanus cajan interplanted with maize in the
300
.?00
4000
00 o000
Twenty-month-old Albizia falcataria. Among the fastest 0 0 c00 00 0
growing trees in the world, they are suitable agro-forestry species in the humid lowland tropics for control of Fig. 45. Pigeon pea canopy structure and weed distriImperata cylinderica grass as well as for site fertility bution in 4 m alley, 3 months after maize harvest, improvement and in the derived savanna as a planted (maize/pigeon pea -, ( pure pigeon pea fallow species. *
40




early rair could provide an effective green ground cover, and sorghum. Heavy shading under cowpea retarded less prone to fire hazard, in the savanna regions. The Leucaena growth somewhat but from the standpoint of plants can be eradicated mechanically with the residue risk-aversive small farmers, suboptimal establishment providing mulch suitable for no-tillage cropping. The po- growth of Leucaena appears to be a small price to pay tential contribution of N to the soil and protein for human for steps taken to obtain an economic yield from their
and animal consumption are other factors that warrant fields in a bad year.
further investigation of Cajanus cajan in savanna crop- Where mechanized yam culture is likely to be an attracping systems. tive way to reduce the labor costs of yam production, the
trials indicate that a 4 m between-row spacing for LeuOn-farm trials-Leucaena-maize caena may be more suitable than the experimental 2 m
systems spacing. Wider spacing may also be preferred where
farming traditions favor multiple vine support by fewer The main objectives of the long-term, alley cropping stakes such as in the "tent staking" system of Tawari, farm trials that began in 1980 are as follows: (1) to iden- Nigeria. While allowing for such modifications, it seems tify and enlist the cooperation of a set of participating advisable to maintain a sufficient population of Leufarmers, representing a range of farming systems and caena to provide adequate mulch material, firewood and farmer types involved in the production of yam and other by-products to maximize overall returns. Where maize in Nigeria: (2) to instruct the farmers on the agron- firewood scarcity is a problem, the wood yield may be a omy of the prototypical Leucaena-maize-yam alley crop- significant component of the net economic return from ping system; and (3) to see to the establishment of the the alley cropping system. The possibility of a double Leucaena hedgerows as an understory intercrop with Leucaena row in the 4 m spacings should be investifirst season maize in preparation for use as live in situ gated.
yam vine staking next year.
Trials were established on 9 plots at 6 different locations One potential variant, which aroused considerable inter visits est among farmers for whom the labor costs of weeding across the yam belt of Nigeria (Fig. 46). Periodic arviajrfrmmngeetcnsrititheueo
are a major farm management constraint, is the use of were made to each site to provide basic guidance and
mor poes all aste throvie fas weae nr Leucaena shade to control Imperata cylindrica and other
monitor progress. In all cases, the farmers were encourheliophytic weeds. Although this use of alley cropping
aged to feel free to modify the system to fit their needs. has yet to receive systematic attention, a possibility If Information on such modifications provides for a better h ye to receie systeaticeion, a apo
undestadin ofthe orkng riniple ofappoprate might be to rotate a 1 or 2 year closed-canopy fallow
understanding of the working principles of appropriate through a large field set up for alley cropping. If, for alley cropping systems for smallholder conditions. example, the field were divided into 4 sections, threefourths could be in alley crop production in any given year with the remaining one-fourth in rotating fallow. Research is needed to evaluate the actual weed control potential of this alley cropping modification, but farmers indicate any innovation that minimized the herbicide need would be a major step toward making no tillage and mulch farming techniques more widely acceptable.
Linear programming model of Leucaenal rice alley cropping system
A linear programming model was developed to evaluate the economic attractiveness of an experimental Leucaena/-rice alley cropping system under West African .I smallholder conditions. Using production data representative of upland family rice farms in Sierra Leona and N DERIVED SAVANNA,.
:I1 OUIE SAVANNA response data from the Rokupr Rice Research Station in
v I SOUTH GUINEA SAVANNA
IV NORTHERN GUINEA SAVANNA Sierra Leone, the model was used to explore the relative v JOS PLATEAU profitability of various rice growing activities at 0, 20, 40,
VI SUDAN SAVANNA
VII SAHEL SAVANNA 60, 80 and 100 kg N/ha from 3 different sources: urea;
ammonium sulphate and in situ Leucaena hedgerows. Fig. 46. Map of Nigeria showing ecological zones and The main results indicate that under the conditions of sites of the Leucaena-maize-yam alley cropping system. smallholder production in the model, it is consistently Farm trials in 1980. 1. Lagbe 2. ljaye 3. Osara 4. Tawari more profitable to grow rice with N from Leucaena 5. Yandev 6. Zaki Biam. hedgerows than from either of the 2 mineral N sources.
Furthermore, with labor as the limiting production factor, First year trials have yielded a number of preliminary the 2 components of the Leucaena/rice alley cropping fiis Leucaena has been found to be a priardy system always combine in the same economically optifindings, eceahsbe on ob very hardy mal proportions of 0.37 ha (3,700 linear meters) of Leuplant that established well even under extremely adverse
conditions. A severe first season drought caused failure caena hedgerow to 1.28 ha for rice, grown in the alleys of the maize intercrop in most sites, but Leucaena between the hedgerows, for an optimum field size of 1.65
growth continued at moderate levels. In order to cope ha.
With the economic effects of the drought, farmers at- In order for rice production with urea or ammonium sultempted to substitute other intercrops such as cowpea phate to become competitive with Leucaena-based rice
41




production, subsidy levels of 97 and 99 percent, respec- stover plots but not in the live mulch plots. These results tively, would be required. confirm the finding of last year that live mulch crop proThe results of this modeling exercise must be regarded duction has potential for minimizing the need to control weeds in maize. The highest maize yield was observed in as provisional until validated by field trials, but the ap- a live mulch plot in which Psophocarpus pa/usfris was parent economic attractiveness of the hypothetical Leu- aied. caena/rice alley cropping system would seem to justify the research attention of agronomists. High maize yields in this live mulch cover were not related to weeding treatment. On the other hand, compaLive mulch systems rable yields were obtained in no-tillage and maize stover
plots that were weeded at least twice. Yield in the conLive mulch crop production involves planting a food ventional-tillage plots was significantly reduced even crop directly into a living cover of an established cover when the plot was kept weed free. crop without tillage or destruction of the fallow vegetation. This incorporates the soil conservation features of organic mulch and no tillage and has the advantage of cover on the N fertilizer requirement of maize was investigated. All plots received a blanket application of 30 kg smothering weeds and contributing N in the case of a P20, and K20/ha. Maize responded to varying levels of N legumn live mulch. fertilizer applied in the conventional and no-tillage and
Weed competition, tillage and ground cover. Observa- maize stover plots but not in the live mulch plot in which tions in 1980 showed marked differences in weed bio- Psophocarpus palustris was the ground cover (Table 49). mass due to the weed control method and ground cover type in maize. Although the field dry weight of maize Table 49. Effect of N fertilizer level and ground cover stover used as ground cover averaged 10 t/ha, this was on maize (TZE 4) yield (IITA, 1980 First seanot enough to completely eliminate weed growth. Weed biomass was identical in the unweeded check of the conventional and no-tillage plots, and these were signifi- N-fertilizer Grain yield
cantly higher than in the unweeded check of the live Ground cover (kg/ha) (t/ha)
mulch plots (Table 48). Conventional 0 1.24 d'
60 1.49 cd
120 1.51 cd
Table 48. Effect of weeding frequency and ground No-tillage 0 1.63 cd
cover on weed competition and maize (TZE 60 2.70 a
4) yield (IITA, 1980 First season). 120 2.51 ab
Grain Maize stover 0 1.74 cd
Weed Weed D. wt. yield' 60 1.98 bc
Ground cover control (t/ha) (t/ha) 120 2.49 ab
Conventional Weed free 0 c 1.54 cde Arachis repens 0 1.09 d
tillage Weed x 2 0.8 b 1.65 cde 60 1.72 cd
Unweeded 120 1.45 cd
check 1.49 a 1.05 e Psophocarpus 0 2.47 ab
No tillage Weed free 0 c 2.57 a palustris 60 2.39 ab
Weed x 2 0.83 b 2.43 ab 120 2.41 ab
Unweeded Average stand count of 50.6 x 103/p/ha
check 1.4 a 1.84 bcd 'Means followed by the same letter in the same column
Maize stover Weed free 0 c 2.27 ab are not significantly different at the 5% level of the New Weed x 2 0.91 b 2.39 ab Duncan's Multiple Range Test.
Unweeded
check 1.31 a 1.56 cde
Arachis repens Weed free 0 c 1.36 de ,.A"",k' j* Weed x 2 0.27 c 1.52 cde
Unweeded 94 AX.
check 0.31 c 1.32 de t
Psophocarpus Weed free 0 c 2.63 a
palustris Weed x 2 0.05 c 2.50 a
Unweeded
check 0.11 c 2.14 abc
Average stand concent of 50.27 x 103 pl/ha. 'Means followed by the same letter in the same column are not significantly different at the 5% level of the New ... .. Duncan's Multiple Range Test. CGA 41Z%3
2 0 KG ,iA
The unweeded check, live mulch plots had fewer weeds .(Lo-6oat harvest than in each of the conventional and no-tillage .
and maize stover plots that were weeded twice. Uncontrolled weed growth significantly reduced the maize Typical live mulch stand in one of the field experiments grain yield in the conventional and no-tillage and maize in 1980.
42




Without N fertilizer, the maize yield was twice as high in yield (Table 50). When weed interference was limited to the live mulch plot in which Psophocarpus palustris was the above ground vegetative parts of both weeds and the ground cover as in the conventional-tillage plot. crops, only a slight yield reduction was observed; but Maize yield in the no-tillage and maize stover plots with- when the leachete from the roots of weeds was allowed out N fertilizer was as low as in the conventional-tillage to come in contact with the soil in which the yam grew, plot. The advantage of no-tillage over conventional till- yam tuber yield was significantly reduced. This yield reage under continuous cultivation is demonstrated only duction caused by allelopathy was consistent for the 3 when N fertilizer was applied. Without N fertilizer, the years in which this study was carried out. maize yield was as poor in the no-tillage plot as in either When moisture was not limiting and plants received a the conventional-tillage or maize stover plots. Results of basal fertilizer treatment (30 kg N, PO, and KO/ha), the this study show that the best live mulch treatment elimi- major reasons for the yield reduction caused by weeds nated the need for a fertilizer application, and the N con- 'were allelopathy and other soil related factors. Each of tributed by the legume cover in the live mulch maize was these 2 weed interference factors accounted for over 30 comparable to the 60 kg N/ha in the no-tillage maize. percent of the total loss caused by weeds (Fig. 47). KeepThe study of the live mulch production of maize shows ing crops weed-free appears to be the only way to minithat (a) continuous maize production is possible under mize the yield reduction caused by weeds. the live mulch system without a bush fallow, (b) weed Weed interference studies. A study showed that a sigpressure can be minimized, and (c) the requirement for nificntecline uleaf A ind so)ad tbe y of N fertilizer can be minimized or eliminated. These find- nificant decline in leaf area index (LAI and tuber yield of ings have far reaching implications among small-holder white yam occurred due to weed inference from 8 to farmers who can neither afford the labor for weeding nor about 16 or more weeks after planting in tubers planted the chemicals for fertilizing. Live mulch is an attractive at the onset of rains (Table 51). crop production practice that provides complete, yearround ground cover, thus minimizing soil erosion and soil structure loss.
Weed biology and herbicide A ABOVE GROUND INTERFERENCE
research B ALLELOPATHY
C NUTRIENT/ MOISTURE?) INTERFERENCE
Biology of tropical weeds 60.
Little is known about the biology of weeds in relation to their control in tropical agriculture. Some aspects of N weed research focus on reasons for crop yield reduc- .tions caused by weeds. Allelopathy and other factors of = 40. weed interference have been investigated in selected tropical crops to develop effective weed control strategies.
Allelopathy weeds vs. yams. Since previous observa- 20. tions have shown that yam (D. rotundata) is sensitive to weed interference, studies were carried out from 1978 to 1980 to assess the nature and extent of this sensitivity. A new method was designed and tested for assessing allelopathy in root and tuber crops closely simulating field A B C
conditions. Results of these studies show that full weed interference caused a 75 percent reduction in yam tuber Fig. 47. Weed interference in yam.
Table 50. Mean fresh tuber yield and yield losses of white yam grown in wooden boxes as affected by mode of
- weed interference (1978-80)
Mode of weed Fresh tuber yield Yield loss
interference 1978 1979 1980 1978 1979 1980
kg/std%
Interference by aerial
factors 1.83 abl 0.72 ab 2.48 a 16.44 20.88 11.70
Interference by alleloPathy + aerial factors 1.26 bc 0.58 b 1.38 b 42.47 36.26 37.80
Full interference 0.53 c 0.22 c 0.47 c 75.80 75.82 78.80
Weed-free check +
Polythene 2.19 ab 0.91 a 2.22 a 0.00 0.00 0.00
Weed-free check
Po1lythene sheet 2.52 a 15.07
Means followed by the same letter in the same column are not significantly different at the 5% level of the Duncan's New Multiple Range Test.
43




Table 51. Mean values of yield and yield components of white yam as affected by duration of weed interference in
the field (1978-79).
Time (weeks)
LA I to 100% tuber Fresh tuber
Duration of (at 12 WA P) initiation yield (t/ha)
weed interference 1978 1979 1978 1979 1978 1979
W. F. M. 0.55 ab' 0.58 a 12 a 10 a 20.08 a 17.57 a
W. 1. 4 0.87 a 0.21 cd 14 d 10a 18.75 ab 17.40 a
W. 1. 8 0.29 b 0.18 d 14 a 10 a 13.96 bc 13.91 a
W. I. 12 0.22 b 0.16 d 14 a 10 a 12.47 bc 7.40 bc
W. 1. 16 0.29 b 0.04 d 14 a 10 a 6.47 d 6.06 bc
W. I. 20 0.21 b 0.06 d 14 a 10 a 6.58 d 3.24 c
W. I. 24 0.30 b 0.02 d 14 a 10 a 4.88 d 1.89 c
W. 1. M. 0.24 b 0.06 d 14 a 10 a 6.03 d 2.12 c
Weeded 3 + 8 + 12 0.94 a 0.74 a 14 a 10 a 19.71 a 12.25 bc
Weeded 3 + 8 + 12 + 16 0.82 a 0.55 bcd 14 a 10 a 20.88 a 17.28 a
1Means followed by the same letter in the same column are not significantly different at the 5% level of the Duncan's
New Multiple Range Test.
In the absence of weed interference, dry matter produc- 3.0 and 6.0 kg/ha while pendimethalin was sprayed at 2.5 tion in yam vines and leaves peaks in phase II (vegetative and 5.0 kg/ha on a conventional-tillage plot using a knapgrowth stage) and phase Ill (tuber bulking growth stage). sack sprayer calibrated to deliver 200 1/ha spray volume.
Weed interference during these periods severely reduces Herbicide persistence was monitored by a bioassay growth in those organs, and this reduction during method involving tomato seedlings for atrazine and fluphases II and III (12-16 weeks after planting) interferes ometuron and rice seedlings for metolachlor and pendiwith assimilate production necessary for optimum tuber methalin.
bulking in the later part of phase I1l. Atrazine and metolachlor each at 3.0 kg/ha were less
Mamatter production in weeds peaked at 8-12 weeks persistent than fluometuron and pendimethalin (Fig. 49).
and declined thereafter, indicating that the maximum Generally, all herbicides were more persistent at IITA competition occurred during this period. Decline in tuber (rainfall 1,400 mm) than Onne (rainfall 2,400 mm). Atrayield was also most pronounced at this period (Fig. 48). zine at 3.0 and 6.0 kg/ha was present in soils at IITA at
nonphytoxic levels at 8 and 12 weeks after treatment,
respectively (Fig. 50). The persistence of metolachlor
MOP GOWTH PHASE 0- Weed ftr up to naturcy closely followed that of atrazine while fluometuron and ..4 We -,nfested upto t-y pendimethalin showed pronounced persistence at IITA 01 v I i Iv at both low and high rates. Fluometuron, especially at a
CO> I high rate, persisted beyond the test period in the IITA
c joIs soil. Pendimethalin at 2.5 kg/ha persisted at high levels
oo for more than 8 weeks after treatment, but existed at very
002 I low levels at 12 weeks after treatment.
12 I Prisene!" a as shr
'Persistence ofatrazine in soils at Onne was very short o8 VINE 5 and had declined to non-phytoxic levels by 8 weeks after
S treatment (Fig. 49). Doubling the rate of atrazine did not
I I increase its persistence. Metolachlor applied at a high
180 oo 010 rate persisted in Onne for more than 8 weeks after treatS LEAVES 1 3 ment. Fluometuron could not be detected in the top soil
.20 I 'p, at Onne by 12 weeks after treatment. Persistence of pen120,
0so- dimethalin was identical at both IITA and Onne, irrespec060 tive of rates used.
While an atrazine sensitive crop could be planted at IITA e -. o 1 and Onne locations at 12 weeks after treatment, a flu8 2 6 20 2 28 32 8 2 20 2' 32 ometuron sensitive crop could not be safely grown under
WEED INTERrEPENCE (WAP) WEED INTERFE ENCE (WAP) soil and rainfall conditions at IITA within 12 weeks after
treatment. All herbicides tested in soils at Onne were
Fig. 48. Effect of weed interference on yield and yield either nontoxic or had very low herbicidal activity at 12 components in yam. weeks after treatment.
Herbicide persistence in tropical soils Plantain improvement
Field experiments were carried out at IITA and Onne to Previous greenhouse trials indicated that growth regudetermine the persistence of some selected commonly lators applied close to the meristem of young plantain used preemergence herbicides in the 2 ecologies. Atra- (Musa sp.) suckers changed growth hormone balance zine, metolachlor and fluometuron were each applied at and sucker development. This suggested the potential
44




200 E ONNE
. 180 MI IITA
0
160 zin fluometuron 30Okg/ho
0 (a) 0 (c
140- z
LL 0
0 Ui
S120- ... L
0
400
I-L
280
0 2:
In0 0
wm
0 20
m 0
20
-jj
0
0i 14 2 z 20 1 2 4 8 1
WEK FTRLRAMN
Fi20 89 essec fhriie nsraesis
-45




200- otrozine 1I30 kg/ho al 30kg/ha 180- 60Okg/ho fluometuron 60Okg/ho
0J 0
S160- (a) Q (C)
0140 0
ILL LL
0120- 0
a 1l00 Cl
~80
~60 W
140-C)
U- 20- LL
0 1 2 3 4 8 12 0 1 3 4 8 1
WEEKS AFTER TREATMENT 200 ED 3-0 kg/ho Ol25 kg/ho
180- metolachlor 6-0 kg/ha pendimethalin 5-0 kg/ho
o 0
.160- (b) (d
0 0
0140-0
LL U0 120< <(D 80 I C
~60-lt u
< 40 C.
0 L 2 3 4, 8 12 0. 1 2 3 4 8 12
WEEKS AFTER TREATMENT Fig. 50. Effect or rate (kg/ha) on herbicide persistence in soil.
46




use of growth regulators in shortening the harvest cycle injection of 1 ml of H20 also repeated once after 5 days. interval between first and second ratoon crops. It also Observations were on the mother plant, P1, P2 and P3 at suggested the possibility for rapid multiplication without the time of first injection and 6 weeks after the second loss of yield from the mother plant. Studies in 2 field injection (Table 52). trials were condIucted. The GA treated peepers developed faster than the first
peepers on the same stool and the GA injection did not Physiological study of the suckering significantly stimulate the growth of the first and third
behavior peepers. No effects could be observed on the main pseudostem. This points to a very localized action of GA on In the first trial, the pseudostem of a 7-month-old mother the treated suckers only, and an immediate change in plant was cut off and the apical meristem removed. The the balance is enough to stimulate the growth. The H20growth regulators; GA (Gibberillic acid, GA + NAA (Nap- treated peepers did not show any difference in growth thylene acetic acid), GA + cytokinin and ABA (abscisic compared with the peepers of the neutral control, provacid); were applied on alternative days in lml doses in ing that tissue changes by the injection was not interferconcentrations of 10-2 and 10-4 M. Five peepers per ring with the development of a sucker. mother plant were treated at 2-4 weeks after detopping of the mother plant. Effect of Flurenol on apical dominance
Treatments with GA/NAA were not considered because the structure of the peepers seems to be changed. There Flurenol is known to counteract the apical dominance. were no significant differences in height and girth after This effect on plantain was compared with the neutral 4 weeks among the peepers of the 16 treatments (2 du- control (no treatment) and decapitation (Table 53). Flurations x 2 concentrations x 4 hormones), except for renol was injected into the 4.5.month-old main pseudothe height of the GA 10-2 M treated peepers. Peepers injected with ABA were not significantly different in their growth from the control. Table 53. Mean height (Ho, H) and girth (Go, G,) of the
In the second trial, the mother plant was retained. Treat- second peeper and the main pseudostem at ment began at 6 months after planting. Only 1 peeper per 0 and 6 weeks (in cm). Number of peepers at plant was treated. Treatments with GA/NAA and GA/NAA 0 and 6 weeks (N., NJ. CYT were not considered because of abnormal changes Control Flurenol Decapitation
in the peepers. Suckers injected with GA, GA + flurenol P2 H. 3.7 1.64 1.4
and GA + CYT showed, compared with the control, sig- H6 22.7 16.82 57.7 (1)
nificant differences in height and girth after 6 weeks. Go* 3.8 1.73 1.8
Suckers injected with GA + CYT became significantly G8 13.2 9.27 16.8 (5)
thicker. *Girth at soil level
Also, there were no significant differences between du- Mother- Ho 182.6 146.3 (5) 157.9 rations, but there were between the 2 concentrations. As plant H, 264.5 Tf-1 in trial 1, peepers injected with ABA did not differ signif- G 44.2 36.78 (5) 39.8 icantly from the control. G*0 37.9 31.4 (5) 32.9
The effects of GA on suckers in Trial 2 proved that the G6* 64.5 52 (1) significant growth of injected suckers is not linked to the G6 1o* 52 42.4 (1) duration of the injection but to the hormone concentra- No 2.3 0.9 (5) 1.4
tion. In the next experiment, 1 ml of Ga 10-2 M was in- N, 8.4 6.2 4.6 (1)
jected into the second peeper (P2) at 10 days after *Girth at 50cm and 100cm height emergence on 10 different plants. The injection was re- "Number of peepers peated once on the same peeper at 5 days after the first. Means in the same row followed by (1) or (5) respectively, The controls were no treatment (neutral control), and are significantly different for 1% and 5%.
Table 52. Height and girth of second peeper, mother plant pseudostem and number of peeper at 0 and 6 weeks
after treatment.
Control Water GA
Type of Plant Parameter Start 6 weeks Start 6 weeks Start 6 weeks
Peeper2 Height (cm) 8.1 24.6 8.2 28.6 7.6 87.6*
Girth (cm) 6.7 13.4 8.3 16.4 6.6 20.1"
Peeper, Height 24.8 65.7 16.7 47.7
Girth 13.1 24.9 10.2 19.5
Peeper3 Height 5.2 15.5 4.6 20.8
Girth 5.5 12.9 4.0 13.5
Mother plant Height 200.1 285.9 200.0 277.5 196.6 285.8
Girth (50 cm) 48.8 68.9 47.9 65.1 49.3 66.3
Girth (100 cm) 41.1 54.7 40.1 53.2 41.9 54.1
No. of peepers 4 10 3.8 9.6 3.8 8.6
'Significant differences between treatments at 5% probability.
47




stems of 10 different plants and repeated a second time. sponsibilities of IITA is to conduct an agronomic survey Each time 2 ml was injected. Observations were made on of existing farming systems to determine research priorthe mother plant and the second peeper at the time of ities for the regional food crop research.
the first injection and 6 weeks after the second injection. Agroeconomic surveys. The agroeconomic survey was
Flurenol treated plants did not show faster sucker devel- conducted with the following objectives: (1) to investiopment nor higher peeper production. Differences in the gate and analyze existing farming systems, (2) to gather main pseudostem after 6 weeks are probably due to the information on existing patterns of agricultural producselected plants in the beginning of the experiment. Thus, tion and resource use, (3) to identify major constraints, to counteract the apical dominance, repeated injections and (4) to collect relevant information on marketing and seem to be necessary to maintain the change in hormone output supplies. A report titled, "Survey of food crop balance. Decapitated plants gave bigger suckers farming systems in the Zapi-Est, East Cameroon," was
(maiden suckers) as the control of Flurenol treated published.
plants (peepers). Release from apical dominance was Field trials. The field trials on varietal improvement, culnot stimulating higher peeper-number which suggests tural practices and fertilization were initiated in 1980 at that the maiden suckers are dominating. Bertoua in the savanna zone and at Doume in the forest
zone. The varietal improvement was carried out with
Effect of fertilizer and mulch maize, groundnut, soybean, cowpea and upland and lowland rice. Grain yield of selected maize varieties was 3 to
Plantain in backyards are growing under optimal condi- 4 times that of the local variety; TZB was the top yielder tions because of high level of nutrients, as well as or- with 6.1 tlha. Streak is a major problem for maize producganic matter (Table 54.) In this experiment, the tion in the area. nutritional requirements of 2 varieties of plantain (medium and giant False Horn) were compared with that of Among the groundnut varieties tested, Bertoua Blanche a banana (Kparanta). It is clear that in all cases fertilizer gave the highest yield of 2.1 t/ha. Soybeans performed had strong positive effects on growth, whereas the ef- well in the area and yields ranged from 0.6 to 1.9 t/ha. fects of mulch appeared less pronounced. The com- The top yielder was variety 20-67 TB. Though cowpeas bination of mulch and fertilizer showed positive showed good growth, leaf diseases and insect damage interaction. were commonly observed after flowering. Grain yields
varied from 0.2 to 1.4 tha with TVx 1948-OIE giving the
A highest yield. High yields of lowland rice were obtained
fale 54. Height and girth of a medium and a giant with variety IM 16 giving the highest yield of 7.0 t/ha.
False Horn plantain and a banana after 5.5 On newly cleared land at Bertoua, grain yield of maize months. variety Ekona Mixed Color responded significantly to N
Girth (cm) and P applications at rates of 60 kg N/ha and 40 kg P20,/ Height at 50 cm ha. Positive N x P interactions were also observed. ReTreatments (cm) height gardless of level of fertilizer application, maize yield inFalse Horn giant plantain creased significantly as the plant density increased from
No mulch, no fertilizer 85.61 ab* 21.31 abc 33,000 to 66,000 plants/ha.
No mulch, fertilizer 133.45 ab 31.14 abcd Field trials with groundnuts showed significant reMulch, no fertilizer 119.14 ab 27.69 abc sponses to application of lime, N and P (Fig. 51). N appliMulch and fertilizer 181.18 bc 42.92 cd
False Horn medium plantain Relative yield
No mulch, no fertilizer 56.92 a 15.91 a NO
No mulch, fertilizer 116.04 ab 27.30 abc
Mulch, no fertilizer 123.09 ab 29.39 abcd
Mulch and fertilizer 181.15 bc 40.96 bcd I9o O No lime
Banana 180 Lime I I ton of CoO/ha
No mulch, no fertilizer 94.20 ab 22.94 abc 2 2 tons of CoO/ho
No mulch, but fertilizer 188.11 bc 38.78 abcd 170
Mulch, no fertilizer 153.17 abc 31.64 abcd 160I / P I No P2O h5
Mulch and fertilizer 266.95 c 53.14 d / P0 = g P205/ho
'*Means in the same column opposite the same letter(s) 150. 2 1
are not significantly different from each other at 5% lev- 140 / No 11.4O kN N
els. Mulch consisted mainly of Eupatorium. A eN 1 40 kg N/ho
120
Eastern Cameroon Farming ,,o
Systems Project 100 rFruit yield of control treatment s 725 kg/ho.
As part of the effect to develop the Eastern province of
Cameroon in cooperation with Zapi-Est (Zone d'Actions
Prioritaires Integrees de lEst), the World Bank asked
IITA to assist the then Cameroon National Office of Scientific and Technical Research in establishing a food Fig. 51. Response of groundnut (CV Bertoua Blanche) to crop research station in that province. One of the re- N, P and lime applications.
48




cation on long growth-duration rice variety IM 16 showed > maize seeded at 20 days after the seeding of groundthat time of N application greatly influenced yield (Table nut. A similar trend was also observed with groundnut 55). While basal application of 50 kg N/ha at 1 day before yield, and obviously both crops reacted negatively to late transplanting did not increase yield, application of 50 kg planting and intercrop competition. N!ha at panicle initiation did. It appears that the best recommendation may be 50 kg N/ha basal applied and sidedressed with 50 kg N/ha at panicle initiation.
Small farms systems research
Table 55. Grain yield of irrigated rice variety IM 16 (Atebubu District, Ghana)
under different treatments of rate and time The small farms systems research is a component of the of N fertilizer application (Doume, 1980). USAID sponsored project of Managed Inputs and DelivFertilizer rate Grain yield ery of Agricultural Services (MIDAS). The MIDAS project
(kg N/ha) Time of application, (t/ha) is designed to organize and distribute all the inputs nec0 -4.6 essary for food crop production by small farmers of
0 4.6Ghana.
50 50 (B) 4.6
100 100 (B) 5.3 The small farms system research is designed to assist
50 50 (PI) 5.9 the government in establishing an applied multidisciplin100 50 (B) + 50 (PI) 6.6 ary small farms research capacity with the following ob100 40 (B) + 30 (30 DT) 5.5 jectives in mind:
+ 30 (PI) (1) To obtain a sound knowledge of the existing farm,B = Basal, 1 day before transplanting. ing systems, the socioeconomic environment and
PI = Panicle initiation. the positive and negative factors in services.
DT = Days after transplanting. (2) To conduct applied research that is relevant to
these circumstances and responds to the needs of
the small farmers.
Maize/groundnut intercropping has been extensively uti- (3) To identify soil and farm management and produclized by local farmers. A field trial was conducted during tion practices that eliminate constraints. the second season to evaluate the potential of this sys- (4) To increase small farmer production and income. tem and investigate the factors affecting yields of crops when they are planted in association. Among the inter- The project was initiated in 1980 with an appraisal of the cropping treatments, the treatment of maize at 40,000 present food crop farming system practices in the rehills/ha seeded at 20 days after the seeding of groundnut gion. For the survey, a group discussion approach was at 133,000 hills/ha gave the highest monetary return used that was supplemented by the visit to individual (Table 56). farmers and spot harvest checks. The survey covers 7
In general, however, the yield of both crops is reduced subdistricts-Atebubu, Amantiri, Abease, Prang, Kwame when the crops are planted in association. The amount Danso, Kajaji and Yeyi. of yield reduction varies with the seeding time of the The survey was completed in 1980 and a report titled, affected crop in relation to the seeding time of the asso- "An appraisal of the present farming systems of the Ateciated crop. The yield of maize planted in association bubu District of Ghana," was published. During 1980, with groundnut, for example, was reduced in the follow- preparations were also made in selection of a site for ing order: maize seeded at 20 days before the seeding of establishing the experiment station. Actual field experigroundnut > maize seeded at the seeding of groundnut mentation, however, will be initiated in 1981.
Table 56. Yield of sole and intercropped maize and groundnut and their monetary values. (Bertoua, 1980 second
season).
Crop and date of
seeding
Treat- (M = Maize; Plant density "Treat- Crop Yield Kg/ha Monetary value (CFAF)'
ment G = Groundnut) (x 101 hills/ha) ment Maize Groundnut Maize Groundnut TOTAL
C, M on 1 August 50 C, 3066 183,960 183,960
C2 G on 1 August 200 C2 2277 284,625 284,625
M on 1 August, and 40
C, G on 1 August 133 C, 2263 1206 135,780 150,750 286,530
M on 1 August, and 40
C, G on 20 August 133 C 2959 539 177,540 67,375 244,915
M on 20 August, and 40
C., G on 1 August 133 C 831 2122 49,860 265,250 314,790
M on 10 Sept., and 40
C, G on 1 August 133 C6 2245 280,625 280,625
At 60 CFAF/Kg of maize grain and 125 CFAF/Kg of unshelled groundnut (Retain price of PROVIV-ZAPI, 1980). "Maize seeded on this date did not germinate.
Fertilizer rates maize 8N 80 P20; groundnut 2N 6 P20 in Kg/ha.
49




Full Text

PAGE 1

7, E, Ow Aboutil het lnternationa4iiistituit 4.6t -".k.. Tropicai*Agricult6j6* (11.rXg.v -orWJUly# T was established 7!ft' 19677 as& arwautonomouo, nc P rofiftcorporationsbyrattdecreeoo .......... thft Federalt Militaryw GoVernment 0f NigeriaLit'formalikiorganized, 61.""4 v thet first meeting* of B.oards"obA Trustees ins lbadarw duHngv ujyj k 1968tV Thm Federal; Republicto geriaoprovide&11, 000thectarest of lanchfootheti[TA site!randtthciiFord" Foundationthe initial capita buildingstandidevelopmenfl,,, Oak, Support fooresearchtandtdaywt6 w dayr opera tionsi in, 1980, m the* For& Foundationrs. the!,, Canatv _,dianit international DevelopmentK. Agenc)w(CIDA),,.thetiOversea4pe, velopmentzMinistryw oft thet United,&., Kingdorrr. (ODM); thetUiS. Y# 1 Agefic for* Internationak'. D V iopmenW.,-, (USAID)" the.' WorldfBankp international Funds, forAgriculftirallil" DevelopmenV(IFAD) '-an OthetgovV1.1 a ernmentsr oV AustraliamwtBeigium Japan;; The Netherlandig, nk* Norway .and.the, FederahRepublic, 7 of Germanyi Theinstitute& is. governed bjvan, inttemational.Board. otTrustees;.,-.,..,. ........... Correct citation: International Institute of Tropical Agriculture, 1981, Annual Report for 1980. lbadan, Nigeria. Published by the International Institute of Tropical Agriculture Oyo Road, PMB 5320, lbadan. Nigeria



PAGE 1

A knowledge of the thermal diffusivity is necessary in 56predicting the temperature profile of a soil under differ5ent systems of management. Soil temperature affects 2 -" crop growth directly as well as indirectly through its in' 48Rodiotion 59Ogr/Cal cm2 teraction with soil moisture regime and its effects on root ¢ .o. Flat bare / growth, biotic activity and plant nutrients. Soil temperaZ Sybean (196) ture measurements at different depths and times can be 40. -Cowpeo (1.80) cumbersome. With the knowledge of soil thermal prop936 -&Maize (1-41) erties, moisture regime and boundary conditions, the A temperature profile of a soil can be predicted with a rea32 sonable degree of accuracy. An attempt was made to 28 compare the measured soil temperature with that esti24 37 days after seeding mated by the Fourier series solution and the Hanks 24 model. The initial and boundary conditions were mea20sured for the following treatments: bare flat, flat mulched -1'6 2,0 i4 with crop residue ( ut/ha, ridged and flat surface covHo,, of the day ered with transparent polythene. 36 .."Soybean (0-37) Radiation z 230g/Col cm-2 -xCowpeo (0.41) 1 1'" Soil temperature measurements were made for maize, F 4 -34 -Maize (047) 1 cowpea and soybean. The predicted soil temperature at 32 20 cm depth under maize and cowpea was within 0.6C of the measured soil temperature. Soil temperature can E also be predicted from the air temperature (Table 19). 28 0 The effect of canopy cover on diurnal fluctuations in soil 29 doW after seeding temperature at 29 and 37 days after planting for a sunny and a cloudy day, respectively, is shown in Fig. 13. The 24-X ........ higher temperature under soybean compared to the 22 lower temperature under maize and cowpea is partly 1'6 2'0 24 attributed to poor stand because of a supraoptimal temperature regime during the periods of seedling Fig. 13. Diurnal temperature variation at 5-cm depth in emergence and establishment, a flat base soil under 3 crop canopies at 2 occasions The values of amplitude and phase angle for the first 3 following seeding. Numbers in the brackets denote leaf harmonics of the temperature wave for different surface area index on the day of temperature measurements. conditions and crop covers are shown in Table 20. It is Table 19. Calculated and measured soil temperature in control (bare flat) treatment. Air temperature at 1-r' Soil temperature (0C) height Soil (C) depth Maxi. Mini. Maximum Minimum Mean Date (cm) Meas. Calc. Meas. Calc. Meas. CaIc. Difference 4.12.80 1 34.3 22.9 44.2 47.8 26.7 26.0 35.4 36.9 1.5 15 --35.5 34.2 29.5 30.0 32.5 32.1 0.4 8.12.80 1 35.5 23.2 45.8 49.2 27.2 26.3 36.5 37.7 1.2 15 --35.6 34.4 29.6 30.3 32.6 32.3 0.3 Maxi. = maximum; Mini. = minimum; Meas. = Measured; Calc. = Calculated. Table 20. Total variance (OC) of the soil temperatures at various depths under different crops grown on bare flat, mulched flat or bare ridged soil surface and its percentage accounted for by different harmonics. Percentage of variance accounted for by the harmonic Cover Crop-No Crop Soybean Cowpea Maize Harmonic Soil --* Number depth (cm) 5 20 35 5 20 35 5 20 35 5 20 35 BARE FLAT Variance (C) 115.53 19.90 4.45 140.83 28.75 6.45 19.30 3.49 0.56 18.12 3.20 0.57 1 94.8 95.6 92.6 93.2 96.0 93.3 92.3 94.0 91.1 94.4 95.3 89.5 2 3.8 2.1 4.5 5.3 1.7 4.2 4.4 4.3 5.4 3.2 2.5 7.0 3 0.5 0.7 0.7 0.6 0.8 0.5 1.3 0.3 0.4 0.9 0.9 1.2 Total 1 to 3 99.1 98.4 97.8 99.1 98.5 98.0 98.1 98.6 96.9 98.5 98.7 97.7 15



PAGE 1

obvious that 90-95 percent of the variability in soil ternTable 21. Effects of crop cover on runoff and erosion. perature regime under bare, maize, cowpea and soybean Runoff Erosion covers can be explained by the first harmonic while 97Crop cover (mm) (t/ha) 99 percent of the variability can be explained by the first 3 harmonics (Table 20). In general, the second and third Maize with sweet potato 154 2.2 harmonics represented 1.2-7 percent and 0.1-3.8 percent Cassava with melon 380 3.7 of the total variance, respectively. This implies that the Yam with melon 314 4.0 first 2 harmonics alone can explain the diurnal fluctuaSweet potato 223 4.4 tion in soil temperature up to about a 30 cm depth. The Maize 197 4.4 third harmonic explains about 1 percent of the total varMelon 303 6.3 iance. Weed fallow 251 4.4 Yam 186 2.0 Runoff and erosion under root crops. Table 21 shows Cassava with maize 457 9.2 the effects of canopy cover of tropical root crops at difCassava 462 8.0 ferent growth stages on runoff and erosion. Cassava, although a closed canopy crop, takes a relatively longer time to provide a complete ground cover than maize or trol, the yield was significantly higher in conventional cowpea. Staked yams leave the ground surface exposed than no-tillage plots (Table 23b). to raindrop impact even when the canopy is fully develWeed control in conventional and no-tillage cassava was oped and, therefore, renders the soil more susceptible to evaluated in an Alfisol in a subhumid climate. The vegeerosion. Sweet potato, with quick and effective ground nation was a 2-year fallow consisting of perennial grass, cover, is a soil-conserving cover since it results in miniPanicum maximum; and broad leaves, Eupatorium odormal runoff and erosion compared to cassava and yam atum, Alchornia laxiflora and Ficus spp. The fallow veg(Table 21). Mixed croppings of maize with cassava, caselation was first slashed in late February before the onset sava with melon and maize with yam resulted in a signifof rains, and its regrowth was then sprayed in April with icant decrease in both runoff and erosion. Simultaneous 3.0 kg glyphosate/ha a.i. In the conventional-tillage secmeasurements of canopy cover made at different growth tion, the fallow vegetation was plowed under, and the stages will provide necessary information for computing cassava was planted at 1.0 m x 1.0 m spacing on the flat "C" values that can be used in the "Universal Soil Loss after harrowing. In the no-tillage section, the cassava Equation" for predictive purposes. was planted directly into sprayed regrowth. Cassava Table 22a. Effect of residue management and tillage Weed control in no-tillage system practices on maize yield and crop perforThe effect of residue management and tillage on maize mance (kenne, 1980). production using chemical weed control was investiWeed Grain gated on a 1-year Eupatorium odoratum fallow. (Tables Management Plant Lodging F.W. Yield 22a and 22b). Maize yield was significantly lower in the practice HI. cm % g/m2 t/ha no-tillage plot than conventional (plow and harrow) and reduced tillage (disc harrow) plot. Also, significantly Residue present 81.4 33 a 6.1 2.06 a more weeds grew at 4 weeks after planting where the Residue removed 81.7 29 a 6.7 2.06 a moewed ge a wesResidue burnt 87.9 a' 35 a 12.2 a 1.98 a residue was burnt off than where the residue was rePlow and harrow 86.7 a 41 a 8.5 ab 2.12 a tained or raked off. The method of weed control affected Disc harrow 89.8 a 34 a 11.4 a 2.20 a grain yield but not lodging in maize. The maize yield was No tillage 74.5 22 5.0 b 1.78 significantly lower in the unweeded than weeded plots (Table 23a). When plots were hand weeded, the crop Means followed by the same letter in the same column yield was identical in the 3 tillage practices. However, are not significantly different at the 5% level of Duncan's when a preemergence herbicide was used for weed conNew Multiple Range Test. Table 22b. Effect of residue management and tillage practices on maize yield and crop performance (Ikenne, 1980) Grain Residue Tillage Plant ht. Lodging Weed F.W. Yield Management Practice cm. % g/m2 t/ha Residue present Plow and harrow 83.8 ab' 45 a 5.8 b 2.13 ab Disc harrow 88.7 ab 31 abc 7.2 b 2.20 ab No tillage 71.7 c 24 bc 5.2 b 1.84 ab Residue removed Plow and harrow 87.0 ab 38 ab 8.8 b 2.17 ab Disc harrow 86.3 ab 32 abc 7.0 b 2.34 a No tillage 71.9 c 18 c 4.2 b 1.68 b Residue burnt Plow and harrow 89.3 ab 41 ab 10.8 ab 2.07 ab Disc harrow 94.3 a 39 ab 20.2 a 2.05 ab No tillage 79.9 bc 25 bc 5.7 b 1.83 ab 'Means followed by the same letter in the same column are not significantly different at the 5% level of Duncan's New Multiple Range Test. 16



PAGE 1

E!,ecI of tillage methods and mulches on crop producTable 15. Effects of tillage methods and N on rice grain -.on in heavy-textured soils in Zanzibar. In collaboration yield (IITA, 1980 Second season). -e University of Morogoro, Tanzania, tillage studies 7 :-ducted in Zanzibar to investigate the effects of Grain yield t/ha -s of seedbed preparation and mulching on crop N rate Conventional ,r Different treatments investigated were black and kg/ha No Tillage Tillage ,,. coiythene mulch, no tillage, ridges and bare flat i,.e Differences in plant growth among different 0 1.99 3.75 -rents were attributed to differences in soil temper30 2.94 4.79 .1!,v' ., and moisture regimes. These soils have high clay 90 4.02 5.05 c --,nt and low permeability. High soil moisture content .1a :} oor aeration were more serious in no-tillage and LSD (5%) 0.82 rucned of ots during periods of frequent and heavy !a.-_ consequently, crops were more chlorotic than in -cnventional-tillage plots. Effects of different treatments Effects of tillage methods and mulches on yam producen yield indicate that the highest maize yield was tion. An experiment was initiated at Onne to investigate .;,mined under white polythene mulch treatments and the effects of tillage methods and mulches on yam pro,c'vwest under no-tillage treatments (Table 13). These duction. Treatments consisted of planting on either the c fencess in grain yield were attributed to the number flat or ridge with and without residue mulch. Observa-, !;rains per row and the unit grain weight. The yields of tions were made for soil temperature, moisture and bulk co uea and soybean also observed a similar trend. density and root and shoot growth and tuber yield. Leaf area was generally more in mulched than unmulched plots with the least leaf area measured in the unmulched Table 13. Effects of seedbed preparation on soybean grain yields in Zanzibar. Grain yield t/ha K Treatment 1979 1980 lack polythene 0.8 1.1 Vhite polythene 1.5 1.0 W Nc-tillage 0.5 0.4 Ridges 0.6 0.5 Bare 0.8 0.6 LSD (506) 0.4 0.2 Effect of tillage on paddy rice production. In 1976, grain -i. \ ;* yields of rice were similar in conventional and non-tillage va., methods. Yields were significantly affected only by the level of N application. The yields were not affected by the .. tillage treatments for about 3 years. Since 1979, the plant height, tillers and grain and straw yield have been less in conventional-tillage than no-tillage for sandy soils (Table 14). In clayey soil, the differences due to tillage treatNo-till rice (top) had a poor stand and low yield after 6 ments were not pronounced (Table 15). Comparisons of consecutive crops compared to good stand and high rice yield in Tables 14 and 15 indicate that an optimum yield in conventionally puddled paddy (below). level of N for rice for a clayey soil is about 60 kg N/ha; whereas, the effect of N on rice yield in a sandy soil was not significant. The adverse effect of no-tillage on rice may be due to many factors, including nutrient imbalance. In a sandy soil, leaching losses of applied fertilizer on an untilled paddy may be more than in a puddled soil. Table 14. Effects of tillage methods and N on rice grain yield (IITA, 1980 First season). Grain yield t/ha Conventional Fertilizer No Tillage Tillage P11 With 2.50 3.82 Without 1.98 3.13 LSD (5%) 1.00 13



PAGE 1

(c) A single-row rolling injection planter with a fertilizer band applicator attached behind it is too heavy to balance. ____(d) A single-row rolling injection planter with a fertilizer band applicator attached beside it acting as a row marker as well as placing basal fertilizer works very well and eliminates using strings or poles for ______marking the rows. This system does not work well on heavy mulch, which contains a lot of fine grasses or weeds and hides the fertilizer. Neither the single-row rolling injection planter with a fertilizer band applicator nor the 2-row rolling injection planter are able to work on traditionally cleared land that has stumps left standing more than 15 cm in height. This -is because the amount of deviation from the row to be able to go around the stump is too great. The working rates of these machine combinations are shown in Table 25. Table 26 gives the amount of time to move row marker stakes when planting or spraying. Table 25. Rate of work of planter combinations (IITA, Fig. 14. Two-view drawing of the metering wheel cover 1980). for the rolling injection planter. Machine Man-h r/ha* caused increased numbers of seeds in some hills and Single row rolling injection planter 10 none in others. For crops requiring 3 or 4 seeds to be Single row rolling injection planter plus planted per hill, the small metal plate is not necessary. 2-rwrlingr ianjeto aplatr 63 The seeds distribute themselves fairly evenly between 2-row rolling injection plusftzer ban *the hole openers giving good crop stands. appwrliato ineto lsfetlzrbn A simple guard was designed that covers the opening Single row rolling injection with fertilizer between the opener lever and the injection wheel to preband applicator row marker 13 vent trash from passing between. The guard does not *These times to not include time to move row marking add significantly to the cost or weight of the machine, stakes. Previously, the small pieces of wood left in the field after clearing fallow were found to catch in the hole opener lever and wedge between the injector wheel and the hanTable 26. Time for machine operator to move row dles. This makes the planter almost impossible to push. marker stakes per hectare (IlTA, 1980). The solution to the planter sinking in soft soil was to fit a Width of machine man-hr/ha wider hole opener lever. This lever is 5 cm wide, adding 0.75 m 5 3 cm to the width of the previous hole opener lever. The 1.5 m 4 extra width provides for more contact with the soil, thus, 20m distributing the weight over a greater area and keeping2.m3 the planter from sinking. Finally, the requirement of *Estimate. planting in the row at other than 250 mm spacing has partially been solved. Seed rollers were made that give 2 and 3 seeds per revolution of the rolling injection wheel. This will give seed spacing of 500 mm (3-hole roller) and 750 mm (2-hole roller). A single hole roller could be made to give 1,500 mm spacing. This development brings a much wider range of activities for the rolling injection planter, giving it adaptability to various crops. In an attempt to increase the rate of planting, differentV configurations of planters and fertilizer band applicators were also designed and tested as follows: (a) A 2-row rolling injection planter was designed to in crease the field capacity and make a more stable planter by eliminating the sideways tipping of the machine. (b) A 2-row rolling injection planter with a fertilizer band applicator attached behind it acting as a press wheel worked very well, but lumpy fertilizer ~ slows down the planting operation as the lumps Guard on opener lever. Planter on right has guard inmust be broken up before being used. stalled. 18



PAGE 1

use of growth regulators in shortening the harvest cycle injection of 1 ml of H20 also repeated once after 5 days. interval between first and second ratoon crops. It also Observations were on the mother plant, P1, P2 and P3 at suggested the possibility for rapid multiplication without the time of first injection and 6 weeks after the second loss of yield from the mother plant. Studies in 2 field injection (Table 52). trials were condIucted. The GA treated peepers developed faster than the first peepers on the same stool and the GA injection did not Physiological study of the suckering significantly stimulate the growth of the first and third behavior peepers. No effects could be observed on the main pseudostem. This points to a very localized action of GA on In the first trial, the pseudostem of a 7-month-old mother the treated suckers only, and an immediate change in plant was cut off and the apical meristem removed. The the balance is enough to stimulate the growth. The H20growth regulators; GA (Gibberillic acid, GA + NAA (Naptreated peepers did not show any difference in growth thylene acetic acid), GA + cytokinin and ABA (abscisic compared with the peepers of the neutral control, provacid); were applied on alternative days in lml doses in ing that tissue changes by the injection was not interferconcentrations of 10-2 and 10-4 M. Five peepers per ring with the development of a sucker. mother plant were treated at 2-4 weeks after detopping of the mother plant. Effect of Flurenol on apical dominance Treatments with GA/NAA were not considered because the structure of the peepers seems to be changed. There Flurenol is known to counteract the apical dominance. were no significant differences in height and girth after This effect on plantain was compared with the neutral 4 weeks among the peepers of the 16 treatments (2 ducontrol (no treatment) and decapitation (Table 53). Flurations x 2 concentrations x 4 hormones), except for renol was injected into the 4.5.month-old main pseudothe height of the GA 10-2 M treated peepers. Peepers injected with ABA were not significantly different in their growth from the control. Table 53. Mean height (Ho, H) and girth (Go, G,) of the In the second trial, the mother plant was retained. Treatsecond peeper and the main pseudostem at ment began at 6 months after planting. Only 1 peeper per 0 and 6 weeks (in cm). Number of peepers at plant was treated. Treatments with GA/NAA and GA/NAA 0 and 6 weeks (N., NJ. CYT were not considered because of abnormal changes Control Flurenol Decapitation in the peepers. Suckers injected with GA, GA + flurenol P2 H. 3.7 1.64 1.4 and GA + CYT showed, compared with the control, sigH6 22.7 16.82 57.7 (1) nificant differences in height and girth after 6 weeks. Go* 3.8 1.73 1.8 Suckers injected with GA + CYT became significantly G8 13.2 9.27 16.8 (5) thicker. *Girth at soil level Also, there were no significant differences between duMotherHo 182.6 146.3 (5) 157.9 rations, but there were between the 2 concentrations. As plant H, 264.5 Tf-1 in trial 1, peepers injected with ABA did not differ signifG 44.2 36.78 (5) 39.8 icantly from the control. G*0 37.9 31.4 (5) 32.9 The effects of GA on suckers in Trial 2 proved that the G6* 64.5 52 (1) significant growth of injected suckers is not linked to the G6 1o* 52 42.4 (1) duration of the injection but to the hormone concentraNo 2.3 0.9 (5) 1.4 tion. In the next experiment, 1 ml of Ga 10-2 M was inN, 8.4 6.2 4.6 (1) jected into the second peeper (P2) at 10 days after *Girth at 50cm and 100cm height emergence on 10 different plants. The injection was re"Number of peepers peated once on the same peeper at 5 days after the first. Means in the same row followed by (1) or (5) respectively, The controls were no treatment (neutral control), and are significantly different for 1% and 5%. Table 52. Height and girth of second peeper, mother plant pseudostem and number of peeper at 0 and 6 weeks after treatment. Control Water GA Type of Plant Parameter Start 6 weeks Start 6 weeks Start 6 weeks Peeper2 Height (cm) 8.1 24.6 8.2 28.6 7.6 87.6* Girth (cm) 6.7 13.4 8.3 16.4 6.6 20.1" Peeper, Height 24.8 65.7 --16.7 47.7 Girth 13.1 24.9 --10.2 19.5 Peeper3 Height 5.2 15.5 --4.6 20.8 Girth 5.5 12.9 --4.0 13.5 Mother plant Height 200.1 285.9 200.0 277.5 196.6 285.8 Girth (50 cm) 48.8 68.9 47.9 65.1 49.3 66.3 Girth (100 cm) 41.1 54.7 40.1 53.2 41.9 54.1 No. of peepers 4 10 3.8 9.6 3.8 8.6 'Significant differences between treatments at 5% probability. 47



PAGE 1

inputs and good water control for intensive wetland rice .RVOM production. More fertile rice soils in terms of soil-texture 2. LAFIA and nutrient status comprise about 15-20 percent of the 4.TUAGU 37 soils studied (Tables 2 and 3). Wetland areas with .GMANA "T 30 S. Wu¢, such soils may be developed into productive rice land. 3Zo 6. TUMBIu •More detailed and systematic soil surveys by national 7. SAU organizations are needed in order to determine the extent and distribution of the more fertile wetlands and their feasibility for rice production. Table 2. Texture, organic C and pH of 37 hydromorphlc soils from the forest region of West Africa (southern Nigeria, Sierra Leone and Liberia). Properties Distribution, % and range Surface soil Subsurface soil pH (H20, 1:1) 4.0-5.0 78 73 5.1-5.5 14 22 5.6-6.0 8 5 1 2 3 & 5 676910 20 30 405060 80100 200 AWO am B O TIME (I) Organic C, % 0.5-1.5 38 97 Fig. 9. Accumulative infiltration for some Nigerian soils. 1.6-2.5 24 3 2.6-5.0 30 0 > 5.0 8 0 gime. While hydromorphic soils mean the same to all soil scientists in terms of morphological and chemical propClay content, % erties, the different soil classification systems are still 0-15 46 19 not very interchangeable because their respective con16-25 19 35 cepts are based on different premises. 25-35 14 16 According to U.S. Soil Taxonomy, hydromorphic soils 35-45 8 16 have an aquic soil moisture regime, and they appear 45-55 5 3 under the suborder as well as the subgroup levels of 55-65 8 11 classification. The FAO/UNESCO soil classification emphasizes the concept of gleization, and Gleysols and the Silt content, % "gleyic" subunits constitute the major hydromorphic 0-10 16 24 soils of the "FAO Soil Map of the World." For both the 11-20 22 24 French and Belgian classification systems, the pro21-30 16 14 cesses of gley and pseudogley play prominent parts in 31-40 24 19 the definition of hydromorphism. Moreover, Vertisols in 41-50 12 14 West Africa are also subjected to seasonal waterlogging 51-80 10 5 (pseudogley) and should come under the realm of hydroSource of Data: IITA Soil Information Bank, Njala Univermorphic soils. From the agronomic viewpoint, wherever sity College and University of Illinois Soil Survey Report excess soil moisture is the most important feature in land for Sierra Leone (1974), Ministry of Agriculture and use for crop production, soils in such areas will come USAID Soil Survey Report for Liberia (1977), Manor River underthe realm of hydromorphicsoils. Union Soil Survey Report, Liberia (1979). An approximate correlation of the hydromorphic soils among the 3 international soil classification systems has Forest/savanna transition zone (southern Nigeria). Hybeen prepared at IITA. dromorphic soils in the forest/savanna transition zone A preliminary inventory of some chemical and physical occur mainly in inland valleys and depressions. As the properties of hydromorphic soils from Nigeria, Sierra region is characterized by a rolling topography and Leone and Liberia may be summarized as follows: quartzose-rich basement complex rocks, the wetland valleys or depressions are generally small, and the soil Forest zone (Liberia, Sierra Leone and southern Nidepths are shallow. All the same, they occupy about 10geria). Hydromorphic soils occur in coastal, fresh-water 15 percent of the total land surface. Most of the hydroswamps, inland swamps and small river valleys. Some morphic land in this region presently is not used for food areas are subject to deep flooding during the rainy seacrop cultivation. son. • An extensive study on the soil and land characteristics of Approximately 80 percent of the hydromorphic soils (exseveral inland valleys in the region was completed in cluding acid sulfate soils in this region) are acidic (pH 1979. In this study, groundwater regime throughout the below 5.0) and coarse-textured (loamy sand and sandy year was used as an important criterion for the hydroloam) with low P and K reserves and low clay activity morphic land quality evaluation. N deficiency and Fe tox(Tables 2 and 3). Such soils would require high fertilizer icity associated with groundwater or seepage water were 6



PAGE 1

Table 1. Summary of climatic data (IITA, 1980). Total Total Solar Temperature Relative humidity rainevaporadiation 0C % fall ration G"-cal/ Month mm mm cm2/day Min. Mean Max. Min. Mean Max. Jan. 0 121.22 391.90 23.3 27.8 32.2 46 71 96 Feb. 26.2 150.51* 504.14 23.8 28.9 33.9 39 68 39 Mar. 52.0 174.22 513.70 24.3 28.7 33.0 46 72 98 Apr. 50.1 164.70" 490.16 23.8 28.4 33.0 49 72 95 May 193.0 126.17" 401.85 23.3 26.7 30.1 61 79 97 June 132.2 129.50 421.16 23.2 26.3 29.4 64 80 95 Jul. 203.8 88.04* 318.53 22.2 24.6 27.0 72 84 96 Aug. 367.8 92.07* 325.16 22.4 24.9 27.3 73 85 97 Sept. 248.9 104.10* 384.56 22.3 25.5 28.7 67 83 98 Oct. 208.4 107.88 418.96 22.5 26.0 29.4 64 81 97 Nov. 66.7 107.10* 408.28 22.5 26.3 30.0 56 76 97 Dec. 0 117.58 386.25 20.1 25.3 30.5 41 67 92 Walues adjusted for days with missing data. 400-IITA, 0 24"---University of Ibadan 20-year ----o RAINFALL 380-~~ avrg x EVAPORATION average 360.20340 "E S 16 320' 4I., 3_ 0 2z 8 260" -6 240 -2 22 JAN F'EB MAR APR MAY' 'JN JU;L' 'AUIG ~ 'C E T IME 180 Fig. 2. Weekly mean rainfall and evaporation (IITA, I. -\ 1980). o-\ This spell of moisture deficit was followed by a period of / V excessive and persistent rainfall, which culminated in 100the highest monthly peak for the year and the third highest on record (367.8 mm for.August). A singular rain in80/ cidence on 31 August resulted in an unprecedented and catastrophic flood in the general area. / With the failure of the 'normal' August break in the rains, 40first-season crops, with potential yields already depressed, suffered further damages through lodging, im2 / proper drying and delayed harvesting. The excessive moisture also caused considerable delay in land prepa'JAN 'FEB MAPR'MAY'JUN JUL USEP' CT NOVDEC ration and second season planting and persisted TIME through September and October. The rainy season Fig. 1. Mean monthly rainfall, ended in early November. The cumulative total for the year was 1,549 mm or 22 percent above the long-term mean. Pan evaporation, on the other hand, was 7 percent of May, about a full month later than normal (Fig. 2) with less than its mean, totalling 1,483.09 mm. a corresponding delay in the start of the first cropping Sky conditions and solar radiation. January and Februseason. ary were unusually cloudy, but solar radiation (Table 1, Precipitation remained quite regular through the first 2 Fig. 3) was higher than normal. Cloudiness increased, weeks of June but tapered off again in the last third of particularly in the second half of April, in spite of the the month and continued so through the first of July. limited rainfall. 2



PAGE 1

G. Moldboard plowing followed by 2 harrowings (resMaize grain yield declined sharply in a field with its idue on the surface). twelfth continuous crop of maize (2 crops per year). H. Moldboard plowing followed by 2 harrowings, folHowever, the rate of decline was much greater in conlowed by ridging (residue plowed in). ventional than in no-tillage plots (Fig. 11). Soil erosion The effects of these treatments were investigated for soil has been a severe problem for yield decline in convenphysical properties and crop response. Preliminary retional-tillage plots while soil compaction has been a sesuits indicated that for a sandy loam soil, tillage methods vere problem in no-tillage plots. had no significent effects on grain yield (Tables 11 and 12). However, it will be useful to conduct a similar study for different soils and ecological conditions to provide 50 guidelines for appropriate tillage methods. BLOCK 00 NO TILLAGE UNTERRACED Table 11. Effects of tillage methods on grain and stover CONVENTIONAL TILLAGE-TERRACED yield (IITA, 1980 First season). 40Grain yield Stover yield Harvest index Treatment tlha tlha % A 2.52 A 6.3 AB 39.8 A B 2.00 A 5.2 B 39.8 A C 2.69 A 7.7 A 35.0 AB 030 D 2.41 A 6.7 AB 36.0 AB 0 E 2.14 A 5.9 AB 36.2 AB F 2.62 A 7.3 AB 36.5 AB G 2.07 A 7.7 A 27.0 B H 2.25 A 6.9 AB 33.0 AB LSD (5%) 0.82 2.1 10.5 9 20 Table 12. Effects of tillage methods on grain and stover yield (IITA, 1980 Second season). Grain yield Stover yield Harvest index o Treatments t/ha t/ha % A 2.40 A 4.80 A 51.8 A B 1.81 A 4.32 A 43.5 A C 2.00 A 3.65 A 53.3 A D 2.12 A 3.53 A 60.2 A E 1.80 A 4.22 A 42.3 A 1975 1976 4977 1978 1979 1960 1981 1982 1983 1984 F 1.81 A 3.58 A 49.9 A YEAR G 1.97 A 4.54 A 43.6 A Fig. 11. Maize grain yield (Cv TZPB) under conventional H 2.05 A 4.27 A 49.3 A and no tillage on a kaolinitic Alfisol cleared from secLSD (5%) 1.01 1.81 17.7 ondary forest. Wt D11. M|Il ( Poor stand In conventionally tilled maize with water Maize planted at the same time as the conventionally standing for up to 2 weeks after rain because of comtilled maize, it has a good crop stand and compaction paction and low permeability, was not severe enough to reduce infiltration. 12



PAGE 1

2000 2000 plication at the rate of 60 kg K/ha. N application only LSO (5%) slightly increased tuber yield. TMS 30572 also gave higher tuber yield than TMS 30555 at this particular location. There is no difference in the percentage dry matIter yield between the 2 varieties. Differential P response of cowpea and soybean varieties. Investigations were carried out at Ikenne -in the A humid region of Nigeria to determine differences in the -internal and external P requirements of 4 cowpea varinz I000: -0-Table 35. Effect of fertilizer and preemergence herbicide on maize grain yield (IITA, 1980). ABkg/ha c Full basal & side dressing with Primextra 3333 a. Basal P only and side dressing with Primextra 2970 a b No fertilizer and No Primextra 2727 b c Side dressing alone-No Primextra 2545 b c Full basal & side dressing without Primextra 2303 c Basal P only & side dressing without SPrimextra 2182 c d 0 o50 K s T5ONS 0 TONS No Fertilizer and with Primextra 1727 d N. RATE LEUCAENA RATEHA Side Dressing alone with Primextra 1091 e KG N/HA Desn *Means having common letters are not significantly difFig. 25. Effect of N and Leucaena rates and application ferent at 5% level according to Duncan's Test. methods (A and B Leucaena tops banded at, respectively, 25 and 50 cm width and C Leucaena tops broadScasted) on maize grain yield on Apomu soil series Table 36. Effect of fertilizer and preemergence herbi(Psammentic Usthorthent). cide on weed weight at maize harvest (IITA, 1980). In-situ mulch from cover crops and fertilizer levels on kg/ha maize yield. In-situ mulch from cover crop residue has No fertilizer with Primextra 1803 a' been found to be a feasible method of establishing Side dressing alone with Primextra 1741 a mulch for effective no tillage or mulch-conventional tillBasal P only and side dressing with Primextra 1437 a age production. Using Mucuna utilis as the cover crop, Full basal & side dressing-No Primextra 1144 a the effect of fertilizer and preemergence herbicide on Basal P only and side dressing alone-No maize yield in mulch-conventional tillage was evaluated. Primextra 1112 a The design was a split plot in which main plots were (a) Side dressing alone-No Primextra 1061 a no herbicide and (b) a basal application of 60 kg N/ha, 60 No fertilizer and without Primextra 1032 a kg P/ha and 60 kg K/ha with a side dressing of 30 kg N/ Full basal & side dressing with Primextra 877 a ha at 4 weeks after planting; (ii) a basal application of 60 Means having common letters are not significantly difkg P/ha with a side dressing of 30 kg N/ha; (iii) a side *Means having common letters are not significantly difdressing of 30 kg N/ha only and (iv) no fertilizer. The ferent at 5% level according to Duncan's Test. cover crop was killed with 2.0 kg glyphosate/ha a.i. Side dressing alone and no fertilizer were found to produce of N and P application on tuber yd significantly lower yields than a full application of basal Table 37. Effect of N and P application on tuber yield with a side dressing and a basal application of P with a of cassava cultivars TMS 30555 and TMS side dressing (Table 35). The preemergence herbicide 30572 on an Alfisol (Oxic Paleustalf), Ikenne, had no significant effect on yield. The results indicate 1980. that some fertilizer is needed in mulch-conventional tillFertilizer Treatment TMS 30555 TMS 30572 age with in-situ mulch from a leguminous cover crop. A kg/ha Tuber yield t/ha preemergence herbicide is not necessary as weed N K Fresh Dry Fresh Dry suppression by the mulch is an effective weed control (Table 36). 0 0 19.52 8.42 19.50 8.02 0 120 21.48 9.42 23.85 .9.51 60 120 23.29 9.39 21.14 9.07 120 120 21.35 9.60 25.42 9.95 Nutrient requirement of crops on Alfisols 120 0 18.29 8.47 21.83 9.79 N and K responses of cassava. The N and K response 120 60 21.56 9.78 26.92 11.68 Study of cassava conducted on an Alagba series (Oxic Mean 20.92 9.18 23.11 9.67 Paleustalf) was initiated in 1978 at Ikenne. The 1979-1980 LSD (5%) Variety means fresh tuber 3.08; dry tuber 1.85. cropping season results are shown in Table 37. On this Between fertilizer treatments within variety: plot, which is in the second cycle of continuous cropfresh tuber 3.09; dry tuber 2.11. ping, the fresh tuber yields of both varieties, TMS 30555 Between fertilizer treatments among variety and TMS 30572, showed a significant response to K apfresh tuber 4.37; dry tuber 3.74 27



PAGE 1

t..ft opener /ever. Planter on left has 50 mm wide lever, -jni .:er on right a 25 mm wide lever. A-,." :.. Single-row rolling injection planter with the fertilizer .... ... 7, -:band applicator attached. : : -, .Single-row rolling injection planter with fertilizer band .! "..' !applicator row marker. Two-row rolling injection planter. : i U" AI / -I,-.,. Swo-row rolling injection planter with the fertilizer band applicator attached. ac 19 Sigerwrligijcinpatrwt2etlzrbn



PAGE 1

production, subsidy levels of 97 and 99 percent, respecstover plots but not in the live mulch plots. These results tively, would be required. confirm the finding of last year that live mulch crop proThe results of this modeling exercise must be regarded duction has potential for minimizing the need to control weeds in maize. The highest maize yield was observed in as provisional until validated by field trials, but the apa live mulch plot in which Psophocarpus pa/usfris was parent economic attractiveness of the hypothetical Leuaied. caena/rice alley cropping system would seem to justify the research attention of agronomists. High maize yields in this live mulch cover were not related to weeding treatment. On the other hand, compaLive mulch systems rable yields were obtained in no-tillage and maize stover plots that were weeded at least twice. Yield in the conLive mulch crop production involves planting a food ventional-tillage plots was significantly reduced even crop directly into a living cover of an established cover when the plot was kept weed free. crop without tillage or destruction of the fallow vegetation. This incorporates the soil conservation features of organic mulch and no tillage and has the advantage of cover on the N fertilizer requirement of maize was investigated. All plots received a blanket application of 30 kg smothering weeds and contributing N in the case of a P20, and K20/ha. Maize responded to varying levels of N legumn live mulch. fertilizer applied in the conventional and no-tillage and Weed competition, tillage and ground cover. Observamaize stover plots but not in the live mulch plot in which tions in 1980 showed marked differences in weed bioPsophocarpus palustris was the ground cover (Table 49). mass due to the weed control method and ground cover type in maize. Although the field dry weight of maize Table 49. Effect of N fertilizer level and ground cover stover used as ground cover averaged 10 t/ha, this was on maize (TZE 4) yield (IITA, 1980 First seanot enough to completely eliminate weed growth. Weed biomass was identical in the unweeded check of the conventional and no-tillage plots, and these were signifiN-fertilizer Grain yield cantly higher than in the unweeded check of the live Ground cover (kg/ha) (t/ha) mulch plots (Table 48). Conventional 0 1.24 d' 60 1.49 cd 120 1.51 cd Table 48. Effect of weeding frequency and ground No-tillage 0 1.63 cd cover on weed competition and maize (TZE 60 2.70 a 4) yield (IITA, 1980 First season). 120 2.51 ab Grain Maize stover 0 1.74 cd Weed Weed D. wt. yield' 60 1.98 bc Ground cover control (t/ha) (t/ha) 120 2.49 ab Conventional Weed free 0 c 1.54 cde Arachis repens 0 1.09 d tillage Weed x 2 0.8 b 1.65 cde 60 1.72 cd Unweeded 120 1.45 cd check 1.49 a 1.05 e Psophocarpus 0 2.47 ab No tillage Weed free 0 c 2.57 a palustris 60 2.39 ab Weed x 2 0.83 b 2.43 ab 120 2.41 ab Unweeded Average stand count of 50.6 x 103/p/ha check 1.4 a 1.84 bcd 'Means followed by the same letter in the same column Maize stover Weed free 0 c 2.27 ab are not significantly different at the 5% level of the New Weed x 2 0.91 b 2.39 ab Duncan's Multiple Range Test. Unweeded check 1.31 a 1.56 cde Arachis repens Weed free 0 c 1.36 de ,.A"",k' j* Weed x 2 0.27 c 1.52 cde Unweeded 94 AX. check 0.31 c 1.32 de t Psophocarpus Weed free 0 c 2.63 a palustris Weed x 2 0.05 c 2.50 a Unweeded check 0.11 c 2.14 abc Average stand concent of 50.27 x 103 pl/ha. 'Means followed by the same letter in the same column are not significantly different at the 5% level of the New ... .. .Duncan's Multiple Range Test. CGA 41Z%3 2 0 KG ,iA The unweeded check, live mulch plots had fewer weeds .(Lo-6oat harvest than in each of the conventional and no-tillage and maize stover plots that were weeded twice. Uncontrolled weed growth significantly reduced the maize Typical live mulch stand in one of the field experiments grain yield in the conventional and no-tillage and maize in 1980. 42



PAGE 1

'Table 5. Estimated average labor utilization for differtheir landuse. The dominant and most widespread types ent food crops per operation in Nigeria. of traditional farming systems are permanent compound farming, rudimentary sedentary agriculture (with or withFrom literature out short fallow periods), bush fallow cultivation and sources From field surveys traditional permanent tree crop farming. Shifting Crop mandays/ha mandays/ha cultivation hardly exists, and taungya is hardly used in Upthe region. Livestock is relatively insignificant. The bush Casland CocoSoyfallow cultivation and the rudimentary sedentary agriculOperation Maise Yam sava rice yam bean ture are the most widespread systems of land use. Permanent compound farming is comprised of tree and Land preparation 24 95 40 55 36 35 arable crop farming enterprises with livestock as a third Planting 10 35 13 30 14 15 and minor activity. Land allocation is essential for homeFert. application 15 15 5 12 steads, and tree and arable crop farming with livestock Weeding 25 70 45 50 38 28 is confined to homesteads and villages. Harvesting 16 60 70 35 60 35 Outside the permanent compound farms, except for reTotals 90 275 168 175 148 125 tained (wild) economic tree crops such as oil palm, planted tree crops and arable crop farming are not spatially integrated. Hence, tree crops, once planted, occupy Agroforestry surveys the land permanently while the arable crops are largely An agroforestry field survey was undertaken in southern grown in the traditional bush fallow and/or rudimentary Nigeria to identify major types of landuse in traditional sedentary agriculture with a land use factor for uplands farming systems and establish the degree of integrationI between (a) forest and tree crop plantations and small (L -ranging from 2.5 to 3.6 (Table 6). Thus, outholder tree crop farming and (b) traditional arable crop farming and livestock production systems. Economic side the permanent compound farms and traditional perwoody species, both cultivated and noncultivated, and C their association with traditional arable crop farming manent tree crop farms, the land use ratio (R = C x received particular emphasis during the survey. The 6 survey locations-Ezzamgbo, Umudike, Onne, Ikom, 100) in eastern Nigeria is less than 34 percent for uplands /$Imkpa and Uyo-are in the humid tropical zone, exand 51 percent for bottomlands (Table 6). cept Ezzamgbo, which is in the transition from the tropiIn terms of land allocation, planted tree crops (including cal wet to the tropical wet-and-dry climatic zone. plantains and bananas) accounted for about 67 percent Landuse in traditional farming systems. Both traditional of the cultivated land; tree and arable crops mixture for agriculture and rural settlements are largely confined to 7 percent; and arable crops for only 25 percent (Table 7). uplands with bottomlands seldom used for agriculture. About 76 percent of the farmers had arable crop farms All farmers in the survey areas have rain-fed upland under 4 ha, 45 percent of which were under 2 ha. Simifarms. Only 23, 13 and 23 percent of the farmers in the larly, about 56 percent of the farmers had tree crop farms derived savanna, lowland forest and coastal lowlands, under 4 ha. Tree and arable crop mixtures were largely respectively, have additional farms located in bottomconfined to farmers with the smallest amount of cultilands. Permanent compound farm plots, common in the vated land. central zone of the region and in the settlement sites, are Cultivated areas devoted to the important tree and arable typically located on uplands or well-drained sites. The crops areasesed forah trey ation. most common type of settlement is the dispersed homecrops grown were assessed for each survey location. stead type, which is also correlated with the prevalence Tree and arable crops most frequently grown with considerable acreage, in the order of importance for each of permanent compound farms. group, were (a) oil palm, cocoa, plantain/banana, kola, Major types of traditional systems of agriculture and rubber and citrus and (b) cassava, yam, maize and rice. Table 6. Small holder land use in eastern Nigeria: Type of land cropped and cropping/fallow duration (1980). Derived Savanna Lowland Forest Coastal Lowland Ezzamgbo Ikom Umudike Akamkpa Onne Uyo BotBotBotBotBotBotUptom Uptom Uptom Uptom Uptom Uptom land land land land land land land land land land land land Enterprise/Farmer Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Yr. Cropping year, average 2 1.5 1.8 -2.20 1 2.10 1 2 1.3 2.15 Cropping year, range 1-5 1-8 1-3 -1-3 1 2-3 1 1-3 2-8 2-3 Fallow year, average 3.2 1.3 5 -4.9 1.25 4.9 1.35 4.9 1.3 4.1 Fallow year, range C + F' 2-5 1-3 3-7 -5 6 3-7 1-5 2-7 1-8 2-7 Land use factor (L = C ) 2.5 1.9 3.6 -3.25 2.25 3.33 2.35 3.45 2.08 2.9 Frequency of upland and bottom lands (% of farmers using) 100 35 100 -100 5 100 20 100 50 100 5 '(See Okigbo and Greenland, 1977 for classification) 8



PAGE 1

Farming Systems Program Introduction A primary goal of the Farming Systems Program is B sln aac leto the development of methods of land use and crop B s ln aac le to management that will enable efficient, economic and analysis and sustained production of food crops for the humid and subhumid tropics. Research is directed primarily at Research on baseline data collection and analysis-inproblem solving the constraints of small farmers, many cludes the following areas: agroclimatology, soil and of whom still rely on bush fallow systems for producing ln hrceiainadeauto n oieooi the bulk of the food in the humid and subhumid tropics landaratrzainansvlatoindscoeooi of Africa and elsewhere. Emphasis is given to developing anls. technologies that are scale neutral so that they can be Research in 1980 emphasized crop water requirements used by a range of farmers. as well as the development of a soil evaluation system for highly weathered soils in West Africa and studies of Because of the wide range of farming system forms in soil erodibility. Socioeconomic analyses included food the humid and subhumid tropics, the program will not crop and agro-forestry farming systems surveys. develop specific local blueprints for improved methods of land, soil and crop management; rather, the program's concern is to develop and make available preliminary 1ITA general weather conditions technologies and subsystems that can be modified and adapted by national and regional organizations to the The nonoccurrence of the late *July-August dry season agronomic and economic as well as political conditions was the dominant feature of the weather in 1980. As with of their own areas. Considering the diversity of the farmsimilar cases in 1973 and 1979, the only other 2 of the ing systems in the Institute's mandate area, baseline data decade, this feature was again associated with a general collection and analysis are undertaken to better delineshift in rainfall pattern; the second rather than the first ate the major benchmark areas with relevant bioclimatic season constituted the major rainy season. There were and soil parameters in relation to prevailing cropping consonant departures in isolation, the heavy rainfall pesystems. The benchmark sites will be used for determinriod generally receiving less than normal global radiaing typologies and testing principles of land and soil tion. The mean temperatures were comparatively high management and cropping systems. along with the mean relative humidity while the evapoThe eserchempasi oftheproramis n asising rative demand was lower than the long-term mean. A farmers in the move from the subsistence shifting cultismayo h anvralsi ie nTbe1 vation and particularly bush fallow systems, to more continuous and productive systems of cultivation with Rainfall and evaporation. A streak of rainless days appropriate land and soil management practices, which extending back to November, 1979, was broken with a will maintain soil productivity and minimize soil erosion 26.2 mm rain on 14 February. There were 4 other rains and soil degradation. before the end of March (Table 1, Fig. 1), and the first quarter of the year was, thus, marked by a pronounced To achieve more impact in the immediate future, the promoisture deficit with a -44 percent departure in cumugram focused its activities in the following research lative rainfall in conjunction with a -7 percent differareas: baseline data collection and analysis, land and ence in cumulative evaporation. soil management and cropping systems. No rain of substance fell in April, and total rainfall was Cooperative programs have been established in Camefully 66 percent below normal. It was a clear case of a roon and Ghana. Reports of these projects are included "false start" in the rains, causing most of the late March in cooperative projects, and individual reports are availplantings to succumb to drought stress (Fig. 2). Favorable from the respective organizations. able water balance was first established at the beginning



PAGE 1

EXCH. CA, meq/100g grown. Thus, it is recommended that for Ultisols (Typi 0 0.5 1-0 1.5 2.0 2.5 3.0 3.5 Paleudult) or for Ultisols with similar mineralogy and tex 0-15 ture, annual rates of 200-400 kg/ha should suffice t( maintain maize and cowpea yield. Under such circum stances, lime could be regarded as a form of fertilizer E 15-30. rather than a major soil amendment. 30-45Initial Lime Rate (1976) Leaching of nitrate under maize. /An experiment was established at Onne in 1980 to study S-2 tlh the pattern of nitrate leaching under field conditions, S5"60 --Z4 t/ha Maize was planted in the first season and upland rice in -jthe second season. Three methods of N application were X 60-75 used, 1 application at planting, 2 split applications: one0. half at planting and one-half at 4 weeks after planting and 3 split applications: one-third at planting, one-third 75-901 at 4 weeks after planting and one-third at 8 weeks after planting. A N rate of 150 kg/ha as calcium amonium nitrate was used. A bare fallow treatment was included. EXCH. AL, meq/100g The experiment was a split-plot design with 4 replica0-150 0.5 0 1. 2.0 25 30 3 tions. Lime and unlimed treatments were the main plot. Downward movement of inorganic N (NH4-N and NO-N) 150 in the soil was monitored periodically up to a 120 cm 15-30 depth (Fig. 31). At the end of 4 weeks (129 mm rainfall), E the pattern of nitrate movement when all N was applied Ul .30-45 at planting in the unlimed plots was similar to the bare S3-fallow and the crop plots. The peak concentration oc2: curred between 30-60 cm. At the end of 8 weeks (477 mm 45-60 rainfall), the nitrate peak in the cropped plots occurred z at a depth of 60-90 cm, but the nitrate peak in the bare fallow plots was considerably deeper and broader, indi60-75cating greater leaching. At the end of the first season W (940 mm rainfall), the nitrate peak moved to a depth of 0 1105-120 cm. Liming increased the rate of nitrate leaching 75-90. in both bare fallow and cropped plots. When N was split into 2 applications, nitrate peaks occurred in the upper Fig. 30. Vertical distributions of exchangeable Ca and Al layer between 0-30 cm both at 4 and 8 weeks after plantat 3 years after lime application. ing, indicating less leaching. Splitting N into 2 applications significantly increased grain yield and N-uptake by the plant (Table 40), but furN03-N in Soil,jg/g 0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25 X-00o 0 15 ii t \ ,, 15 .0 -U.S I, 60 J > / t> U I I I 45. 0 X 0 o O / C E 60. ,os5 0) I o,',,. '" o~ 0 '. NYC > I/t 90 > t I Bore Follow, No Bare Follow, N, Maize, N, Maize ,N2 Fig. 31. Leaching of nitrate in unlimed plots under cropping and bare fallow. (Nono N applied, N,-150 kg N/ha applied at planting, N,-150 kg N/ha split application, one-half at planting, one-half at 4 weeks after planting.) 30



PAGE 1

100 90 80 6. 00" 70. ,5 350_660 -WEEKLY MEAN W o <250" --K-. MONTHLY MEAN 200..30. _0__ MAXIMUM JA FB'AI lY U UL AGSE C NO CI20 -MEAN T.,% IM E 10 M IN IM U M Fig. 3. Weekly and monthly mean solar radiation (IITA, .-A Y 1980). 'JJUI EII AG ''AS i OCT ' 0 i C' Fig. 5. Mean, maximum, and minimum relative humidity During the period May through August, which in effect (IITA, 1980). covers the first cropping season, incident radiation dropped to a mean of 363.7 g-cal cm-2 day-1 or 6 percent siderable lodging and stalk breaking observed during the less than the multiannual mean for these 4 months. The period. rest of the year, except for the month of November, saw Dew/light precipitation. The contribution of dew to the a return to higher than normal insulation regimes. water budget was much less than in the previous year. Temperature and relative humidity. Higher temperaThis is not surprising in view of the record high minimum tures prevailed on the average, mainly as a result of temperatures and the high incidence of rainfall. The higher night-time temperatures, an apparent result of the measured total for the year was 11.59 mm, ranging from more humid and cloudier conditions (Fig. 4). Monthly a monthly maximum of 1.57 mm in November to 0.48 mm average minimum temperatures were record-high in in August. many cases. Agroclimatic analysis 40. Agroclimatic zones of West Africa 30 ... ..To provide a uniform framework for research to facilitate Z the choice of priority areas and determine the range of ,W applicability of research results, an agroclimatic zona20 tion of West Africa was undertaken. It is based on the MAXIMUM concept of water-balance and Franguin's method of in...... MEAN terception of rainfall potential evapotranspiration cure. Months of positive water balance were defined as months with rainfall greater than or equal to potential JAIN'FES I MAR "APR IMAY 'JUN I JUL I AUG I SEP OCT I NOV I DECI evapotranspiration. On the basis of the number of the TIME months determined for the network of approximately 88 stations over the region, 7 generalized agroclimatic Fig. 4. Weekly mean, maximum, and minimum mean air zones were defined (Fig. 6). temperature (IITA, 1980). Commonly used terminologies were retained in describing these zones. It is believed that the basic definition Daytime hours were by contrast cooler. The only aboveused in this classification is not only more meaningful average maximum temperature occurred in April (deparfrom the cropping point of view but also avoids complex ture: +0.60CY, a month of subnormal rainfall. With the indices and is, thus, easier for the user to follow. delay in the onset of the rains, soil temperatures remained rather high well into the month of April with a mean maximum of 41.30C at 5 cm. Crop water requirements. Observed relative moisture contents of air were also There is a scarcity of research information regarding the higher than normal, except for April (Fig. 5). These water requirements for tropical food crops. It is imporhigher humidities evidently account, in part, for the tant, therefore, to define the optimum water requirelower values of evaporative demand. ments of these crops to maximize their productivity An unusual prevalence of southerly and southwesterly under different ecological conditions. winds in January and February constituted the basis for Evapotranspiration and cowpea yields. Many studies on the observed cloudiness noted earlier. In general, wind a variety of crops have tended to show a linear relationspeeds also remained above average throughout the ship between actual evapotranspiration and, more paryear. The incidences of gusts during the wet period, particularly, actual transpiration and yield. Using cowpea, ticularly in late July and in August, account for the conTVu 3629 and TVu 4557, yield figures from a series of 3



PAGE 1

15W low 5W 0 5E IOE 15E LEGEND 20N 1 -20N ( PERHUMID (P!ETp for months or more) uI HUMID: (P2ETp for period between 6-8months) TRANSITION HUMID/SUBHUMID: (P_ ETp for period 5-6 months) SUBHUMIDO. (PETp for period between 4-5 months) SEMI-ARID:(P>ETp for period between 2-4 months) 4.5 ARID: (PETp for period between 1-2 ION + ION months) DESERT:(P ETp for period less thanone 7 month) NATIONAL BOUNDARIES ISOLINES OF NUMBER OF HUMID (P>ETp) 5N 4 + 5N MONTHS BOUNDARY OF BIMODAL RAINFALL/CROPPt O7 IN0R A 5 ow 5w 0 I E8 7 ISE ING AREA STATIONS Fig. 6. Generalized agroclimatic map of West Africa. (Source: T. L. Lawson, IITA, 1979). previous sequential planting experiments (IITA Annual Pest and disease problems are believed to be partly reReport, 1976), a preliminary attempt was made to model sponsible for this. Possible inaccuracies in the evapocowpea yield using estimated evapotranspiration. transpiration estimates may be equally important. Evapu ranspiration estimates were obtained by assuming that the potential evapotranspiration is equal to Class A pan evaporation, and the actual mean weekly evapooo1600 1977PLANTINGSEASON transpiration is proportional to the ratio of the actual VARIETY ,, IOWEEKS available soil moisture to the available soil moisture at 400.oo. field capacity. The 1976 results show a much better relationship than the 1977 results between the estimated zoo ev'Apotranspiration and cowpea yields (Figs. 7 and 8). 2000 000 0800 1976 PLANTING SEASON x VARIETY I, ISt -10 weeks a *o VARIETY2, I -I0 weekS --1600 400 1400 200 $oo 1200 20 40 60 8o 100 120 1410 160 leo 200 TOTAL EVAPOTRANSPIRATION (SET) Z X oo000 fig. 8. Cowpea yield vs. evapotranspiration (1977). W 600 -Dry season evaporative demand and cassava yields. Cassava planted in a given year normally grows through 600. / the dry season and is harvested at the beginning of the SA rains the following year or thereafter. Although the plant / 400o o reduces its canopy during the intervening dry season, no / e sign of wilting is observed on the leaves of the remaining / foliage. On the hypothesis that the plants probably 200 o. achieve this only by drawing on reserve resources, it 100 ,would appear that yields might tend to decline in propor20 ..2 2' tion to the length and severity of the dry season. Using 0 20 40 60 80 100 120 140 160 180 200 220 240 260 TOTAL EVAPOTRANSPIRATION ( ET ) cumulative Class A pan evaporation (Ep) as a measure of the intensity of the dry season, an analysis was made to Fig. 7. Cowpea yield vs. evapotranspiration (1976). relate the reported decline in the average yield of the 10 4



PAGE 1

Table 40. Maize grain yield and N recovery as affected by methods of application (IITA Onne substation, 1980). Estimated fertilizer N removed by crop N in soil (0-120 cm) Estimated recovery of Grain yield (grain + stover) at harvest + applied N** Treatment# kgiha kg/ha kg/ha % LoN, (Bare) 70.6 47.1 LoN, 2,492 c 54.5 d 58.9 75.5 LoN2 3,208 ab 73.7 bc 43.0 78.5 LoN, 3,376 ab 82.7 ab 55.3 92.0 L N, 2.853 bc 61.4 cd 32.9 62.5 L N, 3,499 a 81.6 ab 32.6 76.3 L N, 3,717 a 92.4 a 59.1 100.8 #Lo = unlimed: L = limed, 2t/ha; N rate 150 kg/ha; N, = one application at planting; N2 = 2 splits; N3 -3 splits. + + Corrected for mineral N (NH4-N and NO3-N) in soil (0-120 cm) from unfertilized bare fallow plot (190.2 kg/ha). **Crop removal plus estimated fertilizer N in soil (0-120 cm) at harvest. ther splitting into 3 applications had no significant effect the Ultisol (Typic Paleudult) from Onne was completed on yield though it reduced leaching loss. in early 1980. The primary objectives are to study leachEstimated percentage recovery of applied N is also given ing of nutrients under high-rainfall conditions and deterin Table 40. Loss of applied N in the form of calcium mine crop water use. The high-rainfall condition, which ammonium nitrate under cropping was not as high as prevails at Onne, was simulated during the experimental expected although more reliable measurements on N reperiod by supplementing natural rainfall with irrigation cover can only be obtained with N-i5 tagged fertilizers, using the 1979 rainfall regime at Onne as a basis. Nutrient losses over the 2 cropping seasons were deterWhen N was applied in 3 split applications, nearly all the mined by measurements of fertilizer ions in the leachate applied N could be accounted for by plant uptake and collected as samples of the drainage through the profile, the amount present in the soil (0-120 cm) at the end of and evaporation/evapotranspiration was assessed by the season. At 2 split applications, about 80 percent of the water balance method. applied N was recovered in plant and soil. Maize was planted as a test crop on 3 of the lysimeters It is important to point out that the unfertilized bare faland in the surrounding area to maintain adequate fetch low plot contained 60-80 kg/ha of mineralized N (NH4-N while the remaining 3 were left bare. All except 1 of the plus N03-N) in the surface 30 cm layer for both the unlysimeters was tension drained by applying a suction of limed and limed treatments, suggesting relatively high 0.2 atmosphere at the base of the column. Fertilizers rates of mineralization during the early part of the cropwere applied in each of the 2 seasons to all 6 lysimeters ping season. It appears that application of N fertilizer as follows: may not be necessary at the time of planting during the N (Calcium nitrate): 150 kg/ha, 3 split applications first season if sufficient mineralized N is already present P (Monocalcium phosphate): 60 kg/ha, basal in the surface layer. K (Potassium chloride): 150 kg/ha, basal MG (Magnesium sulfate): 30 kg/ha, basal Lysimeter studies No lime was applied during this preliminary run. As a result, there was poor growth of maize inside the lysiThe installation at ITA of 6 monolith lysimeters (80 cm in meters due to strong soil acidity. On the basis of the diameter and 130 cm deep) with undisturbed profiles of preliminary results, total drainage through the seasons averaged 1,153 mm or 58 percent of the total equivalent rainfall (2,012 mm) received by the bare (uncropped) lysimeters and 977 mm or 49 percent of the total equivalent rainfall received by the cropped lysimeters. The corresponding values of evaporation and evapotranspiration were 832 mm and 1,035 mm, respectively, averaging 3.70 mm/day and 4.60 mm/day, respectively, over the period. Preliminary leaching data of nutrient cations and anions calculated as percentage of total amount applied are given in Table 41. These data indicated that nitrate and magnesium ions leached readily; whereas, potassium ions, were preferentially retained. Further studies are being conducted under maize and upland rice cropping where adequate amounts of lime have been applied to ensure normal crop growth. i& =i' ; :. *, K and Mg response of cassava Monolith lysimeter cutting and installation at IITA Onne Investigations on the K and Mg response of cassava vasubstation. rieties TIMS 30395 and 30211 were initiated on an Ultisol 31



PAGE 1

5-00 -., -IBADAN I ITA 4-OOONNE 3-00 0*00 5 .j .9,U 20 Africa, associated in natural bush fallow for soil fertility of restoration. IC) The results of the first 2 years of screening and evaluation indicate that, in general, the establishment and Fig. 42a. Relative growth performance of selected tree growth performance of the shrub species were better at species at IITA and Onne after 18 months of establishIITA while the tree species were better at Onne (Figs. 42 ment. a & b). Some of the shrub species that performed well are being included in the alley cropping trials. Light regimes in potential tree-shrub species for alley 500IBADAN IITA cropping. Competition for light remains a key factor in i the suitability of different species in mixed stands. Tests -4-00 ONNE were made to determine the light characteristics within E alleys of different tree-shrub species established over the -3.00 past 2 years. The tree rows are oriented roughly eastwest. C m 2-00.100 ... SESENtA GRANDIFLORA 1 OO9 0, SALCHo RNEA CORDoFOLIA O 0 000 0 ALBIEUCAENA LCATAR' '00 z <~C .-o -z U) o GMLN IAOA g th -Al i Z ani rQ .j (D IL U m -. -, Fig. 42b. Relative growth performance of selected treeshrub species at IITA and Onne after 18 months of es-,tablishment. ALBIZIA FALCATARIA GM • -E Figure 43 shows that Albizia falcataria and Gmelanai ar* borea planted in 4 m rows are clearly incompatible with alley cropping once they are fully established because of 20 their strong depletion of incident light. They could be maintained possibly only as planted fallow. ,C. Crops planted at less than 2 m from established Alchoro 2 3 nea cordifolia will also suffer severely from shading. ReaDISTANCE FROM TREE ROWS (m) sonable performance can only be expected at distances of at least 3 m from the tree rows. This is also true of Fig. 43. Light transmission through the canopies of esLeucaenea leucocephala. The distance from Sesbania tablished fallow/alley cropping species in relation to grandiflora can be reduced to 2 m and even 1 m for crops distance from tree rows (trees in 4 m-rows, E-W; OBS in with a fair degree of shade tolerance. In general, for all November). the species, except the latter, cutting back after the first 39



PAGE 1

1980 ANNUAL REPORT INTERNATIONAL INSTITUTE OF TROPICAL AGRICULTURE



PAGE 1

early rair could provide an effective green ground cover, and sorghum. Heavy shading under cowpea retarded less prone to fire hazard, in the savanna regions. The Leucaena growth somewhat but from the standpoint of plants can be eradicated mechanically with the residue risk-aversive small farmers, suboptimal establishment providing mulch suitable for no-tillage cropping. The pogrowth of Leucaena appears to be a small price to pay tential contribution of N to the soil and protein for human for steps taken to obtain an economic yield from their and animal consumption are other factors that warrant fields in a bad year. further investigation of Cajanus cajan in savanna cropWhere mechanized yam culture is likely to be an attracping systems. tive way to reduce the labor costs of yam production, the trials indicate that a 4 m between-row spacing for LeuOn-farm trials-Leucaena-maize caena may be more suitable than the experimental 2 m systems spacing. Wider spacing may also be preferred where farming traditions favor multiple vine support by fewer The main objectives of the long-term, alley cropping stakes such as in the "tent staking" system of Tawari, farm trials that began in 1980 are as follows: (1) to idenNigeria. While allowing for such modifications, it seems tify and enlist the cooperation of a set of participating advisable to maintain a sufficient population of Leufarmers, representing a range of farming systems and caena to provide adequate mulch material, firewood and farmer types involved in the production of yam and other by-products to maximize overall returns. Where maize in Nigeria: (2) to instruct the farmers on the agronfirewood scarcity is a problem, the wood yield may be a omy of the prototypical Leucaena-maize-yam alley cropsignificant component of the net economic return from ping system; and (3) to see to the establishment of the the alley cropping system. The possibility of a double Leucaena hedgerows as an understory intercrop with Leucaena row in the 4 m spacings should be investifirst season maize in preparation for use as live in situ gated. yam vine staking next year. Trials were established on 9 plots at 6 different locations One potential variant, which aroused considerable intervisits est among farmers for whom the labor costs of weeding across the yam belt of Nigeria (Fig. 46). Periodic arviajrfrmmngeetcnsrititheueo are a major farm management constraint, is the use of were made to each site to provide basic guidance and mor poes all aste throvie fas weae nr Leucaena shade to control Imperata cylindrica and other monitor progress. In all cases, the farmers were encourheliophytic weeds. Although this use of alley cropping aged to feel free to modify the system to fit their needs. has yet to receive systematic attention, a possibility If Information on such modifications provides for a better h ye to receie systeaticeion, a apo undestadin ofthe orkng riniple ofappoprate might be to rotate a 1 or 2 year closed-canopy fallow understanding of the working principles of appropriate through a large field set up for alley cropping. If, for alley cropping systems for smallholder conditions. example, the field were divided into 4 sections, threefourths could be in alley crop production in any given year with the remaining one-fourth in rotating fallow. Research is needed to evaluate the actual weed control potential of this alley cropping modification, but farmers indicate any innovation that minimized the herbicide need would be a major step toward making no tillage and mulch farming techniques more widely acceptable. Linear programming model of Leucaenal rice alley cropping system A linear programming model was developed to evaluate the economic attractiveness of an experimental Leucaena/-rice alley cropping system under West African .I smallholder conditions. Using production data representative of upland family rice farms in Sierra Leona and N DERIVED SAVANNA,. :I1 OUIE SAVANNA response data from the Rokupr Rice Research Station in v I SOUTH GUINEA SAVANNA IV NORTHERN GUINEA SAVANNA Sierra Leone, the model was used to explore the relative v JOS PLATEAU profitability of various rice growing activities at 0, 20, 40, VI SUDAN SAVANNA VII SAHEL SAVANNA 60, 80 and 100 kg N/ha from 3 different sources: urea; ammonium sulphate and in situ Leucaena hedgerows. Fig. 46. Map of Nigeria showing ecological zones and The main results indicate that under the conditions of sites of the Leucaena-maize-yam alley cropping system. smallholder production in the model, it is consistently Farm trials in 1980. 1. Lagbe 2. ljaye 3. Osara 4. Tawari more profitable to grow rice with N from Leucaena 5. Yandev 6. Zaki Biam. hedgerows than from either of the 2 mineral N sources. Furthermore, with labor as the limiting production factor, First year trials have yielded a number of preliminary the 2 components of the Leucaena/rice alley cropping fiis Leucaena has been found to be a priardy system always combine in the same economically optifindings, eceahsbe on ob very hardy mal proportions of 0.37 ha (3,700 linear meters) of Leuplant that established well even under extremely adverse conditions. A severe first season drought caused failure caena hedgerow to 1.28 ha for rice, grown in the alleys of the maize intercrop in most sites, but Leucaena between the hedgerows, for an optimum field size of 1.65 growth continued at moderate levels. In order to cope ha. With the economic effects of the drought, farmers atIn order for rice production with urea or ammonium sultempted to substitute other intercrops such as cowpea phate to become competitive with Leucaena-based rice 41



PAGE 1

All machines were designed to either push or pull. OpThe results on the no mulch treatments for both maize erators generally chose to pull. When conditions made and cowpea support the conclusion of the earlier test on penetration a problem, it was improved if the planter was covering of seed. When the seeds are well covered, there pushed. Other trials were carried out to judge the variis higher germination. During planting, it was noted that ability of the rolling injection planter in stand establishseeds are covered better when planting on soil with conment with no tillage. This variability may be caused by ventional than no-tillage. The soil falls back better onto the amount of mulch cover on the soil or by how well the the seed with conventional than no-tillage. On no-tillage seeds are covered. To determine whether the planter was planting, there is usually a small indentation above the covering seeds well enough, maize was planted with no seed where the opener is inserted. This may allow birds tillage. Ten pairs of rows were selected at random. One and rodents to find the seeds more easily and, thus, reof the pairs of rows was carefully dug up and the seeds duce the stand establishment. Further tests will be concounted, then carefully replaced in their holes and.covducted to determine the causes of lower stands. ered with soil. The next row was left and not counted. Out of the total number of rows tested, the average stand The hole openers of the rolling injection planter were count for the seeds dug up was 82 percent. The count tested to know the maximum speed for pushing before for the rows left alone was only 51 percent. It is clear seeds are left on the ground's surface and the advanfrom this test that more work is required on seed covertages and disadvantages of the different types of hole ing. It was also noted that the heavier the mulch cover, openers. Table 29 shows the maximum speed for a the lower the stand count when using the rolling injecformed hole opener, a wedge hole opener and a split tion planter. A test to determine the effect of the amount type hole opener. (Also see Fig. 15.) Both wedge and the of mulch cover on the effectiveness of the rolling injecsplit hole openers had a maximum speed of 3.4 kph while tion planter was carried out on small plots with maize the formed hole opener had only 2.6 kph. This lower and cowpea, separately, replicated 20 times. speed for the formed hole opener was because a small cup near the end of the opener held the seed after it was Tables 27 and 28 show that there is a consistent reducopened causing a delay before the seed was dropped. tion in stand establishment as the amount of mulch cover increases for both maize and cowpea. This reduction in stand as mulch weight increased could be attributed to greater bird and rodent activity on the heavily i mulched fields and/or shading from the mulch. Table 27..Percent stand of maize on varying mulch + weight per hectare (IITA, 1980). = Mulch t/ha Stand percent 0 (Control) 74.4* Fig. 15. Side view of different hole openers for the roll0 47.4 lng injection planter; left to right, formed, wedge and 4 35.2 split. 6 32.4 -8 23.2 10 22.1 Table 29. Performance test of 3 types of hole openers *The control seeds were planted by hand and well covat different speeds of planting. ered between the rows planted and the rolling injection % of Seed placement' planter. Wedge" Split, Kph Cup type 2 Type Type Remarks Table 28. Percent stand of cowpeas on varying mulch 1.69 100 100 100 weight per hectare (IITA, 1980). 2.14 100 100 100 Mulch t/ha Stand percent 2.61 100 100 100 0 (Control) 84.0* 3.00 75.02 97.85 100 0 84.4 3.43 50.07 98.60 97.37 Limiting 4 59.7 4.00 3.2 80.66 83.31 -speed 6 52.9 4.29 13.05 15.00 17.00 8 51.5 'Percentage (%o) seeds deposited inside the hole. 10 41.1 2Formed hole opener carries the seed out, harder to pencontol eedswer platedby hnd nd wll oyetrate the soil; clogged with soil very often. *The coto ed eepatdb adadwl o-3Wedge hole opener frequently clogged with soil. ered between th Ie rows planted by the rolling injection 4 Split hole opener; split opener squeezes out dirt. planter. In the cowpea plots, no mulch treatments were rotovated CDA sprayer. A 2 head, hand-carried sprayer, which had before planting. The control seeds, which were well coybeen built previously at IITA, was modified to give more ered, had the same germination rate as the seeds planted even flow between the heads. The original sprayer had 1 with the rolling injection planter. In the maize plots, the feed line from the bottle to the cross supporting the no mulch treatments were not rotovated before planting. heads. The tube had a T placed in it at the cross support The control seeds had 27 percent more germination than to feed each of the heads (Fig. 16). The combination of the seeds planted with the rolling injection planter. low pressure in the spray bottle and the resistance to 20



PAGE 1

grain yield was maintained at about 3.5 t/ha for the 2 LSD(5%) seasons (total yield in 1979 was 3.6 t/ha.) The addition of 60the Leucaena prunings has a distinct effect on soil orI MAIN SEASON ganic matter and N levels (Table 46.) MINOR SEASON Effect of legume species and alley width. Legumes es50MINOR SEAON. tablished between maize did not reach sizes at which the ( quantity of leaves was regarded as sufficient to contribute significantly to soil nutrient levels. Alley width had no a 40significant effect on legume development during early stages. The leaf yield and potential N contributions are z shown in Table 47. The leaf and N yield is related to alley 3 owidth. Because of the low potential N contribution, the plot was fertilized with 60 kg N/ha, 20 kg P/ha and 30 kg K/ha after = 2 oa soil test. The plot was Rome plowed and maize planted with a single-row rolling injection planter in the 225 cm alley and with a 4-row rolling injection planter in the 10 wider alleys. The spacing was 75 cm x 25 cm. The legumes were 75 cm from the nearest maize. Maize yields calculated with a correction for land devoted to legumes ON ON 40N 8ON are shown in Fig. 41. Neither alley width nor legume spe-LEUCAENA 3N 6N cies had a significant effect. It would appear that it is too PRUNINGS early to observe the effects of the different treatments. N -RATE, KG/HA Selection and evaluation of woody species for alley Fig. 40. Effect of N rates on maize grain yield from cropping systems. The usefulness of fast growing maize/Leucaena alley cropping system on Apomu soil woody and herbaceous legumes in alley cropping sysseries (Psammentic Usthorthent). (Main season N rates: tems is being evaluated on the Alfisols at IITA and on the 0, 40 and 80 kg N/ha; Minor season N rates: 0, 30 and 60 kg N/ha). LSD (5%) LSD (5%) MAIN SEASON MAIN SEASON 250 6.0MINOR SEASON MINOR SEASON 5.0200 2.0 _150 LU3O0 2 IU00>50 1'O ON 40+ 80+ ON 40+ 80+ 30N 60N 30N 60N DRY MATTER YIELD NITROGEN YIELD Fig. 39. Leucaena dry matter yield and N yield from maize/cassava alley cropping on Apomu soil series (Psammentic Usthorthent). (Main season N rates: 0, 40 and 80 kg N/ha; minor season N rates: 0, 30 and 60 kg N/ha). 37



PAGE 1

200otrozine 1I30 kg/ho al 30kg/ha 18060Okg/ho fluometuron 60Okg/ho 0J 0 S160(a) Q -(C) 0140 0 ILL LL 01200 a 1l00 Cl ~80 ~60 W 140-C) U20-LL 0 1 2 3 4 8 12 0 1 3 4 8 1 WEEKS AFTER TREATMENT 200 ED 3-0 kg/ho Ol25 kg/ho 180metolachlor 6-0 kg/ha pendimethalin 5-0 kg/ho o 0 .160(b) (d 0 0 0140-0 LL U0 120< -<(D 80 I C ~60-lt u < 40 C. 0 L 2 3 4, 8 12 0. 1 2 3 4 8 12 WEEKS AFTER TREATMENT Fig. 50. Effect or rate (kg/ha) on herbicide persistence in soil. 46



PAGE 1

Without N fertilizer, the maize yield was twice as high in yield (Table 50). When weed interference was limited to the live mulch plot in which Psophocarpus palustris was the above ground vegetative parts of both weeds and the ground cover as in the conventional-tillage plot. crops, only a slight yield reduction was observed; but Maize yield in the no-tillage and maize stover plots withwhen the leachete from the roots of weeds was allowed out N fertilizer was as low as in the conventional-tillage to come in contact with the soil in which the yam grew, plot. The advantage of no-tillage over conventional tillyam tuber yield was significantly reduced. This yield reage under continuous cultivation is demonstrated only duction caused by allelopathy was consistent for the 3 when N fertilizer was applied. Without N fertilizer, the years in which this study was carried out. maize yield was as poor in the no-tillage plot as in either When moisture was not limiting and plants received a the conventional-tillage or maize stover plots. Results of basal fertilizer treatment (30 kg N, PO, and KO/ha), the this study show that the best live mulch treatment elimimajor reasons for the yield reduction caused by weeds nated the need for a fertilizer application, and the N con'were allelopathy and other soil related factors. Each of tributed by the legume cover in the live mulch maize was these 2 weed interference factors accounted for over 30 comparable to the 60 kg N/ha in the no-tillage maize. percent of the total loss caused by weeds (Fig. 47). KeepThe study of the live mulch production of maize shows ing crops weed-free appears to be the only way to minithat (a) continuous maize production is possible under mize the yield reduction caused by weeds. the live mulch system without a bush fallow, (b) weed Weed interference studies. A study showed that a sigpressure can be minimized, and (c) the requirement for nificntecline uleaf A ind so)ad tbe y of N fertilizer can be minimized or eliminated. These findnificant decline in leaf area index (LAI and tuber yield of ings have far reaching implications among small-holder white yam occurred due to weed inference from 8 to farmers who can neither afford the labor for weeding nor about 16 or more weeks after planting in tubers planted the chemicals for fertilizing. Live mulch is an attractive at the onset of rains (Table 51). crop production practice that provides complete, yearround ground cover, thus minimizing soil erosion and soil structure loss. Weed biology and herbicide A ABOVE GROUND INTERFERENCE research B ALLELOPATHY C -NUTRIENT/ MOISTURE?) INTERFERENCE Biology of tropical weeds 60. Little is known about the biology of weeds in relation to their control in tropical agriculture. Some aspects of N weed research focus on reasons for crop yield reduc._ tions caused by weeds. Allelopathy and other factors of = 40. weed interference have been investigated in selected tropical crops to develop effective weed control strategies. Allelopathy weeds vs. yams. Since previous observa' 20. tions have shown that yam (D. rotundata) is sensitive to weed interference, studies were carried out from 1978 to 1980 to assess the nature and extent of this sensitivity. A new method was designed and tested for assessing allelopathy in root and tuber crops closely simulating field A B C conditions. Results of these studies show that full weed interference caused a 75 percent reduction in yam tuber Fig. 47. Weed interference in yam. Table 50. Mean fresh tuber yield and yield losses of white yam grown in wooden boxes as affected by mode of -weed interference (1978-80) Mode of weed Fresh tuber yield Yield loss interference 1978 1979 1980 1978 1979 1980 kg/std% Interference by aerial factors 1.83 abl 0.72 ab 2.48 a 16.44 20.88 11.70 Interference by alleloPathy + aerial factors 1.26 bc 0.58 b 1.38 b 42.47 36.26 37.80 Full interference 0.53 c 0.22 c 0.47 c 75.80 75.82 78.80 Weed-free check + Polythene 2.19 ab 0.91 a 2.22 a 0.00 0.00 0.00 Weed-free check Po1lythene sheet 2.52 a 15.07 Means followed by the same letter in the same column are not significantly different at the 5% level of the Duncan's New Multiple Range Test. 43



PAGE 1

Table 23a. Effect of tillage and weed control on maize tillage. The hand weed plot under no-tillage compared yield and crop performance (Ikenne, 1980). favorably with the preemergence herbicide treatments Stand under conventional-tillage. This is an indication that count good cassava yield is possible under no-tillage condi(x 103 Lodging Grain yield tions provided that weeds are controlled. The best weed Treatment pl/ha) % t/ha control in cassava was obtained when a mixture of metolachlor and fluometuron was used. This yield did not Plow and harrow 25 a' 40 a 1.91 ab differ significantly from that of the formulated mixture of Disc harrow 25 a 37 a 2.06 a atrazine and metolachlor, which has previously been No tillage 18 21 1.81 b shown to be safe for maize/cassava intercrops. Herbicide 22 33 a 2.03 a Hand weeding x 2 25 a 34 a 1.99 a Unweeded check 20 32 a 1.76 Table 24b. Effect of weed control on weed biomass 'Means followed by the same letter in the same column and cassava root yield in an Alfisol (IITA, are not significantly different at the 5% level of Duncan's 1980) New Multiple Range Test. Cassava Weed D. Wt. root yield Table 23b. Effect of tillage and weed control on maize Weed control t/ha t/ha yield and crop performance (Ikenne, 1980). Atrazine plus metolachlor 3.18a' 19.53 b Stand Fluometuron plus metocount Grain lachlor 3.88 a 20.89 b Tillage Weed X 103 Lodging yield Diuron plus paraquat 3.97 a 19.58 b practice Control pl/ha) % t/ha Weed free 0 31.79 a Unweeded check 3.39 a 12.27 c Plow and 'Means followed by the same letter in the same column harrow Herbicide 26 a, 41 a 2.12 a are not significantly different at the 5% level of Duncan's Handweeding New Multiple Range Test. x 2 27 a 40 a 1.96 abc Weedy 22 bc 40 a 1.66 c Disc harrow Herbicide 25 ab 34 a 2.20 a Handweeding Small tools development and evaluation x 2 27 a 39 a 2.08 ab Weedy 22 bc 39 a 1.92 abc Rolling injection planter. The rolling injection planter No tillage Herbicide 16 e 22 b 1.78 bc was modified to improve its performance in both the Handweeding conventional and no-tillage systems. First, it was prox 2 20 cd 23 b 1.94 abc vided with a small metal cover to hold the seed on the Weedy 18 de 19 b 1.71 c metering wheel to prevent uneven distribution of seed. This cover holds the seed in the metering wheel until the 'Means followed by the same letter in the same column seed is just over the opener, and the seed is dropped are not significantly different at the 5% level of Duncan's almost directly on the ground through the opener (Fig. New Multiple Range Test. 14). Previously, the cutoff device on the rolling injection planter had a tendency to pinch maize seeds in such a storage root yield was significantly depressed by tillage way that the seeds jumped out of the seed hole in the (Tables 24a and 24b). However, preemergence herbimetering wheel. This happened just as the seed had alcides were more effective in conventional than no-tillage most completely passed under the cut-off device. Also, cassava. Crop yield was significantly higher in the hand the vibration of the machine had a tendency to shake weeded than in the chemically weeded plot under noseeds out of the metering wheel prematurely. This Table 24a. Effect of tillage and weed control on weed biomass and cassava root yield in an Alfisol (IITA, 1980). Treatment Cassava root Tillage Weed control Weed D. Wt. t/ha yield t/ha Conventional tillage Atrazine + metolachlor 3.0 kg/ha 1.85 de' 25.08 b Fluometuron + metolachlor 2.0 + 2.0 kg/ha 1.94 de 28.62 ab Diuron + paraquat 3.0 kg/ha 2.71 bcd 27.27 b Weed free -0 e 35.83a Unweeded check2.13 cde 16.43 c Mean 1.73 26.65 No tillage Atrazine + metolachlor 3.0 kg/ha 4.51 abc 13.98 cd Fluometuron + metolachlor 2.0 + 2.0 kg/ha 5.81 a 13.17 cd Diuron + paraquat 3.0 kg/ha 5.24 ab 11.90 cd Weed free -0 e 27.75 b Unweeded check4.65 abc 8.10 d Mean 4.04 14.98 'Means followed by the same letter in the same column are not significantly different at the 5% level of Duncan's New Multiple Range Test. 17



PAGE 1

(main season crop) followed by cowpea (minor season Data on the soil K-status with K application and cropping crop) annual rotation was practiced during the last 6 showed significant changes on both soil types (Fig. 21). years. Though P and N responses on the sandy loam The soil K-status showed faster depletion with continuEgbeda soil were observed, respectively, since the first ous cropping compared to a bare uncultivated treatand third years following land clearing, a significant rement, particularly on the Egbeda soil. Annual application sponse to K application on the maize crop was not obof 80 kg K/ha was able to maintain the soil K-status of the served until 1980, the ninth year after land clearing (Fig. Egbeda soil at a level comparable to that observed under 20). bush fallow. However, on the Apomu soil, the K-status The significant response to an annual application of 40 was observed to be lower than that observed under bush kg K/ha was rather unexpected since the soil K-status fallow even with an annual application of 80 kg K/ha. with this treatment at the start of the trial in 1980 was The effect of maize crop residue retention and removal more than adequate at 0.32 me K/100g soil (Fig. 21). This with and without fertilizer application in this long-term may result from an early drought affecting the crop, trial, which initially did not show any particular trend on which may have less effect on the crop receiving the the maize grain yield in the last few years began to result higher K application. in definite effects (Fig. 22). On the Egbeda soil, removal On the sandy textured Apomu soil, where significant K of the maize residue, particularly without fertilizer appliresponse was observed already in the fourth cropping cation, significantly reduced maize grain yield. On the year, the main yield response was observed to be consisApomu soil, removal of maize residue significantly retently higher at 40 kg K/ha than 80 kg K/ha. Though no duced maize grain yield, particularly with fertilizer applidefinite explanation can be given for this observation, it cation. may in part be due to the higher acidity build up with the The detrimental effect of continuous removal of maize higher K rate, which could have an indirect negative ef-crop residue with continuous fertilizing and cropping fect on the maize crop. was observed to be more pronounced with the cowpea crop. In 1980, the cowpea variety used in the experiment was changed from VITA-4 to VITA-5. VITA-5 gave a lower yield (Fig. 23) and is apparently more sensitive to Mn 0. Toxicity. With continuous application of NPK fertilizers, There was a significant soil pH depression with both the 2! o Egbeda and Apomu soils. The lower pH resulted in S .... 0 higher mobilization and extractable Mn levels, particu0 -larly on the Apomu soil, resulting in lower cowpea yields P _mainly due to Mn toxicity combined with K deficiency. 0-2-. 0 0V CD 03 2? Plant residue management 2t Despite the importance of maintaining adequate soil orE --. ganic matter under the traditional techniques of shifting S .cultivation or bush-fallowing, the plant residue during o01: seedbed preparation is commonly burnt off. The longterm effect of this practice compared to plant residue management has not been adequately studied. A trial z was, therefore, initiated on an Alfisol at Ikenne, Nigeria, I EGBEDA SOL 0 to study the effects of burning and mulching of plant o residue in relation to fertilizer application on the perfor, ,mance of maize. Results of the trials during 1978-80 are 0-. shown in Table 34. Though some responses to N and P S ...... were observed following land clearing from fallow, sig-nificant responses to P and N were only observed, re04 spectively, in the second and third years following Clearing. The main response was to P application. 0-3 Though the mean maize grain yields do not show any 0 significant differences between burning and mulching of 2!! the plant residue during the 3 cropping years, some ef02 fects were observed among the various fertilizer treatlments. Without fertilizer application, PK, NP and NPK treatments combined with mulching gave a higher maize 0.1 .yield, particularly in the third cropping year. However, O 40 80 BARE BUSH with the NK treatment, burning gave a higher maize yield, K-RATE, KG/HA SOIL FALLOW which may be attributed to a quick release of P from burning of the plant residue. Fig. 21. Changes in Soil potassium status of Egbeda Nitrogen-tillage interaction on maize (Oxic Paleustalf) and Apomu (Psammentic Usthorthent) soils in long-term fertility trials at Ibadan, Nigeria, with Despite the importance of no-tillage systems in soil conK application (i = 0; K, = 40 and K, = 80 kg K/ha) and servation, little information is available on the best nuwithout cultivation. trient management aspects of the no-tillage system of 24



PAGE 1

80 ItoII0 60 I-/ /, too. 0 L. 06.\Y z9i I-./.,,,X X A z --,Z MAIZE CASSAVA 650 80 goo4o. =3\ CASSAVA -.. 8o-: \ A N -A' .;MAIZE 2 2 w 70. ., I\o z 60 0 'O 20 30 40 5o 60 7o so 9o oo PLANT POPULATION 2n100O's) 50, Fig. 35. Light transmission through mixed cassaval maize crop canopy at full maize development as a function of plant populations. 6 12 18 24 30 32 Induced micro-climate and crop response. In a related 10/0 12/7 2/9 14/10 25/11 0/I experiment in 1980, using cassava of relatively prolific WEEKS AFTER PLANTING growth, plant populations were maintained constant in Fig. 36. Relative soil moisture content in different crop 3 treatments at 12,500 plants/ha (cassava: 3001) and cix ture ote in dieferiod of Q 25,000 plants/ha (maize: TZPB). In a fourth treatment, the crop mixtures as a function of time during the period of maize population was increased to 40,000 plants/ha, and growth the cassava population was reduced to 10,000 plants/ha. 45. Pure cassava with 12,500 plants/ha constituted a sixth treatment, and pure maize at 25,000 and 50,000 plants/ MAIZE/CASSAVA ha were treatments 5 and 7, respectively. 40. A... SOLE MAIZE Analyses of the partial results show that the amount of SOLE CASSAVA light intercepted by the upper canopy of the maize (to G35 cob level) is significantly related to yield. "" Observations on soil moisture and temperature in the o / experiment also show marked differences in both varn,. / .... ables under the different crops and crop combinations. Fig. 36 compares the 6 weekly average mean relative soil o25 moisture contents under the pure maize, pure cassava and maize/cassava combinations, respectively. Evidently, the presence of cassava in the mixture has a benQ 20 eficial effect on the moisture available to the maize while maize helps reduce the soil temperature as a result of J rapid canopy development (Fig. 37). I 15 Within the mixed crops, appreciable differences were 0 observed with changes in planting geometry (Table 43). o i2 1'6 2o 28 32 36 Similarly, air temperature and humidity as reflected by WEEKS AFTER PLANTING the wet bulb thermometer were also affected by the planting pattern (IITA Annual Report, 1979). Fig. 37. Soil temperature in different crop/crop mixtures as a crop function of time during the period of growth. Table 43. Comparative mean relative mixture content*. Partial mechanization of maize/cassava intercropping. Yields of maize planted mechanically on ridges were not Treatment 1 2 4 affected by cassava planted on the same ridges at 0, 7, Period* 1 2 3 1 2 3 1 -2 3 14 and 21 days after maize (IITA Annual Report, 1979). Moisture content 72 67 102 94 85 123 83 70 112 Herbicide, however, depressed maize yield significantly *The moisture content is in % of value at FMC. at the 5 percent probability level. The effects on cassava *Period 1 = 2-6 WAP; Period 2 = 7-12 WAP; Period 3 show that delay of up to 21 days in planting cassava was = 13-18 WAP. not critical to cassava yield (Table 44). However, signifi34



PAGE 1

100 Y 8.846 + 0.397X 38 (r=889*) 90 x Y 1735 X0851 34 80 so.. TO 30 60 X S26 50 hi > 22 140 KU) 30 18. 20 14 10 10 20 O30 40 50 60 70 o80 90 PLANT POPULATION (inlOOO's) ,I0 20 30 40 50 O 80 Fig. 32A. Light intercepted by mixed maize/cassava crop PERCENT LIGHT TRANSMITTED (%) canopy (to cob level) at full maize development (10 wap) as a function of plant population. Fig. 33. Cassava yield in a maize/cassava mixed crop as affected by light transmission through the mixed crop canopy at maximum maize height. Y 5 694+0'3617X 100 38 (r -0888 *) Yg .5501X0 622 x 90 34 so 30 30. To. 70 26 a a 26 60 3 22 -"I 6 V) 50 U x 1840 1441 3010 20 0to 20 30 40 50 60 70 80 90 PERCENT LIGHT TRANSMITTED(%) 10 0 FIg. 34. Cassava yield in a maize/cassava mixed crop as S0o 2b SO b do o 7 s o So affected by light transmission through the upper maize PLANT POPULATION (n OOO's) canopy at maximum maize height. Fig. 32B. Light intercepted by mixed maize/cassava crop canopy (to ground level) a full maize development (10 wap) as a function of plant population. 33



PAGE 1

ridged treatment. Table 16 shows that 20 percent more volumetric heat capacity, generally increases with an inyam tubers were produced in mulched than unmulched crease in soil moisture content up to a certain value. The plots with least tuber yield measured in the unmulched exact relationship depends on soil texture. Thermal difridge treatment. The beneficial effects of residue mulch fusivity measurements were made for 20 different soils, may partly be attributed to favorable soil temperature and the relationship between soil moisture content and and moisture regimes. Observations on soil moisture thermal diffusivity for some soils is shown in Fig. 12. In content made 40 and 90 days after planting indicated general, thermal diffusivity is higher in coarse-textured that unmulched ridges had the least soil moisture resandy soils than heavy-textured clayey soils. The regresserves. sion equations between sand, silt and clay on thermal diffusivity are given in Table 18. Table 16. Effects of tillage methods and mulches on yam production (Onne, 1980). Ridge Flat Ridge plus Flat plus no mulch no mulch mulch mulch Length of 3 2 tuber 20.7 20.7 23.5 21.7 6 cm Diameter of tuber 13.3" 13.9a 16.7b 17.1b Cly o.C cm I so" Clay Loam 61 27 12 Yed2 Clay Loom 47 33 :.56 Yield 14.1 16.3 19 18.9 2c3. Loom 33 22 47 tlha o 2 ,4 4 Lom 27 23 1.61 5 Lo, 78 Is 039 ow 67 020 030 ..0. o40 045 Effects of soil bulk density on root and shoot growth of WATER CONTENT, C._.m3 cassava. A root-box study was conducted to investigate the effects of 3 soil bulk density treatments, 1.4, 1.6 and Fig. 12. Thermal differsivity of various soils as a function 1.8 gcm-3, on cassava root and shoot growth. Cassava of the water content. Size of glass beads used 50-75 U. was grown in collapsable boxes 48 cm x 46 cm x 100 cm, with sides that could be removed for monitoring root growth at different stages. Measurements on root growth were made for cassava cultivar TMS30572 at 48, 78, 108, 132 and 185 days after planting. Root density, measured. Table 18. Regression equations relating thermal diffuas length and dry weight, was not significantly affected sivity, D (cm2 sec-') to percent sand, silt, clay by the soil bulk density treatments investigated. Neither or organic matter in a soil sample. were there significant differences in plant height, leaf area nor shoot dry weight. Cassava can withstand soil Simple compaction more than grain crops such as maize, cowpea or soybean. Nevertheless, the shoot: root ratio was Multiple generally higher at soil bulk density of 1.6 gcm 3 than 2. D = 0.01182 -01099 X 10-3 r =0.844** other densities (Table 17). The optimum density for the cl -0.1680 x 102 cm high feeding: tuberous root ratio was 1.6 gcm-3.These 3. D = 0.0121 -0.3066 x 10-4 r = 0.860** results imply that effects of soil compaction on cassava Si -0.1104 x 10-3 ci -0.1479 x 10-2 may be highly dependent on soil texture as the latter Om affects both the intensity and capacity factors for nu4. D = 0.001025 + 0.1104 x 10-3 r = 0.60** trient and water availability. Sa + 0.798 x 10 4Si -0.1479 x 10-2 Om Effects of tillage methods and mulches on soil temper-. ature regime and its effects on crop production. TherSa = Sand, Si = Silt, Cl = Clay, and Om = Organic mal diffusivity, the ratio of thermal conductivity and matter. Table 17. Effects of soil bulk density on cassava's shoot: root ratio and feeding root: tuberous root ratio. Bulk density Bulk density Bulk density 1.4 1.6 1.8 Root dry wt Root dry wt Root dry wt Days after Shoot dry wt Tuber dry wt Shoot dry wt Tuber dry wt Shoot dry wt Tuber dry wt planting Root drywt x 10 1 Root drywt x 10 4 Root drywt x 10 1 48 134.72 58.89 -50.00 77 188.24 84.53 258.89 305.08 126.04 185.90 106 140.13 67.47 128.55 70.40 110.4 98.81 134 150.17 42.44 267.41 41.90 348.91 12.60 185 401.96 299.79 249.39 537.66 200.01 329.94 14



PAGE 1

200 E ONNE .180 MI IITA 0 160 zin fluometuron 30Okg/ho 0 (a) 0 (c 140z LL 0 0 Ui S120... L 0 400 I-L 280 0 2: In0 0 wm 0 20 m 0 20 -jj 0 0i 14 2 z 20 1 2 4 8 1 WEK FTRLRAMN Fi20 89 essec fhriie nsraesis -45



PAGE 1

Table 44. Effect of relay times and herbicides on cascropping and highlight the potential hazard of herbicide sava yield in maizelcassava intercropping. failure. Cassava relay Herbicide No Herbicide Mean N response in maize/cassava intercropping. The N rein days M/C' C2 M/C C M/C C sponse of inlercropped maize/cassava on an Alagba soil 0 16.8 7.6 22.6 25.7 19.7 16.6 (Oxic paleustalf) at Ikenne that was initiated during the 7 22.4 12.2 22.9 27.0 22.7 19.6 1978/1979 season was repeated during the 1979/1980 14 16.7 13.9 23.5 22.3 20.1 18.1 season. During the second year, the maize crop was af21 13.0 9.1 23.1 20.6 18.0 14.8 fected by drought resulting in low yield (Table 45). DeMean 13.9 23.5 spite the low grain yields, the sole and intercropped LSD between herbicide 2.7 maize showed significant responses to N application. As Relay time 5.2 observed during the first year of cropping, intercropping Relay time same herbicide 7.3 with cassava had no effect on maize grain yield. In casRelay time different herbicide 7.3 sava, high tuber yields were observed with later harvesting (13 months after planting), but no significant Maizelcassava response to N application was observed. This indicates 2 Sole cassava that the N requirement is lower for the cassava than the associated maize. High N rates (120 kg N/ha)also tend to decrease the tuber yield of cassava. cant cassava yield reduction was observed where herbicide was used. This yield reduction was ascribed not Effect of maize population on cassava yield. An experionly to the sensitivity to the herbicide but also to weed meant was conducted to determine effects of maize competition resulting from failure of the herbicide. This population on cassava development and yield in observation is substantiated by a tendency toward lower intercropping where maize is planted in hills or clusters. yields by herbicide treated pure cassava. Shading by It simulated farm situations where many seeds are maize appeared to have retarded weed growth, thus, placed in 1 position, and the number of surviving plants reducing the competitive effect on cassava. The results depends on chance or are deliberately selected by the confirm the compatibility of maize/cassava in interfarmers. The maize was spaced at 100m x 100 m with Table 45. Effect of N application on yield of intercropped maize (variety TMS 30395) grown on Alagba soil (Oxic paleustalf) (1979-80). Cassava fresh N-Rate tuber yield Maize grain kg N/ha Cropping mixture tlha yield LER* Maize (1 x .33 m, 0 1 plant/hill) 1.80 Maize (1 x 1 m, 3 plants/hill) -1.74 Cassava(1 x 1 m) 30.12 Maize (1 x .33 m) + cassava (1 x 1 m) 29.11 1.93 1.90 Maize (1 x 1 m) + cassava (1 x 1 m) 28.89 1.87 2.00 Maize (1 x .33 m, 60 1 plant/hill) -2.30 Maize (1 x 1 m, 3 plants/hill) -2.40 Cassava (1 x 1 m) 30.90 Maize (1 x .33 m) + cassava (1 xl m) 27.10 2.45 1.95 Maize (1 x 1 m) + cassava (1 x 1 m) 29.50 2.39 1.95 Maize (1 x .33 m, 120 1 plant/hill) -2.28 Maize (1 x 1 m, c plants/hill) -2.12 Cassava (1 x 1 m) 28.65 Maize (1 x .33 m) + cassava (1 x 1 m) 24.85 2.52 2.05 Maize (1 x 1 m) + cassava LSD (1 x 1 m) 29.68 2.10 2.03 3.40 0.40 Maize crop slightly affected by drought. Cassava harvested at 13 months. Land equivalent ratio. 35



PAGE 1

Table 34. Effect of plant residue management and fermaize production. Most of the fertilizer trials carried out tilizer application on main season grain yield in the tropics have dealt with conventional tillage. Studof maize variety TZPB grown on Alfisol (Oxic ies were, therefore, carried out on an Alfisol (Alagba sePaleustalf), Ikenne, 1980. ries, Oxic Paleustalf) at Ikenne and on an Entisol (Apomu 1978 1979 1980 series, Psammentic Usthorthent) at Ogbomosho, Nigeria, to determine the N requirements for maize producFertilizer M* B*" M B M B tion under conventional and no tillage. At both locations, treatment kg/ha maize yields from the control plots without N application Control 4228 3706 2864 2675 2841 3055 were higher with conventional tillage than no tillage. this PK 5069 4830 4493 4386 4341 4080 was also observed during the first cropping year in 1979 NK 4939 4812 3016 3385 3243 3601 (Fig. 24). However, maize yields with N application, parNP 5205 4720 4360 4526 5378 4857 ticularly at high N rates, were higher with no tillage than NPK 5327 5275 4454 4448 5289 4936 conventional tillage. NPK Mg Zn 5838 5824 4316 4509 4968 4724 Mean 5101 4861 3852 3988 4343 4208 Planted fallow as an alternative N source LSD (5%) (1)** 835 612 567 (11) 822 536 802 Use of Leucaena prunings. In looking at low cost alter(111) 738 764 1263 natives to N sources, experiments were carried out using Leucaena leuocephala top prunings. Leucaena not only *M= Plant residue applied as mulch. can serve as a potential N source but also can supply **B = Plant residue burnt before each cropping. fuel wood. In a trial carried out on an Apomu soil series ***LSD () = Between plant residue management at IITA, the effects of rate and placement methods of the means. Leucaena top prunings were compared to urea. LeuLSD (11) = Between fertilizer treatments within resicaena top prunings were banded or broadcasted once at due management. maize planting, while urea was banded twice, one-third LSD (111) = Between any two fertilizer treatments with N at maize planting and two-thirds N at 4 weeks after different residue management. maize planting. Figure 25 shows that banding the Leucaena top prunings at 25 cm widths was most effective, and banding at rates of 5t or 10t was, respectively, 52 and 67 percent as effective as urea. Though the prunings can be used as an N source, application at planting time only was apparently not too effective. LSD (.05) LSD (.05) 5.0. 0 NT 'o NT < 4.0 CT NT YCT >"3.0 z IKENNE OGBOMOSHO J OXIC PALEUSTALF PSAMMENTIC XJ O N 1980 USTORTHENT 1980 S2-0 b 30 60 90 .120 150 0 30 60 90 120 150 N-RATE, KG N/HA Fig. 24. Effect of tillage and N rates on maize grain yield. 26



PAGE 1

Foreword T he year 1980 was one of financial uncerFarming systems research was reorganized tainty for the International Institute of to focus on three major components-land Tropical Agriculture. A high rate of inflaclearance and development, post-clearance tion, a doubling of the minimum wage, and a management, and cropping systems. The strengthening of the Nigerian naira against maize improvement program included some the American dollar had a very serious and exploratory work on the development of hynegative impact on the real research capacbrids. The development of lowland rice vari -ity of IITA.i. eties and the problems of production in Research program budgets and staff had to central and eastern Africa were given greater be reduced, and a general belt tightening left emphasis. Breeding for resistance to insect theremaining scientists with insufficient suppests of cowpeas was intensified, and casporting staff and operating allocations to ensava research was concentrated on, host ab~them to work in a fully effective and plant resistance to the green spider mite and productive manner. mealybug. It was through a very generous special conIn the resolution of the etiology of virus distribution of 1 million naira, equivalent to 1.9 eases, techniques of indexing the presence million U.S.dollars, by the Federal Republic of sweet potato virus was. developed.. As a of Nigeria in December 1980 that this precarresult, it is now possible to export virus-free ious financial situation could be redressed. sweet potato clones in tissue, culture form Despite.. the disruptive economic condiinternationally. tions, IITA's research community produced a Previously, the Institute's research was alscientific product of commendable quality most exclusively pointed toward the small, and quantity as reflected in this Annual traditional, resource-poor African farmer. But Report. because of increasing interest in commercial As in previous years, the 1980 Annual Refarming in many African countries, the Instiport is a compendium of salient, research tute is devoting a greater share of its reconducted in cereal, root, tuber, and grain search effort toward the solution of problems legume improvement programs, and in farmencountered by medium and large-scale ing systems for the humid and subhumid farming enterprises. tropics. This report and its sister publication, We are sure you will find this Annual Report 1980 Research Highlights, together provide a informative, and we will be pleased to recomprehensive summary of findings for both spond to all requests for additional informathe scientist and the generalist., tion about on-going research at the Institute. This year also marks the first time the AnA listing of journal articles and IITA's scinual Report is being published in French. entific staff is appended to this report to Long overdue, we are confident it will assist you. be enthusiastically received by our Frenchspeaking colleagues. The structure of the Institute's research program remained unchanged in 1980, but the focus of the work was somewhat Dr. E. H. Hartmans modified. Director General VI



PAGE 1

85 Cereal Improvement Program 85 Introduction 85 Maize 86 Genetic improvement 95 Entomology 96 Semi-Arid Food Grains Research and Development Project (SAFGRAD) 100 Rice 100 Genetic improvement 109 Pathology 109 Physiology 117 Grain Legume Improvement Program 117 Introduction 117 Cowpea 117 Genetic improvement 124 Entomology -130 Upper Volta Food Legumes Program 131 Semi-Arid Food Grains Research and Development Project (SAFGRAD) 134 Tanzania Food Crops Research 137 Brazil Cowpea Program 137 Soybean 137 Genetic improvement 143 Entomology 149 Agronomy 155 Tanzania Food Crops Research 155 Microbiology 161 Training Program 163 Research Support 163 Introduction 163 High-rainfall substation 163 Research station operations 163 Screenhouses and the plant growth facilities 163 Genetic Resources Unit 166 Virology Unit 171 Analytical Services Laboratory 171 Biometrical Unit 171 Conference Center 171 Library and Documentation Center 172 Communications and Information Office 173 Special Projects 173 Introduction 174 Training and collaborative research projects 174 Core supplementing projects 175 Personnel 177 Collaborators and Training 180 Publications 182 Conference and Seminar Papers 185 Major Seminars IV



PAGE 1

LSD (5I) LSD (5%) III III N2 P2 K2 R N 2 P2 I, R N2 P2 KIN2 P2 K No P6 Ko + R No Po Ko-R P0 SEGBEDA SOIL APOMU SOIL No Po Ko -R I p p I I I _7 2 73 74 75 76 77 78 79 80 2 73 74 75 76 77 78 79 80 YEAR YEAR 1" 22 Ftfect of continuous removal of maize crop residue and fertilizer application on maize grain yield in longoii s pial on Egbeda (Oxic Paleustalf) and Apomu (Psammentic Usthorthent) soils. LSD (5%) LSD (50/%) 600 10 -2 22 K % -R 20 N2 P2 K R 4N2 P2KIR :*30 N2 P2K4+ R. No Po Ko R SNoPoKo R No Po Ko +N R 600 "\No Po KoR '400200 EGBEDA SOIL POMU SOIL 01 76 77 78 79 80 76 77 78 79 80 YEAR YEAR Fq 23. Effect of continuous removal of maize crop residue and fertilizer application on grain yield of cowpeas 9 ongqterm trial on Egbeda (Oxic Paleustalf) and Apomu (Psammentic Usthorthent) soils. 25



PAGE 1

Table 7. Small holder land use in eastern Nigeria: Land allocation to enterprises per farmer (1979-80). Derived Savanna Lowland Forest Coastal Lowland Row Total Enterprise Ezzambgo Ikom Umudike Akamkpa Onne Uyo Av. % Average No. of plots 4.5 3.7 5.20 4.95 4.45 5.50 4.91 Area under Tree Crops (ha) 1.34 7.10 15.05 13.79 6.30 11.87 9.24 67.0 Area under Arable Crop (ha), 3.00 4.43 2.30 2.47 5.00 3.78 3.50 25.0 Area under Tree/Arable Crop (ha)' 0.12 1.01 0.56 0.81 3.03 0.51 1.01 7.0 Av. total Cult. land (ha)l 4.46 12.54 17.81 17.07 14.33 16.16 13.75 I Fallow periods are often short and contain some cassava resulting in some short fallows being considered as cultivated plots by farmers. This has tended to put the total cultivated land slightly higher along with the problem of double counting of same plots due to intercropping and relay cropping. Small ruminant (goats and sheep) and poultry proProposed agroecological regions of the survey area: a duction is common with a general trend of more small synthesis. A map at a scale of 1:3,000,000 delineating ruminant production in areas of higher demographic agroecological regions has been prepared using the data densities where goats are more common and important of this survey (Fig. 10). Climate, geomorphology and in both humid and subhumid zones, while sheep and soils, agricultural land use (farming systems), demopoultry are more important in the subhumid (derived sagraphic conditions (density and settlement patterns) and vanna) zone. The number of domestic animals kept by the interaction of these have been considered in identieach family is small. fying these regions. Resource allocation to enterprises and production conAccordingly, 5 major agroecological regions were identraints. Family labor and arable land are the 2 principal tified: resources. Farmers are dependent on family labor and only use hired labor during labor peak periods. A cropI. Basement complex (cocoa/forest) with 4 subping calendar of the 13 most important tree and 18 araregions. ble crops showed that upland tree and arable crop I1. Sandstone and coastal sand complex (oil palm/ farming, as parallel systems of production, compete for root crop) with 5 subregions. farm labor during land clearing, preparing and weeding. Ill. Niger delta and coastal swamp complex with 2 It was observed that there is a better utilization of family subregions. labor by having both tree and arable food crops than by IV. Upland moist savanna complex with 4 subreeither alone, and the 2 enterprises fulfill both cash and gions. food needs of the family. The competition for labor beV. Mangroves and coastal sand complex. tween the tree and arable food crop enterprises could be reduced, if not eliminated, if both are integrated on the Region V was not surveyed, for it is considered a nonsame land and, when possible, simultaneously operated. agricultural zone. Similarly, region Ill was not included in However, the practice of burning the bush in land prepthe survey because it is nonzonal in that land use is govaration induces the physical or spatial separation of the erned by the Niger River as the environmental factor. 2 enterprises and causes arable crops to be grown in Only the lower part of Region IV was included in the distant farms away from the homesteads (Table 7). survey, the derived savanna part, and this region, often referred as the middle belt is agriculturally (crop and State of tree and arable crop farming. In general, there livestock) important and is currently experiencing inappears to be less expansion of tree crops on new lands. creased agricultural developments with a corresponding Any increase, especially in cocoa, kola and citrus, is due increase in population. to rehabilitation and an increase in density of tree crops. Most farmers in the drier savanna and coastal lowlands, Region I and II are the most important for rain-fed tradiparticularly under high-population conditions, thought tional farming. The major distinction between these is tree crop farming to be declining. Major constraints to their difference in geology and soils. In general, soils of tree crop replanting and expansion were credit and lack Region I are geologically better soils (mostly Alfisols) of production inputs. and can support both tree and arable crop farming. The r zone is currently used for cocoa, kola, citrus, etc., and a Farmers' view toward farming. Many traditional farmers great variety of food crops is grown. On the other hand, themselves view farming as a nonprofitable enterprise Region II has soils that are chemically poor but with and will abandon it if given other options. Eighty-six pergood physical properties (mostly Ultisols) that make tillcent of the farmers stated that they will continue farming. age easy. Both soils are erodable and easily degradable Of these, 68 percent stated that they will continue bewith intensive cropping, particularly soils of Region II cause they have no other option available to them, 12 where the situation is already serious. percent stated that they have great family responsibility that requires them to continue farming and 6 percent The agroforestry potentials (alley or strip cropping, stated that they do not wish to buy foods from the marplanted fallows or taungya, multi-storey farming and ket. Only 14 percent stated that they will continue farmagro-silvo-pastoral) are higher for Regions I and IV ing because it is profitable. Those wishing to discontinue where such new or improved systems of land use may farming, especially arable crops farming, often gave old not be difficult to adopt. But, the same cannot be said age as their main reason. for Region I1. 9



PAGE 1

cation on long growth-duration rice variety IM 16 showed > maize seeded at 20 days after the seeding of groundthat time of N application greatly influenced yield (Table nut. A similar trend was also observed with groundnut 55). While basal application of 50 kg N/ha at 1 day before yield, and obviously both crops reacted negatively to late transplanting did not increase yield, application of 50 kg planting and intercrop competition. N!ha at panicle initiation did. It appears that the best recommendation may be 50 kg N/ha basal applied and sidedressed with 50 kg N/ha at panicle initiation. Small farms systems research Table 55. Grain yield of irrigated rice variety IM 16 (Atebubu District, Ghana) under different treatments of rate and time The small farms systems research is a component of the of N fertilizer application (Doume, 1980). USAID sponsored project of Managed Inputs and DelivFertilizer rate Grain yield ery of Agricultural Services (MIDAS). The MIDAS project (kg N/ha) Time of application, (t/ha) is designed to organize and distribute all the inputs nec0 -4.6 essary for food crop production by small farmers of 0 4.6Ghana. 50 50 (B) 4.6 100 100 (B) 5.3 The small farms system research is designed to assist 50 50 (PI) 5.9 the government in establishing an applied multidisciplin100 50 (B) + 50 (PI) 6.6 ary small farms research capacity with the following ob100 40 (B) + 30 (30 DT) 5.5 jectives in mind: + 30 (PI) (1) To obtain a sound knowledge of the existing farm,B = Basal, 1 day before transplanting. ing systems, the socioeconomic environment and PI = Panicle initiation. the positive and negative factors in services. DT = Days after transplanting. (2) To conduct applied research that is relevant to these circumstances and responds to the needs of the small farmers. Maize/groundnut intercropping has been extensively uti(3) To identify soil and farm management and produclized by local farmers. A field trial was conducted during tion practices that eliminate constraints. the second season to evaluate the potential of this sys(4) To increase small farmer production and income. tem and investigate the factors affecting yields of crops when they are planted in association. Among the interThe project was initiated in 1980 with an appraisal of the cropping treatments, the treatment of maize at 40,000 present food crop farming system practices in the rehills/ha seeded at 20 days after the seeding of groundnut gion. For the survey, a group discussion approach was at 133,000 hills/ha gave the highest monetary return used that was supplemented by the visit to individual (Table 56). farmers and spot harvest checks. The survey covers 7 In general, however, the yield of both crops is reduced subdistricts-Atebubu, Amantiri, Abease, Prang, Kwame when the crops are planted in association. The amount Danso, Kajaji and Yeyi. of yield reduction varies with the seeding time of the The survey was completed in 1980 and a report titled, affected crop in relation to the seeding time of the asso"An appraisal of the present farming systems of the Ateciated crop. The yield of maize planted in association bubu District of Ghana," was published. During 1980, with groundnut, for example, was reduced in the followpreparations were also made in selection of a site for ing order: maize seeded at 20 days before the seeding of establishing the experiment station. Actual field experigroundnut > maize seeded at the seeding of groundnut mentation, however, will be initiated in 1981. Table 56. Yield of sole and intercropped maize and groundnut and their monetary values. (Bertoua, 1980 second season). Crop and date of seeding Treat(M = Maize; Plant density "TreatCrop Yield Kg/ha Monetary value (CFAF)' ment G = Groundnut) (x 101 hills/ha) ment Maize Groundnut Maize Groundnut TOTAL C, M on 1 August 50 C, 3066 -183,960 -183,960 C2 G on 1 August 200 C2 -2277 -284,625 284,625 M on 1 August, and 40 C, G on 1 August 133 C, 2263 1206 135,780 150,750 286,530 M on 1 August, and 40 C, G on 20 August 133 C 2959 539 177,540 67,375 244,915 M on 20 August, and 40 C., G on 1 August 133 C 831 2122 49,860 265,250 314,790 M on 10 Sept., and 40 C, G on 1 August 133 C6 -2245 -280,625 280,625 At 60 CFAF/Kg of maize grain and 125 CFAF/Kg of unshelled groundnut (Retain price of PROVIV-ZAPI, 1980). "Maize seeded on this date did not germinate. Fertilizer rates maize 8N -80 P20; groundnut 2N -6 P20 in Kg/ha. 49



PAGE 1

Table 41. Percentage of applied cations and anions in mixed cropping systems. As a follow up on previous the leachate after 2,012 mm of water under studies showing or establishing a range of modifications bare fallow (uncropped, weed free) IITA, that may be brought about in the light regime in maize 1980). canopies through modifications of planting geometry Tension drained and density (IITA Annual Reports 1978 and 1979), studies Free drained (0.2 Atm) were carried out in 1980 with the aim of quantifying and lons Lysimeter 1 lysimeter 6 standardizing some of these relationships partly to provide a basis for optimal design in mixed cropping systems. Attempts were also made to relate the effects NO,78 70 of the modified light climates within the established C" 35 30 crop (maize) on the lower or slower growing intercrop Mg -64 60 (cassava). K, 9 9 Results show that at full development of the maize canopy in a maize/cassava mixture, the percentage of inciat Onne in 1978. Results of the second year cropping dent radiation depleted at cob height is a power function during the 1979-1980 season are shown in Table 42. In of the combined plant population. The relevant equation the second year of cropping, the soil K-status even with is as follows: K application was very low. There was a distinct effect of Y., = 1.735Xos8t r = .917"* Mg application in increasing the soil Mg status. The tuber yields of both varieties showed a more pronounced where Y., is the amount of light intercepted (to cob level), and significant response to K application. Despite the and X is the combined plant population (Fig. 32A). A presence of Mg deficiency symptoms without Mg applisimilar relationship was obtained at ground level (Fig. cations, similar to the first year results, there was no 32B). definite response of tuber yield to Mg application. As previously reported (IITA Annual Report 1979), plant populations in this experiment were varied from 10,000 Cropping systems plants/ha in the pure cassava to 80,000 plants/ha in the maize/cassava mixtures. The maize (TZPB) population In 1980, research in cropping systems or crop managevaried from 10,000 to 70,000 plants/ha by increasing the ment focused on the following areas: mixed cropping, number of maize plants per stand from 1 to 7. Spacing alley cropping, live mulch system and the role of agrowas maintained at 1 m x 1 m with the plants along the forestry in food crop systems. The ultimate goals are to same row. achieve high and stable crop yields while maintaining For the lower or slower growing cassava, which was long-range soil productivity. Several projects were initishaded through much of the growth of the maize, the ated to investigate the inclusion of leguminous cover crop yields were very significantly affected by the crops and managed tree and shrubs into food crop proamount of light reaching it through the maize canopy duction systems in an attempt to find more efficient, low(Fig. 33). Again, a similarly linear relationship was obenergy input and stable alternative systems to traditional tained between the cassava yield and the total light bush fallow cultivation. Special emphasis was given to transmitted through the combined maize/cassava canthe production system of plantain in the humid and peropy at full maize development (Fig. 34). These amounts humid regions. of transmitted light can also be systematically related to the combined plant population as shown in Fig. 35. The Intercropping agronomy and relationship between the 2 factors is expressed: meso/micro-climatic studies Y = 82.198e 001sx r = -0.903* Light regime and productivity in mixed crops. Competiwhere Y. is the amount of light transmitted through the tion for light and moisture are clearly 2 key factors in canopy, and X is the combined plant population. Table 42. Effect of K and Mg application on tuber yield of cassava cultivars TMS 30395 and TMS 30211 and K and Mg status of an Ultisol (Typic Paleudult), Onne, 1980. Fertilizer Treatment Soil K and Mg status TMS 30395 TMS30211 kg/ha Me/i OOg Tuber yield t/ha K Mg K Mg Fresh Dry Fresh Dry 0 40 0.09 0.33 15.84 6.47 14.99 5.66 30 40 0.09 0.27 19.41 8.08 18.67 6.11 60 40 0.11 0.33 19.97 8.84 18.20 6.36 120 40 0.10 0.30 20.13 8.52 19.08 6.73 120 0 0.11 0.22 21.47 9.39 17.33 6.69 120 20 0.11 0.22 19.47 .8.06 17.50 5.97 Mean 19.38 8.23 17.63 6.25 LSD (5%) Variety means fresh tuber 0.63; Dry tuber 0.78. Between fertilizer treatments within variety: fresh tuber 3.48; dry tuber 1.83. Between fertilizer treatments among variety: fresh tuber 5.29; dry tuber 2.88. 32



PAGE 1

Fig. 10. Agro-ecological regions in the humid and sub-humid zones of Nigeria. 1 Basement Complex Regions (Cocoa/Forest) llCb Cross River Basin-sedimentary sands-/ow IA Northern sub-humid zone (W. Nigeria)-pop. pop. high liD (Coastal) Sedimentary Sandy Plains-(E. NiI~a Central Humid, Sub-humid Uplands-(W. Nigeria)-high pop. geria)-pop. high. IBb Eastern Humid Uplands (E. Nigeria) baseIIl. Niger Delta and Coastal Swamps ment complexlvolcanic-low pop. IliA Middle-Upper Delta-medium to high pop. IC Southern Humid Lowlands (W. Nigeria)IllB Lower Delta and Coastal Swamps-low pop. basement complex/coluvial, alluvial deposits .high pop. IV. Upland Moist Savanna IVAa Western Moist Savanna-mostly Basement II. Sandstone and Coastal Sand (Qilpalm/Root Crops! Complex, medium-high pop. Forest Complex) IVAb Eastern Moist Savanna (sandstone hills)IIA S.W. Sub-humid lowlands-sandy lowlands high pop. -medium pop. IVAc Eastern Moist Savanna-medium pop. liB Lagos and Adjacent Lowlands-sedimentary IVB. North Eastern (Guinea) Savanna-Sandsands-high pop. stone medium to low pop. IICa Benin Lowlands-Sedimentary sands-medium to low pop. V. Man groves/Coastal Sands" Landandsoi ma age ent Effect on soil bulk density. Deforestation resulted in a ~Lan an sol ma age ent significant increase in soil bulk density (Table 8), and there were slight differences in bulk density among varResearch on land and soil management includes the folious methods of land clearing. Maize, being an open-row lowing areas: land clearing and development, tillage syscrop grown immediately after deforestation, increased terns and small tools development and management of soil bulk density (data of 1979), while cassava, being a kaolinitic Alfisols and siliceous Ultisols. close canopy crop, decreased soil bulk density. Moreover, tuber development just beneath the soil surface may have contributed to decreasing the bulk density of Land clearing and development the layer above the tuber and increasing the bulk density Hydrological investigations and characterization of soil oftelyrbowheues.Dfrncsisiluk physcalproprtis asinfuened b mehodsof and density were'also reflected in the infiltration rate and clearing and post-clearing soil management were conpntoee eitne tinued in 1980 for cassava planted in the 1979 second Effect on total water yield. Water yield from the cleared season and harvested toward the end of 1980. watershed treatment was 259 mm while the forested 10



PAGE 1

both cowpea cultivars is as high as 60 percent when Residual effect of lime chemical fertilizer is not applied in the second season (Fig. 28). But, with moderate rate of fertilization, the critBecause lime is scarce in many parts of the humid tropical level of Al saturation is reduced considerably, and ics such as southeastern Nigeria, it is necessary to know the 2 cowpea cultivars show differential response to soil what minimum rates of lime are needed for optimum acidity. i.e., 30 percent for VITA-4 and 43 percent for crop production and what residual values may be exVITA-1 (Fig. 29). pected. Results from an Onne trial are summarized as follows: Relatively low rates of lime would be adequate to sustain 100 -crop yields under a maize/cowpea rotation system. It is interesting to point out that with balanced fertilization, 90" respectable levels of maize and cowpea yields were Vita-I maintained for 5 years in the unlimed plots (pH 4.3) 80. where no severe A toxicity effects were observed. Both 80 toict cowpea cultivars nodulated well with indigeneous acid70" Fertilized tolerant rhizobia, and nodulation was only slightly re1979, 2nd Season duced without liming. Vita-I 1.28Btlha Applied Ca in the form of lime leached readily from the 60Vita-4 1-50 tlha Vta -surface layer, which is accompanied by the subsequent reappearance of exchangeable Al (Table 39). The in50 crease in CEC values due to liming is also short-lived. The downward movement of Ca has little effect on the 100subsoil pH and exchangeable Al, suggesting that Ca Bleached in the form of neutral salts. The vertical distribution of exchangeable Ca in the field profile measured 903 years after lime application follows the theory and.formulation of ion-exchange chromatography (Fig. 30). This 60No Fertilizer is in agreement with a previous laboratory leaching study 1980 2nd Season using undisturbed soil columns. Over 90 percent of the 70Vita -t 1-48 tlha applied Ca may be found between 0-90 cm depths 3 Vita-4 1.36 1lha years after liming because of the strong subsoil acidity (pH 4.3 and greater than 50 percent Al saturation). Re60 covery of the subsoil Ca, however, would require deep10 20 i0 40 50 60 7b 90 rooting species tolerant to high exchangeable Al levels. EXCH. AL SATURATION,% Moreover, because a high lime rate in the surface soil has little effect on the subsoil acidity, there would be no Fig. 29. Relative yield of cowpea as affected by percent advantage of applying a rate of lime more than required exchangeable Al saturation in the soil. to reach the exchangeable Al levels for the crop to be Thus, liming is not necessary unless the exchangeable Table 38. Concentration of Al in soil solution of a Ultisol Al saturation of the soil has reached a level beyond the (Typic Paleudult) as affected by fertilization critical level of the crop to be grown. The critical level of (Onne, 1980). exchangeable Al saturation in such coarse-textured, kaAl in olinitic Ultisols depends to some extent upon the rate of Exch. At saturation fertilizer to be applied. The variability in the critical level Soil pH saturation extract of exchangeable Al saturation is because of the large (H20) % ug/ml increase in soluble Al in the soil solution after fertilization. Such effects have not been taken into account beUnfertilized 4.3 60 2.7 cause soil sampling is normally done before planting. Fertilized and incubated The effect of fertilizer salts on soil solution Al is shown 6 weeks at field in Table 38. capacity 3.9 46 17.1 Table 39. Changes in soil properties in the surface layer (0-15 cm) 5 years after lime application (IITA Onne substation, 1980). Initial lime pH (H20) Exch. Ca, meq/100g Exch. A, meq/100g ECEC, meq/100g rate (1976) t/ha. Initial # 5 Yrs. Initial 5 Yrs. Initial 5 Yrs. Initial 5 Yrs. 0 4.7 4.2 0.57 0.27 1.26 1.51 2.77 2.47 0.5 5.0 4.3 1.37 0.31 0.82 1.43 3.08 2.40 1 5.2 4.4 1.58 0.42 0.45 1.27 2.77 2.36 2 5.6 4.5 2.36 0.61 0.06 1.01 3.10 2.28 4 6.3 4.8 5.71 1.03 0.04 0.40 6.53 2.12 #Initial soil samples taken one month after lime application. 29



PAGE 1

-dli rr -71W, hi:M '113 -~ -~l........ -I ~ ~I, <, Zip ~



PAGE 1

1-7 plants hill to give 10,000-70.000 plants/ha. Cassava Alley cropping was planted at 100 m x 100 m in the same rows as maize but spaced 50 m from the maize along the same row to give a population of 10,000 plants/ha. The land was Alley cropping systems in forest zone plowed and harrowed but not ridged. The fertilizer recEffect of N application on maize/Leucaena alley cropommendation used was for maize and consisted of a ping. Results of a field experiment conducted at IlTA on basal application of N.P.K. 15:15:15 at 400 kg/ha and a the effect of N application in the maize/Leucaena leucoside dressing of 30 kg N;ha as urea 4 weeks after planting cephala alley cropping system are shown in Fig. 39. Subof maize. stantial Leucaena dry matter and N yields are produced The cassava growth was significantly affected by maize with a total of 6 prunings. The total annual N yield of over population with tuber yields being significantly lower at 180 kg N/ha was quite remarkable. the higher maize populations. The marked differences During this cropping year, there was a significant effect between maize populations suggested that maize popuof the low N rates applied to the maize on the Leucaena lations should not exceed 30,000 plants/ha if good casdry matter and N yields. There was, however, only a slight sava yields are expected. The maize yield indicates no effect from the high N rate applied. serious yield reduction of maize population if kept at 30,000 plants;ha. The results also support the practice of The effect of N rates and removal of Leucaena prunings thinning maize to 3 plants hill in maize/cassava interon maize grain yield is shown in Figure 40. The results cropping in West Africa. clearly show that despite the high amount of N yield, there was still a need for application of low N rates for Weed control in maize/cowpea intercropping. The effect obtaining high yields. Removal of the Leucaena prunings of early weed interference in maize/cowpea intercrops is reduced the yield to about 54 percent. Without N applimodulated by the growing season. During the 1980 first cation but with Leucaena prunings only, the total maize season, a maize/cowpea intercrop was more sensitive to weed interference than each of the crops grown in pure culture (Fig. 38.) However, during the 1980 second season, crop sensitivity to early weed interference peaked in both the sole and intercrop at 2 weeks after planting. This may be related to moisture conditions. Weed biomnis at the 2-week stage was higher in the second seasWhan the first season, thus, accounting for the more severe weed interference during the second season. 16 EARLY 1979 Sole Maize (40 x 103 pit/ho) o-o Sole Cowpeo (50 x IO pit/ho) 14 SMoize/Cowpeo Intercrop (30/40xO3 pit/ho) 12 10 o10 LATE 1979 Pod of Leucaena leucocephala, a tree type legume S] being used for soil fertility restoration. 66 6 60 2 0 2 4 6 8 6 WEED INTERFERENCE DURATION (WAP). Fig. 38. The effect of weed interference on food energy values of maize/cowpea intercrop. -, Collecting leaves from Leucaena leucocephala. The dried leaves will be used as a source of plant nutrient and organic matter, the stems as stakes and firewood. 36



PAGE 1

Table 46. Some characteristics of Apomu surface (0-15 cm) soil before and after alley cropping. Treatment* Org. C Total N Extr. P IN Am. Acetate Extr. kg N/ha pH-H,20 (%) (%) (ppm) K Ca. Mg. Na ECEC me/100g 1976 (Before alley cropping) 6.2 0.98 0.12 24.7 0.25 2.63 1.02 0.06 4.40 End 1980 (with alley cropping) ON" 5.7 0.96 0.11 19.4 0.16 5.07 0.35 ON 5.7 1.47 0.12 21.5 0.16 5.33 0.43 30N 5.5 1.24 0.13 18.7 0.14 5.72 0.34 60N 5.3 1.41 0.13 22.8 0.18 4.43 0.28 LSD (5%) 0.1 0.67 0.01 3.3 0.07 1.18 0.08 *Major season 1980 N rates 40 and 80 kg N/ha; Minor season 1980 N rates 30 and 60 kg N/ha. **Leucaena prunings removed from this treatment. Table 47. Legume leaf yield and potential N yield in (2) Fast growing pulp, fuelwood and/or soil amelioratearly stages of alley cropping. ing species, such as Cordia alliodora, Albizia fa/Alley width Leaf yield N yield cataria, Cassia siamea, Sesbenia grandiflora, Legume cm kg/ha dry wt. kg/ha Flemingia congesta, Samanea saman and Gmelina arborea. Pigeon pea 225 1321 47.6 (3) Important multipurpose trees and shrubs, such as 375 793 28.6 Treculia africana, Irvingia gabonensis, Dacryodes 675 440 15.9 edulis, Pterocarpus spp., Chrysophyllum albidum, Leucaena 225 1084 45.5 Afzelia spp. Blighia sapida and Parkia clapperton375 651 27.3 iana. 675 361 15.2 Tephrosia 225 600 22.8 375 360 13.0 675 200 7.65OT Tephrosio Condido C Coanu.s Cajon / L Leucoeno Leucocephalo G Glircidio Sepium 40 2"25 (m) alley width 20 W -J 30 .,. -U 675(m) z -.J 0 W )~O ,.0 0 T C L G Fig. 41. Effect of shrub legume species and alley widths on maize grain yield. strongly acidic Ultisols at Onne. Emphasis is given to those species that are established easily and can be maintained through basal sprouts and coppices when cut back periodically. The screening and evaluation includes more than 30 woody specimens as listed below: (1) Planted and/or retained woody fallow species, such as Acioa barteri, Anthonatha macrophylla, Alchornea cordifolia, Dalium guineense, Gliricidia sepium, Leucaena leucocephala and Tephrosia Acioa baterii, one of the shrubs being tested for imcandida. proved bush fallow. 38



PAGE 1

year of growth may be the only practical way of using them for alley cropping. Establishing leguminous trees or shrubs for alley cropPure Maize ping in the derived savanna. Establishment and early 2.0MaizePigeon pea maintenance costs appear to be major deterrents in the Moize -Leuceeno use of leguminous trees or shrubs for nutrient recycling Maize -Tephrosia in alley cropping. To overcome these problems, the leMaize -Gliricidia gumes are being established by interplanting through maize. A trial evaluating the potential of this method in .: the derived savanna was initiated in 1980. In the interplanting, the legumes are spaced at 400 cm x 50 cm and the maize at 100 cm x 100 cm with 3 plants/hill. Seeds "iC were used for Tephrosia candida, Cajanus cajan and 0 1Leucaena leucocephala and cuttings for Gliricidia. Because of adverse rainfall during the first season, maize 0 yields were lower than expected (Fig. 44). The legumes had no significant effect on maize yield, and the maize had no significant effect on legume yield; legumes planted without maize were the same size as those planted with maize. Fig. 44. Effect of shrub legumes on single maize crop. Establishment of Tephrosia was very poor while that of the other legumes was good. Growth of both Leucaena and Gliricidia was poor even though adequate weed control was practised during the intercropping phase. By A' ~.~ November (6 months after sowing), these 2 legumes were growing slowly with no marked effect on the adjacent vegetation. On the other hand, Cajanus cajan grew N~ 4 vigorously and by November dominated the field. A well A/ developed canopy shaded out weeds preventing many from flowering (Fig. 45). A closer between row spacing (about 300 cm) could produce better weed control although a closer spacing may hamper tractor drawn implements. The slow growth of Leucaena and Gliricidia suggests that these legumes may require more than 1 growing season to significantly affect field ecology. Vigorous growing Cajanus cajan interplanted with maize in the 300 .?00 4000 00 o000 Twenty-month-old Albizia falcataria. Among the fastest 0 0 c00 00 0 growing trees in the world, they are suitable agro-forestry species in the humid lowland tropics for control of Fig. 45. Pigeon pea canopy structure and weed distriImperata cylinderica grass as well as for site fertility bution in 4 m alley, 3 months after maize harvest, improvement and in the derived savanna as a planted (maize/pigeon pea -, ( pure pigeon pea fallow species. -* 40



PAGE 1

Table 32. Changes in soil properties of a forest Alfisol 1 year after forest clearing (0-10 cm) (IITA, 1979-80). NOF3-N Property and At clearing After one year # standard error (1979) (1980) .Bulk Density, g/cm 1.04 0.05 1.22 0.02 Organic C,% 1.55 0.07 1.24 0.06 Total N, % 0.171 0.007 0.146 0.007 Organic P, ppm 1947 1607 4,' #Crop residues were removed after each harvest. Feb Mar Apr.I My Jue July Se Oct Nov mo,. 98 Table 33. Average grain and dry matter yields of Fig. 19. Seasonal fluctuation of nitrate and ammonium soybean and maize from 32 subplots (IITA, -N in a newly cleared forest Alfisol (0-10 cm). 1979-80). Grain Yield Dry Matter (12% moisture) (Oven Dry) the N03-N and NH4-N content in the soil were low, indiCrop kg/ha kg/ha cating that a higher rate of N fertilization would be required for the second maize crop. There were significant Soybean (TGM 294) changes in soil organic C, N and P as well as bulk density (2nd season, 1979) 1759 5 4605202# 1 year after clearing (Table 32). The decline of soil orMaize (TZPB) 2639 183 4709 20" ganic C, N and P is primarily due to mineralization as soil #Total dry matter yield sampled at maturing stage. erosion and surface run-off on this carefully cleared plot **Stover yield only at harvest. (64 subplots) are minimal. The second season maize, however, suffered severe streak virus disease. SGrain and dry matter yields of soybean and first season Continuous fertilization and cropping maize are given in Table 33. The grain yield of the first Continuous fertilization and cropping season maize was considerably lower than would be preThe long-term fertility trials initiated in 1972 on an Alfisol dicted from its early growth condition. This is prob(Egbeda Series, Oxic Paleustalf) and Entisol (Apomu Seably because of the long dry spell during the grain-filling ries, Psammentic Usthorthent) to investigate the long stage that severely affected the grain yield of the 1980 term N, P, K, Mg, S and Zn responses of these soils folfirst season crop without supplementary irrigation, lowing land clearing were continued in 1980. Maize LSD (5%) LSD (5%) I I I I I I Il I II J I I I I 8-0 7.0 6'60 N2 P2 K2 S50 N2P2K1 5.0N N2 P2 KI 54"0-N 2 P2 K2 0. N2N2 P2Ko SNP2K0 2 P2 K0 S 30W 20 EGBEDA SOIL APOMU SOIL N(OXIC PALEUSTALF). (PSAMMENTIC USTHORTHENT) 1.0" 0 72 73 7i4 75 76 77 78 79 86 72 73 74 75 76 77 78 79 806 YEAR YEAR Fig. 20. Potassium response of maize grown on Egbeda and Apomu soils in long-term fertility trials at Ibadan, Nigeria (K-rates; K, = 0; K, = 40 and K, = 80 kg K/ha). 23



PAGE 1

most promising cassava clones at IITA between 1973/ the last 2 columns refer to: 1974 and 1977/1978 (IITA Annual Report 1977). The rePhysical condition modifiers: suits, indeed, show a negative and highly significant linw -low available water reserve ear relationship between the cumulative pan evaporation r -high soil erosion hazard from 1 December through 28 and 29 February (the dry c -high soil compaction hazard season for IITA) and the cassava yield. The relevant equaChemical condition modifiers: tion is as follows: Y = 176.69-0.3421, r = -.978*, where k -low potassium reserve is the yield of cassava and I (1 = Dec., n i -high phosphate fixation d (Ep) (1 t -secondary/micronutrient deficiencies and/or imbalFeb. 28/29). This relationship could prove useful in estiances nation of the performance of cassava across climatic a -aluminum toxicity for most legume crops m -manganese toxicity to most legume crops zones. -potential soil toxicity and/or secondary and micronutrient deficiencies and imbalance due to continuous cultivation with conventional chemical fertilSoil and land characterization and ization. evaluation Tentative quantitative limits of the "soil condition modifiers" have been defined, and refinement of these can be introduced in the future on the basis of further research Technical soil evaluation system based information. on soil mineralogy It is proposed to integrate this soil evaluation system in Two distinctive approaches have evolved in recent years the land types and agroclimatic information so as to esregarding the management of tropical soils for higher tablish comprehensive guidelines for land clearing and agricultural productivity. The high-energy input, extenmanagement for different regions of the tropics. Close sive food and cash crop production systems are well collaboration with FAO and other development agencies suited for the fine-textured oxidic Oxisols, Alfisols, Ultiand interested national soils research institutions will be sols and Inceptisols; whereas, agroforestry and low-enfruitful in further development of the system. ergy input, food crop production systems are more suitable for the kaolinitic and siliceosis Alfisols, Ultisols and Inceptisols. Soil erodibility characterization A key factor differentiating these 2 major categories of Field experiments have been established to directly monsoils and their response to agricultural exploitation is itor soil erodibility at 3 locations in Nigeria-Onne, Ikom soil mineralogy. A technical soil evaluation system using and Jos. In addition, soil detachability and transportabilmineralogical characteristics as the main criterion is ity measurements on 20 soils collected from different being developed at IITA with the primary objective to parts of Nigeria are being made with a laboratory rainfall provide agricultural planners in the tropics with a set of simulator. The accumulative infiltration measured for 3 simple guidelines for agricultural soil utilization. It is inlocations prior to establishing the runoff plots are detended to provide supplementary information to the esscribed by the following equations: tablished soil classification systems with special I = 30.8t,/2 + 26.1t ... Onne reference to agricultural soils in which variable charge 1= 16.0t,/2 + 93.3t ... Ikom colloids dominate. Vertisols, Alfisols, Ultisols and IncepI = 2.1t1/2 + 6.3t ... Jos tisols dominating in high activity clays or constant Where I is the accumulative infiltration in centimeters, charge minerals are excluded from this system since the and t is the time in minutes. agricultural soil evaluation systems for such soils are well established for the temperate and subtropical reField measurements of soil bulk density for 20 soils indigions. cated a range of 0.70g cm-3 for Ikom to 1.5g cm-3 for Bakura and Tumu, Nigeria. Similarly, the soil-water The proposed system may be briefly summarized as foltransmissivity ranged from 236 cm/hr for Ikom to 1.4 cm/ lows: hr for Samaru, Nigeria. A plot of the accumulative infiltration vs. time for some Nigerian soils is shown in Fig. 9. 'Condition modifiers Estimates of erodibility for some Nigerian soils by using (or soil fertility the USDA Nomogram indicate a range of 0.05 f/acre/footlimitation) ton for Ikom to 0.56 t/acre/foot-ton for steep lands near Soil group Sub-group Physical Chemical Abakaliki, Nigeria. A majority of the soils, however, have a low erodibility of about 0.1. Kaolinitic soils Eutric w, r, c t*, (m*) Dystric w, r, c t, k, a, (m) Siliceous soils Eutric w, c t*, k*, (m*) Hydromorphic soils in West Africa Dystric w, c t, k, a (m) A study was carried out to assess the quality and distriEutric w i bution of hydromorphic soils in various wetland areas in AIophanicsoils Dystric w i, t, k, a West Africa and their suitability and limitations for Dyssoils trick -i, krainfed and irrigated rice production. Dystric i, t, k Hydromorphic soils could be defined simply as those The eutric subclass refers to high "base" status, and dyssoils where water can gather in sufficient volume and tric to low "base" status. The soil fertility limitations in time to produce the effects of gleying or reducing re5



PAGE 1

e7,,.,nt was only 2 mm. Deforestation contributed to a Effect on cassava growth and yield. Seedling mortality .-. ::nt amount of seepage or ground water flow. The was about 50 percent in these treatments because of .-..., :en: stream with only traces of flow during peshading by trees in the traditionally managed plots and -.,eavy rains is now a perennial stream with meaby maise in the no-tillage plots (Table 10). Since cassava ..ow throughout the year. The maximum surface was planted in no-tillage treatments through 6-8 week&as observed in September; the subsurface flow old maize, row spacing was often more than 1 m (some.' er was almost equal to surface runoff. Continutimes 1.5 m), which also contributed to a low plant pop-.:,..:..through the dry season also indicates the posulation. Cassava tuber yield varied by a factor of 2-2.5 cf sizeable ground water storage that may be among treatments with the lowest yield in the tradition.... nt at least in analyzing the hydrological balance ally managed plots. Conventional tillage plots were .ce~~red watershed. planted at least 6 weeks later and harvested about a month after no-tillage plots. Tuber yield in conventional tillage plots would have been even more if they had been Table 8. Effects of land clearing methods on soil bulk harvested 2-3 months later. Nevertheless, tuber and stalk density in glcm for the 0-5 cm depth (IITA, yield from the no-tillage plots were comparable with 1978-80). those from the plowed and ridged plots. T, itment 19781 1979 1980 -.1ni.onal farming 0.64 1.06 1.07 Table 10. Effects of land clearing methods on cassava ',' inual clearing-no tillage 0.68 1.14 1.05 growth and yield (1ITA, 1980) ,,'inual clearing-conventional tillage 0.68 1.20 1.29 Sretar blade 0.70 1.19 1.37 Yield Tr,?e pusher-no tillage 0.66 1.25 1.37 Plants! Tubers! t/ha Tree pusher-conventional tillage 0.53 1.22 1.32 Treatment ha ha Tubers Stalks ,'ean of 25 replications. Traditional farming 4,540 31,570 7.7 14.6 1)78 data prior to clearing. Manual clearingno-tillage 5,230 36,640 15.0 22.4 Elfect on surface runoff and erosion. Soil erosion and Manual clearing,.atter runoff were lower for cassava than maize. Moreconventional-tillage 4,850 36,180 11.7 26.1 ;';er. there were no significant differences among treatShear blade-.fnts either for runoff or erosion, except for the tree no-tillage 5,420 30,550 14.1 14.6 pusher-conventional tillage treatment. In spite of the Tree pusherprotective cover of the cassava canopy, this treatment no-tillage 5,800 39,580 16.8 had 48 mm of water runoff and 4.2 t/ha/annum of soil Tree pushere2rosion (Table 9). In comparison to this, runoff and eroconventional-tillage 12,700 59,940 17.5 23.0 sion from the shear blade-no-tillage cassava was 5.0 Table 9. Effects of land clearing methods on surface Tillage systems and small tools runoff and erosion (IITA, 1980) development RunErooff, sion, Effects of tillage methods on maize production. NoTreatment mm tha tillage methods with residue mulch have proven useful Traditional farming 0 0 for some row crops in kaolinitic Alfisols in the forest Manual clearing-no tillage 0.2 0.001 zone of Western Nigeria. However, long-term studies at Manual clearing-6.n0 tlag 0 0.06 IITA have indicated that soil compaction can be a probSha clearing-conventional tillage .0 0.0 lem in no-tillage plots within 3-4 years. Moreover, soil Sre blade-no tillage 5.0 0.08 compaction is more severe on mechanized than manTree ousher-no tillage 3.0 0.04 ually cultivated plots. Crop residue mulch is needed for Tree pusher-conventional tillage 48.0 4.2 many other uses (fodder, building houses and fences, fuel, etc.) and, therefore, may not be always available in mm and 0.08 t/ha. On the contrary, runoff and erosion the quantity required for effective soil and water conserfrom the shear blade-no-tillage maize treatment was 53 vation. Chiseling in the row zone rather than plowing the mm and 1.9 tlha'annum, respectively. These results on entire field, which makes the soil vulnerable to erosion, the effects of methods of land clearing and post-clearing may be an alternative to ameliorate the soil of the comsoil management have important practical agronomic paction hazard. Plowing at the end of the rainy season Implications. It seems that, in the long run, post-clearing may be another method. With this background, an expersoil management has the most important effect on soil iment was carried out at IITA with the following treaterosion, runoff and decline of soil physical and chemical ments: properties. This is not to say that the methods of land A. No-tillage with residue mulch. clearing are not important because a combination of B. No-tillage with chiseling in the dry season. harmful land clearing and post-clearing soil manageC. Moldboard plowing followed 2 harrowings (residue ment methods, such as land clearing with tree pusher/ plowed in). root rake followed by conventional plowing and harrowD. Disc plowing (residue disced). ing, results in the most losses in water runoff and soil E. No-tillage with residue removed. erosion and in rapid degradation of soil physical and F. Moldboard plowing at the end of rainy season and chemical properties. harrowing at planting. harwn t lnig



PAGE 1

although to a much lesser extent compared to the no 10 crop residue plots (Table 30). Moreover, properties of soils under continuous mulch, no-tillage maize for 8 9 years were by no means comparable to that under bush No-,il Stow, Muhed and grass fallow. Planted fallows such as Lucaena with V.7182.23756667X--I2X2 both leaves and branches returned as surface mulch compared favorably to that of natural bush fallow; whereas, results of pigeon pea (bush type) were inferior. 7 The latter result is mainly due to severe disease problems that cause the crop to die off after 1 year. Lack of effeco6 NO-1,S .. w ... tive ground cover during or before the reestablishment .629 25.M56X652592 '_1 of the new crop may have caused the loss of Ca, Mg and R2. 0,7%9 K from the surface soil by eroding and leaching. \ Effect on earthworm activity. There was a remarkable N difference in earthworm activity between the fallow and cultivated plots. Earthworm activity monitored during a 4-day period in July, 1980 (Table 31), showed that the natural bush fallow plot had a significantly higher Hyper2 iodillus activity (columnar casts) than the 2 planted fallow plots (Leucaena and Guinea grass). The extremely low earthworm activity in the continuously cropped plots (8.5 years) is apparently due to many factors. Soil compaction, continued use of pesticides and herbicides and k2 Pin Ik 175 197 d77 198 079 980 the decline in soil organic matter are probably among the important ones. The results of Eudrillus activity Fig. 18. Calculated yield curves of maize for a period of (granular casts) taken during the same period were less 9 years under no-tillage with and without stover mulch indicative. Although the difference among the fallow and on a kaolinitic Alfisol. cropped plots was not statistically significant, the Eudrillus activity under Leucaena was considerably greater 0an that under bush and Guinea grass as well as the Mineralization of soil organic matter. cropped plots (Table 31). A small-scale, manual, land clearing project (0.25 ha) was carried out to study the soil organic matter decomRole of crop residue mulch on maize yield under noposition and nutrient release after forest clearing. A sectillage system. Respectable grain yields under a mulchondary objective of the experiment was to demonstrate no-tillage system on manually cleared land were mainwhether improved soil and crop management practices tained up to 4 years; but yield declined steadily, thereat the small farmer's level could, in fact, increase the after, despite adequate fertilization and plant protection length of the period under cultivation before the land is (Fig. 18). The reason for the yield decline is a complex returned to bush fallow. The improved practices inone. The decline in soil biomass activity due to cultivaeluded no burning, no tillage and the use of mulches, tion, soil compaction and possible Mn toxicity due to soil fertilizers, preemergence herbicides, pesticides and an acidification are probably among the more important improved crop variety. Soybeans were planted immedigrowth-limiting factors that cannot be simply remedied ately after land clearing during the 1979 second season by conventional fertilization. The beneficial effect of crop followed by maize during the 1980 first season. Fertilizer residue mulch in the no-tillage maize system is evident, was not applied to either crop. Excellent growth of both particularly in areas of ustic soil moisture regime and the soybean and maize was observed, and tissue analysis frequent incidences of dry spells during the early cropindicated no nutrient deficiency. Results of weekly monping season. These results suggest that to avoid perma-, itoring of NO,-N and NH4-N in the surface (0-10 cm) soil nent degradation of the "superficially" fertile Alfisols in are given in Figure 19. Substantial amounts of soil orthe forest/savanna transition zone of West Africa, cultiganic N were mineralized during the onset of the rainy vated land needs to be returned to an effective fallow or season. These data further suggest that, in spite of riskresting period after 4-5 years of cropping under recoming a drought stress, early planting of maize is preferred mended soil management practices, i.e., the mulch noso that the crop can effectively utilize the high levels of tillage system. mineralized N in the soil during the second season, both Table 31. Earthworm activity in fallow and cropped plots taken during a 4-day period In July, 1980 (IITA). Hyperiodillus (columnar cast) Eudrillus (granular cast) Treatment g/250 cm2 No./250 cm2 g/250 cm2 No./250 cm2 Natural bush 10.46 a 12.98 a 0.01 a 7.66 a Guinea grass 7.83 ab 10.02 ab 0.04 a 3.22 a Leucaena 3.98 bc 6.85 abc 2.36 a 29.30 a Maize/Cassava 2.15 c 5.78 abc 0.01 a 8.48 a Maize + residue 0.60 c 1.79 c 0.02 a 2.35 a Maize -residue 0.004 c 0.02 c 0.00 a 2.67 a 22





PAGE 1

Board of Trustees Mr. Guy Vallaeys (Chairman) Dr. Willi Ehman Institute de Recherches Agronomiques Tropicales et de Ministry of Economic Cooperation, 53 Bonn 12, Cultures Vivrieres (IRAT), Paris, France Germany Mr. D. lyamabo (Vice Chairman) Prof. R. J. Olembo Ministry of Science and Technology, Ibadan, Nigeria Div. of Ecosystems and Natural Resources, UNEP, Nairobi, Kenya Dr. R. C. McGinnis Dean of Agricultute, University of Manitoba, Winnipeg, Citoyen Mukendi Mbuyi Tshingoma Canada FAO, Bamako, Mali Dr. Hidetsugu Ishikura Dr. F. F. Hill Director General, Institute of Environmental Toxicology, Ford Foundation, Ithaca, New York 14850 Tokyo, Japan Dr. John J. McKelvey, Jr. Alhaji M. Liman Associate Director, Agricultural Sciences, The Federal Ministry of Agriculture, Lagos, Nigeria Rockefeller Foundation, New York, N.Y. 10036 Dr. E. H. Hartmans Prof. C. A. Onwumechili Director General, IITA, Ibadan, Nigeria Vice Chancellor, University of Ife, Ile-lfe, Nigeria V



PAGE 1

Table 3. Nutrient status of 37 hydromorphic soils from Table 4. Properties of 27 hydromorphic soils from the the forest region of West Africa (southern Nisavanna zone and the forest/savanna transigeria, Sierra Leone and Liberia). tion zone of Nigeria. Properties Distribution, % Surface soil Subsurface soil and range Surface soil Subsurface soil Properties Range Mean Range Mean Exch. Ca, meqIlOOg Forest/savanna transition zone (17 soils) 0-0.50 41 59 pH (H20, 1:1) 4.7-6.4 5.6 5.0-6.5 5.8 0.51-2.00 35 22 Organic C, % 0.37-1.96 0.91 0.07-1.36 0.42 2.01-4.00 5 5 Clay, % 4-68 15 3-70 21 4.01-8.00 11 8 Silt, % 4-51 21 5-53 18 > 8.00 8 5 Sand, % 13-89 64 10-84 61 Exch. Mg, meqIlOOg Exch. cations, meq/100g 0-0.50 70 78 Ca 0.67-12.83 3.48 0.47-11.13 3.34 0.51-1.50 11 5 Mg 0.29-7.40 1.53 0.21-7.89 1.67 1.51-2.50 8 5 K 0.03-0.26 0.13 0.02-0.46 0.11 2.51-3.50 8 3 ECEC, meq/100g 1.51-23.60 5.99 1.06-21.26 6.14 > 3.50 3 8 Bray P1, ppm 2-14 5 0.2-4 2 Savanna zone (10 soils) Exch. K, meqIlOOg pH (H20, 1:1) 4.2-6.1 5.3 4.6-7.4 5.7 0-0.05 24 65 Organic C, % 0.29-2.70 1.04 0.07-0.75 0.42 0.06-0.15 35 22 Clay, % 6-50 22 5-70 28 0.15-0.25 30 14 Silt, % 8-51 37 6-55 29 0.25-0.35 11 0 Sand.% 4-82 42 5-89 43 Extractable P (Bray 1), ppm Exch. cations, meq/100g 0-5 51 89 Ca 0.55-7.94 3.50 0.40-6.84 3.34 6-15 35 8 Mg 0.20-3.36 1.32 0.06-3.25 1.30 16-25 8 0 K 0.04-1.92 0.40 0.04-0.29 0.14 25-35 0 3 ECEC, meq/100g 1.73-11.31 6.26 0.99-10.61 6.13 > 35 5 0 Bray, P1, ppm 2-39, 9 0.4-5 2 Source of Data: IITA Soil Information Bank, Njala University College and University of Illinois Soil Survey Report for Sierra Leone (1974); Ministry of Agriculture and this region are montmorillonitic and, thus, of high clay USAID Soil Survey Report for Liberia (1977), Manor River activity. Hydromorphic soils with more favorable clayey Soil Survey Report, Liberia (1979). and silty texture are found in some extensive areas in the river basins of central Nigeria. But, because of generally low nutrient status and seasonal flooding, effective water also stressed. Utilization of these lands at traditional and management and fertilization are required in order to improved levels of management for year-round crop prodevelop such areas for rice production. duction using wetland rice as the main season crop was proposed. An estimate is being made of the distribution of hydroFurther inventory of properties of 17 selected soils from morphic soils of selected areas of West Africa using the region are given in Table 4. Again, the predominant LANDSAT imageries and a few large-scale soil maps that hydromorphic soils in this region are coarse and medium are available. The selected study areas include the textured, but their chemical quality is slightly better than wetland tracts along rivers of West Africa and inland their counterpart in the high-rainfall region in terms of swamps, such as those in Liberia and Sierra Leone and available Ca, Mg and K in the surface horizons. Available the wet sedimentary belts of southern Nigeria. Vertisols P status is invariably low although these soils have very and the vertic subgroups are widely scattered but not low P fixation capacity. fully documented on a national and regional basis. Savanna zone (central and northern Nigeria). Hydromorphic soils in the savanna zone occur in a wide range Socio-economic analysis of land forms such as inland depressions, river valleys and ancient and recent flood plains. The sedimentary Labor utilization plains of the Niger and Kaduna Rivers and their tributaries comprise a significant portion of potential rice Literature was reviewed on labor utilization in cassava, land in central Nigeria, which is yet to be fully developed yam, maize and upland rice production. Additionally, and utilized. from short regional crop production surveys-preliminary labor utilization data were derived for cocoyam and The inventory of hydromorphic soils in this region is exsoybean production in Nigeria (Table 5). ploratory due to lack of soil survey information. A preliminary inventory of soil properties of 10 selected sites in In 1980, further progress was made in the collection of central Nigeria (excluding Vertisols) is given in Table 4. time series data for food crop prices in West African Hydromorphic soils at these locations show wide varicountries. In addition to food crop prices in Nigeria and ability in texture, organic matter and soil reaction. MinerCameroon, market prices were obtained from Ghana, alogical studies showed that many hydromorphic soils in Ivory Coast and Togo. 7



PAGE 1

Table 51. Mean values of yield and yield components of white yam as affected by duration of weed interference in the field (1978-79). Time (weeks) LA I to 100% tuber Fresh tuber Duration of (at 12 WA P) initiation yield (t/ha) weed interference 1978 1979 1978 1979 1978 1979 W. F. M. 0.55 ab' 0.58 a 12 a 10 a 20.08 a 17.57 a W. 1. 4 0.87 a 0.21 cd 14 d 10a 18.75 ab 17.40 a W. 1. 8 0.29 b 0.18 d 14 a 10 a 13.96 bc 13.91 a W. I. 12 0.22 b 0.16 d 14 a 10 a 12.47 bc 7.40 bc W. 1. 16 0.29 b 0.04 d 14 a 10 a 6.47 d 6.06 bc W. I. 20 0.21 b 0.06 d 14 a 10 a 6.58 d 3.24 c W. I. 24 0.30 b 0.02 d 14 a 10 a 4.88 d 1.89 c W. 1. M. 0.24 b 0.06 d 14 a 10 a 6.03 d 2.12 c Weeded 3 + 8 + 12 0.94 a 0.74 a 14 a 10 a 19.71 a 12.25 bc Weeded 3 + 8 + 12 + 16 0.82 a 0.55 bcd 14 a 10 a 20.88 a 17.28 a 1Means followed by the same letter in the same column are not significantly different at the 5% level of the Duncan's New Multiple Range Test. In the absence of weed interference, dry matter produc3.0 and 6.0 kg/ha while pendimethalin was sprayed at 2.5 tion in yam vines and leaves peaks in phase II (vegetative and 5.0 kg/ha on a conventional-tillage plot using a knapgrowth stage) and phase Ill (tuber bulking growth stage). sack sprayer calibrated to deliver 200 1/ha spray volume. Weed interference during these periods severely reduces Herbicide persistence was monitored by a bioassay growth in those organs, and this reduction during method involving tomato seedlings for atrazine and fluphases II and III (12-16 weeks after planting) interferes ometuron and rice seedlings for metolachlor and pendiwith assimilate production necessary for optimum tuber methalin. bulking in the later part of phase I1l. Atrazine and metolachlor each at 3.0 kg/ha were less Mamatter production in weeds peaked at 8-12 weeks persistent than fluometuron and pendimethalin (Fig. 49). and declined thereafter, indicating that the maximum Generally, all herbicides were more persistent at IITA competition occurred during this period. Decline in tuber (rainfall 1,400 mm) than Onne (rainfall 2,400 mm). Atrayield was also most pronounced at this period (Fig. 48). zine at 3.0 and 6.0 kg/ha was present in soils at IITA at nonphytoxic levels at 8 and 12 weeks after treatment, respectively (Fig. 50). The persistence of metolachlor MOP GOWTH PHASE 0Weed ftr up to naturcy closely followed that of atrazine while fluometuron and ..4 We -,nfested upto t-y pendimethalin showed pronounced persistence at IITA 01 v I i Iv at both low and high rates. Fluometuron, especially at a CO> I high rate, persisted beyond the test period in the IITA c joIs soil. Pendimethalin at 2.5 kg/ha persisted at high levels oo for more than 8 weeks after treatment, but existed at very 002 I low levels at 12 weeks after treatment. 12 I Prisene!" a as shr 'Persistence ofatrazine in soils at Onne was very short o8 VINE 5 and had declined to non-phytoxic levels by 8 weeks after S treatment (Fig. 49). Doubling the rate of atrazine did not I I increase its persistence. Metolachlor applied at a high 180 oo 010 rate persisted in Onne for more than 8 weeks after treatS LEAVES 1 3 ment. Fluometuron could not be detected in the top soil .20 I 'p, at Onne by 12 weeks after treatment. Persistence of pen120, 0sodimethalin was identical at both IITA and Onne, irrespec060 tive of rates used. While an atrazine sensitive crop could be planted at IITA e -. o 1 and Onne locations at 12 weeks after treatment, a flu8 2 6 20 2 28 32 8 2 20 2' 32 ometuron sensitive crop could not be safely grown under WEED INTERrEPENCE (WAP) WEED INTERFE ENCE (WAP) soil and rainfall conditions at IITA within 12 weeks after treatment. All herbicides tested in soils at Onne were Fig. 48. Effect of weed interference on yield and yield either nontoxic or had very low herbicidal activity at 12 components in yam. weeks after treatment. Herbicide persistence in tropical soils Plantain improvement Field experiments were carried out at IITA and Onne to Previous greenhouse trials indicated that growth regudetermine the persistence of some selected commonly lators applied close to the meristem of young plantain used preemergence herbicides in the 2 ecologies. Atra(Musa sp.) suckers changed growth hormone balance zine, metolachlor and fluometuron were each applied at and sucker development. This suggested the potential 44



PAGE 1

Contents v Board of Trustees vi Foreword 1 Farming Systems Program 1 Introduction 1 Baseline data collection and analysis 1 General weather conditions 3 Agroclimatic analysis 5 Soil and land characterization and evaluation 7 Socio-economic analysis 10 Land and soil management 10 Land clearing and development 11 Tillage systems and small tools development 21 Management of kaolintic Alfisols 28 Management of siliceous Ultisols 32 Cropping systems 32 Intercropping agronomy and meso/micro-climatic studies 36 Alley cropping i 42 Live mulch systems 43 Weed biology and herbicide research 44 Plantain improvement 48 Eastern Cameroon Farming Systems Project 49 Small farms systems research (Atebubu District, Ghana) 51 Root and Tuber Improvement Program 5.1 Introduction 51 Cassava 51 Genetic improvement 54 Entomology 59 Pathology 61 Nematology 62 Tissue culture and virus indexing 62 Zaire's National Manioc Program (PRONAM) 64 Cameroon National Root Crops Improvement Program (CNRCIP) 64 Yam 65 Genetic improvement 66 Pathology 70 Nematology 72 Quality evaluation 72 Tissue culture and virus indexing 72 Cameroon National Root Crops Improvement Program (CNRCIP) 72 Sweet Potato 73 Genetic improvement 74 Entomology 76 Nematology 77 Quality evaluation 78 Tissue culture and virus indexing 80 Cameroon National Root Crops Improvement Program (CNRCIP) 83 Cocoyam 83 Genetic improvement 83 Agricultural economics 84 Cameroon National Root Crops Improvement Program (CNRCIP) III



PAGE 1

SPRAY BOTTLE 3 Poleustalt II TA, Ibadan After 8years DELIVERY TUBE DELIVERY TUBES 2--/ E / U MAIZERESIDUE Li 2-0-" 2,A TEE I FIRST MESXG SECOND DESIGN E MAIZE. RESIDUE Fig. 16. Schematic diagram of delivery tubes for 2 head I~ CDA sprayer. 2: S .'0PANICUM MAX 0 flow in the 6 mm tubing would at times allow 1 head to uJ take all the flow, leaving the other with no flow at all. w This would happen normally when 1 head was higher nL 0.5than the other. This was mostly overcome by having a BUSH FALLOW separate tube for each head coming from the spray bottle. The sprayer was set up with the spray heads 0.75 m apart to match the row width when working on standing 0 maize stubble. This 2 head sprayer was used in conjunc5 10 1,5 2) tion with the 2-row planter with every other row of maize stover knocked down. The time required for using the 2 SOIL DEPTH, cm heads was approximately 0.6 of the single head. This Fig. 17. Penetrometer readings of the surface layers of increase in the rate of work would probably justify the kaolinitic Alfisols under fallow and after 8 years of conextra expense of the 2 head machine, approximately 1.5 tinuous cultivation. times the price of a single head machine. crop residue return resulted in an increase in soil bulk Management of Kaolinitic Alfisols density and acidity and a decrease in soil organic matter levels, and, consequently, a reduction in CEC and exA long-term experiment established at IITA since 1972 changeable Ca and Mg status in the soil (Table 30). shows that fallow remains a vital component in maintainPenetrometer readings of the cropped plots after 8 years ing the soil productivity of the highly erosive kaolinitic indicated a severe compaction of the surface horizons Alfisols on a predominantly rolling topography in the forcompared to the soils under grass and bush fallow (Fig. est/savanna transition zone of West Africa. This is, in 17). part, demonstrated by the alarming decline of some Continuous no-tillage maize with crop residue returned chemical and physical properties of the soils after 8 as surface mulch twice a year was able to maintain soil years of continuous cropping compared to those under organic matter (i.e., total N) levels comparable to that of natural and planted fallow, bush fallow. But, the decline in soil pH, CEC, exchangeEffect on soil chemical and physical properties. Continable Ca and Mg levels and the increases in soil bulk uous cropping for 8 years with insufficient amounts of density and penetrometer readings were considerable Table, 30. Properties of surface soil (0-15 cm) under 8-year continuous cultivation and fallow after clearing of secondary forest (Oxic Paleustalf, Egbeda Series, IITA, 1980). Bulk density Total (Fine earth) pH N ECEC Exch. cations, meg/100g g/cm3 Treatment (H20) % meq/1OOg Ca Mg K 0-5 cm Continuous cropping with minimum tillage Maize, residue returned 5.0 0.18 3.23 2.19 0.41 0.35 1.20 Maize, residue removed 4.7 0.11 1.81 1.13 0.24 0.11 1.31 Maize/Cassava 5.6 0.15 3.40 2.04 0.42 0.32 1.25 Soybean, residue returned 5.0 0.11 3.05 1.65 0.42 0.28 1.23 Natural and planted fallow Natural regrowth 6.5 0.19 5.14 3.53 0.91 0.41 0.88 Guinea grass 6.7 0.26 7.69 4.75 1.28 0.91 1.01 Pigeon pea 6.0 0.23 3.42 2.18 0.64 0.32 1.10 21



PAGE 1

eties with different growing habits. The relationship beeties have similarly low external P requirements of about tween adjusted P concentrations in the soil at planting .018 ppm P in soil solution. These low external P requiretime and cowpea yields is shown in Figure 26. The 4 ments may be attributed to other growth-limiting factors varieties showed distinct differences in their external P as indicated by grain yields. In spite of the low yield, data requirement. Without P, a local variety, Shaki, gave the in Fig. 27 indicate the relatively higher efficiency of P highest yield and has an external P requirement of about utilization by variety TGM 51 x TGM 344 over variety 0.06 ppm P while with P application, TVx 1193-7D, an TGM 479. improved erect variety, gave the highest yield and has the lowest external P requirement of about 0.016 ppm p. Management of siliceous Ultisols Ife Brown and VITA-4 gave the lowest yields, and Ife Brown has the highest external P requirement of about 0.10 ppm P while VITA-4 has an external P requirement Maize-cowpea rotation of about 0.06 ppm P. From these results, it thus appears A 5-year management trial at Onne shows that a producthat though variety TVx 1193-7D is quite efficient in utiliztive maize-cowpea rotation system is possible on coarseing applied P in low P status soil, the local variety Shaki textured deep and permeable Ultisols (Typic Paleudult) does better. on a predominantly flat to gently undulating coastal land The differential phosphate requirements of 2 soybean form. Maize was sown in early March and cowpea in late varieties, TGM 51 x TGM 344 and TGM 479, were comSeptember or early October. Maximum grain yield pared in a field experiment conducted on an Alfisol at throughout the 5 years ranges from 3.5 to 4.5 t/ha for Mokwa in the subhumid region of Nigeria. Results of the maize and from 1.3 to 1.5 t/ha for cowpea. trial as shown in Fig. 27 clearly indicate that variety TGM A well-balanced fertilization scheme is required for the 479 has a significantly lower yield than variety TGM 51 maize crop, N, P, K, S. Mg and Zn, and fertilization is x TGM 344. Despite their differences in yield, both varigenerally not needed for cowpea in the second season as the residual fertility from the maize crop with residue return as surface mulch is adequate to support the cowpea crop. Reduced tillage with residue mulch is recomvx IM-70 mended because the coarse-textured, kaolin-dominated ,oo o (sssoil may be easily compacted. It is important to point out that the maize, TZPB, and VA -.cowpea cultivars, VITA-1 and VITA-4, tested are fairly tolerant to soil acidity. The critical level of exchangeable Al 10 saturation, i.e., a level required to attain 90 percent of IFE maximum yield, for the maize cultivar ranges between < I0 30-45 percent depending upon the rate of chemical ferSI I tilizer used (Fig. 28). The critical level of AI saturation for 60 .TRIAL At IKENNE OXIC PALEUSTALF MINOR SEASON 1980 00 95. OF 100 X-. -MAXIMUM I -. 200 -.y ,YIELD '_ _' _ ,_ 90" 0Os oI O so 1 ADAIUSTED P CONCENTRATION AT PLANTING. ppm 80 60\ 1 Fig. 26. Relationships between adjusted P concentra. \ \ tions and cowpea yield.. 70\ • 50kg K 60 (1980) 150kg K N 50 (1979) 140 TOM SI (X) TGM 3 A W 40 Maize (TZPB) Max Yield 4-30398 tlha 1979 ist season o000 4.56 tihq 1980 istseason TM 47920 S800-0 0 0 TRIAL At MOKWA 600OXIC PALEUSTALF 0 1980 /95% OF MAXIMUM 980 s005 0 0 0 10 20 30 40 60' 70 90 ADJUSTED P CONCENTRATION Al PLANTING, ppm EXCH. AL SATURATION, % Fig. 27. Relationships between adjusted P concentraFig. 28. Relative yield of maize as affected by percent tions and soybean yield. exchangeable AI saturation in the soil. 28



PAGE 1

stems of 10 different plants and repeated a second time. sponsibilities of IITA is to conduct an agronomic survey Each time 2 ml was injected. Observations were made on of existing farming systems to determine research priorthe mother plant and the second peeper at the time of ities for the regional food crop research. the first injection and 6 weeks after the second injection. Agroeconomic surveys. The agroeconomic survey was Flurenol treated plants did not show faster sucker develconducted with the following objectives: (1) to investiopment nor higher peeper production. Differences in the gate and analyze existing farming systems, (2) to gather main pseudostem after 6 weeks are probably due to the information on existing patterns of agricultural producselected plants in the beginning of the experiment. Thus, tion and resource use, (3) to identify major constraints, to counteract the apical dominance, repeated injections and (4) to collect relevant information on marketing and seem to be necessary to maintain the change in hormone output supplies. A report titled, "Survey of food crop balance. Decapitated plants gave bigger suckers farming systems in the Zapi-Est, East Cameroon," was (maiden suckers) as the control of Flurenol treated published. plants (peepers). Release from apical dominance was Field trials. The field trials on varietal improvement, culnot stimulating higher peeper-number which suggests tural practices and fertilization were initiated in 1980 at that the maiden suckers are dominating. Bertoua in the savanna zone and at Doume in the forest zone. The varietal improvement was carried out with Effect of fertilizer and mulch maize, groundnut, soybean, cowpea and upland and lowland rice. Grain yield of selected maize varieties was 3 to Plantain in backyards are growing under optimal condi4 times that of the local variety; TZB was the top yielder tions because of high level of nutrients, as well as orwith 6.1 tlha. Streak is a major problem for maize producganic matter (Table 54.) In this experiment, the tion in the area. nutritional requirements of 2 varieties of plantain (medium and giant False Horn) were compared with that of Among the groundnut varieties tested, Bertoua Blanche a banana (Kparanta). It is clear that in all cases fertilizer gave the highest yield of 2.1 t/ha. Soybeans performed had strong positive effects on growth, whereas the efwell in the area and yields ranged from 0.6 to 1.9 t/ha. fects of mulch appeared less pronounced. The comThe top yielder was variety 20-67 TB. Though cowpeas bination of mulch and fertilizer showed positive showed good growth, leaf diseases and insect damage interaction. were commonly observed after flowering. Grain yields varied from 0.2 to 1.4 tha with TVx 1948-OIE giving the A highest yield. High yields of lowland rice were obtained fale 54. Height and girth of a medium and a giant with variety IM 16 giving the highest yield of 7.0 t/ha. False Horn plantain and a banana after 5.5 On newly cleared land at Bertoua, grain yield of maize months. variety Ekona Mixed Color responded significantly to N Girth (cm) and P applications at rates of 60 kg N/ha and 40 kg P20,/ Height at 50 cm ha. Positive N x P interactions were also observed. ReTreatments (cm) height gardless of level of fertilizer application, maize yield inFalse Horn giant plantain creased significantly as the plant density increased from No mulch, no fertilizer 85.61 ab* 21.31 abc 33,000 to 66,000 plants/ha. No mulch, fertilizer 133.45 ab 31.14 abcd Field trials with groundnuts showed significant reMulch, no fertilizer 119.14 ab 27.69 abc sponses to application of lime, N and P (Fig. 51). N appliMulch and fertilizer 181.18 bc 42.92 cd False Horn medium plantain Relative yield No mulch, no fertilizer 56.92 a 15.91 a NO No mulch, fertilizer 116.04 ab 27.30 abc Mulch, no fertilizer 123.09 ab 29.39 abcd Mulch and fertilizer 181.15 bc 40.96 bcd I9o O No lime Banana 180 Lime I I ton of CoO/ha No mulch, no fertilizer 94.20 ab 22.94 abc 2 2 tons of CoO/ho No mulch, but fertilizer 188.11 bc 38.78 abcd 170 Mulch, no fertilizer 153.17 abc 31.64 abcd 160I / P I No P2O h5 Mulch and fertilizer 266.95 c 53.14 d / .P0 = g P205/ho '*Means in the same column opposite the same letter(s) 150. 2 1 are not significantly different from each other at 5% lev140 / No 11.4O kN N els. Mulch consisted mainly of Eupatorium. A eN 1 40 kg N/ho 120 Eastern Cameroon Farming ,,o Systems Project 100 rFruit yield of control treatment s 725 kg/ho. As part of the effect to develop the Eastern province of Cameroon in cooperation with Zapi-Est (Zone d'Actions Prioritaires Integrees de lEst), the World Bank asked IITA to assist the then Cameroon National Office of Scientific and Technical Research in establishing a food Fig. 51. Response of groundnut (CV Bertoua Blanche) to crop research station in that province. One of the reN, P and lime applications. 48