TropSoils triennial technical report

Material Information

TropSoils triennial technical report
Portion of title:
Triennial technical report
Soil Management Collaborative Research Support Program
Place of Publication:
[Raleigh N.C
Management Entity Office for the Soil Management Collaborative Research Support Program]
Creation Date:
Publication Date:
Physical Description:
1 v. : ill. ; 28 cm.


Subjects / Keywords:
Soil management -- Periodicals -- Tropics ( lcsh )
Soil science -- Periodicals ( lcsh )
serial ( sobekcm )


Dates or Sequential Designation:
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.

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 ( with any additional information they can provide.
Resource Identifier:
12545890 ( OCLC )
sn 90040075 ( LCCN )

Related Items

Preceded by:
Agronomic-economic research on soils of the tropics
Succeeded by:
TropSoils technical report

Full Text
*.......... .. *

Charles B. McCants, Editor
January, 1985
TropSoils is a collaborative research program whose goal is to develop improved soil management technology for developing countries in the tropics.
Primary funding is provided by the U.S. Agency for International Development through Grant DAN 1311-G-SS-1083-00. This action is in support of Title XII "Famine Prevention and Freedom from Hunger" of the Foreign Assistance Act.
The formal collaborators are: Agency for International Development-USA; Center for Soils Research-Indonesia; Cornell University-USA; Empresa Brasileira de Pesquisa Agropecuaria-Brazil; Institut National de Recherches Agronomiques du Niger-Niger; Institute d'Economic Rural-Mali; Instituto Nactional de Investigacion y Promocion AgrariaPeru; International Crops Research Institute for the Semi-arid Tropics-India; North Carolina State University-USA; Texas A&M University-USA; University of Hawaii-USA.

Goal and Organization 9
Admi ni str at ion 11
Humid Tropics Program **** Indonesia 15
Organi zati on 17
Technical Personnel 19
Overview 21
Research Projects
Detail soil survey of the site for land clearing 3
research 3
Soil spatial variability in newly cleared forest land 33
Reclamation of unproductive, abandoned land 37
Management of organic matter in Indonesian farming 4
systems 4
Potassium dynamics in cropping systems of weathered
soils of West Sumatra, Indonesia 45
Phosphorus rates and method of application 47
Source and method of lime application 51
Residual and maintenance rates for lime 55
Pasture grass and legumes for the humid tropics 59
Soybean variety evaluation 63
Farmer and researcher designed and managed cropping
systems 67
Cooperator farmer interview series 71
Nutrition/diet/income survey 75
Time allocation study 79
Humid Tropics Program *** Peru **** Brazil 83
Organi zation 85
Technical Personnel 87
Overview 89

Research Projects
Stability of Fertilizer-Based Continuous Cropping Systems
Effect of land clearing methods on soil properties
and crop performance 105
Reclamation of bulldozed land 109
Soil fertility dynamics after clearing a tropical
rainforest 113
Continuous cropping of annual crops 117
Soil nutrient dynamics and fertility management
for sustained crop production on Oxisols in
the Brasilian Amazon 123
Phosphorus fertilization alternatives for continuous
cropping systems in Amazon Oxisols 127
Potassium fertilization of Ultisols in Peru 129
Potassium fertilization for annual cropping systems
in Amazon Oxisols 133
Minimum tillage and phosphorus, sulfur, calcium
and magnesium interactions 135
Field estimation of phosphorus retention by Andepts 139
Chemical weed control in upland rice 141
Weed population shifts under high input cropping
systems 143
Low Input Crop Production Systems Aluminum tolerance: cultivar screening 145
Integrated low input cropping system 149
Minimum tillage X residue management X potassium
rates with Al-tolerant cultivars 153
Downward movement of calcium and magnesium in soils 157
Managed kudzu fallow 159
Weed control for low input systems 163
Legume-Based Pasture Production Systems Pasture germplasm adaptation to acid soils of the
humid tropics with minimum inputs 165
Grass-legume mixtures under grazing 169
Fertilizer requirements for pasture establishment 173

Nitrogen contribution of legumes in mixed pastures
in the humid tropics 175
Dynamics of potassium and nutrient cycling in pastures under grazing 179
Effect of sulfur on quality and palatability of
Desmodium Ovalifolium in association with Brachiaria
Decumbens under grazing 181
Tree-Based Production Systems
Forest and soil regeneration 183
Alley cropping 185
Improved fallow 189
Nutritional requirements of peach palm (Guilielma
gasipaes) 191
-Nutritional requirements of Gmelina arborea 195
Guarana fertilization 197
Intensive Management of Alluvial Soils 199
Soil Characterization
New areas: characterization, classification and
interpretation of soils in southeastern Peru
Puerto Maldonado 207
New Areas: Characterization, classification and
interpretations of soils in northeastern Peru 209
FCC refinement and testing 211
Soil Management Research Network for the Humid Tropics 215
Semi-Arid Tropics Program **** Niger **** Mali 223
Organization 225
Technical Personnel 227
Overview 229
Research Projects
Characterization of Soil-Water-Atmosphere-Plant System
Preci pi tati on
Statistical analysis of rainfall records 247
Measurement of the physical characteristics of
rainstorms 251

Studies on soil detachment by raindrop impact and soil transport and deposition by overland
flow 253
Estimating percentage runoff from crusted forest
soils 255
Water and Energy Balance
Moisture retention properties of soils at the
ICRISAT Sahelian Research Center 257
Characterization and modification of the soil
temperature regimes in Niger 261
Water and energy balance of a Nigerien soil 265
Soil physics laboratory 269
Gamma and neutron moisture meter calibration 273
Measurement and simulation of water use under
dryland conditions 275
Soil Physical, Chemical and Mineralogical Properties
Soil survey of the ICRISAT Sahelian Center 277
Soil-geomorphological relationships fo the DallolBosso in Niger 281
Role of amorphous and crystalline iron in the formation of surface crusts and ironstones of
soil in West Africa and Texas 285
Properties and genesis of Vertisols and associated
soils in the Maroua region of northern Cameroon 289
Soil classification and soil loss from steeplands
in Haiti 293
Soil Microvariability
Spatial variability in Niger soils 297
Spatial variability of soil properties 301
Spatial variability in soil chemical and physical
properties in relation to barren and adjacent
forested soils 303
Plant Nutrient-Soil Water Interactions
Soil preparation techniques and the interaction of soil preparation and nitrogen and phosphorus
fertility on sorghum yield 305
Relationship between soil water utilization and
soil plant nutrient status 307

Modification of the Soil-Water-Atmosphere-Plant System
Evaluation of the effects of a neem shelterbelt/ windbreak plantation on wind velocity, soil moisture content and yield of millet 311
Influence of windbreaks on water evaporation from
a bare soil 314
Soil Properties
Effect of low-input soil conservation practices
on erosion losses, runoff and soil moisture status 317
Evaluation of the effectiveness of a sandfighter 319
Rejuvenation of crusted, barren, forest soils
in Niger 323
Acid Savannas Program **** Brazil 327
Organization 329
Technical Personnel 331
Overview 333
Research Projects
Nitrogen availability from legume crop residues and
green manures to the succeeding non-legume crop 339
Laboratory mineralization studies as a soil test for
nitrogen and their correlation with field results 342
Fertilizer nitrogen movement in Cerrado soils 345
Influence of soil texture and liming on phosphorus
and zinc soil test levels and fertilizer management
of Cerrado Oxisols 347
Morphological evidences of a seasonally high water
table in a red-yellow Latosol and its identification
on Lansat images 351
Abbreviations 355

Goal and Organization
The need to increase world food production for the present and foreseeable future is an accepted fact if famine is to be prevented, hunger alleviated and minimum dietary needs met. The actions necessary to meet
this challenge are clear: increase production on existing fields and bring new lands into the system. For either approach to be successful, limitations to be plant growth caused by soil constraints must be mini.mized. The goal of TropSoils is to address this issue. Formally defined, it is to develop and adapt improved soil management technology which is agronomically, ecologically and economically sound for developing countries in the tropics.
TropSoils is an acronym for the Soil Management Collaborative
Research Support Program. It is one of several programs developed to implement Title XII, "Famine Prevention and Freedom from Hunger" of the U.S. Foreign Assistance Act. This legislation sets forth the framework for a collaborative research program involving: (1) U.S. Agency for International Development, (2) U.S. universities, (3) international agricultural research centers and (4) developing country institutions. Within TropSoils, adherence to the principle of collaboration has been a
requirement and maximizing the collaborative effort from all the participants a major goal. Identifiable inputs by all components have been substantial and have had a major impact on the structure, operations and accomplishments of the program.
The TropSoils approach is to focus segments of its program on specific agroecological zones, with primary and secondary research sites within each zone. The zones are: the humid tropics with primary sites in Peru and Indonesia and a secondary site in Brazil, the semi-arid tropics.with a primary site in Niger and a secondary site in Mali, the acid savannas with the primary site in Brazil and the steeplands, which currently is inactive. This approach evolved from the two-year planning process which preceeded its implementations.

Goal and organization
The leadership for detailed program development and for project execution is provided by four U.S. universities. Their identification and the agroecological zone of activity follow:
Cornell University acid savannas
University of Hawaii humid tropics (Indonesia)
North Carolina State University humid tropics (Peru, Brazil,
Indonesi a)
Texas A&M University semi-arid tropics
TropSoils was officially initiated in September, 1981. The dates at which individual programs became fully operable varied considerably due to the time required to complete the necessary administrative actions with collaborating institutions. The humid tropics program in Peru utilized an existing activity and thus was the earliest to become fully operational. The approximate time at which the different programs came on-stream are:
Humid Tropics (Peru): September, 1981
Semi-Arid Tropics (Niger): February, 1983
Acid Savannas (Brazil): June, 1983
Humid Tropics (Indonesia): June, 1983
The fact that TropSoils is a new program which involves collaborative research on complex problems, operates primarily in the field and
under difficult situations in developing areas and has been functional for a maximum of three years at one site and less than eighteen months at the others, has necessarily limited its technical output to-date. Nevertheless, a broad base of well-conceived and scientifically-sound projects have been designed and initiated. Many of them already are having an impact on local actions.
The purpose of this report is to describe the individual TropSoils research projects that are active and to provide a rationale for their place in the program and their current status. Inclusion of data intentionally is omitted because it was not necessary for the purpose of the report and such citations would be premature for most of the projects.

Administrative philosophy and procedures for TropSoils reflect the collaborative nature of the program. Thus, it involves identifiable and significant inputs by all participants.
The Agency for International Development has delegated overall program and fiscal responsibility for performance to the Management Entity, an administrative unit prescribed in the organization of all collaborative research support programs. It receives guidance and recommendations on policy issues from a Board of Directors and on technical matters from the Technical Committee. The Board of Directors is composed of an administrative official from each participating university and collaborating country institution. The university representatives compose an Executive committee of the Board. The Technical Committee is formed from the Program Coordinators from each university.
Research projects are developed by Program Coordinators in collaboration with campus and field-based faculty and in consultation with research and administrative personnel from the collaborating country institution. Advice and concurrence is requested from the respective USAID office to insure that the objectives are consistent with its goals
and priorities.
Program reviews are conducted periodically by an External
Evaluation Panel composed of persons with international agricultural development experience and with no affiliation with any of the participating institutions. Its reports are used by the Management Entity, the Board and the Technical Committee in assessing needs for adjustments or revisions in objectives and approaches.
The following persons have had a major involvement in the administration of TropSoils:

Management Entity Office
Charles B. McCants, Director NCSU
Kim S. Stevens, Administrative Assistant NCSU
Neil Caudle, Editor NCSU
Board of Directors
Morris Bloodworth, Chairman (until 10/83) TAMU
Ada Demb, Chairman (after 10/83) UH
Wenceslau J. Goedert (after 7/84) EMBRAPA
Mamadou Ouattara INRAN
Robert H. Miller NCSU
D. Muljadi (until 1/84) CSR
Edwin B. Oyer CU
Victor Palma INIPA
E. C. A. Runge (after 10/83) TAMU
M. Sudjadi (after 1/84) CSR
Elmar Wagner (until 7/84) EMBRAPA
Technical Committee
Frank G. Calhoun, Chairman TAMU
Douglas J. Lathwell CU
John J. Nicholaides NCSU
Pedro A. Sanchez NCSU
Goro Uehara UH
External Evaluation Panel
John Coulter, Chairman World Bank
Peter Hilderbrand UF
Marlowe Thorne U1

Agency for International Development
John Malcolm, Program Manager AID/S&T
David Bathrick USAID/Lima
Allen R. Hurdus USAID/Jakarta
Adolfo Jurado USAID/Lima
Howard Lusk (after 1/83) USAID/Brasilia
S. K. Reddy USAID/Bamako
Samuel Taylor (until 12/82) USAID/Brasilia
Wilbur Thomas (until 7/83) USAID/Niamey
Frederick Vigil (after 7/83) USAID/Niamey




Humid Tropics Program**** Indonesia
Lead Institution
University of Hawaii Support Institution
North Carolina State University Collaborating Institution
Center for Soils Research Linkage Institutions
Bogor Agricultural Institute
Centro Internacional de Agricultura Tropical
International Board of Soil Research and Management
International Fertilizer Development Center
International Rice Research Institute
Sukarami Research Institute for Food Crops Research Site
Sitiung, West Sumatra, Indonesia Principal Investigators
Lead Institution
Goro Uehara
Support Institution
John J. Nicholaides, III

Humid Tropics Program **** Peru **** Indonesia
Representatives on Board of Directors
Lead Institution
Ada Demb
Collaborating Institution
M. Sudjadi

Humid Tropics Program **** Indonesia
Name, Degree TropSoils Responsibility Affiliation
Goro Uehara, Ph.D. Soil Physics1/ UH
John J. Nicholaides, Ph.D. Soil Fertility= NCSU
Cahyono, B.S. Soil Conservation CSR
D. Keith Cassel, Ph.D. Soil Physics NCSU
Carol J. Colfer, Ph.D. Anthropology UH
Carl Evensen, M.S.* Soil Science UH
Robert L. Fox, Ph.D. Soil Fertility UH
Dan W. Gill, M.S.* Soil Fertility NCSU
Ronald F. Guyton, Ph.D. Agronomy UH
Heryadi, B.S. Soil Fertility CSR
Eugene J. Kamprath, Ph.D. Soil Fertility NCSU
A. Karim Makarim, M.S.* Soil Physics NCSU
Harold McArthur, Ph.D. Anthropology UH
E. Santoso, B.S. Biochemist CSR
D. Santoso, M.S. Soil Fertility-3 CSR
A. Sofyan, B.S. Soil Fertility CSR
Subagio, M.S.* Soil Classification CSR
S. Sukmana, Ph.D. Soil Physics3/ CSR
Sutji, B.S. Meteorologist CSR
John R. Thompson, Ph.D. Agronomy UH
B. Tori, B.S. Soil Fertility CSR
Gordon Y. Tsuji, Ph.D. Soil Physics4/ UH
Michael K. Wade, Ph.D. Soil Management NCSU
Russell S. Yost, Ph.D. Soil Fertility UH
-/Also, Principal Investigator, Lead Institution.
-/ Also, Principal Investigator, Support Institution.
-/Also, Principal Investigator, Collaborating Institution.
-4/Also, Project Manager.
*Enrolled in a graduate program leading to next highest academic degree.


Humid Tropics Program **** Indonesia
The Scope of Work for this activity as set forth in the grant which established the Soil Management CRSP includes the following:
A. Characterize the soil of experimental sites.
B. Test promising methods of land clearing and select one or more
appropriate for existing conditions.
C. Monitor the effects of clearing methods on soil physical properties and identify or devise means for correcting undesirable
D. Determine the amounts of fertilizer and lime needed to produce
satisfactory crops and to sustain yields at levels profitable
for the farmers.
E. Evaluate the potential of grass/legume pasture mixtures in the
farming system.
F. Apply and assess the efficacy of soil conservation measures to
typical areas.
G. Find management systems which minimize energy needs to the
extent possible given local limitations on land,, manpower and
H. Assess the likes, dislikes, needs and resources of farmers of
the area to guide research along lines likely to be beneficial
because of the adoption of results.
I. Disseminate the results of the research to other areas in the
humid tropics.
Prepared by Goro Uehara, Principal Investigator.

Humid Tropics Program **** Indonesia
The principal goal of the program is to uncover principles which will enable resource-poor farmers to adopt soil management practices that will increase family income and farm productivity and at the same time preserve land quality. The research strategy is designed to insure that social, cultural, economic and environmental factors that enhance adoption of a soil management innovation are made an integral part of the research plan. To achieve its goal, the project conducts a major portion of the soil management research with farmers and in farmer fields.
The Setting:
A 100,000 hectare transmigration site in Sitiung, West Sumatra, Indonesia, serves as the project's research area. Six thousand transmigrant families and 1500 indigenous families live in the area. Large cultural and language differences between the Javanese and Sudanese transmigrants and between the transmigrants and indigenous groups present unparalleled opportunities to study the responses of different ethnic groups to soil management innovations.
The soils of the region range in quality from moderately fertile
Inceptisols on river terraces to highly leached and impoverished Oxisols and Ultisols of the dissected peneplain. Mean annual rainfall is 2800 mm and mean annual air temperature is 260C. The tropical rain forest is gradually giving way to rubber plantations and subsistence farming by new settlers.
The first large group of transmigrants settled in Sitiung in 1976. A modest home, 1.25 hectares of recently cleared land and a year's supply of food, fuel, other living essentials, seed and fertilizer awaited each family upon its arrival. Since then, five additional areas in Sitiung have been settled. Bulldozer crews continue to clear more land to accommodate new settlers from the densely populated Islands of Java and Bali. The productive land on the river terraces has long been settled and the newest transmigrants are being placed on the less desirable lands of the dissected peneplain.
The damaging effect of land clearing by bulldozers is a serious
problem on the fragile and infertile soils of the peneplain. In response

Humid Tropics Pro gram **** Indonesia
to rising costs and negative effects of mechanical land clearing, the government currently is leaving the bulk of land clearing to the farmers. The farmers from Java and Bali, however, have no experience in land clearing and regard felling and burning of large trees as dangerous. Even after the land is cleared, the Javanese and Sundanese farmers, accustomed to tilling the soil with a hoe, encounter a root mat too thick for this tool. The indigenous farmer, on the other hand, employs a no-till farming practice and places seeds in holes formed with a pointed stick. The immigrant farmers soon discover that the knowledge and experience that worked so well on the rich volcanic soils of Java and Bali do not necessarily apply in Sitiung. For this reason, they are enthusiastic participants of the TropSoils project. Developing a Collaborative Project:
Transmigration is a major economic and development goal of the
Indonesian government. Sitiung is typical of many transmigration sites in Kalimantan, Sulawesi and Sumatra. It is the intention of the government to transfer soil management and farming systems principles discovered in Sitiung to other transmigration sites in the country., Similarly, TropSoils is also interested in uncovering principles that will enable it to respond to requests for technical assistance for management of tropical soils worldwide.
Sitiung was selected as the research site after detailed discussions between Indonesian agencies and representatives of TropSoils. The site is a representative microcosm of virtually everything that is possible in the humid tropics. To exploit Sitiung's unique biophysical and social setting, the Indonesian government has invested heavily in the program. In addition to the administrative leadership and support from CSR in Bogor, it has committed 15 scientists and technicians to the projects. The senior Indonesian scientist holds a Ph.D. degree; he is assisted by four Bachelor of Science and ten high school level technicians.
In addition to the research and support staff, the Indonesian
government has provided equipment, office facilities and living quarters for its staff. The costs of seeds, fertilizer and chemicals are shared

Humid Tropics Pro gram **** Indonesia
among the collaborating units. The installation, monitoring and harvesting of all TropSoils' experiments are done by Indonesian technicians. An operating budget of $88,000 was allocated to the TropSoils project by the Indonesian government in 1983-84 fiscal year. A similar amount has been budgeted for the current fiscal year.
For its component, the U.S. institution provides three on-site
senior scientists to the project--an agronomist who also serves as team leader, a social scientist who deals with the farming systems component, a soil scientist--and two graduate students. The project is also assisted by university-based scientists from the U.S. institutions who share research responsibilities with scientists in the field.
The major research sites have been chara cterized for soil properties and are continuously monitored for climate. The thorough, ongoing soil and climate inventory of the Sitiung area will be valuable in efforts to transfer the technology to other regions of the humid tropics.
Although research is currently conducted on farmers' fields, the Indonesian government has reserved 220 hectares of forested land for development into a permanent research station. It is located on the peneplain and represents the full topographic range of the region. About a third of the area is relatively flat and the remainder varies in steepness. It offers a range of landscapes for conducting land clearing experiments, which will be a primary initial emphasis.
The Indonesian government has structured its project efforts so
that a fully-staffed research station will be operating in Sitiung when the contribution of AID and the U.S. institution comes to an end. Team Building:
To prepare the U.S. scientists for work in Indonesia and to ensure that the group functions as a cohesive unit, a team building and language training course was conducted in Honolulu between November 7 and December 3, 1982. The group met together daily over a six-week period with approximately half of the time spent in language study.
The team building included an introduction to the concepts and
methods of farming systems research, seminar sessions with people from a

Humid Tropics Program **** Indonesia
variety of subjects on campus who were involved in complementary research and introduction to conditions in Indonesia and the research site. The team also took a Myers-Briggs personality test which was designed to make explicit some of the differences in approach team members might expect from each other and the particular strengths that each team member brought to the group.
Since the farming systems approach requires continuing collaboration and communication, it was essential that participants overcome the kinds of disciplinary barriers that often interfere with team efforts. The team building period served to acquaint the team members with each other and to set a stage that has facilitated continuing collaboration-both between disciplines and across cultural boundaries.
A trip to Yurimaguas, Peru, was made by the U.S. team members to
acquaint them with the valuable and closely related work underway there. This trip also contributed to the team building process, as team members coped together with the challenges of remote areas. Yurimaguas provided a concrete situation in which team members first became alerted to each other's interests and areas of expertise.
The value of the team building has been verified on site in
Sitiung. The necessity to prepare housing for themselves, participate in a soil survey and adjust to the new conditions simultaneous constituted a real test of team cohesian. The staff functioned effectively during that critical period and has continued to do so. Training:
Although Indonesia has quality soil scientists, it lacks the quantity needed to meet the country's development needs. The Center for Soil Research has stated its desire to use the TropSoils project to identify potential candidates for advanced training. The objective is to select from the large staff posted in Sitiung, young people who demonstrate leadership qualities and interest in science. The intent is to select high school graduates for entry into Bachelor of Science program, holders of B.S. degrees for M.S. programs and individuals with M.S. degrees for Ph.D. programs.
As part of this program, the Center for Soil Research is sending two of its staff for Ph.D. training. Mr. D. Santoso, the site coor25

Humid Tropics Pro gram **** Indonesia
dinator and lead Indonesian scientist for the TropSoils project will be matriculating in an Australian University, and Mr. M. Subagio will be going to North Carolina State University. Mr. Santoso will be replaced by Dr. Soleh Sukmana who holds a Ph.D. degree in soil physics.
The number of Indonesian students qualified to enter U.S. institutions is low because most lack proficiency in English. To overcome this problem, the project has hired a teacher who serves as tutor for the expatriate children and language instructor for the Indonesian technicians. Daily contact with the U.S. project staff adds to the language learning process for the Indonesians. Research Strategy:
Early in the project, Indonesian and U.S. scientists and administrators agreed on a research plan for TropSoils activities in Sitiung. The plan is summarized as a flow chart in Figure 1.
It calls for an initial attention to survey and characterization of soil resources and farming systems in the Sitiung area. That effort is to be followed by testing of improved, alternative soil and crop management systems on farmer fields.
Owing to serious soil erosion in parts of Sitiung and the need to prevent worsening of the situation, the plan also called for soil management research to reclaim the abandoned, eroded lands.
Since the method of land clearing has a decided effect on land
quality, the plan also called for a major effort in identifying suitable land clearing methods that could be successfully transferred to other parts of the humid tropics. Work in this area includes the selection,
survey and characterization of 220 hectare land clearing experimental site which the Center for Soil Research intends to use as a permanent research station.
Factors Affecting Progress of the Program: During the planning process for the Soil Management CRSP, a visit
to Indonesia was made by personnel from the Planning Entity and AID/W to discuss with Indonesia administrators and scientists information needs and level of interest in a collaborative research program. The response

Develop Soil/Management Systems Appropriate to the Humid Tropics
Characterization of Soil, Present Farming Systems, and Land Clearing Practices
Action feline Survey
N) Cleared AesInitial Best Estimate UceareAra
Soil/Crop Management Action Restoration Systems____ and___Land_ Action Prevention
EprmnaReerhClearing Practices Experimental Research
Action Farmer Collaboration +and Assessment
Improved Farming Systems (Economically, Environmentally and Socially Workable) Figure 1. Flow chart of activities for the Indonesia TropSoils Project.

Humid Tropics Program **** Indonesia
on both issues was positive and enthusiastic. A "Letter of Intention" was signed in February, 1980, by representatives of the Soils Research Institute, the Central Research Institute of Agriculture and Bogor University expressing concurrence with proposed research activities, work location and institutional coordination.
The subgrant from the Management Entity to the University of Hawaii and North Carolina State University to actually implement the program was executed in January, 1982. In March, 1982, representatives from the two institutions and the Management Entity visited Indonesia to discuss formalization of the program and develop the appropriate legal documents. There was anticipation that all necessary approvals would be obtained by August, 1982, and, thus, recruiting of personnel and other plans were initiated by both universities. Difficulties occurred in obtaining the required actions. Considerable uncertainty on timing occurred and resulted in a number of go-stop actions on personnel posting. Final approval was obtained in June, 1983; senior scientists arrived in the country within a few weeks thereafter.
In spite of these developments, with the exception of a six-month period between January and June, 1983, most of the time was used in effective start-up activities.
Major constraints to on-site activities can be linked to isolation of the project area. Isolation results in real and perceived constraints in security, communication, housing, power supply and health.
Security. Concern for the security and safety of team members was a major factor in Indonesian administration's reluctance to permit the team to enter Sitiung prior to formal approval of the Agreement. The concern for security exists to this day, and team members are required to sign in at the District Attorney's office when they leave and return to the area. The District Attorney's office would be in serious trouble with the Central Government if the whereabouts of a team member, injured outside the Sitiung area, were not known to the Office. The security problem is perceived differently by the team members and the district officials. The team feels secure in the village and fields, but the officials are fearful that team members will be harmed by unfriendly people from outside.

Humid Tropics Program **** Indonesia
Housing. Since original housing reserved for team members was
located in Sumani, 2.5 hours from the research area, the team arranged to rent and upgrade homes owned by local settlers. The resulting savings in time and fuel more than compensates for the inconvenience of locating adequate housing.
Communication. There is no direct telecommunication link with the team. The only telephone in Sitiung is located a few kilometers from the team's residences. Cables are received at the post office in Sitiung. A battery operated radio communication system is being developed to link team members with each other and to Padang and Bogor. In cooperation with the IADS group in Sukarami, direct telephone, cables and telex linkages have been established to a one person office in Padang. Messages received in Padang are either called or mailed into Sitiung by the office staff.
Power Supply. There is no municipal electricity or water supply. Except for the kerosene refrigerators and gas stoves, household appliances and lighting are provided by diesel generators. The type of equipment purchased by the project to support a soil characterization laboratory is constrained by lack of steady and reliable power supply.
Health. Two members have been afflicted with unknown ailments. One member of the University-based staff who visited Sitiung in March, 1984, was suspected of having contracted Dengue fever. Although proper precaution can be maintained at home, it is difficult for team members to reject food and drink offered by cooperating farmers. One team member has been ill for over six months and returned to Hawaii for medical help. The doctors suspected a virus and recommended a long rest, but the patient returned to Sitiung after only a short stay.
Although isolation and the constraints that go with it appear formidable, team members treat them as minor inconveniences and continue their work with remarkable enthusiasm. They recognize the reality of working in a developing country is to deal with these kinds of constraints. The connection between soil management and quality of life is not lost because the constraints that affect project activities are the same constraints that face everyone in Sitiung.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Harijogjo, CSR
Bambang Mahmudi, CSR
Yayat Hidayat, CSR
Anggana, CSR
Date Research Initiated:
May, 1984
Rationale for Investigation:
220 hectare, forested site has been selected to serve as a permanent research station. Instead of simply clearing the land, the land clearing effort will become part of a study to assess land clearing methods (see project on land clearing). A detailed soil and plant inventory of the site will enable the land clearing project to locate suitable research plots and to plan a long term research strategy for
the site.
To prepare a detailed soil map (1:5,000) of the TropSoils land clearing research site. Experimental Approach:
This research is conducted in the Sitiung transmigration area of
West Sumatra, Indonesia. Transects through the site will be prepared and soil and vegetation type recorded. Soil samples will analyzed in Bogor and the field and laboratory data will be used to classify the soil
according to the Indonesian system of soil classification and Soil taxonomy. A large scale map (1:5000) will be prepared.

Humid Tropics Pro gram **** Indonesia
State of Progress:
A preliminary report has been reported. Constraints of Progress:
See Overview
Applicability of Results to TropSoils Goal:
Detailed characterization of the land clearing site is necessary to render the research results useful elsewhere in the humid tropics. This site will be one of three "benchmark" land clearing sites planned under a cooperative effort with IBSTRAM. Training Component:
Soil survey and classification is a continuing training aspect of the TropSoils project. Its main training will be in the use and adoption of Soil Taxonomy as a common international language for soil management technology transfer.

Humid Tropics Pro gram **** Indonesia
Leadership Personnel:
Bruce Trangmar, UN
Goro Uehara, UN
Djoko Santoso, CSR
Date Research Initiated:
July, 1982
Rationale for Investigation:
A typical farm in the transmigration area varies in productivity from bare spots to green strips. The bare spots are sterile subsoil exposed by the bulldozer and the green strips correspond to the ash lines of burnt trees. This type of variability is a problem for the farmer and researcher. In farmers' fields, the bare spots are the first to erode. Bare spots produce nothing and, therefore, are neglected. Erosion feeds on neglect, and the land is eventually abandoned.
If a farmer applies lime and fertilizers uniformly over a field, too little is applied to the bare spot and an overdose is given to the green strips. But because the Sitiung farmers spread farm chemicals by hand, they are in a position to vary the application rate according to need. The farmer must learn or be taught to recognize various forms of variability that causes inefficient use of scarce resource.
The researcher needs variability but wants a field with minimum
natural variability so that the effect of the treatment variable will be clearly expressed. In Sitiung the range of natural soil variability is almost always as large as the imposed treatment range. The aluminum saturation of the surface layer of a freshly cleared and burnt forest ranges from 0 to 90%. This is the same range a researcher would impose

Humid Tropics Program **** Indonesia
in a liming experiment to assess the effect of aluminum toxicity on crop performance.
Fortunately, a new technique for dealing with soil variability is now available to soil scientists. This technique, called geostatistics, enables soil scientists to estimate the value of a soil property at unsampled locations from an analysis of neighboring samples. Objectives:
A. To assess soil variability in a small field and measure its
influence on crop performance.
B. To identify the soil, chemical and physical properties most
responsible for yield variability.
C. To test the occurrence of structure in the variance of soil,
chemical and physical properties by means of geostatistical
D. To establish the relationship between soil spatial variability and spatial variability in crop performance. Experimental Approach:
This research is conducted in the Sitiung transmigration area of West Sumatra, Indonesia. The theory of regionalized variables has been developed by mining engineers to extract the maximum amount of information from a minimum amount of sampling data. The sample data are used to generate a semi-variogram which shows the existence or non-existence
of spatial relationships among neighboring samples. If such relationships exist, the information in the semivariogram can be used to estimate values of soil properties in unsampled locations. This additional information can be used to prepare more accurate soil maps so that problem areas can be more precisely pinpointed.
The theory of regionalized variables will be applied to two sets of data. The first set consists of 88 soil profiles analyzed by the Center for Soil Research. The samples were collected from the 100,000 hectare transmigration site. The second set of data was collected from a plot in

Humid Tropics Program **** Indonesia
a farmer's field by the TropSoils team. The samples were analyzed by the soil characterization laboratory of the Center for Soil Research in Bogor.
State of Progress:
The results show that natural soil variability can be exploited to answer key agronomic questions. Examples are available to illustrate the variability of aluminum saturation in an experimental plot and its corresponding effect on the rice crop. The data show that rice yields were significantly higher on burn sites than on exposed subsoil. A more detailed analysis shows that difference in organic matter, phosphorus and nitrogen accounted for less of the yield increase than differences in aluminum saturation, calcium, magnesium, potassium or zinc. Thus, geostatistics enabled project scientists not only to map spatial variability of soil properties to extract from the data set agronomic information that relates soil productivity to soil properties.
The same technique has been employed to map agronomically important soil properties for the entire 100,000 hectares research area. The spatial variability of the amount of lime needed to correct aluminum toxicity in the Sitiung area illustrates how geostatistics can be used to identify soil constraints.
Constraints to Progress:
See Overview
Applicability of Results to TropSoils Goal:
Soil variability which cannot be accounted for by traditional soil survey methods can be dealt with by application of the theory of regionalized variables. This type of soil variability occurs in every agroecological zone and can be accommodated by existing computer software. These softwares are available in a dissertation that has been distributed to all SM-CRSP projects.

Humid Tropics Program **** Indonesia
Training Component:
Use of these techniques requires access to mainframe computers. It is the intent of this project to train an Indonesian graduate student in this technique when he arrives in Hawaii. Two Master of Science candidates from Indonesia have been proposed for matriculation in the University of Hawaii.

Humid Tropics Program **** Indonesia
Leadership Personnel:
Karim McKarim, NCSU
D. Keith Cassel, NCSU
John J. Nicholaides, III, NCSU Date Research Initiated:
September, 1983
Rationale for Investigation:
Many transmigration areas in Indonesia have been hastily planned and administered. The consequence is degraded landscapes that cannot produce sufficient food crops. Sometimes even cassava will not grow. These areas were improperly cleared and improperly protected. Thus, topsoil and organic matter have been pushed and/or eroded away. A major challenge facing the governments agencies is how to reclaim these barren lands. This research is an initial step in identifying what must be done, both chemically and physically to rectify the situation. This
process, of course, is partially site and soil specific, but what will work in one area will give clues to what may work in others. Objectives:
A. To identify soil management practices that will improve chemical and
physical properties of degraded soil .
B. To study the effect of lime and fertilizer on crop performance at
low and high critical levels for lime and nutrients.
C. To study the effect of various tillage methods on soil physical properties and crop yields.
D. To study the effect of green manures on crop yields and soil chemical and physical properties.

Humid Tropics Program **** Indonesia
Experimental Approach:
The research is conducted in the Sitiung transmigration area of West Sumatra, Indonesia. A two pronged approach was developed to reclaim the
damaged soils by modifying both chemical and physical properties. To alleviate suspected compaction and infiltration problems, the following treatments were installed: (1) hoeing to 15 cm, (2) hoeing plus applying mulch to the soil surface, (3) hoeing with incorporation of
organic matter, (4) turning the soil by spading fork to 30 cm depth, (5) spading in alternate 40 cm strips to the 30 cm depth and (6) rototilling to 15 cm. Three subplot treatments of the following soil fertility levels were included on each main plot: FO, no lime and fertilizer; F1, application of lime, N, P, K, Mg, S, Cu and Zn at rates of 1500, 120, 40, 72, 70, 93, 1 and 4.5 kg/ha, respectively; and F2, the application
of the above nutrients at rates of 6840, 150, 572, 144, 140, 187, 2 and
6 kg/ha, respectively.
The experiment is a randomized complete block design with a splitplot arrangement of treatments with four replications. The study site will be continuously cropped for two years. Soil physical and chemical properties will be monitored periodically throughout the study. State of Progress:
The first two crops (rice and soybeans) of the first cycle have been harvested. The rice crop responded dramatically to lime and fertilizers. The unamended soil was nearly barren. Application of low rates of lime and fertilizers resulted in moderate yields. The high rates increased yields on the average by 17%. Green manuring, incorporated 12.5 T/ha of fresh calapagonium sp, showed a very
substantial increase in rice yields at all three levels of fertility compared with the nonmanured plots. Low rates of lime and fertilizers with green manure produced considerably more than high rates without green manure. There was little difference among tillage methods, except for the strip tillage which was not as effective as the others.
The soybean crop, which is much more sensitive to low P and high Al responded quite differently than the rice. Again, no amendments meant

Humid Tropics Program **** Indonesia
no yield, and even the low rates produced rather poorly. The high rates, however, yielded very well, on the order of three times the low rates. With this crop the green manure was not reapplied and there was no residual effect on the bean yield. Deep spading produced a marked
positive effect on yields at low rates, but gave a negative effect at high rates of lime and fertilizers.
Deep spading had the most favorable effect on compaction, bulk density and water infiltration. Organic matter additions generally improved the soil physical condition compared to simple hoeing but not as dramatically as the deep tillage.
Soil chemical analysis of the fertility main plots after the first crop of rice showed that the low rates of lime and fertilizer increased bases and P, while reducing Al to about 50 per cent of ECEC. The high
rates completely eliminated exchangeable Al and brought available P to a very high level. The dramatic soybean response reflects this favorable nutrient status.
Constraints to Progress:
Equipment for measuring soil physical properties is needed but is
currently unavailable in Sitiung.
Applicability of Results to TropSoils Goal:
The Center for Soil Research, as the national soil institute of the Ministry of Agriculture, will be (and is) called upon to recommend practices that will prevent soil degradation and reclaim land that is eroded. This research provides information that will assist it in developing a strategy for soil conservation and reclamation. This project also helps the U.S. institutions by providing information on the basic principles involved in wfiat amounts to "rebuilding" a soil. We see
firsthand the soil forming processes at work in a dynamic and dramatic situation, unlike perhaps any field condition that can be found in the U.S.

Humid Tropics Program **** Indonesia
Training Component:
Karim Makarim is an Indonesian who is a candidate for the Ph.D.
degree at NCSU. This research will be used, in part, for his doctoral thesis. He is gaining valuable research training and will return to Indonesia.

Humid Tropics Program **** Indonesia
Leadership Personnel:
Carl Evensen, UH Russell Yost, UH
John Thompson, UH
Date Research Initiated:
October 1984
Rationale for Investigation:
Organic matter is an important soil component, especially in highly leached and weathered tropical soils. Soil organic matter is associated with increased cation exchange capacity, improved soil structure, and is a major source of many plant nutrients. The soils in much of the Sitiung area developed under rain forests which, as a rule, maintain a tight nutrient cycle between litter decay, root uptake, and plant growth. The clearing of rain forests breaks this cycle and often leads to soil degradation through increased organic matter decomposition, leaching of nutrients, soil compaction, and erosion.
Sustainable, low-input farming systems in Sitiung Indonesia must provide for maintenance of soil organic matter. This project is designed to compare farming system technologies in terms of inputs and persistence of organic matter and their feasibility and attractiveness to subsistence level farmers. An assumption to be tested is that a permanent soil cover of vegetation or mulch will best protect the soil and create a stable equilibrium between organic matter and decomposition. Hedges of legume trees cut frequently to provide mulch and legume cover crops providing vegetative soil cover will be assessed in intercropping with food crops.

Humid Tropics Program **** Indonesia
A. To evaluate the importance of soil organic matter in agricultural production in Indonesia, determine its role in
nutrient cycling and effect on physical and biological processes in the soil.
B. To identify optimum farming systems for the management of
organic materials.
C. To identify the major groups of soil organisms involved in
organic matter decomposition in Sitiung and to study the
effects of management factors on their numbers and activities. Experimental Approach:
The field investigations will be conducted in the Sitiung area of West Sumatra, Indonesia. The research will consist of two major phases. The first phase will involve tree and cover crop species selection, determination of lime and fertilizer requirements, and observation of interactions between food crops and organic matter producing plants. In the second phase, promising tree or cover crop, fertilizers, and food crop combinations will be tested in farmer managed trials to determine farmer acceptance. Since this research will be conducted as part of a farming systems project, the trials will be open to adaptation as the research needs and farmer situation in Sitiung are understood better. State of Progress:
Project just initiated.
See Overview.
Applicability of Results to TropSoils Goal: Organic matter is expected to be an important soil component in other transmigration sites with highly leached and weathered soils. Findings from these studies in Sitiung should contribute to better management of soil organic matter in similar sites. Of particular interest

Humid Tropics Pro gram **** Indonesia
will be the identification of useful tree and cover crop species and determination of their interactions with intercropped food crops. Training Component:
Management of organic matter will be part of the farming system and will be incorporated in training aspects of farming systems research. The data arising from this project will also be used in training for nutrient dynamics and nitrogen transformation in crop modeling.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Dan Gill, NCSU
Eugene J. Kamprath, NCSU
John J. Nicholaides, III, NCSU
Mike K. Wade, NCSU
Date Research Initiated:
June, 1984
Rationale for Investigation:
The soils in Central Sumatra are quite acidic with aluminum saturations of 80 to 90 percent and have very low levels of potassium and other bases. Sustained crop growth on these soils will require additions of fertilizer potassium. The relatively high rainfall in the area enhances the potential for leaching losses of potassium. Such losses can be reduced to some extent by neutralization of aluminum which increases the accessibility of exchange sites for potassium. Relatively little information is available on the potassium requirements for sustained.production of cropping systems in Sumatra. Objectives:
A. To determine the effect of K rates at three levels of base saturation on the yield and uptake of K and basic cations of upland crops
in the Sitiung area of West Sumatra.
B. To determine the effect of different levels of base saturation on
the retention and movement of K applied at various rates. Experimental Approach:
The research will be conducted in the Sitiung transmigration area of West Sumatra.

Humid Tropics Program **** Indonesia
Two cropping systems will be used: (1) corn, soybeans and mung
beans and (2) rice, peanuts and cowpeas. Lime rates will be equivalent to 500 kg CaCO 3/ha and rates that will give 50 and 0 percent aluminum saturation. The rates of K applied, as KCl, to each crop will be 0, 20, 40, 80, 120 and 240 kg K/ha.
Plant leaf samples will be taken at flowering to determine the K, Ca, Mg and N concentration. Total plant content of these nutrients in the harvested grain and yield of grain will be determined at maturity.
Soil samples will be taken at 15 cm intervals to a depth of 60 cm after each crop to determine the movement of K and the basic cations.
Laboratory studies will be done on the quantity-intensity relationships of K in these soils.
State of Progress:
The plot area is being prepared and lime treatments are now being
applied. The first crops will be planted in September, 1984. Constraints to Progress:
The lack of equipment for conducting field experiments such as
drying facilities, plant choppers and small hand operated tractors could hinder the quantity and quality of research.
Applicability of Results to TropSoils Goal:
Potassium availability and reserves in the weathered soils of the humid tropics are generally very low. Addition of fertilizer K is the
only means for supplying the needed K. These studies will provide needed information as to the amounts of K required and the soil K levels that will result from given rates of fertilizer K.
The results of this study will be important for developing a soil
management program which provides for the efficient use of K fertilizers in highly weathered soils.
Training Component:
Mr. Gill is a candidate for the Ph.D. degree at North Carolina State University and is interested in pursuing a career in international agricultural research. This research will be used for his doctoral thesis.

Humid Tropics Program **** Indonesia
Leadership Personnel:
Mike K. Wade, NCSU Dioko Santoso, CSR
Date Research Initiated:
January, 1984
Rationale for Investigation:
Current transmigration policy is to supply triple superphosphate (TSP) at the rate of 100 kg/ha twice per year, irrespective of soil type, soil analysis, or crop grown. Most soils in the Sitiung area are
extremely deficient in available P and doses of 100 kg TSP (20 kg P) per ha are not likely to be adequate. The need for crop response and soil test correlation data is great for all areas of Indonesia, especially for upland crops. This research will aid in the development of a soil
test correlation data bank that can assist scientists and policy makers in making scientifically-based decisions for P fertilizer rates and
It has been observed that farmers generally have poor P fertilizer management practices regarding method and timing of application. The current recommendation is that P fertilizer be banded, a very labor intensive operation that farmers are reluctant to follow. A more labor efficient application method that can achieve equal or better crop production is surely needed; especially for these farmers who do all field operations with a few simple hand tools. Objectives:
A. To determine optimum rates of TSP fertilizer on a newly cleared
clay loam Ultisol
B. To determine cost:benefit ratio of various methods of applying TSP
fertil izer
C. To study long-term effects of various P management schemes 47

Humid Tropics Program **** Indonesia
D. To determine critical P soil test values for rice, peanut and cowpea Experimental Approach:
The research is conducted in the Sitiung transmigration area of West Sumatra.
On-farm, researcher-managed, low input trials are used. Three replications, of the following treatments were applied.
* No P
* 20 kg P/ha broadcast and incorporated o 40 kg P/ha broadcast and incorporated 80 kg P/ha broadcast and incorporated
* 20 kg P/ha banded 40 kg P/ha banded 80 kg P/ha banded
o 20 kg P/ha placed in dibble hole 5 cm to the side of the seed
* 20 kg P/ha placed in hole with seed
* 20 kg P/ha banded inbetween every other row
o 80 kg P/ha broadcast and incorporated plus 2T lime, 50 kg K and
100 kg MgSO4.2H20 per ha
The source of P was triple superphosphate
It is hypothesized that if Al tolerant crops are grown (prior to treatment the soil had 40% Al saturation), then P would be the major nutrient limitation. Therefore for this low input approach only P fertilizer has been applied, except for one 11 which was given lime and all macronutrients to test the hypothesis that available P indeed is the primary limiting soil factor for crop production. State of Progress:
There was a marked response to P, as yields increased from about
zero to nearly a ton/ha at the highest rate. There were no significant yield differences between the banded and broadcast treatments. Three
additional methods of applying the fertilizer were tested at the 20 kg rate. Again there was no significant difference among the methods.
Yield levels were low, even with high inputs. The response curve did not reach plateau yields, indicating P may still be limiting. In

Humid Tropics Program **** Indonesia
addition, the variety (locally obtained from new transmigrant farmers) did not appear to be well suited to this region. At flowering and early pod set stages, the crop suffered from both leaf and wilt diseases.
The current crop is mungbean. Because peanut did not show a plateau yield, all P treatments were reapplied prior to planting. Unlike peanut, this crop is exhibiting considerable within-plot variability. From other experiments we now know that mungbean is quite Al sensitive, and it is suspected that soil acidity is responsible for the variabi1 i ty.
Soil test levels of P increased with P rates. The relationship
indicates little initial P fixation and that predication of available P should not be difficult.
Constraints to Progress:
This experiment requires rapid turnaround on soil and plant analyses
to better predict or evaluate the rates of lime and fertilizer should be used on subsequent crops. Growing three crops a year with virtually no break puts great and constant demand on laboratory facilities. We have a small functional lab (approx. 20 samples per day of pH, bases and P) for soils. This must be upgraded if it is going to meet the local demand. Also equipment for doing plant samples is very much needed.
The peanut crop was strongly affected by an apparent poor choice in variety selection. There is a serious shortage of developed or improved varieties of all the upland crops. Sometimes seed is even difficult to
find in Bogor. It appears that the best solution to this problem is for us to have our own seed storage facilities. High oil content seeds such as soybeans and peanuts will maintain good viability in ambient conditions for only 2-3 months.
The peanuts did not nodulate. We had no inoculant at planting time for peanuts. Only soybean inoculant is regularly produced in-country.
No other seed or forage legume inoculant is available. Applicability of Results to TropSoils Goal: This research can help build a nationwide P soil test correlation data bank for making site-specific P fertilizer recommendations.

Humid Tropics Program **** Indonesia
Current national production programs usually call for a blanket P rate for all growers, ignoring soil test levels and soil type. Such a practice is inefficient as some soils need little or no P fertilizer, while others need large doses. Accurate methods to predict optimum rates of P
will encourage farmers to use fertilizers and to use them effectively.
More labor efficient methods of P application are needed for these
non-mechanized farmers. The experiment showed that the currently recommended banding method required 275 person-hours/ha of labor, while broadcasting took only 40. Nearly all Javanese farmers hoe their land between crops anyway, so incorporation is not additional labor. Even if no tillage is done, the method of adding fertilizer in the seed hole only required 80 person-hours/ha and was equally effective as banding in improving yields (at the 20 kg P/ha rate).
Training Component:
This project is helping to train CSR personnel involved in methods
of determining fertilizer rates. Pre-treatment lab incubation studies, post-harvest soil analyses and correlation with treatment and crop yield exhibit the process involved in choosing and evaluating P fertilizer rates. The foundation for soil test correlation is being laid.
General use of the laboratory in P determination and incubations is training the technicians in useful techniques that can be expanded and carried-on in future years.

Humid Tropics Program **** Indonesia
Leadership Personnel:
Mike K. Wade, NCSU Agus Sophian, CSR
Date Research Initiated:
September, 1983
Rationale for Investigation:
The Indonesian government has made a new commitment to developing
their "palawija" crops, i.e., food crops other than paddy rice. To provide information for this objective, a large scale liming program was
initiated in 1983. At present the most common, if not the only, available source of lime in many areas is burned lime. It has certain advantages as liming material: (1) It can be produced as a home industry with little or no capital investment (an earthen or dug kiln and firewood), (2) it is very reactive and therefore relatively low doses are adequate to reduce toxic level of Al and (3) it is an established industry in many places (although its end-use has not been for agricultural purposes). Its main disadvantage may be in its short residual effect. The other possible or likely source of lime is ground limestone, either calcitic or dolomitic. The advantages of ground lime are: (1) it has a longer residual effect and (2) it ultimately may be
cheaper if produced on a large scale. Its principle disadvantage is that it takes a fairly significant capital investment to purchase the grinding equipment. Also, quality control may be more difficult as just about any rock can be ground and not visually distinguished from lime. only two carbonaceous materials can be heated and subsequently slaked by hydrolysis.
Farmers use a variety of means to till their soil, and due to the differing reactivity of these lime materials, it would be beneficial to know if one or some methods were superior to others in bringing about

Humid Tropics Program **** Indonesia
incorporation and subsequent reaction of lime in the soil. Objectives:
A. To compare burned lime and ground limestone for effectiveness in
neutralizing soil acidity and improving crop production
B. To compare the residual effect of the two liming materials C. To study the effect of tillage on effectiveness of lime D. To study the effect of tillage on crop growth and soil physical
Experimental Approach:
On-farm, researcher-managed trials are used in the Sitiung transmigration area of West Sumatra.
The treatments are:
A. Lime Source
1. none
2. burned lime
3. ground limestone
B. Application method
1. surface broadcast no till
2. hoe incorporation 15 cm 3. cattle drawn plow 15 cm
4. rototiller 15 cm
5. deep spading 30 cm
The burned lime was applied at 1.2 x exch. Al (1.33 t/ha), while the limestone, due to low solubility, was applied at 1.5 x exch. Al (2.25 t/ha). For the deep spading treatment the lime was applied at 2x the above rates as incorporation was to 2x the depth of the other treatments. In the no-till plots, 1/3 the lime rate was applied before planting of each crop. Three crops were grown during the season (rice,
peanuts and mungbeans) and thus the same annual amount of lime was used. State of Progress:
Rice did not respond to lime or tillage treatments. Initial soil reaction was pH = 4.8 with 55% Al saturation. Upland rice is expected in most cases, to tolerate such levels of acidity.
Peanuts are not as tolerant to soil acidity as rice but this crop too showed little response to lime or tillage. Although there were no

Humid Tropics Program **** Indonesia
significant differences (P > .05) the limed plots generally gave higher yields than the no-lime treatment. The deep tillage (spading) treatment tended to give higher yields but again, no significant differences among tillage treatments were obtained. Yield level of all plots was quite low. The rice straw from the previous crop was used to mulch the peanuts. However, rainfall during the 85 day.growing period of the peanuts was almost 1000 mm. The combination of high rainfall, mulch, and drainage from higher areas kept the soil almost constantly saturated. The peanuts did not grow well, looking stunted and yellow from apparent lack of oxygen throughout the growing cycle. This may well have suppressed a lime response. The deep tillage treatment may have permitted better drainage and explains the somewhat higher yields
under thi s treatment.
The final crop of the 1983/84 season was mungbeans. This crop is very sensitive to Al toxicity and showed a strong response to lime. Harvest is currently under way, so no data is available. However, most no lime plots are bare or at best have very stunted and low- or nonproducing plants. Visually the no-till treatments are the poorest and the deep tillage are the best. Subsequent soil chemical and physical property measurements should help explain why these responses occurred.
Soil samples are being taken at 0-5, 5-10, 10-15 and 15-30 cm depth every 6 months. The pending analyses will allow monitoring of lime
reaction and distribution of source by tillage. Constraints to Progress:
The primary constraint of this experiment is laboratory facilities.
The sampling with depth produces 180 samples per crop. Although this is not so many by itself, coupled with the other experiments a backlog of samples are developing. on-site facilities are capable of analyzing a maximum of 20 samples per day, when there are no chemical or personnel shortages. Frequent interruptions due to other responsibilities prevent the technician from doing more than about 50-60 samples per week. Also lack of plant sample preparation and analysis creates a serious problem
in having available plant tissue analysis data in-hand.

Humid Tropics Program **** Indonesia
Applicability of Results to TrapSoils Goal:
This research is aimed at testing the effectiveness of the two most common sources of lime in Indonesia. It will also test the importance of tillage or method of incorporation on lime reaction. Soil analysis will show the distribution and movement of lime within the soil as affected by source and tillage. Liming agricultural land is a relatively new practice in Indonesia. This study should provide information for deciding on the kind of lime processing that should be developed. The general characteristics of both types are known, but field data are needed to quantify the relative reactivity and residual effect of the two sources. Burned lime may not be an economically feasible source in the U.S. or other developed countries, but can be rendered potentially competitive in Indonesia as a home or smiall-scale industry. This has obvious benefits for a country with high unemployment, and low investment funds.
Training Component:
This experiment provides the opportunity for the personnel to gain both field and lab experience with these two major sources of lime. it also allows them to experience and consider alternative tillage operations, which is a very important question for the farmers. This is in
opposition to conventional research station experience where tillage is usually done by hand or wheel tractor.

Humid Tropics Program **** Indonesia
Leadership Personnel:
Mike K. Wade, NCSU
Eugene J. Kamprath, NCSU
Djoko Santoso, CSR
Date Research Initiated:
September, 1983
Rationale for Investigation:
Soils in the Sitiung area, as well as many in the outer islands of Indonesia in general, are quite acid. Most of these unamended soils have pH values less than 4.5 and exchangeable acidity greater than 2.5 meq/100 ml soil, with a resulting base saturation greater than 60% (many reach 80-90%). To successfully grow such food crops as peanuts, soybeans, mungbeans, and corn, liming is necessary. Even acid tolerant
crops such as cassava and upland rice might be expected to respond to lime on the more acid soils. Although much is understood about the
science of liming soils for agricultural purposes, on-site verification of rates and establishing critical soil test acidity levels for the major crops is desirable. Beyond determining critical levels and initial rates for a given soil and crop, it is necessary to study the residual effect of applied lime under the local rainfall and soil conditions to determine how much and how often lime will have to be reapplied in
order to maintain a desired level of base saturation. Objectives:
A. To determine the critical level of soil acidity parameter(s) for
optimum production of upland rice, soybeans, and mungbeans
B. To determine a method of predicting lime rates necessary to achieve a specified level of soil acidity
C. To determine the annual lime application rate required to maintain a

Humid Tropics Program **** Indonesia
specified level of soil acidity
D. To monitor the residual effect of various rates of one-time lime
applications on a rotation of annual food crops Experimental Approach:
The general procedure involves on-farm, researcher-managed studies in the Sitiung transmigration area of West Sumatra. It is designed for
a minimum duration of three years.
The experiment is 2-factor factorial with 4 replications. The treatments are:
Lime rates
1. u 7Ta
2. 1/2 T/ha (3/8 x exch Al)
3. 1 T/ha (3/4 x exch Al) 4. 2 T/ha (1112 x exch Al)
5. 4 T/ha (3 x exch Al)
Lime maintenance
1. Residual only
2. Annual application to maintain first crop levels of Al
The source of lime was burned limestone.
All plots receive a blanket application of N, P, K, Mg and S at rates estimated to eliminate deficiencies. State of Progress:
Upland rice, the first crop grown, showed little yield increase due to lime. Unlimed plots averaged 1.6 T/ha while all limed plots yielded
2.0 T/ha. However, this difference was not statistically significant at 5% level of probability and there was poor correlation between yield and soil acidity parameters such as percent aluminum saturation, pH or Ca + Mg. This is not surprising, as upland rice is known to tolerate acid soils. The unlimed plots had an average pH of 4.6, with acid saturation of 66%.
Soybeans, the second crop, showed a marked response to lime, as it is much more sensitive to acid soil.
Mungbeans, the third crop, has been grown and recently harvested.
Although this crop responded sharply to lime, it grew poorly and yields

Humid Tropics Program **** Indonesia
are low even at high rates of lime. Marginal rainfall and leaf disease seemed to be contributing factors.
The residual effect of the lime treatment when burned lime is the form applied appears to be relatively short. Seven months after its application there was a decrease in the available soil Ca and Mg and an increase in the exchangeable Al and Al saturation. The residual effect of calcitic and dolomitic lime will probably be longer because of more coarse materials being present compared with the powdered form of the burned lime. Thus, in areas of high rainfall and potential for leaching of Ca and Mg ions, the limestone should probably contain some coarser fractions in addition to some fine particles to provide a longer residual effect.
The soil analysis for the final crop in the rotation, mungbeans, are
pending. Based on those analyses, estimates will be made for maintenance doses to re-achieve first crop acidity levels in the designated
maintenance plots. Then the second year's rotation will begin with a comparison of the lime response on residual vs. maintained rates. The rotation will be the same as the first year, i.e. rice, soybeans, mungbeans.
Constraints to Progress:
In addition to the need for improved soil analytical facilities and the creation of plant analytical facilities, more and better information on crop protection is needed. Disease in the rice and insects in the soybeans caused reduced yields. Low yields were achieved in mungbean
without any clear cut problem being positively identified.
As with varieties, crop protection for upland crops is under the
jurisdiction of the Food Crops Research Institute. But it is, as is our
collaborating CSR, relatively new in working in these upland, acid and infertile soils of the humid tropics. Much progress has been made in the central producing areas of Java, but materials and methods used there don't necessarily work here. We have good contact and rapport with the West Sumateran Food Crops people from Sukarami, but do not have tried and proven management practices yet developed.

Humid Tropics Program **** Indonesia
Applicability of Results to TropSoils Goal:
This experiment is providing information to begin building a database for correlating soil acidity parameters with crop growth. Eventually, we hope to define critical values for the major crops of the area so that more precise recommendations for correcting soil acidity can be passed on to government agencies.
The red-yellow podzolics (Ultisols and Oxisols) are prevalent on many transmigration sites throughout Indonesia. It is expected that these results, i.e. critical values and lime rate predictions, can be extended to other areas of similar soils. Once sufficient data are collected, CSR can develop a lime recommendation scheme for growers through national commodity production programs. Our results can be compared with findings from other acid soil regions to establish or confirm acid soil/lime/crop management principles. Training Component:
The researchers involved in the experiment can learn about the dynamics of lime application; its effect on soil parameters and its reaction and movement in the soil. Once again the importance of simple lab procedures in quantifying and predicting lime requirements are being
learned by the researchers and technicians involved. Also, this serves as another example of how to define and establish crop yield and soil test correlations as well as critical levels. As a service to their farmers and government agencies, this is probably one of the most lacking aspects of CSR's work. If the value of such information can be seen, then expansion and development of a nationwide program can begin.

Humid Tropics Program **** Indonesia
Leadership Personnel
John Thompson, UH
D. S. Gunawan, CSR
Date Research Initiated:
January, 1984
Rationale for Investigation:
Ground cover is a logical and sound way to protect cleared land, reclaim eroded soil, provide feed for livestock and serve as green manure for the resource-poor farmers of Sitiung and the humid tropics. Pasture grass and legume species that perform well and serve multiple uses can become permanent and inexpensive components of the farming
system. This project is designed to match the environmental requirements of pasture grass and legumes to (1) the environmental characteristics of the land and (2) the resource characteristics and preference of the farmer.
A. To evaluate germplasm in a range of environments representative
of the humid tropics.
B. To identify suitable cover crops for eroded lands. Experimental Approach:
This research is conducted in the Sitiung transmigration area of
West Sumatra, Indonesia. CIAT has a collection of pasture grass and legume germplasm ready for testing. These cultivars will be tested in Sitiung under minimum input situations. Cultivars will be selected for vigor, growth rate, pest resistance, palatability, productivity and survivabil ity.

Humid Tropics Program **** Indonesia
State of Progress:
These experiments were previously arranged through contacts with CSR and CIAT personnel. Jose Toledo, Coordinator of the forage and pasture project for CIAT, provided seedstocks of several species of grasses and legumes.
James Spain, pasture agronomist from CIAT, has observed the growth and was much impressed with the performance of several of the species being evaluated. His wife, who is a microbiologist, collected several
samples for classification of mycorrhizae. This adds to the information on mycorrhizae in the Sitiung area which Russell Yost and his graduate students in Hawaii are compiling.
About two months ago the team also had the opportunity to discuss
the native legumes of Sumatra with Ranier Schultz-Kraft who is the legume germplasm collector for CIAT. Schultz-Kraft was much impressed with the broad range of forage legumes in Sumatra and promised to return for a detailed collection and classification of local legumes. The team will be working closely with him and may utilize some of these species in the legume evaluation in Sitiung.
One grass and four legume species show promise as cover crops to
reclaim eroded land and at the same time serve as animal feed.
The grass species which performs very well is Brachiaria dietyoncura. The legumes are:
" Aeschinamene histrix This is a very vigorous legume with quick
recovery. It should be compatible with a vigorous grass such as
Brachi aria.
" Centrosema maerocorpem; C. Pubescence and C. sp. Both appear to
be well-adapted and vigorous.
* Pueroria phaseolides In spite of some insect problems, this
species now appears to be well-established.
* Desmodium ovalifolium This legume was slow to establish but it
now appears to be well-established. Some of the Stylosanthes and
Zornia species also appear to be well-adapted to the Sitiung

Humid Tropics Pro gram **** Indonesia
Constraints to Progress:
See Overview
Applicability of Results to TropSoils Goal:
Carol Colfer has discovered from her time allocation studies that
Sitiung families require considerable time to cut forage from roadside and abandoned fields for their farm animals. The quality of the feed is very poor and the CIAT collection has the potential to significantly reduce the time needed to harvest forage and measurably improve quality
of the harvested feed.
This work also has relevance to the small ruminant CRSP which is active in the country. It is highly likely that future tests of CIAT's pasture grass and legume collection will uncover many more high performance species that will serve as ground cover to protect soil from erosion, green manure, forage for farm animals and a land reclamation cover crop.
Training Component:
Pasture grasses and legumes serve as training elements in erosion control, green manuring, biological nitrogen fixation, ecology of mycorrhizae, animal feeding, land reclamation and ground cover for
rubber plantings. The varietal testing also serves to illustrate genotype-environment interactions and the principles of matching crop requirements to land characteristics. The multiple uses of grasses and legumes in the farming system provide the basis for systems thinking in
soil management research.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Mike K. Wade, NCSU
Heryadi, CSR Martono, CSR
Date Research Initiated:
February, 1984
Rationale for Investigation:
Soybean is one of the major food crops in Indonesia. Any research, soils or otherwise, will be affected by variety used. Improved soybean
varieties suitable for the low elevation humid tropics are relatively scarce. Soybean varieties are notorious for site specificity and few data are available from trials under soil and climate conditions similar to those in Sitiung. It is deemed quite beneficial to use tested varieties in our soil management trials. To lose an experiment due to poor yields or erratic response because of an unknown or poorly chosen variety is very costly in both human and financial resources. It is
perhaps ultimately more beneficial to actually screen varieties and lines for Al and/or low P tolerance. However, this initial trail is needed to help select good genetic stock for use in research. Objective:
Test available soybean varieties in the Sitiung area under high
input technology.
Experimental approach:
Preliminary screening of common soybean varieties in a thrice
replicated trial on an Oxisol was conducted in the Sitiung transmigration area of West Sumatra. It was limed to 1.5 times the exchangeable Al (3.25 T/ha) and received 40 kg P and 100 kg K/ha. Seed was inocu63

Humid Tropics Program **** Indonesia
lated with rhizobium received from NIFTAL. State of Progress:
Six varieties were tested. Wilis, obtained from the old Benchmark project, produced very well. It grew very well vegetatively, matured in 90 days, and had large seeds. A line (B-3038) from Food Crops in
Bogor grew well also, even taller than Wilis, but the seed was very small and yields were lower than Wilis. Orba, the most common improved variety in Indonesia, did poorly.
The Wilis variety was subsequently tried at another location that previously had been limed and fertilized and planted to Orba. Orba growth had been poor and yield was less than 500 kg/ha. Early vegetative growth of the Wilis, with no additional lime or fertilizer, was markedly superior to what Orba had been (height at 30 days was 28 cm for
Orba and 45 cm for Willis).
The variety trial is completed, but will be repeated as more
varieties are collected from whatever sources for testing in Sitiung. Constraints to Progress:
Availability of good quality varieties is quite limited even on a research level. Contacts with other Indonesian agencies and projects need to be improved. Better yet would be contacts with international sources, such as INTOSOY, AVRDC, etc., to help introduce and test more
germplasm in Indonesia.
Applicability of Results to TropSoils Goal:
Varieties that perform well in Sitiung would be possible candidates for other analogous areas of Indonesia, as well as humid tropical areas
of South America and Africa. Initial screening here can help select varieties suitable for testing in other tolerance-type trials such as soil infertility, insects, diseases and water stress which can be an effective means of increasing production. Also using tried and proven varieties in our management trials can help insure good yields, which, in turn, improves the credibility of our research results.

Humid Tropics Program **** Indonesia
Training Component:
Conducting variety trials is a relatively simple matter but has two important functions. One, it teaches those involved how to analyze a crop. The close observations usually made in variety trials, e.g. date of fl lowering, days to harvest, rate of growth, fruiting characteristics, forces the researchers to observe a crop closely at all its development
stages. This provides a useful background for later evaluating the performance of the crop in other type trials, such as a fertility trial. Second, variety trials usually exhibit the large differences that may occur among varieties. It illustrates the importance of knowing and
choosing good varieties for our research, and it shows how the variety chosen can seriously affect or influence the results and conclusions of a given experiment.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Mike K. Wade, NCSU
Carol J. P. Colfer, UH
Djoko Santoso, CSR
Date Research Initiated:
October, 1983
Rationale for Investigation:
Part of the philosophy of the TropSoils/Indonesia program is utilization of the FSR approach, which infers close involvement of both agricultural and social scientists with each other and with farmers. When working in our own culture, especially those with a farm background, there is usually an inherent understanding of the farm culture. However in a foreign setting, that situation does not necessarily exist. In a strict agricultural research sense it does not matter, as we can conduct experimentation on the research station that deals with soil management problems. But such an approach can have shortcomings. Since the ultimate recipient of our agricultural research
is the local farmer, it seems that in order to develop effective and acceptable technology for him, we must gain an understanding of his situation, economically, agriculturally, culturally, and his thinking, goals, conflicts, government influence, etc. Such information can then help the agricultural research develop in a way that will hopefully help optimize our work toward development of effective and acceptable technol ogy.
This research project then is not aimed so much at answering scientific questions, but is done to enhance interaction with the new transmigrants. A few treatments have been selected to test various input packages that might be used by the transmigration program. These are done on farmers' fields and managed by the farmers to see how well

Humid Tropics Program **** Indonesia
the packages perform under "real world" conditions. We have a good idea
of what to expect of these on a station with researcher management but now we test them by the user.
Last, but not least, the interaction of the farmers and researchers allows the researchers to learn from the farmer as well as vice-versa. His years of experience have been a good teacher in how to manage his
land, and we should be able to learn from it. Objectives:
A. To compare fertiizer packages, including rock phosphate or lime
against the current government supplied package of urea and triple
superphosphate (TSP)
B. To enhance interaction with local farmers so as to better understand
a) their problems and b) their goals
C. To learn and get ideas from the farmers, drawing on their knowledge
and experience to help generate more appropriate research Experimental Approach:
On-farm, farmer-managed trials are conducted with 19 participating farmers, in the Sitiung transmigration area of West Sumatra. There is
one replication per farmer; the plot size is 10 x 20 in2. The initial plan is for a one year study, with the intention of making use of established contacts and rapport with these farmers that can be used to test positive results from component trials in years to come. Four packages tested are:
1. no inputs
2. government supplied fertilizer, i.e., 100 kg urea + 100 kg TSP/ha
3. 800 kg rock phosphate/ha + government suppl ied fertil izer
4. 2.5 T lime + 100 kg urea + 200 kg TSP/ha. State of Progress:
An intercropping pattern was used consisting of rice with relay
planted cassava. After the rice harvest, peanuts and chili peppers were
planted between the cassava. To date only the rice harvest is complete.
Tillage decisions were left to the farmer for the no-fertilizer and

Humid Tropics Program **** Indonesia
government packages. The rock phosphate and lime plots were to be hoed after broadcasting the rock and lime. TSP was banded in all cases.
Three farmers were selected not to hoe the rock and lime.
Although tillage was not a factor in the four treatments, it turned out to be a very critical management decision. There was no increase in production by fertilizers when the land was hoed. However if not hoed,
the lime treatment was considerably better than the other packages. In the non-limed treatments, hoeing gave an approximate 50% yield increase. The area for this trial was all newly cleared land, i.e., first planting. Hoeing was very difficult due to the thick mass of roots from the previous forest. But, apparently, incorporating the forest litter (there was no general burn) had a fertil izing effect that was even greater than the urea and TSP and rock phosphate. Most of the Javanese
farmers thought it best to hoe, as they traditionally do in Java. However, the root barrier deterred some of them. Also, the local
(Minang) transmigrants traditionally slash and burn for upland rice cultivation and they advocated not hoeing. They did not think it was detrimental merely a waste of time. However the situation here was somewhat different from the usual slash and burn. This forest had been felled in the rainy season and not burned. The farmers arrived and cleared the felled trees several months later during the following dry season. By then there was little dry leaf litter or twigs suitable to fire a generalized burn. Thus they had to hand cut and pile the limbs and logs.
Currently peanut and cassava harvests are being completed. Pepper failed mostly due to disease and infertility. Pl ans are being made to amend and continue the treatments. Next year' s cropping pattern will be a simple rotation of rice and soybeans or peanuts. Constraints to Progress:
Because of the nature of the FSR, with farmers managing the plots,
the work is not particularly demanding. We only do annual soil analysis and measure only crop yields so facil ity demand is al so minimal The main constraint is having time to observe the plots and interact with farmers in their fields or homes. No fixed schedule is used and so this

Humid Tropics Program **** Indonesia
is done at our convenience when not involved in other research activities.
Appl icability of Results to TropSoils Goal:
This research offers package suitability and management implications for farmers arriving at newly cleared sites. It also gives the researchers the opportunity to monitor soils and farmers from their time of arrival. Information gained can be used for making recommendations and designing research in other new areas. Training Component:
Farmers newly arriving at transmigration sites are often working in a vacuum. They have just been moved from an area with very different soils and climate. They, like the researchers, are in a new and different situation. They are very open, and searching for new information and technology. By working closely with such farmers, the researchers
can find more appropriate and useable technologies that can be of immediate use to other new transmigrants in neighboring or similar areas. By the close association with them, the researchers also gain insight into the goals, problems, and thinking of transmigrants. This interaction in turn should help guide the component research so as to provide more culturally and economically acceptable technology.

Humid Tropics Program**** Indonesia
Leadership Personnel:
Carol Colfer, UH
Barbara Chapman, EPC
Veronica Kasmini, CSR
Bartholomeus Wied Apriadji, IPBNP
Liek Irianti, IPBNP Date Research Initiated:
November, 1983
Rationale for Investigation:
The principle of matching the requirement of a soil management
innovation to the cultural and resource characteristics of the farmer entails knowledge of the requirement of the innovation and the characteristics of the user of the innovation. This interview series is designed to understand the cultural and resource characteristics of the farmer so that soil management innovations can be tailored to match the needs and absorptive capacity of the farmer. Objectives:
A. To monitor characteristics of the farming families with whom
the team is working.
B. To compare their lives with those of a randomly selected
group of "control" farming families. Experimental Approach:
This research is conducted in the Sitiung transmigration area of West Sumatra, Indonesia. The approach involves development of rapport with 40 families, and periodic interaction with them over time. Effort was made to develop a sense of trust and understanding of team goals, and thereby increase the likelihood of getting accurate responses to

Humid Tropics Program **** Indonesia
queries. The families that were surveyed by the nutritionists included those involved in this interview series.
The following represent kinds of questions that have been incorporated into the interview series: Since the people appeared to be quite interested in fruit trees, questions on the kinds, numbers and locations of fruit trees were included. In this way the existence of a variety of fruit trees already planted in people's houselots was verified, as well as the people's interest in these crops. This information is then shared with soil scientists.
The team wanted a sense of whether the people in the community were experienced in agriculture so a question on land holdings and previous agricultural experience was included. About half of the farmers interviewed had been landless in Java, but virtually all had agricultural experience as laborers.
At another point the team wanted to get an idea of how much money the people had brought with them. The purpose of this question was to ascertain whether they had funds to buy agricultural inputs. The whole series was designed to monitor and provide ongoing information as the team decided what would be useful.
State of Progress:
The most relevant benefit of this particular interview series was to provide the team with timely, focused information that relates to ongoing agricultural experimentation in the community. In the future the team expects to use this method to ascertain the profitability of the soil management innovations recommended to and adopted by the farmers. Constraints to Progress:
See Overview
Applicability of Results to TropSoils Goal:
The applicability of this interview series derives more from the
process of determining what questions need to be answered than from the specific questions deemed important by this particular team in this particular location. It is hoped that soil scientists working in this team

Humid Tropics Program **** Indonesia
will leave the project understanding that information from farmers is relevant and is also accessible by relatively straight-forward procedures. If these persons do not try to gain this information themselves in the future, at least they will understand that it is important and seek help from social scientists. Training Component:
The primary training impact of this series to date has been
periodic input from Colfer into decisions about agricultural experimentation.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Carol Colfer, UH
Barbara Chapman, EPC Liek Irianti, IPBNP
Bartholomeus Wied Apriadji, IPBNP Date Research Initiated:
April, 1984
Rationale for Investigation:
A purpose of this study is to compare the nutritional status and
income levels of the people now with what is obtained at the conclusion of the TropSoils project. In the interim, the team hopes to compare various survey questions in this location and in the urban context of
Bogor. It may be possible to make comparisons with other countries which participated in the Street Foods Project. Chapman is quite interested in comparing the findings in Sumatra with those of her previous research in Central Java several years ago. The analysis will provide the team with a means of assessing the appropriateness of proposed new crops for experiments over time. Entry of these data into a computerized data base management system will improve their usefulness. Objectives:
Establish a baseline for nutritional status, dietary patterns and income levels in the area of field research investigations. Experimental Approach:
This research is conducted in the Sitiung transmigration area of West Sumatra, Indonesia. The dietary patterns were of particular interest since the team hoped to select food crops for use in management experiments which would be usable by transmigrants in their subsistence efforts and would supplement nutritional deficiencies. Data were also

Humid Tropics Program **** Indonesia
collected on the agricultural production of families. The team wanted a reading on the relative wealth of Sitiung I compared to II (irrigation and no irrigation, respectively); and Sitiung V which, in contrast to I and II, is newly settled.
The approach was to utilize experienced researchers combined with knowledge of local conditions to construct a survey instrument that
would provide the information desired. People were interviewed in their homes, usually in their language. Each of the 80 families was interviewed twice, on consecutive days, so that actual food consumption data would be as accurate as practical and relying as little as possible on
memory. Income data were obtained to include agricul tural production that was consumed and not sold, using market prices that were obtained at the time of the interviews.
State of Progress:
The results showed Sitiung I and II to be grouped together, incomewise, at the top and bottom of the ladder. Sitiung V was generally in the middle, from an income standpoint. Incomes range from approximately $8.00 to $200.00 per month. Nutritional status was marginally adequate. The lack of variety in the diets was suggested as a possible contributor to their nutritional deficiencies. Virtually no meat was consumed by the
people in any of the locations. The nutritionists warned the team of a danger in focusing on high value crops, the reason being that the people
will sell them rather than consume them to improve their nutritional status. They also suggested that the team introduce some of the variety
of seeds for edible plants available in Java. Contraints to Progress:
See Overview
Applicability of Results to St4-CRSP Goal: The applicability of these findings should be reasonably great for
other transmigration sites. The findings from Sitiung V, where the people had only recently stopped receiving their government subsidy, should be of relevance for assessing the adequacy of the subsidy. The

Humid Tropics Program **** Indonesia
description of the subsistence adjustment that has been made by the longer term residents can also be of use to the team in suggesting agronomic improvements that are consistent with existing patterns. Training Component:
Training activities were limited to the informal interaction that occurred between team members and the visiting researchers. Some team personnel were alerted to nutritional inadequacies and possible agricultural solutions; but the regular seminars that had been planned could
not be arranged due to Colfer's illness.


Humid Tropics Program **** Indonesia
Leadership Personnel:
Carol Colfer, UH
Atin Kurdiana, CSR
Edi Santoso, CSR
Veronica Kasmina, CSR
Date Research Initiated:
September, 1983
Rationale for Investigation:
Many promising soil management innovations and practices are not
adopted by farmers because the effort required to implement the innovation conflicts with the work habits and schedule of the farmer. Farmers generally invest their greatest effort on activities which are critical to their survival and quality of life. Time allocation studies of farmer activities are efficient ways to identify priority soil management research that corresponds to farmer needs. Objectives:
A. To ascertain the current division of farm labor, by sex and
B. To determine how people were choosing to use their labor.
C. To determine important seasonal variation in activities.
D. To maintain ongoing communication between researchers and
farming famil ies.
Experimental Approach:
This research is conducted in the Sitiung transmigration area of West Sumatra, Indonesia. The approach was somewhat different from many time allocation studies in that people were not asked to remember how much time they devoted to some preselected tasks. Rather a randomized schedule was drawn up at the beginning of the study for the entire year. Visits were then made to the scheduled households, and the activities of

Humid Tropics Program **** Indonesia
all members of the household were noted. Four households were visited each day.
State of Progress:
The data are currently being entered into the computer to form a data base from which specific soil-management-related questions can be asked related to people's usual activities. The large number of observations and their random nature permit reasonably accurately general izations to the general populace. The fact that two very different transmigr-ation sites were used, one representing a long-established one and the other a newly-settled site, allow the team to ascertain important differences based on length of residence. These data will also provide a measure of the changes that occur over the course of the research.
The team has already had cause to question the frequency with which people must search for grass for their cattle and goats, the division of agricultural labor between the sexes, the incidence of off-farm
employment, monthly variation in productive activities of adults, among others. When the data are fully entered into the computer system, they
will be much more accessible to team members. Constraints to Progress:
Refer to Overview
Applicability of Results to TropSoils Goal:
The applicability of these findings will depend on the similarities between this location and others to which one might wish to generalize. However, the process is an easy one to replicate, and other researchers all over the world are conducting time allocation studies (using this method as well as others). There is considerable probability that transmigrants of similar ethnic groups in other tropical rainforest
locations will have similar options for spending their time.
Besides the benefit of allowing the team to answer specific,
research-related questions about the population with which they are working, this time-allocation study contributes to a growing body of

Humid Tropics Program **** Indonesia
information about how people spend their time. In recent years, there has been an increasing recognition that unpaid work has been underestimated and undervalued. This particularly affects poor people and women, and researchers have begun to develop a body of information on how people spend their time when they are not being paid. The TropSoils studies can contribute to this body of knowledge.
Training Component:
The training of CSR personnel in this kind of study can contribute to a more people-oriented research agenda for the future, as well as
teaching soil scientists one of the techniques for maintaining awareness of community concerns and interests. A number of CSR personnel have been involved in data collection for this study. This provides a mechanism by Aich they regularly interact with farmers, becoming attuned to their concerns, constraints and goals. This in turn effects their decisions about the kinds of experimentation that are likely to yield agricultural technology that can be used by farmers.



Humid Tropics Program **** Peru **** Brazil
Lead Institution
North Carolina State University Collaborating Institutions
Instituto Nacional de Investigacion Agropecuaria
Unidade de Execucao de Pesquisa de Amrnbito Estadual de Manaus Linkage Institutions
Centro Internacional de Agricultura Tropical
Centro Internacional de la Papa
CorporaciOn de Desarrollo de Loreto
Empresa Brasileira de Pesquisa Agropecuaria
International Council for Research in Agroforestry
International Fertilizer Development Center
International Institute for Tropical Agriculture
Instituto Interamericano de Coorperacion para la Agricultura
Instituto de Investigaciones de La Amazonia Peruana
Instituto Veterinario de Investigacion del Tropico y Altura
Potash and Phosphate Institute
Proyecto Especial Alto Huallaga Proyecto Especial Madre de Dois
Proyecto Especial Pichis Palcazu
Red de InvestigaciOn Agroecologica para la Amazonia
Rockefeller Foundation
Universidad Nacional Agraria la Molina
Universidad Nacional de la Amazonia Peruana

Humid Tropics Program **** Peru **** Brazil
Universidad Nacional de la Selva Tingo Marta Research Sites
Yurimaguas Agricultural Experiment Station, Yurimaguas Peru
Unidade de Execucao de Pesquisa de Ambito Estadual de Manaus,
Manaus, Brazil
La Esperanza Experiment Station, Pichis Valley, Peru
IVITA Principal Tropical Station, Pucallpa, Peru
Tulumayo Experiment Station, Tingo Maria, Peru
Puerto Maldonado Experiment Station, Madre de Dios, Peru Principal Investigators
Pedro A. Sanchez
John J. Nicholaides, III
Representatives on Board of Directors
Lead Institution
Robert H. Miller
Collaborating Institution
Victor Palma

Humid Tropics Program **** Peru **** Brazil
Name, Degree TropSoils Responsibility Affiliation
Pedro A. Sanchez, Ph.D. Soil Management/ NCSU
John J. Nicholaides, Ph.D. Soil Fertility1/ NCSU
Julio C. Alegre, M.S. Soil Physics NCSU
Miguel A. Ara, M.S. Soil Management NCSU
-Luis Arovalo, M.S. Soil Chemistry INIPA
Miguel A. Ayarza, M.S. Soil Management NCSU
Andres Aznaran, B.S. Agric. Engineering INIPA
Dale E. Bandy, Ph.D. Soil Management NCSU
Joaquim B. Bastos, M.S. Soil Management EMBRAPA
Jose R. Benites, Ph.D. Soil Management NCSU
Stanley W. Buol, Ph.D. Pedology NCSU
Eber Cardenas, B.S. Pastures UNAS
Rafael Chumbimune, M.S. Soil Fertility INIPA
D. Keith Cassel, Ph.D. Soil Physics NCSU
Jose Carlos Correa, M.S. Soil Management EMBRAPA
Charles B. Davey, Ph.D. Forest Soils NCSU
Jose R. Davelouis, M.S. Soil Management UNA
Dennis del Castillo, Ph.D. Soil Management NCSU
Rolando Dextre, Ing. Agri. Soil Management INIPA
Helmut Elsenbeer, M.S. Soil Physics NCSU
Martha Gallo, Ing. Agr. Soil Fertility INIPA
Expedito U. Galvao, M.S. Crop Science EMBRAPA
Marco Gavez, Ing. Agr. Agronomy INIPA
Mwenja P. Gichuru, M.S. Soil Fertility NCSU
Robert E. Hoag, M.S. Pedology NCSU
Hemilce Ivazeta, Ing. Agr. Pastures INIPA

Humid Tropics Program **** Peru **** Brazil
Lisa B. Katz, M.S. Soil Fertility NCSU
Jonathan LOpez, Ing. Agr. Agronomy INIPA
Robert E. McCollum, Ph.D. Soil Fertility NCSU
Jane Mt. Pleasant, M.S. Weed Control NCSU
Ruben Mesla, Ing. Agr. Extension INIPA
George C. Naderman, Ph.D. Soil Management NCSU
Laurie R. Newman, B.S. Pedology NCSU
Marco A. Nurena, Ing. Agr. Extension INIPA
Cheryl A. Palm, M.S. Soils-Agroforestry NCSU
Jorge R. Perez, Ing. For. Agroforestry NCSU
Beto Pichanasi, Ing. For. Agroforestry INIPA
Melvyn Piha, M.S. Soil Fertility NCSU
Antonio Polo, Ing. Agr. Agronomy INIPA
Carlos Pomareda, Ph.D. Economics INIPA
Martti Poutannen, B.S. Agroforestry IDSF
Jamie Powell, B.S. Soil Chemistry NCSU
Alfredo Rachumi, Ing. Agr. Agronomy INIPA
Kenneth Reategui, M.S. Pastures NCSU
Olga Rios, M.S. Soil Chemistry UNP
Alcibidades Sanchez, Ing. Agr. Perennial Crops PEPP
Rodolfo Schaus, Ing. Agr. Soil Management NCSU
Paul C. Smithson, M.S. Soil Chemistry NCSU
Thomas J. Smyth, Ph.D. Soil Fertility NCSU
Lawrence T. Szott, M.S. Agroforestry NCSU
Johannes Van Diepen, M.S. Soil Management NCSU
Jorge Vela, Ing. Agr. Pastures INIPA
Manuel Villavicencio, Ph.D. Soil Management2/ INIPA
Adelo Vivanco, Ing. Agr. Soil Fertility INIPA
/Also, Principal Investigator.
/Also, Director YAES.
* Enrolled in a graduate program leading to next highest academic

Humid Tropics Program **** Peru**** Brazil
This program is rooted in the experiences, results and infrastructure developed during the preceding 12 years of collaborative soils research in Peru and other countries in Latin America. A recognition of the relationships, organization and operational procedures which evolved from these activities is important to an understanding of this TropSoils
Since 1971 The Tropical Soils Program at NCSU has operated as the international component of the Soil Science Department for teaching, research and extension. Although it has its own office and laboratory facilities, its activities are woven through the fabric of the Department, with participation of faculty members who carry both domestic and international responsibilities. Of the 41 present faculty members in the Department, 20 have had significant involvement in the Tropical Soils Program. Approximately 38 percent of the M.S. and Ph.D. degrees awarded by the Soil Science Department from 1973-1981 are based on thesis research related to tropical soils. Many of these graduates occupy key leadership positions in soil science around the world.
Field research started in 1972 with the establishment of cooperative experiments at primary sites in the humid tropics (Yurimaguas, Peru), acid savannas (Brasilia, Brazil) and the volcanic highlands of Central America (Turrialba, Costa Rica). Cooperative relationships were
-Prepared by Pedro A. Sanchez and John J. Nicholaides, III,
Principal Investigators.

Humid Tropics Program **** Peru **** Brazil
developed and formal agreements were signed with the predecessor agencies of INIPA in Peru, EMBRAPA in Brazil and CATIE in Costa Rica. The Central American program produced the first multiple cropping research focus in this region and led to the farming systems program of CATIE. Work in the Cerrado of Brazil, conducted jointly with Cornell University, developed basic soil management technology for savanna Oxisols and was instrumental in establishing the research program at EMBRAPA's Cerrado Research Center.
The humid tropical component of the program has operated for 12 uninterrupted years in spite of funding shortages and political upheavals. A strong commitment by the Peruvian research institutions, the Office of Agriculture of AID/Washington and NCSU, permitted an opportunity for continuous direct involvement of NCSU at one key location, Yurimaguas, Peru. Core funding for the program has been provided by AID through Contract AID/csd 2806 "Agronomic-economic research on tropical soils" from 1970 to 1975, by Contract AID/ta-C-1236 "Discovery of new management systems for tropical soils" from 1976-1981, and from the Soil Management CRSP since 1981.
As NCSU's Tropical Soils Program became better known, several
institutions requested assistance in various ways. A change in government priorities in Peru towards development of the Selva prompted USAID/Lima and various Peruvian institutions to request our involvement in country programs involving research, extension and training in tropical soil management. The question of whether the Yurimaguas technology is applicable to the drier humid tropics in Brazil prompted a cooperative project in Manaus jointly funded by the Rockefeller Foundation, the Potash-Phosphate Institute and EMBRAPA. The emergence of REDINAA carried a request to develop an Amazon-wide soils project. A similar development for IBSRAM led to involvement in designing a worldwide acid tropical soils network.
Funding availability necessarily limited the extent to which a
favorable response could be given to such requests. Thus, a decision was made to seek other sources of funds to supplement those provided by the CRSP. Several donors responded positively, some granting funds directly

Humid Tropics Program **** Peru **** Brazil
to NCSU and others to our counterpart national institutions such as INIPA. The "collaborative" word in the CRSP became a reality in terms of funding. What formerly was a single donor-sponsored program, therefore, became a multi-donor program. Because all funds are for programs that support TROPSOILS objectives, funds from the Soil Management CRSP are considered as "core," while those from other donors are considered as "special projects." In addition to INIPA and EMBRAPA's core budget contributions, the other donors supporting this program are USAID/Lima, the Rockefeller Foundation, Potash-Phosphate Institute, Corporacion de Desarrollo de Loreto, Proyecto Especial Pichis Palcazu, and the International Development and Research Centre. Stages of Program Development:
NCSU's Tropical Soils Program activities in the humid tropics have gone through several stages characterized by technical focus, involvement of collaborative institutions and funding. A summary of this development is shown in Table 1.
Stage 1: The establishment phase took approximately two years between contract awarding and the initiation of field experiments. Much of the effort was geared towards reviewing the available literature and extensive travel through the Latin American tropics to establish the research priorities and a work plan. This effort culminated in a stateof-the-art publication, "A Review of Soils Research in Tropical Latin America," published in 1973 in English and Spanish.
Soil characterization to understand the properties and variability of Amazon soils was also a major undertaking. The selection of the Yurimaguas Station was a consequence of such studies. Time has proven that it is indeed a site representative of the kinds of climatic, soil, vegetation and socioeconomic constraints of the humid tropics. The degree of expression of these constraints, however, varies widely within the region. During this stage, the Fertility Capability Classification (FCC) system emerged as a means for interpreting Soil Taxonomy data for agronomic purposes.
Stage 2: The field phase started with the posting of NCSU staff at Yurimaguas in July 1972 with work on mapping, land clearing and

Table 1. Stages of development of NCSU's Tropical Soils Program in the humid tropics.
Stage Establ ishment Continuous Cropping Management Options
and dates: 1970-71 1972-79 1980-85
Focus Establish program Text continuous cropping Develop different options for
hypothesis soil-l andscape- infrastructure
Main activities Priority development Soil characterization Stability of high input systems
Site selection Land clearing Low input technology
Agreements Fertility dynamics Legume-based pastures
Multiple cropping Agroforestry
Paddy rice for alluvial soils
Main Review of soils Feasibility of continuous Quantified long-term effects
accomplishments research in tropical cropping proved, with on soil properties
Latin Pmerica proper land clearing, Developing promising options
Yurimaguas Station fertilization and liming for low input cropping, pastures
established and agroforestry
Paddy rice technology applied
Research output FCC concept Basic understanding of Long-term effects on soil
dynamics after clearing properties
rainforest attracts Managed fallow concept
international attention Components of different options
FCC gains wide acceptance
Extrapolation None San Ignacio, Bolivia Manaus, Sumatra
Around Yurimaguas INIPA Selva Program
REDINAA network development
IBSRAM network development

Table 1. (continued)
Stage Establishment Continuous Cropping Management Options
and dates: 1970-71 1972-79 1980-85
Training None Limited to NCSU graduate NCSU, UNA and European Univ.
students Short-term training of research
and extension workers
LDC None Virtually none until 1978 Major Peruvian involvement, also
participation Strong moral support, but Brazilian, Indonesian, USAID
no funds Missions, IARC's
Funding 100% AID-Washington 100% AID-Washington 70% AID-Washington
30% LDC, USAID Mission and Others

Humid Tropics Program **** Peru **** Brazil
continuous cropping. Efforts concentrated on testing the hypothesis that continuous cultivation of annual food crops is possible in acid infertile soils of the humid tropics. Nutrient dynamics were followed and a series of experiments on fertilization, liming, multiple cropping, structure regeneration, land clearing options were conducted. The FCC system as a precursor of extrapolation efforts continued to be tested. Modest extrapolation efforts began in 1978 at San Ignacio, Bolivia, and on farms around Yurimaguas. This phase culminated with the finding that after eight years, continuous cropping is indeed feasible in agronomic terms provided the marketing infrastructure is present.
The main impact of this phase was the elimination of the myth that these soils are not suitable for cropping, a finding that attracted international attention because it provided a new outlook for world food production. Land clearing research provided the basis for fundamental changes in Brazil's land development approach in the Amazon.
Stage 3: The advent of the Soil Management CRSP and the change in Peruvian government priorities towards the Selva cast the Program in a very different light. Worldwide assessments recognized the importance of research on marginal lands vis-a-vis the green revolution approach. Ecological concerns about deforestation in rainforest areas brought additional attention to soil management research.
It became clear that providing the continuous cropping hypothesis with solid, long-term data was a necessary but not a sufficient condition for the utilization of this technology. The program then began to focus on several soil management options suitable for different soil, landscape and infrastructure conditions. A model emerged which considers low input approaches such as use of acid-tolerant germplasm and managed fallows, legume-based pastures and agroforestry (Figure 1). Attention began to be given to the fertile alluvial soils as an important component. The Tropical Soils Program broadened its scope from one option (fertilizer-based continuous cropping) to five.
Participation of Peruvian institutions, both in terms of funding and personnel, mushroomed. Paddy rice research on alluvial soils began with direct support from CORDELOR. INIPA and USAID/Lima assigned signi94

Humid Tropics Program **** Peru **** Brazil
ficant PL-480 funds to support both the physical plant and operations at the Yurimaguas Station. The strengthening of INIPA under the Integral REE Program financed by USAID/Lima, the World Bank and the Interamerican Development Bank resulted in the development of a National Selva Program with 197 research and extension professionals stationed throughout the Selva of Peru. On-the-job training of 46 Peruvian professionals resulted in an extrapolation program operated on-site and financed from sources other than the CRSP. INIPA and USAID/Lima requested a major increase in the training/extrapolation activities. Outside Peru, two major research sites began to test the Yurimaguas technology on a systematic basis: at EMBRAPA's Manaus Station and at Sitiung in West Sumatra, Indonesia. A third site, the Pichis-Palcazu in Central Peru, started in 1984. In addition to the National Selva Program, two additional networks are at advanced stages of development, REDINAA for the Amazon and IBSRAM for acid tropical soils on a worldwide basis. International testing and adoption of the FCC system is in progress in Asia, Africa, Latin America and the United States.
Current Status of the Tropsoils Work Plan:
The Work Plan presented in May 1982 to the TropSoils Technical
Committee was based on the CRSP grant documents, an AID-sponsored review in October 1981 and an NCSU internal review in October 1982. The final version dated 15 February 1983 constitutes our research commitment to the CRSP for the five-year period. The Work Plan includes 72 experiments organized into six projects.
The Work Plan defines the overall purpose of NCSU's Tropical Soils Program as follows: "to develop and transfer, together with national institutions and other TropSoils universities, improved soil management technologies for productive and sustained farming systems in humid tropics and acid savanna ecosystems on an agronomically, economically and ecologically-sound basis."

Humid Tropics Program **** Peru **** Brazil
The Work Plan lists six specific objectives for 1982-1986. A short discussion of the accomplishments and shortcomings at approximately the midpoint of this period follows.
A. Determine the Long-Term Stability of Continuous Crop Production
Systems Based on Judicious Lime, Fertilizer and Management Inputs
for Soils of the Humid Tropics.
Data on crop performance and soil fertility dynamics during the first eight years of the continuous cropping experiment were analyzed during this period and published in articles in Science and the Soil Science Society of America Journal in 1982 and 1983, respectively. The agronomic and economic data plus a hypothesis on their ecological implications were included. This high input-approach was recognized as being limited to areas with ready accessibility to inputs and a good market infrastructure. Examples of such conditions are found in the Alto Huallaga region of Peru and in West Sumatra, Indonesia.
During the 1982 internal review, it was recognized that for such continuous cultivation to be truly stable, it has to be mechanized to provide a realistic management of crop residues, weed control and maintenance of physical properties. Deeper plowing resulted in higher phosphorus requirements caused probably by mixing part of the subsoil with fertilized topsoil. Weed control is being studied for the first time on a systematic basis.
B. Develop Soil Management Practices for Continuous Crop Production
Systems Based on Low Lime, Fertilizer and Tillage Inputs.
Concentration on low input technology was strongly recommended by the 1981 AID Review of the previous program. A conceptual paper was developed, together with CIAT, summarizing current information on low input technology for acid soils and providing a conceptual basis for the approach. The paper was published in-Advances in Agronomy in 1981 and the Spanish translation in 1983 by the Colombian Society of Soil Science. The strategy is based on selection of acid-tolerant cultivars, minimum tillage and fertilization, weed control and managed allows. The more immediate potential applicability of this approach to small farmers in the humid tropics resulted in major efforts directed toward this objective since the CRSP was initiated.

Humid Tropics Program **** Peru **** Brazil
Several key components have come to fruition in the past two years. Upland rice and cowpea cultivars from IITA that combine high yields and acceptable tolerance to pests and diseases with tolerance to high levels of Al saturation have been identified and are now being widely tested through our INIPA collaborators. We have failed, however, to obtain similar results with corn, soybeans, peanuts or winged beans. Screening continues and peanut work is now involving specific crosses made in North Carolina which are tested in Peru. Work with minimum tillage, residue management and their fertility interactions have yet to provide clear evidence of the superiority of minimum or zero tillage, unlike what has happened on Alfisols of IITA. The promotion of faster downward movement of Ca and Mg is being tested. Fertility-weed control interactions are now being studied. A farm level low input central experiment has shown the feasibility of upland rice-cowpea rotations starting with forest clearing with zero tillage, proper spacing and only small rates of N and K have produced a total of 10.3 tons/ha of rice and 2.1 of cowpeas during the first two years.
We suspect that the low input system, however, is unstable as far as continuous cropping is concerned. It may last for more than two years but then a decision will have to be made about the permanent role of the field. Alternatives are high input cropping, managed fallow, grasslegume pastures, or agroforestry. One way to maintain a "continuous" low input system is to rotate one or two years of crops with one to three
years of legume fallow.
C. Develop Soil Management Practices for Different Landscape Positions, Including Legume-Based Pastures, Agroforestry and
Flooded Rice on Alluvial Soils.
The pasture project has provided very positive results. Three years of grazing trials using germplasm from CIAT have demonstrated the high productivity and persistence of mixtures of acid-tolerant grasses and legumes such as Andropogon gayanus-Stylosanthes guianensis and Brachiaria decumbens-Desmodium ovalifolium, with annual liveweight gains of 700 kg/ha with rotational grazing management and with minimal fertilizer inputs. This provides a promising alternative for degraded

Humid Tropics Program **** Peru **** Brazil
pastures in the udic tropics with low animal productivity which decreases and leads to overgrazing and subsequent soil erosion. Pasture establishment problems have been alleviated by experience and better seed sources. Regenerating degraded Panicum maximum pastures on farmer fields was simply accomplished by broadcasting Bayovar rock phosphate.
The principal problem seems to be the low nutritional quality of some grass:legume mixtures which is reflected in relatively low liveweight gains per animal. Initial work in this direction has identified high tannin content as a constraint. Phytotron work in Raleigh is testing the hypothesis that sulfur deficiency may be responsible. Plans are underway to validate initial observations of incipient nutrient recycling in well managed grass-legume pastures at Yurimaguas. The question of whether to use legumes at all is to be studied in a nitrogen transfer experiment.
Research on agroforestry continues to increase. After observational trials on Gmelina arborea intercropping and peach palm (Guilielma gasipaes)-legume interrows, work has centered on establishing the nutrient response patterns of Gmelina and peach palm on a systematic basis. The first, to our knowledge, fertilizer trial of peach palm in acid soils shows a very strong N response and also to other nutrients. Germplasm collection of this valuable species is taking place, as Yurimaguas is one of its centers of origin. These studies have led to the awarding of an IDRC grant through ICRAF for agroforestry research in Yurimaguas.
Alley cropping work began in 1983 with the overall purpose of finding species that can do in acid soils what Leucaena leucocephala does in high base status soils. One trial with several promising species is ongoing and germplasm collection from sources in the Amazon, Africa and Southeast Asia continues.
The question on how fallows regenerate soil productivity is being evaluated on a preliminary basis and more intensive research is now being done in the library to develop hypotheses and future plans.
Flooded rice production on high base status alluvial soils has
been an unqualified success. Five crops of irrigated rice have been

Humid Tropics Program **** Peru **** Brazil
grown during two years with an average production of 16.5 tons/ha/yr. Replacing transplanting with pre-germinated broadcasted seed depressed yields by 20% but decreased labor input for crop establishment from 20 to 3 man-days/ha.
D. Extrapolate, Validate and Adapt Research Results to Other Humid
Tropical Areas, Including the Peru Selva Network, Manaus,
Indonesia and REDINAA Soils Network.
The request for assistance in the Selva of Peru resulted in consideration on the adaptation of the Yurimaguas results to special project areas. We expressed the need to validate the results first to make sure that local adaptations could be incorporated. It became apparent that research workers needed prior training to validate and transfer the results in a sound manner. The frequent request for short-term (1-2 weeks) training made it necessary to make on-the-job training a formal activity. During 1982 and 1983 a total of 58 professionals received onthe-job training at Yurimaguas. The 51 Peruvians came from most areas of the Selva and include researchers, extension specialists, farmers and representatives of native communities. Other training activities include graduate students from Argentina, Finland, Germany, the Netherlands and the United Kingdom. Ten NCSU graduate students are in different stages of conducting their theses research at Yurimaguas or in adjacent areas.
Participants of the 1983 Tropical Soils Management Course designed validation trials for their regions. As a result, a network of 29 cooperative trials were established throughout the Selva, on specific soil management options the participants considered appropriate for their localities (Figure 2). These trials are now coordinated through INIPA's newly established National Selva Program. NCSU staff has assisted INIPA in the development of two additional components of the
National Selva Program: the National Tropical Pastures Network and the Agroforestry Project. Internationally recruited staff from CIAT and ICRAF will participate in these projects.
In 1980 an agreement was reached between EMBRAPA, IICA and NCSU to post an NCSU senior scientist at EMBRAPA's Manaus Station to test some