TropSoils, the first three years

Material Information

TropSoils, the first three years
Caudle, Neil
North Carolina State University -- Department of Agricultural Communications
Place of Publication:
Raleigh, NC
TropSoils, North Carolina State University


Subjects / Keywords:
Farming ( LCSH )
Agriculture ( LCSH )
Farm life ( LCSH )
Africa ( LCSH )
Spatial Coverage:


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.

Full Text

the first three years


the first three years

About TropSoils

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 in the program 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'Economie Rural-Mali
Instituto Nactional de Investigacion y Promocion Agraria-Peru
International Crops Research Institute for the Semi-arid Tropics-India
North Carolina State University-USA
Texas A&M University-USA
University of Hawaii-USA

The Cover

Nowhere is the dependence of people on soil more critical or apparent today than in Africa's Sahel, where drought, land degradation and population growth have brought famine and hunger to millions. In view of this crisis, we present a Nigerien woman and a Sahelian landscape in this composite photograph. But in developing nations throughout the tropics, this link of people to soil is no less real, the research no less vital.


TropSoils, the first three years was written and designed by Neil Caudle, Department of Agricultural Communications, North Carolina State University. Production was by Paragraphics, Inc.
Zones maps and diagram, page 42, by Anne Marshall Runyon. Photos by: Neil Caudle, cover, pp. 6, 7, 8, 10, 11, 12, 14 (left), 15, 16, 18, 19, 20, 21, 22 (right), 23, 24, 40, 43; Frank Calhoun, p. 14 bottom; Walter Bowen, p. 22 (left); Pedro Sanchez, pp. 26, 27, 28 (bottom), 29, 34, 36, 37 (left), 39, 44, 46 (right); James P. Blair (c) National Geographic Society, pp. 30, 31, 35; Herman Lankford, p. 33; Larry Szott, p. 38; Jot Smyth, p. 41 (top); Carol Colfer, pp. 46 (left), 47, 49, 50, 51, 53; Mike Wade, p. 52.
Copies of this report, and a companion book, ropSoils Triennial Technical Report, can be obtained from TropSoils, Box 7113 Williams Hall, North Carolina State University, Raleigh, N.C. 27695-7113.


]Earth. One name for two things:
the planet we share, the soils at our feet. Both are in danger. We look around the world and find ourselves needy, angry, starving. The dusts of Africa descend on us all.
In the tropics, our good land is crowded, there isn't enough. We are leaving the riverbanks, coming down from the mountains, up from the coasts. We sew the dry edges of deserts with grain. We torch the stiff brush of savannas so cattle can graze. We slash at the jungle, raise a scant crop in its ash, and move on.
These are the thin lands, the parched lands, the acid, infertile and
erodable lands, where we're planting with sticks and stooping to weed. There is no easy wealth in these lands. Still we must have them, millions more hectares each year. There are so many more of us eating scarce food, wearing scarce fibre, burning scarce firewood and dung. We consume what has rooted in soils.
And now that we're staking our futures on these marginal lands, what next? Can they sustain us? Can we both use and conserve them? What must we learn?

''prevent famine and establish freedom from hunger"
This is the heart of Title XII of the
United States Foreign Assistance Act. Title XII recognized the role of U.S. land grant universities in the development of modern agriculture, and provided for a new partnership in which universities would join government in an effort to help poor nations develop their agriculture and secure their food supply. Title XII also recognized that, in order to apply science to solving food and nutrition problems, universities would need long-term, dependable support to develop strong programs in research, institution-building and technological assistance.
Much of this work would focus on ways to produce more food from plants and animals. Because soil-related problems often seriously constrain this productivity, some of the new research must logically focus on soils. And, since the greatest human need, and the greatest potential resource base from which to meet that need, are found in the tropics, that was where the soils research began.

TropSoils is one of the Collaborative Research Support Programs (CRSPs) created by the Board for International Food and Agricultural Development (BIFAD) to implement Title XII. Nearly two years in the planning, TropSoils is an international collaboration that links developing nations, the U.S. Agency for International Development, international research centers and U.S. universities. The collaboration ensures that partner nations have a stake in 'YopSoils' research, that it meets their needs, and that the program draws on the broadest possible base of knowledge and expertise.
To keep this collaboration working smoothly toward its goals, the program is guided by four groups: the management entity, established at North Carolina State University to administer the overall program; a board of directors, consisting of one representative from each university and collaborating nation; a technical committee, composed of the principal investigator from each U.S. university; and an external-evaluation committee of three scientists from organizations that are not TropSoils participants.


TropSoils research is keyed to agro-ecological zones in the tropics

"agronomically, economically and ecologically sound"

TropSoils' goal is to develop and adopt soil-management technology that is agronomically, economically and ecologically sound for developing countries in the tropics. Because it has been impossible to do this in every tropical nation or region at once, the program has sought to situate and develop its research projects in a way that would make their results applicable over broad areas having similar soils and environments.
These areas, or "agro-ecological zones," became the basic units of TropSoils' organization. Each participating university took a lead role in one of these zones:
e Texas A&M University, with its expertise in dry-land agriculture, was chosen to lead the program in the semi-arid zone, and began research at sites in two African nations, Niger and Mali.
Cornell University, with a fine record of research in Brazil's Cerrado region, took the leadership role in the acid-savannas zone, and began work near Brasilia. North Carolina State University (NCSU), which had also conducted extensive research in the Cerrado, assumed a support role.
o The humid tropics, because of their vast area and potential, were assigned two primary research sites, each with its own lead university. NCSU, which had already established a strong program of tropical soils research in Peru since it began working there in the 1950s, took the lead there, continued developing a research station at Yurimaguas, and began work on a second site near Manaus, Brazil. The University of

Hawaii, because of its strong standing in tropical agriculture and a long record of experience in Asia, was the choice to lead a new program based in the Sitiung transmigration settlements of West Sumatra, Indonesia. NCSU provides support, extending and testing some of its findings from the work in Peru.
Several of these universities have begun or have planned work in another important zone, the steeplands, where hilly or mountainous terrain makes erosion a serious environmental and agronomic concern. The steeplands zone, which overlaps the other three zones, will certainly figure in future collaborative research programs.

the first three years
TropSoils was formally initiated in September, 1981, and the job of developing agreements with governments and institutions began. Now the teams are on-site, the research moving ahead. It's a slow business, equipping and fielding a team in remote regions, bridging the cultures, breaking new ground. But already there are results, successes. Basic information is accumulating about such things as soil types, fertilizer responses and water requirements. And at each site, in every zone, scientists and cooperating farmers are showing that simple, low-cost techniques can improve soil fertility and productivity and boost yields in many cases.
So the effort is well underway, not just to cover more tropical soils with good crops, but to make sure that the new technology fits agronomically, ecologically and economically the regions and peoples it serves. That is the story of this report.



Semi-Arid Tropics

The tropics of Asia, Africa and Latin America all have substantial areas of semi-arid lands, with a dry season of six to nine months and soils that are dry almost half the year. Shifting cultivation, nomadic herding, livestock ranching and subsistence tillage share these regions, which, under the pressure of expanding populations, are facing severe problems with drought, desertification, erosion, soil-crusting, sand-storm damage and declining soil fertility. Most of TropSoils' work in the semi-arid SEMI-ARID TROPICS/NIGER AND MALI
tropics is set in Africa in Niger, Mali, and Cameroon but the research here will apply throughout the semi-arid zone. The reports that follow focus on the Sahel, that broad swath of rolling, sandy lands just south of Africa's Sahara. It is a region where crops, and the people who grow them, exist on the narrowest of margins. It is, in the words of researcher Bob Chase, "life on the edge."

Primary Collaborators
National Institute of Agronomic Research for Niger (INRAN); International Crops Research Institute for the Semi-Arid Tropics (ICRISAT); U.S. Agency for International Development; Texas A&M University.

Principal Investigator
Frank G. Calhoun, soil scientist (pedology), Texas A&M University.


Two faces of the Sahel: parched during droughts and dry seasons-

The Sahel

The Sahel. Here every part of life, even the light itself, is colored by soil. Red dust rises on the harmattan and through it sunlight filters amber and diffused. It is land on the verge of desert, dry soil thirsting for rain, waiting for its yearly ration of a few hundred millimeters, which will come in brief bursts sometime from May through September. Each kilometer north toward the desert, a millimeter less will fall.
For centuries, these semi-arid lands fed only wildlife and the animals of nomadic herders. Agriculture was a practice for the riverbanks and the somewhat moister southern regions, where the Sahel gives way to the Sudan.
The nomads still ride their camels across the northern dunes and the windswept valleys of ancient, vanished rivers, driving their cattle and goats. But recently the farmers have been pushing northward, out to where the earth is mobile in the wind, where great, rolling sand storms blast and bury crops, and the dry months often outnumber the wet. April is a brown, parched and hungry time, waiting for green. And when the season turns at last, the rains

may fall too little or too late to sprout the millet seed, or bring the heads to flower. With the red dust swirling outside his window, seeping in to cover his books, his desk, his work, Mamadou Ouattara talks about the land.
"Before, we had an equilibrium between man and nature," he says. "The soils were not fertile but we were able to manage them. Now, with the population growing, and the recent droughts, the farmers have been expanding their fields. They have increased their area so that when a crop fails they won't be left with nothing. The traditional fallow has almost disappeared, the fertility is decreasing, and the soils are very weak, very subject to erosion, both from water and wind. The level of production is really quite low."
Eighty-seven percent of Niger's people are farmers, working the land by hand, with almost no fertilizers or irrigation. The mean yield from Niger's millet lands 400 kg per hectare is perhaps a fourth of what it could be if such problems as soil fertility, erosion and sand-storm damage were solved. Big ifs.
For Ouattara and his fellow Nigeriens, the crisis is here and now. More drought could bring the country famine. To them it's not enough to sit and wait and count on food aid. They must find a way, not only to


-Green with millet after the rains. Thatched huts are granaries.

feed and sustain their society, but also to reverse the degradation of their resources, to stop the encroaching desert, to rebuild the soils and shape a land that can feed its population.
In another office Moussa Saley points to a huge chart fastened on his wall. It is a chart of all the jobs that must be done, the scientific skills that must be marshalled to the task. For some of these jobs, Saley has a leader, someone, like Ouattara, with the training to begin a program, organize research. These are boxed with solid lines. For too many others, there is simply no one qualified. These have dotted outlines.
Moussa Saley says that what he likes about TropSoils is that it has brought to Niger badly needed expertise, in a spirit of collaboration. And even though the work is just beginning, he sees progress, promise. In short, TropSoils has allowed him to draw a few more solid lines. He is glad for that.

Mamadou Ouattara, chief, Division of
Ecological Research, International Institute of Agronomic Research for Niger
Moussa Saley, director general, International Institute of Agronomic Research
for Niger (INRAN)

Women pounding millet in village



farmers, theory help tackle big differences
over small spaces

Spatial variability. It's a term applied to the marked differences in crop yields over relatively small areas of land, a phenomenon that confounds farmers and scientists all across the Sahel. On experimental plots, variability can mask the

Chase in uniformly treated plot

effects of fertilizers or other treatments, rendering the data almost meaningless. While variations in topography, soil types, storm damage and organic matter all contribute to variability in crops, there seems to be another, more problematic factor as well: history. In most cases, researchers simply don't know how the land has been treated.
Bob Chase decided to ask the farmers. He was looking for their opinions about the reasons for variability, matching their responses to data gathered by sampling in the farmers' fields. As expected, the farmers had some excellent, though sometimes mystical, insights into why some soils produced better crops than others. They have even developed their own system of soil classification.
"Farmers identify God, manure, rainfall and wind roughly in that order, as the overriding factors in millet production," Chase says. Manure has been harder to come by recently, especially for farmers who have put more land into cultivation farther away from the villages. Chase found that, in general, the pattern of yield on the farms in his study was as in the table below.
Chase is continuing these studies at ICRISAT's Center for Sahelian Research, collecting samples and analyzing their chemical and physical properties in an attempt to define the factors that cause variability in Sahelian soils. In a companion project at the Texas


Agricultural Research and Extension Center in Lubbock, Robert Lascano and Charles Wendt have been collaborating with Leo Stoonsnijder of the Netherlands on another approach to the variability problem: geostatistics. Geostatistics, a theory developed in the 1960s, can be applied to describe variability for each observation or location in the experiment. Classical statistical methods can account for variation within an experimental unit, but cannot address the distribution of those variations in space. In the field, it is more important to know where to expect drastic dips or peaks in yield than it is to know how often to expect them.
It's a job more of algorithms than
agriculture. But when it's done, the results

will be of use not only in the battle against variability, but also in such things as soil characterization and fertilizer placement. It may even help with some of those "overriding factors in millet production" at least with the manure, the rain, and the wind.

Bob Chase, soil scientist (physics), Texas
A&M University
Robert Lascano, soil scientist (physics),
Texas Agricultural Experiment Station Charles Wendt, soil scientist (physics),
Texas Agricultural Experiment Station Leo Stroonsnijder, soil scientist (physics),
Wageningen Agricultural University, the

Soil Survey

Sizing up the soil's capacity as a bank for nutrients and water is fundamental to improving agriculture both at the research site and on the farm. A soil survey enables scientists to evaluate differences in results when, for example, tests crops are grown on varying soils. And such surveys are essential when the goal is to transfer methods or technology from one site to another.
So when a team of 'ropSoils scientists led by Larry West finished a detailed soil survey of ICRISAT's Center for Sahelian Research, it was a key step, not only for research at the center, but also for future work across the Sahel. The survey, which took about 15 man-weeks for the field work, mapped soil types and their boundaries by first analyzing aerial photographs for patterns of landform and vegetation, then checking and correcting the map with direct sampling. Lab work established physical, chemical and mineralogical properties, and assessed fertility. The soils are strikingly different from most semi-arid soils in the U.S. They are more acid, and have properties that indicate they evolved under wetter conditions, that there was a climatic reversal, and that they began eroding relatively recently in geologic terms.
The soils are developed in sandy parent materials that blanket a previously eroded laterite surface. They are dominantly red to yellowish-red and their texture is loamy fine sand or fine sand. Infiltration rates are favorable for the deeper soils that cover most of the center, but there is likely to be significant runoff during very heavy rains. There is very little chemical suffering; aluminum is present in levels toxic to plants, and soluble nutrients are easily leached.
A 66-page report, Soil Survey of the ICRISAT Sahelian Center, by L.T. West, L.P. Wilding, J.K. Landeck and F.G. Calhoun, is available from Frank Calhoun, Soil and Crop Sciences Department, Texas A&M University, College Station, Texas.

(%3 E

,=1- 0Y O T11


The Sandfighter

rig from West Texas
helps save crops
at research station

To a Texan, it may look peculiar, strapped to a donkey. After all, this rig the sandfighter got its start ripping along behind big, fast diesels on the dunes of West Texas. In Africa, it goes at a plod. But plodding or not, the sandfighter that Bob Chase modified for animal power has

Donkey pulls sandfighter on test plot

shown that it can improve the survival rates of young crops at ICRISAT's Sahelian Research Center near Niamey, Niger.
"In some cases it has meant the difference between having a crop to study and not having one," Chase says.
Used soon after a rain, the sandfighter's tines dig the damp sand into shallow depressions and small, tight clods. The broken surface traps windblown sand and reduces the damage to young crops, which are susceptible to sand blast and burial.
Chase has fitted his experimental sandfighter so that it also punches holes for seed and fertilizer all in one pass. Millet, a staple in the semi-arid regions of Africa, has figured in most of the tests.
Andre Batiano, who has been doing soilfertility studies at the Center, is one of the researchers most familiar with the struggle to stabilize soils on research plots.
"Last year we lost our main experiment because of sand blast," Batiano says. "It is very important to find a solution."
Chase's first series of experiments, in 1983, used a tractor-drawn sandfighter, pulled at three speeds (5, 10 and 15 km/hr.). Clod and hole dimensions were measured in randomly selected sites. At the ICRISAT center, researchers have used the sandfighter to control sand movement over most of the open fields until each was planted, with good results. Findings from these experiments indicated that, under conditions like those at the center, the sandfighter should be pulled more slowly than expected, possibly because of the low percentage of clay in Niger's sandy soils.
"At first we thought tractors would have to be used to achieve the high speeds required for optimum results when sandfighters are used in Texas," Chase says. "Fortunately, the slowest speed we tested worked best, and that means animal traction will be practical. Tractors are still a rarity here, and they're too expensive for most farmers."
Another unexpected result of the tests


Hard winds dry soils, bury young plants in sand

may also prove important. Chase found that sandfought fields showed no standing water or evidence of runoff after a heavy rain, while untreated fields nearby showed both. Because of the critical need to encourage rainwater penetration into Sahelian farm soils, Chase plans to measure the sandfighter's effect on soil moisture and rainfall redistribution in future research. He is also exploring its use in weed control.
So far the sandfighter has been used primarily for experimental work. But Nigerien and American experts are enthusiastic about the potential for its use on Niger's beleaguered farms.
James Lowenthal says that USAID has begun coordinating sandfighter work on extension farms in the region. AID is also looking for ways of engineering and manufacturing the sandfighter locally for use with animals.
"We have all the pieces in place to move from research to on-farm utilization," Lowenthal says.

Bob Chase, soil scientist (physics), Texas
A&M University
Andre Batiano, soil chemist, International
Fertilizer Development Center
James Lowenthal, USAID Mission, Niger

can windbreaks
increase crop yields?

Far down the Majia Valley, great rows of trees stand up to comb the wind, to break its force and reclaim airborne soil. Since the first 13 rows of neem trees (Azadirachta indica) were planted there ten years ago, 160 kilometers of the valley have been crossed with windbreaks. It is one of the Sahel's most ambitious conservation projects.
And by several indications, the work a joint venture of CARE International and USAID, with help from Nigerien foresters
- is already a success, although its economic and social impacts are still being evaluated. In the 100-meter spaces between the double rows, millet yields have risen 22 percent over yields outside the breaks. The trees, for the most part, are surviving. And local farmers, who have found the soils damper and millet greener between the breaks, are convinced that the trees have somehow attracted more rain. So when 'fTropSoils researcher Naraine Persaud arrived this year and began planning his work in the windbreaks, the questions wanting answers were concerned not so much with how to make the windbreaks work, but rather, how do they work and why? What were the key relationships
continued next page

Chase and Batiano in the field


of wind velocity, soil moisture and millet yields? Could the planting pattern be improved? And, perhaps most importantly, could the work be replicated in other windswept valleys of the Sahel? Niger's forest service wants to plant windbreaks in other locations, but such projects are expensive and, without quantitative information about their performance, risky. The first step was to understand the soils. Larry West made a pedological examination of the site, the results of which Persaud is using to space plots along the transects he is laying out both along and across the rows. Persaud is measuring wind-velocity profiles inside and outside the windbreaks, and will analyze these profiles using one-dimensional momentum balance equations, in order to find what factors control the effect of the rows of trees on wind velocity. He will also attempt to quantify the effect of the windbreaks on soil moisture storage and on the yield of millet.
Persaud says that while the research
will have a strong practical application, not only for the CARE/USAID project but for future windbreak plantings across the Sahel, he has also planned an approach that he thinks will make a contribution to methodology in this field. In place of the traditional randomized-block technique for measuring changes along the transects, Persaud will apply new techniques for analyzing interrupted space, a method he feels will greatly improve the validity of the results.
According to Steve Dennison, a forest economist in charge of evaluating the windbreaks project for CARE: "Naraine's study will capture what we call the 'ensemble variations' of the site. It will allow us to tie in crop production, and to look at water loss inside and outside the windbreak."
A TropSoils study closely related to Persaud's work is underway at the Texas A&M University Agricultural Research and Ex-

Naraine Persaud

tension Center in Lubbock, Texas, where Charles Wendt and Robert Lascano are studying the influence of windbreaks on water evaporation from bare soil. Their work, which combines measurements of temperature, weather variables, soil water and evaporation with simulations generated by a numerical model, is expected to expand the theoretical understanding of a windbreak's effect on soil moisture, an understanding essential to the successful design of future windbreaks in the semi-arid tropics.

Naraine Persaud, soil scientist (physics),
Texas A&M University
Larry West, soil scientist (classification),
Texas A&M University
Steve Dennison, forest economist, CARE
International Majia Valley Project Charles Wendt, soil scientist (physics),
Texas Agricultural Experiment Station Robert Lascano, soil scientist (physics)
Texas Agricultural Experiment Station


wilting crop may need
fertilizer, not water

It seems so obvious, watching sorghum or millet wilt in the sun: dry weather's killing the crops. But across the southern Sahel, scientists are contradicting the obvious. They have found that, many times, plants die of thirst even when there is moisture enough in the soil. The real trouble, they say, is nutrients.
Previous research in Mali has shown that, when nutrients are deficient, waterstressed plants don't use all the moisture available to them in the soil. It has been estimated that fertilizers could perhaps increase production five-fold, with only normal rainfall. The threshold seems to fall somewhere near the 300mm mark. Below that amount of annual precipitation, water appears to be the primary limit to crop production.
But in developing nations like Mali, the solution is not as simple as prescribing a bigger dose of phosphorus or nitrogen. Cost is a factor, and so is availability. Clearly, knowing more about the relationship between plants; water-use and nutrients in semi-arid soils would help soilmanagers decide how to make better use of the fertilizers available, and thereby increase food production.
To TropSoils researchers at Texas A&M, who have seen the same connection between infertility and water stress on the Texas High Plains, it seemed a natural subject for international collaboration. Their studies, just getting under way at the Cinzana Experiment Station in Mali and the Texas A&M Agricultural Research and Extension Center in Lubbock, Texas, are designed to benefit both regions. At the site in Mali, Mamadou Simpara, Mamadou Keita, Arthur Onken and Charles Wendt will be testing a number of soil-preparation techniques, including ridg-

ing, subsoiling and tied ridges (furrow dikes), with nitrogen and phosphorus applications as indicated by soil tests. The goal is to compare various combinations of fertilization and soil preparation for their effect on plants' water-use and on sorghum yield.
In Lubbock, Onken, and Wendt will attempt to establish several important principles in the water/nutrients relationship:
Will fertilizer applied to a nutrientpoor zone in the soil profile improve wateruse by plants in that zone?
Where in the soil should nutrients be applied, and at what rates, to promote optimum water-use by plants?
9 Some breeding lines of plants grow larger (produce more dry matter) than others in nutrient-poor soils. Do these varieties also use water more efficiently?
The team will also analyze phosphorus reactions in soil samples from Mali and the Texas High Plains, in order to improve the basis for research transfer. The work is also expected to mesh with water-use and fertility studies underway at ICRISAT's Center for Sahelian Research in Niger.
It's the kind of international effort that, in the beginning, meets with a host of logistical problems and delays. But it also offers the promise of real advances in food production on semi-arid soils the sort of advances that will likely pay off not only on the dry lands of Mali, but on the dry lands of Texas as well.

Mamadou Simpara, soil scientist (physics),
Institute of Rural Economy, Mali
Mamadou Keita, soil scientist (pedology),
Institute of Rural Economy, Mali
Arthur Onken, soil scientist (fertility),
Texas Agricultural Experiment Station,
Lubbock, Texas
Charles Wendt, soil scientist (physics), Texas Agricultural Experiment Station, Lubbock, Texas


Tiger Bush,

Leopard Bush

From the air it is clear why people call this landscape tiger bush, or leopard bush. The Guesselbodi forest, like much of the woodland of the semi-arid tropics of Africa, has receded into stripes and spots of scrub trees and brush surrounded by crusted, barren ground. While scientists are unsure what causes the forest to recede in these two distinctive patterns, one thing is clear: The forest has decreased. After studying aerial photographs, Eric Boudouresque, botanist at the University of Niamey, has estimated that 30 to 50 percent of the Guesselbodi Forest's vegetation has disappeared in the last 30 years.
The region's urgent need for fuel has pressed woodcutters deeper into the forests each year, hacking away with axes, loading the wood onto donkey carts and camels. The topsoil erodes; bare crusts enlarge and harden; water runs off without soaking into the soil. When seeds fall there, they blow away. Nothing grows.
To foresters and officials, it has been one

of Niger's most pressing problems: how do we regenerate the badly depleted forests and reverse desertification? While standard methods, using transplanted seedlings, cultivation and fertilizer, are being tried on a small scale, they are too expensive for widespread use in Niger. The search is on for more practical approaches. Bob Chase may have found one. Chase knew that when there was new growth in the forest it usually began in patches of fallen branches or brush, which trapped seeds and windblown sand, and attracted termites. Termites improve soils in the tropics in some of the ways earthworms do in temperate regions increasing soil porosity and incorporating organic matter.
Chase set up a series of trials to test the natural regeneration of forest vegetation on bare, crusted soils. He laid out 20 plots, half of them protected from grazing animals by a fenced enclosure. On each site he prepared plots with three treatments: cultivation with a hoe to about 10 cm, cultivation plus a mulch of branches, and branches alone. The branches were the wastes of woodcutting nearby. Chase also left a part of each site bare, as a control.
In July of 1983, eight weeks after the

The Guesselbodi Forest of Niger

Chase and mulched plot


A woodcutter on crusted soils in the Guesselbodi Forest

beginning of the rainy season, there was new growth on each of the treated plots. And soil moisture in the treated plots was 2.7 to 3.8 times higher than in the control. The volunteer plant species included a large number of tree seedlings and a legume suitable for the grazing animals that herders bring to the forest. Not surprisingly, the combination of branches and tillage performed best, with an average vegetative cover of 96 percent by September.
But then, after the next dry season, there was a surprising development. Vegetation on the mulch-only plots began to catch up with vegetation on those with both tillage and branches, and the tillage-only plots had lost their plants and formed new crusts. Extensive termite activity in the mulched plots created stable "macropores," which have significantly increased water movement into the soil. These trends seemed to indicate that mulching alone simple, practical and inexpensive might be sufficient treatment to promote natural reforestation. Chase points out that his studies are only beginning, and it is too early to predict what the long-term success of mulching and tillage techniques will be. Amounts of rainfall and new plant growth vary widely year-to-year in the Sahel, and studies have

shown that survival of young trees will depend on their being protected from grazing during the first few years of life. Chase plans to prepare new trials each year, while recording the progress of the plots established earlier.
In the meantime the work offers hope that, in semi-arid forests like Guesselbodi, tomorrow's trees may emerge from the wastes of today's harvest.

Bob Chase, soil scientist (physics), Texas
A&M University





Control Mulch Tillage Both with by Tillage
Branches Hand and Mulch
Average soil water content to 45 cm depth onJuly 7, and biomass of annual plants on Oct. 7, 1983

96.1 -




60 3

40 u)
E 20
20 M


Herders in search of forage in the Guesselbodi Forest

A Team Effort

With the threat of crop failure and hunger ever-present in the Sahel, why the emphasis on forestry? Maurice Tardieu puts it this way: "When you have millet in your granaries, you also have to cook it." In Niger, wood is the primary fuel. Reforestation is also important in the fight against desertification.
TropSoils' research in the Guesselbodi Forest is one part of a cooperative effort by the International Crops Research Institute for the Semi-arid topics (ICRISAT), the National Institute of Agronomic Research for Niger (INRAN), and also by the U.S. Agency for International Development (USAID), which is backing forestry studies in the Sahel.
USAID's Forest Land Use Project has begun an inventory forest resources, set up data banks and a library, established model sites for trying forest-management programs, and is researching techniques for erosion control and "water harvesting" directing runoff so as to irrigate plants. The project has also organized a cooperative of villagers living near the forests, with the goal of training its membership to take over some of the management of grazing, woodcutting and reforestation. The work stresses the use of native species.
John Heermans says TropSoils is helping to fill a gap in the research portion of the project.
"We are working with Bob Chase to help create a research effort in natural forest management," Heermans says. "He's supplying soil physics expertise where it's badly needed."

Bob Chase, soil scientist (physics), Texas A&M University Maurice Tardieu, regional director, International Crop Research Institute for the Semiarid Tropics
John Heermans, forester, Land Use Project


Acid Savannas

Acid savannas are found in the tropics of Latin America, Africa and Asia. These vast, tree-dotted regions are dry four to six months of the year, with rainfall totals ranging from 1200 to 2000 milimeters. The acid savannas offer great promise for the expansion of agriculture in many developing nations. Their soils, while nutrient-poor and acid, are easily worked and potentially very productive. Gone are the days when the world could afford to leave its acid savannas to the ACID SAVANNAS/BRAZIL occasional grazing herd. One of the world's great savanna regions the Cerrado of Brazil, has become an agricultural frontier where, with the proper management, farms are producing dramatic results. The reports that follow examine TropSoils' efforts to learn the principles of soil management that will help put acid savannas to good use around the world. They are set in the Cerrado, a region that has proven to be, as Elmar Wagner puts it, "an excellent laboratory for studying soil constraints."

Primary Collaborators
The Brazilian Agricultural Research Enterprise (EMBRAPA) and its Cerrado Agricultural Research Center (CPAC); Cornell University (lead U.S. institution); North Carolina State University (support U.S. institution).

Principal Investigator
Douglas J. Lathwell, soil scientist (fertility), Cornell University.


Cerrado range, twisted trees

The Cerrado

Flying toward Brasilia on an August night, you will see the bright fires like tiny hoops and crescents of flame, dotting the dark land as far as the horizon. The people are burning the Cerrado. At night, when the air is cool and calm, and the coral snakes are mostly idle, the people go out on the savanna and set fire to the tough brush, hoping that green shoots will spring up in the ash and feed their cattle. They have burned it this way for as long as anyone remembers.
They have burned but few have plowed. There are 112 million hectares of arable land in these rolling savannas called Cerrado. And while Brazil has hungered for food and agricultural exports, most of this land has remained unused. The real problem was not climate or economics or culture. It was soil. The soil was infertile and acid, so saturated with aluminum that it withered most crops. The native vegetation showed the stress of living here: low shrubs and tough grasses; twisted, stunted trees.
But for all its bad chemistry, the land is physically sound. The soils are deep, well drained. Their clay particles are bound by iron into larger grains that behave almost like sand so nicely structured that they

can often be worked the day after a hard rain. The chemistry, scientists suspected, could be corrected.
And even more importantly, this savanna land is vast and available an economic and agricultural frontier and Brazil has needed it. Through the 1960s airplanes shuttled in huge loads of beams and girders, and a gleaming new capital, Brasilia, rose in the heart of the Cerrado. The government offered incentives to farmers, and the farmers came. Now, on new farms in the Federal District around Brasilia, huge, irrigated fields of potatoes, rice and soybeans are greening the landscape. Improved pastures have multiplied the numbers of cattle Cerrado land can feed. A great variety of cash crops, including coffee, citrus and avocado, are showing great promise. Edson Lobato, technical director of the Cerrado Agricultural Research Center (CPAC), sums up the region's potential this way: "With good management, we can grow almost anything we can think of." Agriculture has come to the Cerrado.

Amadeu Tsuno stands in the door of his warehouse, with stacked bags of fertilizer and drums of chemicals behind him, telling a story more and more familiar in Brazil. He arrived in the Federal District six years ago, with his wife, one tractor and


little else besides ambition. He came from San Paulo, in the south, where his family, Japanese immigrants, still runs a small farm.
Today Tsuno farms 293 hectares, with
another 1000 hectares on the way into production. He has planted 220 hectares of potatoes, and they have paid him well. Tsuno has several tractors, buildings, and he hires 150 people from a village nearby to help him harvest potatoes.
Tsuno's story is like those of many other hard-working farmers who have come to settle here from the south of Brazil. His farm is about average; some are as large as 3,000 hectares. Brazil is becoming a power in agriculture, with this great resource, the Cerrado, barely tapped.
What made such gains possible? Fertilizer and lime, which offset severe phosphorus deficiencies and aluminum toxicity. Government incentives. Farmers' hard work. And, perhaps most importantly, knowledge.
Elmar Wagner, who, during his tenure as director of the CPAC, helped bring TropSoils to Brazil, says that science has played the critical role in opening the Cerrado.
"At the beginning of the Seventies, there was no hope to think about crops on the 203 million hectares we call Cerrado," Wagner says. "But now, with improved management, we're seeing excellent results, and we've found that this region will support much more than just this extensive grazing."
Much of the credit for these strides,
Wagner says, belongs to an international collaboration a collaboration whose U.S. representatives included scientists from universities such as North Carolina State and Cornell, with financial support from the U.S. Agency for International Development. The teams showed that while the Cerrado soils require large initial doses of phosphorus and lime, the residual effects of these applications can sustain crops for years the investment does pay off over time.
The collaboration has been so successful that the Brazilian Agricultural Research Enterprise (EMBRAPA), which created the CPAC, continues to support it as a fundamental part of its programs to develop the Cerrado.
But with this boom of agriculture in the Cerrado, there are still serious problems. The region's soils erode rapidly unless properly covered or contained, and gullies

Amadeu Tsuno

several feet deep can appear in a single season. The high cost of chemical fertilizers puts them beyond the reach of many small farmers. Rainfall is irregular. Insects and disease attack crops. And, while mechanization and high-input agriculture are showing dramatic results for now, there are already worries about its long-term impact on the land.
So the research goes on, not only for the sake of Brazil, but for the extensive regions of acid savannas throughout the tropics. And that's where TropSoils' role comes in. The Cerrado region, with its base of soils and agronomy data, its well-equipped laboratories and skilled scientists and technicians, offers the maximum of research results for a minimum investment, under conditions that will encourage the direct transfer of knowledge to other acid savannas around the world.
"This is an excellent laboratory for studying soil constraints," Wagner says, "not only because of the variety of soils and soil conditions, but also for the diversity of agriculture and the social and economic variety."

Edson Lobato, technical director, Cerrado
Agricultural Research Center Amadeu Tsuno, farmer Elmar Wagner, Inter-American Institute for
Cooperation on Agriculture


Brazilian extension agents walking barley field

Reds and Yellows

Walking the Cerrado, the easiest way to tell when the soils change is to watch the termite mounds. When the soils are red, the termite mounds are deep brick-red. When the soils are red-yellow, the mounds are golden amber.
They share common traits, these acid, iron-rich clays. But their differences are important, especially when it comes to water. For some undiscovered reason, some of the red-yellow soils are sometimes wet, with seasonally high water tables. This is both good and bad. On the minus side, high water tables worsen compaction and erosion problems, and have to be tilled with care. On the plus side, high water tables can supply moisture to the root zone during drought. The biggest problem is how to tell which areas are damp and which aren't, before you plow. Mapping such features usually means a lot of digging and boring of holes, using familiar tools of the soil scientist's trade, the shovel and auger. But Ray Bryant and Jamil Macedo are studying ways of using another, very high-flown tool to do the job: the satellite.
Bryant and Macedo are analyzing the photo-like maps generated by remotesensing equipment on the LANDSAT satellite, and correlating the vegetation patterns the images show with data the team collects on the ground. If they can develop a method of accurately predicting areas with high water tables from features on the satellite images, the technique could probably be adapted not only across the

: vast land, vast potential

Cerrado but in other regions as well.
But even if its images prove useful, the satellite won't get the TropSoils team out of the hole-digging business altogether. It is still important to understand the morphology, or soil features, in the wet areas, and to determine such things as when the water table rises and how it fluctuates.
In most clays, persistent dampness
shows up in gray mottles. But curiously, the red-yellow soils of the Cerrado don't show any of the classic signs of poor drainage. Using data and sampling sites from previous research by Walter Couto, Bryant and Macedo extended the study area and recorded the gradations of color, or hues, in their excavation pits.
Their samples are still being analyzed,
but early indications are that color may yet be the key to predicting areas of high water tables. The team has found a shift in hues with depth that seems to correspond to fluctuations in the water table. And there are even some clues to the origin of the soils' yellow tint.
Studying the root zone in the sample
area, Bryant and Macedo found a coating of reduced iron on the roots, perhaps deposited there during wet periods. If the iron coating oxidizes when the zone dries out, the yellow geothite mineral formed by the oxidation would probably mix with the soil, gradually changing its color.

Ray Bryant, soil scientist (pedology), Cornell University
Jamil Macedo, research assistant, Cornell
Walter Couto, soil scientist (management),

Termite mound



Of the three principal nutrients necessary to grow crops nitrogen, potassium and
phosphorus nitrogen is often the costliest and the first to disappear from the field. And, partly because it alters and moves so rapidly in the soil, nitrogen can also be the most difficult nutrient to study and understand.
Drawing on a strong program of nitrogen studies underway at its New York campus, Cornell University began its TropSoils work in Brazil with a number of studies into the role of nitrogen in acid-savannas soils. The studies sought ways of measuring nitrogen in the soil, tracing its movement, and following the transformation of organic nitrogen, which is found in plants and animal manures, as it is mineralized to the inorganic nitrogen available to crops. The team is also seeking a soil "test" for nitrogen. Much of this work focuses on green manures, cover crops that not only help prevent erosion but also manufacture nitrogen for succeeding food crops.

Green Manures
Green manures aren't new. Farmers have known for ages that some cover crops, particularly the legumes, which fix nitrogen from the air in their root nodules, can improve the soil and supply great quantities of nitrogen to succeeding food crops.
But chemical fertilizers fast, efficient, and simpler to manage with machinery have largely replaced green manures on the big farms of industrialized nations. It was a matter of economics: green manures tied up land, labor and capital with crops that yielded little or no income. They came to be known as a luxury few farms could afford.
But the bloom may be coming back on green manures. And part of the reason is the serious scientific attention green manures are getting in developing nations, where chemical nitrogen is often extremely expensive and where farming is not so mechanized. Even in the U.S., green manures and animal manures are attracting more notice.
Douglas Lathwell of Cornell University, principal investigator in TropSoils' acidsavannas program, says that TropSoils is pulling together extensive research from both New York and the savannas of Brazil in order to develop a more basic understanding of green manures and nitrogen management.
"Farmers in New York make extensive use of manure and legumes as sources of nitrogen, and in fact these sources of nitrogen are far more important than fertilizer nitrogen on most farms," Lathwell says. "The research program in New York has evaluated the components of cropping sequences and manure management. In

Brazil and other areas of the tropics, we must develop cropping sequences and evaluate the various sources of organic nitrogen."
Lathwell points out that many important food crops take up great quantities of nitrogen. Corn, for example, requires as much as 250 kg of nitrogen per hectare, about half of which winds up in the grain and is carried out of the field. Lathwell says that unless a farmer corrects the nitrogen deficit with legumes, fertilizer or both, succeeding crop yields will be low.
"Our aim is to furnish the input-output relationships necessary for a rational analysis of the role of organic nitrogen," Lathwell says. "We want to develop practical management and cropping sequences in which organic nitrogen is a feasible means to improve crop yields, possibly as a supplement or substitute for fertilizer nitrogen."
Lathwell says the 'TropSoils projects are attempting to determine several things: the total nitrogen content of crop residues, manures and green manures; the rate at which these organic sources are mineralized into inorganic nitrogen during cropping, and the residual effect of this nitrogen over several crops. Another goal is to match the pattern of mineralization to continued next page


Team threshes and weighs legume before returning it to test plots (above). Walter Bowen and mucuna that survived dry season (right).

the crop's demand, so that the nitrogen is used and not lost.
Walter Bowen and Shaw Reid of Cornell have begun studying methods for managing green-manure nitrogen, using a legume called mucuna, common in the Cerrado region of Brazil and in much of the tropics. The work has only begun, but so far mucuna appears to have several strengths as a green manure, not the least of which is its apparent tolerance of drought.
"The question is, could you harvest your food crop in March (near the end of the rainy season), plant mucuna, and have it survive the dry season to feed the next crop," Bowen says. "This year, even though the dry season began early, the mucuna in our test plots is still green. And the dry season is almost over."
Bowen and Reid's work is a follow-up to a study in New York, where Dave Bouldin, Lathwell and J.K. Jallah found that another common green manure, alfalfa, can supply most if not all of the first succeeding corn crop's nitrogen needs if the alfalfa is properly incorporated into the soil. The team measured the mineralization of organic nitrogen over five years of continuous corn cropping.
During the first cropping season, the green manure sometimes provided even more nitrogen than the corn could use. But each succeeding corn crop received less nitrogen from the alfalfa residues, and, by the fourth year, the nitrogen supply was

almost exhausted. These results, along with methods of analysis developed during the study, will help researchers at the CPAC and elsewhere in the tropics predict the contribution of nitrogen by green manures.
But because supplies of organic nitrogen vary tremendously from field to field and crop to crop, the TropSoils team has turned considerable attention to the development of a key tool in the management of tropical soils: a reliable "soil test" for nitrogen.
Work at several U.S. research stations has tackled various methods for testing soil nitrogen, and the most promising of these seems to be a technique of incubating in the laboratory soil samples containing organic nitrogen. Bouldin, Jorge Quintana, Eric Stoner and Alert Suhet are using samples from Bowen and Reid's work to conduct such incubation studies in Brazil, measuring the rates and amounts of nitrogen mineralization and correlating the measurements with those in the field.
If the method works, scientists will have a way of quickly and effectively screening various green manures and nitrogenmanagement schemes. And, perhaps most importantly, the technique could probably be applied throughout the tropics.
Eric Stoner says that while the major focus of such studies is nitrogen, cover crops such as mucuna have more to offer than nitrogen alone.


'Erosion is a serious problem on these soils," Stoner says. "Green manures can help control erosion and recycle nutrients such as potassium, so that they aren't leached out of the system." Stoner points out that green manures might also help relieve the compaction and hardening of soils cultivated by heavy machinery, since incorporating organic matter tends to improve soil structure and discourage the formation of the hardened layers called pans.
Wenceslau Goedert, the CPAC's research leader in the TropSoils acid-savannas program, says that these physical problems may be more important than researchers first thought.
"It has been a general observation that Cerrado soils have no problem with compaction," Goedert says. "But after four or five years of cultivation we are finding that there is some compaction a hardening." But the big advantage of green manures is still their ability to provide low-cost nitrogen. Stoner says that the high price of urea, the nitrogen fertilizer available to Brazilian farmers, has made green manures very appealing.
"Many places in Brazil where you have a frost-free dry season, and where moisture is the only limitation to growing green manures, people are starting to use green manures, especially the small farmers, who are not going to go out and spend money on urea."

Fertilizer Nitrogen
and Gypsum
While green manures may satisfy some of the acid savannas' appetite for nitrogen, fertilizer nitrogen is still considered necessary to successful large-scale agriculture in the Cerrado. And using it efficiently can mean the difference between a crop that makes money and one that doesn't.
One of the advantages of fertilizer
nitrogen is that it is readily available to plants in a soluble form as soon as it is applied. Organic nitrogen must break down, or mineralize, before a plant can use it.
But this solubility of fertilizer nitrogen is a disadvantage too. Inorganic nitrogen is all too rapidly leached from the root zone, especially during heavy rains, and much of the expensive nitrogen applied to fields never gets used by a crop.

Jeff Wagenet and Elias de Freitas are attempting to adapt and simplify a method for predicting what will happen to fertilizer nitrogen in Cerrado soils. The product of their TropSoils study, a revised numerical model designed to simulate nitrogen movement under various conditions, would take into account such things as nitrogen movement, leaching rates, nitrogen uptake by plants, and the transformation of nitrogen compounds. The goal is to provide a model simple and adaptable enough to help agriculture devise guidelines for the management of nitrogen over a wide range of conditions. Another complication in the management of nitrogen is that not all sources of nitrogen fertilizer are equal. In Brazil, the
continued next page

Geodert at soil pit: roots can go deep when red soils' chemical problems are solved


Eric Stoner, research plot

major source of fertilizer nitrogen is urea, which, according to Wenceslau Goedert, has one significant drawback. "In Brazil, we are almost self-sufficient in nitrogen. But the source is urea, and urea does not contain sulfur, which is important to crops," Goedert says. "We are trying to work with Brazilian industry to improve urea with calcium sulfate gypsum."
But while the gypsum is available, inexpensive and sulfur-rich, little is known about its long-term effect on Cerrado soils. Incorporating the results of studies by Djalma de Souza and Dale Ritchey at the CPAC, Wagenet and de Souza will study the chemical changes in Cerrado soils after gypsum is applied. As in the nitrogenmanagement work, their goal is to develop a model that could predict the fate of gypsum amendments under a range of conditions, on soils outside the study areas.

Douglas Lathwell, soil scientist (fertility),
Cornell University
Walter Bowen, research assistant (soil fertility), Cerrados Agricultural Research
Center and Cornell University
Shaw Reid, soil scientist (management),
Cornell University
Dave Bouldin, soil scientist (fertility), Cornell University
J. K. Jallah, research assistant, Cornell
Jorge Quintana, research assistant, Cornell
Eric Stoner, soil scientist (management),
Cerrados Agricultural Research Center
and Cornell University
Alert R. Suhet, coordinator, resource
utilization program, Cerrados Agricultural Research Center
Wenceslau Goedert, research leader, Cerrados Agricultural Research Center
Jeff Wagenet, soil scientist (physics), Cornell University
Elias de Freitas, soil scientist (physics),
Cerrados Agricultural Research Center Djalma de Souza, soil scientist (management), Cerrados Agricultural Research
Dale Ritchey, soil scientist (management),
Cerrados Agricultural Research Center



Humid Tropics

Tropical America, Africa and Asia all have humid regions, where rainfall is abundant at least 1500 milimeters a year and there is a temperature difference of only a few degrees between the warmest and coolest months. On the humid tropics' more fertile and accessible lands, civilizations have risen and fallen for thousands of years. But on the highly weathered, acid and infertile soils of the vast rain forests, farming and settling are still a tenuous business. Agriculture, if it exists there at all, is likely of the variety called shifting cultivation.
Today, as crowding and hunger drive people into these forests to clear and plant new fields and pastures, there is worldwide alarm over vanishing forests, and the fact that they are sometimes replaced by eroding and badly used soils. And while the myth of the rain forests' role as the "lungs of the world" has been exposed they consume almost as much oxygen as they produce the threat to wildlife, to new sources of germplasm, medicines and synthetics, to the myriad and awesome wonders the forests hold, are real. There is no doubt that man will need land in the humid tropics to raise his food. The task is to choose that land wisely, use it well, and cut no more than we must. Because of the size and importance of this job, TropSoils has two prominent programs underway in the humid tropics, one based in Peru with a secondary site in Brazil, the other based in Indonesia. Together they are striving to put together soilmanagement programs that not only will produce more food, but further the stewardship of tropical forests and lands.

Primary Collaborators, Peru
National Agricultural Research and Extension Institute (INIPA); U.S. Agency for International Development (USAID/Lima); North Carolina State University (NCSU).

Primary Collaborators, Brazil
Brazilian Agricultural Research Enterprise (EMBRAPA); Rockefeller Foundation; NCSU.

Principal Investigators, Peru and Brazil
Pedro Sanchez, soil scientist, NCSU, and John Nicholaides, soil scientist, NCSU.



The Amazon drains a varied expanse of humid tropics, from river valleys'- -1 A . .

i ne iimazon

The conventional wisdom ran something like this: You cannot farm the soils of the humid tropics. They are poor and infertile. When you clear the jungle and break the land, the soil turns to laterite, or brick. So in the early 1970s, when North Carolina State University (NCSU) began testing ways of farming the soils of the Peruvian Amazon, a lot of people thought the whole idea was, well, quixotic. But in the humid tropics, nations were running out of room for people and crops on the most desirable soils. Fertile ground was not available ground. Hunger for new croplands was pushing more and more people into the rain forests, where they would slash and burn a swatch of jungle, exhaust the soil with one or two crops, and move on to clear again. Not that this slashing and burning was so bad in itself. Shifting cultivation had for ages been the earth's dominant agriculture. As long as there was plenty of jungle, and not too many farmers, the jungle had plenty of time to restore itself, and the soil. It was a sociable system, too, based on the notion of choba-choba, or "You help me, I help you." Neighbors pitched in to clear a field. The women brought food. And later, there was often a party, with plenty of chuchuhuasi, a drink distilled from sugar cane and tree barks.

But increasingly, in those bands of jungle accessible by rivers or roads, there were too many people. The fallow between crops was cut short. Sometimes people cleared the land with bulldozers, stripping the topsoil, compacting the land so badly the jungle wouldn't return. For better or worse, people were using jungle lands to grow food crops and graze their animals, and their governments, pressed to provide new food, new commerce, new opportunities, were showing the way. But the cleared land was often too steep or badly managed; there was erosion and neglect. Too many tropical rain forests were being wasted. In 1982, a study by two United Nations agencies estimated that 7.5 million hectares of tropical forests were being cut each year; other estimates have been even higher.
And wasting a rain forest was no small thing. Two-thirds of the world's species of plants and animals live only in the humid tropics, and the great majority have not even been described by science. Many of our medicines and much of the germplasm for new plants and materials come from tropical rain forests, which have been called nature's library of experience. To scientists from NCSU, who had been working with Peru to improve its rice production on the coast, it seemed like a good time to move into the jungle, to test the conventional wisdom.


-to slopes and mountains, where erosion threatens cleared soils

They began with the soils. To anyone who'd walked a farm in eastern North Carolina, the resemblance was remarkable. Same sandy loam surface, same red clay subsoil. Same acid infertility. In fact, these soils might have been found everywhere in the coastal plain of the southeastern U.S., where agriculture was the foundation of many a state's economy. The only real differences were related to latitude and climate: the land was warm all year, and moist.
Some of the NCSU team felt, in the problems of the tropics, echoes of their own state's history. Stan Buol did a little digging in the library.
In 1822, a Professor Mitchell stood before the North Carolina Agriculture Society to condemn a practice that wasted both land and the great eastern forests. He described how generations of American farmers had cut down the forests, quickly exhausted the land's native fertility a fertility built by centuries of decaying leaves and trees and then moved on to clear another forest, abandon yet another barren field.
"This process," Mitchell said, "has been going on till most of the tracts whose situation and soil were most favorable to agriculture, have been converted into old fields, and in our search after fresh ground to open, we are driven to such inferior ridge-land as our ancestors would have

passed by as not worth cultivating."
Professor Mitchell's solution? Manure. Fertilize the land, he said, and you can cultivate your best farmlands continuously, and leave the poor lands and steep slopes to the trees. Across the southeastern U.S., just such a philosophy has made the soils
- which Mitchell acknowledged were of "middling quality" more productive than ever before.
In 1971, NCSU and several Peruvian
agencies, with financial support from the U.S. Agency for International Development, began the Tropical Soils Research Program in Yurimaguas, Peru, to determine whether continuous cultivation of food crops would be possible in the acid, infertile soils of the Amazon Basin. Scientists from NCSU, and Carlos Valverde of Peru's continued next page

Woman with produce in a Peruvian market


National Agricultural Research and Extension Institute (INIPA), reasoned that if the acid, infertile soils of the southeastern U.S. could be cropped year-in, year-out, then perhaps so could similar soils in the Amazon.
They were right. In the May 1982 issue of Science, Pedro Sanchez, Dale Bandy, Hugo Villachica and John Nicholaides reported that, after 21 consecutive crops from a field first planted in 1972, a rotation of rice, corn and soybeans produced an average of 7.8 tons of grain per hectare per year. The key was a careful program of fertilizers and lime, which offset the soil's nutrient deficiency and its high levels of aluminum, and improved its cation exchange capacity its ability to retain nutrients. Unfertilized fields yielded no grain after the third crop. But after a decade of continuous cropping, soil fertility in the fields with adequate fertilization didn't decrease. It improved. And, contrary to the many people's expectations, the soils didn't turn to laterite or brick.
The authors concluded: "We believe that the continuous cropping technology can have a positive ecological impact where it is practiced appropriately, because for every hectare that is cleared and put into such production, many hectares of forest may be spared from the shifting cultivator's ax."
But the team warned against rushing
large tracts into continuous cultivation, for several reasons. While the technology worked well on level or gently sloping land (of which there are some 207 million hectares of well-drained soils in the Amazon Basin), continuous cropping on slopes

1902 photo of shifting cultivation in the Appalachian Mountains of the U.S. (left), and a recent view of the same practice in the Peruvian Amazon

might bring erosion. Insects, weeds and plant diseases, controlled through the rotation of crops during the studies, would probably step up their attack as the technology became more widespread, and it would be necessary to select pestresistant varieties of crops and use controls wisely.
Just as important was the danger of encouraging the clearing and planting of vast areas with crops for which there was no market, or no ready access to a market an ever-present problem in the remote regions of Peru. And, before continuous cultivation would work in the humid tropics, it would have to suit the farmers, bankers and buyers it would have to find its place in the social and economic structure of each region.
As Stan Buol puts it: "I haven't found a farmer yet who will go to the trouble to grow an extra cassava root, and send it off to market, if he knows it won't sell."
And, the research team said, it would be disaster to clear and plant huge fields, if farmers had neither the know-how nor the capital to manage them. Crop varieties and rotations, fertilization schemes, pest controls all must be tested and adapted to the conditions in each farming area. Careful soil testing would help determine which lands could support food crops and which would be better planted in trees or pastures.
What was needed was a middle ground
a bridge between the shifting cultivation of yesterday and the high-input, continuous cultivation foreseen for tomorrow. And that's where much of TropSoils' effort in the humid tropics has been applied building that bridge.

Stan Buol, soil scientist (pedology), North
Carolina State University
Carlos Valverde, soil scientist, formerly of
National Agricultural Research and Extension Institute (INIPA); now at International Services for National Agricultural
Research (ISNAR), Netherlands
Pedro Sanchez, soil scientist (management),
North Carolina State University
Dale Bandy, soil scientist (management),
North Carolina State University
Hugo Villachica, soil scientist (management), National Agrarian University, La
Molina, Peru
John Nicholaides, soil scientist (fertility),
North Carolina State University


Paddy Rice

new farms spring up
from research in Peru

In 1979, it was jungle. Now there's a church, a bar with cold beer, and a thriving frontier community that takes its name from a hero of Peru, Thpac Amaru. Like Amaru, who led a rebellion against the Spanish, these pioneers are doing something revolutionary, but not with weapons. With rice. They are growing paddy rice where it hadn't been grown before, except on research plots: in the Amazon Basin. They came from the Andes, at first only a few families, looking for work. Some found jobs on the research station at Yurimaguas, where they planted rice in research plots Dale Bandy had laid out on fertile, alluvial soil by the Shanusi River, learning the techniques. Soon they crossed the river with some seed, began planting rice of their own. It grew.
"They are getting darn good yields," says Pedro Sanchez. "The farmers are growing two and a half crops a year five crops over two years and averaging five tons of rice per hectare a year. Now there are thirty or forty families, each with twenty or thirty hectares."
Sanchez says this new rice industry, which sprung spontaneously from the research at Yurimaguas, underscores the potential of rice in the Amazon, and also says a lot about the eagerness of many Peruvians to adapt new technologies and improve their standard of living. He points out that most of Peru's rice has in the past been grown on scarce and dry coastal lands, which are costly to irrigate. On the rich soils along the Amazon's rivers, rice grows well with little or no fertilizer.
"This puts rice where ecologically it belongs," Sanchez says, "and would allow you to plant high-value crops on the coast!'
Sanchez says that the collaborative research at Yurimaguas has been undertaken
with the view that soil-management was important not only for the acid, infertile soils, but for the fertile, alluvial soils as well for the entire humid-tropics landscape.
Researchers at Yurimaguas cut and burned riverside jungle, leveled the site and constructed the dikes, paddies and water-distribution systems. Then, in a series of paddyrice studies, Bandy, Jose Benites and Luis Arevalo selected varieties, tested tillage and planting methods, developed weed-control techniques, and matched rice plants' nutrition reqirements with soil fertility.
The result was a set of recommendations for managing paddy rice on the Amazon's alluvial soils, recommendations Peru's farmers and extension workers have quickly put to use.
"This year, Peru was self-sufficient in rice for the first time in many years," Sanchez says, "largely because of the expansion of flooded-rice agriculture into the Amazon."

Dale Bandy, soil scientist (management), North Carolina State University Pedro Sanchez, soil scientist (management), North Carolina State University Jose Benites, soil scientist (technology transfer), North Carolina State University, based
at Yurimaguas
Luis Arevalo, soil scientist (chemistry), North Carolina State University

New fields, Tupac Amaru



keeping yields up,
costs down

Modern agriculture in developed nations is based largely on what has become known as "high-input" technology: mechanized cultivation and large quantities of lime and fertilizer are used to overcome the soil's limitations and produce good yields. But in developing nations, where capital is scarce, markets are ephemeral, and fields are commonly small and tended by hand, high-input agriculture very often won't work.
For this reason, a substantial part of
TropSoils' research in the humid tropics of Peru has been aimed at developing "lowinput" technology that might improve yields and conserve soils without demanding huge capital investments or a drastic break with tradition. At the Yurimaguas Experiment Station, collaborating scientists from NCSU and INIPA began a series of experiments aimed at testing various lowinput techniques in a field cleared by slash and burn:
e Researchers selected cultivars of such

Carrying palm leaves to thatch a roof: where farmers travel by foot and weed by hand, low-input agriculture fits local economies. (c) National Geographic Society

crops as upland rice and cowpeas developed and supplied by the International Institute of Tropical Agriculture (IITA) that combined good yields with tolerance to pests, diseases and high levels of aluminum (see related story, page 32). Many of these cultivars are now being tested by INIPA.
e Experiments with three tillage
treatments no-till, minimum tillage with a hoe, and rotovation with a tractor showed that planting no-till and recycling crop residues produced good yields of rice and cowpeas, required low fertilizer inputs, gave excellent weed control, and helped prevent erosion. The no-till system offered another advantage over mechanical cultivation: a farmer wouldn't have to remove the tree stumps, which could simply rot in place. Techniques gleaned from this work are being introduced to farmers through Peru's extension service.
In another study using both no-till and rotovation, the research team tested a cropping rotation of acid-tolerant upland rice and cowpeas, and compared the crops' responses to several phophorus fertilizers. The results showed that rock phosphate performed very favorably compared to costlier forms of phosphorus, perhaps because the soil's natural acidity helped make more of the phosphorus in the rock available to plants. Again, the no-till approach produced results at least as good as the results on rotovated plots.
New experiments focused on such
problems as weed-control, and on methods of improving the downward movement of


Peanuts (mani) in early trials at Yurimaguas: High-input cropping succeeded, and the next step was to find a middle ground-2'low-input." (c) National Geographic Society

calcium and magnesium in the soils, to offset subsoil acidity and promote root development.
In 1982, Jose Benites and Marco Nurena combined many of these techniques into a central experiment conducted under onfarm conditions. The results were promising. Starting with a field freshly slashed and burned from secondary forest, the team used no-till, returned crop residues to the soil, spaced plants carefully, controlled weeds with herbicides, and applied very small rates of fertilizers. The field produced a total of 10.3 tons of rice and 2.1 tons of cowpeas per hectare during the first two years. The rice yields were about five times what the typical farmer's crop produces without inputs.
"We have found that this low-input system will produce good yields for the first few years, while the stumps rot," Benites says. "It's a practical alternative to the high-input approach."
But the researchers say it is possible that low-input systems may prove to be unstable after the first few years, and might need to be treated as a transition to other systems high-input cultivation, agroforestry or pastures. Another possibility may be to alternate low-input cropping with two or three years of managed fallow.

Benites believes that when this low-input "package" is thoroughly tested, it will have a broad appeal to farmers in Peru and elsewhere in the humid tropics: "It will give them good results with a low investment and little risk."

Jose Benites, soil scientist (technology
transfer), North Carolina State University, based at Yurimaguas
Marco Nurena, agronomist, National
Agricultural Research and Extension Institute (INIPA)


'Unknown African'

new method ranks
aluminum tolerance
of crop cultivars

It's called Africano Desconocido, the
Unknown African. And across the Amazon Basin, this mysterious rice is taking root and thriving. It grows chest-high farmers don't have to stoop to cut the heads. It won't blow over as easily as other tall varieties. It resists disease. And, it can triple the yields of the native species. But the most remarkable thing about the Unknown Afridan is that it does all this in acid soils soils very high in aluminum without demanding lime. "Upland, we can get three tons of rice per hectare, even under acid conditions," says John Nicholaides, who has been studying aluminum-tolerance in this and other crop species. "For the native variety, the average upland yield has been one ton."
Nicholaides says that Melvyn Piha, now at the University of California at Davis, found and named the Unknown African while working with TropSoils projects in Yurimaguas. There Piha "rogued" the line from a test variety whose seeds produced a few atypical plants that grew well on acid soils.

While it will take more research to determine if the African should be classed as a true variety, and to assess its market potential, its credentials don't seem to matter much to the farmers and scientists who are using it successfully all over the Amazon Basin.
"It has a lot of appeal," Nicholaides says. "Extension and research people from southern Peru came to workshops at Yurimaguas and took the seed away with them. We've had requests for seed from Brazil, Indonesia and Sierra Leone, and from a number of research organizations."
Ground limestone, the material commonly used to neutralize acidity and aluminum, is expensive and hard to transport in much of the humid tropics, where aluminum toxicity in plants has been called the number-one constraint to crop production. Many farmers simply can't afford to lime their fields. High levels of aluminum in the soil stunt the plant's growth, restrict its roots, and reduce its tolerance to drought and disease.
Nicholaides says the TropSoils research has been influenced by field trials with farmers around Yurimaguas, who, despite their willingness to adopt new seed types and fertilizers, have been very reluctant to invest in lime.
"It's an important part of our work not just to modify the soil to the crop, but also to fit the crop to the soil to evaluate species and cultivars for their ability to adapt," Nicholaides says.

78% Al Saturation
- pH 4.0

. 0


Tolerant, Tolerant, Low Yield High Yield Potential Potential
Sensitive Sensitive Low Yield High Yield Potential Potential

Melvyn Piha (Above) with Unknown African, the taller rice on his left. (Left) Piha and Nicholaides' method for rating cultivars' aluminum tolerance and yield potential


2 40

20 -

0 0.5 1.0 1.5 2.0 2.5 3.0
Yield on Acid Soil (tons/hectare)






Nicholaides and Piha have developed a technique for evaluating cultivars for aluminum tolerance and yield potential under stress. In addition to rice cultivars, the team has tested cowpeas, peanuts, sweet potatoes, soybeans, and winged beans, with encouraging results for rice and cowpeas. INIPA is using the technique to test rice varieties from all over the world. And the work is showing promise, not only for agriculture in the tropics, but for crops on acid soils in the U.S. as well.
"For all practical purposes, the soils
we're working with in Peru are the same as those on the Coastal Plain of North Carolina," Nicholaides says. This similarity enables the cultivar-testing to go on yearround, an advantage to investigators like Lisa Katz, whose work with Nicholaides showed aluminum tolerance to be a dominant trait in peanuts.
"When we planted the peanuts in Yurimaguas, it was winter here," Nicholaides says. "If we had had to wait for the growing season in North Carolina, we'd have lost six months." Nicholaides says that while the goal of the TropSoils research is to develop aluminum-tolerant crops that can be used to increase food production and help reduce hunger in developing nations in the tropics, other regions stand to gain as well. "In North Carolina, farmers are in many cases not applying the lime the soils re-

quire. Some are leasing land year-to-year, and don't know which fields they'll lease until it is too late for a lime application to do much good. Also, they are hesitant to make the capital investment one that will be good for several years on land that doesn't belong to them."
Nicholaides says that aluminum-tolerant cultivars, when they are developed, could give such farmers better yields at lower costs. Experiments have shown that, using aluminum-tolerant plants, some farmers could get by with less than one-half ton of lime per hectare (about two and one-half acres), applied once every three years.
"Without a doubt, this work will have some application here in North Carolina," Nicholaides says. "It's a good example of how international collaboration helps each partner."

John Nicholaides, soil scientist (fertility),
North Carolina State University Melvyn Piha, soil scientist (fertility),
formerly of North Carolina State University; now at the University of California
at Davis
Lisa Katz, former research assistant (soil
fertility), North Carolina State University; now a soil scientist on a kibbutz in



soils research
improves forage crops,
boosts weight gains

Across the Amazon, pastures seeded in the ash of burned forests thrive for a year or two, exhaust the soil's fertility, then grow very thin. Often the grasses are badly adapted. Cattle overgraze them. Erosion begins. More forests fall. It's a serious problem, not only in Peru, but throughout the Amazon and elsewhere in the humid tropics: a great many hectares feeding too few animals. And the animals are valuable, integral to most of the Amazon's farming systems. People depend on their Brahma cattle for both milk and beef.
To the TropSoils team at Yurimaguas, the need for better pastures was clear, especially since much of the region's terrain hills and acid soils might be better managed as grazing land. But the constraints were just as obvious: Farmers could not afford to spend much money on lime and fertilizers. The forage plants must persist despite heavy grazing, insects, disease and such soil-related problems as acidity and low fertility.

grams/animal/day (four year average)

Unimproved Pastures
inn 200 300 400

Cattle on improved pastures: important for both milk and meat in the tropics
Using forage varieties developed by the International Center for Tropical Agriculture (CIAT), Pedro Sanchez, Dale Bandy, Miguel Ara, Rodolfo Shaus, Kenneth Reategui and staff from several Peruvian institutions developed a broad program of pastures research based at Yurimaguas. The team tested and selected the best-adapted CIAT varieties of acidtolerant grasses and legumes, studied their persistence under grazing, and measured the weight gains of animals grazing them.
"The pasture project has provided very positive results," Sanchez says. "After three years of grazing trials, we have a stable pasture, and live weight gains six to eight times greater than with native pastures, with only minimal fertilization. Even under heavy grazing pressure, there is complete soil cover, eliminating the risk of erosion."
Soil tests showed that, with small applications of fertilizer, soil fertility generally improved in the managed pastures, especially in the topsoil, where the dense network of roots and the effective cycling of nutrients probably helped prevent nutrients from being leached away.

Live weight gains of cattle on improved pastures


Pastures performed well when they were alternately grazed and left idle for 42-day periods, allowing grasses and legumes time to recover. But when the cattle grazed a pasture continuously, a problem arose: legumes began taking over the pasture, apparently because the grasses were more palatable to the cattle than were the legumes, some of which contained concentrations of tannins.
Sanchez points out that in other
ecosystems, the value of legumes is mainly as forage during the dry season, when grasses stop growing. New studies underway at Yurimaguas are designed to find what role legumes should serve in the Amazon, where grasses are green yearround. The researchers also want to know how much nigrogen the legumes transfer to the grasses. Meanwhile, Miguel Ayarza has established studies in NCSU's phytotron a lab specially equipped to simulate the plant's environment to investigate the relationship between nutrients in plants and their forage quality for animals.
While some questions remain about the best mixes of forage plants, the project has demonstrated that better pastures are possible in the humid tropics. And in Peru, these findings are already having an impact. Extension specialists have studied the techniques and are testing them at new locations across Peru. Demonstrations have impressed Peruvian officials with the potential productivity of well-managed pastures. The results are being shared with other countries through the International Center for Tropical Agriculture (CIAT). And most importantly, the research results suggest that, with some know-how and relatively low-cost inputs, the declining quality of pastures in the humid tropics can be reversed.

Pedro Sanchez, soil scientist (management),
North Carolina State University
Dale Bandy, soil scientist (management),
North Carolina State University
Miguel Ayarza, research assistant (pasture
management) North Carolina State
University, based in Yurimaguas
Miguel Ara, research assistant (pasture
management) North Carolina State University, based in Pucallpa, Peru Kenneth Reategui, animal nutritionist,
North Carolina State University, based in
Puerto Bermudez, Peru
Rodolfo Shaus, soil scientist, National Agrarian University, La Molena, Peru

Pedro Sanchez. (c) National Geographic Society



tillage improves compacted soils
In the humid tropics, great swaths of cleared land are abandoned, eroding and infertile, and much of the blame has been laid squarely on the broad blade and wide tracks of the bulldozer. Some Indonesian transmigrant farmers, unable to raise crops on land bulldozers had stripped of topsoil, have preferred to clear new land by hand, rather than risk more damage (see story, page 47).
But while the drawbacks of mechanical land-clearing have been plain to many, there was until recently little hard information about the effects of bulldozers on tropical soils, or about how damaged lands might be reclaimed.

Clearing by slash and burn

Work begun in 1972 at Yurimaguas,
Peru, by Chris Seubert and Pedro Sanchez showed that land-clearing with a bulldozer, used in place of the native slashand-burn method, seriously damaged the soils' chemical and physical properties. During a two-year period following clearing, crop yields on bulldozed soils were only 33 percent of those on soils cleared by cutting and burning when no fertilizer was applied, and 80 percent when the test plots were adequately fertilized.
Sanchez says that negative effects of
bulldozing were caused by a combination of soil compaction, the stripping of topsoil, and the loss of nutrients stored in the cleared vegetation. He adds that the native slash-and-burn method returns some of these nutrients to the soil as ash, does not cause as much compaction, and leaves the topsoil in place.
"After two years of continuous cropping, crop yields on the bulldozed land became so poor that it was abandoned," Sanchez says. No secondary forest developed during the six years following abandonment.
In 1980, eight years after the original clearing, Julio Alegre and Keith Cassel began studies to evaluate ways of reclaiming that land. They found distinct evidence of compaction between 15 and 45 cm below the surface, even eight years after the bulldozing.
Alegre and Cassel found that while tillage of the top 20 cm of soil greatly improved crop yields, results were even better when tillage was combined with chisel-plowing to a depth of 35 cm. Cassel says that chisel-plowing helped break up compacted layers and increase the soil's porosity, improving root development. The team also compared several landclearing methods for their effects on soil physical properties slash-and-burn, bulldozing with a standard blade, and bulldozing with a blade designed to shear off trees at the stump, called the "KG blade" after the initials of its developers. On each cleared site, the team also tested a number of soil-management techniques to see what combinations of clearing and


Alegre in test plots compacted by bulldozers. Plot at left was tilled before planting; plot at right was not

tillage worked best.
None of the clearing methods was faultless. While bulldozed plots showed the greatest decline in soil physical properties, even the plots cleared by slash-and-burn had some compaction and decreases in rainwater infiltration. "During slash-and-burn, even though there's no machinery, there are people walking around, and trees falling, causing compaction," Cassel says. After five harvests, Cassel reports that crop yields show the plots performing in the following order: Slash-and-burn produces the best results. Second is a combination of bulldozing with the KG blade, burning of felled trees, and disking to loosen the soil. Third best is bulldozing with the common straight blade, followed by chisel-plowing.
Peru's extension service has adopted these recommendations and is encouraging farmers in the Amazon to clear by slashand-burn whenever possible, or by use of the KG blade when time and manpower are short. Since mechanical land-clearing will probably continue in some areas, the research team has focused on ways to minimize the damage in such operations.
"I believe that, based on our experience in this study, it may be possible to do some

mechanical clearing followed either by disking or some sort of subsoiling operation," Cassel says. "Now, if we remove a lot of topsoil during that operation, then it won't work we won't get good results."
Cassel stresses that the trouble with a
bulldozer is not so much the machine itself but its misuse.
"The experience of the bulldozer
operator and the procedures he uses are extremely important," he says. "Some operators can clear a site with very minimal damage. The soil's water content is also important, because compaction is worse with wet soil." Julio Alegre explains that, in Peru, many of the bulldozers clearing farm fields have been operated by people trained only in road-building. "There is a lot of machinery working on roads. The farmer rents this equipment and just says, 'Clear here.' The operators are used to driving back and forth until the site is clear, compacting the soil. Compaction is good for a road, but not for a farm field."
While the research results show that food crops can be grown on fields cleared mechanically, Cassel and Sanchez both emphasize the need for good soil management after the clearing, especially with the large tracts necessary to make using heavy equipment economical.
"Mechanical land-clearing commits you to covering large areas of ground with a crop right away," Sanchez says. "In some areas, land has eroded because farmers could not afford to cover everything the bulldozer cleared."

Chris Seubert, former research assistant
(soil fertility), North Carolina State
Pedro Sanchez, soil scientist (management),
North Carolina State University Julio Alegre, research assistant (soil
physics), North Carolina State University Keith Cassel, soil scientist (physics), North
Carolina State University



Most farmers working big fields with heavy machinery don't want trees in the way maybe just a windbreak here and there, a shade tree or two for livestock. There's not much call for a forester when the business is corn or wheat or soybeans. But when the farm field is a few acres slashed and burned out of the Amazon jungle, and the crop is worked my hand, trees are part of the cycle of life the fallow that restores the soil.
To Chuck Davey, a forester who admits to taking some kidding for "trading his hard hat for a straw hat," there is a natural partnership between agriculture and forestry. He says this partnership is especially important in developing nations in the tropics, where shifting cultivation is the rule and many a food crop is sewn around charred stumps, to be fed by the ash from burned limbs and trunks.
Davey points as an example to the
Amazon Basin of Peru. "Traditionally, the farmer would go in and knock the jungle down and burn it," he says. "He would grow one or two crops, usually rice, then abandon the field. The jungle would take over. It used to be, in the days when population pressures weren't great, that it might be twenty years before that field was used again."
Increasing population and the press for new cropland have gradually cut the fallow short, Davey says. "When the fallow is as short as four years, the soil is not

Alley-cropping experiment with Inga hedge, mulch of leaves and cuttings

replenished, and the weed seeds haven't died."
The forest fallow restores soil fertility both because tree roots mine nutrients deep in the soil and recycle them to the topsoil in the form of organic matter decaying leaf litter, branches, ash from burned trunks. And, many jungle trees are legumes, fixing nitrogen in the soil. Another advantage of the forest fallow is that it provides a natural control for pests and crop diseases, which don't have time to establish themselves in a field before the next fallow begins. In the humid tropics, where pests and diseases thrive, that's important.
Davey says that TropSoils' agroforestry work is testing ways to manage this forest fallow in a way that replenishes soil fertility more effectively. Meanwile, the program is also looking for new ways to help Peru harvest more food and fuel from forests and tree crops.

Enriched Fallows
Davey and Larry Szott have begun studying simple, low-cost ways of improving the jungle fallow. In one study, the team is creating what it calls an "enriched fallow" by scattering seeds of such trees as the Inga in rice fields just before harvest. The Inga, common in the region, grows rapidly, and is a nitrogen-fixer. Sewn in the rice crop, favored species would get a head start on the rest of the jungle vegetation, which, in an unmanaged fallow, would typically include only a few nitrogen-fixers per hectare. The team is monitoring changes in soil properties in the enriched fallows, comparing them to the soilbuilding progression in natural fallows.
"It's a low-input approach," Davey says. "We're not trying to keep the jungle out, we're just trying to see if we can improve the jungle's ability to rebuild the soil."
Davey says that while the techniques
may indeed improve the fallow, the fallow period will probably still need to be at least four years. "We don't think the fallow period can be less than that," he says. "But, with this kind of management, the succeeding crop will probably be better."

Alley Cropping
Despite their advantages, jungle fallows still leave at least 80 percent of the available farmland in jungle, a practice that may not survive the increasing pressure for


new cropland. One possible alternative, known as alley cropping, simulates the effect of fallow without taking whole fields out of production.
Davey says the technique works this way: The farmer grows his crops in the alleys between rows of woody vegetation. As he works the crop he slashes branches from the row of trees and bushes with his machete, and spreads them in the field as a mulch. This adds organic matter and nutrients to the soil, and helps reduce soil temperatures. If some of the woody plants are nitrogen-fixing, they may also supply some nitrogen to the crops nearby.
"Using this approach might mean a more permanent agriculture, and a more stable population," Davey says.
Alley cropping, though successful on
some fertile soils elsewhere in the tropics, has not been extensively tested on the infertile, acid soils of the humid tropics. In their trials at Yurimaguas, the research team has planted rows of bushes and trees, using a series of spacings to see which alley size works best. They are also testing various combinations of fertilizers, food crops and tree and bush species, while monitoring the chemical and physical properties of the soil.
Like the managed fallows, alley-cropping represents an "intermediate stage" between shifting cultivation and high-input, continuous cultivation.

Tree Crops
Davey says it's hard to interest Peruvians around Yurimaguas, people who spend a good deal of time hacking away at the jungle, in the idea of planting or tending trees for timber.
"You try and talk about trees to most of these people, they're not interested," Davey said. "There's not a large market for lumber and firewood, yet. But if you talk about something like tree fruits or living fences, they might pay attention."
Farmers have told Davey that fences must be "horse high, bull strong and pig tight" a tall order in the tropics, where a fence post rots almost as soon as it's set in the ground. Davey has been studying ways to plant and maintain rows of trees, fence posts that won't rot or need replacement. But some of the most interesting agroforestry at Yurimaguas may turn out to be work with the peach palm, an Amazonian tree Davey and Jorge Perez have been studying. The peach palm, which produces

a fruit high in protein, vitamins and oils, is valued in Peru, where most of the fruit is gathered wild. Perez has collected seed from 1000 trees and is testing them to see if some specimens out-perform others. Research is also underway to determine the peach palm's fertilizer requirements.
The peach palm, an important cash crop in Costa Rica, has adapted well to culture, and could prove to be a profitable crop that is, if a small controversy can be resolved. There are two main varieties of the peach palm. One variety's trunk bristles with long, sharp spines. The other's trunk is smooth.
"Right now, there's an argument over
which is better, the ones with spines or the ones without," Davey says. "It's much easier for a farmer to pick fruit from the spineless tree he doesn't get stuck. But the rats and monkeys can also climb the spineless trees and get the crop first."
Davey says that all these TropSoils
agroforestry projects managed fallows, alley cropping, tree crops are aimed at a pair of general goals: helping farmers grow food, and helping reduce deforestation in the Amazon Basin.
"This kind of agroforestry is intended for developing countries where the trees can be of use in the farming system," he says.

Chuck Davey, forestry scientist, North
Carolina State University
Larry Szott, research assistant (forest soils),
North Carolina State University
Jorge Perez, forester, National Agricultural
Research and Extension Institute (INIPA)

2.0 -

___ Diarmeter

50 100

Peach palm, with spines

12.5 10.0

7.5 ,

5.0 E 2.5 0


Nitrogen (kg/hectare)

Response of 18-month-old peach palms to nitrogen


A New Way of Sorting Soils

If you gather food in the woods and fields, you'll want to know a little something about berries. Now, you might not have the time or inclination to learn all the taxonomic tidbits you would need in order to identify each and every berry by species. Probably you only want to know how to tell which ones are edible, and which aren't.
If you farm in the tropics, you've got a similar problem with soils. You don't want to know all the thousands of combinations of soil characteristics that comprise soil taxonomy. You only want to know where to plant your crops and what to feed them.
In 1973, Stan Buol and his colleagues in the Soil Science Department at NCSU proposed a system of grouping soils, a system they thought might help scientists, extension workers and farmers plan and manage their fields. The system was called Fertility Capability Classification FCC.
The idea of the FCC was to identify the physical, mineralogical and chemical properties in the upper part of the soil profile that were important to agriculture, and to use them in classifying soils. Because the properties used to define a given group of soils were limited, the FCC greatly reduced the number of soil types a planner must consider in order to make an informed decision. A computer program made the system even easier. Keying in lab data and answers to simple questions about traits in a given field, the operator could quickly classify the soil and get a run-down on, for example, its expected fertilizer requirements. After years of testing and fine-tuning, the FCC is rapidly becoming an institution in agronomic research in the tropics, and not only on TropSoils sites. Tests in Indonesia showed that the FCC produced more useful groupings for agronomic purposes than either traditional soil taxonomy or the local classification systems. Thailand has evaluated its soil classification groups using the FCC. It is an official system in Venezuela. Taiwan used the system to select research sites, and also showed how the response of paddy rice to nitrogen fertilizer clearly varied from one FCC soil type to another. An adaptation of the FCC to paddy-rice soils in the Philippines was successful, leading to new tests in 14 Asian countries. Copies of the computer program have been supplied to Malaysia and to The Peoples Republic of China. And, criteria for salinity have enabled planners to evaluate land in California, using the system. Buol says the FCC will be an important tool in applying new soil-management technologies and in transferring research results from one region to another, around the world. But he adds that no classification system replaces basic soil taxonomy, soiltesting and research.
"Classification is not science," he says. "Classification follows science.'
Buol also says that no matter how useful the FCC proves to be, it will continue to need updating and refinement.
"As long as technology keeps changing new cultivation and management practices
- the job of classification is never done," he says. "As our use of the resource changes, the way we describe or interpret that resource changes, too.'

Stan Buol, soil scientist (pedology), North Carolina State University

'As our use of the resource changes,
the way we describe or interpret
that resource changes, too"
-Stan Buol



transplanted research
adapts to new locale

Deep in the Amazon of Brazil, on a flat plateau shaded by great, broad-crowned trees that form a canopy thirty meters above the ground, men walk easily through the open understory, carrying their machetes. They build a buttress at the base of a huge old tree and climb to the top of what will be the stump, the point where the trunk begins its long, straight reach for the sky. The trunk is three meters thick. Through the day, they hack away at the tree, chips flying, until the trunk topples, the branches crash down, and the sun floods in through a gaping hole in the roof of the forest.
In the state of Amazonas, people have always lived near the water, dependent on rivers for transportation and trade. Now, with the riversides crowded, they are pushing into the primary forests, cutting and burning, clearing new land. The government has looked into the future, has seen that some of its vast forests must fall. The forests have little commercial value, they say. The people need commerce. Oil palms and rubber trees will raise the standard of living, develop a needed resource.
But the government has also recognized that, unless the cutting, clearing and planting are managed with care, the land may be ruined, the resource squandered.
"The government sees the Amazon region as a kind of escape valve for Brazil," says Jot Smyth, a TropSoils researcher based at EMBRAPA's (the Brazilian Agricultural Research Enterprise) research station at Manaus. "Brazil wants to know about the soils and how to manage them, to work out some solutions before the pressure's on, before the problems arise. That's the reason for the research center at Manaus."
Manaus was a natural place to test the results from research at Yurimaguas, Peru, on a site in the humid tropics with distinct differences in soils and climate. Both geographically and morphologically, the soils at Manaus represented something of a middle ground between the iron-rich clays (called Oxisols) of the Cerrado and the loamy soils (often Ultisols) common in the Amazon of Peru. The climate is also intermediate: Manaus has a few "dry"

Robust corn of alluvial soils at Manaus: the exception, where much of the land is acid and infertile

months with less rainfall than at Yurimaguas, yet lacks the strong dry season of the Cerrado.
"Our work at Manaus is largely adaptation," Smyth says. "We wanted to know what in the Yurimaguas technology needed to be modified for these conditions. For example, we knew pretty well, from the Yurimaguas work, more or less what fertilizer inputs we'd need. What we didn't know was how much and when the amounts and times of applications are different."
Smyth and Joaquim Bastos have found both similarities and differences between the two sites. Phosphorus fertilizers are often unnecessary at Yurimaguas until the third crop after slash-and-burn. At Manaus, a phosphorus application to the first crop doubles yields, probably because the clays there require much greater doses of fertilizer to satisfy their phosphorus-fixing capacity, though not as much as is required for the Cerrados. On the other hand, potassium fertilization is similar for both Yurimaguas and Manaus.
Smyth says farming practices in the two regions are similar, too, with one exception: "At Manaus they weed with hoes," he says. "In Yurimaguas, the farmer squats and cuts the weeds with his machete."
Some of the research is looking at ways to intercrop corn, rice, cowpeas and other food crops with young oil palms and rubber trees, in order to receive some benefit from the land during the years before the plantations begin to produce income. This is especially important in Amazonas, continued next page


Smyth says, where half the food is imported from outside the state.
While the Manaus extrapolation work
bridges the technologies developing for the Cerrado and those for the humid tropics of Peru, Smyth says it is also developing knowledge that will be useful at the "primary" research sites. He cites as an example studies at Manaus that focus on guarana. Guarana, a bush that produces a coffee-like bean very rich in caffein, has become an important cash crop in the Brazilian Amazon. Its seeds are ground to produce a national soft drink, also called guarana, a drink so popular its supply cannot meet demand.
But despite its importance as a cash crop, the guarana crop and its growth habits have been poorly understood. Smyth, Bastos and Jose Correa have been studying three promising clones of guarana, measur-

ing their response to fertilizers and describing what Smyth calls the "soil-plant nutrition relationships." Results of the studies, Smyth says, will be useful in testing the guarana elsewhere in the humid tropics.
"It's not just a matter of taking information from a primary site and applying it at the extrapolation site," Smyth says. "Many of the studies we're doing at Manaus will support those in Indonesia and Peru. It's a two-way street."

Jot Smyth, soil scientist (fertility), North
Carolina State University
Joaquim Bastos, crop scientist, Brazilian
Agricultural Research Enterprise
Joe Correa, crop scientist, Brazilian Agricultural Research Enterprise

After years of research in Peru, collaborating scientists have tested and adopted a range of soil-management options for the Amazon Basin landscape (from Sanchez)


- -

Sticking With It

commitment pays off
for people, research

Julio Alegre was born in the mountains of Peru, the son of a potato farmer who, like so many others, left the mountains for the capital city, Lima. Soon after this report is published, if all goes as expected, he will be Dr. Julio Alegre. He will have earned a Ph.D. in soil science from NCSU. Alegre is quick to say what this means to him: "In Peru, we don't have enough people trained at this level. If I can go back and train, for example, six M.S. students, then those six may train twenty B.S. students, and those twenty may go out and train a hundred farmers. It is important that while we are doing this research, we are also building something for the future." With support from TropSoils and its predecessor programs, Alegre spent four years conducting soil-physics research (see story on page 36), improving his English, and pursuing his degree. In 1982, he came to NCSU to complete his course work and compile the results of his research. He says that the research at Yurimaguas is vital, not only for people like him, who are directly involved, but for Peru as a whole.
"About half the population of Peru is around Lima, on the coast," he says. "There is crowding. Sometimes there is drought. We need more room. In the highlands, there are big problems. They have erosion, and there are only a few vallies suitable for production. Peru has decided we must develop the jungle. There is enough area, and rainfall all year. Before, people were scared of the jungle; it was taboo. Now they are beginning to come. Initially it's hard for them, changing their customs, learning to eat new foods.
"At first, production was very poor in these new areas. We didn't have any research. Now, we can recommend fertilizers, plant densities very simple kinds of management that farmers can use to double their production." Alegre feels strongly that such gains would have been unlikely without collaboration, both among scientists of different nations, and among scientists and farmers.

"What these farmers feel is very important. If scientists don't have any knowledge of what the farmer feels, it's much more difficult to succeed. It's not just a matter of bringing in new technology. We need the interchange of information." Alegre points out that the success of the work at Yurimaguas has come largely because of the long-term commitment to research there, a commitment that has given Peru the basis for a solid program of research and extension in the Amazon and him the consistent support he needed to complete his education. Pedro Sanchez says that this continuity of effort more than a decade of sustained research has many such benefits.
"There are some very good technical reasons for maintaining a long-term program," he says. "The fact that we have answered some of these basic questions has allowed us to move now from 'what' to 'why.' We know, in many cases, what works, but we need to know why, so that we can better transfer these results to other areas. And there is a lot of concern about sustained food production. We want to know if these systems will be stable over the long term. The only way I know to find out is to stay there and farm it for a long time."
The continuity has helped the collaborative research program to earn credibility in the country, and establish continued next page


what Sanchez terms "a measure of confidence" in the developing technology. He says the program has been able to link successfully with many agencies and research networks and compound its effectiveness, largely because of this credibility factor. "The soil-management practices for the humid tropics that we have developed in Yurimaguas are being tested elsewhere, in Sitiung (Indonesia), for example. So far they have been found to be valid." One measure of the worth of a collaborative research program is the degree to which the partner nation supports, or buys into, the program. Victor Palma points out that Peru's contribution to the program, through INIPA, has risen steadily.
"Peru is planning to invest over $800,000 dollars in the TropSoils program (over the next five years). Under Peru's present economic conditions, this is really an enormous investment."
He summarizes the program's impact

The Yurimaguas research station has grown steadily during the decade of collaboration among Peruvian institutions, NCSU and USAID, becoming one of Peru's best

this way: "TropSoils' soil-management recommendations are being tested and put into extension packages throughout the Peruvian jungle. The TropSoils program is the main database for this region. And, since 1972, the program has helped INIPA in establishing and equipping the research at station Yurimaguas, perhaps the best in the country, and a training center in Yurimaguas for research and extension workers in the Peruvian jungle. We consider this project not only oriented to Peru's agricultural development, but it is also international in its objectives."

Julio Alegre, research assistant (soil
physics), North Carolina State University Pedro Sanchez, soil scientist (management),
North Carolina State University
Victor Palma, chief, National Agricultural
Research and Extension Institute (INIPA)



Primary Collaborators, Indonesia
Agency for Agricultural Research and Development (AARD) and its constituent Center for Soils Research and Center for Food Crops Research; USAID; University of Hawaii and NCSU.

Principal Investigator, Indonesia
Goro Uehara, soil scientist, University of Hawaii.


The Settlements

From Java and Bali they've come by the thousands, making new homes under a blazing sun and the tin roofs of the Sitiung transmigration settlements, set up on West Sumatra by the Indonesian government. Often their neighbors were strangers, of different ethnic groups, speaking different languages. The land was a stranger as well
- rolling, acid, and infertile.
But the people, most of whom had never farmed land of their own, were pleased with their plots, their allotments of seed and fertilizer, their dirt-floored wooden houses. Soon the trees were falling, the land cleared and planted with rice, cassava, peanuts and soybeans.
The rain, though it fell in abundance, fell furiously and much of it ran off. Even in the humid tropics, soils can be dry. Plants wilted. Some crops failed. Slopes eroded. Too much land was abandoned, barren.
Soon many farmers saw that bulldozers were spoiling their new land, stripping the topsoil, opening the slopes to erosion. But when they tried to clear the forests other ways with slash and burn the root mat was so thick they often couldn't cultivate with hoes, the method many had used in Java.
To the Indonesian government and scientists, and their U.S. collaborators in research, the challenge was clear: Develop better land-clearing methods that conserve soil and prevent erosion. Find ways to reclaim lands too eroded and infertile to support crops. Fashion new cropping systems suited to the soils, climate and economy of the region. And, make sure

Javanese transmigrant couple (left) working among roots and stumps of new land at Sitiung: a long way from the neatly terraced and fertile fields of Java (above)

that these new methods and tools will be ones the farmers can use, that they will work not only in Sitiung but in other such settlements as well in Kalimantan, in Sulawesi and in other parts of Sumatra. The first step was to build a team: Indonesia committed 15 scientists and technicians, vehicles, some offices and living quarters, and funds. The University of Hawaii and North Carolina State University assigned three senior scientists and two graduate students, some support personnel and equipment. The universities offered the aid of campus-based scientists in the U.S. The formal agreements between governments were signed, U.S. AID supplied TropSoils support, and, in 1983, the work got underway.
It was a new program, just breaking ground. But it was one designed to fill an immediate need, and its objectives were clear. Gordon Tsuji puts it this way: "The primary goal of the project is to uncover principles that will enable resource-poor farmers who cultivate the fragile and impoverished soils of the humid tropics to adopt soil-management practices that will increase family income and farm productivity and at the same time preserve land quality for future generations."
The first steps? Test techniques that have worked in Yurimaguas and Manaus, and some new ones as well. Begin providing Indonesia with facts it needs to strengthen its soil-management and agriculturalassistance programs. Get to know two of Sumatra's most important resources: the soils, and the people who use them.

Gordon Tsuji, project manager, University
of Hawaii


Life in Sitiung

Indonesia's transmigration program has been intended to help relieve overcrowding, especially on the island of Java, while at the same time developing new farmland in undeveloped areas such as West Sumatra. But such mass relocations are notoriously difficult, especially when farmers must sink new roots into infertile, unfamiliar soils.
It seemed to the TropSoils team that
work in the six settlements in the Sitiung area of West Sumatra would be better if it blended soil science with social science. So when researchers from the University of Hawaii and North Carolina State University arrived in the settlements, they included Carol Colfer, an anthropologist.
Colfer, whose previous work included a development-oriented study of shifting cultivation on the island of Kalimantan, says her role in the TropSoils work is to help researchers tailor their soilmanagement studies to the needs of the settlers of Sitiung. But she's also looking for ways to "sensitize" soil scientists to social factors important to their work, not only in Indonesia, but in other developing nations as well.
A time-allocation study, conducted with collaborating staff from the Indonesian Center for Soil Research, used a randomized schedule of visits to settlers' homes over a year's time. The team collected data on the division of labor by age and sex, and on the frequency and seasonal variations of such activities as rice cultivation, home gardening, labor for wages and home industries. The data are being entered into a computer, Colfer says, so that any researcher can tap the data base.
"The team has already asked how often people must search for grass for their cattle and goats; how farm labor is divided between the sexes; the incidence of offfarm employment, and the monthly variation in the productive activities of adults." Here is some of what Colfer and her assistants learned about life in Situing:

The Settlements
The 100,000 hectare transmigration site is home for some 10,000 transmigrant families and 1500 indigenous families. The first large group of settlers arrived in 1976, and each family received a modest home,
1.25 hectares of land and a year's supply of

food, fuel, seed and fertilizer. Six areas have been settled, all called Sitiung after a nearby village of that name. Each major settlement is designated by Roman numeral, and is subdivided into blocks.
Families in the six Sitiung settlements have been assigned small, dirt-floored, wooden houses with two bedrooms, a kitchen and living room. The houses aren't pretty, says Carol Colfer, but most of the settlers are happy with them. "Most of these people came from huts of woven bamboo, which are very picturesque but leak and have no status. They are for the poorest of the poor."

"The vast majority are Muslim," Colfer says, "but they're not particularly fussy. They will eat pork, and the women aren't as modest as some Muslims."
There are three primary ethnic groups in the settlements, each with its own language.
Family Life
The day begins at 5 a.m. Women nurse their babies, haul water from the well, and continued next page

East Javanese children in Sitiung V, after a performance in honor of their village's new leader


make a wafer of fermented soybeans. The workday is broken for rest during the hottest part of the afternoon. The settlers have been surprisingly open to government birth-control programs, but there are still a great many children in most households.
"The men are in charge of the farm work and the women are in charge of the children, housework and cooking," Colfer says. "But there's a good bit of overlap. Men cook and care for children; women farm; children help. Often we see kids of seven or eight walking around with twoyear-olds strapped to their hips. Child care is viewed as a community job. Children have a lot of freedom." Most households have at least one member working outside the home. Some women work as maids.
Role of Women
Colfer was surprised to find that women took a lesser role than she expected in agriculture and decision-making, and were less involved in government-sponsored extension programs. She says such programs could help women improve some of the activities they are most involved in, such as food-processing and marketing. And, because women work very hard in home gardens, Colfer sees a potential for them to begin raising the same high-value crops requiring intensive cultivation.
Diet and Income
Nutritionists interviewed people from 80 families in their homes, twice. Preliminary results of the study include:
Incomes range from about $8 to $200 a month.

Nutrition was marginally adequate.
A lack of variety in the diet suggested a possible cause of nutritional deficiencies.
e The people ate virtually no meat.
The nutritionists conducting the survey advised the research team not to concentrate on high-value crops: the people would sell them rather than eat the better foods themselves.
Even so, most of the families felt better off than before. "Some of them had had lifestyles which involved simply scavaging, wandering around with everything they owned on their backs," Colfer says.
Skills and Education
"These are very low by Indonesian standards," Colfer says. "In one group of seventy-five families, two people had junior-high educations."
Most of the people had been working in agriculture, but under vastly different conditions of soils and climate. On the plus side, Colfer says that most are very open to learning about new agricultural methods.
"In working with these farmers we've tried to emphasize that we're all experimenting together," Colfer says. "One big problem for everyone is predicting rainfall, and timing the planting with regard to the beginning of the rainy season. In Java, they were used to distinct seasons, wet and dry. In Sitiung, there's never a clear demarcation."
Information Exchange
The settlers live among strangers, and speak several different languages, and yet news about seed, crops and cultivation somehow spreads through the settlements.
"It's informal," Colfer says. "Information

Research team in the field, from left to right: Djoko Santoso, Mike Wade, John Thompson, Carol Colfer and Karim Makarim


Settler in her kitchen

seems to travel through families and ethnic groups through kinship channels. I was surprised to find at Sitiung Five how much new people talk to those who have been there a long time. They find that different areas have different strengths. Sitiung Four is a source of seeds for rice and peanuts, and Sitiungs One and Two have knowledge about soils."
Indonesian extension agents are young and energetic, Colfer says, and are providing useful information despite some resistance from farmers, who sometimes suspect them of "controlling" checking up on transmigration violators.

"Javanese see rice as absolutely essential
- the base," Colfer says. "But rice doesn't grow well here. Casava does, and there is a factory nearby that buys cassava and makes flour. But after transportation costs and other expenses, you get about one and a half cents per kilo, maybe less. So people aren't jumping to grow cassava. Also, their general perception is that cassava depletes the soil."
The government is promoting soybeans, and peanuts are planted as well. Indiginous families have done well with tree crops,

but for settlers the problem is waiting for tree crops to produce.

Ann Wade, a sociologist and wife of Mike Wade, soil scientist from North Carolina State University, conducted a survey in Sitiung for the Small Ruminants Collaborative Research Support Program. The survey showed that a lot of settlers liked owning cattle, even though the cattle were too valuable to eat. A good cow is worth about $400. Settlers view them as a source of milk and fertilizer, and as a kind of insurance against hard times. A government program entrusts the care of a cow to a family. The family must give away the first calf, but may keep the second. The family spends about an hour a day gathering grass for its cow.
"People also keep goats, but they save them for parties," Colfer says.
For the most part, it's been good. Colfer says there were some early problems because people thought the researchers were officials checking-up on them. "But once they understood what we wanted, they answered our questions quite readily," she says.
If anything, the settlers tried to be too
kind and helpful. "Sometimes they will say what they think you want to hear, just to be polite," Colfer says. "That's why, in Indonesia, it's important to hang out a while, observe, and listen to what they say to each other."
Colfer says work with the farming
families has already had considerable influence on research design. Local farmers suggested that the team replace a legume grown to improve soils in one of the experimental cropping rotations with mucuna. The farmers pointed out that mucuna, which is also a legume, would both improve the soil and provide a food.
"Throughout the process we have altered things to make them acceptable to the farmers without really negatively influencing the research," Colfer says. Colfer has lived with farm families in one of the settlements at Sitiung, doing what she calls participant observation to learn about such things as family life and customs, work habits, information exchange and farming practices. These obsercontinued next page


Clearing logs from new field

vations, along with surveys and experimental collaboration projects with groups of farmers, have impressed her with the settlers' ability to contribute to soil-science activities.
She found that, even though many of the transmigrants were accustomed to the fertile, volcanic soils of Java, they have quickly learned that their new land needs special care. Most of the fertility lies in the topsoil, and the subsoil is practically sterile. In the beginning, poor land-clearing practices led to severe erosion and some areas were abandoned.
"The thing that has impressed me most so far is the amount of knowledge people seem to have about soil," Colfer says. "They're worried about erosion. They're aware of the importance of terracing; they understand that bulldozers can be very bad for the soil. In Sitiung Five, for example, they had the opportunity to have bulldozers free, to clear all those trees off their land, and most of them would not take it because they didn't want their land ruined. They'd rather go through the terribly hard labor of clearing it by hand." Colfer says she has been surprised by the optimism with which the families view their transmigration. Despite such problems as low incomes and marginal nutrition, most of them, she's found, feel they are better off than they were, largely because they have been provided with land and shelter. On Java, they typically had tiny parcels of less than a hectare, or no land at all.
"Their biggest complaint is that it's too quiet here," she says. "There's a shortage of singing and dancing."

Note on the research team
Carol Colfer has led TropSoils research into social structure and farming systems in the settlements, but she hasn't been working alone. A large team of scientists and assistants has contributed to the efforts described on these pages. The team includes:

Atin Kurdiana, field assistant, Center for Soil Research
Suwandi, field assistant, Center for Soil Research
Edi Santoso, field staff, Center for Soil Research
Heryadi, field staff, Center for Soil Research Sarmi, junior high school student, Sitiung I Veronica Kasmini, field assistant, Center for Soil Research
Barbara Chapman, project director, Bogor Street Foods Project
Liek Irianti, Institute Pertanian Bogor Nutrition Department
Bartholomeus Wied, Institute Pertanian Bogor Nutrition Department
Harry Apriadji, Institute Pertanian Bogor Nutrition Department Ann Wade, sociologist

Woman holds rice she harvested selectively from her yard



Barren Land

Seeing the damage, it almost seems hopeless. Topsoil stripped or eroded. A hard, compacted surface. Bare subsoil, acid and infertile, where not even cassava will grow. Bulldozers, hasty clearing, poor erosion control they've all contributed to the degradation of large areas of land in the settlement area.
Indonesia wants to stop the degradation and reclaim abandoned land. It's a big job. How does man go about rebuilding soils it took nature ages to build?
The TropSoils team of Karim Makarim, Keith Cassel, John Nicholaides, Mike Wade and Gunawan set up an on-farm trials to evaluate various chemical and physical treatments for an unproductive hillside in Sitiung II. In a cropping cycle of rice, soybeans and mung beans, the team tested low and high rates of lime and fertilizers, along with various tillage methods, against a control.
While rice production improved from almost nothing on the control plots to moderate yields with the addition of lime and fertilizers, the best results came when green manures were added to the soilamendment recipe. After turning under a legume cover crop, the team found substantial gains in rice yields at all three levels of fertility.
In fact, low rates of lime and fertilizers, coupled with green manure, produced considerably more rice than high fertilizer rates without green manures 2.71 tons per hectare vs. 2.15 tons. Adding organic matter to the soil also improved the soil's physical properties, though not as markedly as with deep tillage. (For a related story on green manures, see page 21.)
Unlike the rice, which yielded only
about 17 percent better with high fertilization rates than with moderate rates, the soybean crop tripled its yields with the larger doses. The team found, also, that fertilization increased the residual level of phosphorus in the soil, and the heavy liming completely eliminated exchangeable aluminum both factors in the increased yields. These residual effects will benefit succeeding crops.
While these food-crop tests continue, John Thompson and D.S. Gunawan began working on the land-reclamation problem from a different angle: forage. As Carol

Bulldozed land in Sitiung

Colfer found in her time-allocation studies, Sitiung families spend considerable time every day cutting forage from roadsides and abandoned fields for their animals. The feed is poor, the time costly. The TropSoils team, working in association with scientists from the Center for International Agriculture in the Tropics (CIAT) and the Center for Soils Research, gathered and tested a number of grasses and legumes to see which would perform best on degraded soils.
So far, four grasses and one legume show promise as cover crops that can feed livestock and reclaim eroded land, and there will likely be more.
The results of these studies will help Indonesia mend some of the damage to its soils, not only in Sitiung, but in other transmigration sites as well. And, as the teams point out, the information will also be welcome in the U.S., since it will shed some light on the basic principles of rebuilding a soil. The scientists are seeing first-hand the soil-forming processes at work under dynamic conditions perhaps unknown in the U.S.

Karim McKarim, research assistant (soil
physics), North Carolina State University Keith Cassel, soil scientist (physics), North
Carolina State University
John Nicholaides, soil scientist (fertility),
North Carolina State University
Mike Wade, soil scientist (fertility), North
Carolina State University
D.S. Gunawan, research assistant, Center
for Soil Research
John Thompson, agronomist, University of


Soils, Crops and


To hoe or not to hoe. To lime or not to lime. To fertilize or not to fertilize. For the transmigrant farmers of Sitiung, strangers in a new land, finding answers isn't always a simple matter of consulting a wise old relative or neighbor. It's a new place, largely unstudied, almost without agricultural tradition. But a great deal rides on the answers crops, incomes, food.

On-Farm Trials
The idea was for TropSoils projects to answer basic questions about soils and cropping systems, as quickly as possible, taking some of their cues from the farmers themselves. Mike Wade's team which included Carol Colfer, Djoko Santoso, Suwandi, and Atin Kurdiana set up on-farm trials with 19 farmers who took charge of tillage, crop-protection and planting patterns. The scientists wanted to test various combinations of fertilizers against the government's recommendations. At the same time, the team wanted to build a working rapport with a group of Indonesian farmers a rapport that would benefit both.
Intercropping relay-planted cassava with rice and, after the rice harvest, with chili peppers and peanuts, the researchers and farmers found that tillage was a very important soil-management consideration. Hoeing forest litter into land recently cleared, but not burned, greatly improved

Cowpeas in research plots, simulating onfarm conditions

yields on both fertilized and unfertilized plots.
But there was one exception to this trend: Liming a field increased yields and almost offset the advantage of hoeing. This was a point in favor of lime, since farmers working newly cleared land quickly tire of hammering away at the thick mat of tree roots that lies just under the surface. While these trials are likely to help scientists formulate fertilizer and soilmanagement recommendations, they also benefit the research in other ways. Researchers have a rare opportunity to monitor soils and farming practices, beginning with a first crop on new land. They have some good, basic information to use in designing new studies. And, they have laid a foundation for more collaboration.

Liming is a relatively new practice in Indonesia, and such questions as when, where and how much to apply are unanswered. What was clear was the need for more research; the acid, red-yellow soils of Sitiung were very similar to those on many transmigration sites throughout Indonesia, and what worked on one site would likely work on others. TWo TropSoils projects began to sort things out.
In one of these projects, Wade, Agus
Sophian and Kasno began comparing two sources of lime, burned lime and ground lime, for their effectiveness in neutralizing soil acidity and improving crop production. Burned lime, which is made from ground lime, is not considered economically feasible in developed countries, may be potentially competitive in Indonesia as a home or small-scale industry.
In initial trials, upland rice, which is known to tolerate acid soils, showed almost no response to lime. Peanuts showed a slight response, and mung beans, a crop sensitive to high levels of aluminum found in acid soils, exhibited a very definite response.
In a companion project, Wade, Gene Kamprath, Djoko Santoso and Rumawan are trying to determine what soil acidity levels are optimum for upland rice, soybeans, peanuts, corn and mung beans. They want to know what liming rates are needed to maintain these levels, and the residual effect of various rates on a rotation of food crops. Early tests have shown soybean yields climbing steadily as lime was increased and soil acidity decreased.


About the Soils

Although the soils are deep and permeable, without physical barriers to root growth, practically all of their fertility lies in a very shallow layer on the surface. Native vegetation has accumulated most of the available plant nutrients into its biomass, releasing them slowly as the vegetation dies and decomposes, directly feeding the roots. Therefore, removing organic material from a site, by bulldozing, for example, greatly reduces the site's fertility.
Under this surface layer of organic material, the soils, which are generally more than half clay, are acid, with levels of aluminum toxic to plants. They are also nutrientpoor, with a lack of weatherable minerals. Unless these chemical constraints are corrected, cleared land will be barren. While there are some areas of relatively level land, much of the transmigration site is rolling, with short, steep hills prone to erosion.

Both projects, which are continuing, will help Indonesia's Center for Soil Research develop a set of guidelines for lime applications that may be extended to areas with similar soils.
It's a matter of the right dose for the situation. Some soils need almost no phosphorus. Others need large doses. But as yet no accurate method for matching the dose to the need is available for the settlements. The need varies not only with the soil but with the crop as well. The 100 kilograms-per-hectare alloted to Sitiung farmers are fine for upland rice, but apparently far too little for peanuts.
In another continuing project designed to establish basic guidelines for the use of soil amendments, Wade, Santoso and Suwandi began studying rates, methods and longterm effects of phosphorus applications using triple super phosphate. The first test crop, peanuts, showed a dramatic response to phosphorus, and yields increased from almost nothing to nearly a ton per hectare.
The project is continuing, but some early results suggest that the government's recommendation that phosphorus fertilizer be banded laid in bands along the rows
- adds unnecessary labor. In the TropSoils trials, which were managed by researchers in farmers' fields, simple broadcasting performed as well as fertilizer banding, when both applications were incorporated by hoeing. If long-term studies confirm this trend, a change in policy would benefit the farmers.

Mike Wade, soil scientist (fertility), North
Carolina State University

Carol Colfer, anthropologist, University of
Djoko Santoso, soil scientist (management),
Center for Soils Research
Suwandi, research assistant, Center for
Soils Research
Atin Kurdiana, research assistant, Center
for Soils Research
Agus Sophian, research assistant, Center
for Soils Research
Kasno, research assistant, Center for Soils
Gene Kamprath, soil scientist (fertility),
North Carolina State University
Rumawan, research assistant, Center for
Soils Research

Farmer hoeing a new field


Strips and Spots

sorting the puzzle
of spatial variability

Rolling, acid and variable, the soils of the transmigration area of West Sumatra challenge farmers and researchers alike. One of the first tasks for the new TropSoils team was to survey and assess the soils in order to establish a baseline of information for its studies. Another was to try and understand spatial variability the sometimes extreme differences in soil conditions and crop responses over relatively small areas.


g/m2 300 200 100

Why is the typical farm in Sitiung alternately dotted with bare spots and striped with green bands? ftropSoils researchers found that the bare spots were areas where acid, sterile subsoil had been exposed by bulldozers, and the green strips were usually the ash lines of burned trees. Many of the bare spots were abandoned by farmers, and have rapidly eroded.
Accommodating such variation in a farm field means carefully measuring and applying lime and fertilizers in differing rates to suit conditions a hard job, even for farmers who apply their chemicals by hand. On research plots, where the goal is to measure the effect of treatments under reasonably uniform and replicable conditions, variability greatly complicates results.
Like scientists from TropSoils' Semi-Arid Tropics Program (see page 8), the TropSoils team in Indonesia has looked to geostatistics for help in analyzing variability in soils, and its effect on crops. Using geostatistics to estimate soil properties at unsampled locations from analyses of neighboring samples, the team, which includes Bruce Trangmar, Mike Wade, Djoko Santoso, John Thompson and Russel Yost, has shown that natural soil variability can actually help research by revealing answers to key agronomic questions. For example, data from a variability trial showed that differences of rice yields were more related to aluminum (which can vary from zero to 90 percent saturation on freshly cleared land), potassium and micronutrients than to such things as organic matter, nitrogen and phosphorus.

Bruce Trangmar, research assistant (soil
physics) University of Hawaii
Mike Wade, soil scientist (fertility), North
Carolina State University
Djoko Santoso, site coordinator, Center for
Soil Research
John Thompson, agronomist, University of
Russell Yost, soil scientist (fertility),
University of Hawaii

Diagrams showing soil acidity-aluminum saturation-(below) and grain yields (above) on the same test plot


Wade in uniformly
treated test plot

Variability, Piece by Piece

It looked like a failure, a key experiment and a lot of hard work down the drain. The object was to find the response of mung beans to phosphorus, but the test plot, a neat rectangle, contained a wildly disorganized pattern of growth. Patches of mung beans were thriving, others were hardly alive.
"Normally, when the micro-variability is that extreme, the tendency is to plow it under and start over, somewhere else,' says Mike Wade. "With that kind of variability, you can't tell much about the crop's response." But the experiment was crucial, and losing it would cost valuable time. Wade decided to call on some campus colleagues for help. Gene Kamprath and Russel Yost arrived to study the plots and talk things over. The diagnostic approach they agreed on was based on a time-honored scientific principle: examine the cause-and-effect. "We came up with a plan:' Wade says. "We marked off any uniform spot in the field, whether good or bad, and sampled it. We harvested and measured yield and took soil samples one each plot, just as if it were a field in itself:' Jig-sawed by stakes and string, the field became a puzzle, each piece yielding a different set of data. Piece by piece, the puzzle came together, a picture formed the relationship between growth and several soil characteristics.
"We found out that where there was poor growth the soil acidity was high,' Wade says. "And where growth was good, soil acidity was low and the bases, calcium and magnesium, were high. Looking at all these factors, we could explain the growth response."
Wade says that, because his team could draw on the expertise of soil scientists with different backgrounds and experience, the program found a new way of dealing with variability, and the experiment was salvaged.
"We actually used the variability to get more information than we'd expected," he says.

Mike Wade, soil scientist (fertility), North Carolina State University Gene Kamprath, soil scientist (fertility), North Carolina State University Russel Yost, soil scientist (fertility), University of Hawaii


In Conclusion

To prevent famine, alleviate hunger and meet the minimum dietary needs of its population, the world must increase its production of food. This is an accepted fact. 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 succeed, the soil constraints that limit plant growth must be minimized. The goal of 'ropSoils is to address this issue. Formally defined, this goal is to develop and adapt improved soil management technology which is agronomically, ecologically and economically sound for developing countries in the tropics.
The level of success achieved in reaching this goal will be influenced by the resources available and the effectiveness with which they are used. The TyopSoils approach, collaboration, is designed to enhance both these factors. Though the program is still in its infancy, identifiable progress has been made. The program is already developing technological practices that have the potential for improving food production in developing countries. Dedicated and capable people are forming and conducting a unified program that is both technically sound and suited to its goals. We are pleased to have had the opportunity to present some of their work.
Charles B. McCants
TropSoils Management Entity

Management Entity Office Charles B. McCants, Director
North Carolina State University
Kim S. Stevens, Administrative Assistant
North Carolina State University Neil Caudle, Editor
North Carolina State University

Board of Directors Morris Bloodworth, Chairman (until 10/83) Texas A & M University Ada Demb, Chairman (after 10/83)
University of Hawaii
Wenceslau J. Goedert (after 7/84)
Brazilian Agricultural Research Enterprise Mamadou Ouattara
National Institute of Agronomic Research
for Niger
Robert H. Miller
North Carolina State University D. Muljadi (until 1/84)
Center for Soils Research Edwin B. Oyer
Cornell University Victor Palma
National Agricultural Research and
Extension Institute
E. C. A. Runge (after 10/83)
Texas A & M University M. Sudjadi (after 1/84)
Center for Soils Research Elmar Wagner (until 7/84)
Brazilian Agricultural Research Enterprise

Technical Committee
Frank G. Calhoun, Chairman
Texas A & M University

Douglas J. Lathwell
Cornell University John J. Nicholaides
North Carolina State University Pedro A. Sanchez
North Carolina State University Goro Uehara
University of Hawaii

External Evaluation Panel
John Coulter, Chairman World Bank Peter Hildebrand University of Florida Marlowe Thorne University of Illinois

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