Back to Home Page of CD3WD Project or Back to list of CD3WD Publications

Home - English - French - German - Italian - Portuguese - Spanish
                          TECHNICAL PAPER #58
                           UNDERSTANDING SOIL
                         CONSERVATION TECHNIQUES
                              Fred Weber
                             Carol Stoney
                           Dr. Edward Pytlik
                             Illustrated By
                           Frederick J. Holman
                             Published By
                     1600 Wilson Boulevard, Suite 500
                       Arlington, Virginia 22209 USA
                  Tel:  703/276-1800 * Fax:   703/243-1865
               Understanding Soul Conservation Techniques
                          ISBN:  0-86619-277-8
              [C] 1989, Volunteers in Technical Assistance
This paper is one of a series published by Volunteers in Technical Assistance to provide an introduction
to specific state-of-the-art technologies of interest to people in developing countries.  The papers are
intended to be used as guidelines to help people choose thechnologies that are suitable to their situations.
They are not intended to provide construction or implementation details.  People are urged to contact
VITA or similar organizations for further information and technical assistance if they find that a
particular technology seems to meet their needs.
The papers in the series were written, reviewed, and illustrated almost entirely by VITA volunteer
technical experts on a purely voluntary basis.   Some 500 volunteers were involved in the production of
the first 100 titles issued, contributing approximately 5,000 hours of their time.  VITA staff included
Suzanne Brooks handling typesetting and layout and Margaret Crouch as editor and project manager.
Co-author Fred Weber, a pioneer in the community forestry concepts presented here, has advised
projects for over 20 years.  He wrote the original edition of the VITA publication Reforestation in Arid
Lands, from which much of this paper is drawn, based on a training manual he prepared for Peace Corps
volunteers in Niger.  Carol Stoney collaborated with Mr. Weber on the revisions for the new edition of
Reforestation..  Dr. Edward Pytlik teaches appropriate technology at West Virginia University.   Frederick
J. Holman, a landscape architect, illustrated both the book and the additional material in this paper.
VITA is a private, nonprofit organization that supports people working on technical problems in
developing countries.  VITA offers information and assistance aimed at helping individuals and groups
to select and implement technologies appropriate to their situations.  VITA maintains an international
Inquiry Service, a specialized documentation center, and a computerized roster of volunteer technical
consultants; manages long-term field projects; and publishes a variety of technical manuals and papers.
                by Fred Weber, Carol Stoney, and Dr. Edward Pytlik
Soil conservation efforts protect the soil from the two primary forces of erosion, wind and water.  A
wide assortment of different soil conservation techniques are being used today.  Windbreaks and dune
stabilization, for example, are effective methods of halting wind erosion.  Terracing, planting trees and
other vegetation in contour strips or along contour ridges, and gully control plantings are techniques
used in combination with physical control measures to reduce soil erosion from water.  Conservation
tilling refers to a variety of methods used to control both wind erosion.
Some of these methods are based on traditional practices that have been carried out by local people
for generations.  Others are relatively new, "invented" by technicians working with local farmers or
pastoralists and still being adapted to varying site conditions.  The methods described here provide a
practical guide for use in the field, rather than extensive coverage of background information, theory,
and reference sources.  The Reference List and Information Source List should be consulted for further
The techniques described here can contribute to the increased productivity and sustainability of land
use systems.  Most can be implemented by rural households or communities using locally available
materials.  Nearly all of the techniques involve the establishment of vegetation cover, primarily trees
and shrubs.  Some also involve physical soil conservation methods as well, such as contour ridges,
terraces, or walls.  This approach is intended to increase awareness of ways in which vegetative
methods can be used in combination with physical methods.
Trees and shrubs play a critically important conservation role.  They can reduce soil surface
temperatures, increase infiltration and retention of soil moisture, provide organic matter, pump
nutrients, fix nitrogen, reduce erosion from water and wind, form live fences, and provide shade, all
of which create better growing conditions for crops and grasses.  In addition, certain tree species may
provide food (fruit, leaves, edible seeds, etc.) not only for people but also for livestock, fuel, building
materials, and other important products.
Soil conservation project planning should not take a cookbook approach.  Rather, the project design
should be adapted to specific site conditions and current land use patterns.  Species trials are required
to meet site requirements.  Demonstration sites using more varied species, including more indigenous
species, are needed so that future selection can be made on the basis of what has worked.
The material in this technical paper is drawn largely from Reforestation in Arid Lands (Weber and
Stoney, 1986), which provides a comprehensive review of reforestation methods including project
design, site and species selection, soil preparation, nursery management, and many of the conservation
techniques presented here.  Additional material, on physical methods to control erosion, were provided
by Dr. Pytlik.
Windbreaks are strips of trees and other vegetation that slow the flow of the wind, reducing wind erosion,
evaporation, and wind damage to crops.   They are sometimes referred to as shelterbelts, although this
term usually implies a wider strip of vegetation, which incorporates more rows of trees and shrubs than
are usually found in a windbreak.
Windbreaks have an especially high potential in farming areas where cereal crops such as millet and
sorghum are grown.  The most successful windbreak projects to date are those found on enclosed farm
lands and in some demonstration or pilot projects under government or private control.  The major
obstacle to windbreak establishment in other areas has been the difficulty and high cost of protecting the
trees against animal grazing.   Some large-scale successes have been achieved in areas where donors,
government agencies, and local people have worked closely together.
Highly impressive results have been observed at a CARE project in the Majjia Valley in Niger, where
crop yields from fields protected by windbreaks are consistently higher than those from unprotected
fields.  Studies indicate that total yields are approximately 20 percent higher, even after accounting for
losses from land that has been taken out of crop production to provide space for the windbreaks
(Dennison, 1986).
The effectiveness of a windbreak depends on how efficiently the wall of vegetation blocks the wind and
confines the wind's turbulence to the zones close to the windbreak.  A vegetation density of, 60 to 80
percent seems to work best in arid zones.   A barrier dense enough to block wind passage completely will
cause turbulence close to the ground, loosening soil particles that can then be picked up by the wind.   As
well as removing needed topsoil, wind that is carrying soil particles causes damage to crops through the
abrasive effect of the sediment on plant tissues.
A row of trees that provides less complete wind reduction will also ensure that the effects of the wind
are felt farther away.  Gaps or openings in the windbreak should be avoided as much as possible.  Wind
is funneled through gaps in the tree rows, concentrating its force and speed, so that the final effect can
be very damaging.

22p02.gif (300x600)

Windbreaks can furnish protection for downwind areas up to 10 times the height of the trees, provided
the windbreak consists of at least two rows of plants of different heights.  Large trees should be chosen
for one row (see A, below).  Fast-growing species can be mixed with slower growing, longer-lived trees,
depending on local preference.   Row B should be composed of shorter species, chosen if possible for their
by-products, and rows C and D are auxiliary rows.   These are planted with lower, bushier trees, shrubs,
and grasses.  A well chosen vegetation mix for windbreak composition will not only provide protection
from the wind, but will yield secondary products as well.  The windbreak trees themselves, if properly
harvested, can also provide significant quantities of fuelwood and poles without jeopardizing their
primary function.

22p03a.gif (600x600)

Windbreaks and shelterbelts can be laid out to include roads, trails, or driveways for livestock.  In this
way, animals and people can benefit from a shaded passageway that otherwise would be very hot.  Any
path through the windbreak should be at an oblique angle rather than perpendicular to the tree rows.  This
will allow people and livestock to move through the windbreak without opening a gap for the wind to
roar through.

22p03b.gif (486x486)

Some other points to consider about windbreaks:
o      Species chosen should obviously be suited to the soil and climate where they will be grown.
       Local species are preferred Good selections can be made from species protected by law.
       Use species local residents themselves have chosen and value.
o      The most efficient windbreaks are those with one or two rows of low-growing
       shrubs or trees on the outside and two or three rows of taller trees on the inside.
o      The utility of wider shelterbelts can be enhanced by the selection of multiple use
       species for the middle rows.  Species that provide locally consumed fruits and
       medicines should definitely be considered.
o      Frequently a combination of planting methods is highly practical when establishing
       windbreaks.  In other words, a combination of nursery transplants, live fencing,
       cuttings, and stumps can be used depending on planting times in the area.
o      Preparation and protection of the site involved are possibly more important for
       windbreaks than for regular plantations.  During the rainy season when crops are
       being cultivated, the fields are effectively protected from livestock, however, after the
       harvest the animals are often allowed to browse the crop residues left in the
       fields.   Keeping livestock away from the windbreaks during this time is difficult, and
       fencing a long narrow strip of land is costly.
o      Where complex land ownership patterns exist, it may not be possible to establish
       continuous straight tree rows across individual fields and parcels.   In this case
       windbreaks may be staggered so that they conform with established boundaries such as
       borders of fields, roads, trails, stream, and other natural or man-made features.   Staggered
       windbreaks also provide the most effective protection around towns and villages, where
       they are laid out in a pattern of overlapping blocks.

22p04.gif (486x486)

o    Another possible planting pattern is to line farm fields with wide windbreaks and to
     Plant dispersed trees inside the field.
o    Many nurseries in arid zones could benefit from the establishment of a windbreak to
     protect the seedlings from drying winds.  The nursery windbreak also serves as a
     demonstration to visitors to the nursery.  If the nursery is very small, however, a tall
     windbreak might cast too much shade on the seedlings.

22p05.gif (270x540)

Sand Stablization
Sand stabilization is an important aspect of revegetation and conservation activities in many arid and
coastal areas.  Shifting and blowing sand causes great damage to farmland, buildings, installations, and
roads.  Entire settlements can be threatened by the movement of shifting dunes.
The best protection against drifting or blowing sand is to prevent the sand from being picked up by the
wind and becoming airborne.  Conservation of existing grass and other vegetation cover is necessary to
hold the sand in place.  Even a small disturbance such as a footpath can start the process of erosion on
fragile dunes.  Once airborne, drifting sand can be made to settle, nevertheless, and can be kept from
further shifting.
The first step is to determine why the natural vegetation has not recolonized the area that is being eroded.
Various options that will remove any constraints to natural vegetation should then be considered.  Often
the problem is being caused by animals.   Under these circumstances, little if anything will be gained by
planting trees, unless access is first controlled.
There are basically two approaches to dune fixation:   biological and physical. The best ultimate results
are obtained when the open area where sand is picked up can be permanently covered by vegetation.
Biological methods include:
o    Fencing off the area to protect it from animals, so that the vegetation can regenerate
o    Establishing hedge rows of species that can be successfully regenerated from
     cuttings even in areas where annual rainfall does not exceed 300-400mm.   Freshly cut
     branches can be partially buried in rows of shallow trenches.
o    Direct seeding, particularly of grasses, but also of woody plants such as vines,
     shrubs, and trees.
Often before grasses and other ground cover can be reestablished, however, the movement of the sand
must be halted.  Physical dune stabilization measures include:
o    Wind-baffles (palisades), which are constructed of a variety of materials, generally
     whatever is locally available.
o    "Fore-dunes," which consist of sand or soil ridges set at right angles to the major
     winds.   They can be 1 to 5 meters high and stretch over hundreds of meters in
     length.   Heavy construction equipment is required for large-scale efforts.
o    Mechanical surface stabilization, which is accomplished by covering exposed areas
     to reduce further erosion.  Plastic sheeting, nets, cloth, or some other fiber are used.
o    Chemical surface stabilization, which involves spraying a binder (rubber, oil, or
     plastic base) on the surface to bind particles together.  Grass seeds and mulch can
     also be mixed with the binder and sprayed on the area to be protected.
Preference should be given to biological control measures whenever possible.  However, some physical
construction is often needed for initial plant establishment.  Usually some type of low-cost materials are
available locally.  This barrier can take many forms and be made of a variety of materials.

22p06.gif (486x486)

Stems and poles (3-8cm in diameter and up to 2m long) can be used to construct a diamond pattern of
criss-cross rows across areas of open sand.   Branches can be staked out in dense rows, or fences can be
woven from branches to construct the palisade.   By breaking the force of the wind, the palisades keep
the exposed sand from being picked up, and the sediment load already carried by the wind is deposited
in or behind the barrier.  Sand will become entrapped in such rows, and ridges will gradually form.  Plant
growth then becomes possible in the protected areas behind the ridges.
Fenced in squares and other sand traps can also be constructed of materials as basic as bundles of grain
stalks or other crop residues.   Additional possibilities include palm fronds, sticks, branches, cardboard,
or any material that is reasonably sturdy, easily available, and low cost.  Some of the problems that may
be encountered in maintaining the barriers include damage from animals and termites that are attracted
to them for food.  Where sand accumulations are heavy, the barriers may have to be raised or added to

22p07a.gif (486x486)

Before beginning a sand or dune stabilization project, planners should consider the following:
o     Dune fixation is not an appropriate conservation investment if the area that is being threat
      ened by shifting sands has no inherent value.  Unless some benefit will accrue in terms of
      protection of farmland, homes, or other property, the cost is prohibitive.   Furthermore, those
      who will gain the most from the project should also be willing to exert the most effort, particularly
      in terms of sustaining and protecting the vegetation cover.

22p07b.gif (486x486)

o    Dune fixation projects should not be undertaken without first carefully evaluating traditional
     and current land use attitudes, especially those governing grazing and wood cutting.   If these
     are incompatible with the restrictions needed to protect the vegetation, then changes in land
     use policies and practices must take place before dune fixation activities are initiated.
o    The shifting of live dunes is influenced by a complex set of variables, and may
     change with the seasons.  It is worthwhile to observe and measure dune movements for a
     period of 12 months before starting stabilization activities.

22p08.gif (600x600)

o    Except under extreme desert conditions, it is more effective to stabilize the zone of
     origin of the shifting sand, rather than concentrating efforts on the areas where the sand is
     being deposited.  It is important, therefore, to determine the location from which the sand is
     being removed by the wind.
o    Project sites that are close to or within actual desert zones will require more
     intensive efforts to stabilize shifting dunes.  Maintenance inputs will also be higher.
o    The more exposed a specific location is to the wind (near the crest of large dunes, or in
     saddles between ridges), the more difficult it is to establish vegetation.   Physical protection is
     often needed.  If it is not possible to use physical control measures, however, the area can still
     sometimes be stabilized after the top has been lost to wind erosion.
o    Locally occurring trees and shrubs have great resiliency.   In species selection, the indigenous
     vegetation should receive priority over exotics, particularly for large-scale projects.
Contour Strips
The most likely, logical place to use trees and shrubs to halt erosion caused by water is across slopes,
particularly where hillside cultivation is practiced.   Properly maintained trees and shrubs, planted in
combination with grasses and other vegetation, can effectively control surface runoff, thereby reducing
soil losses.  One successful technique involves establishing parallel vegetation bands along contour lines.
These contour strips will reduce runoff from the slopes above if they are designed and maintained to
ensure a dense, multi-layered permanent ground cover.   The ground surface is protected by successive
layers of litter, grasses, other ground plants, bushes, and trees.  A dense vegetation belt will not only stop
or slow down runoff, but will also trap soil particles suspended in the water that have been removed from
the more exposed areas between the strips.

22p09.gif (540x540)

Correct dimensioning of the D and W variables indicated in the illustration above is important.  Many
factors affect the spacing of the strips, but the degree of slope is the most important.  If previous efforts
to establish contour strips in the area are available for study, these sites should be observed for evidence
of erosion to determine if the dimensions are in proportion.  Conservation services may also have tables
or formulas appropriate for local site conditions.   If no information of this kind is available, dimensions
can be calculated using the following table as a rough indication of spacing in arid and semi-arid areas:
             Slope (degrees)          W (meters)          D (meters)
                      0                   2                   50
                      5                   4                   47
                     10                   5                   43  
                     20                   8                   38
                     30                  10                   33
                     40                  13                   28
                     50                  17                   24
                     60                  20                   20
Basis:  0-600mm mean annual precipitation
In areas with rainfall between 600-1,000mm:       increase W by 20%
                                                 decrease D by 10%
In areas with rainfall more than 1,000mm:         increase W by 50%
                                                 decrease D by 20%
Revegetation efforts on these strips can be approached in many ways.  To simply establish some
groundcover, scarification of the ground along the contour may be sufficient site preparation.  Furrows
can be dug by hand or by using a harrow or disc blade.   More intensive effort may consist of additional
seedbed preparation, for instance, loosening up the soil surface and raking along the contour.  Direct
seeding of desirable trees and shrubs may be feasible for some species.  Other trees can be established
by cuttings.  The most direct, but also most costly, method of establishing contour strips is by planting
nursery raised seedlings.
The primary consideration for species selection should be local preference, because the contour strips
take a certain percentage of the land out of cultivation, even though they are intended to increase
productivity of the total area.   Many different species can be used, often in combination. Fruit trees are
frequently a high priority on farmland.   In other areas, trees that produce poles for construction, rafters,
and fences may be preferred.
Particular attention should be given to vegetation layers nearer the ground surface.  Fodder plants, such
as Guinea, napier, clover, or elephant grasses, may be of interest for feeding to penned livestock.
Perennial bean species, produced on small woody shrubs for human consumption, may appeal to the local
inhabitants.  Contour strips can be a good location for introducing new species on a small-scale,
experimental basis as well.
For centuries, farmers living in hilly and mountainous regions of the world have been terracing their
hillsides as a means to prevent soil erosion.   Terraces are simply channels cut into hillsides, embankments
built onto hillsides, or a combination of the two constructed across the slope of the land. They have proven
to be the most effective mechanical means of erosion control on slopes planted in continuous row crops.
As much as 85 percent of the sediment eroded from a field can be trapped by terracing.
There are four basic terracing designs.   In the level bench design the terraces are parallel with the horizon,
whereas in the sloping bench design the terraces are leveled so that their planting surfaces have a slight
downward angle.  The reverse slope or step terracing design has terraces that have planting surfaces that
angle upward slightly.  The fourth terracing design, used primarily in conjunction with flood irrigation,
has terraces that are parallel with the horizon and have a built up outer edge to prevent water runoff down
the hill.

22p10.gif (540x540)

Trees Along Contour Ridges
For information on the various applicable soil conservation measures that involve construction of
contour ridges or terraces, or excavation of infiltration ditches, a number of texts are available.  The
Centre Technique Forestier Tropical (CTFT), the Centro Agronomico Tropical de Investigacion y
Ensenanza (CATIE), the International Council for Research in Agroforestry (ICRAF), and the United
Nations Food and Agriculture Organization (FAO) have all published handbooks and technical materials
on the subject.  In addition, many of the bilateral donor organization have developed standard texts on
the subject during the past decade.   Construction designs and extension materials have been developed
specifically for certain countries, among them Honduras, Kenya, Burkina Faso, and the Philippines.
There is still relatively little information available, however, that deals with the effective combination
of biological and physical erosion control measures.   Vegetation, especially trees and shurbs, can play
a vital role in increasing the effectiveness of soil and water conservation efforts.  Properly established and
managed woody plants can reduce maintenance and costs on hillside erosion control projects as well.
The following sketch shows where trees and shrubs can make an important contribution to physical ridge
or ditch formations along the contour lines of sloping surfaces.

22p11.gif (600x600)

Gully Reclamation
Permanent vegetation, especially shrubs and trees, can reduce bank or channel bottom erosion as long
as the flow of water is not too powerful.   Vegetation can also help stabilize mechanical protection
materials, such as large rocks positioned along banks or bottom (rip-rap), wire mesh baskets filled with
rocks (gabions), or bales of straw or branches staked in place to reduce water velocities.
Gullies present special problems, because they often occur on steep slopes, and even brief peak flows
can cause serious damage.  Gully erosion is difficult to reverse once it has gotten started, and it can quickly
destroy valuable agricultural land.
To prevent the formation of gullies along waterways, the banks should be lined with trees and shrubs.
Trees, shrubs, and other vegetation can be established within the gullies to control further erosion and
to help rebuild the soil layers that have been removed.   Improperly placed trees can, however, have the
undesired effect of narrowing the channel and increasing the speed of stream flow.  The following sketch
shows how to combine vegetation with mechanical gully erosion control methods for optimal results.

22p12.gif (600x600)

Conservation tilling is a general term that includes a number of tilling methods, used alone or in
combination, to conrol erosion caused by both wind and water.  The methods have in common the goal
of disturbing the surface of the soil--as by plowing--as little as possible.
In general, the dominant factor in determining the effectiveness of conservation tillage practices is the
amount and distribution of crop residue left on the soil surface.  However, the amount of crop residue
mixed into the soil during tilling, the type of soil, size and location of untilled residue strips, contour
ridging, and surface roughness are all important factors contributing to soil loss prevention.
Wind erosion control can be established by developing vegetative and non-vegetative land cover,
reducing field lengths along the prevailing wind direciton, roughening or clodding the land, and terracing
slopes and hilltops where converging winds increase velocity and shear stress.
Crop residue and mulches help to reduce both water runoff and the amount of sediment contained in the
runoff.  Ground roughness and clods created through tilling increase water absorption and reduce water
runoff velocity.  Ridging on the contour also substantially reduces runoff velocity and soil loss.
Low-till and no-till farming practices combined with residue mulch cover and contour planting can
reduce the soil loss ratio of a field from .63 in conventional across-slope plowing and planting to .12 in
the first year of implementation and to .04 by the end of the fourth year.  Examples of no-till farming
systems include the following:
1.   Sod-planting--in which maize, for example, is grown in combination with cool-season
     perennial grasses.
2.   Sod-strip planting--in which six rows of maize are alternated with 8m parallel
     strips of established grasses across the general slope of the land.   Each year 1 1/4
     rows of maize is advanced down the slope and the upper border is reseeded to a
     mixture of grass and legumes.
3.   Complete pasture renovation--sod-planted maize method extended into entire
     fields where erosion is too severe to permit conventional tillage.
4.   Interseeding legumes and/or grasses into established grass.
5.   Planting in winter cover crops.
6.   Planting in crop residues.
7.   Multi-cropping systems--maximizes production by providing three crops in two
     years or five crops in four years.

22p13.gif (600x600)

Contour plowing and planting are more popular than terracing because of their lower production and
maintenance costs (both real money and time).   In the contour system, both the plowing and planting are
done across the slope and follow the natural contour of the land.  Contour strip cropping is an even ore
efficient means of soil erosion, but this efficiency is offset by a loss of the major crop yield.  However,
the alternating forage crop compensates somewhat for this loss.
Buck, L.E. (ed.). 1983.  Proceedings of the Kenya National Seminar on Agroforestry,
      Nov. 1980.   Nairobi:  International Council for Research in Agroforestry and the University of
Delehanty, J., J. Thomson, and M. Hoskins.   1985 Majjia Valley Evaluation Study:
      Sociology Report.  Niamey:   CARE International Report.
FAO. 1977.  Guidelines for Watershed Management.  Rome:  Food and Agriculture Organization
      Conservation Guide Series No. 1., 298 pp.
FAO. 1977.  Conservation in Arid and Semi-Arid Zones.   Rome:  FAO Conservation Guide
      Series No. 3.
FAO. 1977.  Special Readings in Conservation Techniques.  Rome:  FAO Conservation Guide
      Series No. 4.
FAO. 1983.  Management of Upland Watersheds; Participation of the Mountain
      Communities.  Rome:   FAO Conservation Guide Series No. 8.
FAO. 1985.  Sand Dune Stabilization:  Shelterbelts and Afforestation in Dry Zones.  Rome:
      FAO Conservation Guide Series No. 10.
FAO. 1985.  FAO Watershed Management Field Manual:  Vegetative and Soil Treatment
      Methods.   Rome:  FAO Conservation Guide Series No. 13.
Flannery, R.D. 1981.  Gully Control and Reclamation.  Arlington, Virginia; Volunteers in
      Technical Assistance (VITA), 26 pp.
Gulick, F.A. 1984.  Increasing Agricultural Food Production through Selected Tree
      Planting Techniques:  A summary memorandum with selected references.
      Washington, D.C.:  USAID/Bureau for Africa, 149 pp.
Hagedorn, H. et al. 1977.  Dune Stabilisation:  a survey of literature on dune formation and
     dune stabilization.  Eschborn, W. Germany:  GTZ 193 pp.
Ilaco, B.V. (ed.) 1981.  Agricultural compendium for Rural Development in the Tropics and
     Subtropics.   New York: Elsevier Scientific Publishing Co., 239 pp.
Kunkle, S.H. 1978.  Forestry Support for Agriculture Through Watershed Management,
     Windbreaks and Other Conservation Actions.  Position Paper, Eighth World Forestry
     Congress, Jakarta, Indonesia, 28 pp.
Le Houerou, H.N. (ed.) 1980.   Browse in Africa:  The Current State of Knowledge.  Addis
     Ababa: ILCA, 491 pp.
McGahuey, M. 1986.  Impact of Forestry Initiatives in the Sahel on Production of Food,
     Fodder, and Wood.  Washington ,D.C.:   Chemonics International, 25 pp.
Nair, P.K.F. 1980.  Agroforestry Species:  A Crop Sheets Manual.  Nairobi:   ICRAF, 83 pp.
Phillips, R.F. and Phillips, S.H. 1984.   No-Tillage Agriculture:   Principles and Practices.
     New York:   Van Nostrand Reinhold Co., 306 pp.
Poincelot, R.P. 1986.  Toward a More Sustainable Agriculture.  Westport, Ct.; AVI
     Publishing Co. 241 pp.
Soil Conservation Society of America. 1973.   Conservation Tillage.   Ankeney, Iowa;
     SCSOA, 241 pp.
Sprague, M.A. and Trippet, G.B. 1986.   No-Tillage and Surface-Tillage Agriculture.  New
     York:   John Wiley and Sons, 467 pp.
USDA/SCA. 1962.  Soil Conservation Manual.  Paris:   USAID/Centre Regional d'Editions
     Techniques, 359 pp. (Also available in French).
Vergera, N.T. (ed.) 1982.  New Directions for Agroforestry:  The Potential of Tropical
     Legume Trees.  Honolulu Environment and Policy Institute, East-West Center.
Weber, F. and M.W. Hoskins. 1983.   Soil Conservation Technical Sheets (Fiches
     Techniques de Conservation du Solss).  Moscow, Idaho:  University of Idaho for
     USDA (OICD), 112 pp.
Weber, F. with Carol Stoney. 1986.   Reforestation in Arid Lands.  Arlington, Virginia: