SENEGAL AGRICULTURAL RESEARCH II PROJECT
TRIP REPORT FOR ISRA/DRPF
AGROFORESTRY RESEARCH IN SENEGAL:
PROBLEHS AND PDTENTIAL
by
James R. Seyler
Michigan State University
Department of Forestry
East Lansing, MI 4882.4
Lansing, October 8, 1988

---

TABLE OF CONTENTS
Page
1.0 Introduction
1
2.0 Historical Perspective
1
3.0 ISRA's Current Agroforestry Research Program
2
3.1 Projet de Recherche - "Développement sur le Rôle
2
de l'Arbre en Exploitation Agricole"
3.1.1 The Northwest Peanut Basin
3
3.1.1.1 Agroecological Zone Description
3.1.1.2 On-Station Research (Thienaba)
:
3.1.1.3 On-Farm Research (Khayes)
4
3.1.1.4 Comments/Observations
5
3.1.2 Sine Saloum
7
3.1.2.1 Agroecological Zone Description
3.1.2.2 On-Station Research (Nioro du Rip)
ii
3.1.2.3 On-Farm Research (Sinthiou Kahel)
B
3.1.2.4 Comments/Observations
9
3.1.3 Casamance (lower)
1 0
3.1.3.1 Agroecological Zone Description
3.1.3.2 On-Station Research (Djibelor)
:o
3.1.3.3 On-Farm Research (Boulandor)
3.1.3.4 Comments/Observations
;:
4.0 Suggestions for Strengthening ISRA/DRPF's Agroforestry
13
and Natural Resource Program
4.1 Determination of Agroforestry (and Forestry)
13
Research Priorities
4.2 Linking Research with Policy, Prices and Institutions
14
4.2.1 Markets
4.2.2 Land and Tree Tenure
ii
4.3 Suggestions for Expansion of ISRA/DRPF's Current
1 5
Program
4.3.1 Erosion Control (Water)
1 5
4.3.2 Erosion Control (Wind)
4.3.3 Natural Forest Management
::
4.3.4 Supporting and Component Research
17

---

4.4 Suggested Areas for Additional Long-term Training
17
4.5 Short-term Training
18
4.6 Collaboration with the IARCS
19
4.7 Information/Documentation
19
4.8 Short-term Technical Assistance
20
5.0 Themes for Discussion by ISRA’s Scientific and
2 0
Technical Committee
5.1 Intra-agency Collaboration
2 0
5.1.1 Problems
2 0
5.1.2 Suggestions for Imporved Intra-agency
Collaboration
2 2
5.2 Inter-agency Collaboration
2 3
5.2.1 Problems
2 3
5.2.2 Suggestions for Improved Inter-agency
2 4
Collaboration
ANNEXES
ANNEX 1
Summary of ISRA/DRPF On-Station Species Related
2 7
Research
ANNEX II
Suggested Criteria to Prioritize Research Needs
3 1
in Agroforestry
ANNEX III Partial List of Current MPN/DCSR Projects
3 3
ANNEX IV Persons Contacted
35
ANNEX V
Field Trips
37
ANNEX VI
Bibliography
39

---

AGROFORESTRY RESEARCH IN SENEGAL
1.0 INTRODUCTION
In the context of refining and expanding its agroforestry research
program, the Senegalese Agricultural Research Institute’s Department of
Forest Production Research (ISRA/DRPF) requested that the USAID-Michigan
State University Senegal Agriculture Research Project (SARII) fund a
consultant whose scope of work would include:
(1) a review of forestry and agroforestry research programs in view
of ISRA/DRPF's proposed research directions and ideas on
interdisciplinary collaboration;
(2) discussions with ISRA/DRPF of agroforestry research directions;
(3) identification of themes for discussion by ISRA’s Scientific and
Technical Committee.
This review was undertaken from September 9-24, 1988 in collaboration
with the Director of the DRPF.
Given the limited duration of the visit, the author was only able to
partially tap sources of agroforestry information outside of ISRA, such as
projects being implemented by the Ministry of Nature Protection's
Directorate of Soi1 Conservation and Reforestation (MPN/DCSR) and many non-
governmental organization (NGO) activities.
While the comments and observations in this report may seem at times
over-critical, they are made in the context of helping to strengthen what
is already one of the best agroforestry research efforts in sub-Saharan
Africa.
2 .O HISTORICAL PERSPECTIVE
While the term agroforestry has only corne into vogue in the past
decade, what could be considered to be agroforestry research in Senegal has
had a long tradition.
Beginning in the early to mid-1960s, much of the
original research on the Acacia albida agro-silvo-pastoral system was
conducted in Senegal.
This is one of the best known "traditional"
agroforestry systems and is widely practiced from Senegal to Eastern Sudan.
In 1966, Poulain reported that millet yields near Bambey averaged
about 930 kg/ha under the crown of A. Albida and 450 kg/ha outside the
crown.
Other important contributions to understanding this species include;
Charreau and Vidal (1969), CTFT/Senegal (1966), Gautreau (1967),
IRHO/Senegal (1966), Jung (1971), Lebrun (1968), Wickens (1969), and
Charreau (1970).
In addition to supporting Poulain/s original research,
these studies also indicate that fallow requirements are greatly reduced
when the species is present and that livestock carrying capacity cari be
doubled with a minimum density of trees.

---

2
During this same period, considerable effort was also placed on
agroforestry-related methods of soi1 regeneration including improved fallow
systems and trop rotations (CNRA/Bambey 1964, De Haut de Sigy 1962, Tourte
et al. 1964) and erosion control (Charreau 1970). Much of this original
work, however, was abandoned by the early 1970s due to financial and
personnel constraints as well as a general shift in emphasis to minera1
fertilizer research (See Dancette and Sarr 1985). According to Dancette and
Sarr, agroforestry and related activities in Senegal in the 70s and early
80s were limited to:
millet, bean and peanut intercropping trials with
Eucalyptus SPD;
soi1 regeneration trials with A. tortilis and Prosoois
juliflora at Ndiemane;
living hedges trials of 15 species in Sine Saloum;
windbreak trials (between and within line spacing and orientation) at Bandia
begun in 1982 but with no trop associations to determine windbreak impact on
production.
The results of many of these trials are either inconclusive or
generally unsuccessful.
In the mid 198Os, GOS's and ISRA’s attitudes to agroforestry and
related research began to change.
Evidence was accumulating that minera1
fertilizer use on farms was generally low (Kelly 1986 and Gaye 1987) and
that improving soi1 fertility in Senegal would depend on increased use of
trees in farming systems to supply much needed organic matter and
microbially-mediated nitrogen fixation.
Moreover, many of ISRA's trop
breeders began to recognize that improved varieties of trop seed alone may
not be effective in increasing agricultural production without the improved
fertility provided by agroforestry technologies.
As a result, agroforestry research has been given high priority by
ISRA because of its present and potential impact on rainfed cereals
production particularly with regard to improving and maintaining soi1
fertility.
ISRA recognizes that the use of minera1 fertilizers alone Will
not be an economically viable method for the GOS to achieve its goals in
rainfed cereals production at least in the short to medium term.
Because of
its potential to increase soi1 fertility, ISRA also recognizes that
agroforestry has the potential to complement varietal improvement research
which to date has met with limited success due to poor soi1 fertility and
plant-water environment.
3.0 ISRA'S CURRENT AGROFORESTRY RESEARCH PROGRAM
3.1 Projet de Recherche - Développement sur le Rôle de l'Arbre en
Exploitation Agricole
This activity, designed to support research on the role of trees in
farming systems, is currently the main component of ISRA/DRPF's
agroforestry program.
The project, funded by FAC at a level of
150,000,000cfa began in March 1987 and is scheduled to terminate in
December 1988.
The objectives of this project are to:
.
(1) i;ni;Sa:; village level agroforestry activities in Senegal in
.
.
-reintroduce trees in Senegalese farming systems.

---

3
-determine the biological, social and economic importance
of trees in these systems.
(2) Reinforce individual and collective agroforestry actions by a
program of complementary on-station and on-farm research.
(3) Assure farmer participation in project activities.
ISRA and FAC initially chose to work in three agroecological zones
each with its distinctive set of natural resource and production problems;
the Northwest Peanut Basin, the Southern Sine Saloum region, and the Lower
Casamance Region.
These regions also benefit from a fairly extensive ISRA
forestry and agriculture research presence as ISRA research stations are
located in a11 three regions.
It was hoped that the presence of these
stations as well as personnel from ISRA's other divisions (DSAEA, Animal
Production) would facilitate program implementation. The project's two
primary modes of action are:
(1) to conduct diagnostic surveys to describe the rural milieu, to
identify problems and determine with the farmers what
agroforestry techniques are most appropriate;
(2) to develop a research development component comprising three
types of actions:
- on-station research
- on-farm research
- development/extension of technologies;
According to the FAC project paper, ISRA/DRPF was to collaborate with
ISRA's DSAEA and Animal Production Departments, with ORSTOM and with the
Directorate of Soi1 Conservation and Reforestation under the Ministry of
Nature Protection (MPN/DCSR).
Technical assistance to the project is provided by CTFT and includes
the project coordinator and one researcher. ISRA/DRPF has assigned four
researchers (including two "stagiares"), one Agent Technique des Eaux et
Forets and four "encadreurs/observateurs"
to the project.
The following section describes in detail the project's actions in
each region along with some observations and suggestions for program
improvement.
3.1.1 The Northwest Peanut Basin (Thienaba)
3.1.1.1 Agroecological Zone Description
This region is characterized by poor sandy soils, with organic matter
content between 0 and 3%.
Intensive millet-peanut culture in the area has
had a severe. impact on soi1 fertility. Moreover, this is the Acacia albida-
millet zone, but natural regeneration of A. albida has decreased
substantially over the past several years.
Most ISRA/DRPF researchers
attribute this decline to increased use of animal traction in the area and
increased demographic pressure, although these hypotheses have not been
substantiated.
However, older trees are systematically being harvested for
fuelwood and/or lopped for fodder in the dry season. Rainfall appears to be

---

4
declining;
the average for 1931-1960 was 670mm compared to 395mm for
1986-87.
3.1.1.2 On-Station Research (Thienaba)
Past and present on-station species trials and protocols at Thienaba
are presented in Annex 1.
In 1985, this station also initiated alley
cropping/line planting trials.
Five species (Acacia albida, Acacia
tortilis, A. nilotica, A. senesal and Prosopis iuliflora); and four
treatments (with and without rhizobium inoculation, with and without
nematocides) in 4 repetitions.
Tree lines were originally spaced at 5m but
light competition forced ISRA to eliminate one line of 2 for a final 10m
spacing.
Within line tree spacing was 2.50m.
Crops for the intercropping component are based on the traditional
system of rotation;
cowpeas (year l), peanuts (year 2) and millet (year
3).
While these trials are still on-going, preliminary results indicate
that:
in years one and two, when the lines were expanded to lOm, there was
a slight decrease in trop yield.
Nematode treatment appeared to have the
biggest positive influence in trop yield. However, there was some
interaction between nematode treatment and rhizobium particularly with
peanuts (decrease in production). In terms of tree species, it is too early
to determine species performance although A. tortilis. and A. senesal appear
promising.
The results in A. albida are inconclusive as the provenance used
may not be adapted to Thienaba conditions.
3.1.1.3 On-Farm Research (Khayes)
This is the only one of the three agroforestry sites in which the
diagnosis was conducted independent of ISRA/DSAEA. Apparently, ISRA/DSAEA
did not consider this area to be a priority in terms of its program. As an
alternative approach, ISRA took advantage of its "memoire de confirmation"
process to use an ISRA/DRPF stagiare to examine physical and socio-economic
conditions in a village (Khayes) near the station in view of establishing an
eventual agroforestry management plan.
This study was conducted in early
1988 by Samba Ndiaye and is entitled "Etude des Facteurs Physiques et
Socio-economiques Utiles a 1'Etablissement d'un Plan d'Amenagement
Agroforestier" (Ndiaye 1988).
While Mr. Ndiaye would have undoubtedly
benefitted from participation in an inter-disciplinary systems team, his
study nevertheless makes an important contribution to understanding the
rural milieu from a natural resource perspective rather than from the more
traditional ISRA/DSAEA cropping systems perspective.
In particular, the study documents the decline of Acacia albida in
the area and suggests that the increased use of animal and traction may be
responsible for this decline.
The study also identifies certain regions
where regeneration of A. albida has been maintained or increased implying
that farmer perception and cultural practices have an important role in
maintaining the A. albida - millet system. Mr. Ndiaye was also able to
identify five distinct vegetation zones in the study area (Borassus
aethiooium, Adansonia diqitata, Parinari macroohvlla, Detarium microcaroum
and Prosoois africana) which interestingly enough, correspond neither to

---

5
soi1 type nor rainfall.
Again, this would imply that predominant vegetation
in the study area is again a factor of cultural practices and farmer
perception.
In short, Mr. Ndiaye has identified areas which merit
additional research and would obviously benefit from a systems approach.
Based on Mr. Ndiaye's study, ISRA/DRPF has begun a small on-farm
research - development program in Khayes which to date has focused on:
natural regeneration of Acacia albida - (painting seedlings a
bright orange in order that farmers cari see them and plow around
them).
ISRA/DRPF intends to measure trop yields around these
trees based on age class.
establishment of a village forest and fruit tree nursery and a
village (community) fruit orchard with mangos, lemons, oranges,
guavas.
ISRA loaned the village sufficient fencing to establish
a lha fruit orchard in a "bas fond" donated by the village.
Hedgerow trials are located inside the fente using single
species and/or combinations resulting from the on-station
research.
ISRA also assisted the village in establishing a well
within the orchard.
Villagers paid for fruit tree seedlings
while hedgerow
seedlings were provided free of charge. The
village supplied a11 labor for fencing, tree establishment and
well digging but Will return the fencing to ISRA when the living
fente is sufficiently established to deter animals.
3 . 1 . 1 . 4 COWlENTS/OBSERVATIONS:
On-farm-on station linkaqes:
Unfortunately, the on-station trials
were established prior to Mr. Ndiaye's study. The result is essentially a
top-down research program.
For example, the station researchers are pushing
the Australian Acacia SDD as a "good" hedgerow/windbreak species because of
their relatively fast growth and coppicing characteristics.
However,
initial farmer experience with the species in the area indicates that they
are not compatible with crops.
A closely related species, Acacia manqium,
is known to have allelopathic characteristics and it may be that the
Australian species being tested on the station may also be allelopathic.
Moreover, the form of these acacias (relatively open on the bottom) may not
be the best for trop protection either from animals or desiccating winds.
The on-station-on-farm linkage problem is further compounded by the
fact that there appears to be little communication between the Senegalese
and the expatriate researcher. Mr. Ndiaye apparently has little input into
decisions made on-station.
There is also a question on the ability of farmers to replicate some
of the technologies currently being tested on-station.
It is extremely
doubtful whether farmers in the area Will be able to purchase nematocides or
anti-desiccants, particularly considering their difficulty in purchasing
other, more common inputs such as fertilizer.
A system of research
priorities needs to be established at the station based on Mr. Ndiaye's

---

6
study and research obtained on station to date. This issue Will be
discussed in detail in a later section.
fidditi a1 smcies; While there has been considerable attention paid
to local specyis, some indigenous species which appear to have good hedgerow ;i
characteristics have been overlooked.
For example, Guere seneqalensis
(SP.) is a shrub common to the area appearing as part of the fallow
Farmers note that crops planted where the species was removed tend to
better than other areas.
This would indicate a nutrient rich litterfall.
The species also has a good hedgerow form with dense foliage a11 the way to
ground level and coppices well.
However, the species is usually never
allowed to reach its maximum height as it is continuously being tut for
firewood and other uses (baskets).
Given this shrub's characteristics, it
would appear to be worthwhile to include it in any hedgerow research
protocol.
The Acacia albida-millet svstem:
The peanut basin continues to
be the prime traditional area for the Acacia albida-millet agroforestry
system, but little research is being done on this important system. This is
even more surprising considering that the general consensus among natural
resource professionals in Senegal indicates that the use of the system is
declining.
Thus, there appears to be a need for compromise between research
on the introduction of new agroforestry systems and species, and study or
refinement of viable, old systems.
Questions that ISRA/DRPF could address
on the A, albida system include:
& eh@
Why is the system declining in some areas and being maintained fy~~
or increased in others?
Are there lessons to be learned
(socio-economic, cultural or biological) from areas where the
J& /,s.t
system is being maintained which cari be applied to other areas? I'lw-
What are the best methods for replicating/maintaining the system
on-farm; natural or artificial regeneration?
If artificial, by
what means? How many trees cari be maintained per ha without
interfering with trop production?
A-6
1 +
LD /7,M k-ou t f!
& L a & g &?&t.k Ad Jp1
C'~' /,& da*
LA
What are the constraints on extension and maintenance of the “ &'-- . *
system?
TO what extent has animal traction contributed to the declin
the system? Are farmers willing to plow around naturally seede
seedlings or is it better to establish them in line plantingsî
Are there any other factors which are contributing to the
species decline?
A nationwide, if not rangewide, provenance test of A. albida
appears to be warranted. This could perhaps be done in
collaboration with an IARC such as ILCA or ICRAF.
brimenta desisn: The experimental designs at the Thienaba
station are extremely complex.
Agroforestry research complex, but when it

---

7
is compounded with numerous species,
rhizobium and nematocide trials, the
analysis of variante and determination of statistical
significance of the
various species and treatments become extremely difficult.
(In fact, none l
of the published results of the trials gives the statistical significance r
and level of confidence.)
The general rules of thumb for agroforestry research design are: (i)
simplicity (no more than 2 species - 2 treatments);
(ii) repetition (at
least three);
(iii) randomization;
(iv) control; and ( ) to "block" for
differences in soi1 type and for slope whenever possible Y . On-station
trials should be simplified into smaller units rather than one large tria1
which attempts to address everything.
Considerably more attention should be given to within and between
row spacing in terms of the alley cropping trials,
and in diagonal planting )c.;*, ,nLz
of the hedgerow associate species trials. There Will of necessity be a
i
compromise between the number of trees a farmer cari plant in his field (due
to labor and time constraints, etc.) and the number required to maximize
trop production.
In the trials, parallel planting takes considerably more
space out of production than diagonal planting.
Hedgerow "yield" and
effect may not be influenced by diagonal planting and may even be enhanced.
Research vs. Develooment: The majority of on-farm activities
conducted by ISRA/DRPF have a distinctly development inclination.
Very
little on-farm research (other than the diagnostic study) has been conducted
to date.
Development actions are necessary to gain farmer confidence, but
the DRPF should begin to move some of the more promising species and trials
on-farm as soon as possible.
In this regard, Thienaba is somewhat behind
the other two stations.
J
3.1.2. Sine Saloum
3.1.2.1 Agroecological Zone Description
This area is currently experiencing a rapid population increase, with
reports indicating that population has doubled over the last 10 years. -2
3 dIAA-
t>
Soils are variable but generally poor with low organic matter content. The
population pressure has increased pressure on the natural resource base to
the extent that traditional farming systems are beginning to break down.
Soi1 erosion and decreased fallow periods are becoming more of a problem and
are major contributing factors to declines in agricultural productivity.
Overgrazing has also recently become a problem and also contributes to the
decline in forest caver.
Rainfall has declined in the area from an average
of 927mm between 1931-60, to an average of 640 in recent years with a slight
increase to 850 mm during 1986-87.
lThe Thienaba station has different soi1 types whose effect could have
been removed by blocking.

---

8
3.1.2.2 On-Station Research (Nioro du Rip)
Research on this station has primarily focussed on living fente
trials and species and protocols are found in Annex 1. There are no
intercropping trials at this station.
3.1.2.3 On-Farm Research
As part of his "memoire de confirmation", M. Diatta, an ISRA/DRPF
research trainee, is currently studying:
the correlation between vegetation and soi1 types in the region
(soi1 evolution/regeneration);
m
the influence of the different vegetation types on run
off/erosion in collaboration with ISRA's Soi1 Conservation
Program and ORSTOM;
Results of this study are expected by the end of 1988.
The Sine Saloum ISRA/DPRF team is also conducting research on the
effect of Woody species on trop yields.
In the area, four species are
i'.,-
t3 d2
commonly found in farmers' fields: A. albidq, Cordvla oinnata, Prosoois ]
gfricang and lannea acida
The objective of this experiment is to measure
Iff14'r,+LL-
trop yields under the crains of these species in order to determine optimum A-
JjS
tree density/ha.
Additionally, factors such as litterfall as a function of ?"‘>' ,..--J--
tree size, quality of litterfall/forage, and fruit production per diameter
_CC~-@'
class Will be examined.
Two villages were selected during the course of the diagnostic survey
(with the Kaolack Agrarian Systems Team for research-development type
4
actions:
Ndimb Taba and Sinthiou Kahel . ISRA/DRPF research - development
activities in Sinthiou Kahel include:
the establishment of a village
nursery and training of a nursery technician; the establishment of live
fentes on farmer's fields (using preliminary results from the on-station
trials), and planting fentes along contours whenever possible to help stem
soi1 erosion).
Hedges are also being established to control animal circulation with
the goal of protecting crops in the rainy season and natural regeneration in
the dry season.
Villagers apparently requested this action. Ownership of
hedges and rights to eventual benefits have been determined at the village
level. Each farmer has the rights to the hedgerow north and east of his
2Unfortunately, due to an overflowing river, the author was only able
to visit the village of Sinthiou Kahel.
On farm research at Ndimb Taba is
reported to include activities similar to Khayes', plus windbreak
establishment and soi1 erosion control measures. The diagnostic survey
identified loose soils and absence of vegetation as key problems in Ndimb
Taba, combined with severe water erosion.

---

field.
ISRA/DRPF also provided Eucalvptus camaldulensis seedlings to a
farmer interested in establishing a private woodlot.
As in Khayes, ISRA/DPRF is attempting to maintain natural
regeneration by placing painted stakes near important species.
Their
intention here is to stake not only A. albida, but other species such as
P. africana, Lannea acida, etc.
In practice,
however, the author only saw
staked A. albida and not too many of those.
When sufficient regeneration
has been established, ISRA/DPRF hopes to be able to eventually develop
"lines" of local species (by thinning) to facilitate plowing.
3.1.2.4 Comnents/Observations:
While not as sophisticated as the trials at Thienaba (Khayes),
ISRA/DRPF research at Sine Saloum is progressing quite well. Some areas
for improvement include the following:
On-station trials:
There was no repetition in the species/hedgerow*
t?/ ;?Y
trials on the grounds that these were only aptitude tests. Without
is:
I
p- g
repetition, variation in soils, microclimate, etc. may give misleading
Ik,+ {te'
M
results in terms of species performance.
It is suggested that a11 future
* c c+A' ,
species trials follow the protocol of simplicity, randomization, repetition ~n~~~D*'Y""'
(and control).
& ,'o.J jm., G!R bu, fA.a+'-
It is also suggested that intercropping (alley croppi:g) trials be
initiated both on-farm and on-station.
The station (Nioro du Rip) has
sufficient land to accommodate new line planting/intercropping trials while
ISRA's pilot villages appear to be interested in on-farm tree planting.
Choice of sr>ecies:
Several of the species used in the original
trials should be eliminated.
For example, A. albida does not make a good
1; IA ftt@
+
hedgerow species, while species such as D. melanoxvlon and S. soinosa, while t#,4 -,L!'@"
extremely valuable, are too slow growing to be of use in a hedgerow
. \\," , “..'" tr' 4e.*br
configuration.
Some thought needs to go into species choice prior to @cIf;B 0~ #?'(;y
- c. protocol establishment in order to avoid dispersion of effort and resourtes. F'~~ .i,.;
p acw 4 "'<. /sj+jv$~;:e 404
Villaqe Knowledse, Attitudes, Practices: There appears to be a
0.‘ L-,..-,,-,"
difference in knowledge, attitudes and practices between the villages of
Khayes and Sinthiou Kahel.
In this latter village, ISRA/DRPF was able to ,.
install more private on-farm triais than in Khayes, where the villagers
allocated only a 1 ha area of communal, non-arable land in a "bas fond".
comparative study of these two villages might contribute to a better
understanding of agroforestry technology transfer.
Research vs. development:
As in the case of Khayes, many of
ISRA/DRPF's actions in the village are more development than research
oriented.
ISRA/DRPF needs to reconcile these two roles and initiate closer
collaboration with GOS and NGO development agencies and projects if it is
going to maximize its impact.
(For example, both the Director of the DRPF
and the author were concerned about the establishment of a widely spaced
Eucalvotus camaldulensis plantation in what is currently a productive
millet field.
While the request for this plantation was made by the

---

10
proprietor, it seems a shame to take viable cropland out of production with
a species that is known for its allelopathic effects as well as its high
,,
water requirements.
Alternatively, ISRA/DRPF should have assisted the
i!lA
farmer in establishing a narrow-spaced production plantation, particularly
on the more marginal sites of his farm.)
3.1.3 Casamance (Lower)
3.1.3.1 Agroecological Zone Description
This area still has a relatively rich natural resource base in terms
of forests, estuaries, range and wildlife. Many of the forests, however,
are being rapidly depleted and villagers note the increased absence of
ronier palms (overcutting) and mangroves (salinization). Traditionally
Senegal's prime rice producing area, soi1 erosion is now rampant in some
areas due to inappropriate cultural practices on the watersheds. As a
result, many paddies are now experiencing severe siltation problems.
Rainfall has been decreasing from an average of 1547mm between 1931 and 1960
to 1007mm in 1986-1987.
Soi1 structure in the region is generally weak with a low water holding
capacity and they are extremely susceptible to erosion.
Soi1 fertility is
low where the natural vegetative caver has been removed and where the fallow
period has 'been shortened or absent.
Soils are generally acidic and given
the region's higher rainfall, acidification progresses rapidly once the soi1 .&+- '
caver is removed.
The cost of restoring these soils once acidified cari be ,.:
extremely high.
3.1.3.2 On-Station Research (Djibélor)
Since 1985, ISRA/DRPF in collaboration with ILCA, has been conducting
provenance tests on 13 provenances of Gliricidia seoium. The objectives of
the test are to determine the growth rates of the various provenances, the
effect of the "green manure" obtained through litterfall on soi1 fertility,
and to determine optimum harvesting dates.
The best provenances have been
identified and are now established in a seed orchard.
Alley cropping research is also being conducted at Djibélor with
Gliricidia seoium and leucaena leucoceohala.
The objective of the research
is to determine if alley cropping is an alternative to shifting
cultivation/reduction of fallow periods.
Tests are also being conducted to
determine alley cropping's influence on fertilizer uptake and soi1
evolution.
The experimental design is eight treatments and four repetitions
as described in Annex 1.
3.1.3.3 On-Farm Research (Boulandor)
Prescriptions for on-farm research/village selection stem from
ISRA/DRPF and the Agrarian Systems Team. On-farm trials include:

---

1 1
study of the influence of wooded vegetation strips (E.
çamaldulensis, G. seoium and 1, leucoceohala) on trop
productivity and soi1 erosion.
on-farm alley cropping between line spacing trials with G.
sepium.
initial research on improved fallow systems with the annuals
Teohrosia SDD. and Crotalaria SDD. and the Woody species A.
manaium and @. holosericea.
Research - development type activities include:
training of one village nurseryman and one vi llage fruit tree
"grafter".
establishment of a 1 ha mango orchard. The s ite is initially
fenced with wood stakes in order o protect a live fente of A.
mellifera for the first l-2 years 5 .
on-farm hedgerow trials - Jatrooha curcas, hrvthrina
senesalensis and Morinqa oleifera (by cuttings); A. albida and
z mauritiana (by direct seeding); and z, mauritiana and 0.
melanoxvlon (by seedlings).
protection of natural regeneration (by staking) on a number of
species including A. albida, P. africana and Tamarindus indica.
amenity4 plantings in the village.
3.2.3.4 Comments/Observations
On-station allev croopinq trials: Considerable data has been
collected on these trials to date, but the results/conclusions Will not be
available until the end of 1988. Since these are the first "truc" alley
cropping trials to be conducted in the Sahel and outside of IITA (and of
considerable importance to the Casamance region), ISRA/DRPF needs to
carefully evaluate and statistically analyze them. The design of the trials
is not overly complicated, but future trials should probably be separated by
species and blocked for soi1 and slope if necessary.
n
q
Soaci
: More attention needs to be given to between and within row
spacing.
As stated earlier, there Will of necessity be a compromise between
the spacing required for optimum trop and tree production and what the
farmer Will be able to establish on his fields.
3The physical quantity of wood used in the fente indicates that
firewood is not a problem in the area.
4Shade trees, lining roads, etc.

---

On-farm allev croopinq soacinq triais: While ISRA/DRPF Will
undoubtedly gain some important information from these trials, the results
Will not be valid since spacing was not selected at random but
systematically.
The protocol for measiring trop production was
be done systematically as well (1 15 m plot in the middle) but
discussions during this mission, harvest sites will be selected
thereby adding a degree of credibility to the results.
Veqetation bandq:
ISRA/DRPF calls these "bandes anti-erosifs", but
their ability to stem soi1 erosion- since they have been established on
fallow fields where erosion is prac$cally nil- cari be questioned. (In
1;1
fact, the predominant use of 1. ca a dulensig may pnhance
:
soi1 erosion, as
the species tends to eliminate vegetative caver and channels the water
@
rather than slowing it down and "spreading" it.)
The real erosion problem in the Casamance arises from the traditional
practice of tilling fields parallel to the slope. ISRA/DRPF needs to
change its focus in this critical area from vegetation strips to contour
land preparation and simple "passive" erosion control measures such as
debris strips, and/or QZ_SS strips wit~.,_o__r__w-~~ho.~t~.~r.~es,
An on-farm study
'Y' 6, <
of the effect of alley cro~gX'ï%oil erosion appears warranted since this
method of soi1 protection has proven to be successful in other countries.
L ucaena leucoce~hal~:
While this species grows quite well in the
lower CaEamance, its use on-farm appears extremely problematic due to
browsing.
Basically, the local animal population appears to prefer
leucaena to a11 other species.
DRPF attempts to protect the species in the
vegetation strips by using wooden "cages" Will more than likely result in
A ’ using more wood than the strips Will eventually produce.
Other Dotential allev croooinci soecies: Given the potential problems
/
with L, leucoceohalq on-farm, it is suggested that ISRA/DRPF explore
,
alternative tree species. Of particu ar interest would be Calliandrq a**--'-'
'
calothrvsus, Casuarina eauisetifolia, ?J Sesbania qrandiflora (or S. sesbans)
and perhaps some local fast growing N-fixing trees such as Albizzia SDD.
Imoroved fallow svstems:
The continued use of Teohrosia SDD and
Crotalaria SDD is recommended.
These species have proven quite benéficial
in Rwanda in i&reasing soi1 organic matter content.
However, rainfall Will
ultimately determine the success or failure of these species in the
Casamance.
The use of the Australian acacias in improved bush fallow systems
\\ ,.,
should be discontinued.
A. mansium is known to be allelopathic and farmer
experience with A. holosericea in Thienaba (Khayes) indicates similar
+
potential.
However, research on improved bush fallow systems appears to be!
warranted in the Casamance and elsewhere in Senegal. It is therefore
5An agroforestry project in Kenya has had particular success with this
species in a similar environment.
The problem is that it does not coppice
and therefore has to be completely replaced after harvest.
,; *,, j F ! CI,, &[ ! iJ>'*'

---

suggested that
local species (e.g. site indicators, good pioneer species, etc.) which may
be planted or sown in order to advance succession and reduce fallow periods.
During the course of the visit, the villagers identified several species
(unknown to this writer) which appear to have considerable potential.
ISRA/DRPF should take advantage of indigenous knowledge systems in this
important area.
Ronier Daim:
Discussions with the villagers indicated that they were
dismayed with the rapid disappearance of the ronier palm and the dieback of
the mangroves.
While little cari be done with the latter, perhaps ISRA/DRPF
ir/& A
could initiate a small ronier palm plantation effort, at least for future
Boulandor generations.
Methods of propagation of the species need to be /
explored, however.
On-farm hedqes:
ISRA/DRPF has had considerable success with the
reintroduction of Morinsa oleifera, a multi-purpose tree used for fodder,
fencing, fuelwood, food, and medicinal purposes. The species had
apparently a11 but disappeared in the area and its cultivation in the form
of a hedgerow deserves special attention.
Research vs. Develooment: As in the case of the other two areas,
many of ISRA/DRPF’s actions at Boulandor are development rather than
research oriented.
However, in the case of Boulandor, the villagers appear
to be particularly receptive to new ideas/techniques, etc. However, some of
these techniques, such as vegetation bands, Will have little if any impact
in spite of the many hours of village labor that went into their
establishment.
ISRA/DRPF's key in conducting on-farm research in Boulandor
Will be to sustain a high level of village/farmer interest in agroforestry
activities without jeopardizing the confidence of the villagers towards
DRPF staff as a result of inappropriate technologies.
4.0
SUGGESTIONS FOR STRENGTHENING ISRA/DRPF’S AGROFORESTRY/NATURAL
RESOURCE PROGRAJ4
4.1 Determination of Agroforestry (and Forestry) Research Priorities
The D PF's agroforestry research priorities are for the most part
1
well-defined .
However, given its limited staff and budget and considering
the wave of interest in agroforestry, donor and individual researcher
interests in the future may complicate priority determination.
6With the somewhat arbitrary exception of nematocide and
anti-desiccant research.
Additionally, as Gold (1987) points out, a large
proportion of DRPF research consists of species introduction trials leftover
from a variety of projects.
Limited resources require ISRA/DRPF to evaluate
these trials and eliminate those that are no longer of interest to current
priorities.

---

1 4
In order to assist ISRA/DRPf, Annex II provides a list of "Suggested
Criteria to Prioritize Research Needs" adapted from Wadsworth (1983). The
key to using these criteria is the notion that the best applied research
~ ;J. .I #
stems from local participation in problem identification and research design c.
and implementation, and not from a top-down predetermination of research
I
,+,. i
needs.
;I'r /WI ff*m n,,m,L,&P &J
4.2 Linking Research with Policy, Prices and Institutions hw- ,‘c4** I’ ‘y”*
!
C “‘?7Gc’
For the time being, most of ISRA/DRPF’s research focusses on
\\
{ ‘I j / .,1' 3
biological aspects and neglects the socio-economic and institutional
\\
__..-.-
context and the macro-economic setting. The development of successful '
agroforestry and related technologies requires that biological and
socio-economic on- station and on-farm research be linked to broader
economic and policy issues.
Several examples of the importance of this
linkage in Senegal cari be cited.
4.2.1 Markets
.
Studies by the USAID-funded Senegal Reforestation Project indicate
that there is virtually no demand/market for Eucalvptus SDD. (or other)
poles greater than about 10 cm in diameter. This is supported by the
experience in the Bayotte "forêt classée" where DRPF staff were unable to
market the poles (20cm diameter) from a coppicing trial. Nonetheless, most
DRPF research on Eucalvotus SDD focusses on the rotation periods required
to maximize wood productton rather than rotations for diameter/market
considerations.
A similar problem exists with gum arabic (Acacia senesal). Despite
considerable donor and GOS inputs into gum arabic plantations, the program
is unsuccessful due to depressed world prices, tribal differences over
harvesting and marketing rights and the fact that traditionally, Seneqalese
have never srown gum arabic as a trop as is common in other countries .
Tree tenure also appears to be a problem particularly in the Northern
Peanut Basin.
4.2.2 Land and Tree Tenure
Under Senegal's current forest code a11 trees belgng to the State and
farmers must have permission from a local forestry agent to harvest them.
Currently, a new forest code, which has been awaiting approval from the
National Assembly for over two years, provides for some transfer of
responsibility for tree planting and harvesting from the national to the
local, or at least community level.
However, it is unclear how the code
Will address private/individual tree planting activities which are
necessary for agroforestry program success. As the revised code has been
7According to Mark Freudenberger, UC/Berkeley.
81n most cases, considered to be "policemen" by the farmers rather than
extension agents.

---

15
before the National Assembly for over two years, chances that it Will appear
in its original form appear slim according to most Senegalese officiais.
Additionally, a hypothesis currently being formulated by some
Senegalese and expatriate foresters is that given that a new forest code
Will eventually be instituted, wealthy or better educated farmers are
planting more trees (or protecting natural regeneration) than poor farmers
in order to position themselves for the day when tree establishment may be
equated with security of land tenure. This may partially explain why A.
albida is declining in some areas of the Peanut Basin while increasingin
number in other areas.
4.3 Suggestions for Expansion of ISRA/DRPF'S Current Program
The following areas require significant additional attention by the
DRPF over the next several years.
4.3.1 Erosion Control (water)
Research on dryland water management techniques presents
opportunities for the development of technologies which may have both short
and longer term impacts on trop production.
Considerable practical
knowledge and research data are currently available from the Sahel and other
countries which ISRA/DRPF might use to build a more cohesive watershed
management program. Of particular interest and relevance to Senegal are:
Water harvestinq techniques: - i.e., micro-catchments,
diguettes, etc., which are capable of capturing rainfall and
runoff and channeling it to a specific area for storage and
plant use.
On-farm research in Yatenga, Burkina Faso indicates
that water-harvesting techniques have the possibility of
"extending" the rainy season by up to three months. While labor
intensive, these techniques have been widely accepted by farmers
in the Yatenga area.
While primarily intended for agricultural
crops (sorghum and millet), trees have been successfully
introduced into these systems in Burkina.
v
Soi1 Structure Imorovement:
ISRA/DRPF is currently working on
a number of techniques to increase soi1 organic matter content,
including alley cropping and line plantings. However, one area
that deserves particular attention is improved bush fallow
systems and the use of green manure.
Some work has begun on
this topic, but much more needs to be done.
Soi1 Preparation and Passive Erosion Control Techniques:
Techniques that show considerable promise include contour
plowing, tied ridgesg and post-harvest plowing. Research and
development are quite advanced on these techniques in the Sahel
gICRISAT has an excellent tied ridge program in Burkina Faso.

---

1 6
but may need to be adapted/tested to the conditions of the
Senegalese farmers.
This implies a very strong role for
ISRA/Agrarian Systems Teams.
ISRA/DRPF's role in this type of
research will obviously be limited.- However, the DRPF could
make substantial contributions to the development of other on-
farm erosion control methods such as debris strips, and more
active methods such as grass strips (with or without trees) and
alley cropping.
The point is that research in these areas must
be inter-disciplinary.
4.3.2 Erosion Control (wind)
On-farm windbreak research in the Maggia Valley in Niger indicates
that the use of single line Azadirachta indicq (neem) windbreaks cari
increase millet yields by up to 30% while providing a substantial source of
income through the sale of poles and firewood. Research has shown that
windbreaks significantly reduce plant stress during the critical time of
sowing/germination and that survival of Young plants is improved due to
_--- _-.---.."_," __,- __.___--l_._C_-
reduced damage from moving Sand,
-3
--a-_
Research in Sudan and Egypt has also shown that windbreaks and
shelterbelts cari substantially reduce evaporation from irrigated perimeters
as well.
This may have particular significance in the Fleuve area in the
near future.
ISRA/DRPF has some windbreak trials near Podor and Ndimb Taba, but
they were not visited during this mission. However, before expanding its
windbreak program, ISRA/DRPF should address several questions:
Determination of soi1 types most prone to wind erosion and
selection of the best multi-purpose species for the soi1 type.
Determination of the optimum planting configuration and species
mix with particular reference to protection of the windbreaks
from browsing during initial establishment.
'
Determination of the optimum harvesting schedule and technique
for the particular tree species.
4.3.3 Natural Forest Management
The emergence of natural forest management as a technique for
fuelwood production is only beginning to gain some recognition in the Sahel.
However, almost a11 of the fuelwood in the Sahel as well as many secondary
and tertiary products corne from the natural forests and woodlands. While
these forests have been used for multiple and sustained purposes by rural
people over the centuries, little has been accomplished in terms of
research on biomass productivity, cutting yields, rotations as well as
socio-economic considerations involving use of forest resources.
However,
research in Niger (USAID's Forestry Land Use Planning Project- FLUP)
indicates that when rationally exploited, natural forests and woodlands cari

---

17
be more productive than traditional Eucalvptus SDD. block plantations
while yielding a, wide variety of other economic benefits from secondary and
tertiary forest products not found in intensive plantation culture.
Additionally, research on the FLUP project indicates that the key to
"rational" management of natural forests and woodlands is the involvement of
local people in the decision making process, and transferring management
responsibilities from the capital city to the local population.
However, these type of actions are in direct conflict with Senegal's
current natural forest management strategy of "mise en defense" and
enrichment planting strategy.
1 t has generally been proven (Burundi and
Kenya) that it is impossible to keep the rural population from using
protected natural forests (forets classees). Moreover, evidence is
accumulating that increased yields obtained through enrichment plantings do
not caver establishment costs (South America, Burundi).
Considering that Senegal has over 50 "forêts classées," most of which '1
are severely degraded, a small pilot natural forest management research
project, similar to the FLUP project in Niger, could be initiated in a foret.(
classee near one of ISRA/DRPF’s stations or villages.
‘.. i I
,,c" f
.‘
4.3.4 Supporting and Component Research
~
ISRA/DRPF's collaboration with other research institutions such as
ORSTOM appears instrumental in providing the basic research for improved
agroforestry systems.
ORSTOM/DRPF has already made substantial
contributions in isolating effective rhizobial strains for Prosopis SPD
Albizzia SDD
and Leucaena leucoceohala which have been incorporated inéo
many of the DRPF's on-station and on-farm trials.
Time did not permit a review of ORSTOM's research nor a review of ISRA’s
trop production/soils/agronomy research, and its relevance to the
agroforestry program.
However, of particular interest to DRPF's current
program would be research on tree-plant water relationships, the
allelopathic potential (positive or negative) of agroforestry tree species,
research on Frankia SPD. for Casuarina SPD., the potential for use of
vesicular-arbuscular mychorrizae (VAM) on agroforestry species as a means
of enhancing water and phosphorus uptake, as well as additional work on the
impact of trees on soi1 characteristics such as moisture content, organic
matter content, temperature, structure, infiltration rates, etc.
ORSTOM Will also be adding a micro-propagation (tissue culture) lab
to its facility and it is foreseen that this Will also play an important
role in the development of more viable agroforestry systems through the
propagation of genetically superior material.
4.4 Suggested Areas for Additional Long-Term Training
In addition to the two candidates currently scheduled for long term
training in genetics and agroforestry, long-term training in the following
areas would contribute substantially to a more problem-responsive DRPF.

---

18
Forest/Natural Resource Economist:
The DRPF is in desperate need of :
an economist capable of addressing both micro (farm issues) as well as the 'i f y'
aforementioned macro-economic policy issues.
One of the complaints most
often heard about ISRA/DRPF is that research priorities and technologies
/
developed in ISRA/DRPF have been done with little regard to their financial
and economic viability.
In particular, it is feared by many "development"
foresters that some of the more intensive agroforestry technologies may not
be viable, based on farm labor and capital constraints.
ISRA/DRPF'
first
priority in terms of long-term training should be to fil1 this post 40.
Michigan State University has excellent agriculture and natural resource
economics programs.
Social Forester/Rural Sociolowist: On-farm socio-economic
conditions, constraints and opportunities on the research and development of
improved agroforestry systems was to have been the responsibility of
ISRA/DSAEA in the original FAC-funded agroforestry project. However, for a
number of reasons, as Will be discussed later in this report, the Systems
Teams have had relatively little input in terms of agroforestry research and
development.
For this reason, the author suggests that ISRA/DRPF make every
effort to "acquire" and train its own "social forester".
Again, MSU could
provide this type of training.
Watershed Manaqement/Soil Erosion Soecialist: ISRA/DRPF's program
would benefit considerably by a person specialized in arid land watershed
management/soil erosion control.
Washington State University and the
University of Arizona both have good programs for this specialization.
WindbreaksDhelterbelts Soecialist: Again, the DRPF's program would
be considerably strengthened by the addition of a MSc-level researcher
specializing in windbreaks and shelterbelts. The Universities of Kansas and
Nebraska (and most "plains states" universities in the U.S.) a11 have
excellent windbreak/shelterbelt programs.
Natural Forest Manacrement Soecialist: Long-term training in natural
forest management should be oriented away from the traditional U.S. focus on
plantation and monocultures and instead concentrate on sustained yield and
multiple use of natural forests.
Yale and Oxford (UK) have natural forest
management programs that might be worth exploring.
4.5 Short-Terra Training
The following short-term training activities would contribute
directly to DRPF's immediate needs:
-
Natural Forest Manasement: A visit to the FLUP project in Niger
by Ibrahima Thomas.
loNote that ISRA/DRPF at one time did have a forest economist on their
staff but economists are in such high demand in Senegal that he resigned for
more lucrative employment.

---

19
Forest Administration Short-course: This course is sponsored by
USDA/USAID's Forestry Support Program (FSP). It is intended for
top level forest administrators and researchers and would
present an excellent opportunity for the Director of the DRPF to
strengthen his research administration skills.
Brochures on the
course should be available at the AID Mission or through the
Forestry Support Program.
Manioulation and Analvsis of Svstems Teams Data:
Should the
difficulties involved in the intra-departmental use of this data
be resolved, it is suggested that the DRPF request USAID
assistance to organize a workshop for already computer-cognizant
DRPF staff in the manipulation and analysis of Agrarian Systems
Team data.
4.6 Collaboration with the IARCS.
ISRA/DRPF’s experience wlth ILCA in the provenance testing of
Gliricidia sepium has been very favorable with benefits occurring on both
sides.
ISRA/DRPF gains experience in provenance testing, experimental
design and seed orchard development while ILCA benefits from an extended
national presence.
It is suggested that the DRPF explore with ILCA the
possibility of provenance testing of other forage species, including AA
albida.
ISRA/DRPF should also explore collaborative research arrangements
with IITA,
ICRISAT/SAFGRAD and ICRAF. For example, IITA is currently
attempting to strengthen its own alley cropping research program by
developing stronger linkages with national research institutions.
USAID is currently financing a major portion of ICRAF's Agroforestry
Research Network for Africa (AFRENA) project. While the first phase of this
project was to concentrate on Africa's bimodal highlands, the second phase
Will focus on the arid and semi-arid areas.
Furthermore, the author
understands that ICRAF and ICRISAT have entered into a collaborative
agreement for supporting national agroforestry research in the Sahel.
ISRA/DPRF should explore this possibility as well.
4 . 7 Information/Documentation
There is currently a very low level of information exchange between
Senegal and other Sahelian countries.
This is particularly distressing
because,of the many similarities in sector problems that exist among these
countries.
While there is no substitute for site-specific research, there
is a great deal that cari be learned both in terms of approaches to problem
solving and in basic technologies which transcend national boundaries.
As a
result, the DRPF library is several years out of date in both Sahelian and
other relevant literature.
The author has developed an annotated bibliography on agroforestry in
the Sahel which was supposed to have been forwarded to ISRA/DRPF but was not
available during the course of ,the mission.
Another copy Will be forwarded

---

2 0
to both USAID/Senegal and DRPF in order that photocopies of the more
relevant materials cari be provided to DRPF. DRPF should also request
support from USAID for subscriptions to relevant scientific journals such as
Aqroforestrv Svstems Journal and International Tree Crops Journal. (MSU's
Forestry Department could ensure that photocopies of other relevant
literature is made available to the DRPF over the remaining life of the
project.)
Finally, while the francophone-anglophone issue remains a critical
issue in information dissemination, it is understood that the USAID Africa
Bureau NRMS project Will provide funding for the translation
(English-French, French-English) of documents, articles and reports which
have wide application.
ISRA/DRPF and USAID/Senegal should be able to
benefit from this relatively new "innovation" in information exchange.
4.8 Short-term Technical Assistance
During the course of the author's visit to ISRA/DRPF's field stations
and on-farm research sites, the most pressing need for short-term advice is
in the area of experimental design and statistical analysis of results. In
almost every trial, at least one of the four laws of simplicity,
randomization, replication and control was broken.
‘DRPF should request USAID to provide the services of a consultant to
conduct a 2-3 day workshop on agroforestry (and other) experimental designs
and on their statistical analysis.
This consultant should spend an
additional 3-4 days working with individual researchers on developing
appropriate designs for the next season's trials.
5.0 THEMES FOR DISCUSSION BY ISRA’S SCIENTIFIC AND TECHNICAL COMMITTEE
In the context of strengthening ISRA/DRPF's research program and in
addition to the issues and recommendations discussed above, two additional
themes require the attention of ISRA’s Scientific and Technical Committee:
intra and inter-agency collaboration.
5-l Intra-Agency Collaboration
5.1.1 Problems
Gold (1987) provided two suggestions for improved ISRA intra-agency
collaboration;
improved project design protocols and research design for
rural development.
ISRA/DRPF has made considerable progress in
implementing these suggestions.
Planned experiments address national
priorities and focus on key problems, as well as permit a certain continuity
or follow-through, albeit dependant on external funding.
Feedback on research is now enhanced through a series of monthly
meetings in which DRPF scientists present the results of their work.
However, interna1 review of design protocols, at least in agroforestry,

---

21
still appears problematic and perhaps contributes to the generally poor
experimental designs and analyses.
ISRA/DRPF has also made considerable progress in moving research from
the station to the farm, thus enhancing its potential contributions to rural
development.
However, station-farm linkages are weak in certain regions,
financial analyses are almost never conducted on technologies, and many of
DRPF's on-farm actions are better classified as development rather than
research activities.
While DRPF is working to alleviate some of these problems, one of the
major stumbling blocks in conducting more appropriate on-station and on-farm
research has been a lack of collaboration between ISRA/DRPF and the
ISRA/DSAEA Systems Teams in diagnosis, research design, implementation and
evaluation.
While the Systems Teams did provide some assistance to the DRPF
in Boulandor and in Sinthiou Kahel, it was done more on a persona1 and
informa1 (ad hoc) level than by any forma1 obligation.
Issues in achieving closer collaboration include:
Different Geosrahic Priorities: DRPF's and DSAEA's geographical
(regional and village) priorities do not coincide.
For this reason, DSAEA
was unwilling to collaborate with DRPF in the "diagnosis" of the Khayes
village.
However, the village and area in general are representative of
production and natural resource problems in the Peanut Basin as well as
being the prime Acacia albida - millet zone.
Difficult access to DSAEA Datq: Currently, at least in the
-‘\\,
Casamance, DRPF must Zay DSAEA for the analysis of the Systems Team's
diagnostic data.
DSAEA Will not provide DRPF with the data disks for
$2..
;‘-
interna1 use/analyses although DRPF researchers are competent in a wide
:
variety of computer applications.
DSAEA's apparent fear is that DRPF (and
1
other ISRA departments) Will use this data to "publish" reports and papers
prior to the Systems Teams.
Research rivalry and the push to publish are /
academic problems that a nascent research institute cari il1 afford.
Additionally, while the Systems Teams' diagnostic approach was to
have been a compromise between the "rapid appraisal" method favored by U.S.
farming system experts and the lengthy survey method prescribed by the
French, most diagnostic surveys now exceed two years and in some cases are
approaching five years as in the case of the Fleuve region (Montagut 1988).
Delays of this length cari hardly be justified in the context of Senegal's
rapidly declining natural resource base.
A contributing problem is DSAEA's reluctance to initiate on-farm
activities prior to complete data analysis and publication. As a result,
diagnostic information may not be valid in a particular village by the time
the data have been analyzed and published.
No Svstems Team Follow-UD:
Farming (and agroforestry) systems
research needs to be inter-disciplinary in both the diagnostic and the
research design and implementation phases.
However, it appears that in the

---

2 2
case of the Systems Teams, the diagnostic phase is inter-disciplinary but
that the research design and implementation (and evaluation) phases are not.
Diagnostic results are simply given to single discipline researchers for
implementation.
According to DRPF officiais, there is no "systems" input
into the research and development process once the diagnosis has been
completed causing some to refer to the process as "diagnose and disappear".
Crowinq Svstem vs. Farminq Svstem Focus: As is the case of most
farming system activities in sub-Saharan Africa, ISRA's farming systems
research cari be more aptly described as cropping systems research. The trop
components are given considerably more attention than the farming system as
a whole.
When foresters or livestock people are omitted from the teams, the
role of trees and livestock in the farming system (even if considered to be
a major component of the system) are generally ignored or relegated to
"second phase" studies.
5.1.2 Suggestions for Improved Intra-agency Collaboration
The resolution of the above issues cari be a problematic, lengthy and
costly process.
Considering the extent of Senegal's natural resource and
agricultural productivity problems, the question to ask is whether ISRA/DRPF
cari afford to spend the time and resources necessary to promote improved
long-term intra-agency collaboration on agroforestry research and
development or develop its own diagnostic capabilities. Thus, two
strategies are available for the Committee's consideration.
In the first approach, ISRA/DRPF could move gradually toward the
recruitment and training of its own agroforestry systems team composed of a
forestry/agriculture economist, an agronomist, a forester/agroforester and a
rural sociologist/social forester.
Ideally, this team would be able to draw
on other ISRA researchers (animal production, etc.) when particular
situations warrant.
This team would be responsible for a11 agroforestry
diagnosis and design activities and would backstop a11 subsequent on-station
and on-farm research, including monitoring and evaluation activities. The
team would be located in Dakar but would travel to a11 regions for
backstopping purposes.
Additionally, the team could be made available to
the numerous DCSR projects on a time available, contractual basis.
An alternative approach which may at least alleviate some problems in
the short-term and lay the groundwork for improved longer-term
collaboration would focus on stronger joint programming. In this approach,
it is suggested that:
194's Scientific and Technical Committee address the issue of
priority regions for the DSAEA and the DRPF. If DSAEA and DRPF
were at least operating in the same regions, some degree of
collaboration could probably be obtained at least at the
informa1 level.
DRPF should be involved in a11 diagnostic activities conducted
by the DSAEA.
While this may detract from DRPF's current
activities, DRPF participation is seen by this author as an

---

23
investment in the development of agroforestry technologies more
appropriate to farmer needs.
In short, the investment should
pay considerable benefits over the long term.
DSAEA make a11 diagnostic data and analyses available to the
DRPF within a timeframe that reflects the need to act while
farmer interest is high.
Alternatively, DSAEA data disks should
be made available to DRPF researchers. As previously stated, a
short course in systems data analysis may be appropriate for
DRPF researchers.
Questions regarding the length of the diagnostic process as well as
the lack of inter-disciplinary follow-up in the farming systems research
process are beyond the scope of this report, but are nevertheless topics
with the Committee should address.
The issue to keep in mind is that the
financial and economic costs and benefits of both of these approaches need
to be carefully weighed beforc any decision is reached.
5.2 Inter-Agency Collaboration
While not specifically called for in the mission's terms of
reference, another major problem that ISRA’s Scientific and Technical
Committe
needs to address is the lack of collaboration between the DRPF and
the DCSR f 1
5.2.1 Problems
Annex III provides only a partial list of DC 8 projects. Out of
these 21 activities, DRPF is only involved in three T
Considering the
proximity between the DRPF and the DCSR in Hann, visits between DRPF and
DCSR project and administrative personnel are rare. Some of the stated
reasons (real or imagined) for this lack of collaboration include:
“ISRA/DRPF does too much on-station research and their results
are not applicable in the real world";
"the results of ISRA/DRPF's research are never communicated and
when they are, they are never in a form which cari be used by
DCSR agents";
llNote that there is little collaboration among DCSR projects, but a
planning/coordination project currently being designed by FAO Will no doubt
alleviate this situation.
While FAO envisions a role for ISRA/DRPF in this
project, details are not yet available.
12Plantations Irriguees dans la Vallee du Fleuve, Projet d'Amenagement
et de Reboisement des Forets du Centre-Est, and Projet de Reboisement
Villageois in Louga.

---

2 4
"given its location in the Ministry of Rural Development,
ISRA/DRPF is isolated from the DCSR and any15eal world problems.
DRPF needs to be brought back under the MPN ";
-
"ISRA/DRPF researchers are not interested in the
administrative/"negotiative" aspects of collaboration,
preferring research to a11 other activities";
"there is no financial or economic analysis of any of
ISRA/DRPF's research activities. How do we know what the
financial incentives to the farmer are?"
Evidence of this lack of coordination is apparent in the activities
of both organizations.
As previously stated, many of the DRPF's on-farm
activities are more development that research-oriented.
Similarly, many of
the DCSR projects contain a research component, although as the Directorate
of the DCSR stated, these research activities are often a result of donor-
imposed bilateral requirements.
However, DRPF is usually never consulted
on these project-related research activities.
As a result, there is much
duplication of effort between the DCSR and the DRPF, a tendency to "reinvent
the wheel" and a tendency for the DCSR to focus on technologies and species
which the DRPF has already determined to be inappropriate.
The case of collaboration in agroforestry research and development
has its own unique set of problems.
Agroforestry to many DCSR officiais
means everything from composting to village woodlots to the planting of any
combination of crops and trees.
For example, in some areas, the DCSR is
advocating the intercropping of E. camaldulensis with manioc.
The use of
these two species in combination Will render the site completely useless for
any future agriculture or forestry activities.
As difficult as it might be, scarce financial and intellectual
resources in Senegal require close collaboration between the DCSR and DRPF.
DRPF should be able to draw on DCSR projects to help define and refine its
research priorities.
DRPF should be able top provide the projects with
concrete applied results.
Increased collaboration would also avoid
duplication of effort by allowing the DRPF to fulfill its research mandate
while allowing the DCSR and projects to use their resources on other
priority areas.
5.2.2 Suggestions for Improved Inter-Agency Collaboration
A Research-Extension Workshop:
An attempt was previously made by the
DCSR to address research-extension linkages in a one day workshop in Sokone
in April 1988.
However, research-extension linkages was only one of five
agenda items and as a result, there has been very little follow-up to
13Note that bringing DRPF under the MPN would even further distance it
from ISRA's other departments, particularly agriculture/DSAEA. The costs
and benefits of such an action need to be carefully weighed.

---

2 5
workshop recommendations14.
Given this situation, the author suggests that
ISRA/DRPF and DCSR consider conducting a more in depth research-extension
workshop which would serve to:
bring together representatives of a11 forestry-oriented
organizations in Senegal having a research interest;
encourage a high level of collaboration and cooperation among
these organization both in Dakar and in the regions;
work out additional means for increasing research efficiency
both within and among members of the forestry community.
A similar workshop was conducted in Kenya in 1983 where research -
extension linkage problems were very similar to Senegal's. Rather than
discuss this workshop in detail, a copy of the proceedings Will be forwarded
to the DRPF for review and consideratiun.
DRPF Actions:
DRPF needs ta take the initiative in establishing
contacts with DCSR projects.
An initial small investment of DRPF time and
resource
Will more than likcly have considerable medium and longer term
benefits 55.
Furthermore, this type of collaboration needs to be done from
the bottom-up and not through any sort of higher level decree. In a similar
vein, ISRA/DRPF needs to assess the potential and constraints of its
current collaborative efforts with DCSR in view of documenting successes and
alleviating future problems.
Research Coordination:
DCSR needs to at least "pass" a11 project
implemented research activities to DRPF for comment.
Even if DRPF cannot
participate in these activities due to time and resource constraints, the
DRPF needs to be aware of what's going on and have the opportunity to
provide comments and suggestions.
DCSR Aqent Rotation:
Another potential means of enhancing
collaboration between the DCSR and the DRPF would be to have the DCSR agents
spend up to two years of their professional careers in the DRPF as research
assistants as part of DCSR's normal rotation system.
This would have the
advantage of familiarizing DCSR agents with DRPF research which would
hopefully carry over into the agents professional careers with DCSR.
14The workshop stressed the need to create research committees
involving project directors, the "chefs d'inspection" and regional DRPF
researchers in order to better formulate regional research priorities.
However, to the author's knowledge, regional reports and recommendations
have not as yet been forthcoming.
15Note that ISRA/DRPF was to have participated in USAID’s Cereals II
project but negotiations broke down when a request for vehicles by DRPF was
denied.
Collaboration, if only on a small scale, should corne before
material needs in most cases.

---

26
Increased DRPF Comunications:
The complaint that DRPF does not
present its research findings in a form which cari be utilized by technicians
appears to have some validity.
it is therefore suggested that DRPF publish
some "how to" manuals in French and Wolof based on the results of past and
current research programs.
A simple manual on appropriate agroforestry
techniques/species seems particularly important.
Additionally, the DRPF should expand/continue its on-station on-farm
research field days.
An agroforestry field day was held in Thienaba, where
project, ISRA, DCSR and other staff were invited to see ISRA/DRPF actions on
the ground.
This program should be expanded to the other regions and
conducted on a yearly basis.

---

2 7
A N N E X 1
SDWARY OF ISRA/DRPF ON-STATION SPECIES-RELATED
RESEARCH
ON-STATION RESEARCH (THIENABA)
1985 to Present
Living hedges - seedlings
Indigenous
Australian
Bauhinia rufescens
Acacia holesericea
Ziziohus mauritiana
Acacia tumida
Combretum aculeatum
Acacia sclorsperma
Balanites aesvptiaca
Acacia trachvcaroa
Prosoois cineraria
Living hedges - direct seeding
Bauhinia rufescens (1 line)
Ziziphus mauritiana (2 lines-parallel)
C. aculeatum t Z. mauritiana (2 lines-parallel)
Z. mauritiana t C. aculeatum t Z. mauritiana (parallel)
Acacia mellifera t Z mauritiana
(2 lines-parallel)
B. rufescens t Z. mauritiana (2 lines-parallel)
Fruit tree/field trials
Sclerocarva birrea
Parkia bislobosa
Anacardium occidentale
Tamarindus indica
Adansonia disitata
Balanites aesvotiaca
Ziziohus mauritiana
Forage species trials
Acacia trachvcarpa
Combretum aculeatum
Prosoois cineraria

---

2 8
1988 - Trials - species performance:
8 species and four treatments -
Balanites aeovntiaca
Parkinsoni a aculeata
Acacia tortilis
Bauhinia rufescens
Acacia seneoal
Ziziphus mauritiana
Acacia nilotica
Prosopis iuliflora
Treatments include 2 methods of seed treatment (pre-germination and
sulfuric acid) and with and without mulching.
ON-STAION RESEARCH (NIORO DU RIP1
Living fente trials were initiated in 1987 on the following species (a11
single row):
Balanites aesvotiaca
Acacia mellifera
Acacia seval
Z, mauritiana t C, aculeatum (alternating)
Z mauritiana t B. rufescens (alternating)
A: albida - direct seeded
Spacing within lines is 50 cm and between lines 4 m - no repetition.
In 1988, several additional species (single line, living hedges) were
planted with no repetition:
Direct Seeded:
A. seneoal
A. laetq
Dalbersia melanoxvlon
Mavtenus senesalensis
Strvchnos soinosa
Prosoois cineraria
From cuttings:
Commiohora africana
Erithrvna senesalensis
In association (2 lines 50 cm diagonal spacing, no repetitions):
Combretum aculeatum + A. ataxacantha
B. rufescens t & ataxacantha
Commiphora africana t Z. mauritiana
D. melanoxvlon t Z. mauritiana
Xvmenia americana t Z. mauritiana

---

2 9
ON-STATION RESEARCH (DJIBELbQ
Alley cropping trials
1. control:
crops without fertilizer
2. crops with fertilizer
3. Leucaena alley cropping without fertilizer
4. Leucaena alley cropping with 1/2 recommended fertilizer
5. Leucaena alley cropping with full fertilizer
6. Gliricidia - no fertilizer
7. Gliricidia - 1/2 fertilizer
8. Gliricidia - full fertilizer

---

31
ANNEX II
SUGGESTED CRITERIA TO PRIORITIZE RESEARCH NEEDS IN AGROFORESTRY16
1.
Urgency
Leads to solution of one or more immediate, critical problems
:: Early results to be expected
C .
High demonstration and educational value
2.
Lack of Alternatives
Not already solved by experience locally or elsewhere
b: Not already being studied adequately somewhere
C .
Not likely to be undertaken soon otherwise
3.
Compatibility
Not incompatible with local traditions and customs
b: Compatible with prospective land use needs
Utilized familiar techniques and existing talent and skills
d: Enthusiastically supported by affected communities
e.
Has cooperative multi-donor support
4. Returns
Benefits quantifiable
ab: Contiguous
intra-regionally exportable technology of utility to a
large area'or population
High output per unit of land area
i: High output per unit of input
e.
Meets diverse forest product requirements, not just one
f. Large prospective gain in foreign exchange
9.
Large auxiliary multiple benefits: food, fodder, soi1 and water
conservation, timber, other products
h. High leverage, large gain for small donor input
5.
Residual Benefits
Degree to which improved national policies a likely result
i: Degree to which local capability to continue research is enhanced
Technical training value
d: Degree to which door opened to further research
Degree to which self-propelled research is apt to continue
F: Degree to which self-sustaining private efforts stimulated
161t would be an intersting and extremely useful exercise to use these
criteria to weigh ISRA/DRPF's current program.
This could perhaps be
achieved under ISRA’s "memoir de confirmation" process.

---

JJ
ANNEX III
PARTIAL LIST OF CURRENT MPN/DCSR PROJECTS
NAME
R E G I O N
DONOR
Boisements Communautaires
Diourbel
AFRICARE
Fixation des Dunes du Gandiolais
Gandiolais
CANADA/CIDA
Fixation des Dunes et Protection des
Kebemer
USAID
Cuvettes Maraicheres
Gommier Gonakier Vallee du Fleuve
Fleuve
Netherlands
Gommier Podor
Podor
W. Germany
Gommiers et Reboisement Pastoraux
Mbiddi
C R D I
Plantations Irrigu
s dans la
Fleuve
Vallee du Fleuve f f
Projet d'Amelioration des Methodes
Dakar
USAID
de Carbonisation Energies Renouvelables
Projet d'Amenagement e de Reboisement des
Kaffrine
Forets du Centre-Est f8
Projet d'Amenagement et de Reboisement
Fleuve
W. Germany
Senegalo-Allemand
Projet Anacardier Senegalo-Allemand
Sokone
W. Germany
Projet de Boisement Villageois19
Louga
UN/FAO
Projet de Reboisement Villageois de
Bakel
UN/FAO
Bakel
Projet de Co-operation pour la Promotion
Thies
Japan
de la Verdure du Senegal
171SRA/DRPF is currently negotiating a joint research-development
project with DCSR for this activity.
181SRA/DRPF provides assistance to this project in terms of provenance
testing of Acacia nilotica and A. raddiana.
lgISRA/DRPF assist this project with speNcies/provenance trials.

---

34
Projet Poles Verts
Richard-TO11
Projet de Protection Forestiere de
Casamance
Canada/CIDA
Casamance
Projet Reboisement Villageois dans le
Thies
UN/FAO
Nord-Ouest du Bassin Arachidier-Tivaouane
Projet Tobor
Ziguinchor
Reboisement Communautaire dans le Bassin
Fatick
FAO/Finland
Arachidier
Reboisement, Fixation des Dunes, et
Diander
USAID
Protection des Cuvettes Maraicheres
du Kayar
Projet de Reboisement du Senegal
National
USAID

---

35
ANNEX IV
PERSONS CONTACTED
ANAM
FAO Forestry Project (and coordination)
BA Amadou: Chef d'inspection Forestiere, Fatik
BAKHOUM Abdoulaye, Observateur ISRA/DRPF, Sine-Saloum-Sinthiou Kahel
BENDER Ozzie, Marketing Consultant, Projet Reboisement du Senegal (USAID)
BJORKDAHL Goran, Coordinateur, Projet Boisement Villageois (FAO)
BONNER Jim, Assistant ADO, USAID/Sencgal
CAMARA Mamadou, Observateur ISRA/DRPF, Casamance-Boulandor
DIAITE Ibrahima, Chercheur ISRA/DRPF, Casamance
DIATTA Malainy, Chercheur Stagiare ISRA/DRPF, Sine-Saloum
DIONNE , Chef Secteur Forestier, Gossas
ELLIS Jane, USAID/Senegal
FICKES Jim, TA Team Leader (SECID), Projet Reboisement du Senegal (USAID)
FREUDENBERGER Mark, Research Associate (Acacia seneqal), UCLA
GAYE Moctar, Production Economist, ISRA-Kaolack
GAYE Abibou, Researcher, ISRA/DRPF
GUEYE Moumar, Project Coordinator, Project Reboisement du Senegal (USAID)
JOHNSTON Alan, Associate Director/Rural Development, Peace Corps/Senegal
LEWIS Scott, Associate Director/Forestry, Peace Corps/Senegal
MBAYE Ndiaga, Directeur General Adjoint, ISRA
MONTAGUT Gerard, Unite de Planification, Programmation et de Formation, ISRA
NDIAYE A.Samba, Chercheur Stagiare ISRA/DRPF, Thienaba
NDOUR Babou, Chercheur ISRA/DRPF, Sine-Saloum
NEME Jean-Pol, Conseiller Technique (Conservation des Sols), Direction de
Conservation des Sols et du Reboisement
NIANG Moktar, Directeur de Conservation des Sols et du Reboisement, MPN
OUEDRAOGO Ismael, MSU-SARI1
RUELLE Pierre, Thematique, Direction Systemes Agraires et Economie Agricole,
ISRA-Sine-Saloum
SALL Bokar, Directeur des Eaux et Forets (Administration), MPN
SALL Pape, Directeur ISRA/DRPF
SARR Alioune, Observateur ISRA/DRPF, Thienaba-Khaye
THOMAS Ibrahima, Chercheur (Directeur de Station) ISRA/DRPF, Casamance-
Djibelor
VERCHOT Louis, Forestry Advisor, Projet Reboisement du Senegal (USAID)
WADE Momar, Agent Technique Eaux et Forets, Attache a ISRA/DRPF, Thienaba
WILLIAMS Ann, MSU-SARI1

---

3 7
ANNEX V
FIELD TRIPS
September 12
DRPF Station at Thienaba and on-farm
agroforestry trials at Khaye
September 14
DRPF Station at Sine-Saloum (Kaolack)
and on-farm agroforestry trials at
Sinthiou Kahel
September 15-17
Casamance:
DRPF Station at Djibelor,
on-farm agroforestry trials at
Boulandor, and natural foret management
at the Foret Classe de Bayottes
September 20
Gossas:
collaborative field trip with
ISRA/DRPF, DCSB and Projet Reboisement
du Senegal

---

39
ANNEX VI
BIBLIOWWHY
Cazet M (1987) "Plantations en lignes et cultures intercalaires sur les sols
sableux degrades de la zone centre nord du Senegal: premiers
resultats de l'experimentation conduite a Thienaba depuis 1985",
ISRA/DRPF, Dakar.
Charreau C (1970) "Amelioration du profil cultural dans les sols sableux et
sablo-argileux de la zone tropicale seche Ouest-Africaine et ses
incidences agronomiques: association arbres/cultures", IRAT, Dakar.
Charreau C and P Vidal (1969) "Influence de l'acacia albida sur le sol, la
nutrition minerale et les rendements des mils Pennisetum au Senegal"
IRAT, Dakar.
CTFT (1966) "Acacia albida: etude de la croissance - rapport annuel 1966",
CTFT, Dakar.
Dancette C and PL Sarr (1985) "Degredation et regeneration des sols dans
les regions centre-nord du Senegal', ISRA/DRSPTTMR, Dakar.
De Haut de Sigy (1962) "Rapports d'enquetes effectuees dans la zone d'action
du Centre Experimentale de regeneration et de conservation des sols
de Thienaba", CNRA/Bambey.
Gautreau J (1967) "Rapoort de camoasne 1966- essais sur les Caddes",
IHROICNRA; Bambky .
’ ”
Gaye Matar (1988) "Enquete sur l'acquisition et l'util isation des intrants
agricoles dans les regions de Fatick et Kaolack : campagne 1986-1987",
ISRA/DRSAEA, Dakar.
Gaye Matar (1987) "La problematique de l'engrais dans le bassin
arachidier",
ISRA/DRSAEA, Dakar.
Gold MA (1987) "Senegal Agricultural Research adn Planning Project: Trip
Report for ISRA”, MSU/Department of Forestry, E. Lamsing.
IHRO (1966) “Influence de l'Acacia albida sur la culture de l'arachide",
IHRO/CNRA, Bambey.
ISRA (1988) "Strategies et programmation des recherches 1989-1993:
presentation et synthese", Ministere du Developpement Rural, Dakar,
ISRA/CNRF (1986) "Programme de recherche-demonstration prevu par
l’ISRA/CNRF a l'annee 2 du projet agroforestier conservation des sols
et des eaux", ISRA/DRPF, Dakar.

---

40
ISRA/DRPF (1987) 'Rapport technique d'activites 1987', ISRA/DRPF, Dakar.
Jung J (1971) "Variations saisonnieres des caracteristiques
microbiologiques d'un sol ferrugineux tropical peu lessive (dior)
soumis ou non a l'influecne d'Acacia albida", Oecol. Plant., Gauthier
Villars, Paris.
Kelly VA (1986) "Farmers' demand for fertilizer in the context of Senegal's
new agricultural policy:
a study of factors influencing farmers'
purchasing decision, INA, Dakar.
Lebrun J (1968) "A propos du rythme vegetatif de l'Acacia albida",
Collectanea Botanica, Paris.
Linehan Peter (1986) "Cereals II Agroforestry Pilot Project: Final Report",
USAID/Dakar.
Louppe Dominique (1987) "Projet de recherche-development sur le role de
l'arbre en exploitation agricole: Rapport Annuel 1987", ISRA/DRPF,
Dakar.
MDR (1965) "Programme d'aide a la classifica,tion des cultures, Operation
Cadde", Ministere du Developpement Rural, Dakar.
MENR (1983) "Proceedings of the national workshop on strengthening forestry
research in Kenya, November l-4, 1983, Eldoret, Kenya", Ministry of
Environment and Natural Resources, Nairobi.
. MER (1966) "Protocole d'etudes sur l'Acacia (albida", Ministere de
.
1'Economie Rurale, Dakar.
MER (1963) 'Le cadde, arbre miracle au Senegal", Ministere de 1'Economie
Rurale, Dakar.
Ndiaye Samba SA (1988) "Etude de facteurs physique et socio economiques
utiles a l'etablissement d'un plan d'amenagement agroforestier:
cas
de Khayes", ISRA/DRPF, Dakar.
Poulain JF (1966) "Bilan des experimentations realisees au CERCS de
Thienaba" IRAT/CNRA, Bambey.
4
Tourte R et a1 (1964) "Bilan d'une rotation quatriennale sur sole de
regeneration au Senegal", AGROTROP no, 12, Dakar.
Wickens GE (1969) "A study of Acacia albida",, Kero Bulletin.

---