Final repor: activities, findings and conclusions of the Range Management Improvement Project 608-0145, Morocco

Material Information

Final repor: activities, findings and conclusions of the Range Management Improvement Project 608-0145, Morocco
Place of Publication:
Logan, Utah
Utah State University ; USAID ; Ministere de l'Agriculture et de la Reforme Agraire, La Direction de l'Elevage
Publication Date:


Subjects / Keywords:
Africa ( LCSH )
Farming ( LCSH )
Agriculture ( LCSH )
Farm life ( LCSH )
Forage ( jstor )
Ewes ( jstor )
Lambs ( jstor )
Spatial Coverage:


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

Record Information

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


This item has the following downloads:








































































































































































































































































































Full Text
Range Science Department USAID Ministere de 1'Agriculture
Utah State University et de la Reforme Agraire,
Logan, Utah La Direction de l'Elevage
August 1986

Range Science Department USAID Ministere de 1'Agriculture
Utah State University et de la Reforme Agraire,
Logan, Utah La Direction de 1'Elevage
August 1986

Applied Research 2
Extension Program Development. . . . . . . 4
Training 5
The Plant Materials Center . . . . . . . 5
Applied Research Range Forage Production . . . 8 Applied Research Range Livestock Production . . 12 Applied Research Sociology . . . . . . 16
Applied Research Range Economics . . . . 18 Program Development Extension. . . . . . 21
Training . . . . . . . . . . . 22
The Plant Materials Center . . . . . . . 23
Introduction . . . 26
Identification of Major Forage Resources:*Beni Meill. 27 Identification of Major Forage Resources: Midelt . 31 Identification of Major Forage Resources: Oujda. 33 Identification of Major Forage Recources: Timahdite. 40 Forage Production Alternatives . . . . . . 43
Results and Conclusions of Adaptability Trials by Site 45 Seeding and Planting Techniques. . . . . . 52
Land Treatments. 59
Deferment from Grazing ........ 67
Forage Quality of Annual Vegetation. ......... 69
Conclusions and Recommendations. . . . . . 72
Introduction 78
Current Livestock Management Practices: Midelt . . 82 Current Livestock Management Practices: Azrou. . . 93 Current Livestock Management Practices: Beni Mellal. 101 Alternative Livestock Management Practices . . . 104 Conclusions and Recommendations. . . . . . 118
Introduction 128
Sample Selection: El Faija . . . . . . . 130
Characterization of Producers: El Faija. . . . 131 Assessment of Producer Resource Sets: El Faija . . 132 Timahdite . . . . . . . . . . . 147
Sample Selection 147
Characterization of Producers: Timahdite . . . 147 Assessment of Producer Resource Sets: Timahdite. . 149 Ait Rbaa .. . . . . . . . . . 162
Sample Selection ..... .. 162
Characterization of Producers: Ait Rbaa ...... 162 Assessment of Producer Resource Sets: Ait Rbaa .... 165 Ain Beni Mathar. .. ..... ... 174
Characterization of Producers: Amn Beni Mathar .... 174 Assessment of Producer Resource Sets: Ain Beni Mathar. 177 Summary of Model Production Units: Ain Beni Mathar . 179

Assessment of Producer Perceptions of Current Production
Assessment of Producer Expectations . . . . . 191
Assessment of Producer Perceptions of Needs, Limitations,
Issues and Opportunities . . . . . . . 192
Assessment of Social and Cultural Influences on Livestock
and Crop Production and the Acceptance of New
Technology 194
Introduction 200
Results 202
Conclusions and Recommendations. a o e e e * 224 Literature Cited .. 226
Introduction ............... 228
Summary of Informatio n a Development 1:io*Li7vestock
Production System Models for Identification of
Intervention Opportunities . . . . . . 228
Extension Program: Direct Interventions . . . 228
Extension Program: Teaching Revegetation Techniques and
Animal Husbandry 2135
Development of Extension Materials . . . . . 241 Conclusions and Recommendations . . . . . . 242
CHAPTER SIX: TRAINING. .. ...... . . . . . 245
Introduction 246
Long-term Degree Training . . . . . . . 246
Short-term Training 246'
Administrative Short Course . . . . . . . 248
In-country Seminars. .. .. . . . . . . 248
Professional Meetings . . . . . . . . 249
Computer Training Seminars . . . . . . . 253
Sheep Selection Training at Oujda and Safi . . . 253 Conclusions and Recommendations . . . . . . 253
Introduction 256
Development at the PMC . . . . . . 257
On-site Nursery Operation by the Oujda Office . . 264 Conclusions and Recommendations . . . . . . 267
Introduction 270
Project Commodities 270
Plant Materials Center . . 271
Informational Resources. . . 274
Construction ............. 277
Microcomputer Rom......... 279
Conclusions and Recommendations . . . . . . 281
Appendix A: Project Staff and Contributers . . . 284 Appendix B: List of Abbrevations Used . . . . 287 Appendix C: Arabic Words Used in the Text . . . 288
Appendix D: Documents Written by Project Personnel and
Available in the DE/SP Library . . . . . 289


The Range Management Improvement Project (RMIP) (USAID
Project #608-0145) began in March 1981 and ended in August 1986. The project was implemented by Utah State University (USU) under a host country contract with the Direction de l'Elevage (DE) of the Ministry of Agriculture and Agrarian Reform (MARA), and was directed at strengthening the institutional capability of the Service de l'Amenagement et de la Mise en Valeur des Terrains de Parcours (DE/SP).
The RMIP was designed to assist in the planning and
implementation of extension programs in range management and range livestock improvement. The specific goal was to improve both livestock production and production efficiency and thereby increase the incomes of Moroccan livestock producers.
The first project evaluation was carried out in January 1984 and, based on the need for quantified information and recommendations of the evaluation team, resulted in various changes in the project design. A more concretely defined strategy emerged and the RMIP management structure was reorganized to ensure improved coordination. The redesigned project consisted of four components:
* applied research,
* extension,
* long- and short-term training, and
* the development of a Plant Materials Center (PMC).
Project activities were centered in five provincial offices of the Service de l'Elevage: Beni Mellal, Meknes, Midelt, Oujda and El Jadida. The geographic location of these sites of field activity are indicated on the map of northern Morocco presented as Figure 1.
RMIP COMPONENT ONE: Applied Research
It had become obvious in the early years of project implementation that it was necessary to undertake certain research tasks in order to develop a credible extension and range management program. DE/SP effectiveness depended upon further developing the capability to plan, implement and evaluate research efforts and to utilize the research results. In order to assist the DE/SP in meeting information needs, the RMIP applied research activities focused on:
* evaluating current production systems,
* analyzing production problems,
* identifying viable production alternatives, and
* determining methods of information transfer that would
assure acceptance of superior alternatives.

00 ....... ......
.......... .......
................ ......
... ...... % ................... .........
............................. ......
...... .. ........
............... .......
....... ....
...... ........................ .... .. ...
............ ....... ..... .. ........ .....
........ ...... . . . . .. .................
........ .............. . . . . .
........ ....... ... Te
........ ........ o an ............. ......
............... : : : : : : * * :"*,** .... *.*.*
....... ..... .. :: :: : : :: : : : : : : : * , * *
ftJ ......
.... .... ... ... .......................
0 1-4 1-4 ....... ..............
rt rt OQ .......... Larache .............. .......
9: .................. ...................
............. ...
........ ...... Chechaoue ..........
m .................... ..... .......
Zouada Ranch
m 93 Oujda
4- V
. ........ ..
....... Kenitra
lb e
........... Rabat
f-t :J
-4 0 ......... Meknes uercif
H- rt
rt V : .... Ben Slimane Khemisset
14 m
Casablanca Ain Beni Maihar
Tizi Tin
A4. Btt *0%4%
El Jadida Azrou
.......... PtA ,(#A
0 0 ......... Had S ulern. Ranch Adarouch PtumttA w
P. 0
CtAfft 0
o n
Settat Timahdite
0 0
Khenifra ft Faija PtAimCC4 004b ...................
................... 0
................... A;,t Xbda F&AiMU4
.................. Kasba Tadla
.................. M idelt
................ Deroua pt"At de t!A&ud Peuwtvt
................ Beni M ellal

Specifically, evaluating and analyzing current production systems and problems involved:
identifying the sources and seasonal availability of
livestock feed and forage including natural vegetation,
fallow, crop residues and supplemental feeds, and
estimating production parameters, estimates of total
livestock production and production efficiency by level
of management.
Identifying viable production alternatives included:
* estimating current and alternative livestock/crop
production costs and returns,
* assessing potential alternatives including the seeding of
marginal cropland and selected range sites, manipulating
the soil surface to enhance moisture infiltration and
grazing management, and
* balancing animal nutritional requirements with available
forage and feeds, culling and selection programs and
synchronization and timing of lambing.
Determining acceptance of superior alternatives involved:
* identifying producer perceptions of resource availability
and the current production system, and
* determining the social and cultural ramifications for
technological production alternatives.
Models of the current livestock and forage production
systems in the various zones were needed in order to gain more complete understanding of production problems and to provide means of simulating the effects of production alternatives. Through the research effort the project strengthened the ability of DE/SP staff to improve range management practices and assure active producer participation.
The RMIP applied research program was designed to create the basis of a functional extension program. Extension program development relied on:
developing audience profiles from the stratification of
the producer population,
assessing production and profit estimates of current
production systems by audience, and
identifying intervention opportunities and potential
acceptance of alternative strategies.
Once information had been synthesized, extension program
implementation involved:
* formulating technical interventions relevant to the
identified audiences,
* planning an extension program based on those

interventions, and
* evaluating audience response.
Thus, an extension effort was started to address critical producer needs for forage resources and animal management. The extension materials developed from the RMIP research are appropriate for the range livestock production systems in Morocco. DE/SP staff have been trained to continue the program and have been instrumental in its establishment.
RMIP COMPONENT THREE: Long- and Short-Term Training
Since continued program success depends upon the capability and credibility of DE/SP staff, training may well prove to be the most essential component of the project. Training efforts made through RMIP included:
* long-term degree training in the U.S., short-course training in the U.S., and
* administrative, computer and extension seminars in
Specifically, through long-term training,
* RMIP participants were to receive MS degrees in range
management, extension and rural sociology. All of these
participants were to complete their programs and to be
contributing to the range management and range livestock
improvement program of MARA by the end of the project.
Through short-term training,
* non-degree training in range management and extension
practices was to be provided in the U.S. for DE
technicians and administrators,
* two one-week technical seminars were to be sponsored
through the RMIP in collaboration with other Moroccan
institutions, and
* one-day RMIP seminars were to be sponsored at different
sites throughout the country to train local DE
Again, in terms of long-term impact, training may prove to be the most successful RMIP component.
RMIP COMPONENT FOUR: The Plant Materials Center
The Plant Materials Center (PMC) was created as a support facility for the RMIP research and extension efforts. In addition, the PMC was designed with the long-range objective of multiplying seeds of forage species and (as more producer needs were identified) of producing fuel and fodder shrubs for revegetating Moroccan rangelands. Thus, the PMC functions included:

*producing seed of range forage species,
*developing a shrub seed production nursery and increasing
shrub production for the perimeters,
*producing certified seed for the National Seed Marketing
Company (SONACOS) to generate revenue and introduce the
PMC to the certification system, and.
The most essential function of the PMC may prove to be collecting and assessing native and exotic species by:
*establishing nursery plots for collected native species,
*evaluating and multiplying the most promising species,
*establishing pilot production of breeder seed,
*training range technicians in native plant collection
techniques, and
*developing a collection program for all of Morocco.
As a well-equipped operating facility capable of identifying and producing plant materials, the PMC can also assess the possible economic value of native and exotic forage species for Morocco. With its well-trained staff in place, the PMC can contribute leadership and technical expertise to the development of plant material quality standards in Morocco.


Nearly 600,000 families (with an average family size of
eight) or approximately 4.8 million Moroccans depend directly on sheep and goat production for their livelihood. For this reason, the current levels of productivity and efficiency of range livestock production systems in Morocco are grounds for concern at the national level. Not only do these families form one of the lowest income groups in society, but also, the production practices they employ have serious, negative, long-term impacts on land resources that serve the needs of society. These include:
* excessive erosion and sediment production associated with
the continual degradation of rangeland,
* the resulting increase in xeric conditions at the point of
origin, and
* the resulting reduction in the productive life of
reservoirs and intensive agricultural developments
Moroccan rangelands are situated over diverse climatic and
ecological zones and productive rangelands can often be converted to marginal croplands. Therefore, more and more rangeland is being converted to cultivated cereal crops each year by a rapidly expanding population despite the fact that little or no profit from subsequent grain harvests can be realized.
Since crops provide a standing forage crop even as stubble, Moroccan producers compensate for poor range forage productivity by producing a very marginal cereal crop while simultaneously controlling the land for their personal use. Cereal crop subsidies and land tenure policy by the Moroccan Government encourage this practice and, as a result, pastoralism has nearly completely given way to an agro-pastoral way of life to the detriment of rangeland productivity and the communal rangeland users.
Major findings resulting from the RMIP indicate the extent to which rangeland degradation and deterioration of livestock production have occurred in Morocco. Detailed descriptions and recommendations for future actions follow.
RMIP COMPONENT ONE: Applied Research
Findings: Range Forage Production
Where productive rangeland has been converted to poor cropland, pasturage is scarce. What rangeland remains is excessively used, usually in the complete absense of any kind of management or control. A vivid example of this abuse can be noted at Ait Rbaa Perimeter which is located in the midst of a large agricultural area dominated by irrigated and dryland crop

production. Much of the forage which animals at Ait Rbaa consume is comprised of crop
residues and weedy fallow. After harvest, shepherds herd large flocks through cropland to eat stubble and weeds. However ', when planting begins the livestock must withdraw from the cereal fields. They are then herded onto a relatively small area of communal rangeland where they quickly consume emerging vegetation. Under this excessive grazing pressure, the range continues to deteriorate and the herds must receive large amounts of supplemental feeds including barley, straw and residual products from agricultural processing factories such as sugar beet pulp and cottonseed based feeds.
Thus, many Moroccan rangelands are generally devoid of the original stands of palatable, perennial, herbaceous vegetation. Many forage species, both legumes and grasses, which were reported during the earlier USAID Range Improvement Project (60864-4) could not be located by a former team member of that project, Mr. Walter Graves, during his TDY visit in 1985.
Project staff also noted that soils on grazing perimeters were so severely eroded and compacted that moisture rarely penetrated sufficiently enough to establish forage plants even in years of greater than average rainfall.
In attempting to increase water penetration through the soil surface and to revegetate selected rangelands, the following range improvement practices proved to be potentially useful:
* ripping of the soil and construction of contour terraces,
* use of shrubs and herbaceous species (native and
introduced) for revegetation, and
* use of conventional disc, plow and roller seedbed
preparation methods in seeding during years of below
average precipitation.
Specifically, ripping of the soil and the construction of
terraces on the contour were highly successful range improvements installed at Fritiss (Ain Beni Mathar), Gouttitier and Ait Rbaa Perimeters. Although not a year old, ripping and terracing activities at the El Faija Perimeter appear to have markedly
e ncouraged revegetation and increased forage production.
Experience indicates that seeding of most introduced species on small plots of marginal cropland is an intervention that is generally too risky to be assumed by private producers. A farmer would rather have a poor stand of barley or wheat from which he may obtain grain and stubble annually than a year of limited or no use which he considers lost. Furthermore, success in establishing perennial forage species has been limited due to competition from weeds on old fields and low amounts of precipitation. Thus, establishment of perennial, introduced species has been generally difficult and expensive.
Stands which have become established, such as those in the perimeters administered by the Midelt Office, are the result of adequate seasonal precipitation and few problems such as seed quality and competition from weedy species. Although a number of

stands composed of *introduced species from the genus Agropyron are managed at Plaine de l'Aarid Perimeter, the condition of these stands has undergone substantial deterioration in recent years (many of the stands are over 10 years old). Juvenile
plants appear to be lacking and the remaining mature plants are decadent with some stands in need of renovation.
In addition, warm-season grasses, including native species, have proven difficult to establish on Moroccan rangelands. Seeds of warm-season grasses are generally small and relatively expensive. Successful establishment depends upon adequate precipitation during and after germination, and planting depth must be carefully controlled by technicians. The possibility of failure is too great to make this a viable alternative unless techniques are improved.
Species adaptability trials are a vital component to the success of future range seeding efforts. There will always be new species, both introduced and native, that will require testing and evaluation. Species must be evaluated for seed germination and seedling establishment, the keys to successful stand establishment. Adaptability of the species to local conditions, including grazing tolerance, is essential to longevity and persistence of seedings. Forage productivity along with stand longevity affect the economic feasibility of seeding efforts. The results of the RMIP trials have shown substantial differences in adaptability and productivity among varieties within species.
In particular, the overall poor performance of introduced species was in contrast to the relative vitality of the few native species within the trials and found in growing locally in various protected areas. This observation prompted a change in project emphasis from solely evaluation of introduced species to
additional efforts in the research and development of native species.
The potential of plants native to Morocco is the most
promising aspect of the RMIP work with range forage species. Among these plants are ecotypes highly adapted to climatic, grazing and soil conditions on the grazing perimeters. Among accessions from the native seed collection being propogated at the PMC and undergoing trials at the perimeters are annual species which have the ability to produce the large numbers of seed necessary to maintain a presence on managed communal rangeland. These include leguminous annuals, especially from the genus Medicago, which exhibit sprawling forms and subterranian fruits capable of persisting under heavy grazing pressure. These species characteristically produce forage of high quality and large numbers of seeds in protective pods. The use of native annuals may be one key to reducing the risk of failure of a range revegetation investment in some zones.
Several basic range planting techniques were tested in the Midelt area. The conclusion drawn was that conventional seedbed preparation methods using a plow, disc and roller were most successful during the years methods were tested (generally lower

than average precipitation periods). The rangeland drill, which scarifies soil surface and seeds in areas of rough topography, did not prove effective during drought years because seedbeds were inadequate for seed germination and seedling establishment under marginal soil moisture conditions. Where interseeding with the rangeland drill in sagebrush stands was attempted, seedlings could not compete with established plants for limited moisture.
Recommendations: Range Forage Production
The need for range management and the adherence to basic
ecological concepts needs to be observed throughout the grazing perimeters studied. This can only be affected through the cooperation of the people using the resource. Benefits that accrue to producers must be developed and demonstrated in order to reverse the trend toward further resource degradation. Thus, proper management of grazing must occur if the range forage resource is to improve.
Specific conclusions include:
*Range improvement is possible through manipulating the
soil surface and redistributing water. Soil ripping and
construction of contour terraces should be considered
prior to seeding rangelands. Where native residual plant
cover still exists, the treatment can be used to capture
rainfall and increase productivity without seeding.
*Range improvement research is essential. DE/SP must
either decide to dedicate personnel and materials to
research or make the proper contractual agreements to
continue range research in identified topical areas
through the auspices of such existing research
organizations and programs as INRA, INAV9 ENA or the
*Species adaptability trials should continue. Comparisons
of standard introduced varieties and newly released
varieties as well as comparisons among native species
should proceed. Despite unpredictable weather
conditions, it is essential that trials continue to be
planted and evaluated, preferably on a yearly basis.
*Coordination is mandatory between local offices and the
PMC. The PNC research supervisor must keep personnel at
the grazing perimeters informed of the details of the PMC
production program. Otherwise little success can be
expected from the PMC.
A combination of range improvement techniques and development of adapted forage species can improve the possibilities of successful revegetation. Only through reducing the risk of rangeland seeding can seeding of denuded range and submarginal cropland become a viable alternative for producers. The acceptance of rangeland management in the private sector, however, is not the only consideration. Large-scale government

direction and investment is necessary in order to conserve the dwindling Moroccan range resource.
The continual push of dryland farming into more and more
marginally productive areas has accounted for an irreparable loss of topsoil and the local extinction of many forage plant species. Government intervention and expenditure will be necessary to save soil and vegetation even if only to protect watersheds. The continued emphasis on dryland farming by the Moroccan Government and USAID without a balanced effort in the rangeland production sector is a policy which encourages environmental degradation of natural resources in Morocco. Social and economic consequences of policies that foster unwise land use will undoubtedly be felt at the national level through increased rural poverty and urban migration.
Findings: Range Livestock Production
Livestock populations in Morocco consist primarily of sheep with smaller numbers of goats, cattle, camels, horses, mules and donkeys. Livestock production systems and levels of production varied across the country. Based on information gathered through project efforts, the following conclusions can be drawn:
* level of livestock management is reflected by the producer
income levels,
* types of forage resource and times of use vary among
producers and project sites,
* currently applied animal health care practices are more
curative than preventive, and
* particular problems in sheep management and production
exist across the country.
Specifically, livestock management is reflected by the producers' income levels in several respects:
the high-income producers generally provide more
intensive management through use of multiple flocks and
forage sources making feeding of supplements and
administering animal health care more feasible; and
low-income producers maintain only one flock in which all
animals are generally treated equally and animal husbandry
practiced is quite poor.
While all producers feed their livestock supplements
sometime during the year, the degree to which animal species, class and stage of reproduction are differentiated and the amount of supplements fed varies among producers. Producers at the higher income levels utilize supplemental feeds to a greater advantage and often to a greater extent than do poorer producers.
Types of forage resources and times of use vary among
producers and locations. Animal movement and forage used are dictated by what resources are available. Producers use preferred resources first and then move on to less desirable sites as the forage availability diminishes. The more prosperous

the producer, the more abundant and flexible are his opportunities to find adequate forage to meet livestock needs. For example, mid- and high-income producers have access to crop residues and are able to haul or herd flocks to leased land to graze, and-low-income producers are limited to the forage available on communal land where available forage is generally insufficient to meet livestock nutritional needs.
Animal health care practices are most often curative rather than preventive and, throughout Morocco, livestock are penned at night. While these corrals provide good protection from predators, the permanence of the structures and negligence in keeping them clean results in many health-related problems. Animals treated for internal parasites are immediately returned to the same corrals and grazing areas used before treatment. The animals are thus immediately re-exposed to parasite eggs and larvae and soon become infested again. Furthermore, during discussions with producers throughout Morocco a common concern was raised over the quality and effectiveness of the animal health care provided by the government. Why these concerns were voiced is unknown and no effort was made to determine the reasons,
Particular problems in sheep management and production across the country include:
* excessive numbers of low quality breeding rams,
* late weaning of lambs,
* lambing at times when mortality is high,
* low levels of animal nutrition,
* poor sheep selection practices, and
* low quality wool.
Specifically, producers in Morocco generally keep an
excessive number of rams in their flocks year round allowing indiscriminate breeding. This practice is unnecessary resulting in inefficient use of forage and lambing at various times throughout the year with little or no selection for genetic improvement. Ram-to-ewe ratios vary but are often around 1:20.
Intentionally weaning lambs is a management practice rarely used by Moroccan livestock producers. Project studies showed that by early weaning of lambs, producers could receive the health, nutritional and reproductive benefits without a reduction in lamb growth and development. The combination of a grazing management program coupled with early weaning allowed young lambs access to quality forage and reduced the time required to reach a marketable size while ewes remained in better condition allowing them to breed back more readily.
Across project sites, lambs are born throughout the year with a higher incidence of births during the spring, fall, and winter months. Most lambs are born during periods when the risk of mortality is high. The incidence of lambing intervals that exceeded 13 months was common. This primarily reflected a lack of sufficient forage to meet the ewes' nutritional requirements for reproduction.

Death losses occurred year round but a higher incidence of deaths occurred during the late fall and winter when the weather was harsh and forage was limited. Producers reported normal death loss to be 7% to 9% for lambs and 3% to 4% for breeding animals while the project producer study subsample experienced a
6.9% lamb loss.
Poor sheep selection practices and reproduction problems
were common at all project sites. Problems encountered included:
* ram infertility,
* producer failure to identify production and reproduction
related problems in animals for selection of replacement
ewes, breeding rams and cull animals; and
* poor management of animal nutrition.
Specifically, ram fertility problems were found at all project sites and 22% of 217 rams checked for reproductive problems were to be determined sterile or nearly so. Reproductive problems encountered ranged from low sperm motility and concentration to cryptorchidism in rams. Project and SR-CRSP staff trained technicians to recognize ram reproductive problems and a voluntary program for checking producers' rams is in the
initial phase.
The criteria for the selection of lambs as replacement ewes were so different around the country that many undesirable animals remain in producers' flocks. Out of 3,942 breeding animals examined in the project classification and selection program, 26% were rated as inferior animals, and 15% of all animals examined were incapable of reproduction. Unproductive animals require the same investment in forage resources, supplemental feed and medical care as productive animals at the cost of reduced efficiency and profits to the producers.
The low nutritional status which animals suffer as a result of insufficient forage resources produces a severely depressed reproductive capacity. Few, if any, animals are allowed to express genetic potential for production., In addition, measured weight losses of animals were due to inadequate forage quantity and, possibly, quality. Ewes weights were monitored for a year and most lost weight during the last trimester of gestation with .some showing weight loss throughout the entire gestation period.
The weight changes experienced by ewes in the producer study indicated that the ewes gained weight in the spring and lost weight during the remainder of the year regardless of the lambing date. This pattern is in sharp contrast to the weight changes expected in an average ewe's reproductive cycle and has a substantial negative impact on overall reproductive performance.
Wool grades in sheep at project sites were generally low.
Fleece grades ranged from 40 to 60 with most being between 40 and 50. Kemp and/or hair was present in roughly two-thirds of the fleeces. Despite demand for quality wool stemming from the Moroccan carpet industry, livestock producers generally consider

wool a by-product of animal production and have little interest in its improvement because each individual producer deals with a relatively small quantity.
Recommendations: Range Livestock Production
Based upon observations and experience gained at the various project sites, it is recommended that research and extension programs include aspects to increase forage productivity and to improve animal husbandry practices. Significant improvement in forage production and, subsequently, animal production can be expected in many areas through grazing management combined with improved animal management practices.
Given the limited number of trained professionals and the scarcity of program funds in Morocco, it is essential that efforts of the range management personnel be concentrated in areas where an attitude of local concern and cooperation exists. In some areas, livestock producers have recognized the value of formally organizing themselves to try and deal with common range management and livestock production problems. These are the areas where production problems can be successfully addressed. Once groups are formally organized and grazing management is established by the group of producers and technicians, range development projects such as seeding and other forage improvement efforts can be planned and implemented. Under these circumstances the expenditure of limited development funds is justified.
Improved animal husbandry practices that should be considered include:
timing lambing periods to coincide with periods of increased forage production so as to better utilize
available resources, improve ewe condition and reduce
producers' dependence on costly supplemental feeds;
culling unproductive ewes to allow productive animals more
available forage;
separating rams from the ewes except during a specified
breeding season thus enabling producers to better
synchronize lambing periods and other critical nutritional
periods with times of high forage availability;
preventing inferior rams from breeding by such separation
or by practicing the short scrotum technique for
ridding herds of other unproductive members and
encouraging producers to select for high quality
productive animals and not just increased animal numbers; establishing superior breeding herds through a selection and classification program and banding sheep into larger flocks when appropriate to increase livestock production
efficiency; and
establishing extension programs to train producers in preventive animal health care which could ensure that
treatments are scheduled when animals are moved to new
pastures and corrals thereby disrupting parasite life

In addition, extension program efforts should be made in the areas of:
wool quality improvement and marketing, and
goat production.
Since wool quality characteristics (degree of contamination with kemp and hair, fleece weight, staple length and fiber diameter) are some of the most heritable traits in sheep, wool quality could be rapidly improved through herd selection. Improved wool quality, coupled with improved marketing through quality control and pooling of fleeces), could aid Moroccan producers in meeting the local demand for carpet grade wool. The Moroccan carpet industry currently imports approximately 20,000 metric tons of carpet grade wool annually at hard currency cost because Moroccan wool does not meet quality standards established in the world market and the supply is dispersed in relatively small quantities over the entire nation. Development of wool cooperatives for accumulating, grading, pooling and marketing of wool could lead to mutual benefits for both producers and the export carpet industry in terms of employment, prices and raw materials importation issues.
Producers with limited resources may be well advised to
raise animals that can produce on the resources available instead of attempting to raise animals that require supplemental feeds at costs that are difficult to recover. More effort should be expended in exploring the potential of goats for efficiently producing meat and fiber. Although the project effort to compare goat and sheep production on limited resources was not conclusive, it is possible that goats are capable of higher and more efficient production than sheep under severe conditions.
Findings: Sociological Research
The sociological research carried out under the project sought to:
determine producer perceptions of their production
needs and constraints, and
identify social constraints to the adoption of new
Coordinated and systematic project-wide studies were
undertaken in the spring of 1985 and continued throughout the final 18 months of the project. Staffing arrangements allowed Moroccan personnel to gain direct experience in conducting social research in relation to range management and livestock management improvement. The RMIP social research teams were involved in two major work activities:
an on-going study of prices of agricultural commodities in
local souks (weekly -markets) to identify prices relevant
to local producers, and

a series of case studies of producers within producer
categories thought to represent the range of production
syst ems found at each site.
Producer perceptions of the market values of crops and
livestock products were similar to those found in most societies
-- "prices are lower than they were several years ago". There seems to be a tendency to remember the "good" times and to regard them as the norm. In reality prices are generally higher than ever but real profits are lower because of increased production costs and inflation.
Producer perceptions of their production environment
generally centered on the issues of low productivity and their limited ability to provide for proper nutrition and health care of animals due to high costs. Drought and overstocking were identified as major problems associated with low productivity but were deemed beyond the individual's control. Most producers felt that conditions were gradually worsening over time. Producers felt that drought was keeping the production of supplemental feeds low and, therefore, keeping feed prices high. Animal health programs were valued and thought to be needed, however, many producers were sceptical about the effectiveness of existing programs and were not aware of the management aspects necessary for effective disease and parasite control programs.
Social constraints related to marketing of products were few. However, the market limitation affecting controlled breeding of livestock (generally accomplished'through castration of males not kept for breeding) was important. Male animals are marketed with the testicles intact on the carcass and demand a higher price than do female animals. Likewise, docking of tails as a sanitation measure is not appropriate since tails on carcasses are a traditional means by which consumers differentiate sheep carcasses from goat carcasses.
Constraints to acceptance of new technology were generally not found to be cultural. An exception is the aversion within the culture to undertaking any management practices that results in animals bleeding from an open wound. This limitation disallows docking of tails or castration and affects sanitation and controlled breeding efforts. Other constraints to acceptance of technology are generally centered around individual producer resource limitations or scale and producer dependence on common resources. Individual actions involving communally owned resources have essentially no effect. Group actions are needed under these circumstances and, unfortunately, the traditional mechanisms for control and management of communal resources (such as the jemala) have, for all practical purposes, disappeared in many areas.
Recommendations: Sociological Research
Dealing with limitations to the acceptance of new technology is therefore related to introducing techniques for improving management that fit within the existing cultural constraints and

that encourage management of communally owned resources. In the case of controlled breeding of animals, bloodless male sterilization techniques that do not remove the testicles are in order. The short-scotum technique introduced through the project is one appropriate approach to accomplishing controlled breeding. Another approach to this issue as well as many other animal and range management issues is the formation of producer cooperatives on communal land whereby it is possible and practical to form larger herds according to animal class. Through this practice, breeding animals can be herded separately from market animals and managed according to the specific needs of the animal class. This also allows for the control and management of grazing. Such group efforts have been successfully tried at the Fritiss Grazing Perimeter near Ain Beni Mathar. The key to improved range and animal management is involvement of the producers as a group that can and is willing to act in the interest of the entire group.
Findings: Economic Analysis
The objectives of the economic analysis were to:
* identify economically feasible alternatives that would
increase production and production efficiency of livestock
systems in Morocco, and
* estimate the magnitude of production and ultimate net
income increases given a range of producer accepted
alternative production strategies.
The economic analysis process consisted of:
* development of a descriptive model that could be used for
budgeting and partial budgeting based on production
parameters, the costs of resources used in production and
the prices of products;
* integration of biological and socioeconomic information
in order to identify feasible alternative range and
livestock management practices; and
comparison of alternatives in terms of investment
costs and the subsequent increases in forage production
required to produce a given real rate of return on
Data pertaining to the productivity of existing livestock production systems were collected in each of the four project areas and included estimates of:
herd structure and productivity,
feed and forage resources used,
revenue from livestock production,
livestock production costs,
crop productivity,
revenue from crop production,
crop production costs,
income and costs from other activities,
production system net revenue, and
forage use efficiency.

The analysis of the production systems identified basic
economic problems in Moroccan livestock production including high production costs and low productivity. High production costs result from the dependence on expensive sources of feeds for maintaining animals and supporting unproductive animals. Low levels of productivity result from:
abusive land use practices,
outdated animal and grazing management practices, and
resource limitations (biological and financial).
While these problems of high production costs and low productivity were common to both livestock and crop production, the project emphasized livestock production alternatives thereby balancing other more widespread efforts in Morocco directed toward solving agronomic problems. Alternatives addressed livestock production costs in two major areas:
improved animal husbandry and management, and
increased production of range forage as a substitute
for high cost feeds.
Production costs could be reduced through culling of nonproductive animals, adjusting management to reduce disease and parasite infestation, and managing feed and forage resources based on animal requirements (supplemental feeding programs and practices such as early weaning).
When r ange forage is valued as a substitute for supplemental feeds and sites having adequate inherent potential are treated, economically feasible increases in range forage production are possible through various practices. Based on studies initiated as project activities, management practices considered as viable approaches on specific sites include:
grazing management,
contour terracing,
shrub planting,
seeding with the rangeland drill,
seeding by conventional methods, and
combinations of these practices.
The degree to which improvement practices can be applied on rangeland is determined to a great extent on the capacity of the land to produce forage. In the RMIP analysis, approximately 70% of the rangeland in Morocco was considered to have limited potential for forage production because of aridity. These lands were considered for improvement only through investment in extensive practices such as grazing management. The expected average increase in available forage production through grazing management alone was conservatively estimated to be 57%. The remaining 30% of the land was considered to to have adequate production potential to qualify for more investment intensive improvement practices. The potential average forage production

increase on these lands was conservatively estimated to be 200%.
Recommendations: Economic Analysis
There are probably many biological and economically feasible ways in addition to those attempted through the RMIP by which animal and forage production can be increased. The critical element is the degree of interest and desire that producers have in improving their situation. Producers can be successfully assisted in improving their animals and range resources but it is highly doubtful that improvement can be handed to them. Producer involvement and participation in all phases of development including investment is needed to ensure that the commitment to improvement exists.
The effect of improved management of livestock production systems can be important at the local and broader regional or national levels. For example, even at a relatively low yet perhaps realistic level of producer acceptance of 25%, substantial improvement in average production (33%), production efficiency (44%) and, more importantly, net revenue could be expected. The average increase in net revenue per family member is projected to be 49% at the 25% adoption level and is estimated to be much higher as producer acceptance increases.
Estimates of average national annual production and product value based on 25% producer acceptance would result in substantial increases in production (33%) and product value (32%) thus contributing to the long-term possibility of reestablishing export markets for meat and meat products and so strengthening the national economy.
The following actions are needed to strengthen the efforts to improve range management and livestock production:
a national strategy accompanied by clear statements of
policy should be formed to serve as a guide for the
development of programs that are of high interest to
producers and have high levels of improvement potential;
a systematic approach to planningand evaluation of programs should be established to ensure that local
programs are effective in addressing the policy goals of the national strategy and that a systematic approach be
used to analyze problems and alternatives;
development efforts should be directed in those areas where producer interest and commitment is high and at
those investment opportunities that directly and quickly
increase production;
recipients of development assistance should be required to participate in the financing of developments in some form
to ensure their involvement and commitment;
efforts to increase the amount of biological, social and
economic information on range and livestock production systems should be strengthened along with the lines of communication among professionals working in research,
education, extension and management;. and

a national commitment should be made to continue the
search for better development applications in semiarid and
arid zones.
Findings: Extension Program Development
The project extension program development relied on:
determining audience profiles,
assessing current production systems, and
identifying potential interventions and producer
acceptance of alternatives.
In regard to audience profiles, essentially two producer
audiences exist in the extensive livestock sector in each region:
high-income producers who comprise a small group of
families with most of the wealth and income-producing
capability, and
low-income producers who make up the majority and control
only small resource sets with a low capability of
producing income.
This resource distribution pattern affects the design and
implementation of range and livestock extension programs as both audiences have divergent needs and varying abilities to respond to proposed interventions. Producers controlling most of the private land can benefit from programs providing crop and rangeland improvements including improved crop varieties, range seeding, terracing, water spreading and development, herd improvement, animal health and animal selection. Producers with less control overresources would benefit from cooperative organizations through which they can improve the management of collective land to which they have access allowing programs such as animal health, animal selection, range improvements, and record-keeping and analysis to be implemented.
In assessing current production systems, the principal
production constraint for the majority of Moroccan producers does not appear to be a lack of interest in new technology nor a desire to produce by traditional methods but rather a resource base so small that it cannot produce a profit that allows the producer to raise or, in many cases, even maintain his standard of living.
Small ruminants form the basis for the production system in the arid and semiarid zones of Morocco. In these zones, the potential for major increases in grain production is limited by inherently low productive potential determined by climate and soils and of local significance only in the context of its integration into the livestock production scheme. Project extension efforts were therefore directed primarily towards livestock production and improving forage crop production. This

was a reversal of current Moroccan government programs in which the focus of extension has been directed primarily on crop production for grain. Producer interest was the criteria thought to be the most'important determinant to acceptance.
Recommendations: Extension Program Development
Based on project experience, the necessary ingredients to successful range extension program development in Morocco are:
*organization of collective rangeland users into entities
that are capable of managing the range resource,
*continuous effort to develop and test new range and
livestock improvement practices that benefit producers,
*development of programs that introduce technology that
provides benefits to the producers in the short-term and
that will open the door to introduction of technology that
provides long-term benefits to the producers and society.
An example of an appropriate extension program initiated by RMIP personnel is the animal selection program carried out with the producers at the Fritiss Perimeter near Ain Beni Mathar. The producers quickly learned how to apply animal selection criteria for herd improvement and realized benefits of the program through savings in production costs and improved animal production efficiency and quality.
The training component was one of the most successful
aspects of the project. Combined with the training achievements realized through other assistance projects, a substantial number of individuals who received either short-term or long-term training in disciplines pertinent to range and livestock development are now in place. These people will have an important influence on future development efforts in Morocco.
Findings: Long-Term Training
Through RMIP efforts, a total of 11 people were trained to the MS level in range management, extension or rural sociology. .These individuals have all returned to be assigned in various locations around Morocco.
Recommendations: Long-Term Training
Training efforts should continue in order to fill specific areas where trained people are needed (such as sheep and wool production), to replace individuals leaving the workforce and to increase the total number of trained professionals involved in range and livestock development. The number of individuals needed should be projected based on the national strategy for

development and management set by MARA and DE.
Findings: Short-Term Training
Short-term training was an effective training tool in the overall training effort. Special workshops, seminars and shortcourses provided the kinds of "hands on" experience needed by many of the technicians involved. A total of 98 person-months of training in range and livestock management, extension, plant materials production or farm management were provided for DE technicians and administrators. In addition, two one-week technical seminars and numerous one-day seminars were held at various locations in Morocco.
Recommendations: Short-Term Training
While short-term training was not an adequate substitute for degree training, it did serve a very useful purpose. It provided technicians with experience that builds confidence and technical credibility. Short-term training, both in Morocco and in other locations, should be continued. In-country technical seminars or in-service training sessions raise skill levels, give technical staff an opportunity to learn of different approaches to solving similar problems and provide the mechanism for open discussion of technical and professional issues.
RMIP COMPONENT FOUR: The Plant Materials Center
The development of the Plant Materials Center represents the primary effort in development of physical structures. Certain philosophical aspects like determination of goals of such an activity and function of the facility were equally important parts of the development process. At the close of the project some confusion remained regarding the goals and function of the PMC due to some confusing signals transmitted to MARA by USAID. Because of difficulties in adapting government financial support to a farm production situation, USAID suggested that private sector involvement or "privatization" of the facility was preferred due to the cost of public operation of such a facility. This line of reasoning essentially removes the public service aspects which were the driving force behind the original concept of the PMC. Originally, the PMC was envisioned as a facility that would lead the way in the development and production of seed and plant materials which would be useful in solving forage production and soil and water conservation problems in semiarid and arid zones of Morocco and North Africa.
Findings: Plant Materials Center
The PMC facility was completed in 1986. It included
offices, housing, shops, seed cleaning equipment and facilities, seed laboratory and storage facilities, seed production equipment and irrigated agricultural land for pla :nt materials production

and research. Some of the land was devoted to a shrub nursery and research trials on native and introduced species. At the close of the project, most of the land area was scheduled for the production of the seed of legume species which have been imported in recent years.
Recommendations: Plant Materials Center
While the original goals of the PMC emphasized the
production of seed of cool-season forage species, based on production difficulties with these species, this emphasis should be changed. Legume species (both native species and introduced cultivars) are well adapted to production at the Khemis M'touh area. Other goals such as production of foundation seed, development of seed certification standards and promotion of an extension program for seed production are still valid. Five additional recommendations for strengthening the role of the PMC in Morocco are as follows:
* Continue the program of research and development of native
plant species for forage and conservation use even if
satellite locations are needed for the production of some
* Obtain budget autonomy in order to allow the flexibility
required for the PMC to function in the role that has been
* Use the PMC facilities to generate revenue by producing
high quality seed of forage or agronomic crops for more
humid zones to the extent that such activities do not take
place to the detriment of the established program.
* Integrate the PNC seed laboratory and the DPVCTRF program.
*Develop a laboratory for inoculum production that will
complement the program of research and development of
native legume species.


In order to determine acceptable strategies of intervention into the livestock production sector, project participants working in the field at local offices had to answer the following questions: 1) What are the forage resources and level of production available to livestock at each site? 2) What specific range improvements are available to range managers and private producers at each location? 3) How will these range improvements affect forage production?
Project technicians collected range forage and drop
production data from local administrative sources. In each perimeter, they planned a program of field research to determine range improvements which might increase forage production and halt the loss of soil and vegetation resources. The following chapter provides the objectives, methods, results, and conclusions of range research which was implemented on the grazing perimeters during the term of the Project.

Abiotic Influences on Production Site Topography and Location
The Ait Rbaa Perimeter, under the responsibility of the Beni Mellal Office, is located on an outwash plain of the Middle Atlas Mountains about 40 kilometers north of Beni Mellal and 13 kilometers north and west of Kasba Tadla (Figure 1.1). The average elevation at the perimeter is approximately 560 meters above sea level.
.". 4 .
rr th
mel ll a rfeeic a r at "
A Seototan soa ete e
0 3S he Paddeck
.1- II~
Figure 1.1. Map of Ait Rbaa Grazing Perimeter showing locations of watering points and experimental grazing exclosures.
Soils in the area are primarily calcareous clays formed over an extensive limestone layer. Soils vary in depth from a few centimeters on hilltops to more than one meter on the bottomland. The soil is mixed with a large proportion of rock, which is mostly composed of chert, ranging in size from gravel to small boulders. Rocky outcroppings and gravel fields are common throughout the area.
The climate is true Mediterranean characterized by hot, dry summers and moist, cool winters. Summer daytime temperatures reach as high as 45+oC and there are occasional days of freezing

temperatures in winter (Figure 1.2). The distribution of precipitation is variable over the perimeter and varies greatly from year to year. The pattern of rainfall is generally bi-modal with the peaks usually falling in November and April (Figure
1.3). Recorded amounts of precipitation range from a low of 95 mm annually to a high of 600 mm, averaging near 350 mm per year (Figure 1.4).
40 35 30
2 25
S 20
Figure 1.2. Average monthly temperatures (oC) for the Beni Mellal region (Beni Mellal Office).
a so

o 500
a 500
71 72 73 74 75 76 77 78 7g 80 81 82 83 84 85
Figure 1.4. Annual precipitation (mm) for the Beni Mellal region for 1971-1985 (Beni Mellal Office).
Forage Crop Production Estimation: BENI MELLAL REGION
Native Range
The original vegetation of Ait Rbaa was probably composed
primarily of annual and perennial grasses and legumes, but heavy grazing has reduced the composition of the plant communities to mostly annual forbs. Major plant species include Cynodon dactylon, Asphodelus tenuifolius, Malva parviflora, Medicago laciniata, and Stipa retorta. A major perennial component of the vegetative cover is a thorny shrub, jujubier (Zizyphus lotus), which is somewhat resistant to grazing, but is heavily harvested for firewood and for materials for construction of corrals.
Forage production on the perimeter was estimated during a 1984/85 forage study, in whichfnpalatable plant species were also collected. The following table has been corrected to estimate only usable forage (Table 1.).
Cultivated Lands
Cultivated lands are an important source of forage for
livestock raised in the area. The following table (Table 1.2) presents estimates of forage availability for the Kasba Tadla area from forage crops and crop residues. These figures were computed from a table on cultivated land use. In order to estimate crop residues the following formulas were used:
-Roughage from wheat or barely..= Production x 1.08 -Beans.......................... = Production x 1.50
-Peas............. ...... .........= Production x 1.00
-Sugar beets.................... = Production x 0.20

Table 1.1. Average annual herbage production estimates by land category on the Ait Rbaa Grazing Perimeter (Beni Mellal Office).
Slightly Eroded 4,615 1,500 6,922.5
Moderately Eroded 1,680 1,100 1,848.0
Highly Eroded 877 350 307.0
Bottomland 2,500 600 1,500.0
Rock Outcrop 112 0 0.0
TOTAL 9,784 1,081 10,577.5
Table 1.2. Estimated average annual crop production-and crop residue available for livestock in the area surrounding the Ait Rbaa Grazing Perimeter (Beni Mellal Office).
Forage Crops
Alfalfa 428 3,809.2
Bersim 40 180.0
Oat-Vetch 43 309.6
Pea-Barley 148 310.8
Corn 67 254.6
Others 71 269.8
SUBTOTAL 797 5,134.0
Cereal Grains
Hard Wheat 14,280 28,560.0 39,984.0
Soft Wheat 22,827 59,350.2 63,915.6
Barley 16,959 28,830.3 32,222.1
Corn 88 176.0 378.4
Legume Grains
Broad Beans 1,664 1,830.4 2,496.0
Peas 4,013 2,809.1 3,611.7
Chick Peas 789
Lentils 694 138.8 13.9
Green Beans
Sugar Beets 1,587 64,432.2 12,466.2
Fallow 8,997 20,693.1
TOTAL >72,695 1 1 >175,780.7
Two years of data (1983/84 1984/85)
** One year of data (1984/85) + Crop residue estimates:
Straw from wheat or barley = grain production x 1.08
Aftermath from beans = grain production x 1.50
Aftermath from peas = grain production x 1.00 Dry pulp from sugar beets = production x 0.20

Abiotic Influences on Production
Site Topography and Location
Two recognized grazing perimeters fall under the
administrative jurisdiction of the Midelt Office of the Service des Parcours: El Faija Perimeter, a 4000 ha area near Itzer, and Plaine de l'Aarid Perimeter, a 12000 ha area 45 km west of Midelt. Both grazing perimeters are situated on high plateaus located along the eastern edge of the Middle Atlas Mountains and are found at elevations ranging from 1600 to 1800 m.
Soils have developed from old sedimentary deposits and
overlying alluvial outwash. Soils have been described as mostly clays, containing a calcic horizon.
The climate is semi-arid Continental, characterized by relatively cold winters (Figure 1.5). Annual precipitation averages approximately 300 mm, 90% of which falls between October and June (Figures 1.6 and 1.7).
Figure 1.5. Average monthly temperatures (oC) for the Midelt region (Midelt Office).

go so
Figure 1.6. Average monthly precipitation (mm) for the Midelt region (Midelt Office).
71 72 73 74 75 78 77 78 79 80 81 62 83 84 55
Figure 1.7. Annual precipitation (mm) for the Midelt region (Midelt Office).
Forage Crop Production Estimation: MIDELT REGION
Native Range
Native rangelands of the area are commonly characterized as the sagebrush type (dominated by Artemisia herba alba) and the alpha-grass type (dominated by Stipa tenassicima). The sagebrush 32

type generally occurs on areas of even to rolling topography with more mesic site characteristics while the alpha-grass type occurs on the areas of greater topographic relief exhibiting more xeric site characteristics. Estimates of production on native rangelands following drought are presented in Table 1.3.
Table 1.3. Estimated annual herbage production on grazed native rangeland near the El Faija and Plaine de l'Aarid Grazing Perimeters following drought (Midelt Office).
Sagebrush/Grass 133,000 23.0 3,059.0
Alpha-grass 131,000 5.0 655.0
TOTAL 264,000 14.0 3,714.0
Cultivated Land
Cultivated lands in the area are primarily devoted to cereal production. Some fruit orchards and forage crops are also produced on irrigated lands. Estimates of forage produced on cultivated lands are given in Table 1.4.
Table 1.4. Estimated average annual forage production from Midelt area crops and crop residues for a five year period (198084) including drought (Midelt Office).
Grain Crops
Hard Wheat 1 16,270 42,302.0 45,556.0
Soft Wheat 1,001 2,803.0 3,003.0
Barley 1,060 636.0 742.0
Corn 716 787.6 1,217.2
Forage Crops
Alfalfa 1,247 890.7
TOTAL 20,294 47,419.3 50,518.2
Abiotic Influences on Production
Site Topography and Location
The Fritiss Grazing Perimeter is located 17 km south of Ain Beni Mathar (Figures 1.8 & 1.9). It is situated between the

plains of Berguent to the north and Tendrara to the south. Approximately 60% of the surface area (total: 10,000 ha) occurs as hills, breaks and plateaus. The remainder of the area is comprised of lowland plains, basins and drainages. The Gouttitier Farm is located on rolling plains, about 130 km west of Oujda along the road to Taza (Figure 1.10). The Gouttitier Experimental Farm was incorporated into the Bureau des Parcours administration in 1984. This farm was reserved for the needs of future rangeland research for the Oriental Region of Morocco. With some 2000 hectares of degraded rangeland, this area will hopefully become an oasis for the demonstration of advanced agricultural management practices in the region. Of the 2000 hectares, 40 hectares can be irrigated. Rangelands of the region generally occur at elevations ranging from 500 to 1500 meters.
scale 1:1,000,000
\ / [ A.B. Mathas,
0icl B ondar
0 DA.
'A4 - anRa
- Circle Boundary
- --- -- Comune Boundary
Figure 1.8. Map of Oujda Province indicating Fritiss Grazing Perimeter (Oujda Office).

SCALEs 1/50,000
- Main Road Playa Perimeter Boundary
- Unimproved Road -Intermittent Stream Experiment Station Soundary
Railroad Line Cliff or Bank Watershed Management Demo.
Cl: Corral W2t Well/Pumphouse
C2: Animal Handling Facility W3t Concrete Reservoir
Els Stock Pond W4t Concrete Reservoir
E2s Stock Pond W5s Water Harvesting Reservoir
WI: Water Harvesting Reservoir WO Water Harvesting. Reservoir
Figure 1.9. Map of Fritiss Grazing Perimeter (Oujda Office).

Madicago littoralis non inoc I ad
Mixed Grass** Medicago littoralis inoculated edicago littoralis I he.
inoculated 0.33 he.
ftiod Grasses Modicago polymorpha inoculated, KEP0 VXDkA :VXDA 6= 1 1 he.
non non :inoc
Mixed Grass** Modicago polymorpha 0N non-Lnoculated
a Kz" V=A :VIDA KM1 :MZLl I he.
non noo non :1noc non inoc,
Xadicaqo s&tiv rhisonato Mixed Grasses
ua Madica9c, polynorpha
2 he, inoculated%.
madicago littoralis he.
inoculated I he.
Mixed Grasses Medicago littoralis medicago polysorpha inoculated
inoculated I he.
I he.
Mixed Grasses Medicago polymorpha K"icago littoralLs mon-inoculated
non-inoculated I he.
I he.
.ATI Ire
ago Mftltoralis non-inocul.t.A 0.
VANCAUT M*dicago polyworpha VACANT
I he. 0 rAn-inoculated. 0.9 he.
I he.
Medicago littorals inoculated 0.27 he._ [Cool-S.M.On-Grags
Figure 1.10. Map of the Gouttitier Experimental Farm indicating
present plantations (Oujda Office).

Surface soils originated from localized sedimentary sources, varying in texture from coarse gravels, to sand, and then clay. Underlying this entire region is a petro-calcic horizon (caliche layer) which varies in thickness and depth throughout the area. Thickness of the caliche layer ranges from 10 cm to 50 cm and depth ranges from barren outcrops to several meters deep in bottomland sites. This horizon can limit infiltration of precipitation but contributes to the lateral movement of moisture in the soil. In many areas of eastern Morocco this caliche layer is mined for various construction materials.
The climate in this region of Morocco is semi-arid to arid, extremely variable and is more typically Continental than Mediterranean. Ain Beni Mathar lies in the semi-arid zone with an average rainfall of 200 mm (Figure 1.11). Over the last 15 years annual precipitation has varied from near 100 mm to almost 300 mm. There are two peak precipitation months, October and April, but these peaks vary considerably (Figure 1.12). Temperatures peak in July at about 250 C., and are lowest in January (Figure 1.13).
a 800
E 3004
0* co I- I
71 72 73 74 75 78 77 78 79 S0 81 82 83 84 85
1 Mn Sent Mothar + Taourfrt
Figure 1.11. Annual precipitation (mm) from 1971-85 for Ain Beni Mathar and Taourirt (Oujda Office).

90 SO
.50 0
302010 0
0 A Mn BaSnl Mathar 4- Taourirt
Figure 1.12. Average monthly precipitation (mm) for Ain Beni Mathar and Taourirt (Oujda Office).
403530 2520 1510 5
Figure 1.13. Average monthly temperatures (oC) for Ain Beni Mathar and Taourirt (Oujda Office).
Forage Crop Production Estimation
Native Range
Areas of topographical relief, such as hills and ridges, are covered with nearly pure stands of alpha-grass (Stipa tenacissima) in this region. The other 40% of the area is 38

composed of transition zones vegetated with mixtures of alphagrass and sagebrush, and drainages and basins of nearly pure stands of sagebrush (Artemisia herba alba). Estimates of forage production on grazed rangelands are presented in Table 1.5.
Table 1.5. Estimated average annual herbage production on native range following drought for Ain Beni Mathar and Gouttitier Grazing Perimeters (Oujda Office).
Alpha-grass 192,700 3.0 578.1
(Domaine d'Etat)
Sagebrush/Grass 358,065 13.8 4,941.3
(Collective Range)!
TOTAL 550,765 1 5,519.4
Cultivated Lands
Crop production is limited to a small portion of the total land area of the region. At Ain Beni Mathar, cultivated land comprises approximately 11% of the local area (1% irrigated and 10% dryland). Barley and wheat are the primary dryland crops produced and forage crops such as alfalfa are the primary crops produced on irrigated land. Estimates of forage production resulting from cropping activities are presented in Table 1.6.
Table 1.6. Estimated annual forage production from crops and crop residues following drought in the Ain Beni Mathar region (Oujda Office).
CROPS AREA (ha) PROD. RATE (kg dm/ha) FORAGE (T)1
Hard Wheat 1,355 150 203.25
Soft Wheat 475 150 71.25
Barley 19,250 300 5,775.00
Oats 600 150 90.00
Hard Wheat 450 1,800 810.00
Soft Wheat 100 1,950 195.00
Barley 20 2,100 42.00
Oats 50 1,800 90.00
Corn 60 3,000 180.00
Alfalfa 632 8,000 5,056.00
TOTAL 22,992 112,512.50

Abiotic Influences on Production
Site Topography and Location
The Timahdite Grazing Perimeter is a loose collection of private and tribal land on which the government technical services have no formal jurisdiction, except for a small experiment station (Touna Station) which has been dedicated to range research. It is composed of upper elevation shrublands and grasslands and is located in the Middle Atlas Mountains, south of the village of Azrou, bordering the Provinces of Khenifra and Boulemane. The perimeter is divided into three major topographical zones: an undulating plateau at an average elevation of 1800 m, a narrow valley along the Guigou River, and a series of rocky peaks, one of which rises to 2400 m.
The Timahdite region is characterized by two major types of parent material: a limestone substratum and, generally above this formation, a layer of basalt resulting from volcanic activity in the Quarternary Period.
The climate of the area can be described as Continental.
The region has an average annual precipitation of 400 mm (Figure
1.14) with hot summers and very cold (Figure 1.15), snowy winters. There is much climatic variation along altitudinal gradients. .Snow falls on the area from 15 to 40 days annually, with the majority falling during December and January.
:" 70
Figure 1.14. Average monthly precipitation (mm) for the Timahdite region (Meknes Office).

10 S
Figure 1.15. Average monthly temperatures (oC) in the Timahdite region (Meknes Office).
Forage Crop Production Estimation: TIMAHDITE REGION
Native Range
Rangeland plant communities of the area are dominated by various shrubs and perennial grasses. Estimates of the forage productivity of these lands are presented in Table 1.7.
Table 1.7. Estimated average annual herbage production on native rangeland following drought in the Timahdite region (Meknes Office).
Oak-Cedar-Juniper 13,100 Not Estimated
IRangelands from the 31,145 424 13,205.5
following types: Genista-Palouse Thymus-Palouse
Helianthamum Tran.
Cultivated Lands
Wheat and barley are the primary agronomic crops in the area although other crops such as corn, potatoes and various vegetable crops are also common, especially on irrigated parcels. Estimates of forage production contributions from cultivated lands are presented in Table 1.8.

Table 1.8. Estimated annual production available for livestock in the form of crops and crop residues following drought in the Timahdite region (Meknes Office).
ICereal Crops
Hard Wheat 4098 4,507.8 4,917.6
Soft Wheat 63 56.7 63.0
Barley 2719 2,175.2 2,447.1
Corn 363 217.8 1,887.6
IForage Crops
Alfalfa 35 175.0
Oats 78 234.0 257.4
Barley .57 171.0
Other 62
Potatoes 39 390.0
IOther 9
lArboriculture 7
ITOTAL 7530 (Incomplete) (Incomplete)

Species Adaptability Trials
Program Development
The objective of these trials was to introduce and test, under Moroccan range conditions, forage species proven to be successful -on rangelands in other countries with similar range environments. The major achievements in this activity were 1) the establishment of a species adaptability trial program as a continuing part of DE programs in the field, 2) the development of a standardized evaluation system for species trials, and 3) the identification of a number of range forage species and varieties which are usable in each location.
At each field office, species selection, requests for seed, and trial establishment were the responsibility of the DE project leader. The PMC staff distributed seed and inoculum, and assisted the DE project leaders when requested. Beginning in 1982, replicated trials were established in single five meter rows, with spacing of one and one-half meters between rows. Evaluation procedures were recommended by Dr. Mel George, ("Range Seeding, Fertilization and Management", Oct. 28, 1981), yet no uniform methods were followed until late in the program.
After seedbed preparation, using manual labor and hand
tools, the trials were planted by hand. Establishment varied each season. Drought was a major impediment to success. Waiting for the first fall rains, the usual practice in'dryland crop production caused considerable delays each year. Germination was delayed even more for trials that were planted in late November and December. In some locations and in some years, trials were not planted at all. The continuing periods of drought during spring months created additional problems for survival.
Species were chosen for trials based upon recommendations from Dr. George's report written after a visit to each of the project sites in 1981. He ordered the seed for the 1982 trials, while seed orders for the subsequent adaptability trials were placed by the PMC.
In the final year of the Project (1985-86) staff at the PMC and grazing perimeters collaborated to update and improve species adaptability trials through five distinct changes: 1) the testing of introduced-species continued, but with emphasis shifted to testing varieties of these species which were "known to be adapted" based upon the experience of earlier Project trials, 2) during the last year, native species were entered for the first time in a controlled experiment, 3) as a result of the unsuccessful establishment of most species in the trials between 1982-84, new methods were implemented. Attention was focused on planting dates and water catchment, 4) a new "SpeciesAdaptability Evaluation Form". prepared in both English (Figure
1.16) and French (Figure 1.17), was devised and simplified to encourage usage, and 5) reporting procedures were standardized, for simple, concise interpretation.

Scientific Name:
Variety Name:
Origin: Maroc USA Other:
Trial Location: Planting Date:
Seeding Rate: High Medium Low
Date of Observation:
Estimate Emergence Date:
Percent of Row Established:
Plant Vigor High Medium Low
Other Observations:
Date of Observation:
Winter Hardiness: High Medium Low
Percent of Row Established:
Other Observations:
Date of Observation:
Foliage Rating: High Medium Low
Percent of Row Established:
Other Observations:
Date of Observation:
Drought Resistance: High Medium Low
Percent of Row Established:
Other Observations:
Seed Maturity Date:
Plant's Potential for
Future Utilization: High Medium Low
Other Observations:
Figure 1.16. English language version of the standardized form
in use for evaluation of the Species Adaptability Trials on grazing perimeters.

Nom Scientifique:
Nom de la Variete:
Pays de Provenance: Maroc USA Autre:
Lieu de l'Essai:
Date de Semis:
Densite du Semis: Forte Moyenne Faible
Date d'Observation:
Estimation de la date de levee:
Nombre de plantes levees par m2 ou ml:
Vigueur des plantes: Forte Moyenne Faible
Autres Observations:
Date d'Observation:
Resistance au froid: Forte Moyenne Faible
Pourcentage de plantes etablies:
Autres Observations:
Date d'Observation:
Quantite de feuillage: Forte Moyenne Faible
Pourcentage de plantes etablies:
Autres Observations:
Date d'Observation:
Resistance la secheresse: Forte Moyenne Faible
Pourcentage de plantes etablies:
Autres Observations:
Date de la maturation des grains:
Potential de la plante pour
une future utilization: Fort Moyen Paible
Autres Observations:
Figure 1.17. French language version of the standardized form in use for evaluation of the Species Adaptability Trials on grazing perimeters.
Midelt Office: Initial Results
Early trials (1972) provided information on which to start a
seed production program at the PMC. The most promising species
from the region were: Agropyron intermedium, A. elongatum, and

A. trichophorum. Subsequent trials on Plaine de l'Aarid and El Fija Grazing Perimeters resulted in the identification of the following varieties as adapted:
1983 A. intermedium, var. Tegmar and Slate
A. elongatum, var. Jose and Largo
A. trichophorum, var. Topar and Luna
1984 A. intermedium, var. Oahe
A. smithii, var. Rosana A. riparium, var. Sodar
Melilotus officinalis Onobrychis viccifolia
Trifolium hirtum, var. Hykon
Medicago sativa, var. Ladak
Vicia dasycarpa, var. Lana
Astragulus cicer, var. Lutana
Midelt Office: 1985 results
Animal pests destroyed the trial at El Faija.
Beni Mellal Office: Initial Results
The 1982 and 1983 trials were planted in December and failed to become established. The 1984 trial was planted in October. By January, due to lack of precipitation, two artificial methods were attempted to promote stand establishment. In the row trials, supplemental water was applied to one replication twice to offset the lack of precipitation. In another trial, mature plants were transplanted from the PMC to determine if a developed plant had a better chance for survival compared to a seedling. Establishment of these transplants was negligible, probably due to inadequate precipitation and to damage to the plants that occurred during transplanting and
Beni Mellal Office: 1985 Results
Perimeter personnel summarized their findings (using the
standardized PMC form) and made recommendations for species with potential. These species should be multiplied at the PMC for additional testing and demonstration plantings.
Grass species with high potential for establishment:
I. Bromus mollis, var. Blando (annual)
2. Lolium rigidum, var. Wimmer-62 (annual)
Grass species with medium potential for establishment:
1. Agropyron elongatum, var. Jose (perennial)
2. Agropyron intermedium, var. Tegmar (perennial)
3. Agropyron elongatum, var. Largo (perennial)
4. Agropyron intermedium, var. Slate (perennial)
5. Dactylis glomerata, access. GR-659 (perennial)
6. Agropyron trichophorum, var. Mandan (per.)

Legume species with high potential for establishment:
1. Medicago laciniata, access. Ait Rbaa (annual)
2. Trifolium cherleri, access. GR-423 (annual) 3. Trifolium cherleri, access. GR-307 (annual) 4. Medicago laciniata, access. GR-215 (annual) 5. Medicago laciniata, access. GR-222 (annual) 6. Medicago laciniata, access. GR-201 (annual)
Legume species with medium potential for establishment:
1. Medicago truncatula, access. GR-153 (annual) 2. Medicago littoralis, access. GR-221 (annual)
3. Medicago laciniata, access. GR-776 (annual) 4. Medicago laciniata, access. GT-8101 (annual)
5. Medicago laciniata, access. GR-211 (annual)
6. Medicago aculeata, access. GR-138 (annual) 7. Medicago aculeata, access. GR-196 (annual)
Of special interest in the legume trial (listed above) was the introduction of locally collected species. The Medicago laciniata accession from Ait Rbaa was originally collected at the perimeter in July 1984 and planted at the PMC for multiplication in November 1984. The seed, harvested in June 1985, was then returned to the perimeter for this trial. The most interesting difference between the two samples was that the original collection had an average of five to six seeds per pod and produced small seeds. The PMC generation produced nine to eleven seeds per pod, with a seed size three times larger. This seed size could be a major factor in the success of its return to the perimeter. Seed pods harvested from this trial at Ait Rbaa in May 1986 contained seven to nine seeds per pod, with an increasein seed size over existing native plants.
Oujda Office: Initial Results
No trials were planted.
Oujda Office: 1985 Results
Ain Beni Mathar trials were installed within the contour
system of the "America Plantation", southwest of town on the road to Merija. No seed from native species was available for planting.
Due to the cold temperatures and very low.precipitation
during January and February, little growth was observed. Most of the species germinated, but the survival was very low, with only Agropyron elongatum, A. smithii, A. intermedium, A. trichophorum
showing any promise. These trials continue to be monitored on a monthly basis.
Within the contour system at the Gouttitier Experimental Farm, a trial-was established for 38 species. Here, too, cold temperatures and very low precipitation resulted in a lack of extensive growth. The same species of Agropyron which show promise of survival in Ain Beni Mathar appear to be acceptable for the Gouttitier region. In addition, seedlings of some late germinating species were noticed by technicians, prompting them

to continue to monitor trials on a monthly basis.
Meknes Office: Initial Results
In the 1982 Touna Station trial, the following species germinated and persisted:
Agropyron elongatum, var. Largo (perennial)
Agropyron smithii, var. Arriba (perennial)
Agropyron desertorum, var. Nordan (perennial)
Agropyron intermedium, var. Oahe (perennial)
Agropyron trichophorum, var. Topar (perennial)
Bromus marginatus, var. Bromar (perennial)
The 1983 trial was planted in December and failed to establish. No trial was planted in 1984.
Meknes Office: 1985 Results
The trial was established, yet no observations have been recorded.
Summary of Trials
Species entered into these adaptability trials (see Tables
1.9 and 1.10 for seed distribution lists) were not exposed to conditions favorable for germination, establishment, or survival during the life of the project due to drought conditions. Therefore, it would be unwise to eliminate any one species from future experimentation, with few exceptions. Morocco, with its long, hot summers, appears to be ideal for some warm-season grasses. Remnants of a few native perennial grass species (in the genera Cynodon, Pennisetum, Cenchrus) are even found in protected areas. The missing components for stand establishment and survival have been timing and seeding technique. In recent years of low precipitation, moisture has also been limiting in the late spring and early summer when temperatures have been conducive for germination and growth of these species. The expense of these seeds, compared to their limited chance for establishment, minimizes their utility. The situation which may permit their utilization is in plantings behind terraces where rainfall runoff can be captured and soil moisture is retained.

Table 1.9. 1985 adaptability trial seed distribution list.
H I 1W.1LT
Cossaw" ftea Varlety/I.D. SclontifIc ease Cos." ffia Urtetr/1.0.
Grasses Grasses
a I q Tall whestgrass ro cristatue Crested Whestareog Ehprals
W JIM, ,r..Z Rt=
Tell twotgrass I C, a .0 Created whentgrass Ruff
Tell wheatgress Uk r to 4gropyron e I an?& tu, TON Wh.:tgreso Alker
Axropyron 1; rLm-dium Intermediate whestgrose State Agropyron e ng4t-m Tal Vhe tgrose Jose
intermedM_4 intermediate whestgrase Togeor Tsll,.Whostgrsss Largo
Axrovvron Tn't*"dius Intermediate Whe&tgr6W6 Done us into sadists, Wnestgross State
Agropyron rn7Wr_"Tr.G Intormodlat* Wheotgreal, Crooner Apropyron intortedius Intorstdisto Whestgress Teguar
Agropyron as tR-r- Western Wheotgr889 Ar be Agropyron Interse 3T us Intermediate Wh*atgrass Oshe
A ro con saithIT Western Wheatgrass Roden Agropyron Tntor" us Int*rmadlatm, Wheatgrass Greener
ro western whostgross Rosane j&t2gLron -WrtFrr- Western whootgress Roden
ran -sort-M no
ro ran serfm Waster" Whostgrass Flintlock Arropyron saimff western Whestgress Arrlba
A 9 r 0, 1., 111 Er ON! sarrm Western Wheatgrass Barton Agropyron saithil Western WhentgraSs Romano
X.F Puba:cOnt Whtatgrg$t Mandan so thli western Whe4tgreas Flintlock
roDiron Pubs cant Whootgromo Toper xrODVrOn 0 rn Wheatgress Barton
etylle losersta Orchardgra&S CR-526 Axroovron a catud bunch Whootgraws Secor
CR-539 Blu:bunch Whostgress Whitsor
ctyl 6 it 0 Orchardgreas Acroovron se C*tu
Vact 118 Offituo Orchardgres CR-572 Afiroovron frHOWE rum Pubescent WhostgraZa Mandan
c a Ll--Ooer_ta Orchardgree: Pal stint AaroDvron trichophorum Pu acent Whootgrass Toper
W rixidu Annual ryegrass Vim:or-62 ctvlio aloerate Orchordarese CR 659
_u ctvlta Aloetrate Ore Palestine
RMA R'stan:0 Timothy CI Isax hardgrass it*
ctvlis x1omersta rchardgreas Polu
Suls Sweetvatch CR-662 I u ustu* At A Wlldrys, Pre r1olon
Hedyssrus Sveetwetch Tunisia rvzops!FhV oldes Indian Ricegr*se ps lose
t Indian Ricogress Nozpar
u a Dwarf Bledafoot Trefoil Kate ousts an or
Latus re cus, Annual Trtfoll CR-12 dyserus bartalo Northern Sw**tv*tch Common
Re !!,Z
lz-tus c cue Annue I Trefoll CR-560 H!!daZ !!!! 1 ague a Sm"twetch Tunisia
Gwe =ar"a a a a u sets Annual Media CR-514 'us c.r7m=ar a Sula, Sw**tvotch GR-662
Annual Medic Annual "Odic CR:5 5
Mad as& d Annual No Ic C 729
No case locinisto Annual Me le Alt Rb" 1nalls Yellow Sweet Clever Madrid
to Annual Medic CR-509
al = Annual Had Ic 48
dic -- Annum! Medic GR:45
ano orUc-ulario CR 93
ftlllotuf c no Tolle Sweet Clover Madrid
TrlfO us cherlorl Annual clever CR-307
r 0 a suRaranoue Subterranean Clover CR-435
TFMO us _4uFfFrr*n*vu Subterranean Clover GR-448
r us suVEWF'reeus 3ub torroneen Clover CR-494
IE subterran*U Subterranean Clever CA-519
us Z
ffir= us suFforTanium Subterranean Clover CR-567
LUG VOGICULOSUM Arrowlsof Clover Asclo
Scientific Mommi Connote mans, Variety/I.D.
Scientific Nosm, Comem "ads, Verl*ty/I.D. lai3eron elonAet-Um Tall Wheatgrtas Alker
-T-n,.t.. Tell,... Joes,
Crosses Kgropy ran Tell Vhostgress Large
Copyron elonxetum Tell Wheotgreas J06:0 Agropy- "n ;@r!!a1!.!O Intermediate Whostgrass SlateAgropyron *Ionxetum Tell WP*atgraa4 Lar t
Axrovvroa nterwedluft Intermediate Wh*stgrasa Slate Axropyron nt Interoodlets, Whost4gress Toguar
Agropyron in #rmodiui Intermediate Wh.4tgreas Ugsar A ro ran Tnt*ratd1um Intermediate Wheatgrass Oahe
tarn Wh*at4rass Arrtba ro con Inter" a Intermediate Wheatgrass Groaner
Flintlock -M-Ur- western wneotgress Arribe
Western Vnestgress ro ran am
!&ropyron tricnoiholue a. Western Whostgrass Roden
Axr 'EoO -Iihl Pubescent Whestgrass Toper A ran aRTM Western Whoatgress Rossna
ropyron triChoohorus Pubescent Wheatgreas Mandan = -94-TTM Western Whostgrass flintlock
Bros.s malls Soft Crises Blondo Ad on ;s-ithit Western Wheatgress Barton
5;--tv rrs=xoaaratm. Orchardgrass CR-659 Ol.bunch Whostgress Smear
F*otuce arizonica Arizona Fescue Redondo Blu*bun4h Whtstgress Whitser
Annual ryograme r-62 11.1choohorum
11 am Pub scant Wheatgress Toper
Indian Rlcogress VNe1w;:r a....
rvzopste hysenoides n an cogross Polommi trlchopNrus Pw t Whostgress Mandan
vzopsis at faces Bothrichlos Isenseaus Tallow alueswe Canada
Sallograss CT-129 Brame aoUls Sort Chase Blando
vz 02 a a sees ilograss CT-860 Yne -ieetylon Bormu4agrass Common
vZoosts at laces, SaLlograss CT-920 Felt rl -n',- scu* Redondo
I ogress CT F.Tr Ixl
zoosis millacte -940 V_ Yogress Wisimer-62
c ". vr .. Sao r tu Itt a u4stem aseran r : rigncit bp&K* uhly El Veda
xuaws and tartei grg 11 .
tg vasrus borwale N rth@rn v**tv*tCh Gas rose Ps loies
He 3 Oryto a a hmn2ldoo In Lon Alcogress Nexper
vaerue cornooue "Otv*tch Tunisia c.; Sallogrese GT-129
dLc4sO acuLesta A. I "die GN-i96 urvzo
Me us we Orvzooe a RTNcee Sailograss CT-SW
lease sculests Annual die CR-136 ry deals wrFE.-M SeLlograss CT-801
He lease risioula, Annual media CR-145 arr =Oea SmIlograse GT-132
"a lease Intertexts Annual medic rytoosis
CR-188 rytoos a logre" CT-920
dice" Polvearphs, Ann I ::dIc GR:824
at CIS GR 154 ryxoools P
040 Lease 001VOOrpha Anam Smiograse CT-940
case Annual media lacry" u!! scooorlue Little Bluestes Cioneron
Me 921 GA:1a7 SchlzscAnnual modle hy I Inrlus Little, Bluesten Posture
dicame I C ta, Ig
No OR 209 S boi Ire AING&I "canton Salado
M*dIC490 Lacintato Annual medic CR-869 tflRbotH cryptand-rus Send Dropsood Common
Lease lociniato Annual medic GR-215 Xuams an race$
dicato lacinisto Annual madl CR- 22 H*dyssrus borm4le worthem Sw**tvetch Common
He icago Iselniata Annual aodlc A, bee rnoaue Sweetvttch Tunisia
dicemv IacLnIatq CR-776 ""Issrut Annual :dic GUTZ3V! Lague alromfoot Trefoil Coscsd*
dlcaxo lociniste Annual die Lotus corniculatus Dwarf Blrdstoot Trefoil Kalo
Me lease selnista Annual :dIc CR:211 tus otdunculatus Big Trefoil Marshfield
me cato acjnIs GT 6101 dicaso sculoots Annual medic CR-138
alcaso lacLnle Anausi moole GT-8100 d case locinisto us me Ic Alt Rbae
dl a 11,tog, Annual medic GR:221 Me lease rtaidula nual CR
Ann" I medic GR 152 dicoxo Intertoxto us .d I a G 168
Barrel die Akbor dicano Dolveorche Annus G GR:824
!!jgjS!Zo truncatus a., rel madl. CR-153 Annue I medic GR 669
nfri-nall Yellow Sweet Clever Madrid M EM; 1.21H it
1 Cie r GR its I! "!to Annual medic CT-810i
-307 M*41caso, truncatu"14 Barrel Media Akbar
Trip mil fl -::I CI:.wV CR-423 ... ... yellow sweet clever Nsdr14
NM Subterrens" Clever Gereldtan Tr IrWe Rome CI r Mykon
T14191i" 8" remses Clever North"
Trtroliud -1 -ter mter viols aserverss weelyped Vat" Lam
torressea Clever wassarls ftn#t Molinari Palmor F"s- Ce"r

Table 1.10. 1982-84 adaptability trial seed distribution list
including species previously distributed in addition to the 1985
3cl 'Latic fteme C.-"o 014me
rp-ro.,Yon dosertorue Crested Whoet4ras N orda.
Aropyron criaLatum Who*ti;raa: F. It-.,
Aaroovron lntOtmediUs Intermediate Who t8rass A:Ur
Agropyron Slender W hem tgra: a R venue
Slender Whootgross Prisn
-.5. 13 1 F Bluoston, Cosm n
9-rosus N-M*-Ga S-tbrom: Ilebor"
P-roaus inormis Smooth Bromagress ManChar
T-O-us "--=9n-tuG Mountain NtOwogr4*4 Or:mer
-- -- 0 C
C hrus cTTT*rFa Buff*Igr
9* nchru 8 C111arls Butralgre Lol.r"
G.Chrug ZTT-arls Buffolgrass Higgins
H- Suff*lgross NU cab
nchr.s c I a =rs
5r t =s Orchard:resa Berber
Its Orchard rasa Curtis
hrharta ft .*nnial Voldtaress Mission
IF.Eaoo. tunc'suo Russian Wlldry* Common
Cur'. Is Weeping Lovegress Catalina
roxrost a cur Me We ping Lovo:r Common
Erearostis cur-nuTe We:plng Love r: A-84
Er curvuTa Weeping Love
T.W. nn one Lthean" Lov:,Va vato
.. as coo
Erearostis NhianhTane LAhman L4 ,r:, 00
tralrostis R r Wilms Love a Polar
IRRES.. Send Lovegrass Bond
2 al 7term SandhIllcLov*grasa Kasen
Foetuce arm in-C** Tall roe me Demeter
Fostuca, Hard rescue Durar
F* a -tuca ovins Shosip resewe Cover
.-- a "ad rescue Fortress
ftucs ru r;
=4 muj loruis In." I :Y gross Coo n
Pholarls tuberose Harding Me Seadamater
Phalaris tub*ro" Hardinggrass Sir*"
FMarTs Tu to" (Hirt) porlograss Paris
P"- aa-oTa Big Bluegrass Sher n
F.- cman 1 Canby alue"ro .1. Canby
Canada Bluegrass Reuben*
Fr 0 Alkali Socaten Co n
as ke So
ffiroloo 1 go all-NIN At It Saltalk
I a myyros Rattail Slaweekagrega larro
As roaalue'VIoL Clcer "llowetcf. Luca"
P L. Uses sod rorbe
VO case I Uto" Medic Harbinger
OA C400 "dic Peropento
as'C a saff. "I re I to Drylander
A Rambler
Rodicago so =e ,.If*
I I T:n
RWJT-C4-PO Alfalfa R g*land*r
RoWaT-caito A f f to"
A f: to Travels
K*dicar* --a- A, felt& Spreader 2'
Me Ice 0 10 If* I to Lo do k
4 we Altalre RhyxOaO
- P-N, scut*Ttato Snell Medic Snail
ca a t g!!Sjt;UI!! Barrel Ze IC Cyprus
Rod C"o t aE To Farre, I M :1 C Kannatord
Medicate truncatule barrel Magic Joationg
.o Saintain Common
Sorradells Common
Small burnot Dater
Trifo"--- htrtua' Rose Clever Kondinin
fr-170 &=us hiatus, Cl war Wilton
tun Clever
T-rrro7ru0 Thcarne Cries" Not*
T-TraUms EuMErso su:t*rr name clever Clore
R, to:*
rr-rra=4 ou Sa torr:neen Clover Dalisk
Am Fronous, Subterranean Clever Howard
subterrenWum subterranean Clover Mt. Barker
T-rTro=uo &YFt-*rran*v* Subterranean Clover Seaton park
T fuu-eraniun Subterranean Clover Trikkala
;ubtarranoma Subterranean Clover Woogenallup
hear& coo
groosularlosrolla Cooseberrylest Glabomellow,
S Ube
c a cynniffile Acacia
tffo'" I 3oltbush
trial" canescens Fourving Saltbush Karens,
r D as cons"ens rourving Saltbush Rincon
triolex conescons Fourwing Seltbush Wytans
A tlDl*x ZO-nro-rfao I t a Shadescale r 0101 C. to Castle Valley Clover
nardner Car nor Saltbush
x1mis I
Mediterranean Saltbush case Ilse orl Quollbush
as numou Me Oldemn Saltbush
son ccats Australian Saltbush Corte
0. trMnUN Tridents Saltbush
*z un uT&U*onothus sontanus True Mountain Mahogany
FursTre riGnta Bitterbush Antelope

Species adaptability trials are a vital component to the success of future range seeding efforts. Seed germination and seedling establishment are the keys to revegetation. Adaptability of the species to local conditions (including grazing) is the key to longevity and persistence of seedings. Forage productivity along with stand longevity affect the economic feasibility of seeding efforts. There will always be new species, both introduced and native, that will require testing and evaluation. The results of trials in the Midelt region have shown substantial differences in adaptability and productivity among varieties within species. Comparisons of the standard varieties and newly released varieties, as well as comparisons among native species, should continue.
The weather conditions will remain unpredictable, be it a drought or monsoon. Since this risk is inevitable in semi-arid and arid environments, efforts in making this factor manageable by using techniques that are known to be effective should be continued. A certain degree of discouragement will always prevail, yet it is essential that these trials continue to be planted and evaluated, preferably on a yearly basis.
In contrast to the overall poor performance of introduced species was the relative-vitality of the native species, which appeared as "weeds" within the trials. This observation prompted a change in Project emphasis from evaluation of introduced species to research and development of native species.
The total responsibility for these Species Adaptability Trials, and for the native species, should be assigned to one individual as a Research Supervisor. This would require extensive travel to all locations, including the PMC. Coordination and evaluation of all the trials would best be handled by this person.
Guidelines for Research Supervisors
1) A desire and willingness to work is required.
Establishing, maintaining and evaluating trials is not
- an easy task. It requires constant planning,
coordination and consistent attention.
2) Advanced planning and preparation are essential.
Though weather caused delays during the Project, the
most serious problems which occurred were the result of
a lack of interest on the part of Project personnel'in
preparing to undertake the trial. Seed requests and procurement must begin in March, with October as the
target for planting.
3) Coordination is mandatory. There must be continuous
communication between local offices and the PMC. The
Research Supervisor must keep personnel at grazing
perimeters informed of all details of the PMC
production program. Without such communication, little
success can be expected from this program.

Wheatgrass Seedings: Plaine de L'Aarid and El Faija.
Early Field Trials: Plaine de l'Aarid
About 200 varieties of various grass species were planted in row and field trials in 270 to 330 mm precipitation zones in 1970 (Table 1.11). Only a fraction of these remain. Original evaluations of these trials were based on observations one year after establishment in 1971. Unfortunately, dry matter yield data for the original field trial evaluations in 1971 were not available. When the trials were evaluated in 1982, species from the genus Agropyron exhibited the greatest persistence. While none of the 1982 dry matter yields were impressive, it was the first growing season after a severe drought, and therefore, an important indication of drought tolerance, persistence and recoverability.
Tall wheatgrass (A. elongatum var. Largo) performed well during the 1971 evaluations and was subsequently seeded on a large scale at Plaine de l'Aarid. In 1982, yields of four and twelve year old stands of tall wheatgrass were compared on similar soils in the 330 mm precipitation zone. Yield of the four year old stand was almost double that of the twelve year old stand. There appeared to be more plants per unit area in the twelve year old stand due to basal expansion of original plants, but there was a higher degree of mortality and stagnation.
A difference in yield was not detected between tall and pubescent wheatgrasses (A. elongatum var. Largo and A. trichophorum var. Luna) in twelve year old stands at 530 mm average annual precipitation (Table 1.11). Again, there were low yields after drought and previous years of heavy grazing at that site. The growth form of pubescent wheatgrass at 330 mm precipitation was strongly rhizomatous and appeared to withstand grazing. This variety demonstrated longevity and reproduced vegetatively under severe conditions. Tall wheatgrass matured later than pubescent and showed no sign of reproduction. Weeds were not a problem in either stand.
At the location receiving 270 mm average annual
precipitation, tall wheatgrass was not considered a suitable alternative for range seeding. Yield for pubescent was double that of tall wheatgrass. Crested wheatgrass (Agropyron desertorum) yield was between three and four times that of tall wheatgrass. There was considerable stand mortality for tall wheatgrass in the 270 mm precipitation zone even though it was largely protected from grazing.
Crested wheatgrass was the highest yielding species in the 270 mm zone and produced large seed yields, possibly due to an exceptionally wet spring. Numerous crested wheatgrass varieties were impressive in both the 220 mm and 330 mm precipitation zones. However, early seedings of crested wheatgrass did not persist in some areas within the 330 mm precipitation zone. Almost pure stands of sagebrush have become established on these sites.
It appears that tall wheatgrass is a reasonable choice for a

seeding program at 330 mm precipitation and above. Pubescent wheatgrass is rhizomatous and persistent through a wider range of precipitation. What pubescent wheatgrass lacks in productivity could be offset by persistence and survivability. Perhaps a mixture of the two species would reduce the risk inherent in
Table 1.11. Production from row and field trials established in the 1970's at the Plaine de L'Aarid Grazing Perimeter. Values within columns by grouping are not significantly different (P<.05) if they are assigned the same superscript. Samples were oven dried 48 hours at 600C.]
Yiel4 Plantq Infloreqcences
(kg/m4) (#/m') (#/m )
1.) 12 year old stand
(330 mm ppt. zone)
Agropyron elongatum 600a 11a
A. trichophorum 645a 85b
Elymus junceus 180b 6c
2.) 12 year old stand
(270 mm ppt. zone)
A. elongatum 126a 5a 1 2a
A. trichophorum 232b 45b 12.0b
A. cristatum 421c 10c 104.0c
3.) Two A. elongatum stands
(350 mm ppt. zone)
4 year old stand 1136a 7.2a 34a
12 year old stand 600b 11.4b 32a
planting a pure stand of either species. Crested wheatgrass appears to be productive in both the 270 mm and 330 mm zones, however, there is no local yield data for this species in the 330 mm zone or above.

Range Improvement Trials: Plaine de l'Aarid
Five species of perennial grasses from the genus Agropyron were seeded using the rangeland drill and a Belgian made grain drill in the 300 mm precipitation zone on the Plaine de L'Aarid Perimeter during late fall, 1983. The rangeland drill was used to plant directly into native Artemisia herba-alba and an area that had previously been planted to Agropyron intermedium. As a result of heavy grazing pressure the original stand had been virtually eliminated. The land was not mechanically prepared prior to seeding with the rangeland drill. Earlier soil manipulation for the initial A. intermedium stand left a seedbed and eliminated competition from native plants. The more accepted practice of plowing, discing, and rolling prior to planting the seed with a standard grain drill was performed. Treatments of seeded species were: Agropyron cristatum (Fairway), native A. cristatum, A. elongatum (Largo), A. sibericum (P-27), A. smithii (Arriba), and A. trichophorum (Luna) and an unseeded control. Plots treated were five by forty meter strips and replicated three times for each treatment.
The ability of selected grasses to establish, persist and produce when drilled directly into native vegetation or a weakened introduced stand via the rangeland drill was compared to an accepted practice of planting. Low precipitation is probably responsible for the resulting low rate of germination for all species on all treatments. Plant establishment did not occur. Initial treatment with the rangeland drill reduced shrub canopy cover (live canopy intersect) by 28% and shrub density by 27% on the Artemisia herba-alba site. Two years later no significant difference was found between treated and control plots for either shrub canopy or density characteristics.
Range Improvement Trials: El Faija
In September, 1984, two range improvement trials were
established on El Faija Perimeter using seedings of wheatgrasses. These were designed to compare different land preparation and planting methods of Agropyron cristatum, A. elongatum, A. intermedium and A. trichophorum on a variety of soil types.
One of these trials was performed on a sloping area
dominated by Stipa tenacissima (alpha-grass). The site was burned to remove existing alpha-grass, then one third of the plots were plowed, disced and rolled before a grain drill was used to plant either Agropyron cristatum or A. elongatum at approximately seven and 12 kg pure live seed7Ea, respectively. The remaining plots were planted with the rangeland drill with no land preparation prior to seeding. One half of the remaining plots were planted using the rangeland drill with six discs to open furrows. The other half were planted with the rangeland drill using ten discs. The six-disc version has a wider row spacing and heavier discs than the 10-disc version, thus the distance between rows was greater and furrows were deeper. The same species were planted with both the rangeland drill and grain drill. Plots were 10 m x 25 m and treatments were replicated four times.
A second trial repeated in the sagebrush (Artemisia herbaalba) type on both shallow and deep soil sites, compared the

rangeland drill with six and 10 discs and the grain drill with prior seedbed preparation as in the first experiment. The same four Agropyron species were used in the trial. Plots were 10 m x 25 m and replicated three times.
Germination was slow for all species in all trials due to low precipitation in the El Faija area. Stand densities we e measured the following fall (September, 1985) using a one m frame. The conventional method of plowing, discing and rolling prior to planting with a grain drill showed better stand establishment than either of the rangeland drill techniques on the sagebrush shallow soil site. The RD-10 outperformed the RD-6 on this site also. No significant differences were found between treatments on the Artemisia herba-alba, deep soil site and the Stipa tenacissima site. Agropyron cristatum had the lowest stand density for all treatments on all sites. In this trial and other trials previously established, small seeded species were difficult to establish. Though plots were not clipped to estimate production (grazing occurred on the stands prior to their being clipped for production), the most robust plants appeared to be on the plots seeded to A. trichophorum. Stand densities by planting treatment and by species are shown in Tables 1.12 and 1.13, respectively.
Table 1.12. Mean number of plants per meter2 by site and seeding technique for trials at El Faija Perimeter performed in 1984.
Planting Artemisia herba-alba Stipa tenacissima
Treatments Shallow Soil Deep Soil
Conventional 22.55a 7.05a 7.58a
Rangeland Drill
(25 cm row spacing) 12.50b 7.83a 6.65a
Rangeland Drill
(15 cm row spacing) 6.42c 1.98a 4.78a
Values within columns with like superscripts are not significantly different (P>.01).
Table 1.13. Mean number of plants per meter2 by site and species at El Faija Perimeter for trials established in 1984.
Seeded Artemisia herba-alba Stipa tenacissima
Species Shallow Soil Deep Soil
Agropyron cristatum 1.80a 2.31a 0 72a
A. elongatum 20.16b 7.38b 11.95b
A. intermedium 18.64b 624b
A. tricophorum 14.69c 6.56b
Values within columns with like superscripts are not significantly different (P>.01).

Seeding and Interseeding Trials: Plaine de l'Aarid and El Faija.
Field Trial: Plaine de l'Aarid
A field experiment, applied to both shallow and deep soil sites, examined the possibility of seeding a mixture of two legume species (Trifolium hirtum (Hykon) and Vicia dasycarpa (Lana)) with four perennial grass species from the genus Agropyron. This mixture was tested with two seedbed preparation techniques, conventional preparation and use of the rangeland drill. In the conventional preparation treatment, plots were plowed, disced and rolled prior to seeding. On the remainder of plots, the rangeland drill with ten disc openers was used only to create furrows. All plots were hand seeded; one third received the grass seed mixture with T. hirtum, one third received the grass seed mixture with V. dasycarpa, and the remaining third received the grass seed mixture without a legume. Plots were five by ten meters and each treatment was replicated two times.
Germination was slow for all species regardless of the treatment in all trials due to low precipitation. Stand densities weri measured the following fall (September, 1985) using a one m sampling frame.
Total plant density was significantly higher (P<.05) for the treatment with Vicia dasycarpa using the conventional method of seedbed preparation. When seeding in arid zones or during periods of drought, adequate seedbed preparation appears to be critical to stand establishment. Species seeded (adaptability) and seed size also play important roles when moisture available for plant growth is perhaps the most limiting factor. Trials should continue in order to explore the possibilities that may exist with native species such as Dactylis glomerata and the local Agropyron species. Seeding techniques should also be evaluated further and under different environmental conditions since the cost of applying these techniques plays a major role in the determination of economic feasibility of rangeland seeding.
Field Trial: Plaine de L'Aarid
Technicians from the Midelt Office supervised the
interseeding of two legume species, sainfoin (Onobrychis viccifolia) and sweetclover (Melilotus officianalis) within an established stand of Agropyron elongatum on the Plaine de L'Aarid Grazing Perimeter. An experimental area was interseeded with legumes (17 kg pure live seed/ha) and fertilized with triple super-phosphate at three different levels during the first week in October, 1983. A separate trial was performed for each of the legume species. Two treatments for each legume were tested, inoculation and fertilizer rate, in a 3x3 factorial design (split-plot). Inoculation treatments were inoculated seed, seed without inoculum, and no seed planted. Triple super-phosphate was applied at zero, 40, and 80 kg/ha.
Sainfoin displayed limited germination but failed to become established. Sweetclover did not germinate. Plots were clipped in mid-July, 1984. No significant difference in legume cover was found among any of the trials. The lack of establishment of these leguminous species was probably due to the limited

precipitation received during the 1983-84 season.
Seeding with Limited Land Preparation: Ait Rbaa
Experimental Seeding
The 20 hectare parcel used in this experimental range
improvement attempt is located on the extreme west end of the grazing perimeter on deep, flat bottomland soils. The trials were designed to determine the effects of soil treatments and seeding on forage production.
A mixture of five legumes and four grasses was used in the seeding treatments (see Figure 1.18 for plot layout and species mixture). Soil was prepared using a 24 disc offset harrow and a chisel. Land treatment and seedings were both performed on October 30, 1985. The effort benefited from rains received shortly after treatments were installed. By the middle of March, the 20 hectare ripped area was dominated by a solid stand of native annuals, composed mainly of a yellow mustard (Raphanus raphanistrum) which reached over 1.25 meters in height. These native annuals accounted for more than 90% of the cover. The remainder was composed of Vicia dasycarpa (5%), seeded perennial grasses including Agropyron intermedium (Slate) and Dactylis glomerata (Palestine) (3%), and another seeded legume Medicago littoralis (Harbinger) (1%).
----- ---st -- spool,..*$ i ur.
a Percent of Se*ding
SSpectee MLcture Race
_* "(2) (kg) /ha)
(S31te) 6.6 2.5
. ~~Cynodo do.
IiDac ls oral
- Pel-lT1 3,8 2.2
(Kacb'lsier) I3.4 N -.
(Serene) 13.I 7.5
TrLfolius hirtus
N kon n) 1.7 1.0
Srifoligu subterr.
G (Lana) 36.0 20.6
- it dcv.P
(Flinclock) 10.5 6.0
9. TOTAL 100.0 $7.2
15 .1.
S etl cal locks~ Desln Treacs.. t
Pt. Error
0 50 10Q 150 200 Pl trror 0
sal. (eeroe)
Figure 1.18. Plot layout and species mixture seeded at Ait Rbaa Grazing Perimeter.

Small Demonstration Site with Limited Land Preparation
The objectives of this demonstration area were to develop a visual demonstration site showing the effectiveness of developing terraces on natural contours for watershed management and for increasing forage production.
The site of the demonstration was located on the eastern end of the Ait Rbaa Perimeter in fenced exclosures (two exclosures are present at that site, a ridgetop exclosure and an exclosure located in a swale; the demonstration was located in the lowland exclosure). Contours were surveyed by technicians at the Beni Mellal office. Terraces were dug by local hire. The local Centre de Travaux (CT) was contracted to chisel the area behind each terrace. A mixture of six species (Cynodon dactylon, Agropyron intermedium, var. Slate, Trifolium subterranium var. Clare, Dactylis glomerata var. Palestine, Vicia dasyacarpa var. Lana, and Eragrostis curvula) were applied at a rate of four kilos each for a total of 24 kg/ha. Three varieties of Atriplex shrubs were planted behind the terraces facing the upper part of the watershed. Grass clumps provided by the PMC were planted at the junction of the terrace and the chiseled ground. It was hoped that clumps, being mature, live plants, would have a higher rate of survival than hand seeded species. Results of these trials are presented in Table 1.14.
Table 1.14. Names, number and survival of planted species.
Hand Seeded kg Seed Planted Germination
Cynodon dactylon 4 med
Agropyron intermedium 4 med
var. Slate
Trifolium subterranium 4 med
var. Clare
Dactylis glomerata 4 low
var. Palestine
Vicia dasycarpa 4 high
Transplanted Grasses No. Planted Survival
Erharta calycina 37 27
var. Mission
Eragrostis lehmanniana 26 4
Agropyron desertorum 48 43
var. Nordan
Dactylis glomerata 40 38
var. Berber
Sporobolus airioides 37 0
Phalaris tuberosa 58 31
var. Seedmaster
Phalaris tuberosa 46 28
var. Serona
Cenchrus ciliaris 68 8
var. Nueces
Agropyron trichophorum 36 26
var. Lana

Table 4.14 (continued). Names, number and survival of planted species.
Dactylis glomerata 34 32
var. Palestine
Eragrostis curvula 23 20
var. Catalina
Eragrostis superba 52 9
var. Palar
Hand Seeding in Zizyphus Clumps on Degraded Range
This study was initiated to determine whether the area under the Jujubier clumps, due to the high levels of organic matter, visibly less compact soils, and protection from grazing, would be a suitable site to reintroduce perennial rangeland grasses by hand seeding.
Twelve paired Jujubier (Zizyphus lotus) plants were chosen
within the confines of an exclosure on the eastern end of the Ait Rbaa Perimeter. A one m2 plot was staked surrounding each pair of plants. The area was raked of debris and then randomly seeded with one cf the following species: Agropyron intermedium var. Slate, Cynodon dactylon, Cenchrus ciliaris, or Erharta calycina. One plant from each pair was seeded at a rate of 12 kg/ha. The experiment was maintained for only a single season. During the summer months the majority of the shrubs were illegally removed by area herders for firewood and fence building material. The results of this study failed to show that this seeding method is a reasonable alternative with any of the species used during years of below average precipitation.
Ripping and Terracing: Oujda Office
Construction and Evaluation of Contour Terraces
The decreased plant cover on Eastern Moroccan rangelands, caused by animal and human pressure coupled with drought, has accelerated the amount of soil erosion. Technicians at the Oujda Office took measures to capture precipitation runoff, increase plant cover, and forage production through investments in major land treatment. The goal of the construction of contour terraces was to manipulate the soil to insure maximum water retention and survival of plantings of Atriplex nummularia. This shrub species was chosen for planting because of its demonstrated potential for forage production in the harsh climate of Eastern Morocco.
In installing these contours, objectives were threefold: 1) to determine the survival rate of Atriplex nummularia within the contour system, 2) to determine the annual forage production of the species, and 3) to determine the effects of the contour terraces on native vegetation production. It was expected that A. nummularia would survive in the contour system at a level sufficient to warrant large scale use of this method of

plantation. Native vegetation was also expected to benefit from increased soil moisture due to the water harvesting effect of the contour terraces.
The contour terraces were constructed with hand labor, using picks and shovels provided by a Promotion Nationale grant. These grants are to be used for labor on projects for the betterment of the community. In the Ain Beni Mathar area, the site chosen for the demonstration was on the Fritiss Range Perimeter, where contours were used to harvest runoff from the alfa-grass zone and allow infiltration into the productive soils of the sagebrush zone. The contours were surveyed in at 10 meter surface distance intervals.
The first step in the construction process was to dig a trench 20 cm wide x 20 cm deep which would become the rooting zone for the shrubs. The top soil from up the slope (1-1.5 m) was shoveled down to fill the trench and make a furrow of "good soil." After completion, the contour terrace was approximately 40 cm high and packed to resist erosion. In places of expected heavy runoff the contours were fortified with a rock backing covered with soil.
The Atriplex nummularia plants were supplied by the Forest Service nursery in Bouarfa and planted at two meter intervals along the terraces throughout the system. Each of the 5,000 plants was given five liters of water on planting day.
Shrub production data was collected on the Ain Beni Mathar Perimeter after the shrubs were in place for one year. Production was evaluated by ranking each of the 5,000 shrubs into one of five classes, based on visual observation. Rank one was considered poor production and rank four very high production. A zero rank was given to dead plants. To convert these classes into production, four representative plants of each rank were stripped of all forage material and the material was bagged, dried and weighed. The dry matter forage productivity estimates presented in Figure 1.19 are based upon the average of these samples.
0 9 --I 25 17.12,g
I I 23 33.80 g
,III 30 69.85 g
IV 13 245.44 g
Figure 1.19. Production classes of Atriplex nummularia shrubs on contour terraces on the Fritiss Range Perimeter near Ain Beni Mathar.
With a plantation density of 1,000 plants per hectare, the first year production of Atriplex nummularia was estimated at 63.9 kg/ha of dry matter. Although this level may seem low, it represents only part of the total annual production of the site. The average forage dry matter production of sagebrush on contour terraced sites was 447.4 kg/ha making the combined estimated

forage production of sagebrush and A. nummularia on the terraced area 511.3 kg/ha, or over five times the production of sagebrush sites that were protected from grazing. (On grazed sagebrush sites, total production was estimated to be only 17.3 kg/ha. The site yield under protection from grazing was estimated to be 84.0 kg/ha). The water harvesting effect of the contour terraces adds considerably to production of forage on sagebrush sites. It was demonstrated that transplanted shrubs can also be used successfully with this technique to extend the forage base. Longer term effects of terracing and shrub planting will be monitored. As transplanted shrubs reach maturity and full size, their contribution to forage production should increase several fold. The presence of the Atriplex plants should also have a moderating effect on the microenvironment of the site making it more mesic than the surrounding area. If this proves to be the case, production of sagebrush and other forage species could be expected to increase.
The site chosen for contour terrace construction at the Gouttitier Experimental Farm was a moderately sloping shallow soil site. This site was selected because of its close proximity to the main highway.
The construction method used was similar to that employed at Ain Beni Mathar, though modified to fit the shallow soil sites. Construction was by hand labor using picks and shovels. A 20 cm x 20 cm x 20 cm hole was dug at two meter intervals to serve as the planting site. These holes were then connected by a contour terrace of soil from the up slope side. Holes were filled with top soil from up slope. The finished contour terrace was 30 cm in height and was packed to resist erosion.
The contours were planted with Atriplex nummularia supplied. by the Forest Service nursery in Taourirt. Each of the 3,700 shrubs were given five liters of water on planting day.
The Gouttitier contours exhibited many of the problems of the Ain Beni Mathar contours. The major problem was caused by the slope of the site and the nature of the storms in the Taourirt area. Precipitation events in this region are characteristically of high intensity and short duration. This causes considerable runoff which in turn causes damage to the terraces. Many of the problems of structural failure could be solved by proper forethought in placement and consideration of soil types involved. A site with a greater productive potential and less slope could have been chosen for the demonstration at Gouttitier but it would not have been possible to locate it in an area that could be readily seen from the highway.
Production data (Figure 1.20) was estimated using the sample unit method employed at Ain Beni Mathar. With a plantation density of 1000 plants per hectare, the first year production was estimated to be 28.6 kg/ha. This low level of productivity gives some indication of the limitations imposed by the slope and shallow soils at the plantation site.

0 25 --I 26 16.30 g
II 32 26.70 g
III 9 63.80 g
IV 8 126.65 g
Figure 1.20. Production classes for Atriplex nummularia planted on contour terraces at the Gouttitier Experimental Farm.
After a high intensity storm, it was observed that long terraces needed breaks to allow water to pass to the next terrace. Flow of water can only be slowed, not stopped altogether. Breaks at 30 50 m intervals were opened and alfagrass and rocks placed over the exposed ends of terraces to protect them from erosion. Any watershed management program involving contour terracing (or other improvements involving water management) must start at the head of the watershed basin to gain control of the system.
Sites with deeper soils responded considerably better in
overall shrub production than the poorer sites. A petro-calcic horizon very near the surface of the poorer sites appears to inhibit plant growth. Conversely, this same petro-calcic horizon on the deeper soil sites seems to aid in downstream percolation of soil moisture. The water infiltrates and flows laterally on the petro-calcic horizon adding to the available soil moisture
for native vegetation down slope.
One major problem that added to structural failure in an
intensive runoff event was rodent damage. Newly formed terraces of fresh soil provided a perfect nesting place for many of these nocturnal desert dwellers. Many places in the contour terraces were literally honeycombed with burrows.
Additional Terracing Trials
The Bureau des Parcours was allocated funding to reclaim 400 hectares in the Ain Beni Mathar area. The America site (the site of an American Air Force Base during the North African Campaign of World War II) was selected because of its close proximity to a major highway and the local population. The area is divided into two sections, the site of a ten year old Forest Service plantation that failed and a small watershed. The plantation site was planted using a method of impluvium or "hoovera", a pit surrounded with a crescent shaped terrace for water catchment and storage. The small watershed was planted using a system of contour terraces. This will be an excellent chance to compare the two plantation methods for survival and .plant production. The Forest Service method has been widely used in all parts of Morocco in tree plantations. Its utility or effectiveness in the semi-arid and arid zones appears to be negligible however.
The contour terraces were surveyed in at each 20 cm drop in elevation (about 20 meters between contours). The construction was done in two stages: 1) a rooting zone was cut and an initial

terrace was constructed by tractor and three disc plow, and 2) hand labor was used to complete the construction of the terraces. The tractor and plow ran along the contour in one direction to the end of a contour, then reversing direction, ran along the same cut to the starting point of that contour. This method caused a 20 cm trench to be cut and the soil was thrown in a rough terrace on the down slope side. Laborers then filled the trench with top soil from up slope and worked the terrace to stabilize the soil. The contours were broken every 50 meters to allow flow of water to proceed down slope at a slower rate. The ends of the breaks were rocked to protect them from erosion.
The area was planted with Atriplex nummularia supplied by the Ain Beni Mathar Forage Plant Nursery and the Forest Service nursery in Oujda. A total of 260,000 shrubs were planted at two meter intervals with each receiving five liters of water on the planting date.
A disc plow was used to scarify the soil to a depth of five to seven cm at three different sites within the contour system. This was done to open the soil for increased infiltration. Two of the three sites disced were planted with a mixture of Eragrostis lehmanniana and Sporobolus airoides seeded by hand. Seed was then covered with a clump of jujubier branches drawn over the ground by a donkey.
The native vegetation showed somewhat more robust growth in the disced areas than it did in untreated areas. Stipa parviflora was definitely more productive in these areas compared to the untreated sites. Plants broken by the disc seem much more healthy than those untouched. There was also a marked increase in the native shrub growth in the treated areas. As of May 1, 1986, seeded areas had shown no sign of germination of the seeded species. This is probably due to the low temperatures and lack of moisture during the four months after planting. A major sandstorm occurred three weeks after seeding with winds of 150 km/hr. Since the two species used have very small seeds, this storm may have removed much of the seed from the ground.
The Atriplex nummularia shrubs will be evaluated four times annually. The first quarter evaluation showed a death loss of 25%. Much of this loss can be attributed to damage done by windstorms. Many of the impluvium planted shrubs were literally covered with sand.
This range improvement was done by a hired contractor. Many problems could be eliminated if a representative of the Bureau des Parcours was on site at all times to oversee the work.
Fertilization- and Ripping: Ait Rbaa
The cumulative effects of increased animal numbers, drought, uncontrplled use and soil nutrient deficiencies have, over time, reduced available forage on the Ait Rbaa Perimeter. Sulfur fertilization and mechanical chiseling of the soil were two treatments introduced on the perimeter with the objective of providing an inexpensive method of increasing forage production from native species.
The study site was located at the eastern end of the Ait Rbaa Perimeter. Two locations, situated within a fenced exclosure, were assessed to be representative of sloping upland

and bottomland sites found on the perimeter. Three replications
of six different treatments in a randomized block design were installed at each location. Elemental. sulfur was applied at
three different levels (0, 30, and 60 kg/ha of 98% pure sulfur).
Plots were either chiseled or left undisturbed. Soil moisture and water infiltration rates were monitored seasonally. Forage
production was determined at peak season for native legumes,
other forbs, and native grasses.
Graphs of soil moisture data collected over two growing
seasons indicated that, regardless of location, chiseled sites were more capable of retaining soil moisture than non chiseled
sites, both at soil depths between 0 and 10 cm and between 10 and
20 cm (Figures 1.21, 1.22, 1.23, and 1.24). Though statistical analysis has not yet been performed on the data, the values are encouraging. This range improvement practice may be useful at
Ait Rbaa Perimeter and other areas with similar climate and
23 24
22- 2
21 -2
1I, l
Neg-s + ss-o 0*< los M/-** Me/es v-r uo
14 1 .
Loaio. (94-5. oaionI 1948)
13 1
12- 12* ..a* ..
I11 0.
d L 2 I I I 21 29 I i2 2I I I I I 3 I I I
$. jO Ilj8 I g1 j At MW DeC J404 ftb Mwr Sc May JUG
CV0o-10 4 N ca/10-0 e o/-1o a 0/10i2 =/v-I* 4 Ma/ -o c s /vo-1t c/10o-2.
Figure 1.21. Soil moisture, Figure 1.22. Soil moisture,
Location I (1984-85). Location II (1984-85).
20 23
24 21
22 I,
14 12
26 10 29 113 20 IO24Sb2
oat N" IL.724 2 21 IP 19 L71 3 J132 ti. 26 Ij.24 Iob 21 L:I 4, 11 L I- R
l1"0t04 ICV/?-26 4 V-1 eCIIo-z. eve-I* CV10Q26 0 C~V0-1@ 4 V0Figure 1.23. Soil moisture Figure 1.24 Soil moisture
Location I (1985-86). Location II (1985-86).

The results of measurement of soil infiltration (Figures
1.25 and 1.26), at least when plotted, indicate that during the
first year after chiseling, higher mean infiltration rates can be
expected in the chiseled areas than in the non chiseled areas.
During the second year after chiseling, the treated areas were
again showing higher infiltration rates, but the differences
between the chiseled and non chiseled areas were not as
pronounced. This might be expected since both areas were
protected from grazing and plant cover and growth were increasing
annually. This plant cover and biological activity in the soil
and at the soil surface was obviously changing the soil
characteristics in terms of compaction. The actual effect of
chiseling may be only short term, especially if grazing practices
are not changed to reduce trampling and compaction of the soil.
to- 7
7 5
OS-Oa" 16-N" 17-Doe 22--n i-Frb S-Mr 14 .-W 1S--Y 23-Oct 14--g, l*-Om 1t-.Je 21-7eb 14-Mar 17-,p 15-May
o~/ NON/, 1 NO,. Ui IGW P(UOOS
Figure 1.25. Infiltration Figure 1.26. Infiltration
rates, 1984-85. rates, 1985-86.
There was a statistically significant increase (P<.05) in
legume production on chiseled over non chiseled sites. Native
annual legumes provide a large portion of the available palatable
forage produced at Ait Rbaa Grazing Perimeter, especially in dryer years (around 50% in 1984-85). Within each forage type
(leguminous forbs, non-leguminous forbs and grasses), no
significant differences in production were detected among the
three levels of sulfur. Positive effects of the sulfur
application may emerge in future years. Given the data from the
initial two years of sampling, however, sulfur fertilization with
elemental sulfur shows little promise for increasing forage
Fertilization Trials: Plaine de L'Aarid
Stands of perennial grass species, which were established in
field trials planted on Plaine de L'Aarid Perimeter in 1970, have
persisted into the 1980's. The vigor and productivity of these stands, however, has declined substantially. In September 1985,
technicians from the Midelt Field Office applied nitrogen (N) and

phosphorous (P) fertilizer to stands of Agropyron elongatum, A. intermedium, A. trichophorum and Elymus junceus. The trials were designed to show forage production response at various levels and combinations of N and P. An experiment with a 5x5 factorial design and four replications was installed. The N source was ammonium nitrate at 0, 20, 40, 60 and 80 kg/ha. The P source was triple super-phosphate also applied at 0, 20, 40, 60 and 80 kg/ha.
Agropyron elongatum and E. junceus plots were clipped in mid-July 1984 to determine the response in productivity which resulted from fertilizer treatments. No significant differences were found among treatments. Limited precipitation during the trial year made water such a limiting factor that possible responses to fertilizer were unlikely. Current vegetative growth on the A. intermedium and A. trichophorum stands was virtually nonexistent that year. No production data was collected from the plots.
Alpha-grass Trial: El Faija
Trials established on the Moulouya Plain in recent years
have produced little information other than the fact that, during periods of drought, little or no plant establishment and growth occurs due to insufficient soil moisture. Infrequent precipitation and quick runoff when rain occurs, results in low levels of available soil moisture. An increase in available soil moisture could enhance plant establishment and subsequent forage production under dry or more humid conditions. A trial to determine the effects of integrating contour terracing into range seeding efforts on burned Stipa tenacissima sites was established on the El Faija Perimeter during the fall of 1985. The trial was designed to assess the effectiveness of terraces constructed on contours in reducing runoff and to assess the potential for harvesting water (concentrating runoff from precipitation onto productive zones).
Six contours were surveyed on the study area. A two tine
ripper with tines set one meter apart was used to rip soil along contours to a depth of approximately 30 cm apart. Terraces were constructed by hand along the lower edge of the ripped area. A 2x3 factorial split-plot experiment was designed where terraces were tested as primary treatment plots and planting methods were tested within plots. Three planting methods, compatible with the terraces were evaluated: 1) the total surface area between the terraces was planted, 2) the lower one half of the surface area between the terraces was planted, and 3) no planting (control).
Previous research at El Faija had indicated the techniques and species that were most likely to be effective. Conventional seedbed preparation (plow, disc, and roll) prior to planting with a grain drill has produced higher levels of plant establishment than direct planting with a rangeland drill in the majority of recent research and demonstration efforts. A mixture of seven grasses (A. elongatum (Alkar), A. elongatum (Jose), A. intermedium (Slate), A. intermedium (Tegmar), A. trichophorum (Manden), A. trichophorum (Topar) and Dactylis glomerata (PalestineTT and two legumes (Trifolium hirtum (Hykon) and Vicia dasycarpa)) were seeded on the study area. Two species of

saltbush (Atriplex) shrub seedlings were transplanted at two meter intervals in the trough above the terraces (if terraces were not present shrubs were planted in the lower rip along each contour). Atriplex nummularia was planted along the first, second, fifth and sixth contours. Atriplex corrugata was planted along the third and fourth contours of the study.
Terrace construction was continued up the slope to the hilltop above the study area to prevent runoff damage to the trials. Terraces were built along the contour, with one meter increments in elevation between terraces until the hillside became steep. The increments were then increased to one and onehalf meters, and finally to two meters.
Students from INAV Hassan II will monitor the trials for
effects on available soil moisture and corresponding changes in the grass/legume mixture, shrub establishment and production. Four moisture measurements should be taken in the interspace between terraces in each treatment. The first should be taken one and one-half meters above the lower terrace and the fourth should be taken one and one-half meters below the upper terrace. Two readings should be taken at equally spaced points between the first and the fourth. Data collection at these locations would enable the researcher to determine if seepage below the terraces occurs, the effects of terracing and land treatment on soil moisture between terraces and the extent to which terraces increase soil moisture in the catchment formed above the terrace. One meter square sampling frames should be used to sample stand density and production between terraces including the locations where soil moisture is sampled. In addition, shrub survival rate and productivity should be estimated.
Grazing Exclosures: Ain Beni Mathar
Four distinct vegetation types occur on the fenced 300 ha
experimental area at the Fritiss Grazing Perimeter. The need to monitor these plant communities has been recognized due to the changes in plant composition that have taken place over the last two years with near average precipitation. A number of annual forbs and grasses have been collected that were once thought to be locally extinct.
With the idea that the Experiment Station will be used for\ grazing studies of various kinds, four 50 m x 50 m exclosures were constructed in 1985, one in each of the four vegetation types. Type #1 was a water collection area where a dense stand of sagebrush had formed. Type #2 was an upland sagebrush site. Type #3 was the sagebrush/alfa-grass transition zone and Type #4 was a unique site where sand from neighboring fallow fields has been accumulating. Permanent transects were established for monitoring purposes. Though quantifiable data on recovery is not presently available, plans have been made to evaluate each plant community annually.

Grazing Exciosures: Gouttitier Experimental Farm
The land upon which the Experimental Farm was established has a history of mismanagement. With the completion of an enclosure fence, much of the trespass grazing has been eliminated. Since many sites in the region are similar in that an on site native seed source appears to be virtually nonexistent, monitoring the recovery of natural vegetation on the area will give an indication of the response that can be expected in other local reclamation efforts.
Two 50 m x 50 m exclosures were constructed during 1985 on two different range sites on the Farm. Site #1 is in a degraded sagebrush site which, after a single year of deferment from grazing, has shown a remarkable recovery of vegetation. Site #2 is a degraded upland site occupied by a thin stand of sagebrush. This area has also responded dramatically to deferment. Data on recovery of these two sites will be collected annually.

During the 1985-84 growing season, the amount of available forage was estimated weekly by clipping on two distinct range sites at Ait Rbaa Perimeter (20 plots were clipped each week on each site). The clippings from each plot were separated into three components: grasses, legumes, and forbs. Samples were stored on ice in the field then placed in an oven the same day for 48 hours of drying at 800C. After weighing to determine productivity, samples were ground in a mill and combined, resulting in one sample per location per weekly sample date. The milled samples were analyzed for crude protein and ash content (minerals) by Dr. Hamid Narjisse at ENA in Meknes.
The two sites were fenced and excluded from grazing in 198283. They were located on the eastern end of the perimeter and represented: 1) ridgetop sites with shallow soils, and 2) swale sites with deeper soils. The ridgetop site was dominated by Malva parviflora with few grasses. The swale site was dominated by Asphodelus tenuifolia with numerous grasses including Cynadon dactylon and Stipa retorta.
Crude protein was selected as the criteria for evaluating
forage quality because it is the most expensive to supplement and because it is often the most limiting nutritional factor for ruminant maintenance and growth. Unfortunately, data for energy were not available. Energy is usually the second most limiting factor, and in this case, may well have been the primary nutritional constraint. Data was compared using an analysis of variance for a completely randomized block design.
Results and Discussion
Peak season forage production occurred on March 19th for the ridgetop site and March 26th for the swale site (Table 1.15) yielding 831.2 and 754.8 kg of oven dry forage per hectare, respectively [Harkousse et al. (1984) for a discussion of the production data]. Clipping began on December 12th when range forage was actively growing following substantial precipitation. This was the low point in production, but yielded the highest levels of crude protein and ash. The level of crude protein in samples from the swale site ranged from a high of 21.71% on a dry matter basis (DM) on December 12th, to a low of 14.41% DM on April 16th. The range of values for crude protein was from 25.36% on December 19th to 17.44% on April 16th for the ridgetop site. Percentage of Ash (DM) followed the same trend indicating some loss through leaching and maturation through the growth cycle.
The percentages of dry matter crude protein (Figure 1.27) and ash (Figure 1.28) gradually declined as vegetation matured through the growing season. Moisture caused an immediate growth response in vegetation (until the time of seed dispersal) regardless of its stage of growth. This quick growth response to precipitation caused small increases in DM protein content throughout the season.

Table 1.15. Kilos of oven dry forage per hectare, moisture content, and crude protein and ash as a percent of dry matter.
S Moisture Crude I Moisture Crude
Sample of Total Protein Ash of Total Protein Ash
Date Grass Legume Forbs Total Forage z Z Grass Legume Forbs Total Forage I %
Dec. 12 4.00 14.40 159.8 178.2 21.71 18.28 18.0 184.6 202.6 23.44 19.71
19 0.24 .72 207.6 208.6 20.29 21.40 14.0 219.4 233.4 25.36 22.18
26 4.40 15.80 433.6 454.0 86.7 20.03 19.58. 36.4 544.8 581.2 86.7 22.43 20.82
Ja. 2 2.00 18.80 393.2 412.4 86.3 20.54 17.85 55.6 544.0 600.0 86.3 22.90 20.04
9 1.80 28.00 356.8 386.8 85.6 19.55 17.36 74.0 605.6 680.0 85.1 22.29 19.64
16 1.60 24.80 356.0 382.8 83.4 19.46 17.24 56.4 480.4 536.8 83.4 22.03 16.35
23 3.60 25.20 266.8 295.6 50.5 19.3S 16.45 0.6 80.4 616.8 698.0 61.0 23.46 16.50
30 3.40, 32.00 331.0 372.0 86.2 18.39 16.06 45.4 608.4 653.4 70.9 21.56 15.80
ftb. 6 3.20 34.00 336.0 373.2 82.9 18.54 14.58 73.6 654.8 728.4 70.7 20.41 15.e
13 16.00 34.80 346.0 371.2 76.0 17.95 14.22 50.8 523.2 574.0 67.4 21.56 15.61
20 4.40 29.20 352.0 385.6 87.6 18.11 13.30 79.2 698.0 777.0 81.3 19.14 17.41
27 23.60 42.40 323.2 289.6 78.2 17.40 14.97 64.4 522.0 586.4 67.8 19.81 16.79
skr. 5 26.80 54.40 313.6 395.2 75.8 17.63 13.75 84.8 558.4 643.2 76.3 18.54 16.10
12 16.80 54.00 428.4 499.6 71.0 15.71 13.59 72.0 528.4 615.8 71.1 22.65 15.01
19 37.50 77.20 555.4 670.2 65.5 16.31 12.48 0.6 140.4 690.2 831.2 62.1 22.40 14.15
26 56.40 102.80 595.4 754.8 76.8 16.51 14.93 3.2 137.6 649.6 790.4 73.8 19.84 14.91
Apr. 2 27.80 87.80 392.8 508.4 59.7 15.29 10.01 10.0 69.0 499.0 578.0 54.0 18.08 13.01
9 74.60 45.60 575.4 695.6 29.9 16.39 11.03 8.0 110.4 440.4 558.8 49.4 17.74 13.67
16 22.40 40.40 432.4 495.6 54.3 14.41. 9.12 6.0 70.4 396.4 473.2 44.0 17.44 14.93
24. 21
21 7
0 la
16. 14.
17 Q. 1312
012 19 28 J2 9S 1 23M f4O 1 20 27 Me 12 3* 20 A 9 10 012 1. 2 J2 9I 23 e 13 210 27 US t2 11 28 A2 3 14
SAWL"DA 20AA 0*e
a UP.QO Srl1 4 .LLEY W3113 a uP..* 2= 4 YNAZL WM1
Figure 1.27. Forage crude protein Figure 1.28. Forage ash concontent (dm basis), 1983-84. tent (dm basis), 1983-84.
An analysis of variance indicated that there were highly significant differences (P<.01) between locations and between dates of sampling (Table 1.16). Differences between sample dates were normal and followed precipitation patterns and plant development as expected. Real differences between locations indicated that the forage base on sites dominated by Malva parviflora and Medicago laciniata (the ridgetop sitesT were truly superior to those dominated by Asphodelus tenuifolia, Medicago

laciniata, and Cynodon dactylon (the swale sites) as far as DM crude protein content was concerned.
Table 1.16. Analysis of variance of crude protein content of
annual forage.
Location 1-1=1 87.04 87.04 67.47***
Date d-1=18 135.65 7.54 5.84***
Error [1-1][d-1]=18 23.14 1.29
Total ld-1=38 245.83
*** Significant at P<.005.
It must be kept in mind that these crude protein figures
were calculated on a dry matter basis. One kilo of green forage consumed by an animal on December 26 at Ait Rbaa would have, on the average, provided only 133 g of dry matter (86.7% moisture content). The same forage contained 20.03% crude protein DM. This means that one kg of green material only contained 26.64 g of crude protein, and even less digestible protein. A 50 kg lactating ewe requires approximately 136 g of digestible protein daily and 1800 g of dry matter. Therefore, she must consume about 5.68 kg of green forage to satisfy her needs for protein, and about 13.5 kg to satisfy her need for dry matter. It is doubtful that a ewe could consume enough forage during a grazing day at Ait Rbaa to satisfy her protein requirement. A ewe would find it impossible to satisfy her dry matter needs. For this reason, there is an obvious need to provide a supplement to the native forage. Sheep in the area were, in fact, supplemented heavily with various feeds including barley, beet pulp and straw.
There was both a forage quantity and quality problem at the Ait Rbaa Perimeter. This conclusion lends credence to the general opinion voiced by government service range management personal in the region that this communal grazing land is simply a holding area during the cereal production season, and it is not expected to provide the requirements of livestock. This may also provide some insight into the attitudes of right-users, who seem resistant to the prospects of range improvements and who appear unconcerned over the continuing degradation of the natural resources of the Ait Rbaa Grazing Perimeter.

Existing Forage__Resources
Moroccan rangelands are situated over a diverse range of climatic and ecological zones, from desert shrublands to high mountain grasslands. Good rangelands can usually be converted to marginal croplands, and therefore more and more rangeland is being converted to cultivated cereals crops each year by a rapidly expanding population, even though little profit from subsequent grain harvests can be realized. These crops do provide a standing forage crop if only as stubble. By expanding crop production activities into marginal areas, the Moroccan producers have compensated for poor range condition by producing a cereal crop, and at the same time, are controlling the land for their personal use. Cereal crop subsidies and land tenure policy by the GOM encourage this practice, making this practice feasible for the individual at the long-term expense of rangeland productivity and the communal rangeland users.
Pastoralism in Morocco has, nearly completely, given way to an agro-pastoral way of life. In agricultural areas, where good
rangeland has been converted to poor cropland,'pasturage is scarce and the remaining range areas are used heavily (generally in the complete absense of any kind of management or control). Ait Rbaa Perimeter, located in the midst of a large area devoted
to cultivated agriculture, provides a vivid example of this kind of abuse. Sheep depend on crop residues for a large portion of their forage each year. After harvest, shepherds herd large numbers of flocks through cropland to eat stubble and weeds. As planting begins, livestock are withdrawn from cereal fields and herded on communal rangeland until after the harvest. Here they quickly consume emerging vegetation. Most herds, even on native rangeland, receive large amounts of relatively costly supplemental feeds such as barley and by-products from factories which process agricultural products (sugar beets, cotton).
Moroccan rangeland is generally devoid of its original
component of palatable perennial herbaceous vegetation. Many forage species, both legumes and grasses, which were reported as present during the earlier USAID Range Improvement Project (60864-4) could not be found by a former team member of that project, Mr. Walter Graves, during his TDY visit in 1985.
Soils on grazing perimeters were found to be severely eroded and highly compacted. Runoff from precipitation was high, with moisture rarely penetrating the soil in amounts adequate for establishment of perennial forage plants, even in years of greater than average rainfall.
IEverywhere evidence of the need for range management and for the adherence to its most basic concepts was observed. Not until the range manager can control animal numbers and the timing of their access to grazing, can he hope to substantially improve and increase the range forage resource. This can only be done effectively through the cooperation of people using the resource. The benefits to the people must be determined, demonstrated and developed if the trend toward further degradation is to change.

Alternatives for Rangeland Forage Production
Establishment of perennial introduced species has been generally difficult and expensive. Stands which have become established, such as those in the perimeters administered by the Midelt Office, are the result of adequate seasonal precipitation and hard work on the part of range technicians. Though a number of stands, composed of introduced species from the genus Agropyron, are managed at Plaine de l'Aarid Perimeter; the condition of these stands has undergone substantial natural deterioration in the years since their initial planting. No juvenile plants are present and all remaining grass plants are overmature with some stands in need of renovation.
Warm-season grasses, even when of native origin, have proven to be difficult to establish on Moroccan rangelands. Seeds are generally small and expensive to purchase. Successful establishment depends upon good precipitation during and after germination and careful control of planting depth by technicians. The possibility of failure is too great to make this an economical venture unless techniques are improved.
Experience indicates that seeding of most introduced species in range improvement programs is an intervention that is generally too risky to be assumed by private producers. A farmer would rather have a poor stand of barley or wheat from which he may obtain some grain and stubble, than a year lost to barren ground and weeds. Years of adequate seasonal precipitation, sufficient to establish adequate stands of these species, are too rare to guarantee any measure of their success in many areas.
The potential displayed by plants native to Morocco is the most promising aspect of the Project work with range forage species. Among these plants are ecotypes highly adapted to climatic and soil conditions on the grazing perimeters. Among accessions from the native seed collection being propogated at the PMC and undergoing trials at the grazing perimeters, there were annual species which have the ability to produce the large amounts of seed necessary to maintain a presence on managed communal rangeland. Leguminous annuals, especially from the genus Medicago, exhibit many vegetative growth forms which may allow them to persist and produce forage under heavy grazing pressure. These species characteristically produce forage of high quality and produce large amounts of seed. Their use may be one key to reducing the risk of failure of an investment in range revegetation in some zones.
Several basic range planting techniques were tested in the Midelt area. The rangeland drill, which scarifies soil surfaces and seeds in areas of rough topography, did not prove effective during drought years. The need for preparing an adequate seedbed was only accentuated under such environmental conditions. Soil surfaces needed to be disturbed and competition from existing vegetation had to be reduced. Conventional seedbed preparation methods, using a plow, a disc and a roller, were most successful.
Because soils are eroded and compacted, rainfall generally
runs off without significantly penetrating the soil crust. Range improvement methods which increase penetration of water through the soil surface, and prevent runoff are potentially worthwhile investments. Ripping of the soil and the construction of terraces on the contour were highly successful range improvements

installed at Ain Beni Mathar, Gouttitier, and Ait Rbaa Grazing Perimeters. Ripping and terracing activities at the El Faija site, though not yet one year old, appear to have had a marked positive effect on revegetation and forage production.
Attempts to improve range productivity using chemical
fertilizers were unsuccessful. Fertilization with elemental sulfur at Ait Rbaa Grazing Perimeter did not produce detectable improvement after two years. Forms of sulfur that are more available in the soil may produce different results and should be investigated. Adding various amounts of nitrogen and phosphorous to rangeland soils near Midelt did not induce marked differences in plant density or production in drought years. In addition, fertilization is probably too costly to be included in range improvement programs unless very dramatic increases in production can be expected. The adage that "fertilization of soils receiving less than 300 mm of precipitation is valueless", appears to be correct in Morocco.
Range improvement research and species adaptability trials are necessary research which must continue. DE/SP must either decide to dedicate personnel and materials to research, or make
the proper contractual agreements to continue range research in identified topical areas through the auspices of existing research organizations (such as INRA, INAV, ENA, and programs such as SR-CRSP).
The following recommendations are offered based upon observation and research at the Project sites:
1) Continue to move forward toward control of grazing
through organization of people. No range improvement
concerning range vegetation can be made without
adequate control of animal numbers and the timing and
duration of grazing.
2) Improvement of Moroccan rangeland is possible through
manipulation of the soil surface and redistribution of
water. Soil ripping and construction of terraces
should be considered prior to seeding on some
rangelands. Where native residual plant cover still
exists, the treatment can be used to capture rainfall
and increase productivity without seeding.
3) Development and use of native species must be expanded
if a seeding program is to continue. Even the most
adapted introduced species can be difficult to
establish. Among the limited native species collection
initiated by the Project, there are promising forage
species which can be used to produce a nutritious,
highly palatable forage base capable of reestablishment
from seed under adverse conditions.
Only with a combination of range improvement techniques and development of adapted forage species can a range manager reduce the risk of failing to establish a stand. Only through reducing

risk of rangeland seeding can seeding of denuded range be a viable alternative accessible to producers. The acceptance of rangeland management in the private sector, however, is not the only consideration. Large scale government direction and investment is necessary in order to conserve degrading Moroccan range lands.
The continual push of dryland farming into marginally
productive areas has accounted for an irreparable loss of topsoil and the local extinction of many forage plant species. Government intervention and expenditure will be necessary to conserve the natural resources of soil and vegetation, even if only to protect watersheds. The continued investment in dryland farming, by the Moroccan government and USAID, without a balanced effort in the rangeland production sector, is a policy which heralds the deterioration and disappearance of more Moroccan rangeland. It is a program which will undoubtedly provoke dangerous ecological and economic consequences at the national



Animal production activities of the Moroccan Range Management Improvement Project were organized into two components. The first component, the Producer Study, was a survey of local livestock producers' activities at project sites. The study-was designed to gain an understanding of current management systems, production levels and available resources; to identify present and potential problems faced by livestock
producers; and to provide an entree to permit the introduction of improved methods to producers. An understanding of current practices, resources, and production levels realized by livestock producers, more effectively ensured that recommendations addressing observed critical problems were appropriate for producers' operations.
The second component consisted of cooperative demonstration and applied research programs with producers. In general, these activities were geared towards finding a workable solution to a problem detected in the Producer Study (first component), or as a means to introduce new ideas and techniques to producers. Three areas to improve livestock production were investigated in the second component:
- the introduction of the short scrotum technique as a means
of addressing the problem of indiscriminate breeding in
Moroccan sheep flocks,
- a sheep classification and selection program as a method
for eliminating defective animals from sheep flocks and
increasing meat and wool production, and
- an investigation into more efficient use of the limited
resources available to low income producers through
alternative livestock management techniques.
The Producers Study consisted of interviews with, and
observations of, selected livestock producers in order to obtain information on their management systems. Entire flocks owned by selected livestock producers were observed for management practices and general information on herd structure. Quantitative information was derived from subsamples drawn from each flock. Visits to producers' farms were made every four weeks to weigh animals and talk to producers. Intermediate visits were made as necessary. During interviews, flock locations, vegetation types grazed, kinds and quantities of supplemental feeds fed to the livestock, health care practices employed (both prophylactic and curative), and any other activities related to livestock management that had occurred since the previous visit were discussed. In the course of these interviews, information concerning replacement and culling criteria and related programs were gathered. It was also learned what livestock other than small ruminants were raised.

Information acquired through interviews would then be supported by observation of ongoing activities. People associated with livestock production, but not directly involved in the Producer Study, provided aid in keeping track of current livestock production activities and gave insights as to why certain management decisions were made.
The Producer Study design as originally conceived was altered as constraints were encountered. Technicians' work schedules were tight and their time limited. Therefore, in-depth investigations into the quality and quantity of available forage resources utilized and animal behavior were not possible. Rather, the effort was limited to recording kinds of forage grazed and general flock locations. Emphasis was placed on livestock operations and management decisions. The length of time available to conduct the study did not permit the collection of data with seasonal overlaps. Estimates had to be made to project production cycles over a calendar year.
Moroccan counterparts proved invaluable in the initiation of studies and in providing insights. However, counterparts rarely accompanied technicians on routine producer visits (an exception was in Beni Mellal where the Moroccan technicians actually conducted the study). Increased counterpart participation undoubtedly would have enhanced information exchange and the effectiveness of the data collection.
Though these constraints are acknowledged, the study did achieve its goal of providing a documented background understanding of Moroccan livestock production systems. Resources allocated to improve the livestock sector can now be more easily directed towards addressing critical problems that will be appropriate within a producer's operation.
Cooperation-from producers involved was good. Working
within the producers' systems to follow the animals' progress did have its drawbacks, but the benefits of the effort outweighed the negative aspects because it provided a clearer understanding of producers' attitudes and operations. Frustrations did occur. For example, tagged animals that were part of the sample occasionally disappeared, only to reappear in the next period without explanation.
Producers set aside time from their schedules to describe their systems and answer questions. Most producers aided in monitoring their animals and expressed an interest in weight fluctuations, having received copies of updated animal records during each visit. Technicians' advice and assistance was sought to overcome current problems from time to time. Field days on herd performance by TDY specialist, Dr. Jack Ruttle, were well received by producers. In addition, they participated in ram examinations and exchanged views on animal performance and care. The grading of wool generated less producer interest in some cases, though they readily conceded that kemp and hair was a negative factor effecting wool marketing. Nevertheless, the overall perception of producers' attitudes-by Project staff is concern about livestock production problems and genuine interest in methods to overcome them. However, the final acceptance of new methods or suggestions that they invest their own capital, is generally tempered with caution on the part of the producers.
At the inception of the study, one objective was not only to determine the vegetation types grazed by a flock, but the

quantity and nutritional value of the forage available as well. Measures of forage quality and quantity could provide a reference for explaining why weight fluctuations occurred. Initially, plots were clipped and fodder shrub presence estimated to quantify the amount of forage available to herds. Because of time and other constraints, both quality and quantity measurements were dropped from the study shortly after its inception. For the same reasons, studies in animal behavior pertaining to diet selection and animal activity were also eliminated from the study. Facilities were not readily available to analyze forage quality, therefore, data analysis and interpretation centered on forage quantity estimates and existing forage quality information.
It became apparent early in the studies that no excess forage was available (continuous grazing often maintained vegetation height below a level at which it could be clipped).
Diet selection was based on palatable plants present, and behavior was dictated by the herder, except during periods of high temperature when animals would cease searching for forage.
Further information that might have been gathered through the continuation of these studies did not warrant the time required to collect the data. Primary investigations did quantify the extreme conditions under-which livestock are produced, and enabled technicians to examine the areas grazed.
They were able to observe methods and daily procedures in herding the flocks.
During the fall of 1985, producers involved in the Producer Study had their sheep vaccinated against respiratory and gastrointestinal diseases, smallpox and enterotoxemia, and
-treated for parasites by the project. This activity was a demonstration of recommended health care practices and a display of appreciation for the producers' continued cooperation. As animals were treated, their sex and age were determined and recorded. Sex and age ratios were later established for each herd. The opportunity was also utilized to inquire further about producer attitudes towards animal health care and preventive health programs practiced.
A subsample of twenty ewe/lamb pairs was chosen from each producer's herd, identified with ear tags and monitored throughout the study. Lambs were monitored to establish growth rates and marketing tactics. Ewes were monitored for weight fluctuations throughout their reproductive cycles and for lambing frequency during the study. Lambs born to ewes in the subsample had their birth dates recorded, were weighed and ear tagged, and included in the subsample. Breeding dates were fixed by counting back from parturition dates. An effort was made to include lamb numbers that reflected sex and age proportions in the herd. Producers helped select the ewe/lamb pairs. The sample was probably biased towards higher producing animals as a result. One producer's herd contained less than the desired twenty ewe/lamb pairs so the entire herd was monitored. Ewes and lambs were weighed every twenty-eight days. Producers corral their sheep and goats at night without food or water. This management practice served as a shrink period prior to weight collections.
Wool quality and kemp presence were evaluated for ewes and lambs in the subsamples, and for breeding rams by Dr. Ruttle as part of an animal performance field day. Data from these animals

was used to indicate fleece quality in the producers' herds. Dr. Ruttle checked rams for epididymitis and collected and evaluated semen samples to identify ram fertility problems. Ewes in the subsample Were also examined for udder development and fertility problems. Since presence of a lamb was criteria for selection into the subsample, detection of infertility in the ewes was not expected and none was found. Although herd structure information was recorded, no ewes were examined for fertility outside the study subsamples. An indication of ewe fertility was given in another program where fertility was included as a criteria for selection into a breeding herd.
The Producer Study met with various levels of success at the different project sites. Production levels and operation management information gathered for the Middle Atlas and the Moulouya Plain (Azrou and Midelt) was good, and provided insights into limitations and potential of livestock production there. No attempt was made to examine current practices in Oujda. However, in Oujda, a complementary study to determine a possible method to control indiscriminate breeding, and an extension program to improve herd performance through selection, were carried out with project guidance and assistance. At Beni Mellal, the efforts to investigate production systems produced mixed results. A summary of producers' management decisions and operations was written, but initial data on herd performance was limited. Since his return from short course training in the United States in August 1985, Mohamed Driouich (Ingenieur d'Application in Beni Mellal), collected reliable production data. Unfortunately, the available time remaining in the Producer Study was too short to allow complete data collection, thus limited conclusions have been drawn from it.
Producers participating in the Producer Study were not
randomly selected, but were chosen to represent various segments of the livestock producer community. Producer willingness to cooperate during initial visits was also a major factor in final selection. To facilitate data collection and complete the study goals, it was more important to have willing participants than a statistically valid sample. When possible, the Project staff attempted to select producers who were also involved with the Project Agro-Pastoral Systems Study.
The results of the study are presented below. The results are segregated according to project sites to illustrate differences in livestock production that might occur across Morocco. A general description of the selected sample is given prior to results for each study site. If major differences occurred among producers within a site, these differences are noted. Given the method of sample selection and the limited sample size, the results are not statistically valid, but they do serve as indicators of sheep production and associated problems in areas surrounding project sites.

Current Livestock Production
Five producers participated in the study at Midelt.
Categorized by relative wealth, four were considered to be in the middle income category and one was considered to be in the low income category for livestock producers in the region. All have access to communal land. Two of the four middle income producers were also permitted to graze at least a portion of their flocks for part of the year on Plaine de L'Aarid Perimeter. The four middle income producers owned, or had access to cropland, and utilized crop residues as a forage source. Producer land holdings ranged in size from 10 to 25 hectares. Crops varied between producers, but all produced wheat and barley. Barley was grown almost exclusively for livestock consumption. The low income producer did not have access to crop residues and in general, had limited resources for livestock production. Flock numbers reflected the two groups' resource bases. The middle income producers owned between two and three hundred sheep, and the low income producer had less than twenty. Results will be presented separately for the two income levels.
Herd Characteristics
Sheep comprise the bulk of livestock owned in the Midelt
area (Figures 2.1 and 2.2). Goats, equines, and cattle are also present, but in much reduced numbers. The low income producer had roughly as many goats as sheep.
26 24
22 20
Z 14
S 10
Figure 2.1. Livestock species, class, and number owned by the
low income producer in the Midelt area sample.

260 24.0 220
X ISO140
I 100
S 60
Figure 2.2. Livestock species, class, and average numbers owned by the middle income producers in the Midelt area sample.
Low numbers of three year old ewes present in sheep flocks in Midelt (Figures 2.3 and 2.4) may be attributed to the recent drought. A combination of low and middle income producers' herds provided a ram to ewe ratio averaging approximately 1:21. Rams remain with the ewes throughout the year. One producer, however, owned 28 yearling rams that were to be marketed and these were kept separately from the ewes. These speculation rams were included in the determination of ram age distribution, but not to calculate the ram to ewe ratio.
9 k
o 5
2 I
Figure 2.3. Age and sex distribution of sheep owned by the low income producer in the Midelt area sample.

a 40
10 \ \ \ '
0 ilc
Figure 2.4. Average age and sex distribution of sheep owned by the middle income producers in the Midelt area sample.
The average weight for rams throughout the year in the
Midelt area was 33.6 kg, the average weight for ewes throughout the year was 28.8 kg, and the average weights for ram lambs and ewe lambs at nine months of age were 26.7 and 21.0 kg, respectively.
Flock Movement and Management
In Morocco, sheep and goats are generally corralled at
night. When animals are located near the producer's dwelling, the corral is generally an enclosed room. Predation is considered a possible problem. One kid, for example, was reported lost to a fox during the study.
All producers used communal land, keeping herds in areas
dominated by sagebrush (Artemisia herba-alba) from spring to late fall, and moving the animals to alpha-grass (Stipa tenacissima) for winter forage. The low income producer, however, began to graze alpha-grass sites in early fall. Access to communal land was possible for all producers from their permanent dwellings. In addition, one producer rented rights to a mountain pasture (Tacksmiert), where he grazed his flock from late April until January. Generally, animals belonging to this producer are moved
to Tacksmiert because of its abundant early season forage, and are returned to the area around his house when forage becomes scarce at Tacksmiert and severe weather imperils animal health. All the middle income producers had access to crop residues. Except for the producer whose sheep were moved to the mountain meadow, crop residues were utilized from June to August. Flocks spent most of the day grazing residues, and grazed native vegetation while moving between crop residue sites.
Two producers were allowed to graze animals on Plaine de
L'Aarid Perimeter from June to December. Both of these producers took advantage of this right, except for a period in June and

July when one producer pulled his flock off the perimeter to graze crop residues. The other producer maintained a split herd with one flock on the perimeter and the other on crop residues.
The degree to which flocks are divided for management
purposes depends heavily on labor availability. One producer combined his flocks in the fall because of a labor shortage resulting from his children returning to school. While rams are run with the ewes throughout the year, most producers separate lactating ewes from the herd to facilitate supplemental feeding programs. A few of the producers included late gestation ewes in this group. Often, young (less than one month old) and weak lambs were kept at the house during the day and returned to the ewes at night. The flock grazed together during the day and then the older lambs (5 months plus) were penned separately at night. The low income producer stated that the reason he did not wean his lambs prior to their being sold was because he felt that milk is a constant and free supplement for lambs (the literature indicates that the ewe is essentially dry 4 months after parturition, and the stress caused by the constant annoyance of the older lambs often causes ewes to fail to rebreed).
Barley was the most common form of supplemental feed and was used by all producers. Rates ranged from 0.2 to 0.5 kilograms per animal per day, depending upon forage availability and the producer. Other supplements fed to livestock included corn, corn stover, sugar beet pulp, alfalfa, straw and Cicalim (Cicalim is a commercially available protein supplement). Producers fed supplements to the entire herd beginning in December and continued until forage became more readily available in the spring. Lactating ewes and lambs were fed supplements earlier in the fall, as the decline in available forage dictated.
Natality, Mortality -and Marketing
Lambs were born throughout the year, but there was a higher incidence of parturition during certain periods. Fall (Octoberearly November), winter (early mid December), and spring (mid January early March) were noted as the most active lambing periods during the study. All ewes selected for the study subsample had lambs at the time they were selected. Information gathered in another study (Selection and Classification Program) indicated that approximately fifteen percent of the ewe herd could be expected to have reproductive problems and be incapable of producing a lamb. Since presence of a lamb was criteria to be selected for the subsample these infertile animals were not candidates to be included.
Fifty-seven percent of the lambs present at the initiation of the study were born in the Spring '85 period. Thirteen percent of the lambs were born in the Fall '84 period and thirty percent were born in the December '84 period. Ninety-three percent of the ewes in the subsample gave birth again during the study. Four percent of these second lambs were born within eight to ten months after the previous lamb. Fifty percent were born twelve months after the initial study lamb. Thirty-eight percent to as much as forty-five percent were born fourteen to sixteen months later, and one to a maximum of eight percent (if all ewes in the subsample lambed again) were born at least eighteen months after the first lamb. Numbers of second lambs born during the

spring, winter, and fall periods were in similar percentages as the original lambs. Percentages of ewes, whose original lambs were born in t he spring, winter, and fall periods, that lambed again within a twelve month period were 46%, 3% and 82% by period, respectively (Table 2.1). The data indicate that only approximately half of the ewes in the subsample (the productive group) produced a lamb on an annual' basis during the study. When the entire ewe population is considered, the percentage of ewes producing lambs on an annual basis was even lower. This low productivity is a major problem facing Moroccan livestock producers.
Table 2.1. Lambing intervals and number of lambs born during
each interval in the Midelt area sample.
OF THE STUDY 8 10 12 14 16 18 TOTALS
FALL 84 13 PAIRS (13%) 0 ( 0%) 0 ( 0%) 6 (6%) 4 (4%) 2 (2%) 1 (1%) 13 (13%) DECEMBER-04 30 PAIRS (301) 0 ( 0%) 1 ( 1%) 0 (0%) 22 (22%) 2 C2%) *25 (25%) SPRING 84 57 PAIRS (57%) .2 ( 2%) 1 ( 1%) 44 (44%) 8 ( 8%) 55 (55%) TOTALS 100 PAIRS (100%) 2 ( 2%) 2 ( 2%) 50 (50%) 34 (34%) 4*( 4%) 1*( 1%) 9T3 -(93%) A) Percent of the number of ewe-lamb pairs is shown in parentheses.
*The study ended prior to these lamibing intervals.
Information on expected animal death loss percentage was acquired through interviews with the producers. The middle income producers reported that approximately three percent of the breeding herd and seven percent of the lambs (less than nine months of age) died annually. The low income producer said that losses varied with the availability of forage, but that death losses could be expected to be as high as thirty percent of the breeding herd and fifty percent of the lambs annually during a harsh year. Infertile ewes, low annual lambing rates, and death losses combined to produce a limited lamb crop for the Midelt region. The lamb crop, calculated as the number of lambs raised to a marketable age annually divided by the number of ewes in the breeding herd, was determined to be 42% for the middle income producers and 35% for the low income producer in the Midelt sample.
The majority of the middle income producers' lambs that were sold from the subsample were born during the Winter '84 and Spring '85 periods, and sold during July and August, 1985. That period corresponded to Aid El Kbir when prices generally rise. One producer sold 5 one year old rams per week between late June and late August-to benefit from the price increase. Smaller numbers of animals were sold throughout the remainder of the year, as income was required. As a result of-high lamb death loss (Table 2.2), and the need to replace lost ewes to maintain herd size, the low income producer did not sell or slaughter any animals during the period in which the study was conducted.

Table 2.2. Distribution of sales, slaughters and deaths of lambs and ewes in the Midelt area sample. [Numbers of animals are
shown by data collection periods (28-day intervals). The study began April 4, 1985 and ended March 6, 1986.]
Ie sus _aniets. .in j. .0 .i .O 20
DurI n a3 on the health
w l hn l Pi r t p t
o 0 0 0 0 0BT Z 0
ae T t de tIT retments, a dags prod t
producer' sso 0 0oats1were
Octbe A1AT985 ET 0 olde l0
- r- -7 -f 1 Tr --=-ano moT so
-. --. T 'r--.---- I ___,m T T1 -a a a 1a u a isz, o p
ts a t n SOn 0 od a 0so
ISI M -1,
Panicu o hi s f h i 0 0
osoyn ~ 0 0 0 0 0 17 0
0237-3 SOLD a I I 0 0
I P TI 7II C. 0 a
rd -S--OL"-D I I --: I-"' 11=1111 -SO'-LO
.... .... ... s ,io ... !
Health Care
During the study, producers relied heavily on the health care provided by the Project. It is not known what practices
would have been followed if the Project program had not been available. Treatment dates, treatments, and dosages provided the producers' sheep and goats were:
September 1985 Valbazen (2.5 ml/kg of bodyweight) October 1985 Enterovac (adults and older lambs 2 ml; less
than 3 months I ml)
November 1985 Ovipox (0.5 ml to animals over 3 months)
February 1986 Valbazen (2.5 ml/kg of bodyweight)
Sheep that were moved onto Plaine de l'Aarid Perimeter received Enterovac in June 1985. This treatment is required for all animals that use the perimeter and is provided at no cost to the producer. Two of the middle income producers provided other treatments at their own expense. One producer administered Panicur to his sheep for what he described as stomach problems in August, and the other producer treated his animals with Coglovax
8 in June and with Ranide in August. The low income producer treated his lambs and kids twice in January with Enterovac and Combiotic D to treat diarrhea. Death losses were high due to this illness. A list of medicines and their uses is shown in Table 2.3.

Table 2.3. Medicines administered to sheep and goats owned by producers in the Producer Study.
ANTELVETE Lung Parasites Oral C
COGLOVAX 8 Enterotoxemia Injection P
COMBIOTIC D Antibiotic Injection C
ENTEROVAC Enterotoxemia Injection P
ENTEROVAL Enterotoxemia Injection P
EXHELM II Internal Parasites Oral C
MATELVELE Enterotoximia Injection P
MULTIVAC Enterotoxemia Injection P
OVIPOX Smallpox Injection P
PANICUR Gastrointestinal/Resp. Parasites Oral C RANIDE Internal Parasites Oral C
RANIZOLE Internal Parasites Oral C
VALBAZEN Internal Parasites Oral C
Ram Fertility
During the course of the study in the Midelt region,
nineteen rams were checked for fertility problems. Three of the nineteen rams (15.8%) were sterile. Two of these (10.5% of the rams examined) had epididymitis. Five other rams (26.3%) were considered to be of marginal fertility with low sperm motility (50 75% motility) and slow movement and/or poor sperm concentration. Eleven (57.9%) of the rams checked were sound with motility at 80 to 90 percent, and with good movement and sperm concentration.
Replacement and Cull Animals
Three of the five producers in the study selected young
female lambs as replacement ewes based primarily on the criteria of survivability. The other two indicated that overall appearance influenced their choice with weak lambs being eliminated. One producer said thickness of tail and amount of meat on the backbone, signs of a healthy animal, were the criteria he used. As a result of these selection criteria, and a desire to increase herd size, most female lambs were kept and placed in the breeding herd.
Age appeared to be the chief criteria for culling animals. Weak animals may first be given supplements to improve their vigor, and if improvement does not occur, they are sold. One producer listed his criteria for culling animals as age, lamb production, health, appearance, condition of teeth, and eating habits. Culled animals are usually sold at the local souk (market). Few if any producers weigh animals or keep written records.
Weight Gains for Ewes and Lambs
Ewes showed a tendency to gain weight during the period from

April through June and to lose weight during the other months. Weight fluctuations reflected the availability of forage. Contrary to what might be expected, ewes often lost weight throughout gestation (Figures 2.5 and 2.6). This nutritional problem no doubt at least partially explains why reproductive rates and weaned lamb production are so low. In addition, internal parasite load tends to amplify the problem.
W" 0.050.04
0.01 0.00
n -0.01 I
-0.06,P-1 P-2 P-3 P-4 P-5 P-6 P-7 P-8 P-9 P-10 P-1I P-12
Figure 2.5. Average daily weight change during twelve, 28-day periods for ewes owned by the low income producer in the Midelt area sample.
,., 0.0a8
0 --A1
-0.030 03
,. -0.04-0.08P-1 P-2 P-3 P-4 P-5 P-6 P-7 P-8 P-0 P-I0 P-1I P-12
Figure 2.6. Average daily weight change during twelve, 28-day periods for ewes owned by the middle income producers in the Midelt area sample.

Weight changes were similar for both low and middle income producers' lambs though lambs belonging to the middle income producers were generally larger.
The original study lambs gained weight from the study inception until June. Weight gains stabilized from June to December, and then weights declined until February when slight gains were again seen. The low income producer had a summer lamb which failed to gain much weight after the first two months (a point when the ewe probably had little if any milk) and was soon surpassed by fall lambs. New lambs born to the ewes included in the subsample (fall, winter and spring periods) all showed gains during the study (Figures 2.7 and Fig. 2.8).
12 SPRI -85
M 108
FAL 85
2 SUM PR 86
4/4 5/2 5/30 6/27 7/25 8/22 9/1910/11/1412/12 1/6 2/6 3/6
Figure 2.7. Average weights of lambs owned by the low income producer in the Midelt area sample (~28-day intervals; April 4, 1985 through March 6, 1986).
S 26
24- FA 84
18 16
14- SP -85
M 1210 8-,
4,4 5/2 5/30 6/27 7/25 8/22 9/1910/1711/12/12 1/6 2/6 .3/ WEIGHT COLLECTION DATES
Figure 2.8. Average'weights of lambs owned by the middle income producers in the Midelt area sample (~28-day intervals; April 4, 1985 through March 6, 1986).

Wool Production
Sheep were shorn in the Midelt region from mid May to late June, These periods corresponded to just before and after Ramadan. Average raw fleece weight was approximately 1.5 kg per ewe and 2.5 kg per ram, or about 1.6 kg per breeding animal.
Only ewe-lamb pairs in the subsample and rams were evaluated for wool quality and kemp presence. Spinning count wool grades ranged from 40 to 56 for the ewes and lambs (a spinning count of 50 is considered the lower limit of the fine wools). Ewes were evenly spread between the 40, 44, 50 and 54 grades. The majority of lambs and all of the rams fell in the 40 and 44 grade ranges. Kemp was present on 26% of the ewes (Table 2.4), 34% of the lambs (Table 2.5), and 36% of the rams (Table 2.6). No difference in wool quality was observed between the low and middle income producers' sheep. Therefore, the results were aggregated to present the results.
Table 2.4. Ewe wool quality and presence of kemp in the Midelt area sample (0 is no kemp and ++ is abundant).
RANKING 40 42 44 46 48 50 52 54 56 ANIMALS
+ 0 0 0 0 0 0 0 0 0
+ 5 0 1 0 0 0 0 0 0 6 8 0 4 0 0 8 0 5 0 25 0 16 0 30 1 1 21 0 17 2 88
ANIMALS 29 0 35 1 1 29 0 22 2 119
Table 2.5. Lamb wool quality and presence of kemp in the Midelt area sample (0 is no kemp and ++ is abundant).
RANKING 40 42 44 46 48 50 52 54 56 ANIMALS
++ 3 0 0 0 0 0 0 0 0 --3+ 1 0 1 0 0 0 0 0 0 2 23 0 7 0 0 0 0 0 0 30 0 46 0 11 1 0 6 0 3 1 68
AMIMALS 73 0 14 1 0 6 0 3 1 103
Table 2.6. Ram wool quality and presence of kemp in the Midelt area sample (0 is no kemp and ++ is abundant).
RANKING 40 42 44 46 48 50 52 54 56 ANIMALS
++ 0 0 0 0 0 0 0 0 0 0 + 1 0 0 0 0 0 0 0 0 1 3 0 0 0 0 0 0 0 0 3 0 6 0 1 0 0 0 0 0 0 7
ANIMALS 10 0 1 0 0 0 0 0 0 11

The majority of wool produced is kept for domestic use (blankets, sleeping pads, djellabas). Only one producer regularly sells wool, selling half and keeping half for domestic use. Wool sold was marketed at local souks. Another producer indicated that he sold wool when he had extra, and felt the quality was sufficiently good to draw a high price. He judged quality by weight, the heavier the fleece the higher the price. When wool is sorted in the souks prior to being resold, weight is often used as the criteria for quality.

Current Livestock Production
Three producers participated in the Producer Study at Timahdite (Azrou). Categorized according to wealth and resources, all three producers would be considered to be in the middle income range for the Timahdite region. All had access to communal land, owned and leased cropland, and utilized crop residues as a forage source. The size of the land holdings ranged from 20 to 53 ha of dryland, and 9 to 71 ha of irrigated cropland. All producers grew wheat and barley. Areas planted to wheat in 1985 ranged from 8 to 12 ha. Land planted to barley which was grown primarily for livestock consumption ranged from 10 to 14 ha. Flocks owned by producers in the Producer Study ranged in size from 215 to 349-sheep in the fall of 1985.
Herd Characteristics
Sheep account for most of the livestock in the Timahdite
region (Figure 2.9). Goats, equines, and cattle are also owned, but in considerably smaller numbers. Numbers of animals for each age group are fairly consistent for both sexes (Figure 2.10). Although, by animal number count, the ram to ewe ratio is approximately one to twenty-nine, the producers claim that during the breeding season the ratio is one ram for every 32 to 40 ewes. Rams remain with the ewes throughout the year.
280 260 240
V) 220 M 200
140 120
40 20 0
Figure 2.9. Livestock species, class, and average numbers owned
by the producers in the Timahdite area sample.

4 0
0. -0 Z
Figure 2.10. Average age and sex distribution of sheep owned by the producers in the Timahdite area sample.
The average weight for rams throughout the year in the
Timahdite region was 34.7 kg, the average weight for ewes was 29.9 kg, and the average weight for ram lambs and ewe lambs at nine months of age were 23.6 and 21.6 kg, respectively.
Flock Movement and Management
Sheep are corralled at night in an enclosed area next to or in the producer's house, or in a corral constructed next to the herder's tent. From April through July, the producers kept their sheep at their houses and grazed communal land within a 3 to 5 kilometer -radius. In August the flocks were split, with lactating ewes and old, poor condition ewes being kept at the house, fed supplements, and allowed to graze crop residues. The lambs were separated from the ewes during the day and placed with the ewes at night. As the lambs developed they were selected on an individual basis to graze with the ewes. Weaned spring lambs were often kept with this herd. The other animals were moved away from the house and corralled next to tents.
During the study, two of the producers herded their flocks
away from the house on crop residues. They had moved their sheep in order to have easier access to the fields. The third producer moved his animals to a mountain range in order to take advantage of sagebrush and thymus (Thymus capitatus) range that was still relatively high in forage availability. The mountain range was depleted by early August and the flock was returned to the house to join the other sheep on crop residues. In late fall, the producers returned their herds to communal land and grazed the
regrowth of native vegetation. Late gestating and lactating ewes were kept separate from the rams, dry ewes, and older lambs. From January through March, the animals were kept indoors at night, fed supplements and allowed to graze communal land, as weather permitted.

Rams received approximately 0.4 kg of barley per day
throughout the year. Salt blocks are available year round in the
corrals. Starting in late summer, spring lambs and old ewes are fed supplements. The rates and kinds of supplements fed varied
among producers. For example, one producer fed lambs alfalfa
free choice and 0.16 kg of Granulant 90 (Granulant 90 is a
commercially prepared protein supplement) per lamb per day with
some barley. Another producer fed only those lambs that were
eventually to be sold daily rations of 0.1 kg of barley and 0.2
kg of Cicalim per animal. Old ewes received 0.2 kg of barley
each day. The third producer fed lambs and old ewes 0.2 kg of barley per head per day. As the producer prepared to sell the old ewes, he would increase their barley ration to 0.8 kg/day.
In late fall, all of the producers began to feed those ewes that
had lambs 0.15 to 0.3 kg of barley per ewe per day. Once the
weather became severe, producers supplied hay to the entire herd.
Natality, Mortality and Marketing
The producers reported that most lambs were born between
October and November, or between March and May, though lambs could be born at any time during the year. As in the Midelt
area, all ewes selected to be in the study subsample had lambs at
the time that they were tagged. Twenty-five percent of the
initial lambs were born in the fall period (October November),
27% of the lambs were born in the winter period (December early
January) and the remaining 48% in the spring period (March
May). Eighty-seven percent of the ewes in the subsample produced
lambs again during the study. Forty-five percent of the ewes
lambed within six to ten months of their previous lamb. Thirtytwo percent lambed twelve months after the initial study lamb.
Ten percent to as many as 23% lambed fourteen to sixteen months
post the first lamb, and one to a maximum of 15% (if all ewes in the subsampleo lambed again) lambed at least eighteen months after
their previous lamb. Percentages of ewes whose original lambs were born in the fall, winter, and spring periods, that lambed
again within a twelve month period, were 71%, 68% and 83% by
Period, respectively (Table 2.7).
Table 2.7. Lambing intervals and number of lamb: born during
each interval in the Timahdite area sample.
OF THE STUDY,. 6 8 10 12 14 16 TOTALS
FALL 84 14 PAIRS (23%) 0 ( 0%) 4 (7%) 0 (0%) 6 (10%) 2 ( 3%) 1 (2%) 13 (22%) WINTER-84 16 PAIRS (27%) 1 ( 2%) 0 (0%) 6 (6%) 4 ( 7%) 3 ( 5%) 14 (23%) SPRING-85 30 PAIRS (50%) 0 ( 0%) 12 (20%) 4 (7%) 9 (15%) 25 (42%) TOTALS 60 PAIRS (100%) 1 ( 2%) 16 (27%) 10 (17%) 19 (32%) 5*( 8%) 1*( 2%) 52 (87%)
() Percent of the number of ewe-lamb pairs is shown in parentheses.
* The study ended prior to these lambing intervals.

The Timahdite producers reported that approximately 4% of the breeding herd died annually, and that an annual lamb death loss of approximately 9% could be expected. The lamb crop was estimated to be 57% for the producers in the Timahdite sample.
Lambs are primarily sold between June and November. Older
lambs are sold first. Old, and a few young ewes, with or without a lamb, are sold between August and October. Livestock will also be sold throughout the year, as the producer's need for additional income dictates. One producer felt that prices for animals were directly related to available forage. Speculation on Aid El Kbir is also a factor affecting the sale of lambs. One producer saved most of his male lambs in order to sell them when prices were high prior to Aid El Kbir. Another producer bought spring lambs in May and sold them in August prior to the festival (Table 2.8).
Table 2.8. Distribution of sales, slaughters, and deaths of lambs and ewes in the Timahdite area sample. [Numbers of animals are shown by data collection periods (28 day intervals). The study began April 6, 1985 and ended February 11, 1986.]
(14) 112) (30) (5) 117) (13) (3) (60) DIIED -0 -0-0---01 -o o o o
ABATOI 0 0 0U 0* .0 0U 0- 0 FIRST SOLD 0 0 0 0 0 0 0 ABATTOIR 0 0 0 0 0 0 0 0
DIED T T T T 0 u0 0 0-r SIOND SOLD 0 0 0 0 0 0 0 0 ABATTOIR 0 0 0 0 0 0 0 0
FOUTH SOD 0=ir 1 20- 0 T 0 0 2= ABATTOI 0 0 0 0 0 '0 0 0 THIRD SOLD o 0 0 0 0 0 0 1 ABATTOIR 0 0 0 0 0 0 0 0 NED~ 0r- -U- "U- 7 U- 70 FOURTH SOLD 0 1 2 0 0 0 0 2 ABATTOIR 0 0 0 0 0 0 0 0 S T ED 0 0U 0-- -0- 0 FIFTH SOLD 6 0 2 0 0 0 0 3 ABATTOIR .0 0 0 0 0 0 0 0 DIE 7 0 0 ---" 0 U = SIXTH SOLD 1 1 0 0 0 0 0 3 ABATTOIR 0 0 0 0 0 0 0 0 DZ 0 0 0- 0I 0 o
SEVENTH SOLD 0 0 0 0 0 0 0 3 ABATTOIR 0 0 0 0 0 0 0 0
EIGHTH SOLD 1 0 2 0 0 0 0 0 ABATTOIR 0 0 0 0 0 0 0 0 DIED T T 0r 0 T T 0 NINTH SOLD 0 0 0 0 0 0 0 0 ADATTOIR 0 0 0 0 0 0 0 0
-D -T- 7 1 U -r
TENTH SOLD 0 0 0 0 0 0 0 0 ABATTOIR 0 0 0 0 0 0 0 0. DIEDf 0u 0r -V -6 "-- -u U '0-ELEVENTH SOLO 0 0 0 0 0 0 0 0 ABATTOIR 0 0 U 0 0 0 0 0 ;Maximum nudr of T.gnimrTnt sEi 5 pre for ea-c-0 oy
is shown in prenthesis.)
Health Care
One producer belonged to the local ANOC (National Sheep and Goat Association) and received health care for his animals